The World of Maritime and Commercial Law: Essays in Honour of Francis Rose 9781509932429, 9781509932436, 9781509932412

Professor Francis Rose stands at the forefront of maritime and commercial law scholarship. This collection of essays hon

269 34 6MB

English Pages [427] Year 2020

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

English and European Perspectives on Contract and Commercial Law: Essays in Honour of Hugh Beale 9781474201537

The purpose of this book is to honour the influential and wide-ranging work of Professor Hugh Beale, who retires from fu

269 25 3MB Read more

Social Justice and the World of Work: Possible Global Futures Essays in Honour of Francis Maupain 9781509961252, 9781509961283, 9781509961276

In this book, leading international thinkers take up the demanding challenge to rethink our understanding of social just

302 63 6MB Read more

Law, Morality and Society. Essays in honour of H.L.A. Hart

1,282 97 43MB Read more

Commercial Maritime Law 9781849466752, 9781474202954, 9781509901043

The title ‘Commercial Maritime Law’ is a misnomer. There is a patchwork of different commercial maritime laws around the

286 126 2MB Read more

The World of St. Francis of Assisi: Essays in Honor of William R. Cook 9004270981, 9789004270985

In The World of St. Francis of Assisi, thirteen scholars from diverse academic disciplines offer accounts of the way in

904 137 5MB Read more

Law, Life, and the Teaching of Legal History: Essays in Honour of G. Blaine Baker 9780228012252

Exploring Canadian legal history through the life and career of historian Blaine Baker. Using the life and intellectua

115 96 8MB Read more

The Method and Culture of Comparative Law: Essays in Honour of Mark Van Hoecke 9781474201636, 9781849466233

Awareness of the need to deepen the method and methodology of legal research is only recent. The same is true for compar

263 108 5MB Read more

Essays in the History of Canadian Law: In Honour of R.C.B. Risk 9781442620797

The collected essays in this volume represent the highlights of legal historical scholarship in Canada today. All of the

166 78 5MB Read more

Essays in Honor of Louis Francis Solano

245 55 10MB Read more

The Common Law of Intellectual Property: Essays in Honour of Professor David Vaver 9781472565198, 9781841139708

This collection of essays was written in honour of David Vaver, who recently retired as Professor of Intellectual Proper

322 76 3MB Read more

The World of Maritime and Commercial Law: Essays in Honour of Francis Rose
9781509932429, 9781509932436, 9781509932412

Author / Uploaded
Charles Mitchell
Stephen Watterson (editors)

Table of contents :
Preface
Contents
List of Contributors
Table of Cases
Table of Legislation
Table of Treaties and Conventions
1. The Empress of Ireland and Mont Blanc Collisions: Then and Now
I. The Titanic and Limitation of Liability
II. Empress of Ireland
III. The 1917 Halifax Collision and Explosion90
IV. A Twenty-First Century Maritime Law Perspective on the Disasters
2. Do Actual Carriers Require Special Treatment?
I. What is an Actual Carrier?
II. A Special Regime for Actual Carriers under Conventions
III. The English Law Position
IV. Reasoning Behind a Special Regime for Actual Carriers
V. Possible Solutions
VI. Does the Problem Require a New Approach?
VII. Two Further Issues
VIII. Conclusion
3. Multimodal Carriage of Goods by Sea: Time for an International Convention?
I. Introduction
II. Five Problems with Multimodal Carriage of Goods
III. Solutions to the Problems
IV. Conclusion
4. The Three Ages of Utmost Good Faith
I. Introduction
II. The First Age: The Age of Common Law
III. The Second Age: The Marine Insurance Act 1906
IV. The Third Age: The Insurance Act 2015
V. Conclusion
5. English Jurisdiction Clauses in Marine Cargo Insurance and Direct Actions against Cargo Liability Insurers
I. Introduction
II. Party Autonomy
III. Direct Action
IV. The Global Position
V. The European Position
VI. The English Common Law
VII. Conclusion
6. Choice of Law in Determining the Ownership of Ships
I. Introduction
II. The Law of the Underlying Transaction
III. The Law of the Flag
IV. The Law of the Ship's Actual Situs
V. Ship Arrest - Lex Fori
VI. Conclusion
7. Private International Law and the Privy Council
I. Introduction
II. The Effect of Statutes
III. Developing the Common Law
IV. The Limits of the Common Law, and Cambridge Gas
V. The Derivation of Rules from Statutes
VI. The Other Side of the Coin: Trusts and Trustees
VII. Common Laws, Private International Laws and the Privy Council
8. Bulk Goods and Title Transfer
I. Introduction
II. The Seller in Possession
III. Bulk Goods and Oversales
IV. Conclusion
9. The Duty to Take Delivery of Goods
I. Introduction
II. Is there a Free-Standing Take-up Duty to Accept Delivery?
III. The Legal Basis of a Take-up Duty
IV. A Take-up Duty: Fault-Based or Strict?
V. The Take-up Duty: Its Application to Specific Cases
VI. Effect of Breach of the Take-up Duty
VII. Conclusion
10. The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century
I. Introduction
II. Some Brief Historical Comments
III. The Sale of Goods Act 1979
IV. 'Transfer the Property'
V. 'Goods'
VI. Money Consideration
VII. What is Left for the Sale of Goods Act 1979?
VIII. Does it Matter?
IX. Conclusion
11. Are there Any General Principles of Commercial Law?
I. Introduction
II. Is there Such a Thing as 'Commercial Law' ?
III. Modifying Private Law Dogmas
IV. The Leading Advocate of General Principles of Commercial Law
V. Freedom of Contract versus Sanctity of Contract
VI. The Limits of Party Autonomy
VII. Party Autonomy, Certainty and Good Faith
VIII. Facilitation, Intangible Assets and Cryptoassets
IX. A Conclusion of Sorts
12. Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850
I. Introduction
II. Commercial Drafting Practices and Mercantile Usage
III. Construction of Contracts
IV. Implication of Terms
V. Concluding Remarks
13. 'The Obscure, the Implied and the Illegal': English and French Approaches to the Interpretation of Written Contracts, Implication of Terms and Contracts Affected by Illegality
I. Introduction
II. Interpretation of Written Contracts
III. Implication of Terms
IV. Illegality
V. Conclusions
14. The Interpretation of Written Contracts
I. Interpretation as Initially a Type of Translation
II. Relaxation of the Translation Approach Subject to an Ambiguity Constraint
III. Relaxation of the Ambiguity Constraint
IV. 'The Intention of the Parties' as the Underlying Interpretative Principle
V. 'Intention of the Parties' and Ambiguity
VI. 'Intention of the Parties' and Rectification
VII. 'Intention of the Parties' and Implication of Terms
VIII. Conclusion
15. The Boilerplate and the Bespoke: Should Differences in the Quality of Consent Influence the Construction and Application of Commercial Contracts?
I. Introduction
II. Boilerplate – A Short History
III. Incorporation
IV. Contractual Interpretation
V. Integration: Entire Agreement Clauses
VI. Variation
VII. Agreement as Evidence
VIII. Conclusion
16. Illegality: Pleading, Proof, and Presumptions
I. Introduction
II. The Bowmakers Rule
III. What Sort of Rule is the Bowmakers Rule?
IV. The Gascoigne/Chettiar Rule
V. What Sort of Rule is the Gascoigne/Chettiar Rule?
VI. The Gascoigne Authorities
VII. The False Assimilation of the Two Rules in Tinsley v Milligan
VIII. The Gascoigne/Chettiar and Bowmakers Rules Distinguished
IX. Conclusion
17. Negotiating Damages
I. Introduction
II. Morris-Garner v One Step (Support) Ltd
III. The Purpose of Negotiating Damages
IV. The Availability of Negotiating Damages
V. Conclusion
18. The Myth of Common Law Tracing
I. Introduction
II. Leading Cases
III. The Basics of Tracing
IV. Early Common Law Tracing Approaches
V. The Modern Common Law Authority: FC Jones
VI. Reasons against Adopting the Same Tracing Rules at Common Law and in Equity
VII. Conclusion
19. Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies
I. Introduction
II. A Brief Introduction to Cryptocurrencies
III. Contextual Challenges
IV. Conceptual Foundations
V. Conclusion
20. Silence and Solidarity? The Duties of Individual Directors Minded to Speak Out about their Board's Decision-making and Governance
I. Introduction
II. Some High-Level Observations
III. The Two Cases to be Studied
IV. Stobart Group Ltd v Tinkler
V. National Roads and Motorists' Association Ltd v Geeson
VI. Returning to the Substantive Issues and Seeking Answers
VII. Conclusions

Citation preview

THE WORLD OF MARITIME AND COMMERCIAL LAW Professor Francis Rose stands at the forefront of maritime and commercial law scholarship. This collection of essays honours his outstanding contribution. Twenty prominent scholars, judges and practitioners expound and debate a wide variety of important issues and controversies in the areas of maritime collisions, marine insurance, carriage of goods by sea, private international law, contract law, sale of goods, unjust enrichment, and directors’ duties.

ii

The World of Maritime and Commercial Law Essays in Honour of Francis Rose

Edited by

Charles Mitchell and

Stephen Watterson

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2020 Copyright © The editors and contributors severally 2020 The editors and contributors have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2020. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Mitchell, Charles (Charles Christopher James), editor. | Watterson, Stephen, 1975- editor. | Rose, F. D. (Francis D.) honouree Title: The world of maritime and commercial law : essays in honour of Francis Rose / Charles Mitchell, Stephen Watterson. Description: Oxford ; New York : Hart, 2020. | Includes bibliographical references. Identifiers: LCCN 2019055600 (print) | LCCN 2019055601 (ebook) | ISBN 9781509932429 (hardback) | ISBN 9781509932405 (Epub) Subjects: LCSH: Commercial law—England. | Maritime law—England. Classification: LCC KD1629 .W684 2020 (print) | LCC KD1629 (ebook) | DDC 346.4207—dc23 LC record available at https://lccn.loc.gov/2019055600 LC ebook record available at https://lccn.loc.gov/2019055601 ISBN: HB: 978-1-50993-242-9 ePDF: 978-1-50993-241-2 ePub: 978-1-50993-240-5 Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

PREFACE This festschrift for Francis Rose honours a distinguished scholar of maritime and commercial law. As he enters retirement and we look back on his multifarious activities in these fields, spanning a period that is longer than he might wish us to mention, what do we find? For one thing, the authorship of many scholarly works, the enumeration of which would greatly lengthen a book that is already long, but of which we must mention definitive accounts of salvage, general average, marine insurance, admiralty claims, carriage of goods by sea and overseas sales; and, back on dry land, thoughtful studies of doctrinal issues in commercial law and the law of obligations. His interests are broad, his understanding deep, his analysis strong, his writing concise and clear. To discover why English maritime and commercial lawyers are so well regarded around the world, one could do no better than spend some profitable hours – or weeks – reading Francis’s backlist. Distinguished though they are, however, these productions do not give the full measure of Francis’s contribution, nor do they completely explain why he is held in such high esteem by his colleagues. For this, one must also take note of the care, concern and thoughtfulness with which he has pushed his subjects forward through his activities as an assiduous organiser of scholarly happenings and as a truly great editor of books and journals. Particular mention must be made here of his work over many years for the British Maritime Law Association and for the Society of Legal Scholars, as its first Subjects Sections Secretary and as convener of its Restitution Section; of the many workshops on unjust enrichment (and other matters) that he organised with Peter Birks, during a period on which many legal scholars now look back as a golden age of intellectual excitement and productivity; and of his work as editor of the Lloyd’s Maritime and Commercial Law Quarterly (eagerly awaited bedside reading for every aficionado), founding editor of the Company, Financial and Insolvency Law Review (now evolved into the Journal of Corporate Law Studies) and founding editor of the Restitution Law Review (which built the subject for 25 years and changed the landscape of English private law in the process). Gathering together the contributions to this book, we heard many stories of the interest that Francis showed in recognising and cultivating talent in young colleagues, of the imagination with which he commissioned new pieces of writing, matching subjects to authors with a marvellous understanding of who could best do what, and of the kindness, patience and enthusiasm with which he encouraged, enticed and cajoled his colleagues to bring out the best in them. This kind of work is not much done in the public eye, nor is it work for which plaudits are handed out nearly enough in the modern world of legal academia. But it is work that matters and it is work that wins you the admiration, affection and gratitude of your peers. This collection shows that. Thank you, Francis, on behalf of all the contributors to this volume, and of maritime and commercial lawyers everywhere. Charles Mitchell Stephen Watterson

vi

CONTENTS Preface�� v List of Contributors�� ix Table of Cases�� xi Table of Legislation�� xxxix Table of Treaties and Conventions�� lv 1. The Empress of Ireland and Mont Blanc Collisions: Then and Now�� 1 Nicholas Gaskell 2. Do Actual Carriers Require Special Treatment?�� 25 Francis Reynolds 3. Multimodal Carriage of Goods by Sea: Time for an International Convention?�� 41 Simon Baughen 4. The Three Ages of Utmost Good Faith�� 63 Howard Bennett 5. English Jurisdiction Clauses in Marine Cargo Insurance and Direct Actions against Cargo Liability Insurers�� 89 Yvonne Baatz 6. Choice of Law in Determining the Ownership of Ships�� 107 Martin Davies 7. Private International Law and the Privy Council�� 123 Adrian Briggs 8. Bulk Goods and Title Transfer�� 139 Michael Bridge 9. The Duty to Take Delivery of Goods�� 151 Andrew Tettenborn 10. The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century�� 165 Louise Gullifer 11. Are there Any General Principles of Commercial Law?�� 181 Gerard McMeel 12. Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850�� 201 Charles Mitchell

viii Contents 13. ‘The Obscure, the Implied and the Illegal’: English and French Approaches to the Interpretation of Written Contracts, Implication of Terms and Contracts Affected by Illegality�� 215 Richard Aikens 14. The Interpretation of Written Contracts�� 243 James Edelman 15. The Boilerplate and the Bespoke: Should Differences in the Quality of Consent Influence the Construction and Application of Commercial Contracts?�� 259 David Foxton 16. Illegality: Pleading, Proof, and Presumptions�� 279 William Swadling 17. Negotiating Damages�� 297 Andrew Burrows 18. The Myth of Common Law Tracing�� 309 Sarah Worthington 19. Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies�� 329 Stephen Watterson 20. Silence and Solidarity? The Duties of Individual Directors Minded to Speak Out about their Board’s Decision-making and Governance�� 345 Peter Watts

LIST OF CONTRIBUTORS Sir Richard Aikens PC is a retired Lord Justice of Appeal, an Honorary Fellow of St John’s College, Cambridge and a Visiting Professor of Law at King’s College London and Queen Mary University of London. Yvonne Baatz is Professor in Commercial Law, Shipping and Transport at the Centre for Commercial Law Studies at Queen Mary University of London. Simon Baughen is Professor of Shipping Law at Swansea University. Howard Bennett is Hind Professor of Commercial Law at the University of Nottingham and an honorary member of 7 King’s Bench Walk, London. Michael Bridge QC (Hon) FBA is a Professor of Law at the National University of Singapore, an Emeritus Professor of Law at London School of Economics, a Senior Research Fellow at Harris Manchester College, Oxford and a door tenant at 20 Essex Street Chambers, London. Adrian Briggs QC (Hon) is Sir Richard Gozney Fellow and Tutor in Law at St Edmund Hall, Oxford and Professor of Private International Law at the University of Oxford. Andrew Burrows QC (Hon) DCL FBA is a Fellow of All Souls College, Oxford and Professor of the Law of England at the University of Oxford; in June 2020 he will become a Justice of the UK Supreme Court. Martin Davies is Admiralty Law Institute Professor of Maritime Law at Tulane University Law School. James Edelman is a Justice of the High Court of Australia and an Adjunct Professor at the Universities of New South Wales, Queensland and Western Australia. Sir David Foxton is a Justice of the High Court, assigned to the Commercial Court. Nicholas Gaskell is Emeritus Professor of Maritime and Commercial Law at the TC Beirne School of Law at the University of Queensland and a door tenant at Quadrant Chambers, London. Louise Gullifer QC (Hon) FBA is Rouse Ball Professor of English Law at the University of Cambridge, a Fellow of Gonville and Caius College, and an associate member of 3 Verulam Buildings, London. Gerard McMeel QC is Professor of Commercial Law at the University of Manchester, a practising barrister and a door tenant at Quadrant Chambers, London. Charles Mitchell QC (Hon) FBA is a Professor of Law at University College London. Francis Reynolds QC (Hon) DCL FBA is Professor of Law Emeritus at the University of Oxford and an Emeritus Fellow of Worcester College, Oxford.

x List of Contributors William Swadling is a Fellow and Tutor in Law at Brasenose College, Oxford and Reader in the Law of Property at the University of Oxford. Andrew Tettenborn is Professor of Commercial Law in the Institute of International Shipping and Trade Law at Swansea University. Stephen Watterson is Reader in Private Law at the University of Cambridge and a Fellow of Trinity Hall, Cambridge. Peter Watts QC FRSNZ is a Professor of Law at Auckland University and a Senior Research Fellow at Harris Manchester College, Oxford; he is also a practising barrister at Bankside Chambers, Auckland and a door tenant at Fountain Court Chambers, London. Sarah Worthington QC (Hon) FBA is Downing Professor of the Laws of England at the University of Cambridge, a Fellow of Trinity College, Cambridge and an academic member of South Square Chambers, London.

TABLE OF CASES 1406 Pub Co Ltd v Hoare [2001] 23 EG 154 (CS) (Ch)�� 272 20th Century Fox Film Corp v MV Ship Agencies Inc 992 F Supp 1429, 1998 AMC 2514 (MD Fla 1997)�� 121 32Red plc v WHG (International) Ltd [2013] EWHC 815 (Ch)�� 298 A Roberts & Co Ltd v Leicestershire CC [1961] Ch 555 (Ch)�� 252 A Schroeder Music Publishing Co Ltd v Macaulay (formerly Instone) [1974] 1 WLR 1308 (HL)��206, 260–61, 277 A v B (No 2) [2007] EWHC 54 (Comm), [2007] 1 Lloyd’s Rep 358�� 105 A v R [2018] UKPC 4�� 130, 135, 137 A/S D/S Svendborg v Akar [2003] EWHC 797 (Comm)�� 105 Abrahams v Herbert Reiach Ltd [1922] 1 KB 477 (CA)�� 155 AC Ward & Son Ltd v Catlin (Five) Ltd (No 2) [2009] EWHC 3122 (Comm), [2010] Lloyd’s Rep IR 695�� 73 Actionstrength Ltd (t/a Vital Resources) v International Glass Engineering In Gl En SpA [2003] UKHL 17, [2003] 2 AC 541�� 192 Admiralty Commissioners v Owners of the SS Volute (The Volute) [1922] 1 AC 129 (HL)�� 13 Aegeon, The. See Agapitos v Agnew (No 1)�� 77 African Export-Import Bank v Shebah Exploration and Production Co Ltd [2017] EWCA Civ 845, [2018] 1 WLR 487�� 261 Agapitos v Agnew (The Aegeon) (No 1) [2002] EWCA Civ 247, [2003] QB 556�� 77 AIB Group (UK) Plc v Mark Redler & Co Solicitors [2014] UKSC 58, [2015] AC 1503��310, 313 AIB Group (UK) plc v Martin [2001] UKHL 63, [2002] 1 WLR 94�� 269 AIG Europe (UK) Ltd v The Ethniki [2000] 2 All ER 566 (CA)�� 265 Air Tahiti Nui Pty Ltd v McKenzie [2009] NSWCA 429�� 31 Akai Pty Ltd v People’s Insurance Co Ltd [1998] 1 Lloyd’s Rep 90 (Com Ct)�� 103 Aktieselskabet de Danske Sukkerfabrikker v Bajamar Compania Naviera SA (The Torenia) [1983] 2 Lloyd’s Rep 210 (Com Ct)�� 33 Alchemy Estates Ltd v Astor [2008] EWHC 2675 (Ch), [2009] 1 WLR 940�� 267 Alexandra 1, The and Ever Smart, The. See Evergreen Marine (UK) Ltd v Nautical Challenge Ltd��20–21 Alexandros T, The. See Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG��99, 105 Alexion Hope, The. See Schiffshypothekenbank Zu Luebeck AG v Compton��267, 269 Alfred C Beatty v Guggenheim Exploration Co 225 NY 380 (1919)�� 189 Alfred McAlpine Capital Projects Ltd v Tilebox Ltd [2005] EWHC 281 (TCC), [2005] BLR 271�� 277 Ali v Khan [2002] EWCA Civ 974, [2009] WTLR 187�� 286 Aliakmon, The. See Leigh & Sillivan Ltd v Aliakmon Shipping Co Ltd�� 29

xii Table of Cases Alicia Hosiery Ltd v Brown Shipley & Co Ltd [1970] 1 QB 195 (QB)�� 159 All Russian Cooperative Society Ltd v Smith & Sons (1923) 14 Ll L Rep 351 (CA)��154, 162 Allan v Sundius (1862) 1 H & C 123, 158 ER 827�� 211 Al-Sabah v Grupo Torras SA [2005] UKPC 1, [2005] 2 AC 333��126, 137 Amalia, The. See Cail v Papayanni�� 2 Amec Developments Ltd v Jury’s Hotel Management (UK) Ltd (2001) 82 P & CR 22 (Ch)�� 298 Amiri Flight Authority v BAE Systems plc [2002] EWHC 2481 (Comm), [2003] 1 All ER (Comm) 1; [2003] EWCA Civ 1447, [2004] 1 All ER (Comm) 385��262, 264 AMT Futures Ltd v Marzillier [2017] UKSC 13, [2018] AC 439�� 105 Anderson v Pitcher (1800) 2 Bos & Pul 164, 126 ER 1216�� 205 Andres Bonifacio, The [1993] SGCA 70, [1993] 3 SLR(R) 71�� 114–15, 117 Aneco Reinsurance Underwriting Ltd v Johnson & Higgins Ltd [2001] UKHL 51, [2002] 1 Lloyd’s Rep 157�� 79 Angove Pty Ltd v Bailey (Re D&D Wines International Ltd) [2013] EWHC 215 (Ch); [2014] EWCA Civ 215, [2015] 1 All ER (Comm) 36; [2016] UKSC 47, [2016] 1 WLR 3179�� 171 Ant Jurgens Margarinefabrieken v Louis Dreyfus & Co [1914] 3 KB 40 (KB)��144, 150 Antaios Compania Naviera SA v Salen Rederierna AB [1985] AC 191 (HL)��219, 248 Antoni v Antoni [2007] UKPC 10, [2007] WTLR 1335�� 285 AP Moller-Maersk A/S (t/a Maersk Line) v Kyokuyo Ltd (The Maersk Tangier) [2017] EWHC 654 (Comm), [2017] 2 All ER (Comm) 922; [2018] EWCA Civ 778, [2018] 2 Lloyd’s Rep 59�� 46 App Bruxelles, 30.10.75 (1976) 11 ETL 238�� 51 Aqualon (UK) Ltd v Vallana Shipping Corp [1994] 1 Lloyd’s Rep 669 (Com Ct)�� 47 Arbuthnott v Fagan [1995] CLC 1396 (CA)�� 270 Aries, The. See Aries Tanker Corp v Total Transport Ltd�� 277 Aries Tanker Corp v Total Transport Ltd (The Aries) [1977] 1 WLR 185 (HL)�� 277 Armour v Thyssen Edelstahlwerke AG [1991] 2 AC 339 (HL)�� 170 Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2013] Ch 156�� 196–98, 317, 334–35, 337, 343–44 Arnold v Britton [2015] UKSC 36, [2015] AC 1619��216, 220–21, 225, 248 Aspden v Webbs Poultry & Meat Group (Holding) Ltd [1996] IRLR 521 (QB)�� 269 Aspen Underwriting Ltd v Credit Europe Bank NV [2018] EWCA Civ 2590, [2019] 1 Lloyd’s Rep 221��98, 102 Aspen Underwriting Ltd v Kairos Shipping Ltd (The Atlantik Confidence) [2017] EWHC 1904 (Comm), [2017] 2 Lloyd’s Rep 295; [2018] EWCA Civ 2590, [2019] 1 Lloyd’s Rep 221��99, 101 Astro Exito Navegacion SA v Southland Enterprise Co (The Messiniaki Tolmi) (No 2) [1983] 2 AC 787 (HL)��153, 155 Atkinson’s Will Trusts, Re [1978] 1 WLR 586 (Ch)�� 252 Atlantik Confidence, The. See Aspen Underwriting Ltd v Kairos Shipping Ltd��99, 101 Attorney-General for Hong Kong v Reid [1994] 1 AC 324 (PC)�� 322 Attorney-General of Belize v Belize Telecom Ltd [2009] UKPC 10, [2009] 1 WLR 1988�� 216, 227, 256 Attorney-General of Hong Kong v Nai-Keung [1987] 1 WLR 1339 (PC)�� 196 Attorney-General v Blake [2001] 1 AC 268 (HL)��301, 304

Table of Cases xiii Attorney-General v Smith & Cocks [1892] 2 QB 289 (QB)�� 67 Aughton Ltd (formerly Aughton Group Ltd) v MF Kent Services Ltd (1991) 31 Con LR 60 (CA)�� 265 Australian Broadcasting Corp v XIVth Commonwealth Games Ltd (1988) 18 NSWLR 540�� 253 Australian Competition and Consumer Commission v Valve Corp (No 3) [2016] FCA 196, (2016) 337 ALR 647�� 174 Australian Education Union v Department of Education and Children’s Services [2012] HCA 3, (2012) 248 CLR 1�� 247 Automatic Self-Cleansing Filter Syndicate Co Ltd v Cuninghame [1906] 2 Ch 34 (Ch and CA)�� 347 Automotive, Food, Metals, Engineering, Printing and Kindred Industries Union v ALS Industrial Australia Pty Ltd [2015] FCAFC 123, (2015) 235 FCR 305�� 253 AWH Fund Ltd v ZCM Asset Holding Co (Bermuda) Ltd (Bahamas CA, 16 Feb 2017)�� 132 AXA General Insurance Ltd v Gottlieb [2005] EWCA Civ 112, [2005] Lloyd’s Rep IR 369��76–77 AXA Sun Life Services plc v Campbell Martin Ltd [2011] EWCA Civ 133, [2011] 2 Lloyd’s Rep 1�� 191, 270, 272 Baker v Paine (1750) 1 Ves Sen 456, 27 ER 1140�� 202 Ballenita, The and BP Energy, The. See ERG Petroli SpA v Vitol SA�� 162 Bamford v Bamford [1970] Ch 212 (CA)�� 347 Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd (The Good Luck) [1992] 1 AC 233 (HL)�� 75 Banque Belge pour L’Etranger v Hambrouck [1921] 1 KB 321 (CA)�� 314–15, 318 Banque Keyser Ullmann SA v Skandia (UK) Insurance Co Ltd [1990] 1 QB 665 (CA)��72, 77 Barclays Bank plc v Grant Thornton UK LLP [2015] EWHC 320 (Comm), [2015] 1 CLC 180�� 264 Barclays Bank plc v HHY Luxembourg SARL [2010] EWCA Civ 1248, [2011] 1 BCLC 336�� 248 Barclays Bank plc v Unicredit Bank AG [2014] EWCA Civ 302, [2014] 2 All ER (Comm) 115�� 255 Barlow Clowes International Ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 All ER 333�� 357 Barlow Clowes International Ltd v Vaughan [1992] 4 All ER 22 (CA)��148, 325 Barnes v Jarvis [1953] 1 WLR 649 (Div Ct)�� 248 Barnes v Youngs [1898] 1 Ch 414 (Ch)�� 67 Barros Mattos Junior v MacDaniels Ltd [2005] EWHC 1323 (Ch), [2005] ILPr 45�� 309 Barton v Armstrong [1976] AC 104 (PC)�� 84 Base Metal Trading Ltd v Shamurin [2004] EWCA Civ 1316, [2005] 1 WLR 1157�� 359 Bates v Hewitt (1867) LR 2 QB 595 (QB)�� 66 Bayley v Wilkins (1849) 7 CB 886, 137 ER 351�� 212 Bayliffe v Butterworth (1847) 1 Ex 425, 154 ER 181�� 212 Bayoil SA v Seawind Tankers Corp (The Leonidas) [2001] 1 Lloyd’s Rep 533 (Com Ct)�� 267 Beacon Life & Fire Assurance Co v Gibb (1862) 1 Moo PC NS 73, 15 ER 630�� 69 Bean v Stupart (1778) 1 Doug KB 11, 99 ER 9�� 204 Bedouin, The [1894] P 1 (CA)�� 74 Behnke v Bede Shipping Co Ltd [1927] 1 KB 649 (KB)�� 113 Belfast & Ballymena Railway Co v Keys (1861) 9 HLC 556, 11 ER 846�� 67

xiv Table of Cases Bell v Hood 327 US 678, 66 S Ct 773 (1946)�� 119 Bentsen v Taylor, Sons & Co (No 2) [1893] 2 QB 274 (CA)�� 164 Beresford v Royal Insurance Co Ltd [1938] AC 586 (HL)�� 69 Berge Sisar, The. See Borealis AB v Stargas Ltd�� 162 Bernina (No 2), The. See Mills v Armstrong (No 2)��5–8 BGH 17.7.2008, I ZR 181/05, (2009) European Transport Law 196 (Germany)�� 48 Bhatia Shipping & Agencies Pvt Ltd v Alcobex Metals Ltd [2004] EWHC 2323 (Comm), [2005] 2 Lloyd’s Rep 336��46–47 Bigge v Parkinson (1862) 7 H & N 955, 158 ER 758�� 206 Bilta (UK) Ltd (in liq) v Nazir (No 2) [2015] UKSC 23, [2016] AC 1��216, 235 Bishopsgate Investment Management Ltd (in liq) v Homan [1995] Ch 211 (CA)�� 325 Bishopsgate Motor Finance Corp v Transport Brakes Ltd [1949] 1 KB 322 (CA)�� 141 Black King Shipping Corp v Massie (The Litsion Pride) [1985] 1 Lloyd’s Rep 437 (Com Ct)��73, 82 Blackburn Low & Co v Vigors (1886) 17 QBD 553 (CA), (1887) 12 App Cas 531 (HL)��67, 72 Blackett v Royal Exchange Assurance Co (1832) 2 Cr & J 244, 149 ER 106�� 209 Blue Sky One Ltd v Mahan Air [2010] EWHC 631 (Comm)�� 111–12, 126 Blue Whale Corp v Grand China Shipping Development 722 F 3d 488, 2014 AMC 145 (2d Cir 2013)��119–20 Bluett v Osbourne (1816) 1 Stark 384, 171 ER 504�� 206 Blunt v Cumyns (1751) 2 Ves Sen 331, 28 ER 213�� 202 Bols Distilleries BV (t/a Bols Royal Distilleries) v Superior Yacht Services Ltd [2006] UKPC 45, [2007] 1 WLR 12�� 98 Bolton v Lancashire & Yorkshire Rly Co (1865-66) LR 1 CP 431�� 159 Bonner v Cox [2005] EWCA Civ 1512, [2005] 2 Lloyd’s Rep 152�� 79 Borden (UK) Ltd v Scottish Timber Products Ltd [1981] Ch 25 (CA)�� 170 Borealis AB v Stargas Ltd (The Berge Sisar) [2001] UKHL 17, [2002] 2 AC 205�� 162 Boscawen v Bajwa [1996] 1 WLR 328 (CA)�� 319 Boulton v Houlder Bros & Co [1904] 1 KB 784 (CA)�� 68 Bovis Lend Lease Ltd v Cofely Engineering Services [2009] EWHC 1120 (TCC)�� 271 Bow Valley Husky (Bermuda) Ltd v St John Shipbuilding Ltd [1997] 3 SCR 1210�� 16 Bowden Bros & Co Ltd v Little [1907] HCA 14, (1907) 4 CLR 1364�� 162 Bowmakers Ltd v Barnet Instruments Ltd [1945] KB 65 (CA)��233, 279, 281–82 BP Exploration Co (Libya) Ltd v Hunt (No 2) [1979] 1 WLR 783 (QB)�� 333 BP Refinery (Westernport) Pty Ltd v Shire of Hastings [1977] UKPC 13, (1977) 180 CLR 266��227, 256 Bracewell v Appleby [1975] Ch 408 (Ch)�� 298 Bracken Partners Ltd v Gutteridge [2003] EWHC 1064 (Ch), [2003] 2 BCLC 84�� 309 Bridges & Salmon Ltd v Owner of The Swan (The Swan) [1968] 1 Lloyd’s Rep 5 (PDAD)�� 31 Brikom Investments Ltd v Carr [1979] QB 467 (CA)�� 191 Bristol & West Building Society v Mothew (t/a Stapley & Co) [1998] Ch 1 (CA)�� 355 British Telecommunications Plc v Ticehurst [1992] ICR 383 (CA)�� 354 Britton v Royal Insurance Co (1866) 4 F & F 905, 176 ER 843�� 68–69, 76 Brotherton v Aseguradora Colseguros SA (No 2) [2003] EWCA Civ 705, [2003] Lloyd’s Rep IR 746�� 79 Brough v Whitmore (1791) 4 TR 206, 100 ER 976�� 207 Brown v Byrne (1854) 3 El & Bl 703, 118 ER 1304��210, 212 Brown v Edgington (1841) 2 M & G 279, 133 ER 751�� 206

Table of Cases xv Brownlie v Campbell (1880) 5 App Cas 925 (HL)�� 67 Burnet v Kensington (1795) Peake Add Cas 71, 170 ER 198�� 209 Busk v Royal Exchange Assurance Co (1818) 2 B & Ald 73, 106 ER 294�� 203 BV Nederlandse Industrie van Eiprodukten v Rembrandt Enterprises Inc [2019] EWCA Civ 596, [2019] 1 Lloyd’s Rep 491�� 84 Byrnes v Kendle [2011] HCA 26, (2011) 243 CLR 253�� 249 Bywell Castle, The. See London Steamboat Co v Owners of the Bywell Castle�� 15 Cadogan Estates Ltd v McMahon [2001] 1 AC 378 (HL)�� 248 Cail v Papayanni (The Amalia) (1863) 1 Moo PC NS 471, 15 ER 778�� 2 Caine v Horsefall (1847) 2 Car & K 349, 175 ER 144�� 203 Calverley v Green [1984] HCA 81, (1984) 155 CLR 242�� 287 Cambridge Gas Transportation Corp v Official Committee of Unsecured Creditors of Navigator Holdings plc [2006] UKPC 26, [2007] 1 AC 508�� 127 Camden v Cowley (1762) 1 W Bl 417, 96 ER 237�� 203 Canada (Attorney-General) v Fairmont Hotels Inc [2016] SCC 56, [2016] 2 SCR 720�� 252 Canadian Pacific Railway Co v Owners of the Steamship ‘Storstad’ (1917) 34 DLR 1 (Exch); (1918) 56 SCR 324; [1920] AC 397 (PC)��5, 7, 9–10 Canadian Pacific Railway v ‘Storstad’ (The) (1915) 40 DLR 600 (Exch)�� 8 Canterbury-Hurlstone Park RSL Club Ltd v Roberts [2008] NSWSC 845�� 359 Cantor v Cox (1975) 239 EG 121 (Ch)�� 292 Cape Comorin, The. See Carrington Slipways Pty Ltd v Patrick Operations Pty Ltd�� 27, 43–44 Cape Lambert Resources Ltd v MCC Australia Sanjin Mining Pty Ltd [2013] WASCA 66, (2013) 298 ALR 666�� 246 Capital & Counties Bank Ltd v Warriner-Bretherton Ford & Co (1896) 21 Com Cas 314, 12 TLR 216 (QB)��145, 150 Carbotrade SpA v Bureau Veritas 99 F 3d 86, 1997 AMC 98 (2d Cir 1996)�� 120 Cargill International SA v Bangladesh Sugar and Food Industries Corp [1998] 1 WLR 461 (CA)�� 248 Carrington Slipways Pty Ltd v Patrick Operations Pty Ltd (The Cape Comorin) (1991) 24 NSWLR 745�� 27, 43–44 Carr-Saunders v Dick McNeil Associates Ltd [1986] 2 All ER 888 (Ch)�� 298 Carter v Boehm (1766) 3 Burr 1905, 97 ER 1162�� 64–68, 73, 77 Case C-150/80 Elefanten Schuh v Jacqmain [1981] ECR 1671�� 97 Case C-201/82 Gerling Konzern Speziale Kreditversicherungs AG v Amministrazione del Tesoro dello Stato [1983] ECR 2503�� 99 Case C-412/98 Universal General Insurance Co v Group Josi Reinsurance Co SA [2000] ECR I-5925, [2001] QB 68�� 97 Case C-116/02 Erich Gasser GmbH v MISAT SRL [2003] ECR I-14693, [2005] QB 1�� 99 Case C-159/02 Turner v Grovit [2004] ECR I-3565, [2005] 1 AC 101�� 104 Case C-281/02 Owusu v Jackson [2005] QB 801�� 97 Case C-112/03 Société Financière et Industrielle du Peloux v Axa Belgium [2005] ECR I-3707, [2006] QB 251�� 101 Case C-77/04 Groupement d’Interet Economique (GIE) Reunion Europeenne v Zurich Espana [2005] ECR I-4509, [2006] 1 All ER (Comm) 488�� 101 Case C-386/05 Color Drack GmbH v Lexx International Vertriebs GmbH [2007] ECR I-3699, [2010] Bus LR 1044�� 97 Case C-463/06 FBTO Schadeverzekeringen NV v Odenbreit [2007] ECR I-11321, [2008] 2 All ER (Comm) 733��92, 101

xvi Table of Cases Case C-347/08 Vorarlberger Gebietskrankenkasse v WGV-Schwabische Allgemeine Versicherungs AG [2009] ECR I-8661, [2010] Lloyd’s Rep IR 77�� 101 Case C-111/09 Ceská Podnikatelská Pojiìtovna as, Vienna Insurance Group v Bilas [2010] Lloyd’s Rep IR 734�� 97 Case C-412/10 Homawoo v GMF Assurance SA & Ors [2011] ECR I-11603�� 100 Case C-543/10 Refcomp SpA v Axa Corporate Solutions Assurance SA [2013] 1 All ER (Comm) 1201�� 97 Case C-1/13 Cartier Parfums-Lunettes SAS v Ziegler France SA [2014] ILPr 25�� 97 Case C-302/13 flyLAL-Lithuanian Airlines [2014] 5 CMLR 27�� 97 Case C-352/13 Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Akzo Nobel NV [2015] QB 906�� 99 Case C-322/14 El Majdoub v CarsOnTheWeb.Deutschland GmbH [2015] 1 WLR 3986�� 98 Case C-264/14 Skatteverket v Hedqvist [2016] STC 372�� 175 Case C-175/15 Taser International Inc v SC Gate 4 Business SRL [2016] QB 887�� 97 Case C-222/15 Höszig Kft v Alstom Power Thermal Services [2016] ILPr 36�� 98 Case C-340/16 Landeskrankenanstalten-Betriebsgesellschaft – KABEG v Mutuelles du Mans Assurances – MMA IARD SA [2018] Lloyd’s Rep IR 556�� 101 Case C-368/16 Assens Havn v Navigators Management (UK) Ltd (The Sea Endeavour I) [2018] QB 463�� 100–02, 106 Case C-106/17 Hofsoe v LVM Landwirtschaftlicher Versicherungsverein Munster AG [2018] Lloyd’s Rep IR 608�� 101 Cass, 2 fév 1808�� 224 Cass Civ 15 avr 1872, Veuve Foucauld�� 223 Cass Civ 6 mars 1876, D 76.1193�� 230 Cass Civ 1ère, 18 fév 1986, n° 84-12.347�� 224 Cass Civ 1ère, 13 déc 1986, n° 19.068�� 224 Cass Civ 1ère, 1 oct 1996, n° 94-18.876�� 236 Cass Civ 1ère, 7 oct 1998, n° 96-14.359�� 237 Cass Civ 1ère, 15 déc 2011, n° 10-23.528�� 231 Cass Civ 1ère, 28 octobre 2015, n° 14.11498�� 224 Cass Civ 1er, 26 septembre 2018, n° 16-25.184�� 238 Cass Civ 1er, 6 fév 2019, n° 17-20463�� 239 Cass Civ 3ème, 6 fév 2002 n° 00-12.675�� 223 Cass Civ 3ème, 20 av 2017, n° 16-13.462�� 223 Cass Com, 22 mai 2002, n° 99-11.052�� 223 Cass Com, 26 juin 2019, n° 18-13689�� 237 Cattlin v Hills (1849) 8 CB 123, 137 ER 455�� 7 Cavendish Square Holdings BV v Makdessi [2015] UKSC 67, [2016] AC 1172��156, 277 CDV Software Entertainment AG v Gamecock Media Europe Ltd [2009] EWHC 2965 (Ch)�� 277 Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1 AC 1101��216, 221, 250, 252–54 Charter Reinsurance Co Ltd v Fagan [1997] AC 313 (HL)��216, 248 Chatenay v Brazilian Submarine Telegraph Co Ltd [1891] 1 QB 79 (CA)�� 243 Chaurand v Angerstein (1791) Peake 61, 170 ER 79�� 206 Chettiar v Chettiar [1962] AC 294 (PC)��279–80, 283–88, 291–95 Cheverney Consulting Ltd v Whithead Mann Ltd [2007] EWHC 3130 (Ch)�� 273

Table of Cases xvii Childers v Childers (1857) 3 K & J 310, 69 ER 1126�� 292 China Traders’ Insurance Co Ltd v Royal Exchange Assurance Corp [1898] 2 QB 187 (CA)�� 68 China-Pacific SA v Food Corp of India (The Winson) [1982] AC 939 (HL)�� 159 CIC Insurance Ltd v Bankstown Football Club Ltd [1997] HCA 2, (1997) 187 CLR 384�� 247 City of Berlin, The [1908] P 110 (CA)�� 15 Clark’s Patent, Re (1870) 7 Moo PC NS 255, 17 ER 97�� 67 Clarke v Shee & Johnson (1774) 1 Cowp 197, 98 ER 1041�� 340 Clements v Hall (1858) 2 De G & J 173, 44 ER 954�� 67 Codelfa Construction Pty Ltd v State Rail Authority of NSW [1982] HCA 24, (1982) 149 CLR 337��247, 256 Coggs v Barnard (1703) 2 Ld Raym 909, 92 ER 107�� 184 Cohen & Co v Ockerby & Co Ltd [1917] HCA 58, (1917) 24 CLR 288�� 153 Cole v Rawlinson (1702) 1 Salk 234, 91 ER 207�� 244 Collector of Customs v Agfa-Gevaert Ltd [1996] HCA 36, (1996) 186 CLR 389�� 248 Collier v Collier [2002] EWCA Civ 1095, [2002] BPIR 1057�� 286 Collier v Sunday Referee Publishing Co Ltd [1940] 2 KB 647 (KB)�� 153 Colonial Bank v Whinney (1885) 30 Ch D 261 (CA); (1886) 11 App Cas 426 (HL)��196–97 Colour Quest Ltd v Total Downstream UK plc [2009] EWHC 540 (Comm), [2009] 2 Lloyd’s Rep 1�� 23 Colpoys v Colpoys (1822) Jac 451, 37 ER 921�� 245 Comandate, The. See Comandate Marine Corp v Pan Australia Shipping Pty Ltd�� 10 Comandate Marine Corp v Pan Australia Shipping Pty Ltd (The Comandate) [2006] FCAFC 192, [2008] 1 Lloyd’s Rep 119�� 10 Commercial Management (Investments) Ltd v Mitchell Design & Construct Ltd [2016] EWHC 76 (TCC), 164 Con LR 139�� 261 Commercial Union Assurance Co v Niger Co (1921) 6 Ll L Rep 235 (CA); (1922) 13 Ll L Rep 75 (HL)�� 79 Commerzbank Aktiengesellschaft v Liquimar Tankers Management Inc [2017] EWHC 161 (Comm), [2017] 1 WLR 3497�� 98 Commissioner for Prices and Consumer Affairs (SA) v Charles Moore (Aust) Ltd [1977] HCA 38, (1977) 139 CLR 449�� 247 Commonwealth Bank of Australia v Barker [2014] HCA 32, (2014) 253 CLR 169�� 256 Compagnie Générale Transatlantique v Steamship ‘IMO’ (1919) 47 DLR 462 (Exch); (1919) 59 SCR 644; sub nom Steamship Imo v Compagnie Générale Transatlantique (1920) 2 Ll L Rep 188 (PC hearing), 246 (PC submissions) and 536 (PC judgment)�� 14–15, 19 Compania Sud Americana De Vapores SA v Hin-Pro International Logistics Ltd [2014] EWHC 3632 (Comm), [2015] 1 Lloyd’s Rep 301; [2015] EWCA Civ 401, [2015] 2 Lloyd’s Rep 1�� 90 Computer 2000 Distribution Ltd v ICM Computer Solutions plc [2004] EWCA Civ 1634, [2005] Info TLR 147��157, 160 Computer Associates UK Ltd v Software Incubator Ltd [2016] EWHC 1587 (QB), [2017] Bus LR 245; [2018] EWCA Civ 518, [2019] Bus LR 522�� 173–74, 197 Conductive Inkjet Technology Ltd v Uni-Pixel Displays Inc [2013] EWHC 2968 (Ch), [2014] 1 All ER (Comm) 654�� 270 Consorcio MGT v Owner and/or Demise Charterer of the Vessel ‘Min Rui’ (The Min Rui) [2016] SGHC 183�� 117

xviii Table of Cases Container Transport International Inc v Oceanus Mutual Underwriting Association (Bermuda) Ltd (No 1) [1982] 2 Lloyd’s Rep 178 (Com Ct); [1984] 1 Lloyd’s Rep 476 (CA)��71–74 Cook v Fountain (1676) 3 Swans 585, 36 ER 984�� 286 Cooke v Gill (1873) LR 8 CP 107 (CP)�� 283 Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation [1981] HCA 26, (1981) 147 CLR 297�� 248 Cottington v Fletcher (1740) 2 Atk 155, 26 ER 498�� 284, 288–91 Countess of Rutland’s Case (1604) 5 Co Rep 25b, 77 ER 89�� 244 Cox v Bankside Members Agency Ltd [1995] CLC 671 (CA)�� 80 Crédit Agricole Indosuez v BB Energy BV [2004] EWHC 750 (Comm)�� 276 Crédit Suisse Fides Trust SA v Cuoghi [1998] QB 818 (CA)�� 127 Crema v Cenkos Securities plc [2010] EWCA Civ 1444, [2011] 1 WLR 2066��217–18 Crestsign Ltd v National Westminster Bank plc [2014] EWHC 3043 (Ch), [2015] 2 All ER (Comm) 133�� 193 Crichton v Crichton (1895) 13 R 770�� 292 Cropper v Cook (1868) LR 3 CP 194�� 212 Cross v Eglin (1831) 2 B & Ad 106, 109 ER 1083�� 209 Curtis v Perry (1802) 6 Ves Jun 739, 31 ER 1285�� 292 Daewoo Australia Pty Ltd v Suncorp-Metway Ltd [2000] NSWSC 35, (2000) 155 FLR 54�� 175 Dalglish v Jarvie (1850) 2 Mac & G 231, 42 ER 89��66–67 Datec Electronic Holdings Ltd v United Parcels Service Ltd [2007] UKHL 23, [2007] 1 WLR 1325�� 48 Davies v Otty (No 2) (1865) 35 Beav 208, 55 ER 875�� 284, 287–88, 290–91 Dawes v Peck (1799) 8 TR 330, 101 ER 1417�� 160 Dawkes v De Lorane (1771) 3 Wils KB 207, 95 ER 1015�� 203 Dawson v Bell [2016] EWCA Civ 96, [2016] 2 BCLC 59�� 264 DC Merwestone, The. See Versloot Dredging BV v HDI Gerling Industrie Versicherung AG�� 68 De Costa v Scandret (1723) 2 P Wms 170, 24 ER 686�� 64 De Hahn v Hartley (1786) 1 TR 343, 99 ER 1130�� 204 Dean v Macdowell (1878) 8 Ch D 345 (CA)�� 67 Deepak Fertilisers & Petrochemical Corp Ltd v ICI Chemicals & Polymers Ltd [1998] 2 Lloyd’s Rep 139 (Com Ct); (1999) 1 Lloyd’s Rep 387 (CA)�� 191 Deering v Farrington (1673) 3 Keble 303, 84 ER 734�� 205 Dehers v Harriot (1691) 1 Show KB 163, 89 ER 513�� 203 Demby Hamilton & Co Ltd v Barden [1949] 1 All ER 435 (KB)��154, 157 Dennis (A Bankrupt), Re [1996] Ch 80 (CA)�� 315 Deutsche Bank AG London Branch v Petromena ASA [2015] EWCA Civ 226, [2015] 1 WLR 4225�� 97 Deutsche Bank AG v Comune di Savona [2017] EWHC 1013 (Comm), [2018] 1 BCLC 358�� 98 Deutsche Genossenschaftsbank v Burnhope [1995] 1 WLR 1580 (HL)�� 253 Dexia Crediop SpA v Provincia Brescia [2016] EWHC 3261 (Comm)�� 99 Diamond Alkali Export Corp v Bourgeois [1921] 3 KB 443 (KB)�� 43 Diana Properity, The. See Reardon Smith Line Ltd v Hansen-Tangen��188, 219 Diplock, Re [1948] Ch 465 (CA); sub nom Ministry of Health v Simpson [1951] AC 251 (HL)��311, 319

Table of Cases xix DL Electrical Supplies (Mitcham) Ltd v GL Group Ltd 1987 SLT (Sh Ct) 36�� 160 Do-Buy 925 Ltd v National Westminster Bank plc [2010] EWHC 2862 (QB)��263–64 Doe d Gord v Needs (1836) 2 M & W 129, 150 ER 698�� 245 Dominator, The. See Louis Dreyfus & Cie v Parnaso Cia Naviera SA�� 271 Donoghue v Stevenson [1932] AC 562 (HL)��132, 143 Donohue v Armco Inc [2001] UKHL 64, [2002] 1 Lloyd’s Rep 425�� 105 Dornoch Ltd v Westminster International BV (The WD Fairway) [2009] EWHC 889 (Admlty), [2009] 2 Lloyd’s Rep 191�� 109, 111–12 Drake Insurance plc v Provident Insurance plc [2003] EWCA Civ 1834, [2004] QB 601��79, 87 Drumlanrig, The. See Owners of Cargo of the Tongariro v Owners of the Drumlanrig�� 6 Dubai Islamic Bank PJSC v PSI Energy Holding Co BSC [2011] EWHC 2718 (Comm)�� 274 Duchess of Argyll v Duke of Argyll [1967] Ch 302 (Ch)�� 358 Dunlop v Lambert (1839) 6 Cl & F 600, 7 ER 824��46, 160 Dyke v Elliott (The Gauntlet) (1871-73) LR 4 PC 184�� 248 E Clemens Horst Co v Biddell Bros [1911] 1 KB 934 (CA); [1912] AC 18 (HL)�� 161 Eagle Star Insurance Co Ltd v Cresswell [2004] EWCA Civ 602, [2004] 1 CLC 926�� 80 Eagle, The. See Hollingworth v Southern Ferries Ltd�� 23 Early v Garrett (1829) 9 B & C 928, 109 ER 345�� 206 Eastern Distributors Ltd v Goldring [1957] 2 QB 600 (CA)�� 142 Eaton Mansions (Westminster) Ltd v Stinger Compania de Inversion SA [2013] EWCA Civ 1308, [2014] HLR 4�� 298 Éclairs Group Ltd v JKX Oil & Gas Plc [2015] UKSC 71, [2015] Bus LR 1395�� 363 Ecosse Property Holdings Pty Ltd v Gee Dee Nominees Pty Ltd [2017] HCA 12, (2017) 261 CLR 544�� 248 Edgington v Fitzmaurice (1885) 29 Ch D 459 (CA)�� 84 Eichholz v Bannister (1864) 17 CB (NS) 708, 144 ER 284�� 206 El Ajou v Dollar Land Holdings plc (No 1) [1994] 2 All ER 685 (CA)�� 72 Elder Dempster Lines v Zaki Ishag (The Lycaon) [1983] 2 Lloyd’s Rep 548 (Com Ct)�� 43 Elder, Dempster & Co Ltd v Paterson, Zochonis & Co Ltd [1924] AC 522 (HL)��35, 37 Electricity Generation Corp v Woodside Energy Ltd [2014] HCA 7, (2014) 251 CLR 640�� 248 Emery’s Investments Trusts, Re. Emery v Emery [1959] Ch 410 (Ch)�� 292 Empresa Exportadora De Azucar v Industria Azucarera Nacional SA (The Playa Larga and Marble Islands) [1983] 2 Lloyd’s Rep 171 (CA)�� 161 ENE Kos 1 Ltd v Petroleo Brasileiro SA (The Kos) (No 2) [2012] UKSC 17, [2012] 2 AC 164�� 159 Energy Venture Partners Ltd v Malabu Oil and Gas Ltd [2013] EWHC 2118 (Comm)��190, 274 Environment Agency v Churngold Recycling Ltd [2014] EWCA Civ 909, [2015] Env LR 13�� 197 Equatorial Marine Fuel Management Services Pte Ltd v The Bunga Melati 5 [2012] SGCA 46, [2012] 4 SLR 546�� 116 Equitable Life Assurance Society v Hyman [2002] 1 AC 408 (HL)�� 227–28, 230 Equitas Insurance Ltd v Municipal Mutual Insurance Ltd [2019] EWCA Civ 718, [2019] Lloyd’s Rep IR 359��80, 83 Equuscorp Pty Ltd v Glengallan Investments Pty Ltd [2004] HCA 55, (2004) 218 CLR 471��249, 253 ERG Petroli SpA v Vitol SA (The Ballenita and The BP Energy) [1992] 2 Lloyd’s Rep 455 (Com Ct)�� 162

xx Table of Cases Erlanger v New Sombrero Phosphate Co (1878) 3 App Cas 1218 (HL)�� 67 Ermogenous v Greek Orthodox Community of SA Inc [2002] HCA 8, (2002) 209 CLR 95�� 253 Esso Petroleum Co Ltd v Addison [2003] EWHC 1730 (Comm)�� 267 Eurymedon, The. See New Zealand Shipping Co Ltd v AM Satterthwaite & Co Ltd�� 33 Evergreen Marine (UK) Ltd v Nautical Challenge Ltd (The Alexandra 1 & The Ever Smart) [2017] EWHC 453 (Admlty), [2017] 1 Lloyd’s Rep 666; [2018] EWCA Civ 2173, [2019] 1 Lloyd’s Rep 130�� 20 Experience Hendrix LLC v PPX Enterprises Inc [2003] EWCA Civ 323, [2003] 1 All ER (Comm) 830�� 298, 301, 305 Fanti, The. See Firma C-Trade SA v Newcastle Protection & Indemnity Assoc; Socony Mobil Oil Inc v West of England Shipowners Mutual Insurance Assoc (London) Ltd (No 2)�� 92 Fargnoli v GA Bonus plc [1997] CLC 653 (OH)�� 77 Farquharson Bros & Co v King & Co [1902] AC 325 (HL)�� 142 Fawcus v Sarsfield (1856) 6 El & Bl 192, 119 ER 836�� 67 Federal Bulker, The. See Federal Bulk Carriers Inc v C Itoh & Co Ltd�� 265 Federal Bulk Carriers Inc v C Itoh & Co Ltd (The Federal Bulker) [1989] 1 Lloyd’s Rep 103 (CA)�� 265 Federal Commissioner of Taxation v Whitfords Beach Pty Ltd [1982] HCA 8, (1982) 150 CLR 355�� 247 Federal Republic of Brazil v Durant International Corp [2015] UKPC 35, [2016] AC 297�� 325 Fern Computer Consultancy Ltd v Intergraph Cadworx & Analysis Solutions Inc [2014] EWHC 2908 (Ch), [2014] Bus LR 1397��173–74 FG Wilson (Engineering) Ltd v John Holt & Co (Liverpool) Ltd [2013] EWCA Civ 1232, [2014] 1 WLR 2365��153, 171 FHR European Ventures LLP v Cedar Capital Partners LLC [2014] UKSC 45, [2015] AC 250�� 309, 322, 324 Fiona Trust & Holding Corp v Privalov [2007] UKHL 40, [2007] Bus LR 1719�� 248 Firma C-Trade SA v Newcastle Protection & Indemnity Assoc (The Fanti); Socony Mobil Oil Inc v West of England Shipowners Mutual Insurance Assoc (London) Ltd (No 2) (The Padre Island) [1991] 2 AC 1 (HL)�� 92 First Energy (UK) Ltd v Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194 (CA)�� 186 First Tower Trustees Ltd v CDS (Superstores International) Ltd [2017] EWHC 891 (Ch), [2017] 4 WLR 73; [2018] EWCA Civ 1396, [2019] 1 WLR 637; [2019] 1 WLR 1399��133, 135, 191–92 Fitzgerald v Masters [1956] HCA 53, (1956) 95 CLR 420�� 250 Foakes v Beer (1884) 9 App Cas 605 (HL)�� 185 Food Co UK LLP (t/a Muffin Break) v Henry Boot Developments Ltd [2010] EWHC 358 (Ch)�� 278 Force India Formula One Team Ltd v 1 Malaysia Racing Team Sdn Bhd [2013] EWCA Civ 780, [2013] RPC 36�� 298 Ford v Beech (1848) 11 QB 852, 116 ER 693�� 219 Forstater v Dweck [2007] EWHC 1991 (QB)�� 273 Forsyth International (UK) Ltd v Silver Shipping Co Ltd (The Saetta) [1994] 1 WLR 1334 (QB)�� 148 Foskett v McKeown [2001] 1 AC 102 (HL)�� 309–10, 312–13, 316–17, 319–20, 322–25

Table of Cases xxi Four Point Garage Ltd v Carter [1985] 3 All ER 12 (QB)�� 157 Fox v Mackreth (1788) 2 Bro CC 400, 29 ER 224�� 66 Francis v Lyon [1907] HCA 12, (1907) 4 CLR 1023�� 163 Franklins Pty Ltd v Metcash Trading Ltd [2009] NSWCA 407, (2009) 76 NSWLR 603�� 246 Fraser Shipping Ltd v Colton [1997] 1 Lloyd’s Rep 586 (Com Ct)�� 73 Fraser v Canterbury Diocesan Board of Finance (No 2) [2003] EWHC 1075 (Ch), [2003] WTLR 1125�� 271 Fraser v Murdoch (1880) 7 R 694 (IH)�� 67 Frederick E Rose (London) Ltd v William H Pim Jnr & Co Ltd [1953] 2 QB 450 (CA)�� 254 French and Co Ltd v Leeston Shipping Co Ltd [1922] 1 AC 451 (HL)�� 226 FS Sandeman & Sons v Tyzack & Branfoot Steamship Co Ltd [1913] AC 680 (HL)�� 324 FSHC Group Holdings Ltd v GLAS Trust Corp Ltd [2019] EWCA Civ 1363��254–55 G Percy Trentham Ltd v Archital Luxfer Ltd [1993] 1 Lloyd’s Rep 25 (CA)�� 195 Gabay v Lloyd (1825) 3 B & C 793, 107 ER 927�� 211 Galbraith & Grant Ltd v Block [1922] 2 KB 155 (KB)��157, 160 Gamer’s Motor Centre (Newcastle) Pty Ltd v Natwest Wholesale Australia Pty Ltd [1987] HCA 30, (1987) 163 CLR 236�� 156 Gammasonics Institute for Medical Research Pty Ltd v Comrad Medical Systems Pty Ltd [2010] NSWSC 267, (2010) 77 NSWLR 479�� 174 Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (Nos 2 & 3) [2001] EWCA Civ 1047, [2001] Lloyd’s Rep IR 667�� 80 Gandy v Adelaide Marine Insurance Co (1871) LR 6 QB 746 (QB)�� 67 Gardiner v Gray (1815) 4 Camp 144, 171 ER 46�� 206 Gascoigne v Gascoigne [1918] 1 KB 223 (KB)��279–80, 283–95 Gauntlet, The. See Dyke v Elliott�� 248 General Tire & Rubber Co Ltd v Firestone Tyre & Rubber Co Ltd (No 2) [1975] 1 WLR 819 (HL)�� 298 George Cohen, Sons & Co Ltd v Docks and Inland Waterways Executive (1950) 84 Ll L Rep 97 (CA)�� 252 Gerhardy v Brown [1985] HCA 11, (1985) 159 CLR 70�� 247 Geys v Société Générale, London Branch [2012] UKSC 63, [2013] 1 AC 523�� 216 Gibbon v Young (1818) 8 Taunt 254, 129 ER 381�� 208 Gibson v Small (1853) 4 HLC 353, 10 ER 499�� 210 Giedo van der Garde BV v Force India Formula One Team Ltd (formerly Spyker F1 Team Ltd (England)) [2010] EWHC 2373 (QB)�� 298 Gill & Duffus SA v Berger & Co Inc [1984] AC 382 (HL)�� 162 Gill & Duffus SA v Rionda Futures Ltd [1994] 2 Lloyd’s Rep 67 (Com Ct)�� 157 Gissing v Gissing [1971] AC 886 (HL)�� 253 Glasgow Assurance Corp v Symondson & Co (1911) 16 Com Cas 109 (KB)�� 79 Glencore International AG v Metro Trading International Inc (No 2) [2001] 1 Lloyd’s Rep 284 (Com Ct)�� 108 Global 5000 Ltd v Wadhawan [2012] EWCA Civ 13, [2012] 1 Lloyd’s Rep 239�� 104 Globe Motors Inc v TRW LucasVarity Electric Steering Ltd [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601��190, 274 Glynn v Margetson & Co [1893] AC 351 (HL)�� 267 Goldcorp Exchange Ltd, Re [1995] 1 AC 74 (PC)�� 140 Golden Key Ltd, Re [2009] EWCA Civ 636�� 248 Goldman Sachs International v Novo Banco SA [2018] UKSC 34, [2018] 1 WLR 3683�� 98

xxii Table of Cases Goldschmidt v Marryat (1809) 1 Camp 558, 170 ER 1057��67, 69 Gondal v Dillon Newsagents Ltd [1998] NPC 127 (CA)�� 301 Good Luck, The. See Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd�� 75 Goodlife Foods Ltd v Halt Fire Production Ltd [2018] EWCA Civ 1371, [2018] BLR 491�� 262 Gordon v Street [1899] 2 QB 641 (CA)�� 74 Goshawk Dedicated Ltd v Tyser & Co Ltd [2006] EWCA Civ 54, [2006] 1 Lloyd’s Rep 566��80, 82 Graham Joint Stock Shipping Co Ltd v Motor Union Insurance Co Ltd [1922] 1 KB 563 (CA)�� 77 Granville Oil & Chemicals Ltd v Davis Turner & Co Ltd [2003] EWCA Civ 570, [2003] 2 Lloyd’s Rep 356�� 57 Gray v Cox (1825) 4 B & C 108, 107 ER 999�� 206 Gray v Thames Trains Ltd [2009] UKHL 33, [2009] 1 AC 1339��232, 234 Great Berlin Steamboat Co, Re (1884) 26 Ch D 616 (CA)�� 292 Great Elephant Corp v Trafigura Beheer BV [2012] EWHC 1745 (Comm), [2013] 1 All ER (Comm) 415; [2013] EWCA Civ 905, [2013] 2 All ER (Comm) 992�� 156 Great Western Rly and Midland Rly Co v Bristol Corp (1918) 87 LJ Ch 414 (HL)�� 246 Greaves v Ashlin (1813) 3 Camp 426, 170 ER 1433��153, 204 Greenock Steamship Co v Maritime Insurance Co Ltd [1903] 1 KB 367 (KB)�� 75 Groom v Croker [1939] 1 KB 194 (CA)�� 80 Group Josi Re Co SA v Walbrook Insurance Co Ltd [1996] 1 WLR 1152 (CA)�� 71 Groves v Groves (1829) 3 Y & J 163, 148 ER 1136�� 292 Gujra v Roath [2018] EWHC 854 (QB), [2018] 1 WLR 3208�� 239 Gunsbourg, Re [1920] 2 KB 426 (CA)�� 315 Habaş Sinai Ve Tibbi Gazlar Isthisal Endüstri AS v Sometal SAL [2010] EWHC 29 (Comm), [2010] Bus LR 880�� 265 Haigh v Royal Mail Steam Packet Co Ltd (1883) 52 LJQB 640 (CA)�� 5 Hallett’s Estate, Re (1880) 13 Ch D 696 (CA)�� 325 Hansson v Hamel & Horley Ltd [1922] 2 AC 36 (HL)��44, 161 Harding v Bussell [1905] 2 KB 83 (CA)�� 68 Hari Bhum, The. See Through Transport Mutula Insurance Assoc (Eurasia) Ltd v New India Assurance Co Ltd�� 93 Harkness v Commonwealth Bank of Australia Ltd (1993) 32 NSWLR 543�� 362 Harlow & Jones Ltd v Panex (International) Ltd [1967] 2 Lloyd’s Rep 509 (Com Ct)�� 154 Harrison v Shepherd Homes Ltd [2011] EWHC 1811 (TCC), (2011) 27 Const LJ 709�� 262 Hartford Fire Insurance Co v Orient Overseas Container Line (The OOCL Bravery) [2000] AMC 1305, [2000] 1 Lloyd’s Rep 394 (SDNY)�� 58 Haywood v Rodgers (1804) 4 East 590, 102 ER 957�� 74 Hazell v Hammersmith & Fulham LBC [1992] 2 AC 1 (HL)�� 188 Heaven v Pender (t/a West India Graving Dock Co) (1883) 11 QBD 503 (CA)�� 143 Heilbrunn v Lightwood plc [2007] FCA 1518, (2007) 164 FCR 1�� 118 Helby v Matthews [1895] AC 471 (HL)�� 169 Hellenic Lines Ltd v Rhoditis 398 US 306, 90 S Ct 1731 (1970)�� 120 Hercules Insurance Co v Hunter (1836) 14 S 1137 (IH)�� 67 HIH Casualty & General Insurance Ltd v Axa Corporate Solutions [2002] EWCA Civ 1253, [2003] Lloyd’s Rep IR 1�� 75

Table of Cases xxiii HIH Casualty & General Insurance Ltd v New Hampshire Insurance Co [2001] EWCA Civ 735, [2001] CLC 1480�� 263 Hindley & Co Ltd v East Indian Produce Co Ltd [1973] 2 Lloyd’s Rep 515 (Com Ct)�� 161 Hirsche v Sims [1894] AC 654 (PC)�� 355 Hoare v The Queen [1989] HCA 33, (1989) 167 CLR 348�� 247 Hodgson v Marks [1971] Ch 892 (CA)�� 286 Hodgson v Richardson (1764) 1 Wm Bl 463, 96 ER 268�� 65 Hoge Raad 1.6.2012, SS 2012, No 95 (Godafoss)�� 48 Hollingworth v Southern Ferries Ltd (The Eagle) [1977] 2 Lloyd’s Rep 70 (QB)�� 23 Holman v Johnson (1775) 1 Cowp 341, 98 ER 1120�� 232, 238, 280–81, 283–84 Homburg Houtimport BV v Agrosin Private Ltd (The Starsin) [2000] 1 Lloyd’s Rep 85 (Com Ct); [2001] 1 Lloyd’s Rep 437 (CA); [2003] UKHL 12, [2004] AC 715�� 27–30, 32–36, 58–59, 268–69, 271 Hong Kong Fir, The. See Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd�� 215 Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd (The Hong Kong Fir) [1962] 2 QB 26 (CA)�� 215 Hood v Anchor Line (Henderson Brothers) Ltd [1918] AC 837 (HL)�� 5 Horn Linie GmbH & Co v Panamericana Formas E Impresos SA (The Hornbay) [2006] EWHC 373 (Comm), [2006] 2 Lloyd’s Rep 44�� 105 Hornbay, The. See Horn Linie GmbH & Co v Panamericana Formas E Impresos SA�� 105 Houghland v RR Low (Luxury Coaches) Ltd [1962] 1 QB 694 (CA)�� 184 Hounga v Allen [2014] UKSC 47, [2014] 1 WLR 2889�� 235 Howard Smith Ltd v Ampol Petroleum Ltd [1974] AC 821 (PC)�� 355 Humphrey v Dale (1857) 7 El & Bl 266, 119 ER 1246�� 213 Hunter v Moss [1993] 1 WLR 934 (Ch); [1994] 1 WLR 452 (CA)�� 140 Hutchinson v Bowker (1839) 5 M & W 535, 151 ER 227�� 209 Hutton v Warren (1836) 1 M & W 466, 150 ER 517�� 210 Hydarnes Steamship Co v Indemnity Mutual Marine Assurance Co [1895] 1 QB 500 (CA)�� 248 IMS SA v Capital Oil and Gas Industries Ltd [2016] EWHC 1956 (Comm), [2016] 4 WLR 163�� 98 Indian Endurance, The and Indian Grace, The. See Republic of India v India Steamship Co Ltd (No 2)�� 10 Indian Oil Corp Ltd v Greenstone Shipping Co SA (Panama) (The Ypatianna) [1988] QB 345 (Com Ct)��149, 324 Inglis v Robertson [1898] AC 616 (HL)�� 159 Ingram v Little [1961] 1 QB 31 (CA)�� 148 Inntrepreneur Pub Co (CPC) Ltd v Sweeney [2002] EWHC 1060 (Ch), [2002] EGLR 132�� 272 Inntrepreneur Pub Co Ltd v East Crown Ltd [2000] 2 Lloyd’s Rep 611 (Ch)��191, 272 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433 (CA)��194, 262 International Air Transport Assoc v Ansett Australia Holdings Ltd [2008] HCA 3, (2008) 234 CLR 151�� 248 International Finance Corp v DSNL Offshore Ltd [2005] EWHC 1844 (Comm), [2007] 2 All ER (Comm) 305�� 153 International Research Corp plc v Lufthansa Systems Asia-Pacific Pte Ltd [2013] SGCA 55, [2014] 1 SLR 130�� 265 Inverugie Investments Ltd v Hackett [1995] 1 WLR 713 (PC)�� 301

xxiv Table of Cases Investec Trust (Guernsey) Ltd v Glenalla Properties Ltd [2018] UKPC 7, [2019] AC 271�� 133 Investment Trust Companies (in liq) v Revenue and Customs Commissioners HMRC [2017] UKSC 29, [2018] AC 275��323, 342 Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (HL)��216, 218–19, 222, 227–29, 248, 270 Iran Amanat, The. See Owners of the Motor Vessel ‘Iran Amanat’ v KMP Coastal Oil Pte Ltd��116, 119 Iron Trades Mutual Insurance Co Ltd v Companhia de Seguros Imperio [1991] 1 Re LR 213 (Com Ct)�� 73 Isaac Cooke & Sons v Eshelby (1887) 12 App Cas 271 (HL)�� 30 Isle of Wight Railway Co v Tahourdin (1883) 25 ChD 320�� 352 Item Software (UK) Ltd v Fassihi [2004] EWCA Civ 1244, [2005] ICR 450��355–56 Jaggard v Sawyer [1995] 1 WLR 269 (CA)�� 298, 302–03 Jalamahon, The. See Ngo Chew Hong Edible Oil Pte Ltd v Scindia Steam Navigation Co Ltd�� 32 James Roscoe (Bolton) Ltd v Winder [1915] 1 Ch 62 (Ch)�� 325 Jarguh Sawit, The [1997] SGCA 59, [1997] 3 SLR(R) 829��114, 121 Jerome B Grubart Inc v Great Lakes Dredge & Dock Co 513 US 527, 115 S Ct 1043 (1995)�� 119 JI MacWilliam Co Inc v Mediterranean Shipping Co SA (The Rafaela S) [2003] EWCA Civ 556, [2004] QB 702�� 189 Joel v Law Union & Crown Insurance Co [1908] 2 KB 863 (CA)�� 70 John Shaw & Sons (Salford) Ltd v Shaw [1935] 2 KB 113 (CA)�� 347, 351–52 Johnson v Credit Lyonnais Co (1877) 3 CPD 32 (CA)�� 142 Johnson v Taylor Bros & Co Ltd [1920] AC 144 (HL)�� 44 Johnston & Sons Ltd v Holland [1988] 1 EGLR 264 (CA)�� 267 Joint Stock Co ‘Aeroflot-Russian Airlines’ v Berezovsky [2013] EWCA Civ 784, [2013] 2 Lloyd’s Rep 242�� 98 Jones v Bowden (1813) 4 Taunt 847, 128 ER 565�� 210 Jones v Bright (1829) 5 Bing 533, 130 ER 1167�� 206 Jones v Just (1868) LR 3 QB 197 (QB)�� 206 Joscelyne v Nissen [1970] 2 QB 86 (CA)�� 253 Josling v Kingsford (1863) 13 CB (NS) 447, 143 ER 177�� 206 K/S Merc-Scandia XXXXII v Lloyd’s Underwriters (The Mercandian Continent) [2001] EWCA Civ 1275, [2001] 2 Lloyd’s Rep 563��73, 82 Kaefer Aislamientos SA de CV v AMS Drilling Mexico SA de CV [2019] EWCA Civ 10, [2019] 1 WLR 3514�� 98 Kawasaki Kisen Kaisha v Regal-Beloit Corp 561 US 89 (2010)�� 58 Kelly v Fraser [2012] UKPC 25, [2013] 1 AC 450�� 186 Kennedy v Panama, New Zealand & Australian Royal Mail Co (Ltd) (1866–67) LR 2 QB 580�� 65 Kent v Vessel SS ‘Maria Luisa’ (No 2) [2003] FCAFC 93, (2003) 130 FCR 12�� 117 Khedive, The. See Stoomvaart Maatschappy Nederland v P & O Steam Navigation Co�� 16 Khurana v Webster Construction Ltd [2015] EWHC 758 (TCC), [2016] 1 All ER (Comm) 466�� 262 Kingstar Shipping Ltd v Owners of the Ship ‘Rolita’ (The Rolita) [1989] 1 HKLR 394�� 116 Kirkaldy (J) & Sons Ltd v Walker [1999] EWHC 839 (Comm), [1999] Lloyd’s Rep IR 410�� 75 Korea Shipping Corp v Lord Energy SA [2018] FCAFC 201, (2018) 363 ALR 312�� 109, 115–16

Table of Cases xxv Kos (No 2), The. See ENE Kos 1 Ltd v Petroleo Brasileiro SA (No 2)�� 159 Kowloon Development Finance Ltd v Pendex Industries Ltd [2013] HKCFA 35, (2013) 16 HKCFAR 336��253–54 Krys v KBC Partners LP [2015] UKPC 46�� 220 Kwei Tek Chao (t/a Zung Fu Co) v British Traders & Shippers Ltd [1954] 2 QB 459 (QB)�� 161 Kyrgyz Mobil Tel Ltd v Fellowes International Holdings Ltd (No 2) [2005] EWHC 1314 (Comm)�� 105 L’Estrange v F Graucob Ltd [1934] 2 KB 394 (KB)�� 263 Lagunas Nitrate Co v Lagunas Syndicate [1899] 2 Ch 392 (CA)�� 359 Lamare v Dixon (1873) LR 6 HL 414�� 65 Lambert v Co-operative Insurance Society Ltd [1975] 2 Lloyd’s Rep 485 (CA)�� 70 Laurie & Morewood v Dudin & Sons [1926] 1 KB 223 (CA)��141, 159 Lauritzen v Larsen 345 US 571, 73 S Ct 921 (1953)�� 120 Law v Law [1905] 1 Ch 140 (CA)�� 67 Lee v Chou Wen Hsien [1984] 1 WLR 1202 (PC)�� 363 Lehmann Timber, The. See Metall Market OOO v Vitorio Shipping Co Ltd�� 154 Leigh & Sillivan Ltd v Aliakmon Shipping Co Ltd (The Aliakmon) [1986] AC 785 (HL)�� 29 Leni, The. See Transworld Oil (USA) Inc v Minos Compania Naviera SAL�� 261 Leon v Casey [1932] 2 KB 576 (CA)��67, 77 Leonidas, The. See Bayoil SA v Seawind Tankers Corp�� 267 Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] AC 430��216, 234 Lethulier’s Case (1692) 2 Salk 443, 91 ER 384�� 209 Liberian Insurance Agency Inc v Mosse [1977] 2 Lloyd’s Rep 560 (QB)�� 75 Lickbarrow v Mason (1787) 2 TR 63, 100 ER 35; (1794) 5 TR 683, 101 ER 380��43–44, 142, 181 Life Association of Scotland v Foster (1873) 11 M 351 (IH)�� 67 Life Insurance Co of Australia Ltd v Phillips [1925] HCA 18, (1925) 36 CLR 60��243, 251 Lilly v Ewer (1779) 1 Doug KB 72, 99 ER 50��203, 205 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL)��314, 342 Lishman v Northern Maritime Insurance Co (1875) LR 10 CP 179 (Ex Ch)��67, 73 Litsion Pride, The. See Black King Shipping Corp v Massie��73, 82 Lloyd del Pacifico v Board of Trade (No 1) (1929) 35 Ll L Rep 217 (CA)�� 113 Lloyd v Spillit (1740) Barn C 384, 27 ER 689�� 290 Lohia v Lohia [2001] EWCA Civ 1691�� 286 London & Provincial Marine & General Insurance Co Ltd v Chambers (1900) 5 Com Cas 241�� 77 London Allied Holdings Ltd v Lee [2007] EWHC 2061 (Ch)�� 309 London Assurance Co v Mansel (1879) 11 Ch D 363 (Ch)�� 67 London Explorer, The. See Timber Shipping Co SA v London & Overseas Freighters Ltd�� 271 London Steam Ship Owners Mutual Insurance Assoc Ltd v Spain (The Prestige) [2015] EWCA Civ 333, [2015] 2 Lloyd’s Rep 33�� 93 London Steamboat Co v Owners of the Bywell Castle (The Bywell Castle) (1879) 4 PD 219 (CA)�� 15 London Wine Co (Shippers) Ltd, Re [1986] PCC 121 (Ch)�� 140 London, HMS [1914] P 72 (PDAD)�� 16 Lotus Cars Ltd v Southampton Cargo Handling plc (The Rigoletto) [2000] 2 All ER (Comm) 705 (CA)�� 35 Louis Dreyfus & Cie v Parnaso Cia Naviera SA (The Dominator) [1959] 1 QB 498 (Com Ct)�� 271

xxvi Table of Cases Lukoil Asia Pacific PTE Ltd v Ocean Tankers (PTE) Ltd (The Ocean Neptune) [2018] EWHC 163 (Comm), [2018] 2 All ER (Comm) 108�� 222 Lunn Poly Ltd v Liverpool & Lancashire Properties Ltd [2006] EWCA Civ 430, [2006] 2 EGLR 29�� 298 Lusitania, The. See Pierce v Bemis�� 3 Lycaon, The. See Elder Dempster Lines v Zaki Ishag�� 43 M&J Polymers Ltd v Imerys Minerals Ltd [2008] EWHC 344 (Comm), [2008] 1 Lloyd’s Rep 541�� 156 Mackay v Dick (1881) 6 App Cas 251 (HL)�� 155 Macpherson Train & Co Ltd v Howard Ross & Co Ltd [1955] 1 WLR 640 (QB)�� 163 Maersk Tangier, The. See AP Moller-Maersk A/S (t/a Maersk Line) v Kyokuyo Ltd�� 46 Maggbury Pty Ltd v Hafele Australia Pty Ltd [2001] HCA 70, (2002) 210 CLR 181�� 251 Maheno, The [1977] 1 Lloyd’s Rep 81 (NZ SC)�� 44 Mahkutai, The [1996] AC 650 (PC)��35, 58 Mainteck Services Pty Ltd v Stein Heurtey SA [2014] NSWCA 184, (2014) 89 NSWLR 633�� 246 Mamola Challenger, The. See Omak Maritime Ltd v Mamola Challenger Shipping Co�� 306 Manbre Saccharine Co Ltd v Corn Products Co Ltd [1919] 1 KB 198 (Com Ct)�� 162 Manchester, Sheffield & Lincolnshire Rly Co v Brown (1883) 8 App Cas 703 (HL)�� 263 Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd (The Star Sea) [1997] 1 Lloyd’s Rep 360 (CA); [2001] UKHL 1, [2003] 1 AC 469�� 65–66, 69, 72, 76–77, 79, 82, 85 Mannai Investment Co Ltd v Eagle Star Life Assurance Co Ltd [1997] AC 749 (HL)��216, 248 Mapfre Mutualidad Cia de Seguros Y Reaseguros SA v Keefe [2015] EWCA Civ 598, [2016] 1 WLR 905�� 100 Maralinga Pty Ltd v Major Enterprises Pty Ltd [1973] HCA 23, (1973) 128 CLR 336�� 252 Marathon Asset Management LLP v Seddon [2017] EWHC 300 (Comm), [2017] ICR 791��298, 302 Marielle Bolten, The. See Whitesea Shipping & Trading Corp v El Paso Rio Clara Ltda��34, 58 Marks & Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72, [2016] AC 742��216, 228–29, 256 Marlborough Hill, The. See Owners of the Ship ‘Marlborough Hill’ v Alex Cowan & Sons Ltd�� 43 Marley v Rawlings [2014] UKSC 2, [2015] AC 129�� 255 Martin Fierro, The [1975] 2 Lloyd’s Rep 130 (CA)�� 21 Martin v Martin [1959] HCA 62, (1959) 110 CLR 297�� 287 Mar-Train Heavy Haulage Ltd v Shipping.DK Chartering A/S (t/a Frank&Tobiesen A/S) [2014] EWHC 355 (Comm)�� 98 Masters v Cameron [1954] HCA 72, (1954) 91 CLR 353�� 253 Mathias v Pearce (1992) 6 NZCLC 68,102 (NZ HC)�� 356 Mayhew Foods Ltd v Overseas Containers Ltd [1984] 1 Lloyd’s Rep 317 (Com Ct)��46–47 McCann v Switzerland Insurance Australia Ltd [2000] HCA 65, (2000) 203 CLR 579�� 248 McCourt v Cranston [2012] WASCA 60, [2012] ANZ Conv R 12-0006�� 246 McEvoy v Belfast Banking Co Ltd [1934] NI 67�� 292 McPherson v Watt (1877) 3 App Cas 254 (HL)�� 67 Mediterranean Salvage & Towage Ltd v Seamar Trading & Commerce Inc (The Reborn) [2009] EWCA Civ 531, [2010] 1 All ER (Comm) 1�� 228

Table of Cases xxvii Mentink v Registrar of Australian Register of Ships [2015] FCAFC 150, (2015) 234 FCR 458��109, 113 Mercandian Continent, The. See K/S Merc-Scandia XXXXII v Lloyd’s Underwriters��73, 82 Mercuria Energy Trading Pte Ltd v Citibank NA [2015] EWHC 1481 (Comm), [2015] 1 CLC 999�� 159 Messiniaki Tolmi, The. See Astro Exito Navegacion SA v Southland Enterprise Co��153, 155 Metall Market OOO v Vitorio Shipping Co Ltd (The Lehmann Timber) [2013] EWCA Civ 650, [2014] QB 760�� 154 Meters Ltd v Metropolitan Gas Meters Ltd (1911) 28 RPC 157 (CA)�� 298 Meyerstein v Barber (1869–70) LR 4 HL 317��43, 161 Michael Gerson (Leasing) Ltd v Wilkinson [2001] QB 514 (CA)��152, 156 Middle East Tankers & Freighters Bunker Services SA v Abu Dhabi Container Lines PJSC [2002] EWHC 957 (Comm), [2002] 2 Lloyd’s Rep 643�� 31 Milan, The (1861) Lush 388, 167 ER 167��6–7 Mileform Ltd v Interserve Security Ltd [2013] EWHC 3386 (QB)�� 272 Miliangos v George Frank (Textiles) Ltd [1976] AC 443 (HL)�� 333 Miller v Race (1758) 1 Burr 452, 97 ER 398�� 340 Mills v Armstrong (The Bernina) (No 2) (1886) 11 PD 31 (PDAD); (1887) 12 PD 58 (CA); (1888) 13 App Cas 1 (HL)��5–8 Mills v Mills [1938] HCA 4, (1938) 60 CLR 150�� 355 Min Rui, The. See Consorcio MGT v Owner and/or Demise Chartere of the Vessel ‘Min Rui’�� 117 Ministry of Defence v Ashman (1993) 66 P & CR 195 (CA)�� 301 Ministry of Defence v Thompson [1993] 2 EGLR 107 (CA)�� 301 Ministry of Health v Simpson. See Diplock, Re��311, 319 Miramar Maritime Corp v Holborn Oil Trading Ltd (The Miramar) [1984] AC 676 (HL)�� 248 Miramar, The. See Miramar Maritime Corp v Holborn Oil Trading Ltd�� 248 Mitsubishi Heavy Industries Ltd v Gulf Bank KSC [1997] 1 Lloyd’s Rep 343 (CA)�� 248 Molton Builders Ltd v City of Westminster London BC (1975) 30 P & CR 182 (CA)�� 267 Monaghan CC v Vaughan [1948] IR 306 (HC)�� 252 Moon, Re (1887) 19 QBD 669 (CA)�� 67 Moorcock, The (1889) 14 PD 64 (CA)�� 155, 226, 229, 232 Moorgate Mercantile Credit Co Ltd v Twitchings [1977] AC 890 (HL)�� 143 Mopani Copper Mines plc v Millennium Underwriting Ltd [2008] EWHC 1331 (Comm), [2008] 1 CLC 992�� 271 Morley v Attenborough (1849) 3 Ex 500, 154 ER 943�� 205 Morris-Garner v One Step (Support) Ltd [2018] UKSC 20, [2019] AC 649�� 297, 299–307 Moschanthy, The [1971] 1 Lloyd’s Rep 37 (QB (Admlty))�� 116 Motortrak Ltd v FCA Australia Pty Ltd [2018] EWHC 990 (Comm)��259, 271 Motorworld Ltd v Turners Auctions Ltd [2010] NZHC 113�� 357 Mount Bruce Mining Pty Ltd v Wright Prospecting Pty Ltd [2015] HCA 37, (2015) 256 CLR 104�� 248 MSC Mediterranean Shipping Co SA v Cottonex Anstalt [2015] EWHC 283 (Comm), [2015] 2 All ER (Comm) 614; [2016] EWCA Civ 789, [2017] 1 All ER (Comm) 483��195, 240 Muckleston v Brown (1801) 6 Ves Jun 52, 31 ER 934��282, 284, 288, 290–93 Muir v City of Glasgow Bank (1879) 4 App Cas 337 (HL)�� 133 Mutford v Walcot (1700) 1 Ld Raym 574, 91 ER 1283�� 203

xxviii Table of Cases MV Athens, The. See Profindo Pte Ltd v Abani Trading Pte Ltd�� 156 MVF 3 APS (formerly Vestergaard Frandsen A/S) v Bestnet Europe Ltd [2016] EWCA Civ 541, [2017] FSR 5�� 298 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2016] EWCA Civ 553, [2017] QB 604; [2018] UKSC 24, [2019] AC 119�� 185, 190, 274 Napier Star, The [1939] P 330 (PDAD)�� 8 National Crime Agency v Dong [2017] EWHC 3116 (Ch), [2018] BPIR 477�� 286 National Roads and Motorists’ Association Ltd v Geeson [2001] NSWSC 832, (2001) 39 ACSR 401; [2001] NSWCA 343, (2001) 40 ACSR 1�� 348, 359–60 National Westminster Bank Ltd v Halesowen Presswork & Assemblies Ltd [1972] AC 785 (HL)�� 153 National Westminster Bank v Rabobank Nederland RV [2007] EWHC 1742 (Comm), [2008] 1 All ER (Comm) 243�� 105 Nazym Khikmet, The [1996] 2 Lloyd’s Rep 362 (CA)�� 115 Nea Tyhi, The [1982] 1 Lloyd’s Rep 606 (QB (Admlty))�� 29 Nelson v Larholt [1948] 1 KB 339 (KB)�� 318 Nelson v Nelson [1995] HCA 25, (1995) 184 CLR 538��282, 288–89, 293 Neville v Wilkinson (1782) 1 Bro CC 543, 28 ER 1289�� 236 New Hampshire Insurance Co v MGN Ltd [1997] LRLR 24 (CA)�� 79 New York Star, The. See Port Jackson Stevedoring Pty v Salmond & Spraggon (Australia) Pty��33–34 New Zealand Shipping Co Ltd v AM Satterthwaite & Co Ltd (The Eurymedon) [1975] AC 154 (PC)�� 33 Newcastle City Council v GIO General Ltd [1997] HCA 53, (1997) 191 CLR 85�� 247 Newton v Chorlton (1853) 10 Hare 646, 68 ER 1087�� 67 Ngo Chew Hong Edible Oil Pte Ltd v Scindia Steam Navigation Co Ltd (The Jalamohan) [1988] 1 Lloyd’s Rep 443 (Com Ct)�� 32 NHS Commissioning Board v Vasant [2019] EWCA Civ 1245�� 255 Nickoll & Knight v Ashton Edridge & Co [1901] 2 KB 126 (CA)�� 156 Norfolk Southern Railway Co v James N Kirby Pty Ltd 543 US 14, 125 S Ct 385 (2004)�� 119 North Star Shipping Ltd v Sphere Drake Insurance plc (No 2) [2005] EWHC 665 (Comm), [2005] 2 Lloyd’s Rep 76�� 68 NTUC Foodfare Cooperative Ltd v SIA Engineering Ltd [2018] SGCA 41, [2018] 2 SLR 588�� 29 Oatway, Re [1903] 2 Ch 356 (Ch)�� 325 Obestain Inc v National Mineral Development Corp Ltd (The Sanix Ace) [1987] 1 Lloyd’s Rep 465 (Com Ct)�� 33 OBG Ltd v Allan [2007] UKHL 21, [2008] 1 AC 1��197, 335 Occidental Worldwide Investment Corp v Skibs A/S Avanti (The Siboen and the Sibotre) [1976] 1 Lloyd’s Rep 293 (Com Ct)�� 73 Ocean Chemical Transport Inc v Exnor Craggs Ltd [2000] 1 All ER (Comm) 519 (CA)�� 264 Ocean Neptune, The. See Lukoil Asia Pacific PTE Ltd v Ocean Tankers (PTE) Ltd�� 222 Oceanbulk Shipping & Trading SA v TMT Asia Ltd [2010] UKSC 44, [2011] 1 AC 662�� 246 Oceanic Steam Navigation Co v Mellor 209 F 501 (1913); 233 US 718 (1914)�� 2 Ochroid Trading Ltd v Chua Siok Lui (trading as VIE Import & Export) [2018] SGCA 5, [2018] 1 SLR 363�� 280 Office of Fair Trading v Abbey National plc [2009] UKSC 6, [2010] 1 AC 696�� 190

Table of Cases xxix Ohm Mariana, The ex Peony. See Pacific Navigation Co Pte Ltd v Owners of and All Other Persons Interested in the Ship or Vessel ‘Ohm Mariana ex Peony’�� 117 OLG Karlsruhe, 24.5.67 (1967) ULC 289�� 51 Omak Maritime Ltd v Mamola Challenger Shipping Co (The Mamola Challenger) [2010] EWHC 2026 (Comm), [2011] 1 Lloyd’s Rep 47�� 306 OOCL Bravery, The. See Hartford Fire Insurance Co v Orient Oversease Container Line�� 58 Oppenheimer v Attenborough & Son [1908] 1 KB 221 (CA)�� 142 Orakpo v Barclays Insurance Services Co Ltd [1995] LRLR 443 (CA)��75–77 OT Africa Line Ltd v Magic Sportswear Corp [2005] EWCA Civ 710, [2005] 2 Lloyd’s Rep 170�� 105 Overseas Commodities Ltd v Style [1958] 1 Lloyd’s Rep 546 (Com Ct)�� 75 Owen v Homan (1851) 3 Mac & G 378, 42 ER 307�� 67 Owners of Cargo Lately Laden on Board the KH Enterprise v Owners of the Pioneer Container (The Pioneer Container) [1994] 2 AC 324 (PC)��34, 36, 44 Owners of Cargo of the Tongariro v Owners of the Drumlanrig (The Drumlanrig) [1911] AC 16 (HL)�� 6 Owners of Shin Kobe Maru v Empire Shipping Inc [1994] HCA 54, (1994) 181 CLR 404�� 116 Owners of the Motor Vessel ‘Iran Amanat’ v KMP Coastal Oil Pte Ltd (The Iran Amanat) [1999] HCA 11, (1999) 196 CLR 130��116, 119 Owners of the Ship ‘Marlborough Hill’ v Alex Cowan & Sons Ltd (The Marlborough Hill) [1921] 1 AC 444 (PC)�� 43 Oxonica Energy Ltd v Neuftec Ltd [2008] EWHC 2127 (Pat)�� 259 P&O Nedlloyd BV v Arab Metals Co (The UB Tiger) (No 2) [2006] EWCA Civ 1717, [2007] 1 WLR 2288�� 154, 159, 163 P4 Ltd v Unite Integrated Solutions plc [2006] EWHC 2640 (TCC)�� 146 Pacific Carriers Ltd v BNP Paribas [2004] HCA 35, (2004) 218 CLR 451�� 253 Pacific Motor Auctions Pty Ltd v Motor Credits (Hire Finance) Ltd [1965] AC 867 (PC)�� 143 Pacific Navigation Co Pte Ltd v Owners of and All Other Persons Interested in the Ship or Vessel ‘Ohm Mariana ex Peony’ (The Ohm Mariana, ex Peony) [1993] SGCA 43, [1993] 2 SLR(R) 698�� 117 Padre Island, The. See Firma C-Trade SA v Newcastle Protection & Indemnity Assoc; Socony Mobil Oil Inc v West of England Shipowners Mutual Insurance Assoc (London) Ltd (No 2)�� 92 Palmer v Pratt (1824) 2 Bing 185, 130 ER 277�� 66 Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501 (HL)�� 70, 73–74 Pan Ocean Co Ltd v China-Base Group Co Ltd [2019] EWHC 982 (Comm)�� 98 Pangkalan Susu/Permina 3001, The [1977] SGCA 5, [1977–78] SLR(R) 105�� 117 Parkinson v College of Ambulance Ltd [1925] 2 KB 1 (KB)�� 233 Parkinson v Lee (1802) 2 East 314, 102 ER 389�� 206 Parlux SpA v M & U Imports Pty Ltd [2008] VSCA 161, (2008) 21 VR 170�� 47 Pars Ram Brothers (Pte) Ltd v Australian & New Zealand Banking Group Ltd [2018] SGHC 60, [2018] 4 SLR 1404�� 148 Patel v Mirza [2016] UKSC 42, [2017] AC 467�� 216, 232–36, 239–41, 279–80, 282, 294–95 Patmore v Colburn (1834) 1 Cr M & R 65, 149 ER 996�� 273 PCW Syndicates v PCW Reinsurers [1996] 1 WLR 1136 (CA)��71–72 Pearce v Brooks (1866) LR Ex 213 (Ex Ct)�� 233

xxx Table of Cases Peekay Intermark Ltd v Australian & New Zealand Banking Group Ltd [2006] EWCA Civ 386, [2006] 1 CLC 582�� 263 Pegler Ltd v Wang (UK) Ltd (No 1) [2000] BLR 218 (TCC)�� 261 Peiniger Joint Venture v Argos Engineering & Heavy Industries Co Ltd [1994] HKCFI, [1994] 3 HKC 328�� 265 Pell Frischmann Engineering Co Ltd v Bow Valley Iran Ltd [2009] UKPC 45, [2011] 1 WLR 2370�� 298, 304–05 Penarth Dock Engineering Co Ltd v Pounds [1963] 1 Lloyd’s Rep 359 (QB)��154, 164, 298, 301 Pepper (Inspector of Taxes) v Hart [1993] AC 593 (HL)�� 247 Perella Weinberg Partners UK LLP v Codere SA [2016] EWHC 1182 (Comm)�� 98 Permina 108, The [1976] SGCA 12, [1974-76] SLR(R) 850�� 114 Petherpermal Chetty v Muniandi Servai [1908] UKPC 15, (1908) LR 35 Ind App 98�� 236 Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 (CA)�� 225–26, 257 Phillipps v Briard (1856) 1 Hurl & N 21, 156 ER 1101�� 213 Phillips Petroleum Co (UK) Ltd v Snamprogetti Ltd [2001] EWCA Civ 889, (2001) 79 Con LR 80�� 273 Phoenix General Insurance Co of Greece SA v Halvanon Insurance Co Ltd [1985] 2 Lloyd’s Rep 599 (Com Ct)�� 80 Pierce v Bemis (The Lusitania) [1986] QB 384 (QB (Admlty))��3–4 Pioneer Container, The. See Owners of Cargo Lately Laden on Board the KH Enterprise v Owners of the Pioneer Container��34, 36, 44 Pitman’s Patent, Re (1871) LR 4 PC 84�� 67 Playa Larga, The and Marble Islands, The. See Empresa Exportadora De Azucar v Industria Azucarera Nacional SA�� 161 Polemis & Furness, Withy & Co Ltd, Re [1921] 3 KB 560 (CA)�� 16 Pollock v Stables (1848) 12 QB 765, 116 ER 1057�� 212 Port Jackson Stevedoring Pty v Salmond & Spraggon (Australia) Pty (The New York Star) [1978] HCA 8, [1979] 1 Lloyd’s Rep 298; [1981] 1 WLR 138 (PC)��33–34 Porter v Zurich Insurance Co [2009] EWHC 376 (QB), [2009] 2 All ER (Comm) 658�� 69 Portsmouth, The. See TW Thomas & Co v Portsea Steamship Co Ltd�� 265 Power v Whitmore (1815) 4 M & S 141, 105 ER 787�� 203 Poyser v Minors (1881) 7 QBD 329 (CA)�� 283 Prenn v Simmonds [1971] 1 WLR 1381 (HL)��218, 225 Prest v Petrodel Resources Ltd [2012] EWCA Civ 1395, [2013] 2 WLR 557; [2013] UKSC 34, [2013] 2 AC 415�� 140 Prestige, The. See London Steam Ship Owners Mutual Insurance Assoc Ltd v Spain�� 93 Preston v Preston [1960] NZLR 385�� 292 PricewaterhouseCoopers v Saad Investments Co Ltd. See Saad Investments Co Ltd, Re��124–26 Princes Buitoni Ltd v Hapag-Lloyd AG [1991] 2 Lloyd’s Rep 383 (CA)�� 58 Probatina Shipping Co Ltd v Sun Insurance Office Ltd (The Sageorge) [1974] QB 635 (CA)��67, 77 Procter & Gamble Co v Svenska Cellulosa Aktiebolaget SCA [2012] EWHC 498 (Ch)�� 273 Profindo Pte Ltd v Abani Trading Pte Ltd (The MV Athens) [2013] SGHC 10�� 156 Prudential Assurance Co Ltd v Revenue and Customs Commissioners [2018] UKSC 39, [2018] 3 WLR 652��323, 333

Table of Cases xxxi PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2015] EWHC 2022 (Comm), [2015] 2 Lloyd’s Rep 563; [2015] EWCA Civ 1058, [2016] 2 WLR 1072; [2016] UKSC 23, [2016] AC 1034�� 139, 141, 153, 170–71, 178–79 Pyrene Co Ltd v Scindia Steam Navigation Co Ltd [1954] 2 QB 402 (QB)�� 46 Quantum Corp Inc v Plane Trucking Ltd [2002] EWCA Civ 350, [2002] 1 WLR 2678��45, 48 Quinn Building Products Ltd v P&S Civil Works Ltd [2013] NIQB 142�� 98 R (Barclay) v Lord Chancellor & Secretary of State for Justice (No 2) [2014] UKSC 54, [2015] AC 276�� 137 R (Westminster City Council) v National Asylum Support Services [2002] UKHL 38, [2002] 1 WLR 2956�� 246 R v Kensington Income Tax Commissioners, ex parte Princess Edmond de Polignac [1917] 1 KB 486 (CA)�� 67 Rafaela S, The. See JI MacWilliam Co Inc v Mediterranean Shipping Co SA�� 189 Rafsanjan Pistachio Producers Co-operative v Bank Leumi (UK) Ltd [1992] 1 Lloyd’s Rep 513 (Com Ct)�� 74 Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900��216, 220–21, 248 Ramzan v Brookwide Ltd [2011] EWCA Civ 985, [2012] 1 All ER 903�� 301 Randall v Newson (1877) 2 QBD 102 (CA)�� 206 Ravennavi SpA v New Century Shipbuilding Co Ltd [2007] EWCA Civ 58, [2007] 2 All ER (Comm) 756�� 270 Rayner v Ritson (1865) 6 B & S 888, 122 ER 1421�� 68 Rayware Ltd v Transport and General Workers’ Union [1989] 1 WLR 675 (CA)�� 248 RB Antwerpen, 7.12.73 (1976) 11 ETL 295�� 51 Reardon Smith Line Ltd v Hansen-Tangen (The Diana Prosperity) [1976] 1 WLR 989 (HL)��188, 219 Reborn, The. See Mediterranean Salvage & Towage Ltd v Seamar Trading & Commerce Inc�� 228 Reda v Flag Ltd [2002] UKPC 38, [2002] IRLR 747�� 269 Rederiakatiebolaget Navigator v Newcombe [1920] UKPC 26, [1920] 51 DLR 426�� 15 Redfern Resources Ltd, Re 2011 BCSC 771, (2011) 88 BLR (4th) 29�� 152 Redgrave v Hurd (1881) 20 Ch D 1 (CA)�� 65 Reese River Silver Mining Co Ltd v Smith (1869) LR 4 HL 64�� 65 Regie Nationale des Usines Renault SA v Zhang [2002] HCA 10, (2002) 210 CLR 491�� 118 Reigate v Union Manufacturing Co (Ramsbotton) Ltd [1918] 1 KB 592 (CA)�� 226 Relfo Ltd (in liq) v Varsani [2012] EWHC 2168 (Ch); [2014] EWCA Civ 360, [2015] 1 BCLC 14�� 342 Republic of India v India Steamship Co Ltd (The Indian Endurance and The Indian Grace) (No 2) [1998] AC 878 (HL)�� 10 Res Cogitans, The. See PST Energy 7 Shipping LLC v OW Bunker Malta Ltd�� 139, 141, 153, 170–71, 178–79 RGI International Ltd v Synergy Classic Ltd [2011] EWHC 3417 (Comm)�� 276 Richardson Stevedoring & Logistics Services Inc v Daebo International Shipping Co Ltd 2015 WL 1781712 (ED La 2015)�� 120 Richardson v Austin [1911] HCA 28, (1911) 12 CLR 463�� 248 Rigoletto, The. See Lotus Cars Ltd v Southampton Cargo Handling plc�� 35 River Wear Commissioners v Adamson (1877) 2 App Cas 743 (HL)�� 219 Riverlate Properties Ltd v Paul [1975] Ch 133 (CA)�� 252

xxxii Table of Cases Roberts v Roberts (1818) Dan 143, 159 ER 862�� 292 Robinson v Mollett (1874) LR 7 HL 802��67, 212 Rochefoucauld v Boustead [1897] 1 Ch 196 (CA)��287, 291 Rolita, The. See Kingstar Shipping Ltd v Owners of the Ship ‘Rolita’�� 116 Ross v Estates Investment Co (1866) LR 3 Eq 122 (Ch)�� 67 Rowe v Young (1820) 2 Bli 391, 4 ER 372�� 208 Rowland v Divall [1923] 2 KB 500 (CA)�� 141 RTS Flexible Systems Ltd v Molkerei Alois Müller GmbH & Co KG [2010] UKSC 14, [2010] 1 WLR 753�� 276 Rubin v Eurofinance SA [2012] UKSC 46, [2013] 1 AC 236��129–30 RV Ward Ltd v Bignall [1967] 1 QB 534 (CA)�� 164 Ryanair Ltd v SR Technics Ireland Ltd [2007] EWHC 3089 (QB)�� 272 Saad Investments Co Ltd, Re [2013] SC (Bda) 28; sub nom PricewaterhouseCoopers v Saad Investments Co Ltd [2013] CA (Bda) 7 Civ; [2014] UKPC 35, [2014] 1 WLR 4482��124–26 Saad Investments Co Ltd, Re [2013] SC (Bda) 28; sub nom Singularis Holdings Ltd v PricewaterhouseCoopers [2014] UKPC 36, [2015] AC 1675�� 125–32, 134 Saetta, The. See Forsyth International (UK) Ltd v Silver Shipping Co Ltd�� 148 Safeway Stores Ltd v Twigger [2010] EWCA Civ 1472, [2011] Bus LR 1629�� 234 Sageorge, The. See Probatina Shipping Co Ltd v Sun Insurance Office Ltd��67, 77 Salazar v The Atlantic Sun 881 F 2d 73, 1989 AMC 2594 (3d Cir 1989)�� 121 Salvador v Hopkins (1765) 3 Burr 1707, 97 ER 1057��64, 203 Salvage Association v Cap Financial Services [1995] FSR 654 (QB)�� 261 Sanix Ace, The. See Obestain Inc v National Mineral Development Corp Ltd�� 33 Saunders v Vautier (1841) 4 Beav 115, 49 ER 282��283, 294 Savills (UK) Ltd v Blacker [2017] EWCA Civ 68�� 31 Schiffshypothekenbank Zu Luebeck AG v Compton (The Alexion Hope) [1988] 1 Lloyd’s Rep 311 (CA)��267, 269 Schmoll Fils & Co Inc v Scriven Bros & Co (1924) 19 Ll L Rep 118 (KB)�� 161 Schwarz & Co (Grain) Ltd v St Elefterio ex Arion (Owners) (The St Elefterio) [1957] P 179 (PDAD)�� 116 Scottish & Newcastle International Ltd v Othon Ghalanos Ltd [2006] EWCA Civ 1750, [2007] 1 All ER (Comm) 1027; [2008] UKHL 11, [2008] 2 All ER 768�� 161 Scruttons Ltd v Midland Silicones Ltd [1962] AC 446 (HL)�� 33 Sea Endeavour I, The. See Case C-368/16 Assens Havn v Navigators Management (UK) Ltd�� 100–02, 106 Seager v Copydex Ltd (No 2) [1969] 1 WLR 809 (CA)�� 298 Seaman v Fonereau (1742) 2 Str 1183, 93 ER 1115�� 64 Seaton v Heath [1899] 1 QB 782 (CA)�� 70 Sempra Metals Ltd (formerly Metallgesellschaft Ltd) v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561�� 323 SERE Holdings Ltd v Volkswagen Group UK Ltd [2004] EWHC 1551 (Ch)�� 273 Severn Trent Water Ltd v Barnes [2004] EWCA Civ 570, [2004] 2 EGLR 95�� 302 Shagang Shipping Co Ltd v Ship ‘Bulk Peace’ [2014] FCAFC 48, (2014) 314 ALR 230�� 117 Shalson v Russo [2003] EWHC 1637 (Ch), [2005] Ch 281�� 309, 312, 325 Shell-Mex Ltd v Elton Cop Dyeing Co Ltd (1928) 34 Com Cas 39 (KB)��153, 155 Ship ‘Sam Hawk’ v Reiter Petroleum Inc [2016] FCAFC 26, (2016) 246 FCR 337�� 116 Ship ‘Gem of Safaga’ v Euroceanica (UK) Ltd [2010] FCAFC 14, (2010) 182 FCR 27�� 117

Table of Cases xxxiii Shipowners’ Mutual Protection and Indemnity Assoc (Luxembourg) v Containerships Denizcilik Nakliyat ve Ticaret AS (The Yusuf Cepnioglu) [2016] EWCA Civ 386, [2016] Bus LR 755�� 93 Shogun Finance Ltd v Hudson [2003] UKHL 62, [2004] 1 AC 919��187, 281 Shore v Wilson (1842) 9 Cl & Fin 355, 8 ER 450�� 245 Siboen, The and Sibotre, The. See Occidental Worldwide Investment Corp v Skibs A/S Avanti�� 73 Sigma Finance Corp, Re [2009] UKSC 2, [2010] 1 All ER 571�� 220 Signia Wealth Ltd v Vector Trustees Ltd [2018] EWHC 1040 (Ch)�� 277 Simic v New South Wales Land and Housing Corp [2016] HCA 47, (2016) 260 CLR 85�� 248, 253, 255 Singh v Ali [1960] AC 167 (PC)�� 282–84, 293 Singularis Holdings Ltd (in liq) v Daiwa Capital Markets Europe Ltd [2018] EWCA Civ 84, [2018] 1 WLR 2777�� 239 Singularis Holdings Ltd v PricewaterhouseCoopers. See Saad Investments Ltd, Re�� 125–32, 134 Sirius International Insurance Co (Publ) v FAI General Insurance Ltd [2004] UKHL 54, [2004] 1 WLR 3251�� 216, 218, 253 Skanska Rashleigh Weatherfoil Ltd v Somerfield Stores Ltd [2006] EWCA Civ 1732�� 248 Smith v Doe d Jersey (1821) 2 Brod & B 473, 129 ER 1048�� 244 Smith v Hughes (1870-71) LR 6 QB 597 (QB)��63, 85 Smith v Wilson (1832) 3 B & Ad 728, 110 ER 266�� 209 Société Anonyme d’Intermediaries Luxembourgeois (SAIL) v Farex Gie [1995] LRLR 116 (CA)��71–72 Society of Lloyd’s v Robinson [1999] 1 WLR 756 (HL)�� 248 Southern Foundries (1926) Ltd v Shirlaw [1939] 2 KB 206 (CA); [1940] AC 701 (HL)��155, 226 Southwark LBC v IBM UK Ltd [2011] EWHC 549 (TCC), 135 Con LR 136�� 174 Spectra International plc v Hayesoak Ltd [1997] 1 Lloyd’s Rep 153 (County Ct (Central London))�� 44 Speedlink Vanguard, The and European Gateway, The [1987] QB 206 (QB (Admlty Ct)�� 19 Spence v Union Marine Insurance Co Ltd (1867–68) LR 2 CP 427 (CP)�� 324 Spiliada, The. See Spiliada Maritime Corp v Cansulex Ltd�� 104 Spiliada Maritime Corp v Cansulex Ltd (The Spiliada) [1987] AC 460�� 104 Springwell Navigation Corp v JP Morgan Chase Bank (formerly Chase Manhattan Bank) [2010] EWCA Civ 1221, [2010] 2 CLC 705�� 272 St Albans City and District Council v International Computers Ltd [1996] 4 All ER 481 (CA)��172, 261 St Elefterio, The. See Schwarz & Co (Grain) Ltd v St Elefterio ex Arion (Owners)�� 116 St John Shipping Corp v Joseph Rank Ltd [1957] 1 QB 267 (QB)�� 233 Stag Line Ltd v Tyne Ship Repair Group Ltd (The Zinnia) [1984] 2 Lloyd’s Rep 211 (Com Ct)�� 31 Stapylton Fletcher Ltd, Re [1994] 1 WLR 1181 (Ch)�� 141 Star Sea, The. See Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd�� 65–66, 69, 72, 76–77, 79, 82, 85 Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG (The Alexandros T) [2014] EWHC 3068 (Comm), [2015] 2 All ER (Comm) 747; [2014] EWCA Civ 1010, [2014] 2 Lloyd’s Rep 544; [2013] UKSC 70, [2014] Bus LR 873��99, 105

xxxiv Table of Cases Starsin, The. See Homburg Houtimport BV v Agrosin Private Ltd�� 27–30, 32–36, 58–59, 268–69, 271 Statue of Liberty, The [1971] 2 Lloyd’s Rep 277 (HL)�� 21 Steamship Imo v La Compagnie Générale Transatlantique. See Compagnie Générale Transatlantique v Steamship ‘IMO’�� 14–15, 19 Stena Line Ltd v Merchant Navy Ratings Pension Fund Trustees Ltd [2011] EWCA Civ 543, [2011] PLR 223�� 228 Stobart Group Ltd v Tinkler [2019] EWHC 258 (Comm)��347–59 Stoomvaart Maatschappy Nederland v P & O Steam Navigation Co (The Khedive) (1882) 7 App Cas 795 (HL)�� 16 Strand Electric Engineering Co Ltd v Brisford Entertainments Ltd [1952] 2 QB 246 (CA)��298, 301 Stratton Finance Pty Ltd v Webb [2014] FCAFC 110, (2014) 314 ALR 166�� 246 Stuart v Wilkins (1778) 1 Doug KB 18, 99 ER 15�� 205 Studdy v Sanders (1826) 5 B & C 628, 108 ER 234�� 207 Sturz v De La Rue (1828) 5 Russ 322, 38 ER 1048�� 67 Sugar Properties (Derisley Wood) Ltd, Re [1988] BCLC 146 (Ch)�� 140 Super-Max Offshore Holdings Ltd v Malhotra [2017] EWHC 3246 (Comm)�� 358 Surgicraft Ltd v Paradigm Biodevices Inc [2010] EWHC 1291 (Ch)�� 273 Surrey CC v Bredero Homes Ltd [1993] 1 WLR 1361 (CA)�� 302 Sutcliffe v Chief Constable of West Yorkshire [1996] RTR 86 (CA)�� 184 Sutro & Co and Heilbut, Symons & Co, Re [1917] 2 KB 348 (CA)�� 44 Sutton v Tatham (1839) 10 Ad & El 27, 113 ER 11��203, 212 SW Strange Ltd v Mann [1965] 1 WLR 629 (Ch)�� 273 Swallowfalls Ltd v Monaco Yachting & Technologies SAM [2014] EWCA Civ 186, [2014] 2 All ER (Comm) 185�� 155 Swan, The. See Bridges & Salmon Ltd v Owner of The Swan�� 31 Swanwick v Sothern (1839) 9 Ad & E 895, 112 ER 1453�� 159 Swift v Dairywise Farms Ltd [2000] 1 WLR 1177 (Ch); [2001] EWCA Civ 145, [2003] 1 WLR 1606 (Note)�� 196 Syers v Bridge (1780) 2 Doug KB 526, 99 ER 335�� 208 Syers v Jonas (1848) 2 Ex 111, 154 ER 426�� 204, 206, 212 Tamares (Vincent Square) Ltd v Fairpoint Properties (Vincent Square) Ltd (No 2) [2007] EWHC 212 (Ch), [2007] 1 WLR 2167�� 298 Tassell v Lewis (1695) 1 Ld Raym 743, 91 ER 1397�� 203 Taurus Petroleum Ltd v State Oil Marketing Co of the Ministry of Oil, Republic of Iraq [2017] UKSC 64, [2018] AC 690�� 185 Taylor v Bowers (1876) 1 QBD 291 (CA)�� 236 Taylor v Briggs (1827) 2 Car & P 525, 172 ER 238�� 209 Taylor v Johnson [1983] HCA 5, (1983) 151 CLR 422��249, 252 Taylor v Plumer (1815) 3 M & S 562, 105 ER 721��313–14, 318, 324 Team Services plc v Kier Management and Design Ltd (1993) 63 BLR 76 (CA)�� 271 Tele2 International Card Co SA v Post Office Ltd [2009] EWCA Civ 9�� 277 Tennent v Tennent’s Trustees (1870–75) LR 2 Sc 6 (HL)�� 67 Thames Valley Power Ltd v Total Gas & Power Ltd [2005] EWHC 2208 (Comm), [2006] 1 Lloyd’s Rep 441�� 153 The Titanic: In re Oceanic Steam Nav Co Ltd 225 F 747 (2 Cir, 1915)�� 3 Thermo Engineers Ltd v Ferrymasters Ltd [1981] 1 WLR 1470 (QB)�� 52

Table of Cases xxxv Thomas Bates and Son Ltd v Wyndham’s (Lingerie) Ltd [1981] 1 WLR 505 (CA)�� 252 Thomas v Times Book Co Ltd [1966] 1 WLR 911 (Ch)�� 158 Thornbridge Ltd v Barclays Bank plc [2015] EWHC 3430 (QB)��193–94 Thornton v Shoe Lane Parking Ltd [1971] 2 QB 163 (CA)�� 262 Thorogood v Bryan (1849) 8 CB 115, 137 ER 452��6–7 Throckmerton v Tracy (1555) 1 Plow 145, 75 ER 222�� 244 Through Transport Mutual Insurance Assoc (Eurasia) Ltd v New India Assurance Co Ltd (The Hari Bhum) [2004] EWCA (Civ) 1598, [2005] 1 Lloyd’s Rep 67�� 93 Thunder Air Ltd v Hilmarsson [2008] EWHC 355 (Ch)��173, 197 Timber Shipping Co SA v London & Overseas Freighters Ltd (The London Explorer) [1972] AC 1 (HL)�� 271 Timmerman v Nervina Industries (International) Pty Ltd [1983] 1 Qd R 1��153, 155 Tinker v Tinker [1970] P 136 (CA)�� 292 Tinkler v Ferguson [2019] EWCA Civ 819�� 350 Tinsley v Milligan [1992] Ch 310 (CA); [1994] 1 AC 340 (HL)�� 233–34, 238, 279–80, 282–83, 286, 291–95 Tisand Pty Ltd v Owners of the Ship MV Cape Moreton (ex Freya) [2005] FCAFC 68, (2005) 143 FCR 43��107, 109 TNT Express Belgium v Mitsui Sumitomo Insurance Co Europe Ltd (2006) European Transport Law 228 (Belgium)�� 48 Toll (FGCT) Pty Ltd v Alphapharm Pty Ltd [2004] HCA 52, (2004) 219 CLR 165�� 253 Torenia, The. See Aktieselskabet de Danske Sukkerfabrikker v Bajamar Compania Naviera SA�� 33 Transworld Oil (USA) Inc v Minos Compania Naviera SAL (The Leni) [1992] 2 Lloyd’s Rep 48 (Com Ct)�� 261 Travelers Casualty and Surety Co of Canada v Sun Life Assurance Co of Canada (UK) Ltd [2006] EWHC 2885 (Comm)�� 105 Tribe v Tribe [1996] Ch 107 (CA)�� 236, 286, 294 Trollope & Colls Ltd v North West Metropolitan Regional Hospital Board [1973] 1 WLR 601 (HL)�� 227–28, 256 Trustee of the Property of FC Jones & Sons (a firm) v Jones [1997] Ch 159 (CA)�� 309–10, 314 Turf Club Auto Emporium Pte Ltd v Yeo Boong Hua [2018] SGCA 44�� 306 Turner v Harvey (1821) Jac 169, 37 ER 814�� 66 Turner v Jacob [2006] EWHC 1317 (Ch), [2008] WTLR 307�� 325 TW Thomas & Co v Portsea Steamship Co Ltd (The Portsmouth) [1912] AC 1 (HL)�� 265 UB Tiger, The. See P&O Nedlloyd BV v Arab Metals Co (No 2)��154, 163 UCP plc v Nectrus Ltd [2018] EWHC 380 (Comm), [2018] 1 WLR 3409�� 98 Uhde v Walters (1811) 3 Camp 16, 170 ER 1291�� 209 UK Housing Alliance (North West) Ltd v Francis [2010] EWCA Civ 117, [2010] Bus LR 1034��261–62 Ulster Swift Ltd v Taunton Meat Haulage Ltd [1977] 1 WLR 625 (CA)�� 47 United Bank Ltd v Asif (CA, 11 February 2000)��190, 274 University of Nottingham v Fishel [2000] ICR 1462 (QB)�� 354 Vale v Bayle (1775) 1 Cowp 294, 98 ER 1094�� 160 Vallejo v Wheeler (1774) 1 Cowp 143, 98 ER 1012�� 204 Vandervell v IRC [1967] 2 AC 291 (HL)�� 290 Venture, The [1908] P 218 (CA)�� 287

xxxvi Table of Cases Versloot Dredging BV v HDI Gerling Industrie Versicherung AG (The DC Merwestone) [2016] UKSC 45, [2017] AC 1�� 68 Victoria v The Commonwealth [1975] HCA 39, (1975) 134 CLR 81�� 247 Victoria, The (1888) 13 PD 125 (PDAD)�� 9 Virulite LLC v Virulite Distribution Ltd [2014] EWHC 366 (QB), [2015] 1 All ER (Comm) 204�� 274 Vitol SA v Conoil plc [2009] EWHC 1144 (Comm), [2009] 2 Lloyd’s Rep 466�� 154 Vitol SA v Esso Australia Ltd (The Wise) [1989] 2 Lloyd’s Rep 451 (CA)�� 162 Volcafe Ltd v Compania Sud Americana de Vapores SA (t/a CSAV) [2016] EWCA Civ 1103, [2017] QB 915; [2018] UKSC 61, [2019] AC 358�� 29, 35, 46, 184 Volute, The. See Admiralty Commissioners v Owners of the SS Volute�� 13 Vostok Shipping Co Ltd v Confederation Ltd [1999] NZCA 220, [2000] 1 NZLR 37��115, 117 Wainewright v Bland (1835) 1 Mood & R 481, 174 ER 165��67, 69 Wait v Baker (1848) 2 Exch 1, 154 ER 380�� 160 Wait, Re [1927] 1 Ch 606 (CA)�� 140, 152–53 Walford v Miles [1992] 2 AC 128 (HL)�� 194 Walker v Jackson (1842) 10 M & W 161, 152 ER 424�� 66 Walters v Morgan (1861) 3 De G F & J 718, 45 ER 1056�� 66 Ward v Hobbs (1878) 4 App Cas 13 (HL)�� 66 Warinco AG v Samor SpA [1977] 2 Lloyd’s Rep 582 (Com Ct); [1979] 1 Lloyd’s Rep 450 (CA)�� 163 Warren v Gurney [1944] 2 All ER 472 (CA)�� 287 Watson, Laidlaw & Co Ltd v Pott, Cassels & Williamson (1914) 31 RPC 104 (HL)�� 298 WD Fairway, The. See Dornoch Ltd v Westminster International BV�� 109, 111–12 Westbrook Resources Ltd v Global Metallurgical Inc [2009] EWCA Civ 310, [2009] 2 Lloyd’s Rep 224�� 274 Western Export Services Inc v Jireh International Pty Ltd [2011] HCA 45, (2011) 282 ALR 604�� 246 Westmid Packing Services Ltd (No 2), Re [1998] 2 BCLC 646 (CA)�� 345 Wheelton v Hardisty (1857) 8 El & Bl 232, 120 ER 86�� 67 Whitehead v Anderson (1842) 9 M&W 518, 152 ER 219�� 159 Whiteley v Delaney [1914] AC 132 (HL)�� 252 Whitesea Shipping & Trading Corp v El Paso Rio Clara Ltda (The Marielle Bolten) [2009] EWHC 2552 (Comm), [2010] 1 Lloyd’s Rep 648��34, 58 Whitlam v Australian Securities and Investments Commission [2003] NSWCA 183, (2003) 57 NSWLR 559�� 360 Whitwham v Westminster Brymbo, Coal & Coke Co [1896] 2 Ch 538 (CA)�� 298 Whitworth Street Estates (Manchester) Ltd v James Miller & Partners Ltd [1970] AC 583 (HL)�� 178 Wieler v Schilizzi (1856) 17 CB 619, 139 ER 1219�� 206 William Hill Organisation Ltd v Tucker [1999] ICR 291 (CA)�� 153 Williams v Rawlinson (1825) 3 Bing 71, 130 ER 440�� 66 Williams v Roffey Bros & Nicholls (Contractors) Ltd [1991] 1 QB 1 (CA)�� 185 Wilmar Trading Pte Ltd v Heroic Warrior Inc [2019] SGHC 143�� 29 Wilson v Smith (1764) 3 Burr 1550, 97 ER 975�� 203 Winson, The. See China-Pacific SA v Food Corp of India�� 159 Wise, The. See Vitol SA v Esso Australia Ltd�� 162 Wolff v Horncastle (1798) 1 Bos & P 316, 126 ER 924��66, 69

Table of Cases xxxvii Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173�� 216, 221, 248, 250 Woolmer v Muilman (1763) 3 Burr 1419, 97 ER 905�� 204 Worcester Works Finance Ltd v Cooden Engineering Co Ltd [1972] 1 QB 210 (CA)�� 143–44, 146–48, 152 World Fuel Services Singapore Pte Ltd v MV AS Varesia 727 Fed Appx 811, 2018 AMC 686 (5th Cir 2018)�� 121 World Online Telecom UK Ltd (formerly Localtel Ltd) v I-Way Ltd [2002] EWCA Civ 413�� 190 Wrotham Park Estate Ltd v Parkside Homes Ltd [1974] 1 WLR 798 (Ch)�� 297 Wuhan Ocean Economic & Technical Cooperation Co Ltd v Shiffahrts-Gesellschaft ‘Hansa Murcia’ mbH & Co KG [2012] EWHC 3104 (Comm), [2013] 1 All ER (Comm) 1277�� 228 Yallop, Ex parte (1808) 15 Ves Jun 60, 33 ER 677�� 292 Yam Seng Pte Ltd v International Trade Corp [2013] EWHC 111 (QB), [2013] 1 All ER (Comm) 1321��195, 240 Yearworth v North Bristol NHS Trust [2009] EWCA Civ 37, [2010] QB 1�� 196 Youell v Bland Welch & Co Ltd (No 1) [1992] 2 Lloyd’s Rep 127 (Com Ct)�� 266 Your Response Ltd v Datateam Business Media Ltd [2014] EWCA Civ 281, [2015] QB 41��173, 197 Ypatianna, The. See Indian Oil Corp Ltd v Greenstone Shipping Co SA (Panama)��149, 324 Yusuf Cepnioglu, The. See Shipowners’ Mutual Protection and Indemnity Assoc (Luxembourg) v Containerships Denizcilik Nakliyat ve Ticaret AS�� 93 Zhu v Treasurer of New South Wales [2004] HCA 56, (2004) 218 CLR 530�� 248 Zinnia, The. See Stag Line Ltd v Tyne Ship Repair Group Ltd�� 31 Zurich Australian Insurance Ltd v Metals & Minerals Insurance Pte Ltd [2007] WASC 62, (2007) 209 FLR 247�� 80

xxxviii

TABLE OF LEGISLATION UK Legislation Statutes Administration of Justice Act 1982 s 21�� 247 Bankruptcy Act 1914 s 122�� 127 Bills of Exchange Act 1821�� 208 Bills of Exchange Act 1882��183, 189 s 19(2)(c)�� 208 Bills of Lading Act 1855�� 45–46, 188 Bills of Sale Act 1878�� 261 Carriage of Goods by Sea Act 1936��57–58 Carriage of Goods by Sea Act 1992��38, 188 s 1(2)(b)�� 45 s 1(3)(a)�� 45 s 2(1)(b)�� 46 s 3(1)�� 162 s 5(1)�� 45 Colonial Courts of Admiralty Act 1890�� 7 Common Law Procedure Act 1854�� 7 Companies Act 2006��351–52 s 171(b)�� 363 s 172�� 354 Consumer Credit Act 1974��169, 172 Consumer Rights Act 2015�� 165, 176, 179, 197, 260–61 pt 1 pt 1, ch 2�� 169 pt 1, ch 3��173–74 s 2�� 174 s 3(2)�� 169 s 4�� 177 s 4(1)�� 140 s 5�� 177 s 5(2)�� 172 s 8��176–77 s 9�� 170 s 10�� 170 s 11�� 170

xl Table of Legislation s 12�� 170 s 13�� 170 s 14�� 170 s 15�� 170 s 16�� 170 s 17�� 170 s 18�� 170 s 19�� 170 s 20�� 170 s 21�� 170 s 22�� 170 s 23�� 170 s 24�� 170 s 25�� 170 s 26�� 170 s 27�� 170 s 28��170, 177 s 29��170, 177 s 30�� 170 s 49�� 172 Contract (Rights of Third Parties) Act 1999�� 33 s 6(5)�� 36 Criminal Justice Act 1993 s 52�� 233 Equality Act 2010 s 199��285, 287 Factors Act 1877 s 3�� 142 Factors Act 1889�� 149 s 8�� 139, 141–42, 144–45, 148 s 9�� 145 Fatal Accidents Act 1846�� 5 Fatal Accidents Act 1976�� 5 Hire Purchase Act 1964 pt III�� 187 s 27�� 169 Insolvency Act 1985�� 127 Insurance Act 2015�� 83–88, 177 s 2(2)�� 84 s 3��83, 93 s 3(4)(b)�� 86 s4 s 4(3)(b)�� 83 s 4(8)(b)�� 83 s 8��83–84 s 10��74, 93 s 11�� 93 s 12�� 85

Table of Legislation xli s 14

s 14(1)�� 84 s 14(3)(a)�� 84 s 16�� 85 s 17��63, 77, 85 s 21(2)�� 84 s 22 s 22(1)�� 83 s 22(3)�� 84 sch 1, pt 1�� 84 sch 1, pt 2�� 84 Insurances on Ships etc Act 1785�� 66 Judicature Act 1873 s 25(9)�� 7 Judicature Acts 1873–75�� 7 Law of Property Act 1925 s 53(1)(b)�� 286 s 60(3)�� 286 Law Reform (Contributory Negligence) Act 1945�� 20 Marine Insurance Act 1788�� 66 Marine Insurance Act 1906�� 70–83, 87, 183 s 17�� 63, 70–78, 81–84, 86, 88 s 18�� 63, 71–74, 84, 88 s 19�� 63, 71–74, 84, 88 s 19(a)�� 63, 72–74, 84, 88 s 19(b)�� 63, 72–74, 84, 88 s 20�� 63, 71–74, 84, 88 s 31(1)�� 75 Maritime Conventions Act 1911�� 15, 20–21 s 1�� 7 s 1(1)�� 6 s 5�� 7 s9 s 9(1)�� 16 s 9(2)�� 16 Matrimonial and Family Proceedings Act 1984�� 124 Merchant Shipping Act 1854 s 504�� 2 Merchant Shipping Act Amendment Act 1862 s 54�� 2 Merchant Shipping Act 1894 s 429�� 2 s 466�� 4 s 503�� 2, 9–10, 14, 24 s 503(2)�� 2 s 738�� 19 Merchant Shipping Act 1906�� 7 s 69�� 2

xlii Table of Legislation Merchant Shipping Act 1979�� 23 Merchant Shipping Act 1995 s 163A(2)�� 93 s 165 s 165(1A)�� 93 s 165(5)�� 93 ch III�� 93 s 173�� 93 s 174�� 93 s 175�� 93 s 176�� 93 s 177�� 93 s 178�� 93 s 179�� 93 s 180�� 93 s 181�� 93 s 185(4)�� 22 s 187�� 20 s 187(1)�� 6, 20 s 188�� 20 s 189�� 20 s 267�� 19 s 268�� 19 s 314(1)�� 20 sch 1 sch 1, reg 1(1)�� 112 sch 1, reg 1(2)�� 112 sch 12�� 20 Merchant Shipping (Liability of Shipowners and Others) Act 1958�� 5 Merchant Shipping (Safety of Load Lines Conventions) Act 1932 s 29�� 19 Misrepresentation Act 1967 s 3�� 193 s 3(1)��191, 193 Moneylenders Act 1927�� 261 Patents Act 1977 s 30(1)�� 196 Private International Law (Miscellaneous Provisions) Act 1995 s 9�� 100 s 10�� 100 s 11�� 100 s 12�� 100 s 13�� 100 s 14�� 100 s 15�� 100 Proceeds of Crime Act 2002��175, 316 Railway and Canal Traffic Act 1854�� 261 s 7�� 263

Table of Legislation xliii Sale of Goods Act 1893��143, 152, 155, 165, 175, 183 s 12�� 206 s 13�� 206 s 14�� 206 s 15�� 206 s 17�� 281 s 30�� 206 Sale of Goods Act 1979�� 165–180 s 2��165, 167, 172, 174–75 s 2(1)�� 146, 167, 170 s 2(2)�� 140, 146, 167 s 2(4)�� 167 s 3�� 167 s 3(2)��147, 167 s 4�� 167 s 5�� 167 s 6�� 167 s 7�� 167 s 8�� 167 s 9�� 167 s 10�� 167 s 10(2)��163, 167 s 11�� 167 s 12�� 167, 177, 216 s 12(1)�� 147, 167, 216 s 13�� 167, 177, 216 s 14��167, 216 s 15��167, 216 s 15A��167, 216 s 16��139–40, 167, 177 s 17�� 113, 139, 167, 177 s 18�� 139, 160, 167, 177 rule 5(2)�� 139, 160, 167, 177 s 19�� 139, 167, 177 s 20�� 167 s 20(1)��154, 167 s 20(2)��154, 167 s 20A��139, 144–50, 167, 177 s 20A(1)��139, 146, 149, 167, 177 s 20A(1)(a)�� 139, 149, 167, 177 s 20A(1)(b)�� 139, 149, 167, 177 s 20A(4)��139, 148–49, 167, 177 s 20B�� 139, 147, 149–50, 167, 177 s 20B(1)(a)�� 139, 147, 167, 177 s 20B(1)(b)�� 139, 147, 167, 177 s 20B(2)�� 139, 147, 167, 177 s 20B(3)(a)�� 139, 147, 167, 177

xliv Table of Legislation s 21��167, 177 s 22��167, 177 s 23��167, 177 s 24�� 139, 141–42, 144–45, 147–50, 167, 177 s 25�� 145–46, 167–68, 177 s 25(1)�� 147, 167, 177 s 25(2)�� 167, 169, 177 s 26��167, 177 s 27��151–52, 155, 159, 167 s 28��163, 167 s 29�� 167 s 29(2)��157, 167 s 29(4)�� 156, 159, 167 s 30�� 167 s 31�� 167 s 32��160, 167 s 32(1)�� 160–61, 167 s 33�� 167 s 34�� 167 s 35�� 167 s 36�� 167 s 37��163, 167 s 37(1)�� 151–53, 156–58, 164, 167 s 37(2)�� 152, 163, 167 s 38�� 167 s 39�� 167 s 40�� 167 s 41�� 167 s 42�� 167 s 43�� 167 s 43(1)��160, 167 s 44�� 167 s 45�� 167 s 46�� 167 s 47��145, 167 s 48�� 167 s 48(3)��163, 167 s 49�� 167, 171, 177 s 49(1)�� 153, 167, 171, 177 s 49(2)�� 167, 171, 177 s 50��167, 177 s 50(3)�� 167, 188, 177 s 51�� 167 s 51(3)��167, 188 s 52�� 167 s 53�� 167 s 54�� 167

Table of Legislation xlv s 61�� 155, 167, 175 s 61(1)��140, 146 s 62�� 167 s 63�� 172 Sale of Goods (Amendment) Act 1994�� 141 Sale of Goods (Amendment) Act 1995��139, 149 Senior Courts Act 1981��297, 300 s 20 s 20(2)�� 119 s 20(2)(h)�� 118 s 21�� 5 s 21(4)(b)�� 116 s 21(4)(i)�� 116 s 37�� 104 Statute of Frauds 1677 s 7��289–90 Supply of Goods (Implied Terms) Act 1973 s 8��169,172 s 9��169, 172 s 10��169, 172 s 11��169, 172 Supply of Goods and Services Act 1982 s 1�� 172 s 1(1)�� 172 s 1(3)��172, 175 s 2�� 172 s 3�� 172 s 4�� 172 s 5�� 172 s 7�� 169 s 8�� 169 s 9�� 169 s 10�� 169 s 13�� 172 Third Parties (Rights against Insurers) Act 1930�� 92 Third Parties (Rights against Insurers) Act 2010��93, 100 s1 s 1(2)�� 92 s 1(3)�� 92 s 1(4)�� 92 s2 s 2(2)�� 92 s 2(4)�� 92 s 2(6)�� 92 s 2(7)�� 92 s 2(8)�� 92 s 2(9)�� 92

xlvi Table of Legislation s 6�� 92 s 6A�� 92 s9 s 9(5)��23, 92, 94 s 9(6)��23, 92, 94 s 9(7)�� 23 s 12(1)�� 92 Torts (Interference with Goods) Act 1977��173, 197, 320–21 Unfair Contract Terms Act 1977�� 57, 191–93, 260 s 3�� 261 Value Added Tax Act 1994�� 175 Wills Act 1837 s 9�� 255 Statutory Instruments Civil Procedure Rules 1998, SI 1998/3132 r 6.37(3)�� 104 r 58.1(2)�� 183 r 59�� 183 Part 6�� 104 Practice Direction 6B, para 3.1�� 104 Practice Direction 6B, para 3.1(6)(c)��98, 104 Practice Direction 6B, para 3.1(6)(d)�� 104 Practice Direction 6B, para 3.1(8)�� 104 Part 61 Practice Direction 61�� 20 Commercial Agents (Council Directive) Regulations 1993, SI 1993/3053��174, 197 reg 2(1)�� 173 Companies (Model Articles) Regulations 2008, SI 2008/3229 sch 1�� 345 sch 1, Art 7�� 345 sch 1, Art 8�� 345 Copyright and Rights in Databases Regulations 1997, SI 1997/3032 reg 13�� 197 reg 16�� 197 Merchant Shipping (Oil Pollution) (Bunkers Convention) Regulations 2006, SI 2006/1244�� 93 Unfair Terms in Consumer Contracts Regulations 1999. SI 1999/2083 reg 5(1)�� 261 Overseas Legislation European Union Agreement between the European Community and the Kingdom of Denmark of 21 March 2013 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (OJ L79/4)�� 97

Table of Legislation xlvii Convention 88/592/EEC of 25 November 1988 on jurisdiction and the enforcement of judgments in civil and commercial matters [1988] OJ L319/9 (EFTA Convention 1988)�� 97 Convention of 10 June 2009 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2009] OJ L147/5 (revised Lugano Convention)��97, 102 Art 4�� 103 Art 23�� 103 Art 23(3)�� 103 Council Directive 2006/112/EC of 11 December 2006 on the common system of value added tax [2006] OJ L347/1 (VAT Directive) Art 135(e)�� 175 Council Regulation (EC) No 1346/2000 of 29 May 2000 on insolvency proceedings (Insolvency Regulation 2000)�� 97 Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (Brussels I Regulation)�� 97, 100–01 Art 4�� 103 Art 14�� 102 Art 23(3)�� 103 ch II, s 3�� 99 European Parliament and Council Directive 1999/44/EC of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees [1999] OJ L171/12 (Consumer Sales Directive) Art 1(4)�� 177 Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) [2007] OJ L199/40 (Rome II Regulation) Art 18��100, 102 Regulation (EU) 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2012] OJ L351/1 (Recast Brussels Regulation)��104, 125 Art 1 Art 1(1)�� 97 Art 1(2)(b)�� 97 Art 1(2)(d)�� 97 Art 4��97, 99 Art 6�� 103 Art 7�� 97 Art 7(5)�� 99 Art 8�� 97 Art 9�� 97 Art 10�� 100 Art 11�� 100 Art 11(1)(a)��99–100 Art 11(1)(b)��99–100 Art 11(1)(c)��99–100 Art 11(2)��99–100

xlviii Table of Legislation Art 12��99–101 Art 13��100, 102 Art 13(2)�� 101 Art 13(3)�� 100 Art 14 Art 14(1)�� 100 Art 15�� 99, 101–02 Art 15(5)�� 97, 99–100 Art 16��99–102 Art 16(2)(a)��99–100 Art 25�� 97–99, 103 Art 25(1)��98, 103 Art 25(4)�� 101 Art 26�� 97 Art 29�� 98 Art 29(1)�� 99 Art 31 Art 31(2)��98–99 Art 31(3)�� 99 Art 33��97, 103 Art 34��97, 103 Art 66�� 97 Art 71��51, 97, 103 recital 12�� 96 recital 14 recital 14, s 3�� 97 recital 14, s 4�� 97 recital 14, s 5�� 97 recital 18�� 102 recital 19�� 102 recital 22�� 99 recital 22, para 2�� 99 Regulation (EU) 2015/848 of the European Parliament and of the Council of 20 May 2015 on insolvency proceedings [2015] OJ L141/19 (Recast Insolvency Regulation) Art 84 Art 84(1)�� 97 Art 84(2)�� 97 Australia Acts Interpretation Act 1901 (Cth) s 15AA�� 247 s 15AB�� 247 Acts Interpretation Act 1931 (Tas) s 8A�� 247 s 8B�� 247

Table of Legislation xlix Acts Interpretation Act 1954 (Qld) s 14A�� 247 s 14B�� 247 Admiralty Act 1988 (Cth)��7, 118 s4 s 4(2)�� 119 s 4(3)�� 119 s 17 s 17(a)�� 116 s 17(b)��116–17 s 19(b)�� 117 Competition and Consumer Act 2010 (Cth) sch 2 sch 2, s 2(1)�� 174 Corporations Act 2001 (Cth)�� 360 Federal Court of Australia Act 1976 (Cth) s 14(3)�� 118 Goods Act 1958 (Vic) s 22�� 113 Interpretation Act 1984 (WA) s 18�� 247 s 19�� 247 Interpretation Act 1987 (NSW) s 33�� 247 s 34�� 247 Interpretation Act 1987 (NT) s 62A�� 247 s 62B�� 247 Interpretation of Legislation Act 1984 (Vic) s 35 s 35(a)�� 247 s 35(b)�� 247 Legislation Act 2001 (ACT) s 139�� 247 s 141�� 247 s 142�� 247 s 143�� 247 Sale of Goods Act 1895 (SA) s 17�� 113 Sale of Goods Act 1895 (WA) s 17�� 113 Sale of Goods Act 1896 (Qld)�� 155 s 20�� 113 Sale of Goods Act 1896 (Tas) s 22�� 113

l Table of Legislation Sale of Goods Act 1923 (NSW) s 22�� 113 Sale of Goods Act 1954 (ACT) s 22�� 113 Sale of Goods Act 1972 (NT) s 22�� 113 Shipping Registration Act 1981 (Cth) s 77(1)�� 112 Shipping Registration Regulations 2019 (Cth) reg 37�� 112 Succession Act 1981 (Qld) s 33C�� 247 Succession Act 2006 (NSW) s 32�� 247 Wills Act 1968 (ACT) s 12B�� 247 s 12B(b)�� 247 s 12B(c)�� 247 Wills Act 1970 (WA) s 28A�� 247 Wills Act 1997 (Vic) s 36�� 247 s 36(1)(b)�� 247 s 36(1)(c)�� 247 Wills Act 2008 (Tas) s 46�� 247 Wills Act (NT) s 31�� 247 Bermuda Companies Act 1981 (Bermuda)��127–28 s 161�� 130 s 184�� 124 s 195�� 125–26, 130 External Companies (Jurisdiction in Actions) Act 1885 (Bermuda)�� 125 Canada Admiralty Act 1891�� 7 Canada Shipping Act 1906 pt X�� 4 s 781�� 4 Marine Liability Act 2001 pt 2�� 16 Maritime Conventions Act 1914�� 16 Sale of Goods Act (Ont) RSO 1990 c S 1 s 18�� 113

Table of Legislation li Sale of Goods Act (Novia Scotia) RS c 408 s 20�� 113 Sale of Goods Act (BC) RSBC 1996 c 410 s 22�� 113 Cayman Cayman Bankruptcy Law (1997 revn) s 107�� 127 s 156�� 127 Denmark Forsikringsaftaleloven (Law on Insurance Contracts) Art 95�� 100 France Code Civil 1804��216, 239 Art 5�� 238 Art 6�� 236 Art 1108��223, 236 Art 1131��236–37 Art 1133�� 236 Art 1134�� 231 Art 1135��230–31 Art 1156�� 223 Art 1157��223–24 Art 1158�� 223 Art 1159�� 223 Art 1160�� 223 Art 1161�� 223 Art 1162�� 223 Art 1163�� 223 Art 1164�� 223 Art 1902�� 238 Code Civil (new provisions created by Ordonnance nº 2016-131 of 10 February 2016, JUSC1522466R) Title III�� 217 Title III, ch I�� 223 Title III, ch II�� 223 Title III, ch II, s IV�� 238 Title III, ch III�� 223 Art 5�� 238 Art 6��237, 239 Art 1102�� 223 Art 1103��223, 231 Art 1104�� 223, 231, 234, 240 Art 1106�� 215

lii Table of Legislation Art 1128��223, 237, 239–40 Art 1162�� 237, 239–40 Art 1178��237–39 Art 1179��237–38 Art 1180��237–38 Art 1181��237–38 Art 1182��237–38 Art 1183��237–38 Art 1184��237–38 Art 1185��237–38 Art 1188��223–24 Art 1189�� 223 Art 1190�� 223 Art 1191��223–24 Art 1192��223–24 Art 1194��230–31 Art 1352��237–38 Art 1353��237–38 Art 1354��237–38 Art 1355��237–38 Art 1356��237–38 Art 1357��237–38 Art 1358��237–38 Art 1359��237–38 Loi de ratification n 2018-287 of 20 April 2018�� 216 Loi n 70-9 of 2 January 1970 Art 6�� 238 Ordonnance nº 2016-131 of 10 February 2016, JUSC1522466R�� 216 Art 1188�� 249 Report to the President of the Republic on Ordonnance nº 2016-131 of 10 February 2016, NOR: JUSC 1522466P�� 217 Germany German Civil Code BGB § 433�� 151 Guernsey Court of Appeal (Guernsey) Law 1961 s 16�� 137 The Trusts (Guernsey) Law s 42�� 133 s 65�� 134 Hong Kong High Court Ordinance s 12A(2)�� 119 s 12B

Table of Legislation liii s 12B(4)(b)�� 116 s 12B(4)(i)�� 116 Isle of Man Companies Act 1931 (Isle of Man)��127, 129 Jersey The Trusts (Jersey) Law Art 32�� 133 New Zealand Admiralty Act 1973 s 4(1)�� 119 s5 s 5(2)(b)�� 116 s 5(2)(i)�� 116 Companies Act 1993 s 133�� 363 s 149�� 363 s 178�� 364 Consumer Guarantees Act 1993 s 2�� 174 Contract and Commercial Law Act 2017 s 144�� 113 Financial Markets Conduct Act 2013 s 242�� 363 Ship Registration Act 1992 s 80(1)�� 112 Singapore High Court (Admiralty Jurisdiction) Act c 123 s 3(1)�� 119 s4 s 4(4)(b)�� 116 s 4(4)(i)�� 116 s 4(4)(ii)�� 114 Merchant Shipping Act c 179 s 186(1)�� 112 s 186(2)�� 112 Sale of Goods Act c 393 s 17�� 113 USA 9 United States Code § 8�� 120 46 United States Code § 184(b)�� 2 46 United States Code § 12103�� 112

liv Table of Legislation 46 United States Code § 12112�� 112 46 United States Code § 12115(b)�� 112 46 United States Code § 12134(3)�� 112 46 United States Code § 55102�� 112 28 United States Code § 1333(1)�� 119 46 Code of Federal Regulations § 67.53(c)�� 112 46 Code of Federal Regulations § 67.53(d)�� 112 Federal Rules of Civil Procedure Supp Admlty r B(1)(a)�� 119 r C�� 119 r E(4)(f)�� 120 Uniform Commercial Code Art 2��141, 183 s 2-105(4)�� 150 s 2-209�� 191 Art 2A�� 183 Art 3�� 183 Art 4�� 183 Art 4A�� 183 rev Art 5�� 183 rev Art 6�� 183 Art 7�� 183 rev Art 8�� 183 Art 9�� 183

TABLE OF TREATIES AND CONVENTIONS International Conventions Appendix B to the Convention concerning International Carriage by Rail (COTIF) of 9 June 1999 Uniform Rules concerning the Contract of International Carriage of Goods by Rail (CIM), applicable with effect from 1 July 2006 (COTIF/CIM)��46–47, 49, 55 Art 1 Art 1(3)�� 53 Art 1(4)�� 53 Art 5�� 50 Art 23�� 53 Art 38�� 53 Art 40�� 51 Art 44�� 51 Art 46�� 51 Art 48 Art 48(1)�� 50 Athens Convention relating to the Carriage of Passengers and their Luggage by Sea 1974 (Athens Convention 1974) Art 18�� 23 Athens Convention relating to the Carriage of Passengers and their Luggage by Sea 2002 (Athens Convention 2002)��22, 103 Art 4bis(1)�� 93 Art 7�� 23 Art 17(1)�� 93 Budapest Convention on the Contract for the Carriage of Goods by Inland Waterways (Budapest, 3 October 2000) (CMNI)�� 46–47, 49, 59 Art 2(2)�� 52 Art 17(3)�� 51 Art 20(1)�� 50 Art 25�� 51 Convention for the Unification of Certain Rules of Law with respect to Collision between Vessels (Brussels, 23 September 1910) (Collision Convention 1910)��16, 18, 20 Convention for the Unification of Certain Rules for International Carriage by Air 1999 (Montreal, 28 May 1999) (Montreal Convention)�� 46–47, 59–60 Art 13(3)�� 51 Art 18 Art 18(1)�� 53 Art 18(3)�� 53 Art 18(4)�� 49, 53–54

lvi Table of Treaties and Conventions Art 26�� 50 Art 30�� 51 Art 33�� 51 Art 36�� 37 Art 36(3)�� 51 Art 38�� 49 Art 39�� 26 Art 40�� 26 Convention for the Unification of Certain Rules of Law respecting Assistance and Salvage at Sea (Brussels, 23 September 1910) (Salvage Convention 1910)�� 18 Convention for the Unification of Certain Rules Relating to International Carriage by Air (Warsaw, 12 October 1929) (Warsaw Convention)�� 38, 46–47, 50, 59 Art 1(3)�� 26 Art 13(3)�� 51 Art 18 Art 18(1)�� 53 Art 18(2)�� 53 Art 18(3)�� 53 Art 28�� 51 Art 30 Art 30(3)�� 51 Art 31�� 49 Convention on Limitation of Liability for Maritime Claims (LLMC) 1976�� 9 Art 3(e)�� 22 Convention on Limitation of Liability for Maritime Claims (LLMC) 1996��9, 21, 24 Art 3(e)�� 22 Art 6�� 23 Art 7�� 23 Convention on the Contract for the International Carriage of Goods by Road 1956 (Geneva, 19 May 1956) (CMR)�� 41–42, 45–46, 54–56, 58–60 Art 1(1)��47–49 Art 2(1)�� 52 Art 4�� 50 Art 13�� 48 Art 13(1)�� 51 Art 17 Art 17(1)�� 49 Art 17(2)�� 57 Art 17(4)�� 57 Art 28(2)�� 51 Art 32��48, 51 Art 32(1)�� 50 Art 41�� 51 Convention on the Recognition and Enforcement of Foreign Judgments in Civil and Commercial Matters 2019�� 94

Table of Treaties and Conventions lvii Convention Supplementary to the Warsaw Convention for the Unification of Certain Rules relating to International Carriage by Air Performed by a Person other than the Contracting Carrier (18 September 1961) (Guadalajara Convention 1961) Art 1�� 26 EU Declaration to the Choice of Court Convention�� 105 Art 2 Art 2(a)�� 95 Art 2(b)�� 96 Art 2(d)(ii)�� 96 Art 2(d)(iv)�� 96 Art 2(d)(vi)�� 96 Hague Convention on Choice of Court Agreements 2005 (Choice of Court Convention)��91, 94 Art 2 Art 2(2)(f)�� 95 Art 2(2)(g)�� 95 Art 3�� 95 Art 3(a)�� 95 Art 3(b)�� 95 Art 3(c)�� 95 Art 5�� 95 Art 5(1)�� 95 Art 6��95, 99 Art 8�� 95 Art 16�� 95 Art 17�� 95 Art 19�� 95 Art 20�� 95 Art 21�� 95 Art 22�� 95 Hague Rules as Amended by the Brussels Protocol 1968 (Hague–Visby Rules)��28, 31, 34, 36, 38, 41, 46, 49, 52, 55–56, 58–59 Art I(b)�� 47 Art III.8��32, 50, 57 Art IVbis2��25, 51 Art IV(2)(a)–(q)�� 29 International Convention for the Unification of Certain Rules of Law Relating to Bills of Lading 1924 (Hague Rules)��28, 34–37, 46, 51–52, 59, 260–61 Art I(b)�� 47 Art III.8�� 31–32, 50 Art IV Art IV(2)(a)��49, 56 Art IV(2)(b)��49, 56 Art IV(2)(j)�� 57 Art IV(2)(a)–(q)�� 29

lviii Table of Treaties and Conventions International Convention on Civil Liability for Bunker Oil Pollution Damage 2001��22, 103 Art 7�� 93 Art 7.10�� 93 International Convention on Civil Liability for Oil Pollution Damage 1969�� 93 International Convention on Civil Liability for Oil Pollution Damage 1992��22, 103 Art V�� 93 Art VII.8�� 93 International Convention on Liability and Compensation for Damage in Connection with the Carriage of Hazardous and Noxious Substances by Sea 1996, as amended by 2010 Protocol (HNS Convention 2010)��22, 93 Art 1(5)(iv)�� 24 Art 4bis(10)�� 23 Art 4(4)�� 24 Art 7(2)(a)�� 24 Art 9�� 24 Art 14�� 24 Art 14(2)(a)�� 24 International Convention on Tonnage Measurement of Ships (23 June 1969) (Tonnage Convention 1969)�� 23 International Convention on the Establishment of an International Fund for Oil Pollution Damage 1992�� 22 International Convention Relating to the Arrest of Seagoing Ships, 10 May 1952, 439 UNTS 6330 (1962) (Arrest Convention 1952)�� 5 Art 1(1)�� 119 International Convention relating to the Limitation of the Liability of Owners of Sea-Going Ships, and Protocol of Signature (Brussels, 10 October 1957) (Limitation Convention 1957)�� 9, 21 Maritime Labour Convention 2006�� 22 Nairobi International Convention on the Removal of Wrecks (Nairobi, 18 May 2007) (Nairobi Wreck Removal Convention 2007)�� 22 New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards 1958 (New York Convention)�� 94 United Nations Convention on Contracts for the International Carriage of Goods Wholly or Partly by Sea 2008 (Rotterdam Rules) (not in force)�� 25–26, 37, 39, 43, 57 Art 18�� 27 Art 19�� 27 Art 20�� 27 Art 21�� 38 Art 26�� 60 Art 57�� 55 Art 60�� 38 Art 82�� 60 United Nations Convention on Contracts for the International Sale of Goods 1980 (Vienna Convention)�� 192 Art 7�� 196 Art 29(2)�� 191 Art 60�� 151

Table of Treaties and Conventions lix United Nations Convention on International Multimodal Transport of Goods 1980 (MT Convention)�� 43, 55–56 Art 4(1)�� 54 Art 15�� 54 Art 16(1)�� 54 United Nations Convention on the Law of the Sea (UNCLOS), 10 December 1982, 1833 UNTS 3163�� 113 Art 91(1)�� 110 United Nations International Convention on the Carriage of Goods by Sea 1978 (Hamburg Rules)��25, 30, 37, 39, 49, 54–55, 105 Art 1�� 26 Art 1(3)�� 26 Art 1(6)��47, 50 Art 2(3)�� 50 Art 5(2)�� 38 Art 6.1(b)�� 38 Art 7(2)�� 51 Art 10�� 27 Art 10(2)�� 26 Art 10(4)�� 27 Art 21��38, 51 Art 23�� 38 Art 23(1)�� 51 Art 23(2)�� 51

lx

1 The Empress of Ireland and Mont Blanc Collisions: Then and Now NICHOLAS GASKELL*

This is the story of two collisions that occurred in 1914 and 1917, and yielded cases that ended up in the Privy Council. They are fascinating in themselves to any reader about maritime disasters – involving as they do a tragic loss of life, incompetence and questionable political influences. They also illustrate the difficulties faced by maritime lawyers a century ago, some of which are long forgotten and others which still confront us today. The legal assumptions that would have been self-evident to practising maritime lawyers then may now seem puzzling to a modern reader. It is therefore interesting to examine some of the legal fallout from these collisions to illustrate principles that have shaped maritime law, but which have undergone significant change. The two particular examples happen to have a connection with Eastern Canada, but still have a resonance for UK maritime lawyers. It is impossible to set the scene in 1914, though, without mentioning the sinking of the Titanic in 1912.1

I. The Titanic and Limitation of Liability The Titanic is still the most famous maritime disaster of all time, and her story has been repeated in many books and films.2 Some 1,517 souls were killed, 832 passengers and 685 crew; this represented 63 per cent of passengers being carried, but 73 per cent of the crew.3 The sinking also influenced maritime law, mainly through changes to safety regulations, eg as to the number of

* Francis Rose and I have a shared interest in maritime law and legal history. Our grandfathers apparently served together on (the first) HMS King George V during the First World War. This chapter is based on a presentation made on ‘Wrecks, Admiralty Law and Eastern Canada’ given to the Eastern Canadian Admiralty Law Association on 10 September 2018 during a research visit to the Schulich Law School, Dalhousie University, Halifax (Nova Scotia). I am grateful to Professor Aldo Chircop for his support of my visit and hope that he and his colleagues will forgive any errors about Canadian law and history. 1 The Titanic also has a major connection to Eastern Canada as many of the dead were brought to Halifax, as evidenced by the poignant headstones in the city’s cemetery: see also J Boileau, Halifax and Titanic (Halifax, Nimbus, 2012). 2 There is a vast literature, but see, eg www.nationalarchives.gov.uk/titanic/; see also Report of the Formal Investigation into the Loss of the Steamship Titanic (HMSO, 30 July 1912). ‘Titantic’ Disaster – Report of the Committee on Commerce US Senate (62nd Congress, Report No 806) presented a different perspective: see www.titanicinquiry.org/USInq/USReport/ AmInqRep01.php. 3 The crew loss was particularly devastating for her home port of Southampton.

2 Nicholas Gaskell lifeboats to be carried.4 There were also civil claims brought by the passengers, many of which were litigated in the USA. This was hardly surprising, given the number of wealthy US p assengers on board, but although the shipowner asserted that it had contractual defences to cargo and passenger claims, it in fact successfully petitioned to limit its liability under US law.5 There were 666 (mostly personal) claims submitted in the US proceedings and these amounted to US$16,925,687,6 but the shipowner was entitled to limit liability to only US$97,772 (plus interest of $119,525).7 This was because, at that time, the USA operated a limitation system based on the value of the ship, but calculated after the casualty. Although such a system represented in some measure the shipowner’s investment in the maritime adventure, the US Titanic litigation showed its drawbacks as the only property saved consisted of 14 lifeboats and some advance freight.8 The British Merchant Shipping Act 1854, section 504,9 had also based limitation on the value of ship and freight, but the MSA Amendment Act 186210 adopted a new approach of fixing limits based on the tonnage of ships. The limits were set at £8 per ton for property claims11 or £15 per ton for personal claims.12 In fact, it seems that these figures were set at the then average values of general cargo ships (£8 per ton) or passenger ships (£15 per ton).13 Had the Titanic claims been subject to UK law, the shipowner would have been entitled to limit under the Merchant Shipping Act 1894, section 503, which re-enacted in substance the 1862 tonnage approach. Tonnage measurement for registration purposes is of volume, not weight, with gross tonnages generally reflecting the internal capacity of a ship and net tonnages representing the amount of ‘earning space’ – that is, after deducting space used, eg for accommodation, ballast, pumps, storage and propelling machinery. The Merchant Shipping Act 1894, section 503(2)14 adopted the wording for limitation of ‘registered tonnage with the addition of any engine room space deducted for the purpose of ascertaining that tonnage’.15 In effect, the figure for the engine room deduction appearing on the tonnage certificate was added to the net registered tonnage (nrt). It is not possible now to calculate the limitation tonnage of ships from a century ago without seeing their tonnage certificates, because the deduction could vary for each ship,16 but a rule 4 Under Board of Trade Rules made under the Merchant Shipping Act 1894, s 429, ships over 10,000 gross registered tons only needed to carry 16 lifeboats, as ships of the Titanic’s size had not really been contemplated. She was able to carry over 3,000 passengers and crew, but her lifeboat capacity (including four collapsibles) was only 1,178. On the fateful voyage, there were 2,201 persons on board. 5 See N Gaskell, ‘The Amount of Limitation’ in N Gaskell (ed), Limitation of Shipowners’ Liability: The New Law (London, Sweet & Maxwell, 1986) 34. 6 See Final Decree of limitation of Mayer DJ, 28 July 1916, www.titanicinquiry.org/lol/finaldecree.php. 7 ibid. White Star Line’s original petition for limitation was for US$91,806, using an exchange rate of US$4.8665 to £1: www.titanicinquiry.org/lol/wslpetition.php. The US courts rejected the argument that there was unlimited liability for foreign ships and applied US limitation law: see Oceanic Steam Navigation Co v Mellor 209 F 501 (1913, Holt DJ), affirmed 233 US 718 (1914). 8 The USA did provide a minimum limitation tonnage for personal claims in 46 United States Code § l84(b), but only after the later disaster involving The Morro Castle in 1934. There, the owners received US$2.1 million from the hull insurers, but were obliged only to establish a limitation fund of US$20,000 for all the claims against the vessel: see J Donovan, ‘The Origins and Development of Limitation of a Shipowner’s Liability’ (1979) 53 Tulane Law Review 999, 1031; A Rein, ‘International Variations on the Concept of Limitation of Liability’ (1979) 53 Tulane Law Review 1059. 9 For the history of the provision, see Cail v Papayanni (The Amalia) (1863) 1 Moo PC NS 471, 482–83; 15 ER 778, 782. 10 Merchant Shipping Act Amendment Act 1862, s 54. 11 ie ‘loss or damage to ships, goods, merchandise, or other things’. 12 ie ‘loss of life or personal injury’. 13 See Gaskell, ‘Amount of Limitation’ (n 5) 34, fn 8. 14 As amended by the Merchant Shipping Act 1906, s 69. 15 For the evolution of the concept of limitation tonnage, see Gaskell, ‘Amount of Limitation’ (n 5) 45–49. 16 Large tankers could have a much smaller, and small tugs a much bigger, percentage deduction: see Gaskell, ‘Amount of Limitation’ (n 5) 46–47; see further the updated Table 4 in N Gaskell, ‘Limitation of Liability and Division of Loss in Operation’ in S Gault and S Hazelwood (gen eds), Marsden and Gault on Collisions at Sea, 14th edn (London, Sweet & Maxwell, 2016) 985–86.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 3 of thumb was to assume a figure of 32 per cent of gross registered tonnage (grt) to be added to the nrt.17 Under the British tonnage limitation system, the Titanic’s limit can therefore be roughly estimated at about £549,839 (US$2,675,793).18 In fact, the claimants were later allowed to withdraw their claims against the US fund so as to be able to sue in the UK,19 and it appears that the claims were settled out of court.20

II. Empress of Ireland The British registered RMS Empress of Ireland21 had been launched from the Fairfield yard in Govan in 1906 and was the pride of the Canadian Pacific Railway (CPR) Line. On 29 May 1914, while on a voyage from Quebec to Liverpool with 1,057 passengers and 420 crew, she was sunk in the St Lawrence Seaway after a collision at 02:06 in fog with the Norwegian registered collier Storstad. The Storstad was under charter to the Dominion Coal Company and was inbound to Montreal on a voyage from Sydney, Nova Scotia, with a cargo of up to 11,000 tons of coal. The RMS Empress of Ireland has been called the ‘forgotten Empress’,22 because her loss on 29 May 1914 has received relatively little international attention by comparison with that of the Titanic (in 1912) or of the Lusitania (torpedoed on 7 May 1915). Part of the reason for this is probably the outbreak of the First World War shortly after the sinking, yet more passengers (840)23 were lost on the Empress of Ireland than on the Titanic (832)24 or Lusitania (791).25

A. Formal Investigation One unifying factor between all three of these famous sinkings was that formal investigations into them were presided over by Lord Mersey,26 not entirely without controversy at the time27 17 See also J Bes, Chartering and Shipping Terms, 10th edn (London, Baker & Howard, 1977) 219. 18 Based on a grt of 46,328, nrt of 21,831, with an estimated propelling power allowance of 32%, giving a limitation tonnage of about 36,656 limitation tons. The limit would then be £549,839 (36,656 × £15), converted into US dollars at 4.8665 (as used by the White Star line: see n 7 above). From information available to the author, it appears that the actual propelling power allowance for the Titanic was between 25% and 28%; on this basis, the actual limit would therefore have been lower than the estimate given. 19 The Titanic: In re Oceanic Steam Nav Co Ltd 225 F 747 (2 Cir 1915). 20 See, eg J Force and R Zapf, Benedict on Admiralty, 7th edn (rev) (New York, Matthew Bender, 2013) §5, 1–25. 21 14,191 grt, 8,028 nrt. 22 See D Zeni, Forgotten Empress: the Empress of Ireland Story (Tiverton, Halsgrove, 1998) 7. In fact, there have been many who have been keen to keep alive interest in the disaster: see, eg ‘The RMS Empress of Ireland Community’, www. theempressofireland.com. 23 Comprising 51% of first-class passengers, 81% of second-class passengers and 82% of third-class passengers: see Zeni (n 22) 165; Report of Commission of Inquiry into the Casualty to the British Steamship ‘Empress of Ireland’ (Cmnd 7609, 1914) 25. 24 The total loss of life on the Titanic was greater owing to the heavy loss of life amongst the crew. On the Empress of Ireland, 172 crew were lost; some 60% were saved. 25 In total, 1,198 lives, including crew, were lost from Lusitania. There have long been conspiracy theories about whether the ship, now admitted to be carrying munitions, was deliberately exposed to U-boat attack as part of the desire to bring the USA into the war: see, eg C Simpson, Lusitania (Harmondsworth, Penguin, 1972). Litigation in the Admiralty Court as late as the 1980s concerned rights to items recovered from the wreck: see Pierce v Bemis (The Lusitania) [1986] QB 384 (QB (Admlty)). 26 John Charles Bigham, 1st Baron Mersey (1840–1929), President of the Probate, Divorce and Admiralty Division 1909–10 (when he apparently retired through ill health): AWB Simpson, Biographical Dictionary of the Common Law (London, Butterworths, 1984) 53. 27 The ‘literature of maritime disasters has raised some doubts as to Bigham’s manners, competence and even integrity’: Simpson (n 26) 53. Bigham’s experience was more in commercial than Admiralty cases, which would have made him less

4 Nicholas Gaskell and later.28 The investigation into the loss of the Empress of Ireland was convened by the Canadian Minister of Marine and Fisheries and took place under the Canada Shipping Act 1906, Part X.29 The vessels had sighted each other in the early morning from around four miles away, but their stories thereafter differed diametrically, both as to their respective courses and the action taken by each when the fog was sighted. The Investigation Report concluded that the Empress of Ireland’s account of her original course was probably right, but that on either vessel’s version there would have been no collision had the fog not intervened – that is, if they had kept their courses, they would have passed safely and they would not be ‘crossing so as to involve risk of collision’ within the Collision Regulations.30 Thereafter, when the lights of the Storstad began to grow dim in the fog, the Empress of Ireland went full astern. The Commissioners considered that the master might have been better advised to steer away from the Storstad, but surprisingly did not consider that stopping was an unseamanlike act, nor that failure to give a wider berth was a contributory cause of the disaster. By contrast, they found that the Storstad had altered course to starboard, believing that the ships would pass port to port (red to red), whereas the Empress of Ireland was on a course to pass starboard to starboard (green to green).31 This led the Storstad to collide directly with the starboard side of the Empress of Ireland – causing what was a fatal blow, as a result of which the Empress of Ireland foundered and sank within about 15 minutes. The officer of the watch on Storstad was also criticised for not calling his master to the bridge as soon as the fog was seen. The finding that exonerated the British ship in respect of its navigation was not without controversy. There was some evidence that there might have been steering problems with the Empress of Ireland,32 including testimony from her own quartermaster, James Galway, who had been on duty until a couple of hours before the collision. There were allegations at the time, which have continued, that the evidence of the Empress of Ireland had been manufactured and that CPR had attempted to get Galway out of the country so that he could not testify at the enquiry.33 The Commissioners preferred the evidence of the pilot and officers of the Empress of Ireland, and allegations about physical defects are always difficult to establish when the ship has sunk.34 With hindsight, it is possible to wonder whether Lord Mersey was unduly concerned to preserve the reputation of a major British shipowner35 and less inclined to believe the testimony of a Liverpudlian quartermaster (with an Irish name) over the ships’ officers.36

suitable to preside over detailed navigational evidence more often heard in a contested collision action: see Zeni (n 22) 148. By contrast, the Titanic was a single ship casualty, perhaps raising more general issues of evidence. 28 It was claimed in 2015 that in the Titanic Report, Bigham sought to protect fellow Freemasons, such as the President of the Board of Trade and Lord Pirrie Chairman of Harland and Wolff and Director of White Star’s parent company: J Bingham and V Ward, ‘Was the Titanic Inquiry Scuppered by the Freemasons?’ (The Telegraph, 23 November 2015), www.telegraph.co.uk/news/uknews/12010573/Was-Titanic-inquiry-scuppered-by-the-Freemasons.html. 29 Which was based on the Merchant Shipping Act 1894, pt VI, s 466. The Canada Shipping Act 1906, s 781, was amended retrospectively, by an amendment assented to on 12 June 1914 (1914 c 49, s 1), to elaborate upon the obligations of wreck commissioners in particularly grave cases, eg to make a full report to the Minister. 30 See Report of Commission of Inquiry (n 23) 12. 31 Simplified diagrams appear in Zeni (n 22) 150. 32 Report of Commission of Inquiry (n 23) 13. 33 See, eg Zeni (n 22) 151; I Kinder, The 96th Voyage: The Truth about the Empress of Ireland Tragedy (Stoney Creek, I Kinder, 2017). 34 Diving examinations immediately after the sinking had been limited, and an examination of the damage might have been hazardous and perhaps impossible: Report of Commission of Inquiry (n 23) 16. 35 As with the White Star Line (Titanic) and Cunard (Lusitania). 36 Zeni (n 22) 164, considers that the handling by Lord Mersey of Captain Anderson of the Empress of Ireland was deferential and that there was harsher treatment of witnesses such as Chief Officer Toftenes of the Storstadt.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 5

B. Civil Claims Civil claims arising from the sinking of the Empress of Ireland amounted to Can$3,069,484. The majority of these claims were for property loss (eg Can$2 million for the value of the Empress of Ireland), with Can$469,468 in claims for loss of life.37 Unlike modern cruise ships, the Empress of Ireland was also carrying a significant amount of cargo, including mail, passengers’ valuables (locked in the purser’s safe) and 212 bars of silver valued at Can$1,099,000.38 By August 1914, all of the silver was recovered by a US–Canadian salvage team, along with 318 bags of mail.39 Exaggerated claims for loss of belongings were exposed when the purser’s safe was found to contain mostly documents and jewellery – only 0.5 per cent of what had been claimed.40 Both property and personal claimants faced a number of immediate tactical and legal difficulties after the casualty. The exoneration of the Empress of Ireland in the Investigation Report meant that there would be difficulties in proceeding in contract or tort against CPR, especially if those findings were upheld in a civil trial. Even if the Empress of Ireland were partly at fault, the contracts for the carriage of goods41 or passengers might well have contained exclusions for negligence. The personal claimants would clearly have had an action in tort for negligence, but which vessel was negligent, given the vastly different stories? An action in rem against the Empress of Ireland was not possible as she had sunk,42 but it is assumed that an in personam action could have been brought against CPR and that it was likely to be a solvent defendant. The difficulties faced by the personal claimants in suing the Storstad are of interest in providing the background for modern approaches to similar problems. The position of passengers and crew could differ.

C. The ‘Identification’ Issue for Personal Claims and Property Claims By the twentieth century, Lord Campbell’s Act43 had long since dispelled the notion that a person’s claim died with them: as a result of that Act, compensation claims could be made in the event of a wrongful death for the financial losses suffered by persons (such as bereaved families) who were dependent on the deceased. Passengers, though, could face a particular problem in that the contract of carriage could exempt the carrier from negligence of the pilot, master or mariners. Thus, in Haigh v Royal Mail Steam Packet Co Ltd,44 the Court of Appeal held that a passenger killed by the negligent navigation of the crew of the carrying ship could not succeed in a claim under Lord Campbell’s Act because of the exclusion clause in the contract. 37 Canadian Pacific Railway Co v Owners of the Steamship ‘Storstad’ [1920] AC 397 (PC). 38 Zeni (n 22) 169. 39 ibid 172. 40 ibid 174. 41 This might not always be so. In The Bernina (No 2) (1886) 12 PD 58 (CA) (discussed in sections II.C and II.D below in relation to personal claims), the owners of the Bernina were liable to the cargo owners for the safe carriage of the iron ore in the ship, as there was no exception for negligent navigation in the bill of lading and charterparty. 42 The concept of an action in rem against a sister ship owned by CPR did not exist and was only introduced into English law when the Arrest Convention 1952 was given effect by the Merchant Shipping (Liability of Shipowners and Others) Act 1958; see now the Senior Courts Act 1981, s 21. 43 The Fatal Accidents Act 1846, originally introduced in response to railway fatalities. See now the Fatal Accidents Act 1976. 44 Haigh v Royal Mail Steam Packet Co Ltd (1883) 52 LJQB 640 (CA). See also Hood v Anchor Line (Henderson Brothers) Ltd [1918] AC 837 (HL).

6 Nicholas Gaskell This contractual bar could particularly affect passengers where there was a single ship casualty (eg the Titanic). By contrast, in collision cases, they might hope to bring a claim against the noncarrying ship for its negligence. Nevertheless, in the mid-nineteenth century, passengers could face a difficulty even in collision cases because of the application of the common law rule whereby contributory negligence was a complete bar to an action. In this context, there was uncertainty about two issues concerning passengers. The first was whether any negligence could be attributed to passengers when they were on board a ship which was itself adjudged to have been partly responsible, through its crew, for a collision. The second was whether there was a difference between the common law and Admiralty rules generally in relation to contributory fault. The second issue was settled by Dr Lushington in The Milan,45 who stated that the Admiralty rule was that where both ships were to blame, each was liable for half (‘a moiety’) of the damage.46 The Milan particularly concerned the position of shipowners in relation, for example, to claims for collision damage to each ship. However, Lushington also applied the Admiralty rule to owners of cargo on board a ship, so that they could not claim from the other ship any more than half their loss.47 In this sense, the cargo was ‘identified’ with the carrying ship’s negligence, even though the cargo owners themselves could not have been at fault for the collision at all.48 What was not clear until the 1880s was the first issue – ie whether such an ‘identification’ rule also applied to personal injury and death claims, so that crew or passengers were also restricted in the amounts they could claim for the non-carrying colliding ship.49 This first issue was finally resolved in The Bernina (No 2),50 involving a collision between the Bernina and the Bushire which resulted in loss of life of a passenger and crew on Bushire. At first instance,51 Butt J had felt obliged to apply Thorogood v Bryan.52 In that case a passenger was knocked over and killed by the defendant’s horse-drawn omnibus while alighting from another in which he was a passenger. A claim for negligence against the defendant failed before a jury, possibly because of the passenger’s own negligence in getting off too early while his carriage was in motion and not at the kerb. In an application for a new trial on the basis of a misdirection, the court expressed itself in terms suggesting that even if the passenger had not been personally negligent, he was identified with the faults of his own carrier. In a ship collision case argued at the same time, the court appeared willing to apply the same approach to a maritime passenger,

45 The Milan (1861) Lush 388, 389; 167 ER 167, 168. 46 Equal responsibility did not result in no damages being payable, as each ship may have suffered a significantly different loss. Thus, the Empress of Ireland was a total loss after the collision, while the Storstad had relatively minor damage to her (ice-reinforced) bow. If both were at fault, then the latter would have had to pay for half of the former’s loss and so would have been a net payer. 47 The claim for the other half against the carrying ship may have been met by bill of lading terms that excluded liability for negligence. 48 The cargo rule had been approved in Owners of Cargo of the Tongariro v Owners of the Drumlanrig (The Drumlanrig) [1911] AC 16 (HL) and recognised in the Maritime Conventions Act 1911, s 1(1). It persists to this day in the Merchant Shipping Act 1995, s 187(1). Roscoe deprecated this identification with the exact faults of the shipowner: see E Roscoe and H Robertson, The Maritime Conventions Act 1911 (London, Stevens & Sons, 1912) 7, although it is difficult to see that this was not also the effect of the Admiralty law and practice applied by Dr Lushington. US law did not recognise the rule, treating cargo as innocent, and this difference is still reflected today in the use of a ‘both to blame collision’ clause in charters and bills. 49 If there was ‘identified’ negligence impugned to the personal claimants, then a subsidiary question would have arisen under Lord Campbell’s Act: whether to apply the common law contributory negligence rule (a complete bar to the claim) or the Admiralty rule (claim limited to a moiety). 50 Mills v Armstrong (The Bernina) (No 2) (1888) 13 App Cas 1 (HL). 51 Mills v Armstrong (The Bernina) (No 2) (1886) 11 PD 31 (PDAD). 52 Thorogood v Bryan (1849) 8 CB 115, 137 ER 452.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 7 but the matter was settled before final judgment.53 In the Court of Appeal in The Bernina (No 2),54 Lord Esher MR provided a withering analysis of Thorogood v Bryan55 before o verruling it.56 That decision was in turn upheld in the House of Lords,57 which recognised that it was absurd to identify the passengers with the faults of the carrying ship, as they were in no sense an employer. The result was that a passenger (Toeg) on the Bushire could claim against the Bernina (the non-carrying ship) for the whole amount of his claim.58 The upshot is that, by the time of the Empress of Ireland sinking, it seems clear that if both she and the Storstad were negligent, the passengers on the Empress of Ireland could have sued either ship as joint tortfeasors at common law under Lord Campbell’s Act. Yet in the Privy Council, Lord Sumner made the somewhat Delphic statement that the personal claimants were ‘in a position of some embarrassment, for the Maritime Conventions Act [1911] does not apply in Canada’.59 It seems that what he meant by that was a reference to section 5 of the 1911 Act, which conferred jurisdiction on Admiralty courts to hear death and injury cases in an action in rem or in personam.60 This section improved the position of claimants under Lord Campbell’s Act, who could not previously enforce their claims in an action in rem.61 In any event, CPR started an action in rem against the Storstad.

D. Crew Claims Times for seafarers were hard, as their employment contracts ceased with the sinking of the ship on the basis that freight was the ‘mother of wages’. The Merchant Shipping Act 1906 had imposed some obligations on shipowners in relation to repatriation, but employment contracts did not normally extend beyond the voyage. Crew members on a ship such as the Empress of Ireland

53 Cattlin v Hills (1849) 8 CB 123, 137 ER 455. 54 The Bernina (No 2) (CA) (n 41). 55 ibid 63–83. His judgment contains the delightful supporting argument (ibid 70–71) that he had ‘been allowed by the kindness of his grandson to see the volume in which Thorogood v Bryan is reported, in Baron Parke’s own library; and I have seen in his own handwriting, “quaere” written against the case’. Such references were redolent of an era when personal law reporting was still significant to judges and perhaps before the more reliable system of law reporting of the ICLR (post-1865) had become embedded in the profession. 56 He deliberately left open (ibid 83) whether the effect of the Judicature Act 1873, s 25(9), in giving priority to the Admiralty rules ‘in any cause or proceeding for damages arising out of a collision between two ships’ where both ships were at fault, extended not only to the claims of the two ships, but also to the position of cargo, as was held to be the Admiralty practice in The Milan (n 45). The Maritime Conventions Act 1911, s 1 effectively settled that issue in favour of the Admiralty rule, but using proportionate fault rather than the 50/50 Admiralty rule. 57 The Bernina (No 2) (HL) (n 50). Lord Bramwell, in a concurring speech (ibid 11–12), considered that Thorogood v Bryan (n 52) was correctly decided on a pleading point. Bramwell acknowledged his own experience of special pleading, brought in by the Hilary Term Rules of 1834, but ‘happily abolished’ (in part by the Common Law Procedure Act 1854, but eventually by the Judicature Acts 1973–75). The case illustrates the difficulties for modern readers of mid-nineteenthcentury cases in distinguishing issues of pleading from those of legal principle. Bramwell was ultimately driven to accept that, although the four judges that decided Thorogood v Bryan may have been right on pleading, their expression of legal principle was wrong. 58 For the position of crew claims, see section II.D below. 59 Canadian Pacific Railway Co (PC) (n 37) 398. 60 The collision had involved a British and a Norwegian ship over three miles from the coast, ie on the (then) high seas, outside Canadian jurisdiction (today the territorial limit would be 12 miles). The Canadian Admiralty Act 1891 established the Exchequer Court of Canada to operate as a Court of Admiralty, having and exercising all the jurisdiction, powers and authority conferred by the Colonial Courts of Admiralty Act 1890 (Imp). The application of colonial maritime legislation operated as a major complication in the maritime laws of many emerging nations, eg in Australia until the Admiralty Act 1988 (Cth). 61 See Roscoe and Robertson (n 48) 18.

8 Nicholas Gaskell would have faced a particular difficulty in suing their employer in tort because the common law defence of common employment would have barred such a claim if there had been negligence of a fellow crew member.62 A common law suit against the other colliding ship might meanwhile have faced a defence of contributory negligence. As previously explained, The Bernina (No 2) rejected the notion of identifying a personal claimant (including a crew member) with the faults of the carrying ship – this enabled the Bushire’s chief engineer (Armstrong) to make a claim. If, however, the crew member himself was at fault, then contributory negligence could operate as a complete bar. In The Bernina (No 2), the second officer of the Bushire (Owen) had been directly responsible for the negligence causing the collision with the Bernina, and the appeal by his administratrix was abandoned by his counsel before the Court of Appeal.63 Both Butt J and the Court of Appeal concluded that Lord Campbell’s Act did not apply to Admiralty actions, and this view was endorsed in the House of Lords. This meant that the common law rule of contributory negligence applied to provide a complete bar to the action by the officer’s personal representatives, who could not even claim for half the loss (under the then Admiralty rule of divided damages).64

E. The Collision Liability Litigation The collision liability action between the Empress of Ireland and the Storstad was heard by Dunlop J in the Exchequer Court of Canada.65 Evidence from the Formal Investigation was read and used by consent.66 The Storstad did not plead the alleged steering defects on the Empress of Ireland and, not surprisingly, the judge found there to be no evidence to support it. He found that Toftenes, the navigating officer of the Storstad, was wholly to blame in altering course in fog and failing to call the master sooner. There was no appeal from this decision.67 Although the Canadian court endorsed the finding that the Storstad was solely to blame, it is of interest that a ‘Norwegian Inquiry, conducted at the Norwegian Consulate General in Montreal, ultimately exonerated the Storstad and its captain, Thomas Anderson. To this day, the two conclusions are irreconcilable.’68

F. Enforcing the Claims against the Storstad Once it was clear that the Empress of Ireland was not at fault in navigation, and the C ommissioners had dismissed evidence about possible unseaworthiness in relation to the steering, claims fell to be enforced against the Storstad. The problem was that although the Storstad was 100 per cent

62 See, eg The Napier Star [1939] P 330 (PDAD), where the personal representatives of crew killed on the White Star liner Laurentic successfully sued the colliding ship, Napier Star, which had been held to be equally to blame with the Laurentic. Langton J regarded it as ‘difficult to imagine a better or clearer case of agreed acceptance of risk [of a fellow servant’s negligence] than in the ordinary case of a ship’s company’, eg in navigation: ibid 341. This was presumably why the Laurentic’s crew’s claims had been brought against the Napier Star in the first place. The disadvantage of suing the ‘other’ ship was that it would have been able to limit its liability for such crew claims. 63 The Bernina (No 2) (CA) (n 41) 59. 64 The doctrine of common employment did not prevent the claim by Armstrong against the Bernina for that ship’s negligence, although it would have done in a claim against the owners of the Bushire, given the negligence of Owen. 65 Canadian Pacific Railway v The ‘Storstad’ (1915) 40 DLR 600 (Exch). 66 In the Halifax litigation, this approach was criticised in the Privy Council: see section III.C below. 67 There were limitation proceedings up to the Privy Council: see section II.F below. 68 See M Keich, ‘The Collision between the SS Empress of Ireland and the SS Storstad’ (February 2005), www. norwayheritage.com/articles/templates/great-disasters.asp?articleid=99&zoneid=1.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 9 at fault, the claimants faced the difficulty of enforcing any judgment against her owners. There were two possible enforcement difficulties at the time that are still apparent today. The first is the ability of the shipowner of the Storstad to limit liability by statute. The second is the broader question of how to enforce any claim when the negligent ship, in this case the Storstad, is owned by a single ship company whose only asset is the ship itself. CPR’s action in rem against the ship gave CPR the right to arrest the ship and ultimately to have her sold to meet the claims. The legal owners of the Storstad, Aktiesselskabet Maritim, were entitled to limit liability under the Merchant Shipping Act 1894, section 503, to £15 per limitation ton for personal injury and death claims, and £8 per limitation ton for property claims (including collision damage and cargo claims).69 Given that the Storstad was 6,028 grt and 3,561 nrt, an estimate of her limitation tonnage would be about 5,490 limitation tons.70 This would have given a total limit of £82,350 sterling,71 which can be roughly estimated to be Can$400,221 at the date of the collision,72 with no more than Can$210,675 of that being available for property claims.73 The owners of the Storstad could have availed themselves of this defence by establishing a limitation fund, given that the total claims were in excess of Can$3 million, but they chose another course of action. They did not appeal the judgment against the ship in the action in rem, but merely allowed the ship to be sold by court order. The Storstad fetched only Can$175,000. The question was who was entitled to this money? The property interests, mainly CPR and presumably its hull and machinery underwriters, argued that as this sum represented the only asset available to claimants, it should be divided pro rata amongst all claimants (property and personal). In contrast, the personal claimants, who now intervened, argued that it should be distributed in the same manner as a limitation fund under the Merchant Shipping Act 1894, section 503, so as to give some priority to them. Following the liability ruling, in 1916 a deputy registrar ‘collocated the balance [ie the Can$175,000] pro rata in favour of the life claims so far as such funds were sufficient, and excluded all other claimants from participation in the collocation’.74 This was purportedly to give effect to the Merchant Shipping Act 1894, section 503, under which, claimants for loss of life have an absolute privilege and priority over claimants for loss of property or goods to the extent of an amount equal to £7 per ton of the ship held to have been at fault.75

As it was admitted that the £7 per ton figure76 already exceeded the Can$175,000 in court, the personal claimants were entitled to share in all of the latter – to the detriment of CPR in particular. This approach was upheld on appeal to the Exchequer Court.77 In the Supreme Court of Canada,

69 See section I above. 70 (6,028 grt × 32% = 1,929) + (3,561 nrt) = 5,490 limitation tons. See the tonnage explanation in section I and n 18 above. 71 ie at £15 per ton for injury and death claims (whether or not there were also property claims). 72 Until August 1914, Canada adopted the gold standard, and 1 sovereign=Can$4.8666. Until Canada returned to the gold standard in 1926, there were fluctuations caused by inflation, so the calculation is an estimate: see J Powell, ‘A History of the Canadian Dollar’ (Bank of Canada, undated) 18–24, www.publications.gc.ca/collections/Collection/FB2-14-1999E. pdf. 73 Had there been property claims alone, the limit would have been £43,290: ie at £8 per ton. In effect, in a missed disaster, as with the Empress of Ireland, the personal claimants would have sole access to the top £7 per ton (ie £38,430, or Can$187,023) and would share rateably with the property claimants for the balance of £8 per ton: see The Victoria (1888) 13 PD 125 (PDAD). This English preferential system was replicated in the 1957 Limitation Convention and the Conventions on Limitation of Liability for Maritime Claims (LLMC) 1976 and 1996. 74 See Canadian Pacific Railway Co v Owners of the Steamship ‘Storstad’ (1917) 34 DLR 1 (Exch) [2]. 75 ibid. 76 ie the part reserved entirely for the personal claimants. 77 Canadian Pacific Railway Co (Exch) (n 74).

10 Nicholas Gaskell the same basic approach was taken – that is, that the Merchant Shipping Act 1894, section 503, expressed a positive statutory preference in favour of the personal claimants. But the Supreme Court (by a majority) varied the order so that the personal claimants obtained 7/15ths of the Can$175,000 in court (about $81,000) and shared the balance (about $93,000) pari passu with the other property claimants (eg CPR).78 The Privy Council overruled this decision in a brutally short, two-page judgment.79 Limitation of liability was the creature of statute and only gave rights to claimants where a shipowner had elected to establish a limitation fund. Nothing in the MSA 1894 gave any more general rights to different classes of claimants. Had the Storstad been sold for a sum equalling all the claims, everyone would have been paid out in full, unless the shipowners had elected to limit. In fact, the personal claimants had intervened by consent. The Empress of Ireland litigation is a prime and early example of the limits of an action in rem against an owner with no assets other than the ship in the jurisdiction and of the effectiveness of the single shipowning structure in isolating shipowners from liabilities. As Lord Sumner indicated,80 the claimants could have elected to proceed in Norway in a liability action against the shipowning company in personam. Here, they may have found the Norwegian courts less willing to exonerate the Empress of Ireland, but still have been faced with a single ship company with no assets. Such difficult jurisdictional choices confront litigators today, where there may also be the added complication of deciding whether the in personam claim would operate as res judicata to prevent any subsequent action in rem against the vessel.81

G. What Happened to the Storstad There is an interesting coda to the litigation, concerning the fate of the Storstad. When the S torstad was sold at auction on 7 July 1914 for Can$175,000, the buyer was apparently the Prudential Trust Company, acting on behalf of undisclosed foreign clients but ‘believed to be the original owners of the steamer, the Maritime Steamship Co of Norway’.82 It seems that AF Klaveness & Co of Norway was the controlling line behind Aktiesselskabet Maritim,83 and was apparently one of Norway’s largest shipping companies after 1914 until its bankruptcy in 1983 and reformation in 2016.84 A history of Tyne shipbuilding records the Storstad’s ownership as ‘1911 A/S Maritim (AF Klaveness & Co), Christiania; 1915 Klaveness Dampsk A/S85 (AF Klaveness & Co), Christiania’.86 What this shows is that after the Storstad was bought back, the ship was again registered to what was presumably a single ship company controlled by Klaveness.

78 Canadian Pacific Railway Co v Owners of the Steamship ‘Storstad’ (1918) 56 SCR 324. 79 Canadian Pacific Railway Co (PC) (n 37). 80 ibid 398–99. 81 Republic of India v India Steamship Co Ltd (The Indian Endurance and The Indian Grace) (No 2) [1998] AC 878 (HL). Lord Steyn’s refusal to distinguish between an action in rem and an action in personam has been subject to trenchant (probably justified) criticism by Admiralty lawyers: eg Allsop J in Comandate Marine Corp v Pan Australia Shipping Pty Ltd (The Comandate) [2006] FCAFC 192, [2008] 1 Lloyd’s Rep 119 [99]–[130]. 82 See a local newspaper report in The Toronto World (8 July 1914) 6. Presumably the owner’s name was a contemporary translation of Aktiesselskabet Maritim. 83 This appears from historical lists, such as www.sjohistorie.no/en/rederi/111905/, and is assumed by I Kinder (16 March 2017), www.theempressofireland.com/s-s-storstad#!. By 1917, the time of her later sinking, she clearly had a white K on a black funnel, but the 1914 photos also seem to show a dark K on a two-tone funnel. 84 See www.en.wikipedia.org/wiki/A._F._Klaveness_%26_Co; www.klavenessandco.com/who-we-are. 85 The usual abbreviation of Aktiesselskabet. 86 See www.tynebuiltships.co.uk/S-Ships/storstad1911.html.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 11 The significance of all this is that in 1914 – and, indeed, for several decades before that – it was recognised that the use of a single ship company was an effective way to isolate the true owner from extensive liabilities, to the detriment of claimants, particularly personal claimants. To add insult to injury, the shipowner was able, in effect, to buy back the very ship that had caused the damage and continue to deploy her in its trade. The Storstad survived the disaster and, after repair, she continued in the local iron ore trade.87 Later she was employed by a charity, the US ‘Commission for the Relief of Belgium’, founded by Herbert Hoover in 1914 to provide relief from wartime food shortages.88 In 1917 she was engaged by the Belgian Relief Committee89 to carry grain from Buenos Aires to Rotterdam. Thomas Anderson was still her captain and had obtained some assurances from the German Consul at Buenos Aires that the ship would be given safe passage from submarines. Despite being a neutral ship with a neutral crew, and displaying the markings of a relief ship, the Storstad was torpedoed and sunk by a U-Boat on 8 March 1917. Her use as a Belgian relief ship provides a curious link to a second major collision in Eastern Canada, in Halifax (Nova Scotia) in December 1917.

III. The 1917 Halifax Collision and Explosion90 A. The Halifax Collision By 1917, Halifax, Nova Scotia was a vital and booming wartime port. Its Bedford Basin provided a safe anchorage for convoys carrying war cargoes bound for Europe. It was also an inspection point for incoming neutral ships. On 5 December 1917, the former whaler, Imo,91 had received clearance to sail in ballast for New York to pick up a relief cargo for the Belgian Relief Commission, which had also chartered the ship from a Norwegian company.92 Her voyage was delayed by the late arrival of her collier to deliver bunkers, so the departure was postponed to the next morning. On 5 December, another ship was hoping to enter Halifax. This was the Mont Blanc,93 a French cargo ship built in 1899 and, since 1915, owned by La Compagnie Générale Transatlantique (CGT). She was loaded in New York with a cargo of munitions for the French Government, consisting of 2,300 tons of wet and dry picric acid, 200 tons of TNT, 10 tons of gun cotton and 35 tons of benzol (along with 300 rounds of ammunition for use against U-boats). It appears that dry picric acid, in particular, was potentially unstable and sensitive to shocks. Mont Blanc was

87 See Kinder (n 83). 88 See www.baef.be/documents/about-us/history/the-commission-for-relief-in-belgium-1914-.xml?lang=en. The Commission apparently administered over US$900 million during the war, financed by the governments of the USA, UK, France and Belgium, along with significant voluntary contributions. In total, 5 million mt of provisions were distributed in Belgium and Northern France. 89 This is the name (in contemporary photos) painted on her hull to show neutrality; other ships simply displayed ‘Belgian Relief ’. 90 The literature on the Halifax disaster is copious. For some official records, see www.canada.ca/en/heritage-informationnetwork.html. See also www.bac-lac.gc.ca/eng/discover/politics-government/Pages/thematic-guides-halifax-explosion. aspx. 91 5,043 grt, 3,161 nrt. Her curious name obviously has no link to the International Maritime Organisation, established after the Second World War, although it is the Organisation that has addressed some of the risks caused by ships such as the Imo: see section IV.E below. 92 South Pacific Whaling Company Aktieselskab. There is no indication whether this was a single ship company. 93 3,121 grt, 2,252 nrt.

12 Nicholas Gaskell too slow to join a fast convoy from the USA and was sent to join a slower convoy from Halifax. Off Halifax, she embarked an experienced compulsory pilot, Mackey, but had to wait at the entrance to the port because the submarine nets had closed for the night. At about 07:50 on 6 December Mont Blanc left her anchorage to enter the port and then Bedford Basin, following a US tramp steamer, the SS Clara. Mont Blanc proceeded cautiously on the correct starboard side of the narrow channel, next to the town of Dartmouth (opposite Halifax). Imo left the Basin and headed for the channel. She should have stayed to the starboard side of the channel, but the incoming tramp Clara was herself wrongly in that water (apparently adopting a shorter and more convenient course for her pilot). Through signals, Imo agreed to pass this ship starboard to starboard (contrary to the normal rules of the road). This brought Imo more into the centre of the channel, and she veered even more to port to avoid a tug and tow that had also appeared on her starboard side, behind the Clara. From the perspective of Mackey and Mont Blanc’s master, Le Médec, it appeared that Imo was navigating as if to pass (wrongly) starboard to starboard. Mackey and Le Médec then made a desperate decision to turn to port, stopping Mont Blanc’s engines. Meanwhile the Imo had gone full astern, causing her bow to swing to starboard94 where it struck the starboard side of the Mont Blanc – despite the latter’s decision, 20–30 seconds before, to go full astern. A small fire was seen, after friction sparks apparently ignited benzol leaking from barrels on the deck. Le Médec, realising the danger from the cargo, and with no means to control the fire, ordered the ship to be abandoned and his crew headed in boats to the Dartmouth shore. Shouted warnings, in French, that there was a cargo of munitions were not heard. Mont Blanc drifted to the Halifax side, where the fireworks display attracted many onlookers. A local tug tried to take the ship in tow. There were many heroic acts, eg by a railway worker staying at his post to provide warnings to stop passenger trains entering the dockside station. But, at about 09:05, the Mont Blanc exploded, some 20 minutes after the collision.95 The ensuing explosion has been described as one of the worst man-made explosions in history until Hiroshima.96 It is hard now to comprehend the extent of the damage, but the effect on the city was devastating.97 The Mont Blanc disintegrated; parts of her heavy anchor were found over two miles away.98 The shock was felt in Cape Breton, 270 miles away. Some 386 acres of Halifax were totally destroyed, including wooden housing and factories, particularly in the R ichmond district next to the port. Casualty estimates vary, but it seems that some 1,600 were killed instantly and the final death toll may have been up to 2,000; 9,000 were injured (including through blindness caused by flying debris).99

94 A common result with single screw ships with a ‘right-handed’ propeller, especially for a ship in ballast. 95 Le Médec and Mackey were surprised that the ship did not explode immediately: see J Zemel, Scapegoat: the Extraordinary Legal Proceedings Following the 1917 Halifax Explosion (Halifax, SVP Productions, 2014) 60. 96 In December 1943, there was a massive explosion in Bari harbour of ammunition ships (and one containing mustard gas shells), but this was a result of German bombing. 97 An internet search will reveal photographs and maps showing the desolation, compounded by a heavy snow fall that enhanced the misery. Apart from contemporary accounts, there appears to have been relatively little writing about the disaster until the 1980s, with the publication of J Kitz, Shattered City: the Halifax Explosion and the Road to Recovery, 3rd edn (Halifax, Nimbus Publishing, 2008) (first published in 1989). Following this, and leading up to the centenary of the disaster, there have been many books, documentaries and dedicated websites, eg www.cbc.ca/archives/topic/ the-halifax-explosion. 98 Kitz (n 97) 25. 99 ibid 26.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 13

B. The Halifax Formal Investigation Being wartime, there were many conspiracy theories – including that the Germans must have been responsible, eg through sabotage – and Johansen, the Imo’s helmsman, was arrested for a while as a spy!100 There was also a familiar refrain from the newspapers at the time that ‘someone must be to blame’, and this increased pressure to find scapegoats.101 Only six days after the explosion, on 12 December, a Formal Investigation was opened, presided over by the local Admiralty judge, Drysdale J, assisted by two RCN naval assessors.102 The principal witnesses were the Mont Blanc’s master, Le Médec, and her pilot Mackey; the master and pilot of the Imo had been killed, and the only bridge survivor was Johansen. At the investigation (and thereafter) there was some reluctance to criticise the navigation of the dead men on the Imo, while there was aggressive questioning of those who had been on board the Mont Blanc103 and of the naval Chief Examining Officer (CXO), Wyatt.104 On 4 February 1918, a Report was published consisting of 13 short paragraphs.105 It found the pilot and master of Mont Blanc wholly responsible for violating the rules of the road,106 and they were also (wrongly) criticised for not giving warnings when leaving their vessel. Wyatt was censured for not ensuring the pilotage instructions were obeyed. Le Médec, Mackey and Wyatt were immediately arrested for manslaughter, but these charges were in effect dismissed on 15 March by another judge, Russell J, in terms that were diametrically opposed to the blame attributed in the Drysdale Report.107

C. The Halifax Collision Litigation We tend to think of the two world wars as involving a suspension of all normal life, but civil litigation continued and the law reports are full of collision cases. Some of these collisions were caused in part by wartime constraints on navigation, but in the main were resolved by the application of normal principles of liability.108 A collision claim for Can$2 million was lodged by CGT for the loss of Mont Blanc and a counterclaim was brought on behalf of Imo, strangely for exactly the same amount.109 There is no mention of the potentially huge claims in respect of ships damaged in the harbour or of the property damaged on land,110 nor even of claims for the loss of Mont Blanc’s cargo.111 Limitation

100 ibid 172–75. 101 See Zemel (n 95) ch 11. 102 Copies of the ensuing report are rare, but copious extracts appear in Zemel (n 95). 103 Zemel considers that Drysdale had shown ‘deep bias’ against Mont Blanc: ibid 204. 104 He had the power to forbid Imo to leave, but depended on pilots to let him know of departures. A 16-year-old pilotage clerk had effectively stopped reporting to Wyatt’s office, as he thought his reports were being ignored. 105 Zemel (n 95) 236–37. 106 Which rules were not specified, but this was clarified in the later collision action; see section III.C below. 107 See Zemel (n 95) 244–46. The criminalisation of seafarers is a recurrent feature of maritime disasters. 108 Admiralty Commissioners v Owners of the SS Volute (The Volute) [1922] 1 AC 129 (HL) is a notable example. 109 No attempt appears to have been made to involve the Clara in the collision litigation, despite the criticism of her pilot in the Investigation Report, presumably because of the later causative actions of the colliding ships. 110 The insurers of local properties initially declined payments on the basis that the cause of loss was war and explosion, which were not covered by policies. Negotiations led by the Halifax Relief Commission (see section III.D below) resulted in the insurers agreeing to pay between 20 and 30% of losses: see Kitz (n 97) 142. 111 The French Government’s interest appears to have been mainly through the cargo. It has been assumed that it was the owner (although this would not necessarily follow). The ship was apparently operated as a normal commercial vessel,

14 Nicholas Gaskell of liability was not mentioned in the ensuing litigation.112 On my rough calculations,113 the Mont Blanc would have had a limit for property claims of about Can$126,559, while the Imo’s limit would have been about Can$185,895.114 Even stranger is the absence of personal clams in the litigation.115 For personal claims, the Mont Blanc limit would have been about Can$237,299 and the Imo’s about Can$348,552.116 The collision action was decided on 27 April 1918 by the same Drysdale J who had formed such firm views at the Formal Investigation. He was presented with the usual problems of assessing evidence common to collision cases at the time, in particular: which navigators to believe as to respective courses at crucial times; who gave which signals and when; and what was the exact place of impact between the two vessels? The parties had agreed to adduce transcripts from the investigation, with additional evidence of only one other witness, Makiny, the master of the naval tug Nereid. Makiny’s evidence was that the collision had occurred very much on the Dartmouth (ie Mont Blanc’s) side of the channel, but this was contrary to the judge’s pre-existing views that the collision was caused by the improper change of course to port by Mont Blanc, so that the place of collision was more towards the Halifax side.117 Drysdale J adhered to his earlier findings, and in a three paragraph judgment held the Mont Blanc wholly to blame.118 In March 1919, Drysdale J’s decision was appealed to the Canadian Supreme Court, which produced a split decision. Two judges upheld Drysdale J’s finding that Mont Blanc was solely to blame, two found that Imo was solely to blame and one, Anglin J, found them both to blame.119 The result, somewhat curiously, was that Drysdale J’s decision was reversed in part and both ships were found at fault. It is significant that Anglin J rightly identified that Drysdale J had not really made key findings about (i) the exact position of the Mont Blanc (in relation to the channel’s centre line) before she signalled to turn and (ii) the bearing of the Imo at that moment. Nevertheless, he considered that this vital manoeuvre of the Mont Blanc was not yet necessitated by ‘an agony of the collision’,120 and that she could instead have turned a little more to starboard or reversed her engines. Both shipowners appealed against this decision. The appeal reached the Privy Council in February 1920, two years before the Empress of Ireland appeal (even though her collision had occurred three years before the Halifax collision).121 The judgment was delivered by Lord Atkinson122 and the appeals of both ships were dismissed.123 but presumably covered by war risk insurance. It is possible that she had been chartered or requisitioned by the French Government, and the master may have been in the naval reserve: see Kitz (n 97) 4; Zemel (n 95) 96. The French interest in the cargo was part of the political background to the aftermath of the collision. 112 It is possible that it may have been pleaded as a defence. 113 Under the Merchant Shipping Act 1894, s 503, using the £8 per ton limit converted into Canadian dollars and based on the known tonnages of the ships: see also section I and n 18 above. 114 This would surely have been relevant if either ship had been solely at fault: see also the text to n 134 below. 115 But see section III.E below. 116 Using the limit of £15 per ton, of which the personal claimants would have had the top £7, and shared the remainder with the property claimants: see section I and text to n 10, above. 117 See the transcript in Zemel (n 95) 263. 118 Compagnie Générale Transatlantique v Steamship ‘IMO’ (1919) 47 DLR 462 (Exch). There were 10 pages of argument, yet the judgment covers barely two pages. 119 See Compagnie Générale Transatlantique v Steamship ‘IMO’ (1919) 59 SCR 644. Surprisingly for such a major Canadian disaster with an extensive difference of opinions, there is only a partial two-page report of the decision. A summary of the judgments appears in Zemel (n 95) 265–71. 120 For an explanation of this phrase, see n 129 below. 121 Butler Aspinall KC, appearing for the Imo, had also represented the owners in both the Titanic and Empress of Ireland investigations. 122 In strong court also consisting of Viscount Haldane, Lord Dunedin, the Lord Justice Clerk, sitting with two naval assessors. 123 Steamship Imo v Compagnie Générale Transatlantique (1920) 2 Ll L Rep 536 (PC judgment, 22 March); see also (1920) 2 Ll L Rep 188 and 246 (PC hearing and submissions, 9–16 February).

The Empress of Ireland and Mont Blanc Collisions: Then and Now 15 Lord Atkinson echoed Anglin J’s criticism of the focus of the trial, with its ‘economical’ decision to reuse evidence from the investigation: ‘the evidence which would be relevant to the important points in controversy in the action has not been elicited so fully, or with the same precision and directness as it doubtless would have been had it been given orally’.124 The Privy Council therefore felt obliged to examine all the evidence, but without the advantage of a crossexamination of the Imo’s helmsman, Johansen (in particular as to whether the Imo was in her own water).125 It was unable to resolve conflicting evidence about the nature, number and sequence of signals, but was more satisfied with the actual movements to ‘which these signals, at best, ought merely to be the heralds’.126 Here the Privy Council found aspects of Imo’s case to be ‘irrational’, ‘absurd’, ‘incredible’ and ‘impossible’.127 It found that Mont Blanc was proceeding at a ‘not immoderate’ speed in her own waters before the final manoeuvres, while Imo was going faster than her witnesses admitted. The impact may have occurred on the Halifax side of the centre line.128 However, it was not necessary finally to resolve this issue, as the crucial finding by the court was that both masters should have reversed their engines when about 400–500 feet distant, before the final manoeuvres. Thus, although the final manoeuvre by Mont Blanc may have been excusable as being in the ‘agony of the collision’,129 she had failed to prove that the omission to go full astern was good seamanship.130 Looking at the evidence with hindsight is risky, but it seems that, once Drysdale J’s prejudices were put aside,131 the main cause of the risk of collision was the responsibility of the Imo – creating a situation of danger that put Mont Blanc’s master and pilot in a really difficult position. It is hard to criticise Mont Blanc for the last-minute attempt to avoid the impending collision. The change of course appeared to be designed to deal with Imo’s apparent attempt to pass her to starboard. A finding that Mont Blanc should have reversed her engines sooner seems harsh, given the dangerous nature of her cargo and the proximity of the Dartmouth shore close to starboard,132 but one can perhaps sympathise with an appeal court having to deal with the case on documents without hearing the witnesses in person. The court had rejected the evidence, favourable to Mont Blanc, of Makiny – out of deference to its rejection by Drysdale J, who saw and heard the witness.133 124 Steamship Imo (PC) (n 123) 537. 125 The only evidence given at the investigation was a single answer in response to a leading question from Imo’s counsel: ibid 542. 126 ibid 540. 127 ibid 541. 128 ibid 544. 129 See, eg London Steamboat Co v Owners of the Bywell Castle (The Bywell Castle) (1879) 4 PD 219 (CA). This was itself a case about a major disaster (largely forgotten) involving the sinking of the overloaded paddle steamer, Princess Alice, in the Thames, as a result of which 600–700 died. An inquest had found both vessels at fault, but the Court of Appeal ultimately excused the other ship because of the difficulty she had been placed in by the bad navigation of the passenger ship: see www.rmg.co.uk/discover/behind-the-scenes/blog/princess-alice-disaster. It is probably significant that in 1879, prior to the Maritime Conventions Act 1911, the courts had no power to apportion blame other than equally. A voluntary fund was established, which raised some £35,000 for bereaved families, a not insignificant sum for the time: see J Lock, The Princess Alice Disaster (London, Robert Hale, 2013) 166. 130 Perhaps surprisingly, the court found that the burden of proving this omission rested upon Mont Blanc: Steamship Imo (PC) (n 123) 538, citing The City of Berlin [1908] P 110 (CA). 131 It is perhaps unfair to note that the next reported Privy Council decision also reversed Dysdale J (in a prize case): see Rederiakatiebolaget Navigator v Newcombe [1920] UKPC 26, [1920] 51 DLR 426. 132 Le Médec had given evidence that for him to have gone astern would have risked the ship slewing ashore with his sensitive cargo: see Zemel (n 95) 34. The Privy Council did not seem to take account of this evidence when finding that he should have reduced speed earlier. 133 See Steamship Imo (PC) (n 123) 540. One senses that the decision to put aside this evidence was balanced by the fact that the court went on generally to prefer Mont Blanc’s evidence over Imo’s. The court also largely refused to disbelieve crucial evidence from Mayers, an officer on the nearby Middleham Castle, because he had been criticised for being under ‘some fantastic delusion about his being driven through the air for a great distance by the force of the explosion’: ibid 540. In fact, later accounts of survivors also mention this phenomenon: see Kitz (n 97) 32.

16 Nicholas Gaskell The result of the Privy Council decision was that, as the ships were equally to blame, with identical (Can$2 million) damages claims, neither ship would make a recovery – even though the limit of liability of the Imo may have been the higher of the two.134 Although the Maritime Conventions Act 1911 (Imperial) was not in force in Canada,135 by 1917 Canada had enacted its own Maritime Conventions Act 1914.136 Like its UK counterpart, it gave effect to the Collision Convention 1910 and introduced the concept of proportionate fault.137 No argument appears to have been addressed to the Privy Council about the possibility of splitting liability other than equally, and the parties presented their cases on an all-or-nothing basis. A century later, it seems that this would be a prime case to argue for apportionment in favour of the Mont Blanc.138

D. What Happened to the Halifax Personal Claims? What puzzled me in reading the various accounts of the Halifax disaster is the absence of any mention of litigation by the personal claimants against the shipowners. The collision case heard up to the Privy Council seemingly involved only the property claims of the respective shipowners. Lord Campbell’s Act claimants could have been ‘embarrassed’ by the inability to use an action in rem, but there was no value in either ship – one being wrecked, the other having disintegrated. Unlike the Empress of Ireland litigation, there was no sale fund in court in respect of which the personal claimants could have intervened, and both shipowner defendants were foreign – raising difficulties of enforcement. Still, the property claims ultimately heard in the Privy Council would presumably not have proceeded if the parties were thought to be without means.139 What seems to have happened in the aftermath of the Halifax disaster is that the needs of the personal claimants were filled by donations.140 A remarkable amount was contributed by governments141 and charities (particularly those based in the USA).142 The funds were largely

134 See text to n 114 above. This is because the Admiralty rule, settled in 1882, was that limits were only applied to the balance of any claim: see Stoomvaart Maatschappy Nederland v P & O Steam Navigation Co (The Khedive) (1882) 7 App Cas 795 (HL). Where the claims are equal, neither ship will recover anything against any limitation fund, although that may be available to cargo claimants: see Marsden and Gault (n 16) 640. 135 Although the Maritime Conventions Act 1911, s 9(1) extended the Act throughout His Majesty’s dominions, s 9(2) excluded Canada (as well as the other main dominions, such as Australia, New Zealand and South Africa). See also section III.C, text to n 59 above. 136 ie after the Empress of Ireland collision. In fact, the failure to deal fully with Admiralty apportionment rules in non-collision cases continued to cause problems for Canadian law until the 1990s and Supreme Court of Canada decisions such as Bow Valley Husky (Bermuda) Ltd v St John Shipbuilding Ltd [1997] 3 SCR 1210 and the subsequent Marine Liability Act 2001 (Can), pt 2: see A Chircop et al, Canadian Maritime Law, 2nd edn (Toronto, Irwin Law, 2016) 5–7, 844–45. 137 Like the Maritime Conventions Act 1911, it also provided for joint and several liability of ships at fault where there was loss of life or personal injury: see also Chircop et al (n 136) 844. 138 See section IV.C, text to n 186 below. 139 The hull and cargo underwriters (and/or war risks insurers) may have been based in London and the munitions on Imo may well have been owned by the French Government, but I have not yet been able to trace more information about this or whether there were personal claims that were brought and settled. Contemporary notions of causation and remoteness would not seem to have presented a bar to personal claims: see, eg HMS London [1914] P 72 (PDAD); Re Polemis & Furness, Withy & Co Ltd [1921] 3 KB 560 (CA). 140 A voluntary fund had been established after the loss of the Princess Alice in 1878: see n 129 above. 141 The Canadian Dominion Government gave an initial $6 million, rising to $18 million; the State of Massachusetts gave $750,000; the Australian Government gave $250,000; and a London fund raised $600,000: Kitz (n 97) 83 (although not specified by the author, these figures are presumably all Can$-denominated sums). 142 Much volunteer assistance was provided by Boston citizens: Kitz (n 97) 186–87.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 17 administered by a Halifax Relief Commission.143 Kitz144 cites a review in 1918 reporting that expenditure on matters such as emergency relief, claims and temporary housing145 had already totalled Can$18.5 million: Material losses were $35 million, including Dominion government and shipping losses of about $10 million; railway, about $1.2 million; dwellings about $6.5 million; churches, $1 million; personal belongings, about $3.5 million; merchandise, about $1 million; and manufacturing plants, about $3.5 million.146

By the spring of 1918, donations amounted to $23,153,793.147 The Halifax Relief Commission apparently considered that one of its primary roles was to set up and pay out pensions to survivors of those killed, on the basis of ‘relief, not compensation’.148 It appears that this relief was operated as a type of workers’ compensation scheme, albeit not financed by employers. The Commission paid pensions for dependants of Canadians killed on merchant ships not owned by the navy,149 while pensions for service families were paid by the Dominion government; seamen from foreign ships or their families were ‘covered by their native country’s war insurance’.150 In the first four years after the explosion, the Commission disbursed $1 million in pensions and lump sums; in 1920, there were 1,028 pensioners.151 Although increases in payments were made between 1919 and 1921 to account for inflation, the Commission was already indicating that there would not be enough money in the fund to pay pensions for life.152 In fact, pensions continued to be paid until after the Second World War,153 and the Commission also made payments towards civic improvements before it was finally dissolved in 1976.154 CGT, as owners of the Mont Blanc, apparently made a donation of $10,000 to the Commission.155 In retrospect, this seems a very modest payment for what, in modern terms, might have been expected to be a huge liability. I can find no reference to donations made by the Imo owners. It may be that it was easier to think of the human expense of the disaster as another cost of the war, to be picked up by society at large in the same way as with military casualties.

E. Was There a Cover-Up? The Formal Investigation placed much of the blame on the Mont Blanc’s pilot (Mackey) and her master (Le Médec),156 along with the CXO (Wyatt). The focus on individuals deflected criticism 143 ibid 179–86. 144 ibid 179. 145 Some 750 houses had been destroyed, 750 needed extensive repairs and 9,000 needed repairs of some sort: ibid 187. The Commission completed the construction of 328 dwellings between 1919 and 1921, based on a new type of fireproof cement block called hydrostone, and these were made available for rent by about 2,000 people: ibid 188, 189. The Hydrostone Estate is now a heritage area. 146 ibid 179. 147 ibid. 148 ibid 180. 149 ibid 184. 150 ibid 182. 151 ibid 185–86. 152 ibid 186. 153 Numbers dwindled, but in 1989 there were still 30 survivors receiving pensions: ibid 203. 154 ibid 202. By that stage, it had spent more than $30 million on relief and reconstruction; $3,970,808 on emergency relief; $20,231,727 on claims, construction and temporary housing; and $6,420,000 on pensions and medical care. 155 ibid 83. 156 The master was relatively new to the ship, but the evidence shows that he was very cautious about his cargo. There was also some anti-French prejudice in Nova Scotia.

18 Nicholas Gaskell away from more systemic failures, but in wartime there was a reluctance to blame the governments for a ship loaded with a dangerous French cargo in the USA, under British orders and brought into a Canadian port. There is evidence that there was political pressure to avoid criticism both of the administration of pilotage in the harbour and of the Royal and Canadian Navies. The Prime Minister of Canada, Robert Borden (also MP for Halifax), was anxious to minimise criticism of a lack of federal control of pilots in Halifax, particularly with an election looming.157 Any systemic failure may have reflected badly on Borden, but also on Charles Ballantyne, Minister of Marine and Fisheries and Minister of the Naval Service. The CXO, Wyatt, was criticised in part for the lack of communication between the harbour authorities and the naval guard ship at the entrance to Bedford Basin, as the Imo’s departure might have been delayed if it had been known a munitions ship was entering the Narrows.158 Yet, in practice, British naval authorities allowed munitions ships to enter the harbour in wartime even if that might not have been allowed in peacetime. To some extent it is understandable that there was a need not to undermine civilian morale in wartime by highlighting structural failings. In retrospect, it is easy to say that this was a disaster waiting to happen as a result of naval practices regarding munitions ships, inadequate communications about movements between naval and civilian authorities, and a somewhat laissez-faire attitude towards pilotage practices.

IV. A Twenty-First Century Maritime Law Perspective on the Disasters Maritime law in Canada has developed significantly since the disasters of 1912, 1914 and 1917. While many of the Canadian changes have reflected an outgrowing from an imperial system (as has also occurred in other states, such as Australia), they have also echoed the evolution of international maritime law as represented by conventions promulgated by the Comité Maritime International159 and International Maritime Organisation (IMO). Here the broad picture looks very much like that in UK law. By way of a conclusion, it may be of interest to ask what legal lessons have been learned from disasters such as those described above160 and how might the legal cases be decided differently today?161

A. Understanding the Facts One simple illustration of the difficulty of reading the cases is that there was a difference in practice as to helm orders. The British system, based on tillers in the days of sail, was that an order to ‘port’ or to ‘port the helm’ was intended to indicate that the ship was to turn to starboard! This was because moving the tiller to the left (port) would have this effect – and vice versa for 157 See, eg J Maybee, Aftershock: the Halifax Explosion and the Persecution of Pilot Francis Mackey (Halifax, Nimbus Publishing, 2015) 107; Zemel (n 95) 290–94. 158 There may have been some prejudice against Wyatt in naval circles because he had been divorced: Zemel (n 95) 278–82. 159 Like the Collision Convention 1910 and Salvage Convention 1910. 160 Broader comparisons have been made in J Schröder-Hinrichs, E Hollnage and M Baldauf, ‘From Titanic to Costa Concordia – a Century of Lessons Not Learned’ (2012) 11 WMU Journal of Maritime Affairs 151. 161 The scope of this conclusion is necessarily selective and brief, and mainly from a UK perspective.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 19 starboard orders. It was not until the enactment of Merchant Shipping (Safety and Load Lines Conventions) Act 1932, section 29, that the UK adopted the modern system of giving orders, which reflect the intended direction of the vessel. A modern reader of earlier case law, unaware of the earlier differences in practice, might wonder how ships ever came into collision, as they appeared to be steering away from each other. This quirk was also of significance at the time of the Halifax disaster as it appeared that practices differed between masters and pilots – Le Médec (the French master of Mont Blanc) used what we would call the modern system of orders, while pilot Mackey gave his orders to the helmsman using the older system.162 Butler Aspinall KC163 for the Imo suggested in argument164 that this caused confusion aboard the ship, but the argument gained no traction before the Privy Council and does not seem to have been borne out by the evidence.

B. Investigations Accident investigations are now undertaken in private by the Marine Accident Investigations Branch, which has been independent from the Department for Transport since the Herald of Free Enterprise disaster in 1987.165 Formal investigations, chaired by a judge as Wreck Commissioner, still take place for really serious disasters.166 They can satisfy community concerns though a public hearing167 and can often lead to a change in legislation or practice.168 Judges can still use their reports to make trenchant criticisms, including findings of blame, which can be highly influential in settling cases.169

C. Proof and Apportionment in Collision Cases Fact finding in collision cases has always been notoriously difficult, even with a collision like that in Halifax, occurring in broad daylight in a river with numerous apparently disinterested bystanders. Both Canadian collisions illustrate the evidential difficulties, especially when the accounts from each vessel are ‘irreconcilable’.170 The judges felt obliged to choose between the stories, often based on their impressions of a few witnesses who may not have been particularly

162 For an almost farcical extract from the transcript of the Formal Investigation, see Zeni (n 22) 97–101. 163 He also appeared as counsel for CPR in the Empress of Ireland investigation and litigation, and for the White Star Line in the Titanic investigation. 164 Steamship Imo (PC hearing) (n 123) 191. 165 See now the Merchant Shipping Act 1995, pt XI, s 267, replacing the Merchant Shipping Act 1894, s 738, under which Board of Trade inspectors reported on accidents, but might have been conflicted if they needed to criticise the Department of Transport for lack of oversight: see N Gaskell, ‘Merchant Shipping Act 1995’ in Current Law Statutes 1995 (London, Sweet & Maxwell, 1995) 21/285. 166 Now under the Merchant Shipping Act 1995, s 268. 167 The failure to hold a Formal Investigation into the Marchioness disaster in 1989 led to much public criticism: see Thames Safety Inquiry Final Report (Cm 4558, 2000). A Formal Investigation was eventually held by Clarke LJ in 2000: see Marchioness/Bowbelle Formal Investigation (HMSO, 2001) vol 1. 168 eg the two changes introduced in Canada in 1914: see n 29 above and text to n 136 above. 169 In his Herald of Free Enterprise Report, Sheen J notably stated of the shipowner that ‘[f]rom top to bottom the body corporate was infected with the disease of sloppiness’: MV Herald of Free Enterprise Report of Court No 8074 (HMSO, 1987) para 14.1. This was widely interpreted as a pointer for liability claims (and possible evidence to deprive the owner of the ability to limit). It is clear, though, that the results of an investigation are not binding in any subsequent collision action: see The Speedlink Vanguard and The European Gateway [1987] QB 206 (QB (Admlty Ct). 170 Report of Commission of Inquiry (n 23) 11.

20 Nicholas Gaskell articulate.171 The impression is that the fact finding in both cases was influenced at the investigation stage, at least, by wider factors – including the political and social pressures inevitably arising with disasters involving great loss of life. Difficulties of proof were compounded when key navigators were killed, as happened, for example, on the Imo. Today, judges are far more aware about unconscious bias against witnesses, and more concerned to concentrate on objective evidence. A modern collision action looks radically different, as investigators, litigators and judges now have voyage data recorders available to them. In 2017, Teare J explained in The Alexandra 1 and The Ever Smart172 that this enabled parties to agree the tracks of each vessel at case management stage and to produce ‘a schedule showing the course, heading and speed of the vessels from C-26 until collision’.173 The schedule also shows engine orders (from engine loggers) and helm orders (taken from bridge audio recordings).174 Ports and busy coastal areas may now be subject to vessel traffic services (VTS), which enable monitoring of navigation and warnings to be given to ships; VTS radar recordings will also be available after the event.175 While this availability of electronic data presents new difficulties and expense, Teare J was able to say that it was now common in collision actions for there to be little factual dispute concerning the navigation of each vessel (apart from perceptions on the bridge about what the other vessel was doing).176 The significant changes in practice are now reflected in the Civil Procedure Rules, Part 61, Practice Direction 61 and The Commercial Court Guide (incorporating the Admiralty Court Guide (10th edn 2017)). Computer simulations are also available, either to produce a plot or, in some cases, to provide a virtual picture from a full bridge simulator (almost like watching a video game), where alternative scenarios can be tested. Of course, there can still be factual problems when a ship is a total loss and it is impossible to retrieve her data,177 or where the main navigating personnel are dead. While proof of facts in collision cases has changed significantly in the last century, there has been a gradual evolution in the principles of apportionment. As already noted, Admiralty law had been in advance of the common law by not applying contributory negligence as an absolute bar to liability.178 The Admiralty rule of divided damages applied so that the court could apportion liability equally between both ships, ie each would be held equally to blame without any further adjustment. The Maritime Conventions Act 1911 gave effect to the Collision Convention 1910 by allowing a court to apportion liability following the European approach rather than according to the 50:50 Admiralty rule,179 and that is still the rule today.180 At the time of the

171 See, eg the dismissal of the evidence in favour of the Storstad given by Galway on the Empress of Ireland, and of that of independent witness Makiny in favour of Mont Blanc. 172 Evergreen Marine (UK) Ltd v Nautical Challenge Ltd (The Alexandra 1 & The Ever Smart) [2017] EWHC 453 (Admlty), [2017] 1 Lloyd’s Rep 666 [2] (upheld on appeal, [2018] EWCA Civ 2173, [2019] 1 Lloyd’s Rep 130). 173 ibid; ie from 26 minutes before the collision (or from whatever earlier time is relevant). 174 These can reveal VHF communications between ships, a possibility not open to the navigators of a century ago. It can facilitate telling each other what they expect to do next and perhaps agreeing to ignore the Collision Regulations (as also occurred between the Storstad and Clara), but new problems can arise, eg when navigators are confidently communicating with the wrong ship! 175 See further Marsden and Gault (n 16) appendices 6 (VTS), 10 and 12 (electronic charts), 11 and 14 (radar), 18 (voyage data recorders). 176 The Alexandra 1 & The Ever Smart (n 172) [2]. 177 The diving difficulties faced in examining the Empress of Ireland (see n 36) may be fewer today. It may be easier to undertake underwater surveys of equipment or to commission major engineering studies, eg to ascertain whether a rudder had a sufficient surface area to be effective in particular navigational circumstances. 178 The common law rule was changed by the Law Reform (Contributory Negligence) Act 1945. 179 See, eg Roscoe and Robertson (n 48) 6. 180 The Maritime Conventions Act 1911 was repealed and replaced by the consolidating Merchant Shipping Act 1995, s 314(1) and sch 12. The relevant provisions are now found in the 1995 Act, ss 187–89.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 21 1911 Act, expert commentators forecast that English courts would usually continue to divide the damages equally.181 In fact, the most authoritative summary of the development of the principles of apportionment, in a 1975 article by Sir Henry Brandon (as he then was),182 noticed that although after 1911 the courts were cautious as to apportionment, they used broad allocations of liability, such as one-third and two-thirds, or one-quarter and three-quarters.183 In more recent times, courts have been more willing to use percentages, such as 80 per cent and 20 per cent.184 Making a very tentative attempt to apply Brandon’s principles to the facts of the Halifax collision,185 one could suggest that an apportionment of 75–80 per cent against Imo might have been more appropriate.186

D. Limitation of Liability Although after 1862 the UK moved away from basing limitation on the value of the vessel,187 the original 1862 linkage of limitation to values became lost by the time of the International Convention Relating to the Limitation of Liability of Owners of Sea-Going Ships in 1957. The calculation of limits is now merely mechanical. Increases at the IMO in the limits in the Convention on Limitation of Liability for Maritime Claims (LLMC) 1996 are based on largely political and economic grounds unrelated to ship values, even where very high value container or passenger ships are concerned. When the positions of shipowners and potential claimants are balanced at the IMO, shipowners probably fare better. It might have been possible to develop limitation based on pre-accident values, but this could have presented almost insuperable problems where the ship was a total loss, so that no expert valuation was possible, and where insured values might be unreliable evidence.188 It might have been thought that once shipowners began to use limited liability companies to own ships in the nineteenth century, there was no need for an additional form of limitation of liability other than that of the company itself. Nevertheless, maritime limitation of liability has continued to survive in the form of conventions such as the LLMC 1996, with the modern justification for its existence being less about the need to stimulate investment189 and more about the need for insurers such as protection and indemnity clubs (P&I Clubs) to be able to offer liability cover. If maritime limitation of liability were abolished, insurers would almost certainly

181 Roscoe and Robertson (n 48) 6. The authors also cited Lord Gorell in debates; Lord Gorell seemed convinced that in 19 cases out of 20 a judge would be incapable of being sure which of the two would be the more to blame: see HL Deb 31 October 1911, vol 10, cols 16–38. 182 H Brandon, ‘Apportionment of Liability in British Courts under the Maritime Conventions Act of 1911’ (1976–77) 51 Tulane Law Review 1025. This extrajudicial statement by an experienced Admiralty judge has rightly been described as a ‘major contribution’ by Gross LJ in a recent collision case, The Alexandra 1 and Ever Smart (CA) (n 172) [117]. 183 Brandon (n 182) 1041. 184 Brandon (ibid) 1041 added that appellate courts had even made or approved divisions such as 85% and 15%, citing The Martin Fierro [1975] 2 Lloyd’s Rep 130 (CA) and The Statue of Liberty [1971] 2 Lloyd’s Rep 277 (HL). My recollection of Brandon’s lectures to the profession in courses in Southampton in the late 1970s was that he rather deprecated such fine distinctions as they were quite unpredictable when trying to settle cases. 185 See section III.C, text to n 140 above. 186 The ‘blameworthiness’ would seem to fall almost entirely on Imo, although it could be said that the ‘causative potency’ lay significantly with Mont Blanc, given her cargo. On any analysis, an equal division of liability seems harsh. 187 See section I, text to nn 11 and 12, above. 188 The Titanic apparently had hull and machinery insurance cover for £1 million, although the pre-accident value of the vessel was £1.5 million: see C Hewer, A Problem Solved: A History of the ILU 1884–1984 (London, Witherby, 1984) 29. 189 Although this is not an insignificant issue, eg in relation to the newbuilding costs of container ships or cruise liners.

22 Nicholas Gaskell be obliged to offer cover only up to amounts set by themselves in the policy. In fact, even the P&I Clubs have put caps on overall liabilities for many claims, as required by their own reinsurers.190 It might be said that maritime limitation of liability through conventions at least gives states the ability to set the limits rather than private insurers, but the experience at the IMO is that the limits for individual ships are not set at levels that reflect the maximum amount that insurers would offer. If there was widespread denunciation of conventions like the LLMC 1996 and insurers reacted by setting limits on individual cover for ships, states could only react by requiring minimum levels of cover. These would be based on a perceived need, but would be a fundamental change in emphasis from creating a maximum limit of liability, which is often rather low in respect of smaller ship. In fact, since the Torrey Canyon disaster in 1967, the IMO has been moving towards such a system of minimum insurance cover for certain types of claim.191

E. Dealing with the Single Ship Company The IMO has provided a workable solution to the type of problem where shipowners have been able to hide behind single ship companies, as occurred with the Storstad. It has produced conventions creating strict liability for a number of particular types of maritime casualties, such as oil tanker pollution,192 bunker pollution (from non-tankers),193 pollution from hazardous and noxious substances (HNS) carried in ships,194 wreck removal195 and passenger196 claims.197 More importantly, the financial security of shipowners for these claims198 is underpinned by compulsory insurance with direct action against the insurer. In practice, this means that an insurer (usually a P&I Club) can be sued directly by a claimant without the latter having to sue a shipowner that may be insolvent.

(i) Crew Claims Crew claims have been made easier both in relation to limitation of liability199 and, more importantly, through adopting the IMO’s financial security principle. The plight of crews after a collision has been much improved by the International Labour Organization’s Maritime Labour Convention 2006. This provides a much better framework for unpaid wages, repatriation costs (eg where crews are abandoned after a casualty) and compensation for death and injury. Liabilities are also backed by financial security through certificates issued by P&I Clubs. 190 See, eg US$1 billion for oil pollution: www.igpandi.org/reinsurance. 191 See section IV.E below. 192 International Convention on Civil Liability for Oil Pollution Damage 1992; International Convention on the Establishment of an International Fund for Oil Pollution Damage 1992. 193 International Convention on Civil Liability for Bunker Oil Pollution Damage 2001. 194 Convention on Liability and Compensation for Damage in Connection with the Carriage of Hazardous and Noxious Substances by Sea 2010 (HNS Convention 2010). 195 Nairobi Wreck Removal Convention 2007. 196 Athens Convention Relating to the Carriage of Passengers and their Luggage by Sea 2002 (Athens Convention 2002). 197 I have described the effect of these conventions in N Gaskell and C Forrest, The Law of Wreck (Abingdon, Informa Law/Routledge, 2019) chs 2 and 10. 198 Although not for traditional cargo claims. 199 Under the LLMC 1976/1996, Art 3(e), crew claims against the employing ship would now only be subject to limitation of liability by the shipowner if that is allowed by the law governing the contract of service. If that were UK law, then the shipowner could not limit: see the Merchant Shipping Act 1995, s 185(4). Where foreign law is concerned, limitation may well be available. cf n 62 above.

The Empress of Ireland and Mont Blanc Collisions: Then and Now 23

(ii) Empress of Ireland The twentieth century saw much progress in dealing with contractual exclusions in p assenger contracts through judicial activism or legislation,200 but if the latest Athens Convention 2002 were applied to the facts of the Empress of Ireland, her passengers would now be in a far better position. As there was a ‘shipping incident’ (ie a collision), the contracting carrier (eg CPR) would be strictly liable to the passengers under Article 3, although that liability would only extend up to 250,000 special drawing rights (SDR) (about £274,608).201 For any claim between 250,000 SDR and 400,000 SDR (about £439,372),202 the carrier would be liable unless it could prove the absence of fault. On the basis that the Storstad was wholly at fault, CPR, as the contracting carrier, would therefore be liable for the lower amount per passenger. This sum could be recoverable from the insurer that had issued the compulsory Athens Convention 2002 certificate,203 even if the Empress of Ireland was owned by a single ship company. Ironically, the liability position of the Storstad to the passengers in tort would not have changed,204 and she could limit her liability under Article 6 of the LLMC205 for any recourse claim brought by the Empress of Ireland.

(iii) Mont Blanc By modern standards the Mont Blanc was a tiny ship, yet she caused immense damage. While most of the IMO maritime liability conventions have concentrated on marine pollution liability, the possibility of a hazardous cargo exploding is still very real.206 The carriage of large quantities of munitions is obviously much less frequent in peacetime than in wartime, and the Halifax disaster prompted much thought about how such cargoes should be isolated from centres of population. Nevertheless, other hazardous and explosive substances are still carried in merchant ships, in bulk form or packaged, and these ships will often operate in ports that are not so isolated from major population centres. Liabilities to third parties would seem in most cases to be subject to the ordinary rules of tort, eg requiring proof of negligence. Shipowners could still claim to limit liability (eg under the LLMC 1996) even for personal injury or death in circumstances that would not apply to the owner of a factory or warehouse on land. For a ship of the size of the

200 See eg Hollingworth v Southern Ferries Ltd (The Eagle) [1977] 2 Lloyd’s Rep 70 (QB); the Athens Convention 1974, Art 18 (enacted originally in the Merchant Shipping Act 1979). 201 Converted as at 1 July 2019. 202 That is the limit of liability under the Athens Convention 2002, Art 7. In UK law, there would be no further limit applying under the LLMC 1996, Art 7, as the UK has opted to apply only the Athens limits to seagoing ships: see Gaskell, ‘Limitation of Liability and Division of Loss in Operation’ (n 16) 981. 203 HNS Convention 2010, Art 4bis(10). A passenger ship like the Empress of Ireland would now have such a certificate as she would not be allowed to use ports in contracting states without one. 204 Although their ability to sue an insurer under the Third Parties (Rights against Insurers) Act 2010 would be easier, as ss 9(5)–(7) would prevent a marine liability insurer from relying on a ‘pay to be paid’ clause for death or injury claims. 205 To about £4,331,013. This is a very rough estimate, calculated under the LLMC 1996 using the IMF SDR rates for 1 July 2019 and based on a limitation tonnage of 6,028 gt (note that this is the grt of the Storstad in 1914, but would be calculated differently under the Tonnage Convention 1969): see also nn 70 and 72 above. 206 In December 2005, what was apparently the largest ever peacetime explosion in Europe occurred at the (inland) Buncefield oil storage terminal, caused by 300 mt of petrol overflowing and forming a vapour cloud. There was no loss of life but massive property damage, and claims were made in excess of £750 million: see Colour Quest Ltd v Total Downstream UK plc [2009] EWHC 540 (Comm), [2009] 2 Lloyd’s Rep 1 [1]–[3]. This gives an idea of the type of damage (or threat) that might be caused by similar problems involving a ship at an oil terminal.

24 Nicholas Gaskell Mont Blanc, its modern day LLMC 1996 limit could be as low as £7,207,076.207 This would have to cover all injury and death claims, as well as property damage.208 To deal with such risks, IMO produced the HNS Convention 2010. This would provide strict liability of a shipowner for damage209 caused by HNS carried by a ship (even if there had been a collision caused wholly or partly by the other ship, like the Imo in 1917). Amongst the substances classified as HNS are the Mont Blanc cargoes of picric acid and TNT (carried as packaged cargo) and benzol (whether in bulk or packaged form).210 The shipowner, CGT, would be liable up to a limit of about £14,756,007,211 again backed by compulsory insurance and direct action against the insurer (in case the ship was owned by a single ship company). In addition to this sum, an HNS Fund would top up this amount to an aggregate of about £274,607,500.212 The HNS Fund is designed to be contributed to by bulk HNS importers in contracting states and reflects the fact that it is not the ship but the cargo that is the real problem. There is an exception in the HNS Convention 2010 where the damage is caused by an ‘act of war’,213 but that would not have applied to the collision in Halifax (which merely took place in wartime). However, under Article 4(4), the convention would not apply to create shipowner liability for ‘warships, naval auxiliary or other ships owned or operated by a state and used, for the time being, only on Government non-commercial service’. While the carriage of munitions would probably qualify as a ‘non-commercial service’, it is not clear whether the French Government was ‘operating’ Mont Blanc (eg under charter), as opposed to using as an ordinary cargo owner the commercial service provided by CGT.214 In any event, it seems that the HNS Fund would step in here under Article 14 to cover all of that loss. Both the shipowner (if liable) and the HNS Fund would have rights of recourse against a negligent colliding ship (like the Imo in 1917), but that ship would be entitled to rely on its limits of liability, eg under the LLMC 1996.215 There is only one problem about applying the HNS Convention 2010 to a modern Mont Blanc explosion – the convention is not yet in force. Perhaps it would need another Halifax explosion to act as a stimulus for states to adopt it.

207 As explained in n 205 above, this is a very rough estimate based on a limitation tonnage of 3,121 gt (the grt of the Mont Blanc in 1917). 208 But, as under the Merchant Shipping Act 1894, s 503, the personal claimants would be given priority in relation to about £4.8 million of this figure, while sharing the remaining £2.4 million pro rata with property claimants. 209 Including death and injury, and property damage, outside a ship. 210 See HNS Convention 2010, Art 1(5)(iv); www.hnsconvention.org/hns-finder. 211 As explained in nn 207 and 209, this is a very rough estimate based on a limitation tonnage of 3,121 gt (the grt of the Mont Blanc in 1917), calculated under HNS Convention 2010, Art 9. 212 250,000,000 SDR under Art 14. 213 HNS Convention 2010, Arts 7(2)(a) and 14(2)(a). 214 The issue of her armament against U-boats might also be relevant. 215 On the assumed basis of a (modern) gross tonnage of 5,043 gt, she would have an LLMC 1996 property limit of only £3,677,513, using 1 July 2019 conversion rates.

2 Do Actual Carriers Require Special Treatment? FRANCIS REYNOLDS

I. What is an Actual Carrier? The idea of an ‘actual carrier’ refers to a carrier who actually carries goods by sea under some form of subcontract with a ‘contracting carrier’, who makes the main promise that the goods will be carried. The idea that such a party may deserve a special description and special rules is familiar to most English maritime lawyers, if only by reason of the provisions for such carriers in the Hamburg Rules of 1978,1 a convention which has not received much currency internationally, and the related proposals in the Rotterdam Rules of 2009,2 which have not yet been adopted anywhere. These are referred to below. But as will be mentioned shortly, there are clear precedents in other transport law in the UK and elsewhere. The significance of the phrase ‘actual carrier’ in the Hamburg Rules is to identify other persons involved in the carriage of the goods over and above the contracting carrier, in order to impose liability as if under the main contract on some such parties to the shipper or later claimant in respect of the goods carried. Such an idea has no place in the Hague or Hague–Visby Rules, which govern much carriage under bills of lading worldwide and which were drafted on the basis that where a claim was to be made the carrier of goods could be identified as the party liable on the contract. The Visby Protocol added a reference to third parties by conferring the protections of the Hague–Visby Rules on ‘a servant or agent of the carrier’,3 but they went no further than that and most of the utility of the new provision was removed by the addition of the words ‘such servant or agent not being an independent contractor’, which most actual carriers would be. Sometimes the phrase ‘performing carrier’ is preferred in this context. A developed version of this title appears in the Rotterdam Rules, which, consistent with their purpose of covering goods carried wholly or partly by sea, are drafted to take in not only subcontractors who carry the goods, but also subcontractors in the port area who do not perform the carriage but carry out (or promise to carry out) some ancillary role during the maritime portion of the carriage, such as stevedores, for whom the word ‘carrier’ might not be appropriate. Hence the phrase ‘maritime performing party’ is employed. A leading text on the Rotterdam Rules says of the phrase ‘actual carrier’ that ‘the concept is useful but the name “actual carrier” is poorly chosen because it tends to suggest that the “carrier” as defined in the contract of carriage is not “actually” a carrier after all’.4

1 United

Nations International Convention on the Carriage of Goods by Sea 1978. Nations Convention on Contracts for the International Carriage of Goods Wholly or Partly by Sea 2008. 3 Hague–Visby Rules, Art IVbis.2. 4 M Sturley, T Fujita and G van der Ziel, The Rotterdam Rules (London, Sweet & Maxwell, 2010) 131, fn 308. 2 United

26 Francis Reynolds This view is possible. It is clear that the phrase ‘actual carrier’ would not be suited to the Rotterdam Rules because of the non-carrying parties to be taken in. But for the Hamburg Rules, the phrase is quite suitable. There, the definition of ‘actual carrier’ states that the phrase ‘means any person to whom the performance of the carriage of the goods, or of part of the carriage, has been entrusted by the carrier, and includes any other person to whom such performance has been entrusted’.5 This can be taken to include several layers of intermediate contracting parties above the one that actually does the work,6 but the responsibility is confined to the carriage ‘performed’ by the last mentioned.7 From this, the phrase ‘performing carrier’ might again be thought preferable. But the Hamburg Rules use the phrase ‘actual carrier’ and define it, and the usage can be justified on the simple basis that there is a convenient distinction between two types of carrier, the ‘contracting carrier’, who makes the contract for the carriage of the goods, and the ‘actual carrier’, who actually carries them and is only liable in respect of the carriage it performs. In fact, the phrase ‘actual carrier’ has a clear precedent in air law. The Warsaw Convention on International Carriage by Air of 1929 only made provision for ‘successive carriers’, referring to a situation where a transit was regarded as a single operation, even though the passengers or goods were carried by different carriers for different stages of the journey.8 But the Guadalajara Convention of 1961 began by ‘noting that the Warsaw Convention does not contain particular rules relating to international carriage by air performed by a person who is not a party to the agreement for carriage’. It then provided definitions in Article 1 as follows: (b) ‘contracting carrier’ means a person who as a principal makes an agreement for carriage governed by the Warsaw Convention with a passenger or consignor or with a person acting on behalf of the passenger or consignor; (c) ‘actual carrier’ means a person other than the contracting carrier, who, by virtue of authority from the contracting carrier, performs the whole or part of the carriage contemplated in paragraph (b) but who is not with respect to such part a successive carrier within the meaning of the Warsaw Convention.9

Definitions on similar lines now appear in the Montreal Convention of 1999,10 which consolidates and amends earlier conventions. These conventions go on to make the actual carrier liable together with the contracting carrier, but only for what the actual carrier performs. For the purposes of this chapter, which is addressed to maritime law, the Hamburg Rules notion of ‘actual carrier’ will be the one that is initially discussed, though of course such discussion leads on naturally to the notion of ‘maritime performing party’ used in the Rotterdam Rules. To what situation, then, does ‘actual carrier’ reasoning in the maritime context refer? As already stated, it postulates a situation where one party undertakes the carriage (the ‘contracting carrier’), but in fact uses a subcontractor (or subcontractors) to perform the carriage in whole or in part. Under the Hamburg and Rotterdam Rules, that subcontractor and others become liable jointly and severally with the carrier which undertook the carriage, that is to say, as if any one of them was the carrier. Their liability is, however, only in respect of the carriage performed by them. 5 Hamburg Rules, Art 1.3. 6 See J Ramberg, ‘The Vanishing Bill of Lading and the Hamburg Rules Carrier’ (1979) 27 American Journal of Comparative Law 391. Contra, A von Ziegler in A von Ziegler, S Zunarelli and J Schelin (eds), The Rotterdam Rules 2008 (Aalphen aan den Rijn, Wolters Kluwer, 2010) 115. 7 Hamburg Rules, Art 10.2. 8 Warsaw Convention, Art 1.3. 9 As to successive carriers, see also below, text to n 67. 10 Montreal Convention, Arts 39–40.

Do Actual Carriers Require Special Treatment? 27

II. A Special Regime for Actual Carriers under Conventions The Hamburg Rules therefore operate by imposing an additional direct liability on the basis of the contract terms on the actual carrier,11 as do the Rotterdam Rules in respect of ‘maritime performing parties’,12 a potentially wider term. This is not a contractual liability, but rather a statutory liability on the model of a contractual liability: the contracting carrier is liable and entitled in contract, and the actual carrier’s liability is on the same terms. Their liability is joint and several, but that of the actual carrier is derived from statute, not contract.13 To add to the difference from a liability based on contract, the actual carrier, while liable, is not entitled on the contract of carriage to, for example, sue for freight; and any liability of the shipper for shipment of dangerous goods would be remediable only in tort. This is not a new idea. Although it may have started in the context of carriage by air, quite apart from the Hamburg and Rotterdam Rules, a similar liability of an actual carrier or equivalent has been adopted by several countries as part of their transport law.14 In The Starsin,15 Lord Hobhouse referred to section 437 of the German Commercial Code, which in fact did not apply to ocean shipping: a reform of maritime law led to a similar but not identical provision, section 509, effective in 2013. Would something similar, whether by way of domestic statute or adoption of international convention, be of advantage to English law? To answer this, we must first consider the English law position.

III. The English Law Position A. Contract Under English law, there is at present no resort to such reasoning in the maritime context. The analysis assumed by the common law, untouched by any special regime, is perfectly logical, although there are difficulties in applying it. If one party receives or arranges to receive goods for carriage and then entrusts them to a subcontractor, whether by means of a charterparty, voyage or time, or even under a second bill of lading contract,16 the facts and documentation must be examined to see whether the first party undertook carriage or not. If it did, the party actually carrying the goods must operate under some form of subcontract, does not undertake contractual liability to the shipper (though probably does to the main contractor) and is not (normally) liable to the shipper in contract, but could be liable in the tort of negligence (or possibly conversion). This is not the place to rehearse in detail the techniques used by English and other common law courts to determine who was and is the contracting party (part of the so-called ‘identity of carrier’ problem). The traditional inquiry starts by asking whether the bill of lading is a

11 Hamburg Rules, Art 10. 12 Rotterdam Rules, Arts 18, 19. 13 Hamburg Rules, Art 10.4; Rotterdam Rules, Art 20. 14 See F Smeele, Les Règles de Rotterdam: le droit des transports maritimes au XXIe siècle (Marseille, Institut Méditerranéen des Transports Maritimes, 2010) 117–18 (in English). 15 Homburg Houtimport BV v Agrosin Pte Ltd (The Starsin) [2003] UKHL 12, [2004] 1 AC 715 [142]. 16 eg Carrington Slipways Pty Ltd v Patrick Operations Pty Ltd (The Cape Comorin) (1991) 24 NSWLR 745; Cro Travel Pty Ltd v Australia Capital Finance Management Pty Ltd [2018] NSWCA 153.

28 Francis Reynolds ‘charterer’s bill’, binding the charterer who undertook carriage, or a ‘ship’s bill’, under which the subcontractor is the carrier. The matter will normally turn on the wording, and in particular the signature, on any bill of lading or waybill issued.17 The master usually is regarded as signing for the owner who employs him. However, he may be authorised to sign for the charterer. Conversely, the charterer may be authorised to sign for the master and hence for the owner. If, as is common, the signature is by port agents or the like, the question is then for whom they were signing – a question that they might not always be able to answer themselves. Much may also turn on the question of authority, actual or apparent, to sign. The most relevant parts of most bills of lading are the logo at the top and the space for signature at the bottom, which latter is quite likely to have wording already printed on it, ‘for the master’ or something similar. Since the master is employed by the owner or demise charterer, this normally indicates a contract with the owner or demise charterer. That indication can, however, be displaced by other indications. All this can provide considerable uncertainty when disputes arise. Special contract provisions can make a difference. Sometimes, quite likely by accident because such a clause is printed on the back of the bill of lading anyway, there is a ‘demise’ or ‘identity of carrier’ clause operative, which purports to indicate that whatever impression the bill may give on its face, the contract is with the shipowner (or demise charterer) only. It is well known now that this clause was devised for the convoys of the Second World War, to make sure that the carriage contract was not with the charterer, who was likely to be the British Government, but with the owner, in order to secure the owner’s entitlement to tonnage limitation, which did not at that time apply to charterers.18 It has never been clear why such clauses survived the war, or the change in the rules regarding charterers’ right to limit, but they certainly have. Subject to the provisions of relevant legal systems, they may have the advantage of making the shipowner defendant to a contract claim, probably making arrest of the ship easier and facilitating investigation of the causes of any loss or damage while the ship and its crew are available at the port of discharge. The fairly recent leading case of The Starsin,19 which is discussed below, affirmed that where an indication that the contract was with the charterer was absolutely clear on the front of the bill a demise clause could be disregarded, or at least discounted. This may have been a good commercial decision, but the question of how clear the indication must be to justify ignoring clear wording in such a clause raises a further uncertainty beyond the initial inquiry as to who is the carrier. If the result of the inquiry is that the first party did not undertake carriage, ie that the bill of lading is not a charterer’s bill, it is likely that the subcontractor, probably the owner or demise charterer, did. This makes the subcontractor a carrier, very likely a Hague or Hague–Visby Rules carrier, and potentially liable on the contract of carriage accordingly. The first party’s liability is more limited. If it was a time charterer, its only functions would in principle lie in giving instructions to the ship. If it had some relationship with the shipper that could be regarded as contractual, this might be susceptible to interpretation as a contract to procure carriage.20 If the time charterer could be regarded as having possessed the goods, or perhaps having the right to possess them, it might be a bailee who sub-bails.

17 See,

eg R Aikens, R Lord and M Bools, Bills of Lading, 2nd edn (Abingdon, Informa Law, 2016) 183–92. Roskill, ‘The Demise Clause’ (1990) 106 LQR 403. Lord Roskill was involved in the preparation of the clause. 19 The Starsin (HL) (n 15). 20 See J Ramberg, ‘Time Charterer’s Liability against Bill of Lading Holders’ [1966] European Transport Law 874. 18 Lord

Do Actual Carriers Require Special Treatment? 29

B. Tort On this analysis, if the party actually carrying the goods is not the contracting carrier, the initial assumption of most legal systems would be that it can only be liable in tort, which (unless conversion is relevant) would require proof of negligence. This may not in the end be very different from the position in a contract claim, since the tort of negligence or its equivalent elsewhere provides a general action for damaging or causing loss of goods, and it can be said that negligence is the basis of most Hague Rules contract liability. In England there has been a recent analysis of the burden of proof in cargo claims under the Hague Rules,21 which makes heavy use of bailment reasoning, a separate line of inquiry not depending on privity of contract, to which we shall return. It is not clear how the case was pleaded, but the normal rules for proof of negligence claims against non-contracting parties probably remain, though there is also a possibility of bailment reasoning being applicable, as appears below. A more difficult part of the claim in tort is likely to be the establishment of title to sue in tort for such damage or loss, which on the common law as understood in England would be confined to the owner, the possessor or possibly the person with the immediate right to possess the property damaged or lost, at the moment of the damage or loss.22 Thus, in an older case, The Nea Tyhi,23 plywood in a vessel from Malaysia was found to be wet on arrival at Newport, Gwent. Evidence from the Meteorological Office was sought as to rain in the vicinity of South Wales as the vessel approached the port, at the time at which the property passed in the r elevant goods (which could in this particular case have been established). This is not convenient. Fortunately, the judge did not have to reach a conclusion on the issue. This requirement for title to sue in negligence, in fact still uncertain in England at the time of The Nea Tyhi, is, however, a limiting factor that may not be accepted in all common law jurisdictions.24 Another serious problem is that on general grounds it might be assumed that if the actual carrier is sued in tort, it would be entitled to at least the protections contained in its own subcontract, and preferably (since its own subcontract may be a charterparty, the provisions of which might not be useful in the circumstances) those conferred by the main contract under which the goods are to be carried, which would be likely to incorporate the Hague or Hague–Visby Rules. Most international carriers under bills of lading carriers might expect the benefit of the one-year time bar and the package or unit limitation, and as regards the actual carriage of the goods, the exceptions from liability conferred by Article IV.2(a)–(q) of the Rules. Some actual carriers might also expect the benefit of jurisdiction and arbitration clauses in their own contract with the contracting carrier, and possibly those in the main contract of carriage. But the doctrine of privity of contract at first blush prevents the carrier from using against a cargo claimant the protections or other relevant clauses in either its own or the contracting carrier’s contract. This would be to a claimant’s advantage, but actual carriers might seek to avoid such a result. Various techniques have been developed by the common law to deal with this problem for the actual carrier: in particular, the well-known Himalaya clause, which (to varying degrees, dependent on the drafting) purports to extend the terms and protections of the main bill of lading

21 Volcafe Ltd v Compania Sud Americana de Vapores SA (t/a CSAV) [2018] UKSC 61, [2019] AC 358. 22 Leigh & Sillavan Ltd v Aliakmon Shipping Co Ltd (The Aliakmon) [1986] AC 785 (HL); Aikens et al (n 17) 248–49. This was important in The Starsin (HL) (n 15), as will be discussed below. 23 The Nea Tyhi [1982] 1 Lloyd’s Rep 606 (QB (Admlty)). 24 It has been held not to be correct for Singapore: see NTUC Foodfare Cooperative Ltd v SIA Engineering Ltd [2018] SGCA 41, [2018] 2 SLR 588 (concerning loss of business by a food kiosk in a transit lounge at Changi Airport), followed in a maritime context in Wilmar Trading Pte Ltd v Heroic Warrior Inc [2019] SGHC 143.

30 Francis Reynolds contract to servants or agents of the carrier, words which would often cover actual carriers, and in some forms purports to exclude such parties’ liability altogether, and the doctrine of bailment on terms. Reference is made to these below.

IV. Reasoning Behind a Special Regime for Actual Carriers Returning to conventions and legislation which, unlike the common law, specifically impose liability on the actual carrier, we find that the reasoning which underlies them is largely that it is often difficult to know who within a contractual complex should be held responsible, however logical it may be to impose liability on one sole contractual carrier. In this context, the travaux préparatoires of the Hamburg Rules say: the issue is not who should bear the loss. Rather, the issue is establishing the most efficient mechanism to assure that the party who caused the loss should reimburse the cargo owner. In many cases the cargo owner cannot readily ascertain which of successive carriers was responsible for the loss. Indeed, the question may be in dispute among the carriers. The conventions governing road, rail and air reflect that view that it is more efficient for such questions to be settled among the carriers than to force the cargo owner to choose between (1) bringing simultaneous actions against different defendants and (2) running the risk that an initial action will fail on the ground that the wrong carrier was selected.25

V. Possible Solutions A. Both Liable The standard English approach makes little allowance for solutions other than the straightforward one of seeking to find ‘the’ contracting party and then, if necessary, exercising ingenuity to protect the actual carrier when sued in tort. However, some slight support for a looser approach can be gained from the law of agency. It is fairly clear under common law as understood in England that there can in appropriate cases be liability on a contract of both agent (the charterer who was authorised to sign for the owner) and principal (the owner who had given such a uthority).26 This idea was briefly raised by Rix LJ in the Court of Appeal in The Starsin,27 but was not pursued because it had not been formally argued and arose ‘merely from an inquiry from the bench’. The main, if unarticulated, reason for use of such reasoning in agency situations is probably the carelessness with which obligations are undertaken in commerce. It seems a reasonable generalisation that many persons enter into contractual relations without making clear whether they are acting on their own behalf or on behalf of others, and, if so, of whom: it is quite likely that the agents do not know, or at any rate understand, themselves.28 There must be situations where, if the agent has actual or apparent authority, the simplest answer is to conclude that the agent undertakes personal liability as well as his principal. This is especially so if the principal is not

25 See UNCITRAL Yearbook 1973, vol IV, Second Report of the Secretary-General on Responsibility of Ocean Carriers for Cargo: Bills of Lading (A/CN.9/76/Add.I) para 31. 26 See P Watts, Bowstead and Reynolds on Agency, 21st edn (London, Sweet & Maxwell, 2018) Art 98. 27 Homburg Houtimport BV v Agrosin Private Ltd (The Starsin) [2001] 1 Lloyd’s Rep 437 (CA) [70]–[76]. This case is discussed further below. 28 See, eg Isaac Cooke & Sons v Eshelby (1887) 12 App Cas 271 (HL) 272: ‘We had no belief on the subject.’

Do Actual Carriers Require Special Treatment? 31 identified. Since the connection is contractual, under this reasoning there can be no theoretical objection to the agent having the right to sue in addition to being liable, though situations where this would be relevant may not be common.29 The comparatively few clear cases in the modern context on the dual liability of principal and agent usually concern single-person companies or companies with subsidiaries, sometimes often in connection with repairs or supplies furnished to ships, where the purport of a signature is often not clear;30 but other situations must certainly be possible.31

B. Joint and Several Liability of All Relevant Parties If it is valid, does this go far enough? Professor Tetley, in his book Marine Cargo Claims,32 argued for a broader common law approach, stating: The underlying proposition of this text is that the defendant under the Hague and Hague–Visby Rules is (i) the shipowner (and the ship) and (ii) the contracting carrier and (iii) the actual carrier (i.e. the person or persons who carry out the responsibilities of the carrier under arts 2, 3 and 4 of those Rules); in other words, all three parties are jointly and severally responsible.

The basis of this is a slightly mysterious ‘transnational approach’ which presupposes recognition of one basic principle; namely that the Rules are intended to establish a delicately balanced allocation of rights and responsibilities, duties and obligations, as between cargo and carrier – an allocation which cannot be contracted out of by either party, particularly the stronger bargaining party (normally the carrier).33

This is obviously to be accepted. But the text at an earlier point says that To agree that any carrier who does not actually sign the bill of lading is not a carrier despite his responsibilities under the Hague Rules would be to permit an easy method of avoiding the Hague Rules, which would be contrary to art.III.8.34

The way in which this is achieved is usually by invoking further variations of the law of agency: ‘the charterer who issues a bill of lading also contracts for any other charterer and for the shipowner who have responsibilities under the Rules’.35 A consequence of this is that demise or identity of carrier clauses come in for strong criticism from Tetley on the basis that they ignore the reality that carriage by sea is usually a joint venture of ship, owner and charterer(s), and they also are blatant attempts to evade the overriding public policy enshrined in art.III.8, a rule that the drafters of 1924 enacted in the interest of justice and fairness.36

29 There are cases involving auctioneers. See Bowstead and Reynolds (n 26) para 9-008. 30 Bridges & Salmon Ltd v Owner of The Swan (The Swan) [1968] 1 Lloyd’s Rep 5 (PDAD); Stag Line Ltd v Tyne Ship Repair Group Ltd (The Zinnia) [1984] 2 Lloyd’s Rep 211 (Com Ct) (ship repairs); Savills (UK) Ltd v Blacker [2017] EWCA Civ 68 (estate agency); Air Tahiti Nui Pty Ltd v McKenzie [2009] NSWCA 429 (principal and subsidiary companies: air carriers). 31 For a case where the contract expressly made both agent and principal liable, see Middle East Tankers & Freighters Bunker Services SA v Abu Dhabi Container Lines PJSC [2002] EWHC 957 (Comm), [2002] 2 Lloyd’s Rep 643. 32 W Tetley, Marine Cargo Claims, 4th edn (Quebec, Thomson/Carswell, 2008) vol 1, 566. 33 ibid 645. 34 ibid 568. Art III.8 is the provision which renders void clauses which relieve the carrier of, or lessen in its favour, duties and obligations imposed by the Rules. 35 ibid 568. 36 ibid 646.

32 Francis Reynolds Thus, ‘the charterer who does not issue the bill of lading will still be a proper defendant if it undertakes some of the responsibilities referred to in the Hague and Hague–Visby Rules’.37 Tetley in fact spent some 46 pages attacking the demise or identity of carrier clause. The clause is certainly a trap which has caught many lawyers, in that it tends to appear in small type on the back of a bill of lading; and if it has advantages in facilitating actions against s hipowners, these were not part of its original purpose. As Rix LJ pointed out in the Court of Appeal in The Starsin,38 it is not a cargo owner’s clause on a cargo owner’s form but usually a liner company’s clause put in for its own purposes. Nor is there any attempt to give it paramount status. The argument that the clause is contrary to the Hague Rules and so caught by Article III.8 is stronger as regards the original contracting parties, where it can be said that it has the effect of negating the existing liability of one of them, than in the case of a transferee of the bill of lading, who can only go on what appears on the face and reverse of the document.39 But Tetley’s argument seeks to be much more powerful than that: it is that a clause such as that referred to proceeds on the basis that one of the parties involved in the carriage is the only contracting party. This is precisely contrary to his idea that there are several parties sharing duties under the arrangement for carriage, all of whom share joint and several responsibility. The contravention of the Hague Rules argued for by Tetley is something much more far-reaching, and seems to be based on a slightly mystical notion that the Hague and Hague–Visby Rules were intended as a complete regime carrying with it overriding public policy; and further, that the regime envisages more than one party being liable in respect of a contract for carriage by sea. Leaving aside demise and identity of carrier clauses, the difficulty with Tetley’s approach is that the supporting case law cited on liability as carrier, mostly from Canada and the USA, comes out rather variegated, any one case requiring careful study of differences in documentation and analysis before a final view can be formed.40 In the result, it is difficult to avoid the conclusion that the view that all parties are jointly and severally responsible is largely the result of an assessment that consistency in the application of normal contractual principles in this area is impossible to achieve, leading to the result that all parties participating in the carriage should be liable. This is a pragmatic approach, but the argument that the problem was envisaged in the Hague and Hague–Visby Rules and in some way is evidenced by Article III.8 is more difficult to accept. All of this becomes even more difficult to accept where, as sometimes happens, documents are used that give the clear impression that only one party is the contractor. As Lord Millett said firmly in the House of Lords in The Starsin: All the provisions of the bills, both back and front, indicate the existence of a single carrier, though they do not agree upon its identity. Two inconsistent provisions cannot be reconciled by adopting a construction which is consistent with neither.41

Likewise, Lord Hoffmann said: ‘I do not think that any reasonable merchant or banker who might be assumed to be the notional reader of this bill of lading would imagine that there was more than one carrier.’42

37 ibid 580. 38 The Starsin (CA) (n 27) [66]. 39 See Ngo Chew Hong Edible Oil Pte Ltd v Scindia Steam Navigation Co Ltd (The Jalamohan) [1988] 1 Lloyd’s Rep 443 (Com Ct). 40 See Tetley (n 32) 566 ff. 41 The Starsin (HL) (n 15) [190]. 42 ibid [85].

Do Actual Carriers Require Special Treatment? 33

C. Bailment At common law, however, there may be another way of fixing non-contractual liability on an actual carrier. The unique common law concept of bailment, which antedates in common law the formal recognition of the distinction we now know between contract and tort, can possibly be deployed to make the actual carrier liable under its reasoning. The specific part of that doctrine relevant here would be the notion of bailment on terms. It was by this, it appears, that Lord Hobhouse, who was always aware of the possibilities of bailment reasoning,43 thought that the problem addressed as the second issue in the leading case of The Starsin should have been resolved.44 In that case, the judge of first instance, Colman J, and one member of the Court of Appeal, Rix LJ, thought that the bill of lading involved in a shipment of timber and timber products from Malaysia to Northern Europe was, by virtue of the clear words accompanying the signature on it, a charterer’s bill of lading, despite the presence of both demise and identity of carrier clauses (inadequately transcribed) on the reverse. Two members of the Court of Appeal, no doubt following the notion that documents should be read as a whole, decided that it was a ship’s bill, creating a contract of carriage with the owner, Agrosin Pte Ltd.45 But four members of the House of Lords subsequently decided that the clear wording on the face made it a charterer’s bill.46 The result was that a claim against the actual carrier, Agrosin, which would have been straightforward if Agrosin had been also the contracting carrier, became more difficult. Had Tetley’s approach or perhaps agency reasoning such as that discussed above been openly espoused, there might have been no problem. As it was, only bailment might have provided some straight route to the actual carrier’s liability and Lord Hobhouse would plainly have preferred this route. This route was not followed by counsel, who may well have thought it too uncertain and risky, so the matter had to be handled under the normal principles of liability in the tort of negligence. The actual carrier Agrosin was able to start by pleading by way of defence a Himalaya clause which actually appeared in its first part to say that an actual carrier was not liable at all.47 To reach the result that the actual carrier could not rely on this part of the clause required at best an acceptance that Article III.8 of the Hague and Hague–Visby Rules, which render void attempts to reduce the carrier’s liability, somehow struck at that part of the Himalaya clause. But at common law, Himalaya clauses are normally interpreted as part of a collateral contract with the third party.48 (This might not be so in civil law countries, where it appears that the clause would invoke third party contract reasoning;49 to some extent this is true also at common law since the Contracts (Rights of Third Parties) Act 1999, which was not applicable in this case). The required results would probably have been fairly easy under Tetley’s approach. But without such

43 Especially in Aktieselskabet de Danske Sukkerfabrikker v Bajamar Compania Naviera SA (The Torenia) [1983] 2 Lloyd’s Rep 210 (Com Ct), cited in Volcafe (n 21) (discussed at n 57 below); see also Obestain Inc v National Mineral Development Corp Ltd (The Sanix Ace) [1987] 1 Lloyd’s Rep 465 (Com Ct). 44 See The Starsin (HL) (n 15) [137], [138]. 45 The Starsin (CA) (n 27). 46 The Starsin (HL) (n 15). 47 Lord Roskill claimed to have been the draftsman of the original Himalaya clause as well as the demise clause, but never disclosed in print why this first part of the clause appeared in it (as it probably did). He did, however, refer orally to his appearance on the losing side in Scruttons Ltd v Midland Silicones Ltd [1962] AC 446 (HL) as ‘braving the wrath of Lord Simonds for several days’. 48 New Zealand Shipping Co Ltd v AM Satterthwaite & Co Ltd (The Eurymedon) [1975] AC 154 (PC); Port Jackson Stevedoring Pty v Salmond & Spraggon (Australia) Pty (The New York Star) [1981] 1 WLR 138 (PC). 49 See Smeele, Les Règles de Rotterdam (n 14) 124, fn 51.

34 Francis Reynolds reasoning, a subtle formulation of the collateral contract arising under a Himalaya clause was required. Lord Hobhouse’s route to this, involving an analysis of the similar ‘Barwick contract’ in The New York Star,50 is perhaps the most complete: the other three members of the tribunal had their own routes, none of which is free from problems. In particular, two of the Lords51 relied on a ‘deeming’ provision in the relevant Himalaya clause to the effect that servants or agents of the carrier were ‘to this extent deemed to be parties to the contract contained in or evidenced by [the] bill of lading’. But the extent to which parties to a contract can by a contract term effectively ‘deem’ a person who is not a party to the contract (an actual carrier) to be such must surely still be arguable. The second part of the clause entitled the actual carrier to the benefit of the Hague (or Hague–Visby) Rules, which was not controversial; and as only one of the claimants could establish title to sue the actual carrier’s final liability, as limited by the Rules, was not great. The route to a satisfactory, indeed obvious, result on this point proved not to be easy; and there is little prospect of legislating to determine the wording to be used in Himalaya clauses. Would bailment reasoning have provided a more satisfactory and direct answer? The Starsin involved the issue of bills of lading by a time charterer. These were held in the end to constitute the contract of carriage between shipper and time charterer. Lord Hobhouse’s analysis, following that of Lord Goff in The Pioneer Container,52 was that the contracting carrier, the time charterer Continental Pacific, was a bailee for reward, who had authority to sub-bail (which would be normal, but was apparently not in issue in the Privy Council) by the terms of the bill of lading, and to subject the goods to the sub-bailee’s terms. The actual carrier (the shipowner Agrosin) was a sub-bailee carrying the goods on its own terms. It would seem that Lord Goff himself might even have regarded the bailment as a direct one by the shipper to the actual carrier, but on the facts Lord Hobhouse was prepared to accept a bailment and sub-bailment. What were those terms? The sub-bailee’s terms were those of a time charterparty. In The Pioneer Container, the relevant term in them was a Hong Kong jurisdiction clause contained in the charterparty. Once incorporation had been accepted, there was not much difficulty in regarding the clause as relevant. But in many cases, including The Starsin, the shipowner would wish to rely on some clause actually relating to the carriage. At least unless the charter incorporated the Hague or Hague–Visby Rules, a suitable clause for the purpose might not appear in a time (or, indeed, voyage) charter, whichever the contract was. For either line of reasoning, sub-bailment or direct bailment, a key feature in The Starsin was therefore the fact that, in Lord Hobhouse’s words, ‘the time charter stipulates, for the protection of the shipowners, that the Hague Rules are to be incorporated into the contract between the shipowners and the time charterers’.53 This saved the need for a more elaborate argument that the same bailment reasoning could be used to incorporate terms, not of the actual carrier’s contract, but directly of the initial contract of carriage between the shipper and the contracting carrier under a charterer’s bill of lading, an

50 The ‘Barwick contract’ refers to the analysis of Barwick CJ in the High Court of Australia in the same case, Port Jackson Stevedoring Pty v Salmond & Spraggon (Australia) Pty (The New York Star) [1978] HCA 8, [1979] 1 Lloyd’s Rep 298, 304–05, 308–10. 51 Lord Hoffmann and Lord Millett. In Whitesea Shipping & Trading Corp v El Paso Rio Clara Ltda (The Marielle Bolten) [2009] EWHC 2552 (Comm), [2010] 1 Lloyd’s Rep 648, Flaux J analysed the speeches in The Starsin in detail and came to the conclusion that the ‘deeming’ explanation was the majority view of the House of Lords (including that of Lord Hobhouse himself). 52 Owners of Cargo Lately Laden on Board the KH Enterprise v Owners of the Pioneer Container (The Pioneer Container) [1994] 2 AC 324 (PC). 53 The Starsin (HL) (n 15) [137].

Do Actual Carriers Require Special Treatment? 35 argument that was nearly but not quite accepted by Lord Goff in The Mahkutai.54 In that case, an actual carrier sued by the goods owner claimed the benefit of an Indonesian choice of law and jurisdiction clause in the original time charterer’s bill of lading, a contract with which the actual carrier had no obvious connection apart from the fact that all the parties to the carriage were Indonesian. Unfortunately for the actual carrier, the bill of lading also contained a Himalaya clause the terms of which were held specifically not to cover a jurisdiction clause, and therefore to be inconsistent with the idea that the actual carriers received the goods subject to the jurisdiction clause. This draws attention to the risks of drafting. The result was to exclude the clause in question from being incorporated: it would seem that otherwise the bailment reasoning might have been extended to take in provisions of the main bill of lading contract. Thus Lord Goff quoted from the notoriously controversial Elder Dempster case, where cargo was delivered directly to an actual carrier to whom the ship was under time charter, and Lord Sumner said that ‘The obligations to be inferred from the reception of the goods for carriage … amounted to a bailment on terms, which include the exceptions and limitations of liability stipulated in the known and contemplated form of bill of lading’.55 Lord Hobhouse therefore suggested that The Starsin could have been decided on the basis of bailment on terms reasoning, and furthermore that the cargo owner could have gained an additional benefit from it: ‘[the cargo owner] could, by relying on the goods owner/sub-bailee relationship, have put the burden on the shipowners to excuse their failure to deliver the goods undamaged’.56 Reliance on such reasoning as to burden of proof might at the time have been hazardous: there is much law on the burden of proof in cargo disputes, and only some of it uses bailment reasoning. However, the law has subsequently been reconsidered in a way that makes it more attractive to argue bailment reasoning. In the recent case of Volcafe Ltd v Cia Sud Americana de Vapores,57 such reasoning is clearly adopted in the Supreme Court by Lord Sumption in respect of a claim under the Hague Rules, though, as mentioned above, it is unclear how the claim was pleaded. Although the point is controversial and such reasoning might not be adopted in other jurisdictions, for example Australia, where there are High Court decisions from which different conclusions can be drawn, there must now be a prospect of successful argument along these lines in England and Wales. This would remove one problem. Should this bailment reasoning be accepted as generally available, the need for dealing with the position of the actual carrier by means of convention or statute (or, indeed, by detecting joint liability or using looser reasoning, such as that adopted by Tetley) could be diminished. The problem is that there are still plenty of uncertainties.58 There can be complications where there is a Himalaya clause inconsistent with the terms of the actual carrier’s contract with the contracting carrier. The normal view is that it is permissible to choose to rely on one’s own bailment reasoning, as was accepted in The Rigoletto,59 but whether one can chose the Himalaya clause instead is not clear.60 The differences between the duties of a bailee (which can be positive) and those imposed by the general law of negligence nevertheless remain. The application of the doctrine

54 The Mahkutai [1996] AC 650 (PC). 55 Elder, Dempster & Co Ltd v Paterson, Zochonis & Co Ltd [1924] AC 522 (HL) 564; a case now rehabilitated after much hostile comment in former years. 56 The Starsin (HL) (n 15) [137]. 57 Volcafe (n 21). 58 See Aikens et al (n 17) paras 9.33–9.83. 59 Lotus Cars Ltd v Southampton Cargo Handling plc (The Rigoletto) [2000] 2 All ER (Comm) 705 (CA). 60 See The Starsin (HL) (n 15) [136] (Lord Hobhouse).

36 Francis Reynolds requires consent by the head bailor to the bailee’s subjecting the goods to the sub-bailee’s terms, and although such consent may be specified in wide terms in, or inferred as regards normal terms from, the initial contract, some terms may be unusual, particularly as regards jurisdiction and arbitration clauses. It is also said that the sub-bailee must be aware that someone other than the intermediate bailee is interested in the goods, so as to justify the sub-bailee’s assumption of responsibility;61 it is not clear how this is to be established. Where the contracting carrier is a time charterer, who prima facie has no interest in the goods other than to give orders to the ship carrying them, there may be a question whether the time charterer ever had sufficient relationship with the goods to create a bailment, and/or whether it had some sort of additional relationship with the shipper whereby it promised to procure carriage of the goods.62 It should be borne in mind that although section 6(5) of the Contracts (Rights of Third Parties) Act 1999 (which came into effect after the facts in The Starsin) now solves some general problems of incorporation of the Hague and Hague–Visby Rules terms by Himalaya clauses, it applies only to exclusions or limitations of liability, and so does not cover other privity problems. The liability of both relevant parties has never been fully argued. The solution put forward by Tetley is (with respect) rather loose. The contortions manifested in The Starsin to deal with Himalaya clauses that go too far by creating a complete exemption from liability inappropriate to actual carriers (even if acceptable for other subcontractors, such as stevedores) are a confession that drafting cannot be controlled and that the law on its own, while seeking only to implement the intentions of the parties, can reach inappropriate results.63 Lord Bingham recognised this in The Starsin, when he said that any other decision ‘would be to elevate form over substance and to invest what is essentially a legal device with a wholly disproportionate legal significance’.64 The bailment solution has received a recent impetus as regards burden of proof from the Supreme Court in the Volcafe case, and on its face presents the best escape route. However, there is a great deal to work out and bailment may not give appropriate results in all situations.

VI. Does the Problem Require a New Approach? In the result, it may be that legislation would help in a situation where too much intellectual energy has to be spent in seeking to provide a logical answer to a practical problem of compensation recognised as serious, and sought to be solved more directly by other systems of law. How is one to justify this, other than by admitting a weakness in the normal techniques of legal analysis that is not perceived as requiring the intervention of statute in other areas of law – for example, in construction contracts? It is plain that in this latter area, and in other commercial areas, subcontractors can cause considerable problems, not only of liability, but also of right to sue. The main device relied on here seems to be that of a collateral contract. However, in the area of construction law, there are not documents passed from hand to hand, nor (necessarily) international implications, nor a background of international conventions. It seems that the best course may be to accept that in international carriage by sea there are so many conflicting problems of

61 See The Pioneer Container (n 52) 341. 62 See eg Aikens et al (n 17) para 9.36(1). 63 See F Reynolds, ‘Tangling in the Undergrowth’ in PS Davies and J Pila (eds), The Jurisprudence of Lord Hoffmann (Oxford, Hart Publishing, 2015). 64 The Starsin (HL) (n 15) [34].

Do Actual Carriers Require Special Treatment? 37 ocumentation and their drafting, and of the potential interactions between different national d legal systems, that some intervention is both practical and desirable. As already pointed out, there are provisions in the transport law of other European countries.65 It does not seem likely, however, that a case can be made in the UK for legislative reform on such a topic. As the experience with the Carriage of Goods by Sea 1992 showed in a context more clearly needing reform, it is not easy to secure the passage of an isolated piece of private law reform that appears undramatic if nevertheless of considerable importance commercially. The only likely prospect would seem to be the adoption of a relevant international convention, as happened many years ago in air law and the law relating to land transport. It is in fact only in maritime transport that such steps have not yet been taken in the UK. That is presumably because the successful convention66 which created the Hague Rules was adopted at Brussels in 1924, and is still valid, subject to two protocols for those states that have adopted them. It appears that the conventions on air transport were pathfinders in this area. The Montreal Convention provides that both the contracting carrier and the actual carrier are subject to its rules, the former for the whole of the carriage contemplated in the contract, the latter solely for the carriage which it performs. In the result, a claimant can sue either or both, though where one is sued it may require the other to be joined. This type of carriage is different from what the Montreal Convention calls ‘successive carriage’, where a carriage to be performed by successive air carriers is regarded by the parties as a single operation: there, in the case of baggage or cargo, the consignor will have the right of action against the first carrier and the consignee will have the right of action against the last carrier, and each may sue the carrier that performed the carriage during which the loss or damage occurred on the basis of joint and several liability.67 This slightly unusual position is not the same as that for passengers. Maritime transport takes longer, and the problems of documentation are consequently more extended. The Hamburg Rules sought to provide a solution, but are perceived as having drawbacks in this and other respects, and have not attained sufficient currency: although they have been used in this essay by way of example, it is not suggested that they provide any more than a text for discussion. It is said, for example, that they do not make clear whether they cover more layers of actual carriers than the first;68 that they retain an unsatisfactory exception for ‘through bills of lading’, which was itself a compromise;69 and that they do not protect the servants and employees of actual carriers. As has been stated, the Rotterdam Rules70 deal with the problem, and, as already mentioned, they add to the liability of the actual carrier a joint and several liability of other ‘maritime performing parties’ operating within the port area, such as stevedores and firms moving the cargo around within the port. This is consistent with the extended objective of covering situations where the goods are only carried partly by sea, though performing parties operating outside the port area are not affected. They also expressly provide that there is no liability on the master or crew of a ship, an employee of the carrier or of any maritime performing party.

65 See Smeele, Les Règles de Rotterdam (n 14) 118, n 20. 66 In which the problem hinted at in Elder, Dempster & Co (n 55) was not addressed. 67 Montreal Convention, Art 36. 68 See above, text to n 6. 69 See E Selvig, ‘Through Carriage and On-Carriage of Goods by Sea’ (1979) 27 American Journal of Comparative Law 369, 383. 70 See, in general, Sturley et al (n 4) 138–52.

38 Francis Reynolds

VII. Two Further Issues A. Delay One reason for caution in this context is that both the Hamburg Rules and the Rotterdam Rules provide for limited liability in respect of delay, which could be something new. The Hamburg Rules require delivery by the carrier within a time expressly agreed upon, or in the absence of such agreement in (what may be paraphrased as) a reasonable time.71 Agreement on the point proved more difficult under the Rotterdam Rules, which only require delivery within the time agreed,72 and both sets of rules contain a limit on such damages calculated by reference to the freight.73 Both provisions are worded in such a way that this liability attaches also to the actual carrier as regards matters occurring during the part of the carriage performed by it, and under the Rotterdam Rules the application to maritime performing parties in general could make businesses, some of whom may have quite limited roles, liable for delay under terms of which they may be ignorant. It is surprising that actual carriers and (especially) maritime performing parties are subject to this liability.74 Both sets of rules provide that if the carrier undertakes a greater liability than the rules require, the actual carrier or maritime performing party is not liable unless it expressly accepted the additional requirements. But the main duties of the carrier are in both cases expressed to include the delay liability.

B. Mode of Operation It must also be borne in mind that these provisions impose a liability on the actual carrier which is non-contractual and which would require authority from any enacting statute. This is not unusual: the Warsaw Convention was recognised a long time ago as being a partial codification of the law.75 The Hague–Visby Rules have the force of law, but do not purport to provide for the present issue. A further problem arises in respect of actions by consignees, for the wording of present UK legislation, the Carriage of Goods by Sea Act 1992, would create no action against an actual carrier as defined above. The question of the nature of the action may well be relevant to questions of choice of law and jurisdiction. It is true that both conventions themselves contain provisions on jurisdiction (and arbitration); however, the Rotterdam Rules can be adopted without these provisions, and if they are adopted, it seems likely that the appropriate reservation would be made by the UK.

VIII. Conclusion Overall, the provisions of the Rotterdam Rules are clearly better thought through on the basis of previous experience, and if any new solution for actual carriers is to be adopted by means of an 71 Art 5.2. 72 Art 21. 73 Hamburg Rules, Art 6.1(b); Rotterdam Rules, Art 60; and there are restrictive provisions as to notice of delay in Rotterdam Rules, Art 23 which may make liability difficult to establish. 74 See von Ziegler (n 6) 123. 75 See, eg Corocraft Ltd v Pan American Airways Inc [1969] 1 QB 616 (CA); Sidhu v British Airways Plc [1997] AC 430 (HL).

Do Actual Carriers Require Special Treatment? 39 international convention, they would be preferable to the Hamburg Rules. It is to be noted that HH Judge Anthony Diamond QC, in his magisterial discussion of the Rotterdam Rules, said that he exempted from his critique of the provisions on liability of the carrier ‘the provisions on performing parties and maritime performing parties, which in my view should be welcomed’.76 The Rotterdam Rules do require a step to be taken by acceptance of a convention not rigorously confined to sea carriage but taking in certain through transport elements. There are obviously varied arguments for and against adopting the Rotterdam Rules; however, little has happened in the 10 years since they were formulated. If they disappear into obscurity, there seems little left for common law on this topic beyond further refinement of the existing and highly uncertain techniques. This is a matter that needs thinking about.

76 A Diamond, ‘The Rotterdam Rules’ [2009] Lloyd’s Maritime and Commercial Law Quarterly 445, 489–91. Compare Smeele, Les Règles de Rotterdam (n 14) 141, who refers to the concept of the maritime performing party as no more than a ‘moderate step forward’.

40

3 Multimodal Carriage of Goods by Sea: Time for an International Convention? SIMON BAUGHEN

I. Introduction Containerisation has profoundly changed the world of international maritime transport over the last 60 years. Goods stowed in a container can be carried by different modes of transport from the place where they are taken over by the carrier to the final place of destination, without being unpacked for sorting or verification when being transferred from one means of transport to another. In terms of value, global seaborne container trade is believed to account for approximately 60 per cent of all world seaborne trade, which was valued at around 12 trillion US dollars in 2017, with the quantity of goods carried by containers rising from around 102 million metric tons in 1980 to about 1.83 billion metric tons in 2017.1 The computer on which I write this chapter will probably have originated in a factory in inland China. Its journey to me in Swansea will encompass varying modes of carriage performed under a single contract of carriage, crossing continents and crossing unimodal carriage conventions as it does so. From the factory in Wuhan in inland China, where it is packed into a container, it will be carried by road to Shanghai, where the container will be loaded onto a feeder vessel taking it to the great container hub in Singapore. There it will be loaded onto another vessel for the voyage to Marseilles. A bill of lading will be issued noting the reception of the container in apparent good order and condition and saying that it contains so many packages containing computers.2 At Marseilles it will be loaded onto a lorry, which will take it to Southampton via the ferry at Calais. At Southampton, the container will be unloaded from the lorry and stored at the terminal, to be picked up later by another lorry to take it to Swansea. This journey involves multiple legal regimes – domestic law applying to the first and final road legs and the feeder sea leg from China; the Hague–Visby Rules applying to the sea leg from Singapore; and the Convention on the Contract for the International Carriage of Goods by Road (Geneva, 19 May 1956) (CMR)3 applying to the road carriage, involving roll-on, roll-off at the ferry at Calais, from Marseilles to Southampton. 1 Statista Research Department, ‘Container Shipping – Statistics & Facts’, 10 October 2018; see www.statista.com/ topics/1367/container-shipping/. 2 The effect of this clausing is that the statements as to what is inside the container will have no evidential effect, and in the event of damage or loss being apparent after final delivery of the container, it will be for the consignee to provide evidence as to the condition and quantity of the cargo loaded on board the vessel. 3 See n 29 below.

42 Simon Baughen It is more than likely that my computer’s journey, crossing continents and crossing legal regimes attaching to the different modes of performance, will be subject to a single contract of carriage from door to door. Seller and buyers – with containerised cargo it will almost always only be these two parties – and a financing bank, operating under a sale contract usually on Free on Board (FOB) or Free Carrier (FCA) terms, will generally prefer to deal with a single party who arranges for the transportation of goods and assumes contractual responsibility throughout. Enter the multimodal contract of carriage, contained in or evidenced by a multimodal bill of lading or sea waybill. The carrier under such contracts will undertake responsibility for the whole transport chain under one single transport contract. This contrasts with ‘through transport’,4 in which a carrier accepts responsibility for only that part of the total movement performed with its own facilities, but also undertakes to arrange any additional leg acting as forwarding agent, on behalf of the shipper, in which capacity its duty is to select the on-carrier and other service providers with due care. Although there are no internationally accepted definitions for the terms used in multimodal contracting, in most cases the terms are understood as follows. Multimodal transport (or combined transport or integrated transport) is where goods are delivered door to door, by a combination of at least two transport modes – eg road/rail/sea/air. It will involve a multimodal transport operator (MTO), who assumes contractual responsibility as a principal throughout, irrespective of whether it is also the party who actually performs the different stages of the transport.5 The MTO may actually perform only part of the carriage, such as the sea carriage, or may be a non-vessel-owning carrier (NVOC) and actually perform no part of the carriage. The MTO will issue a transport document, a multimodal transport bill of lading (MT Bill) or a sea waybill (MT Sea Waybill) to the shipper which covers the entire transport operation. The MTO will also take, as cargo owner, separate bills or transport documents – such as a CMR consignment note – from its subcontracting carriers. If the cargo is lost, damaged or delayed, the cargo interests will sue the MTO in contract and the MTO will in turn exercise a right of redress against its negligent subcontractor(s) under its separate contract with them, to which the cargo interests will not be party. The MTO will also wish to protect itself from indemnity claims from its subcontractors in the event that the shipper or consignee brings a non-contractual action against them. This will be achieved by the inclusion in the multimodal transport document of the Himalaya and circular indemnity clauses that are ubiquitous in ocean bills of lading. This means that two bills of lading or waybills will be issued: the multimodal bill of lading issued to the shipper by the multimodal carrier; and a bill of lading issued by the sea carrier to the multimodal carrier as shipper for the sea leg or the relevant transport document for the final inland leg by another means of transport. The first of these will be used under the contract of sale and any letter of credit opened to finance the sale. Those entitled to delivery of the cargo will obtain delivery at the place of delivery by presenting the multimodal bill to the actual carrier on the final leg. Delivery of cargo at the port of discharge which is carried under the sea carrier’s

4 Or ‘intermodal transport’, as it is referred to in the USA. 5 The identity of ‘the carrier’ or the ‘multimodal transport operator’ will usually be expressly defined by the standard multimodal bill of lading forms. COMBICONBILL 1995 defines ‘carrier’ in cl 2 as ‘the party on whose behalf this bill of lading has been signed’. MULTIDOC 1995 defines multimodal transport operator in cl 2 as ‘the person named on the face hereof who concludes a Multimodal Transport Contract and assumes responsibility for the performance thereof as a carrier’.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 43 ocean bill will be through presentation of that bill by the shipper or consignee – the multimodal carrier. Problems may arise through the existence of two parallel bills of lading but can be averted by making out the second bill of lading to the order of the first carrier.6 There have so far been two attempts at a convention for multimodal carriage: the 1980 Geneva Convention on International Multimodal Transport of Goods (the 1980 MT Convention) and the 2008 United Nations Convention on Contracts for the International Carriage of Goods Wholly or Partly by Sea (the Rotterdam Rules). The former is dead in the water, and the latter is going that way, having obtained to date only four of the required 20 ratifications necessary for its coming into force. In this chapter I shall consider five problems presented by multimodal transport with a view to assessing whether there is a need for a new convention to regulate this form of international carriage of goods, or whether these problems are adequately resolved by the standard form contractual solutions that have been adopted, such as the two BIMCO bill of lading forms for multimodal carriage, COMBICONBILL and MULTIDOC.

II. Five Problems with Multimodal Carriage of Goods Several potential problems can be identified with multimodal carriage of goods: whether a multimodal bill of lading can function as document of title; whether a third party can obtain title to sue under the contract of carriage contained in or evidenced by the multimodal bill of lading or waybill; whether the unimodal carriage conventions apply to the relevant international carriage elements of a multimodal contract; and, if so, the difficulties in applying the divergent provisions of the unimodal conventions to different segments of the multimodal contract.

A. The Multimodal Bill of Lading as a Document of Title There is some uncertainty as to whether a multimodal bill of lading constitutes a document of title – a document whereby property7 in and possession8 of the goods represented by the bill can pass during the sea voyage by negotiation, without any attornment from the carrier. Such a bill will generally be a ‘received for shipment’ bill of lading. There is no clear authority as to whether such a bill of lading can constitute a document of title. Lloyd J, in The Lycaon,9 assumed that it could be such a document, relying on the decision of the Privy Council in The Marlborough Hill.10 However, the point in issue in that case was the definition of ‘bill of lading’ for the purposes of a statute conferring jurisdiction on the Admiralty Court. Diamond Alkali Export Corp v Bourgeois11 is claimed to be the authority for the contrary proposition that such bills cannot be documents

6 Alternatively, the first carrier could release the second bill to the party holding the first carrier’s bill on surrender by that party of the first carrier’s bill of lading to the first carrier. That party would then present the second bill to obtain delivery. This is what happened in Carrington Slipways Pty Ltd v Patrick Operations Pty Ltd (The Cape Comorin) (1991) 24 NSWLR 745, where a freight forwarder issued a bill of lading to the shipper and then a second bill of lading was issued by a time charterer to the forwarder. 7 Lickbarrow v Mason (1794) 5 TR 683, 101 ER 380. 8 Meyerstein v Barber (1869–70) LR 4 HL 317. 9 Elder Dempster Lines v Zaki Ishag (The Lycaon) [1983] 2 Lloyd’s Rep 548 (Com Ct). 10 Owners of the Ship ‘Marlborough Hill’ v Alex Cowan & Sons Ltd (The Marlborough Hill) [1921] 1 AC 444 (PC). 11 Diamond Alkali Export Corp v Bourgeois [1921] 3 KB 443 (KB).

44 Simon Baughen of title. Again, the issue never directly arose in the case, which merely decided that these documents did not amount to good tender under a Cost, Insurance and Freight (CIF) sale, as they would not evidence the condition of the goods on shipment, when risk passed. A received for shipment bill can be turned into a ‘shipped’ bill of lading by annotating it with the date of shipment. It will then be a shipped bill and will then certainly constitute a document of title. The fact that the bill of lading provides for delivery at an inland location will not prevent it being a document of title for the duration of the sea transit,12 although it may not continue to be a document of title for the final land transit.13 A further possible objection to a multimodal bill of lading amounting to a document of title is that it may well be issued by someone other than the actual sea carrier, such as a freight forwarder. In The Maheno,14 such a document was assumed not to be a document of title, although the actual decision proceeded on the basis that the forwarder had not contracted as the carrier. It has been doubted whether such a document can confer constructive possession in the goods to which it refers when the contractual carrier has never taken them into its physical possession. However, this objection applies equally to time charterers’ bills of lading, which are clearly accepted as documents of title, and also to through bills of lading as regards the second sea leg. It is submitted that what matters is not the physical reception of the goods by the carrier, but its contractual capacity to control delivery by the performing carriers. This is borne out by the finding in Spectra International plc v Hayesoak Ltd15 that a party could become a bailee of goods, even without taking physical possession of them, by obtaining a right to give directions to the warehouseman as to their delivery. Commercially, the contractual carrier does not actually need to be able to hand over the ‘key to the warehouse’; it should be enough that he can direct the party who does have the key as to when it should be turned. In practice, these uncertainties have not affected the widespread use of multimodal bills in international trade. Multimodal bills are acceptable documents for the purposes of Article 19 of UCP 60016 provided that they regulate all stages of the carriage, and it can be argued that this evidences a custom of merchants that a multimodal bill of lading constitutes a document of title. For CIF sales, there would be no problem with the tendering of a multimodal bill of lading, as it would satisfy the requirement of continuous contractual cover from receipt of the goods to their delivery.17

12 G Treitel and FMB Reynolds, Carver on Bills of Lading, 4th edn (London, Sweet & Maxwell, 2017) para 8.087, citing Re Sutro & Co v Heilbut, Symons & Co [1917] 2 KB 348 (CA) and Johnson v Taylor Bros & Co Ltd [1920] AC 144 (HL), although the point was not specifically addressed in either case. 13 The custom found by the merchant jury in Lickbarrow v Mason (n 7) was limited to ‘anytime after such goods have been shipped, and before the voyage performed, for which they have been or are shipped’. 14 The Maheno [1977] 1 Lloyd’s Rep 81 (NZ SC). A similar assumption was made in Carrington Slipways (n 6). 15 Spectra International plc v Hayesoak Ltd [1997] 1 Lloyd’s Rep 153 (Central London County Ct). The point was raised before the Privy Council in Owners of Cargo Lately Laden on Board the KH Enterprise v Owners of the Pioneer Container (The Pioneer Container) [1994] 2 AC 324 (PC) in relation to the Scandutch claimants who had been issued with a bill of lading by an NVOC who had never taken possession of the cargo. Their Lordships did not think it right to allow the point to be raised for the first time before them, but Lord Goff observed (ibid 345): ‘They wish to add however that, on the limited argument on this point which took place before them, it is difficult to see why the shipowners should not, when they received the goods of the Scandutch plaintiffs into their possession, have becomes responsible as bailees to the owners of the goods even if the goods were never in the possession of Scandutch.’ 16 ICC, Uniform Customs and Practice for Documentary Credits – UCP 600 (2007). This is subject to various conditions, principally that the MT Bill should: (i) indicate the name of the carrier and be signed appropriately; (ii) indicate that the goods have been ‘dispatched, taken in charge or shipped on board at the place stated in the credit’; (iii) indicate the place so stated in the credit as well as the place of destination there stated; and (iv) contain or refer to conditions of carriage. 17 Hansson v Hamel & Horley Ltd [1922] 2 AC 36 (HL).

Multimodal Carriage of Goods by Sea: Time for an International Convention? 45

B. Third Party Rights of Suit The multimodal bill will involve a shipper, who has made the contract of carriage with the MTO, and a consignee, who has not. But do multimodal bills of lading and sea waybills come within the ambit of the Carriage of Goods by Sea Act 1992 (COGSA 1992) so as to give rights of suit to the consignee? This is an issue that previously received some academic commentary in relation to the applicability of the Bills of Lading Act 1855 to through bills, with Bateson and Carver arguing that such bills were not included within the Act.18 The joint report of the Law Commission and the Scottish Law Commission19 which led to the enactment of COGSA 1992 gives no real guidance on this issue. The report noted that traders typically treat such documents as traditional bills of lading and that provision is made for tender under the Uniform Customs and Practice for Documentary Credits, and reported that they had received ‘no evidence from consultants that there are particular privity problems which are unique to combined transport bills of lading as distinct from the traditional ocean variety’.20 The report went on to state that ‘since implementing legislation is expressed to cover any bill of lading, including “received for shipment” bills, multimodal documents are capable of falling within its ambit.’21 COGSA 1992 provides no definition of a ‘bill of lading’, although section 1(2)(b) provides that the Act applies to received for shipment bills. However, it is likely that this requires at least a designation of the carrying ship and of receipt by the sea carrier. Carver on Bills of Lading also argues that at common law a bill of lading refers only to a document containing or evidencing a contract for the carriage of goods by sea – a fact bolstered by the title of the Act itself, the ‘Carriage of Goods by Sea Act 1992’.22 As against that, the definition of ‘contract of carriage’ in section 5(1) refers to the ‘contract of carriage contained in or evidenced by’ the bill of lading and makes no reference to that contract being by sea. If multimodal bills of lading do fall within COGSA 1992, the whole of the contract of carriage would be transferred, enabling the lawful holder to sue for loss or damage that occurred outside the period of sea carriage.23 In contrast, a ‘sea waybill’ is defined in section 1(3)(a) as ‘such a receipt for goods as contains or evidences a contract for the carriage of goods by sea’ (emphasis added), which would seem to exclude multimodal transport waybills from the scope of COGSA 1992. Against that is Quantum Corp Inc v Plane Trucking Ltd,24 in which the international road leg of a multimodal contract of carriage involving air carriage was held to be subject to CMR as involving ‘carriage by road’. On this basis, the sea leg of a waybill could be regarded as constituting a contract for the carriage of goods by sea, as regards the sea carriage element of the overall carriage. If that is the case, and multimodal bills of lading are not to be treated as bills of lading under the Act, they could come back in under the provisions regarding sea waybills.

18 HD Bateson, ‘Through Bills of Lading’ (1889) 5 LQR 424; TG Carver, ‘On Some Defects in the Bills of Lading Act 1855’ (1890) 6 LQR 289, 294. 19 Law Commission and Scottish Law Commission, Rights of Suit in Respect of Carriage of Goods by Sea (Law Com No 196 and Scots Law Com No 130, 1991). 20 ibid para 2.49. 21 ibid. 22 Treitel and Reynolds, Carver on Bills of Lading (n 12) para 8.081. 23 Alternatively, the Act might be construed so as to operate only as regards transfers of the contract of carriage that take place between shipment and completion of the sea voyage, during which period the document will be a document of title, provided there is a notation as to shipment. 24 Quantum Corp Inc v Plane Trucking Ltd [2002] EWCA Civ 350, [2002] 1 WLR 2678.

46 Simon Baughen However, it should be noted that in no case has the validity of the operation of the Bills of Lading Act 1855, and latterly of COGSA 1992, been challenged in cases involving a multimodal contract of carriage by sea evidenced by a bill of lading or sea waybill.25 The most recent of these is The Maersk Tangier,26 where Andrew Baker J considered that the receiver had title to sue either as a party to the original contract or as the person to whom delivery was to be made under a waybill pursuant to section 2(1)(b) of COGSA 1992. The title to sue point was not considered in the subsequent decision of the Court of Appeal, which affirmed the first instance decision.27 One suspects that there is a convention among carriers and P&I clubs not to take the point, but that does not mean that there is not a point to be taken. However, a simple way to enable the consignee to sue on the document would be by an assignment of rights by the shipper. If the multimodal bill falls outside COGSA 1992, then the shipper will have the right to claim damages for the consignee’s loss,28 and that right can be assigned to the consignee.

C. The Application of the Unimodal Conventions to the Various Modes of Carriage under a Multimodal Bill of Lading Multimodal carriage gives rise to the problem that one contract of carriage will often involve the application of two or more unimodal conventions. Apart from the three conventions for international sea carriage, there are unimodal conventions in force for international carriage of goods by road, rail, inland waterways and air.29 The conventions clearly apply to the contracts of carriage with the actual carrier, but what about the head contract for the multimodal carriage?

(i) Sea Carriage The Hague and Hague–Visby Rules apply to cross-border sea carriage between convention countries on a ‘tackle to tackle’ basis. This period may exceptionally begin prior to the loading of the goods on to the sea vessel. In Volcafe Ltd v Compania Sud Americana de Vapores SA,30 the sea carrier supplied the containers and loaded the cargo of coffee into them at an inland terminal. It was held that ‘loading’ under the Hague Rules commenced with the start of loading into the

25 For other cases involving such carriage, where the validity of the operation of the legislation was not challenged, see eg Pyrene Co Ltd v Scindia Steam Navigation Co Ltd [1954] 2 QB 402 (QB); Mayhew Foods Ltd v Overseas Containers Ltd [1984] 1 Lloyd’s Rep 317 (Com Ct). 26 AP Moller-Maersk A/S (t/a Maersk Line) v Kyokuyo Ltd (The Maersk Tangier) [2017] EWHC 654 (Comm), [2017] 2 All ER (Comm) 922 [32]. Two of the three contracts of carriage involved onward land carriage from Yokohama to Shizimu. There was no problem with the Hague–Visby Rules applying to the sea leg of these contracts, contrary to dicta of Flaux J in Bhatia Shipping & Agencies Pvt Ltd v Alcobex Metals Ltd [2004] EWHC 2323 (Comm), [2005] 2 Lloyd’s Rep 336. Flaux J was one of the Court of Appeal judges in The Maersk Tangier [2018] EWCA Civ 778, [2018] 2 Lloyd’s Rep 59. 27 The Maersk Tangier (CA) (n 26). 28 Under Dunlop v Lambert (1839) 6 Cl & F 600, 7 ER 824. 29 These are: Convention on the Contract for the International Carriage of Goods by Road (Geneva, 19 May 1956) (CMR); Appendix B to the Convention concerning International Carriage by Rail (COTIF) of 9 June 1999; Uniform Rules Concerning the Contract of International Carriage of Goods by Rail (CIM), applicable with effect from 1 July 2006 (COTIF/CIM); Budapest Convention on the Contract for the Carriage of Goods by Inland Waterways (Budapest, 3 October 2000) (CMNI); Convention for the Unification of Certain Rules Relating to International Carriage by Air (Warsaw, 12 October 1929) (Warsaw Convention); Convention for the Unification of Certain Rules for International Carriage by Air (Montreal, 28 May 1999) (Montreal Convention). 30 Volcafe Ltd v Compania Sud Americana de Vapores SA (t/a CSAV) [2016] EWCA Civ 1103, [2017] QB 915. The issue was not addressed in the subsequent decision of the Supreme Court ([2018] UKSC 61, [2019] AC 358), which reinstated the first instance decision on the burden of proof for the inherent vice exception in Art IV(2)(m) of the Hague Rules.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 47 containers and that the rules would continue to apply to the contract of carriage during the road leg to the port and during warehousing at the port prior to actual loading onto the ship.31 Article I(b) of the Hague and Hague–Visby Rules provides that ‘“Contract of carriage” applies only to contracts of carriage covered by a bill of lading or any similar document of title, insofar as such document relates to the carriage of goods by sea’ (emphasis added). This has been interpreted as a clarification that the Hague and Hague–Visby Rules apply to sea legs of a multimodal transport operation covered by a multimodal bill and not as a prohibition of the application of the rules to multimodal carriage.32 A similar finding has been made in Australia in Parlux SpA v M & U Imports Pty Ltd.33 This seems preferable to the brief dictum of Flaux J in Bhatia Shipping & Agencies PVT Ltd v Alcobex Metals Ltd,34 that ‘[s]ince the relevant contracts of carriage are MTDs, the Hague or Hague–Visby Rules and the 12-month time-limit under those Rules are of no application’.35 The Hamburg Rules apply to cross-border sea carriage between convention countries on a ‘port-to-port’ basis. The position as regards the application of the Hamburg Rules to the sea carriage element of a multimodal contract is made clear by Article 1(6): however, a contract which involves carriage by sea and also carriage by some other means is deemed to be a contract of carriage by sea for the purposes of this Convention only insofar as it relates to the carriage by sea.

(ii) International Carriage by Road, by Inland Waterways and by Rail The scope of the application of these conventions is largely determined by the existence of an international contract of carriage of goods involving the relevant mode of transport, where the place of taking over of the goods and the place of delivery are in two different states, and one36 or both37 of those states are parties to the convention. CMR applies only to cross-border carriage by road, mainly in Europe. A carrier under CMR need not be an actual road carrier and can be an NVOC who is contracting as a carrier.38 Article 1(1) provides: This Convention shall apply to every contract for the carriage of goods by road in vehicles for reward, when the place of taking over of the goods and the place designated for delivery, as specified in the contract, are situated in two different countries, of which at least one is a contracting country, irrespective of the place of residence and the nationality of the parties.

31 The possibility of the parallel application of the Hague Rules and the CMR Convention was raised, but Flaux J in the Court of Appeal was of the view that such concerns were more apparent than real (Volcafe (CA) (n 30) [110]): ‘The CMR Convention could not apply whilst the containers were being dressed and stuffed, since they were not at that stage on board a vehicle. Furthermore, the better view would seem to be that the CMR Convention ceases to apply as soon as “loading” within the Hague Rules begins: Clarke, International Carriage of Goods by Road, 6th edn (2014) pp 40–42.’ 32 See Pyrene v Scindia (n 25) and Mayhew Foods (n 25) 320, where the Hague Rules and the Hague–Visby Rules, respectively, were held to apply to the sea carriage part of the contract of carriage. 33 Parlux SpA v M & U Imports Pty Ltd [2008] VSCA 161, (2008) 21 VR 170. 34 Bhatia Shipping & Agencies PVT Ltd v Alcobex Metals Ltd [2004] EWHC 2323 (Comm), [2005] 2 Lloyd’s Rep 336 (QB). 35 ibid [21]. The case involved a bill of lading which was subject to the provisions of the applicable law of the country where the loss or damage occurred. The bill of lading did not specifically incorporate the Hague or Hague–Visby Rules and the carriage was from India, which was not a Hague–Visby Rules country. The mandatory application of the Hague Rules to outward carriage from India under Indian law would be of no effect in a bill of lading subject to English law. 36 In the case of CMR and CMNI. 37 In the case of COTIF/CIM, the Warsaw Convention and the Montreal Convention. 38 Aqualon (UK) Ltd v Vallana Shipping Corp [1994] 1 Lloyd’s Rep 669 (Com Ct); Ulster Swift Ltd v Taunton Meat Haulage Ltd [1977] 1 WLR 625 (CA).

48 Simon Baughen In the UK and in many other CMR countries, Article 1(1) has been interpreted as applying CMR to the international road leg of a multimodal contract for the carriage of goods falling within Article 1(1), whether before or after the carriage by the other mode of transport. This involves reading the words ‘the place of taking over and the place designated for delivery, as specified in the contract’ as referring to the places which the contract specifies for the taking over and delivery by the carrier in its capacity as international road carrier. In Quantum Corp,39 the Court of Appeal rejected the notion that a carriage involving different modes of transport could only fall within CMR if the carriage by road was the predominant mode. A cargo of hard disk drives was to be carried from Singapore to Dublin. The air carrier flew them to Paris on two airline pallets. From Paris, the pallets were to be carried by road to Dublin, on a roll-on, roll-off basis. The goods were stolen during a staged hijack while being carried by road in England. The question was whether the road carrier was subject to CMR or whether it could, instead, rely on the terms of a ‘Himalaya’ clause contained in the air waybill. The Court of Appeal held that, provided that carriage by road was permissible under the contract of carriage, CMR was capable of applying to the international road leg of a larger contract of carriage. In such circumstances, the places of taking over and delivery to which Article 1(1) referred had to be construed so as to refer to the start and end of the international road leg. CMR therefore governed that part of the contract which was to be performed by road: the journey from Paris to Dublin.40 A similar finding was made by the House of Lords in Datec Electronic Holdings Ltd v United Parcels Service Ltd,41 where domestic road carriage was followed by international air carriage, and then by international road carriage from Germany to Holland, which was held to be subject to CMR. The Quantum Corp interpretation of Article 1(1) superimposes a ‘virtual’ contract for road carriage under CMR onto the overall multimodal contract. This may cause problems with regard to what constitutes ‘delivery’ for the purposes of the CMR time bar. Is it delivery under the ‘virtual’ contract, or under the multimodal contract?42 If the CMR ‘virtual’ contract is for the last stage of the multimodal contract, there may be a clash between the delivery obligations under the MT Bill, which requires presentation of the document, and the consignee’s right to delivery and right to sue for loss under Article 13 of CMR. Are these given to the consignee/indorsee under the MT Bill, to the consignee named in the consignment note or to both parties? A contrary interpretation of Article 1(1) has been reached by the courts in Germany and the Netherlands, which have held that CMR does not apply to multimodal carriage.43 This divergence in the European approach to the CMR and multimodal contracts means that the choice of law clause in the MT Bill becomes critical in identifying what regime applies to the international 39 Quantum Corp (n 24). 40 Had the goods been unloaded from the trailer for the sea leg it is difficult to see how there could be any international element in the road carriage, in which case, CMR would not apply. The Court of Appeal refrained from commenting on this hypothesis. 41 Datec Electronic Holdings Ltd v United Parcels Service Ltd [2007] UKHL 23, [2007] 1 WLR 1325. 42 M Hoeks, Multimodal Transport Law: The Law Applicable to the Multimodal Contract for the Carriage of Goods (The Hague, Kluwer Law International, 2010) s 4.1.2.5.2 convincingly argues that it should be the latter, as the purpose of CMR, Art 32 is to determine that the limitation period starts only after the goods have been delivered to the consignee. 43 BGH 17 July 2008, I ZR 181/05 (2009) European Transport Law 196 (Germany); Hoge Raad 1 June 2012, SS 2012, No 95 (Godafoss). Both decisions were made in the context of jurisdictional challenges as to whether the CMR jurisdiction provisions in Art 31 applied. Both countries have domestic laws for multimodal carriage when their national law is the law of the contract. In Belgium, the Cour de Cassation decided in TNT Express Belgium v Mitsui Sumitomo Insurance Co Europe Ltd (2006) European Transport Law 228 (Belgium) that CMR does apply to the international road leg of a multimodal contract of carriage, but only if the contractual agreement expressly provides for carriage by road or, where the contract leaves the means of transport open, if it can be inferred from the circumstances that the parties intended goods to be carried by road.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 49 road carriage element of the contract. A choice of German or Dutch law will displace CMR and will lead to the application, instead, of domestic German or Dutch law on multimodal carriage. The scope of application of CMNI and COTIF/CIM is framed in similar fashion to Article 1(1) of CMR. There is therefore the possibility of a similar divergence of interpretation as regards the application of these conventions to carriage by international waterways and rail carriage respectively, where this forms part of a multimodal contract of carriage. The air conventions make it quite clear that they will operate as regards the international air carriage elements of a contract of carriage that involves performance by other modes of transport. For combined carriage, Article 38 of the Montreal Convention provides: 1. 2.

In the case of combined carriage performed partly by air and partly by any other mode of carriage, the provisions of this Convention shall, subject to paragraph 4 of Article 18, apply only to the carriage by air, provided that the carriage by air falls within the terms of Article 1. Nothing in this Convention shall prevent the parties in the case of combined carriage from inserting in the document of air carriage conditions relating to other modes of carriage, provided that the provisions of this Convention are observed as regards the carriage by air.

Article 31 of the Warsaw Convention is to similar effect.

D. The Different Liability Regimes of the Unimodal Conventions The unimodal conventions currently in force exhibit a wide range of differences in their application, which leads to complexity when dealing with claims for localised loss or damage arising under a multimodal bill of lading. In this context, ‘localised loss or damage’ is where the loss or damage can be shown to have occurred during one stage of a multimodal carriage; by contrast, ‘unlocalised loss or damage’ is where the loss or damage occurs during the course of the multimodal carriage contract, but it cannot be shown to have occurred during any particular stage of that contract. The basic liability structure of all the conventions is one of presumed fault. If the loss or damage occurs during the carrier’s period of responsibility, there is prima facie liability on the carrier, subject to their being able to prove that the loss was caused by the specified excepted perils.44 These excepted perils differ between conventions, but, except in the case of the Hague Rules, they amount to specific instances in which the loss or damage occurred without the fault of the carrier or its servants, agents and subcontractors. Under the Hague Rules, in contrast, there are two defences on which the carrier can rely which involve fault on the part of these parties – the nautical fault exception in Article IV(2)(a) and the fire exception in Article IV(2)(b). A further difference is that the Hague Rules do not provide for the carrier’s liability as regards delay in the carriage. The scope of the sea carriage conventions is tied to the documentation issued in respect of the carriage. For the Hague and Hague–Visby Rules, this is a bill of lading or similar document of title.45 The Hamburg Rules define contract of carriage as ‘any contract whereby the

44 eg CMR, Art 17(1): ‘The carrier shall be liable for the total or partial loss of the goods and for damage thereto occurring between the time when he takes over the goods and the time of delivery, as well as for any delay in delivery.’ This is not explicit in the Hague Rules, but it is implicit. In Volcafe (SC) (n 30) [20], Lord Sumption described the basic liability structure as follows: ‘I consider that in principle where cargo was shipped in apparent good order and condition but is discharged damaged the carrier bears the burden of proving that that was not due to its breach of the obligation in article III rule 2 to take reasonable care.’ 45 This need not have been issued, but its issue must have been contemplated under the contract of carriage.

50 Simon Baughen carrier undertakes against payment of freight to carry goods by sea from one port to another’46 but excludes its application as regards charter parties.47 The provisions defining the scope of the other unimodal conventions refer to specific documents, such as a consignment note under CMR and an air waybill under the Montreal Convention, but the conventions will still apply to the contract of carriage if that document is not used.48 There are major variations when it comes to limitation of liability. At the lower end of the scale there is sea carriage. Under the Hague Rules, the limits are £100 per package or unit, gold value,49 whilst under the Hague–Visby Rules they are 666.67 special drawing rights (SDR) per package or unit or 2 SDR per kg gross weight lost or damaged, whichever is the higher.50 Under the Hamburg Rules, the limits are 835 SDR per package or other shipping unit or 2.5 SDR gross weight of the goods lost or damaged, whichever is the higher.51 CMNI adopts the Hague–Visby limits, but with special provisions for containers.52 The CMR limit is 8.33 SDR gross weight,53 while the highest terrestrial limit is under the COTIF/CIM for international rail carriage, with a limit of 17 SDR per kg of gross mass short.54 The highest limitation figure is that in the Montreal Convention, which provides an unbreakable limit of 19 SDR per kg for carriage of goods.55 Similar variation can be seen with regard to the time provided for bringing suit. For the sea carriage conventions, under both Hague and Hague–Visby Rules, suit must be commenced within one year from delivery or date when goods should have been delivered, failing which the claim is barred.56 Under the Hamburg Rules, the period is two years.57 CMR provides a one year time limit, but three years in cases of wilful misconduct.58 The Montreal Convention and Warsaw Convention both have a two year time limit.59 CMNI has a one year time limit.60 COTIF/CIM has a one year limit,61 but with a two year limit period in four specific instances.62 All of the conventions have a provision rendering null and void express provisions of the contract that derogate from the convention to the benefit of the carrier – such as Article III.8 of the Hague and Hague–Visby Rules, Article 26 of the Montreal Convention and Article 5 of

46 Hamburg Rules, Art 1(6). 47 ibid Art 2(3). 48 eg CMR, Art 4: ‘The contract of carriage shall be confirmed by the making out of a consignment note. The absence, irregularity or loss of the consignment note shall not affect the existence or the validity of the contract of carriage which shall remain subject the provisions of this Convention.’ 49 Hague Rules, Art IV(5). 50 Hague–Visby Rules, Art IV(5)(a). 51 Hamburg Rules, Art 6(1)(a). 52 CMNI, Art 20(1): ‘If the package or other loading unit is a container and if there is no mention in the transport document of any package or loading unit consolidated in the container, the amount of 666.67 [SDR] shall be replaced by the amount of 1,500 [SDR] for the container without the goods it contains and, in addition, the amount of 25,000 [SDR] for the goods which are in the container.’ 53 CMR, Art 23(3). 54 COTIF/CIM, Art 30(2). 55 Montreal Convention, Art 22(3). The figure under the Warsaw Convention, subject to the 1955 Hague Protocol, is 17 SDR per kg. This was also the figure under the Montreal Convention, Art 22(3), until it was raised at the end of 2009. 56 Hague Rules, Art III(6); Hague–Visby Rules, Art III(6). 57 Hamburg Rules, Art 20. 58 CMR, Art 32(1). 59 Montreal Convention, Art 35: Warsaw Convention, Art 29. 60 CMNI, Art 24. 61 COTIF/CIM, Art 48(1). 62 ibid, providing a two-year period in cases of action: ‘a) to recover a cash on delivery payment collected by the carrier from the consignee; b) to recover the proceeds of a sale effected by the carrier; c) for loss or damage resulting from an act or omission done with intent to cause such loss or damage, or recklessly and with knowledge that such loss or damage would probably result; d) based on one of the contracts of carriage prior to the reconsignment in the case provided for in Article 28’.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 51 COTIF/CIM. Under the Hamburg Rules, Article 23(1) renders null and void any stipulation which derogates directly or indirectly from the provisions of the convention, but Article 23(2) permits the carrier to increase his responsibilities and obligations under it. In contrast, Article 41(1) of CMR precludes any contractual derogation: ‘any stipulation which would directly or indirectly derogate from the provisions of this Convention shall be null and void’. Similarly, under Article 25 of CMNI: any contractual stipulation intended to exclude, limit or increase the liability, within the meaning of this Convention of the carrier, the actual carrier or their servants or agents, shift the burden of proof or reduce the periods for claims or limitations referred to in Articles 23 and 24 shall be null and void.

Most of the conventions contain a provision on third party rights – the right of the consignee to sue the carrier, as well as some form of Himalaya protection for third parties engaged by the carrier. CMR gives rights of suit to the consignee in respect of loss or delay,63 as do the Warsaw and Montreal Conventions,64 and COTIF/CIM.65 However, none of the sea conventions, nor CMNI, give rights of suit to third parties. All the conventions except the Hague Rules provide some form of protection to third parties engaged by the carrier to perform the contract. The Hague–Visby Rules, Article IV bis 2 extends the protection of the rules to servants and agents of the carrier but excludes independent contractors. There is a similar provision in the Hamburg Rules, Article 7(2), in favour of servants and agents of the carrier, or of the actual carrier, who can prove they acted within the scope of their employment, but without the exclusion of independent contractors; and also in CMNI, Article 17(3), and in the Montreal Convention, Article 30. Wider protection to third parties is provided by CMR, Article 28(2): In cases where the extra-contractual liability for loss, damage or delay of one of the persons for whom the carrier is responsible under the terms of Article 3 is in issue,66 such person may also avail himself of the provisions of this Convention which exclude the liability of the carrier or which fix or limit the compensation due.

Similarly, COTIF/CIM, Article 40, extends its provisions to ‘other persons whose services he makes use of for the performance of the carriage, when these servants and other persons are acting within the scope of their functions’ and to ‘servants or other persons for whom the carrier is liable pursuant to Article 40’. Rules on jurisdiction appear in Article 21 of the Hamburg Rules, Article 32 of CMR, Article 46 of COTIF/CIM, Article 28 of the Warsaw Convention and Article 33 of the Montreal Convention,67 but not in the Hague and Hague–Visby Rules, nor in CMNI (which contains provisions on applicable law in Article 29).

63 CMR, Art 13(1). This article does not refer to a claim in respect of damage, but Continental case law is to the effect that this is within Art 13. See, eg OLG Karlsruhe, 24 May 1967 (1967) ULC 289; RB Antwerpen, 7 December 1973 (1976) 11 ETL 295; App Bruxelles, 30 October 1975 (1976) 11 ETL 238. 64 Warsaw Convention, Arts 13(3) and 30(3); Montreal Convention, Arts 13(3) and 36(3). 65 COTIF/CIM, Art 44 gives the consignee the right to ‘bring an action based on the contract of carriage … from the time when he has 1. taken possession of the consignment note, 2. accepted the goods, or 3. asserted his rights pursuant to Article 17 § 3 or Article 18 § 3’. 66 See further CMR, Art 3: ‘agents and servants and of any other persons of whose services he makes use for the performance of the carriage, when such agents, servants or other persons are acting within the scope of their employment’. 67 These rules will take precedence over those in the Recast Brussels Regulation (Regulation (EU) 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2012] OJ L351/1). This follows from Art 71 of the Regulation, which determines that the Regulation ‘shall not affect any conventions to which the Member States are parties and which in relation to particular matters, govern jurisdiction or the recognition or enforcement of judgment’.

52 Simon Baughen

E. Multimodal Operation of the Unimodal Conventions A further complication in the multimodal jigsaw is that some of the unimodal conventions make provision for limited application of their terms to other modes of carriage. Article 2(1) of CMR deals with ‘roll-on, roll-off ’ road carriage over another mode of transport and applies the convention to the whole of the carriage. This is subject to a proviso that applies if the damage, loss or delay in delivery of the goods occurred during the other mode of carriage due to fault of the road carrier or by an event that could not have occurred solely in the course of and by reason of the other mode of carriage. In this event, CMR may be displaced and, instead, the carrier’s liability will be determined in the manner in which the liability of the carrier by the other means of transport would have been determined if a contract for the carriage the goods alone had been made by the sender with the carrier by the other means of transport in accordance with the conditions prescribed by law for the carriage of goods by that means of transport.

This requires the court to construct a hypothetical contract between the sender and the carrier by the other means of transport and then to determine whether or not there would be conditions prescribed by law applicable to that contract. Absent such prescribed conditions, CMR will continue to govern the liability of the carrier by road.68 This complex CMR provision – described as ‘the English nightmare’ because it was inserted at the insistence of English ferry companies – leads to a complex search for the terms of this hypothetical contract that would have been concluded for the carriage by the other means of transport. In Thermo Engineers Ltd v Ferrymasters Ltd,69 maritime-specific damage occurred when a steam heat exchanger on a trailer hit the deckhead of the vessel while the trailer was being loaded onto the vessel. Neill J held that it would not be automatic that the carrier’s liability would be on the terms of the Hague–Visby Rules, given that Article V allows the parties to reduce the shipowner’s defences and increase its obligations. Another possibility, which was not considered by Neill J, was that the ocean carrier concerned may only have been prepared to contract on the basis that a waybill was issued. As the Hague–Visby Rules do not mandatorily apply to waybills, liability would, in the absence of ‘such prescribed conditions’, continue to be governed by CMR. The CMNI’s multimodal aspect is found in Article 2(2), which provides: ‘This Convention is applicable if the purpose of the contract of carriage is the carriage of goods, without transshipment, both on inland waterways and in waters to which maritime regulations apply’. CMNI will not apply where there is transshipment, nor where (i) a maritime bill of lading has been issued in accordance with the maritime law applicable70 or (ii) the distance to be travelled in waters to which maritime regulations apply is the greater. This effectively rules out a conflict with the Hague Rules, which apply to bills of lading for international carriage, and so would be classed as ‘maritime’ bills of lading under CMNI.

68 CMR, Art 2(1): ‘the liability of the carrier by road shall be determined not by this convention but in the manner in which the liability of the carrier by the other means of transport would have been determined if a contract for the carriage [of] the goods alone had been made by the sender with the carrier by the other means of transport in accordance with the conditions prescribed by law for the carriage of goods by that means of transport’. 69 Thermo Engineers Ltd v Ferrymasters Ltd [1981] 1 WLR 1470 (QB). 70 Whatever a ‘maritime’ bill may mean.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 53 COTIF/CIM may expand into other modes of transport through Article 1(3), where there is internal carriage by road or inland waterway in a Member State as a supplement to transfrontier carriage by rail. Article 1(4) provides: When international carriage being the subject of a single contract of carriage includes carriage by sea or transfrontier carriage by inland waterway as a supplement to carriage by rail, these Uniform Rules shall apply if the carriage by sea or inland waterway is performed on services included in the list of services provided for in Article 24 § 1 of the Convention.71

The Warsaw and Montreal Conventions are meanwhile clear as to their unimodal effect. Article 18(1) of the Montreal Convention provides: The carrier is liable for damage sustained in the event of the destruction or loss of or damage to, cargo upon condition only that the event which caused the damage so sustained took place during the carriage by air.

Similarly, Article 18(1) of the Warsaw Convention provides: The carrier is liable for damage sustained in the event of the destruction or loss of, or of damage to, any registered luggage or any goods, if the occurrence which caused the damage so sustained took place during the carriage by air.

Carriage by air is defined in Article 18(3) of the Montreal Convention as comprising ‘the period during which the cargo is in the charge of the carrier’. The equivalent provision in the Warsaw Convention, Article 18(2), contains the additional words ‘whether in an aerodrome or on board an aircraft, or, in the case of a landing outside an aerodrome, in any place whatsoever’. It is likely that the omission of these words extends the scope of the Montreal Convention to the functional limits of the airport – to any airport-related area outside the airport perimeter. A limited multimodal scope is provided by Article 18(3) of the Warsaw Convention and Article 18(4) of the Montreal Convention. Each states that the period of the carriage by air does not extend to any carriage by land, by sea or by inland waterway performed outside an airport, but goes on to provide that: If, however, such carriage takes place in the performance of a contract for carriage by air, for the purpose of loading, delivery or trans-shipment, any damage is presumed, subject to proof to the contrary, to have been the result of an event which took place during the carriage by air.

The operation of the convention outside the airport requires that: (i) the non-aerial segment is covered by a contract for carriage by air; (ii) the non-aerial segment is performed with the aim of delivering the goods to the consignee, loading them on an aircraft or transshipping them; and (iii) the event causing the loss, damage or delay cannot be identified as occurring during either the aerial segment or the non-aerial segment of the carriage. Notwithstanding this, there has been a

71 In this event, COTIF/CIM, Art 38 allows a Member State, by requesting that a suitable note be included in the list of service to which these Uniform Rules apply, to add four additional grounds of exemption from liability to those provided for in Art 23: (i) fire, if the carrier proves that it was not caused by his act or default, or that of the master, a mariner, the pilot or the carrier’s servants; (ii) saving or attempting to save life or property at sea; (iii) loading of goods on the deck of the ship, if they are so loaded with the consent of the consignor given on the consignment note and are not in wagons; and (iv) perils, dangers and accidents of the sea or other navigable waters. Reliance on these additional maritime perils is subject to proof by the carrier that the loss, damage or exceeding the transit period occurred between loading onto and unloading from the ship. The carrier will nevertheless remain liable if the person entitled proves that the loss, damage or exceeding the transit period is due to the fault of the carrier, the master, a mariner, the pilot or the carrier’s servants.

54 Simon Baughen trend in some US courts to give a door-to-door application of the two air conventions to carriage covered by a door-to-door air waybill.72 There is a further extension to non-aerial carriage under Article 18(4) of the Montreal Convention in cases of unauthorised substitution by the carrier of the intended air carriage by another mode of transport.

III. Solutions to the Problems A. International Conventions Multimodal transport gives rise to the problem that one contract of carriage will involve the application of two or more unimodal conventions. There are three ways in which this problem can be addressed either by a dedicated multimodal convention or by mandatory national law. First, the multimodal contract could be subject to a uniform system of liability. This will involve a clash with the unimodal conventions which could be avoided if liability, limitation and time bar provisions were as good for the shipper as those in the most generous of those conventions. Secondly, a network system could be applied where a default regime is applied for the m ultimodal carriage, which in the case of localised loss or damage gives way to the provisions of mandatory unimodal conventions or mandatory national laws governing the particular stage at which the loss or damage occurred. Thirdly, a modified network system could be applied where a uniform liability regime applies, but limitation of liability defers to the provisions of the unimodal convention or mandatory national law that applies to the stage of the carriage at which the loss or damage took place. Two conventions have attempted to address the problems associated with the clash of conventions involved in multimodal transport. The first was the 1980 MT Convention, which applied a uniform liability system for both localised and non-localised loss occurring while the MTO is in charge of the goods, based on the principle of ‘presumed fault or neglect’ – a regime similar to that of the Hamburg Rules. Article 16(1) provided a defence if the multimodal transport operator could prove that ‘he, his servants or agents or any other person referred to in Article 15 took all measures that could reasonably be required to avoid the occurrence and its consequences’. This was subject to Article 4(1), which provided that: ‘This Convention shall not affect, or be incompatible with, the application of any international convention or national law relating to the regulation and control of transport operations.’ Limitation of liability was dealt with under a network system, under which there would be a liability regime under the convention and a separate regime for localised loss – the limits of any unimodal convention or mandatory law applying to that stage of the carriage would apply instead if their limitation figure was higher than those provided by the convention. The convention’s limits were 920 SDR per package or other shipping unit or 2.75 SDR per kg of gross weight of the goods lost or damaged, whichever is the higher, where the multimodal transport, according to the contract, included carriage by sea or by inland waterway. If this was not the case, the CMR level of 8.33 SDR per kg of gross weight of the goods lost or damaged applied.

72 See G Leloudas, ‘Door to Door Application of the International Air Law Conventions: Commercially Convenient but Doctrinally Dubious’ [2015] Lloyd’s Maritime and Commercial Law Quarterly 368. CMR rules out such expansion in Europe.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 55 The MT Convention failed to attract sufficient ratifications to enter into force. This was due to three reasons: (i) the basis of liability, which was modelled after the Hamburg Rules, rather than the Hague–Visby Rules; (ii) the monetary limitation of liability, which some considered as too high; and (iii) the principle of uniform liability, which some considered to raise concerns both in relation to recourse actions by an MTO against a subcontracting unimodal carrier and for introducing mandatory liability levels in relation to transports otherwise not subject to mandatory law, such as road and rail transport not covered by CMR or CIM/COTIF. However, the MT Convention has been influential on the contractual rules which have been developed for multimodal carriage. The second convention is the Rotterdam Rules – a ‘maritime plus’ convention, applying to international door-to-door carriage involving international sea carriage. The Rotterdam Rules provide a 50 per cent uplift on the Hague–Visby limits – 3 SDR per kg or a 875 SDR package or shipping unit, whichever is the higher – and doubled the time for bringing suit or commencing arbitration. As well as carriers, certain of their subcontractors who constituted ‘maritime performing parties’ are brought directly within scope of the convention. Article 57 makes some provision for creating third party rights of suit under bills of lading (or ‘negotiable transport documents’), although commentators are divided as to whether those rights extend beyond the right to delivery under chapter 9 of the Rotterdam Rules and the right of control under chapter 10. The potential clash with the non-maritime transport conventions is addressed, not entirely successfully, through two articles. Article 26 provides a network solution to the problems of competing conventions that occur with multimodal carriage. Where the loss, damage or event causing delay occurs during the carrier’s period of responsibility, provisions of another international ‘instrument’ will prevail over the convention, but only to the extent that they: (i) relate to the carrier’s liability, limitation of liability and time for suit; (ii) cannot be departed from to the shipper’s detriment under the terms of the other ‘instrument’; and (iii) would have applied to a hypothetical contract between the shipper and the carrier for the particular stage of carriage where the loss, damage or event causing delay occurred. Thus, provisions of the convention relating to the right of control and the right to delivery will still prevail over those in the other ‘instrument’, and will also prevail where the claimant is unable to prove where during the carriage the loss occurred. Article 82 defers to the air, road, rail and inland waterway conventions as regards their multimodal provisions. Eleven years on, the Rotterdam Rules seem to be going the same way as the MT Convention – they have attracted only four of the 20 ratifications necessary for their coming into force. Accordingly, there is currently no international instrument in force that deals with multimodal transport, although there are various regional agreements73 and national laws, such as those in Germany and the Netherlands,74 regulating such carriage.

B. Contractual Solutions Although the 1980 MT Convention never entered into force, a modified version of its provisions was to provide the basis for the UNCTAD/ICC Rules for Multimodal Transport Documents,

73 eg Asociacion Latinoamericana de Intergracion [Latin American Integration Association, ALADI] Agreement on International Multimodal Transport, 8 November 1996. 74 See n 43 above.

56 Simon Baughen which entered into force on 1 January 1992. These are incorporated in widely used multimodal transport documents, such as the FIATA FBL 1992 and MULTIDOC 95. They are of a purely contractual nature and apply only if they are incorporated into a contract of carriage, without any formal requirement for ‘writing’.75 They prevail over any inconsistent terms in the multimodal bill of lading,76 but can only take effect to the extent that they are not contrary to the mandatory provisions of international conventions or national law applicable to the multimodal transport contract.77 The UNCTAD/ICC Rules are based on the MT Convention with the same basis of liability and the same defence available to the MT carrier.78 They differ from the MT Convention as follows. For there to be liability for delay in delivery, they require the consignor to make a declaration of interest in timely delivery, which must also be accepted by the MTO.79 In cases of sea carriage or by inland waterways, the UNCTAD/ICC Rules provide two additional defences – the nautical fault and fire exceptions in Article IV(2)(a) and (b) of the Hague Rules, subject to the overriding requirement of the proof of exercise of due diligence to provide a seaworthy vessel.80 The basic limitation figure is the Hague–Visby limitation of 666.67 SDR per package or unit, with its container formula, or the weight alternative of 2 SDR per kg, rather than the higher 2.75 SDR limit in the MT Convention.81 If there is no carriage by sea or inland waterways, the limit is the CMR limit of 8.33 SDR per kg.82 Liability for delay in delivery or consequential loss is limited to an amount not exceeding the equivalent of the freight charged under the multimodal transport contract, with an aggregation of the limits so that they do not exceed the limit of liability for total loss of the goods.83 With localised loss, these limits give way to those provided in an applicable international convention or mandatory national law which would have provided another limit of liability – and not necessarily a higher limit as provided by the MT Convention – if a separate contract had been made for that particular stage of transport. However, as well as limitation, the liability rules of any unimodal convention applicable to the stage at which the loss or damage can be shown to have occurred will prevail over those in the document, both through the mandatory operation of the convention in question and through Rule 13 of the UNCTAD/ICC Rules. The rules provide for a nine month time limit for bringing claims, so as to give the MT carrier time to bring recourse actions against its subcontractors. However, this will be displaced by the time bar provisions in any mandatory international convention or national law governing the stage of transport during which the loss damage or delay occurred. BIMCO have produced two contractual forms for multimodal bills of lading – MULTIDOC and COMBICONBILL – both of which are available in bill of lading and waybill formats. Both forms adopt the presumed fault liability scheme of the MT Convention. MULTIDOC provides that the MTO shall be responsible for the acts and omissions of his servants or agents when any such servant or agent is acting within the scope of his employment, or of any other person of whose services he makes use for the performance of the contract as if such acts and omissions 75 UNCTAD/ICC Rules for Multimodal Transport Documents, r 1.1. 76 ibid r 1.2. 77 ibid r 13. 78 ibid r 5.1: the MTO’s liability is subject to a qualification where ‘the MTO proves that no fault or neglect of his own, his servants or agents or any other person referred to in Rule 4 has caused or contributed to the loss, damage or delay in delivery’. 79 ibid r 5.1. 80 ibid r 5.4. 81 ibid r 6.1. 82 ibid r 6.3. 83 ibid rr 6.5, 6.6.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 57 were his own. There is a similar provision in COMBICONBILL as regards the responsibility of the carrier. Both spell out the common law delivery obligations of the carrier when in bill of lading format. The significant difference between the forms is the reference to liability for delay under MULTIDOC – this may make it unacceptable to P&I clubs, which require terms of carriage to be on a Hague/Hague–Visby basis, which does not provide for liability for delay.84 Under MULTIDOC,85 the front of the bill incorporates the UNCTAD/ICC Rules. It therefore brings in the additional defences for carriage by sea or by inland waterway, adding in the Hague– Visby defences found in Article IV(2)(c)–(p), but not (q), subject to requirement of the proof of exercise of due diligence to provide a seaworthy vessel.86 For limitation of liability, MULTIDOC adopts the twofold sea and road scheme of the UNCTAD/ICC Rules, together with the limitation for delay claims, subject to a limit of $500 per package or customary freight unit where US Carriage of Goods by Sea Act 1936 (COGSA 1936) applies. Under COMBICONBILL,87 which can be used for either port-to-port or multimodal carriage, the carrier is liable for loss or damage to the goods occurring between the time when it receives the goods into its charge and the time of delivery. Eight specific defences are provided, based on the defences specified in CMR Article 17(2), some of the special defences specified in CMR Article 17(4) and the Hague Rules strike clause found in Article IV(2)(j). These eight defences are: (i) the wrongful act or neglect of the merchant; (ii) compliance with the instructions of the person entitled to give them; (iii) the lack of, or defective conditions of, packing in the case of goods which, by their nature, are liable to wastage or to be damaged when not packed or when not properly packed; (iv) handling, loading, stowage or unloading of the goods by or on behalf of the merchant; (v) inherent vice of the goods; (vi) insufficiency or inadequacy of marks or numbers on the goods, covering or unit loads; (vii) strikes, lock-outs, stoppages or restraints of labour from whatever cause, whether partial or general; and (viii) any cause or event which the carrier could not avoid and the consequence whereof he could not prevent by the exercise of reasonable diligence.88 The basic limitation figure is the Hague–Visby weight alternative of 2 SDR per kg gross weight goods lost or damaged, but with no package limitation,89 as well as the UNCTAD/ICC Rules limitation for delay claims if the carrier is held liable for delay. The basic liability regime is displaced in cases of localised loss or damage – in which case the merchant is entitled to have liability determined in accordance with the mandatory international convention regime or national law applicable to that leg of the carriage.90 For localised loss during 84 The carrier will still be liable for delay, but this will be under national law, and the carrier will have complete freedom of contract to limit or exclude this liability. See S Ciger, ‘Claims for Compensation for Delay in Delivery and Notice Requirements under Article 23.4 of the Rotterdam Rules’ (2015) 21 Journal of International Maritime Law 39. 85 cl 19. 86 cl 11. 87 COMBICONBILL, cl 9. 88 For defences (iii)–(vii), a formula applies which is similar to that adopted in Art 18(2) of CMR, for the special risks defences in Art 17(2). If the carrier establishes that the loss or damage could be attributed to one or more of the causes or events, it shall be presumed that it was so caused, with the merchant being entitled to prove that the loss or damage was not, in fact, caused either wholly or partially by one or more of the causes or events. 89 BIMCO’s explanatory notes justify this omission by reference to the carrier’s position in recourse actions against its subcontractors, noting ‘At the same time, since haulers only accept liability on the basis of weight limitation, it was agreed to make no reference to the package limitation’. 90 The Unfair Contract Terms Act 1977 is one such national law. In Granville Oil & Chemicals Ltd v Davis Turner & Co Ltd [2003] EWCA Civ 570, [2003] 2 Lloyd’s Rep 356, the nine-month time limit in such a contract was held to be ‘reasonable’. The 1977 Act will not cover a contract for the carriage of goods by ship, but will cover a multimodal carriage contract, involving sea and road carriage, even when the damage took place during the sea leg. Compare this approach to classification with that of the Court of Appeal in Quantum (n 24). It should be noted that the Hague–Visby Rules did not apply, as carriage was from Kuwait, which is not a party to the Rules. Had the Hague–Visby Rules applied, the nine-month limit would have been struck out by Art III(8).

58 Simon Baughen carriage by sea or inland waterways that is not subject to a mandatory national law or international convention,91 the Hague–Visby Rules apply to all goods, whether carried on deck or not. Their application is contractual,92 and as such their provisions are therefore displaced by other specific provisions in the bill, such as the nine month time limit. Clause 5 of both forms – MULTIDOC and COMBICONBILL – provide for disputes to be determined by the courts of the place where the MTO has its principal place of business and to be governed by the law of that place. This harmonises with those unimodal conventions that have jurisdiction provisions, as this place of jurisdiction constitutes one of the permitted places in all such conventions. Both forms of bill also protect the carrier’s servants, agents and subcontractors by a combined circular indemnity and Himalaya clause.93 In the 2016 version of the forms, a total exemption is given to a wide category of performing parties engaged or employed by the carrier.94 The total exemption granted by the clause is likely to be struck out by the non-derogation provisions of the applicable unimodal convention covering the stage at which the loss, damage or delay occurred. This is certainly the case with the Hague Rules as regards the shipowner as a subcontractor under a charterer’s bill,95 although not as regards non-carrier subcontractors.96 The Himalaya clause provides for the right of ‘servants’ to enforce all carrier’s rights and exemptions, including the right to enforce any jurisdiction or arbitration clause.97 The MULTIDOC and COMBICONBILL forms, or variants of them, are in widespread use, but other contractual forms also exist. Some multimodal bills will extend the provisions of a unimodal convention to the entire multimodal contract of carriage, or to parts of it that would fall outside its ambit.98 In the USA, multimodal bills incorporating COGSA 1936 to the entire carriage have been held to cover inter-state rail carriage within the USA to the place of delivery, notwithstanding domestic legislation on such carriage (the ‘Carmack Amendment’).99 If this results in a clash with the mandatory provisions of another unimodal convention, such incorporation will be rendered nugatory by any non-derogation provisions.100 The current International

91 COMBICONBILL, cl 11(2). 92 If it is argued that an MT Bill is not a bill of lading or similar document of title, the rules would apply contractually as under a clause paramount in a charter with the necessary manipulation to references to ‘bill of lading’. 93 COMBICONBILL, cl 15; MULTIDOC, cl 14. 94 COMBICONBILL 2016, cl 15; 2016, cl 16. The exemption is for ‘servants’ of the carrier, who are defined as including: ‘the owners, managers, and operators of vessels (other than the Carrier); underlying carriers; stevedores and terminal operators; and any direct or indirect servant, agent, or subcontractor (including their own subcontractors), or any other party employed by or on behalf of the Carrier, or whose services or equipment have been used to perform this contract whether in direct contractual privity with the Carrier or not.’ 95 Homburg Houtimport BV v Agrosin Private Ltd (The Starsin) [2003] UKHL 12, [2004] AC 715. 96 Whitesea Shipping & Trading Corp v El Paso Rio Clara Ltda (The Marielle Bolten) [2009] EWHC 2552 (Comm), [2010] 1 Lloyd’s Rep 648. 97 Thereby getting around the Privy Council’s decision in The Mahkutai [1996] AC 650 (PC). 98 Princes Buitoni Ltd v Hapag-Lloyd AG [1991] 2 Lloyd’s Rep 383 (CA). CMR applied contractually to the land leg of a multimodal bill of lading, even though there was no international carriage by road as required by CMR. 99 49 USC 11707 (1979). See Kawasaki Kisen Kaisha v Regal-Beloit Corp 561 US 89 (2010). 100 Unless suit is brought in a country which is not a party to the unimodal convention, such as the USA, for the other mode of transport. In Hartford Fire Insurance Co v Orient Overseas Container Line (The OOCL Bravery) [2000] AMC 1305, [2000] 1 Lloyd’s Rep 394 (SDNY) the limits in COGSA 1936, which were contractually applied to the entire multimodal carriage, were applied, rather than the lower CMR limits, which would have covered the international road carriage in Europe during which the goods were stolen. On appeal, the Second Circuit held that CMR applied due to the floating choice of law clause, which provided that ‘each stage of the transport shall be governed according to any law and tariffs applicable to such stage’. The District Court had not enforced this clause, incorrectly applying COGSA ex proprio vigore, whereas this would only be the case during ‘the period from the time when the goods are loaded on to the time when they are discharged from the ship’.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 59 Air Transport Association (IATA) air waybill applies the Montreal Convention limits to the entire carriage,101 but does not extend either the Warsaw Convention or the Montreal C onvention to the non-aerial parts of the carriage.102

IV. Conclusion The liability regime for multimodal carriage is undeniably complex and untidy. Apart from the differences between the applicable unimodal conventions, there are various gaps in the patchwork of unimodal conventions and domestic laws that govern the performance of this single multimodal contract of carriage in its different modes of performance. What regime governs storage between modes? What regime governs where there is progressive damage to the goods in the container across modes? Is the time limit under the contract dependent on the mode in which the loss, damage or delay occurred? If so, is the starting point of ‘delivery’ under that mode or is it delivery at the end of the overall contractual carriage?103 What should be done about this maddeningly messy patchwork of overlapping unimodal conventions and national laws? One solution would be a dedicated multimodal convention just for multimodal contracts of carriage, applying a uniform regime. Everything would be predictable, with no gaps in the liability regime applicable from when the MTO takes charge of the goods to when the MTO delivers them. A limitation figure would have to be agreed, and here there would be the problem of choosing from the highest common denominator in the unimodal conventions – 19 SDR per kg under the Montreal Convention – or the lowest common denominator – 2 SDR per kg or 666.67 SDR per package or unit under the Hague–Visby Rules, or £100 gold value under the Hague Rules. There would then be the troubling matter of the multimodal operation of the unimodal conventions. These would have to be amended so that their definition of contract of carriage was restricted to international carriage where the entirety of the contractual performance is through that particular mode. Assuming this could be achieved, there would then open up a recourse gap between the MTO and the actual carriers used by the MTO to perform the multimodal contract of carriage. This might be soluble contractually by providing for the higher limits of the multimodal convention to apply, but there is a problem with the two-way non-derogation provisions in CMR104 and CMNI. These would need to be amended as well. Alternatively, one could aim for one single international carriage convention to rule them all – unimodal and multimodal. Every state that is a party to the unimodal conventions

101 IATA Air Waybill, cl 4. 102 ibid cl 2.1: ‘Carriage is subject to the rules relating to liability established by the Warsaw Convention or the Montreal Convention unless such carriage is not “international carriage” as defined by the applicable Conventions.’ 103 Some of these gaps are also present with regard to unimodal sea carriage. With sea carriage, there is the gap in the Hague Rules between the reception of the goods at the port in the carrier’s warehouse and the start of loading operations, mirrored by the gap between discharge and delivery from the warehouse at the end of the voyage, and the question of the application of the Hague Rules during periods of storage pending transshipment. Progressive loss of cargo was in issue in The Starsin (n 95), in which the House of Lords found that a single cause of action in negligence arose when the cargo was first damaged. 104 The problem is more acute with CMR, as Art 1(5) provides: ‘The Contracting Parties agree not to vary any of the provisions of this Convention by special agreements between two or more of them, except to make it inapplicable to their frontier traffic or to authorize the use in transport operations entirely confined to their territory of consignment notes representing a title to the goods.’

60 Simon Baughen would need to denounce them when signing up to the new super convention; otherwise there would be the risk of convention clash. The applicable limitation for the new convention would also need to be addressed. Go high or go low? Agreement on this would be almost impossible. Sea carriers would not want the Montreal Convention limits, which would result in an almost tenfold increase; conversely, air cargo interests would be unimpressed by any lowering of those limits. That leaves the network solution that has been adopted contractually through the UNCTAD/ ICC Rules and then the BIMCO forms – MULTIDOC and COMBICONBILL. This provides a basic liability regime, which would then defer to the unimodal conventions where applicable105 and also to mandatory national laws. Its liability system would, therefore, cover only unlocalised loss and gaps between the unimodal conventions, subject to national mandatory law on those gaps. The Rotterdam Rules have adopted a limited network system, with Article 26 deferring to the unimodal conventions in cases of localised loss as regards liability, limitation of liability and time limits, and deferring to the multimodal provisions of the air, road, rail and inland waterways conventions in Article 82. Without wholesale denunciation of the unimodal conventions, the only scope for a multimodal convention would be as regards unlocalised loss or for localised loss falling outside the scope of any mandatory international convention or national law. The MTO’s recourse for localised loss outside the scope of a mandatory international convention or national law would be subject to the terms of the MTO’s contract with the actual carrier or service provider for that stage. There would be no recourse gap for unlocalised loss as, by definition, the MTO will not be able to be pin the loss or damage on any particular actual carrier.106 It would probably be feasible to secure agreement for a multimodal convention based on some form of presumed fault-based liability, with the Hague–Visby limits where carriage by sea or inland waterways is involved and the CMR limits where it is not, and with the unimodal conventions applying where the loss or damage is located within the scope of their operation. This looks very similar to what is already on offer through the two BIMCO forms – the only advantage of such a multimodal convention would be mandatory application. In any event, the advent of ‘smart containers’107 means that the scope for unlocalised loss with containerised cargo will dwindle in future. Nor is there any compelling economic ground for renewing efforts to create a new international convention for multimodal carriage of goods. A 2001 report to the European Commission, The Economic Impact of Carrier Liability on Intermodal Freight Transport, estimated the total friction cost in Europe of carrier liability for existing intermodal transport operations108 at about €500–550 million per annum.109 However, harmonisation of conditions to facilitate intermodal transport – such as a uniform liability limit for all modes – could yield

105 In some cases, as with the UNCTAD/ICC Rules and the MULTIDOC form, this deference appears to operate only as regards the limitation provisions of mandatory international conventions and national laws, but this is illusory, due to the effect of UNCTAD/ICC, r 13. 106 In case of unlocalised loss, the MTO would incur liability as it would be unable to show how the loss occurred and would therefore not be able to show that the loss came within one of the applicable exceptions. For the same reason, it would be unable to claim recourse from any of its subcontractors. 107 Through an equivalent of an aircraft ‘black box’ attached to the container, information can be captured in real time via a wide range of sensors about their current status (eg GPS position, door opening detection, movements vibrations and shocks, atmospheric conditions). The technology can be used to enable remote changing of container parameter settings (eg temperature setups, remote defrost) anywhere, anytime. See F Stevens, ‘Smart Containers: The Smarter, the More Scope for Liability?’ in B Soyer and A Tettenborn (eds), Maritime Liabilities in a Global and Regional Context (Abingdon, Informa Law, 2018) ch 5. 108 These costs were estimated to amount to less than 0.2% of consignment (cargo) value, varying for different types of journey, depending particularly on consignment value and journey length. The level of risk for national, intra-Europe and extra-European carriage was estimated as amounting to 6.3%, 3.9% and 2.4% of the freight charges, respectively. 109 This comes to about €1 per annum (the 2001 cost of a cup of coffee) for each man, woman and child in the EU.

Multimodal Carriage of Goods by Sea: Time for an International Convention? 61 savings in friction costs to intermodal transport of a much lower figure of €50 million per annum. This is getting very close to de minimis. The EU figures indicate that the economic benefits of a global unified liability system are really not worth the candle; they may be offset by the costs of adjusting to a new regime while still applying the old system; and, as noted by Nikaki, a purely multimodal convention might well lead to a rise in freight rates by multimodal carriers to take account of the recourse gap.110 To answer the question posed in the title of this chapter: no, it is not time for a new c onvention. Far better to leave it to industry contract forms to sort out.

110 T Nikaki, ‘Bringing Multimodal Transport Law into the New Century: Is the Uniform Liability System the Way Forward?’ (2013) 78 Journal of Air Law and Commerce 69, 116–17.

62

4 The Three Ages of Utmost Good Faith HOWARD BENNETT

I. Introduction It is axiomatic that insurance contracts have a different underlying ethos from other contracts, a difference that is conveniently and pithily encapsulated by the concept of ‘utmost good faith’. General contract law does not advocate bad faith, but insurance contract law has higher expectations. The prime manifestation of these expectations lies in the rules that attach to contract negotiation. General contract law requires statements made to be true but affords each negotiating party a largely free choice as to whether to make any statements.1 Insurance contract law imposes a duty to speak and equates silence with a false statement. The resulting duty of disclosure has understandably attracted considerable attention given the onerous nature of the obligation imposed on the assured and, historically at least, the strictness of the remedies for breach. Until relatively recently, however, comparatively little focus was placed on an underlying, pervasive idea or doctrine of utmost good faith. This chapter is concerned with utmost good faith in that broader sense. The chapter considers utmost good faith from a chronological perspective, distinguishing three ages in its evolution. The first age extends from the earliest reference in insurance contract law to utmost good faith throughout an initial period during which its impact was untrammelled by any statutory intervention. The second age is the period of nearly 110 years during which utmost good faith was, in part at least, the subject of sections 17–20 of the Marine Insurance Act 1906 as originally drafted. The 1906 Act transformed utmost good faith from a pervasive, underlying ethos into a formal legal doctrine. Particular problems were generated by the attachment to section 17’s articulation of utmost good faith as the basis of marine insurance contracts of a specific remedy for non-observance, namely avoidance of the contract. The third age was ushered in by the Insurance Act 2015, which repealed most of the relevant provisions of the 1906 Act while amending section 17 so as to remove any reference to remedies. A central theme of this chapter is that utmost good faith affects insurance contracts in a diversity of ways that require remedial flexibility. This imperative drives a proper understanding of utmost good faith in its second age. The welcome removal by the 2015 Act of the section 17 remedial straitjacket restores to the doctrine its full creative potential. Nevertheless, the extent to which insurance contract

1 One exception is the rule in Smith v Hughes (1870–71) LR 6 QB 597 (QB) against taking advantage of known mistakes by the other party as to the terms of the contract.

64 Howard Bennett law truly needs an idea or doctrine of utmost good faith is questionable. Various principles that have developed under the auspices of utmost good faith, or at least with the support of insurance contracts being characterised by utmost good faith, could have evolved at common law as independent, bespoke principles of insurance law sustained by their own merits.

II. The First Age: The Age of Common Law The seminal judgment of Lord Mansfield in Carter v Boehm2 provides a logical starting point for an account of the doctrine of utmost good faith in English insurance contract law.3 Lord Mansfield articulated a general principle and a systematic series of propositions concerning the non-disclosure of information in the course of contractual negotiations. The judgment enjoys a deserved enduring prominence, but there are certain claims that it cannot make. First, Carter v Boehm is not the earliest reported English case relating to pre-formation disclosure in insurance contracts. That honour falls to De Costa v Scandret,4 decided in the Court of Chancery in 1723, in which the assured failed to disclose to the insurer intelligence he had received that the ship proposed for insurance was in danger of being lost. Such ‘concealing’ was considered a ‘fraud’, as disclosure would have led the insurer either to decline the risk or to require a higher premium.5 Another earlier reported case is Seaman v Fonereau6 in 1742, in which an insurer was held entitled to reject a claim for loss by capture on the ground of nondisclosure of information that an accompanying vessel had lost contact with the insured vessel, which was leaking, and that the following day there had been a ‘hard gale’, the materiality of the information being determined as at the time of conclusion of the contract and it being, therefore, irrelevant that the undisclosed matter had no bearing on the casualty. Nor is Carter v Boehm even Lord Mansfield’s earliest reported case on non-disclosure in insurance contracts. Decided one year earlier, conjoined appeals in Salvador v Hopkins and Heaton v Rucker7 were two of nine cases arising out of the insurance of a vessel employed in the East India trade covering a homebound voyage from India to England and prior trading between Indian ports. The underwriters’ allegation of non-disclosure of the assured’s expectations as to the extent of such prior trading was rejected on the basis that the customary trading practices within the East India trade were sufficiently notorious and well known to the insurers as to fall within their deemed knowledge. It was in Carter v Boehm, however, that Lord Mansfield articulated, for the first time, a coherent set of principles for English law. Secondly, neither Carter v Boehm nor the earlier cases just mentioned constitute the origin of the assured’s pre-formation disclosure obligations in English law. The London marine insurance customs drawn up at the instigation of the Privy Council in the 1570s in A Booke of Orders of Assurances within the Royall Exchange (also known as the London Code) provided that the

2 Carter v Boehm (1766) 3 Burr 1905, 97 ER 1162. 3 eg the leading monograph, P MacDonald Eggers and S Picken, Good Faith and Insurance Contracts, 4th edn (London, Informa Law, 2017) para 1.01. 4 De Costa v Scandret (1723) 2 P Wms 170, 24 ER 686. 5 The same point apparently arose the following term in a second, unreported, case called Weaver v Fowler: M Postlethwayt, The Universal Dictionary of Trade and Commerce, 4th edn (London, W Strahan and others, 1774) vol 1, sv ‘Assurance’ and ‘Remarks’. 6 Seaman v Fonereau (1742) 2 Str 1183, 93 ER 1115. 7 Salvador v Hopkins (1765) 3 Burr 1707, 97 ER 1057.

The Three Ages of Utmost Good Faith 65 assured ‘shall not conceale any thing, that may tend to the hurt and hindrance of the Assurer, but with playne and true meaning shall give and continue his Assurreance’.8 This provision, it has been suggested, was derived from Dutch insurance legislation of 1571,9 although the assured’s obligation under that legislation was confined to stating in the policy certain specified information.10 Thirdly, Carter v Boehm is not the origin of the idea that an enhanced concept of good faith applies to insurance contracts, and therefore not the origin of the application to insurance contracts of a turn of phrase so indicating. As is well known, Lord Mansfield invoked a concept of ‘good faith’, unadorned by any epithet, which ‘forbids either party by concealing what he privately knows, to draw the other into a bargain, from his ignorance of that fact, and his believing the contrary’.11 Nor does Lord Mansfield’s judgment break new ground in English jurisprudence in connecting the requirement to disclose information with a principle of good faith. In the unreported 1743 case of Rooke v Thurmond,12 the defendant reinsured a voyage policy on a ship but refused to honour a claim for a total loss by capture on the basis that at the time of taking out the reinsurance the claimant knew but failed to inform the defendant that the insured ship was already overdue. Lee CJ held that in principle the alleged facts established a good defence because ‘these Contracts are made upon mutual Faith and Credit, and … to conceal such Circumstances, which may make any Difference in the Adventure, is fraudulent, for the Insurer ought to have the Advantage of Judgment upon them’. For Lord Mansfield, the injunction against concealment was not confined to insurance contract law: ‘The governing principle is applicable to all contracts and dealings.’13 This proposition, of course, did not survive. Instead, English contract law as a general proposition endorsed the right to profit from information advantage, ‘preferring the benefits of simplicity and certainty which flow from requiring those engaging in commerce to look after their own interests’.14 That obviously did not mean that English contract law endorsed bad faith, either generally or in the specific context of contractual negotiations. Notably, active misleading into a contract triggered rights of recourse for misrepresentation at common law in cases of fraud or where the misrepresentation resulted in a total failure of consideration,15 and more broadly in equity,16 but there

8 MS Harleian 5103, para 35 (likewise MS Additional 48023, para 39), reproduced in G Rossi, Insurance in Elizabethan England (Cambridge, Cambridge University Press, 2016) 546–47. 9 Rossi (n 8) 245. 10 JP van Niekirk, The Development of the Principles of Insurance Law in the Netherlands from 1500 to 1800 (Cape Town, Juta & Co Ltd, 1998) 494–97, 501. This was the approach of most legislation of the time. In contrast, the Ordinance of Bilbao 1573, having specifically addressed insurance after loss or after safe arrival, contemplated a more general disclosure obligation. By virtue of Art XXVIII, the insurer and assured ‘ought, when they shall go sign a Policy, or to treat, and agree on the Premium, to manifest to the Person, who shall intervene, the good, or bad Advices that they shall have of the Ship, and Cargo; that they may thereon treat of an Agreement for the Premium’. 11 Carter (n 2) 1910; 1164. 12 Postlethwayt (n 5) sv ‘Assurance’ and ‘Further Remarks’; N Magens, An Essay on Insurances (London, J Haberkorn, 1755) vol 1, 85. 13 Carter (n 2) 1910; 1164. Even this is foreshadowed in earlier case law. The year before Carter v Boehm, in Hodgson v Richardson (1764) 1 Wm Bl 463, 96 ER 268, the Court of King’s Bench held an insurer was not liable where the assured had stated the port of loading to be Genoa whereas in truth it was Leghorn. Yates J, analysing the case as non-disclosure of Leghorn as the port of loading, stated as follows (ibid 465; 269): ‘The concealment of material circumstances vitiates all contracts, upon the principles of natural law. A man, if kept ignorant of any material ingredient, may safely say it is not his contract.’ Lord Mansfield analysed the case as misrepresentation of Genoa as the port of loading, and therefore did not address non-disclosure. 14 Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd (The Star Sea) [2001] UKHL 1, [2003] 1 AC 469 [45] (Lord Hobhouse). 15 Kennedy v Panama, New Zealand & Australian Royal Mail Co (Ltd) (1866–67) LR 2 QB 580, 587. 16 Reese River Silver Mining Co Ltd v Smith (1869) LR 4 HL 64, 79–80; Lamare v Dixon (1873) LR 6 HL 414; Redgrave v Hurd (1881) 20 Ch D 1 (CA).

66 Howard Bennett was no culpability in failing to volunteer disclosure of information germane to a prospective transaction.17 Insurance contract law, however, made this additional demand, requiring complete openness and the spontaneous disclosure of information material to the risk regardless of potential resulting prejudice to the assured through a refusal of cover or less favourable terms.18 With respect to the origin of the phrase, Lord Hobhouse in The Star Sea19 suggested that: It was probably the need to distinguish those transactions to which Lord Mansfield’s principle still applied which led to the coining of the phrases ‘utmost’ good faith and ‘uberrimae fidei’, phrases not used by Lord Mansfield and which only seem to have become current in the 19th century.

He also attributed the first reported judicial use of such a phrase to Rolfe B in Dalglish v Jarvie20 in 1850.21 This historical account is, however, at least partly inaccurate. Any accolade for first judicial use of the phrase ‘utmost good faith’ or its Latin equivalent belongs not, as asserted, to Rolfe B in 1850 but rather to Buller J in the 1798 case of Wolff v Horncastle.22 The context, moreover, was not the pre-formation disclosure of information to the insurer, but legislative compliance in the wording of marine policies. The Insurances on Ships etc Act 178523 was passed to preclude the issuing of policies in blank that failed to identify the assured, a practice that was conducive to fraud. The terms of the 1785 Act proved unduly restrictive, however, and after only three years it was repealed and replaced by the Marine Insurance Act 1788.24 The issue in Wolff v Horncastle was whether a cargo policy satisfied the requirements of the later statute. Buller J outlined the following approach to the legislation: Now we are bound to say that this second statute must receive the most liberal construction that the words will bear. From the language of the two statutes, as well as the consideration that we are construing a contract uberrimae fide; viz. a policy of insurance, we must avoid bearing harder upon the Plaintiffs than is absolutely necessary.25

17 See notably: Fox v Mackreth (1788) 2 Bro CC 400, 420; 29 ER 224, 234; Turner v Harvey (1821) Jac 169, 178; 37 ER 814, 817–18; Walker v Jackson (1842) 10 M & W 161, 168–69; 152 ER 424, 427–28; Keates v Earl of Cadogan (1851) 10 CB 591, 138 ER 234; Walters v Morgan (1861) 3 De G F & J 718, 723; 45 ER 1056, 1058–59; Ward v Hobbs (1878) 4 App Cas 13 (HL) 26. 18 Although the assured’s disclosure obligation as it was developed in the 19th century came to exhibit a severity beyond that contemplated by Lord Mansfield: R Hasson, ‘The Doctrine of Uberrima Fides in Insurance Law – A Critical Evaluation’ (1969) 32 MLR 615. Notably, for fear of encouraging the deliberate withholding of information, the courts became less receptive to arguments that the insurer ought to have known of a particular circumstance so that it did not need to be brought to the insurer’s attention by the assured: contrast Bates v Hewitt (1867) LR 2 QB 595 (QB) with the facts of and result in Carter (n 2) itself. 19 The Star Sea (n 14) [44]. 20 Dalglish v Jarvie (1850) 2 Mac & G 231, 42 ER 89. 21 Likewise J Botes, From Good Faith to Utmost Good Faith in Marine Insurance (Frankfurt am Main, Peter Lang, 2006) 85. 22 Wolff v Horncastle (1798) 1 Bos & P 316, 126 ER 924. 23 25 Geo III, c 44. 24 28 Geo III, c 56. 25 Wolff (n 22) 1 Bos & P 316, 322; 126 ER 924, 928. Absent this reference to utmost good faith from Buller J, the accolade for first mention would still not go to Rolfe B, but to Best CJ. In Palmer v Pratt (1824) 2 Bing 185, 130 ER 277, it was held that no claim could be made under an insurance policy covering the non-occurrence of a contingency that triggered payment under bills of exchange when the bills in question proved to be a nullity. Best CJ commented (ibid 191; 279) that: ‘I regret that we are compelled to come to this decision, because insurance transactions ought to be conducted uberrimâ fide; but the [use of insurance policies to back up bills of exchange] is a new practice, and the parties who engage in it must take the consequence.’ In Williams v Rawlinson (1825) 3 Bing 71, 130 ER 440, moreover, the provider of a bond alleged deception as to the secured indebtedness. In rejecting this defence on the facts, Best CJ stated: ‘It has been argued, that the Defendant’s undertaking is analogous to an insurance; a transaction, in which, according to Lord Mansfield, there must always be uberrima fides.’ The reference is to Carter v Boehm, as indicated by the continuation ‘but the same learned

The Three Ages of Utmost Good Faith 67 Utmost good faith was invoked, therefore, as an aid to statutory interpretation. There is, moreover, no indication that Buller J considered he was introducing a novelty of either substance or rhetoric. On the contrary, he appears to be praying in aid a characteristic of insurance contracts that requires no explanation and by means of a phrase that will be readily recognised. While the precise timing of and context for the initial association of insurance contracts with the phrase ‘utmost good faith’ remain unknown, it probably pre-dates the emergence of a clear contrast between general contract law and insurance contract law with respect to information advantage. Two further points may be made about the relationship between the idea of ‘utmost good faith’ and insurance contract law in the nineteenth century. First, the phrase ‘utmost good faith’ was not associated uniquely or even primarily with insurance law. Instead, wherever the context rendered it appropriate to require full frankness and openness, to the prejudice, if necessary, of one’s own interests, reference might be made to utmost good faith. In addition to insurance contracts, the phrase was pressed into service in connection with partnerships,26 the obtaining by a fiduciary of full informed consent for a transaction otherwise in breach of fiduciary duty,27 compliance with a trustee’s duties,28 suretyship agreements,29 patent applications,30 the promotion of companies,31 the proposal by a debtor of a scheme of arrangement with creditors,32 an application for an injunction33 and the obtaining of probate.34 Secondly, in the context of insurance contracts, the idea of an enhanced standard of good faith was most commonly associated with pre-formation disclosure obligations,35 but not exclusively. The order for ship’s papers was an order for accelerated, extended disclosure of documents relevant to a claim on a marine policy developed at common law at a time when conventional disclosure orders were available only from courts of equity.36 The commercial justification lay person, upon the occasion in which he established that position, referred also to the maxim, aliud est tacere, aliud celare’. See Carter (n 2) 1910; 1164. For other references to an enhanced standard of good faith pre-dating the judgment of Rolfe B in Dalglish (n 20), see Hercules Insurance Co v Hunter (1836) 14 S 1137 (IH) 1140 (Lord Moncreiff) (‘the strictest equity’, ‘the most rigid good faith’); Wainewright v Bland (1835) 1 Mood & R 481, 486; 174 ER 165, 167–68 (Lord Abinger) (‘the greatest good faith’). 26 Clements v Hall (1858) 2 De G & J 173, 188; 44 ER 954, 960; Tennent v Tennent’s Trustees (1870–75) LR 2 Sc 6 (HL) 10; Barnes v Youngs [1898] 1 Ch 414 (Ch) 419. 27 McPherson v Watt (1877) 3 App Cas 254 (HL) 266; Dean v Macdowell (1878) 8 Ch D 345 (CA) 350–51; Law v Law [1905] 1 Ch 140 (CA) 148. 28 Fraser v Murdoch (1880) 7 R 694 (IH) 711. 29 Newton v Chorlton (1853) 10 Hare 646, 649; 68 ER 1087, 1088–89, referring also to the extensive disclosure obligation stated in Owen v Homan (1851) 3 Mac & G 378, 42 ER 307, where a comparison was drawn with insurance (ibid 397; 315). 30 Sturz v De La Rue (1828) 5 Russ 322, 327; 38 ER 1048, 1050; Re Clark’s Patent (1870) 7 Moo PC NS 255, 262; 17 ER 97, 100; Re Pitman’s Patent (1871) LR 4 PC 84, 86. 31 Ross v Estates Investment Co (1866) LR 3 Eq 122 (Ch) 136; Erlanger v New Sombrero Phosphate Co (1878) 3 App Cas 1218 (HL) 1255. 32 Re Moon (1887) 19 QBD 669 (CA) 672. 33 Dalglish (n 20) 243–44; 94. See also R v Kensington Income Tax Commissioners, ex parte Princess Edmond de Polignac [1917] 1 KB 486 (CA) (writ of prohibition). 34 Attorney-General v Smith & Cocks [1892] 2 QB 289 (QB) 293. 35 Fawcus v Sarsfield (1856) 6 El & Bl 192, 202; 119 ER 836, 839–40; Wheelton v Hardisty (1857) 8 El & Bl 232, 270; 120 ER 86, 101; Belfast & Ballymena Railway Co v Keys (1861) 9 HLC 556, 563; 11 ER 846, 849; Gandy v Adelaide Marine Insurance Co (1871) LR 6 QB 746 (QB) 758; Life Association of Scotland v Foster (1873) 11 M 351 (IH) 359; Robinson v Mollett (1874) LR 7 HL 802, 817; Lishman v Northern Maritime Insurance Co (1875) LR 10 CP 179, 182; London Assurance Co v Mansel (1879) 11 Ch D 363 (Ch) 368; Brownlie v Campbell (1880) 5 App Cas 925 (HL) 954; Blackburn Low & Co v Vigors (1886) 17 QBD 553 (CA) 584. 36 Goldschmidt v Marryat (1809) 1 Camp 558, 170 ER 1057; Leon v Casey [1932] 2 KB 576 (CA) 579–81; Probatina Shipping Co Ltd v Sun Insurance Office Ltd (The Sageorge) [1974] QB 635 (CA) 641. Because it was designed to enable insurers to assess a claim, the order was available as soon as proceedings had been commenced by the assured, with no requirement for the insurer first to serve a defence. The order was directed at the assured but instructed disclosure of relevant documents in the possession of not only the assured but also any other party, or provision of evidence of having tried to obtain them, albeit unsuccessfully.

68 Howard Bennett in the practical disadvantage faced by marine insurers from the potential for the circumstances surrounding the casualty to be predominantly and particularly within the knowledge of the assured. In terms of legal principle, however, reference was made to enhanced good faith to justify the timing and extent of the order,37 and perhaps also to emphasise the limits of the encroachment by the common law courts onto previously equitable territory. An enhanced standard of good faith was invoked also in connection with fraudulent claims. In Britton v Royal Insurance Co,38 a policy against loss by fire provided that a fraudulent claim would render the policy void. In addressing the jury, Willes J emphasised that a claim was no less fraudulent even if the loss had not been wilfully procured but the assured had wilfully inflated the extent of the loss sustained by reason of a genuine casualty: The law upon such a case is in accordance with justice, and also with sound policy. The law is, that a person who has made such a fraudulent claim could not be permitted to recover at all. The contract of insurance is one of perfect good faith on both sides, and it is most important that such good faith should be maintained. It is the common practice to insert in fire-policies conditions that they shall be void in the event of a fraudulent claim; and there was such a condition in the present case. Such a condition is only in accordance with legal principle and sound policy. It would be most dangerous to permit parties to practise such frauds, and then, notwithstanding their falsehood and fraud, to recover the real value of the goods consumed.39

Fundamentally, therefore, an assured could not be allowed to think that a genuine claim could be fraudulently exaggerated without penalty. Law that allowed the assured still to recover an indemnity for the true loss would provide no deterrent against fraud and could even incentivise it. This simple point does not depend on a concept of enhanced good faith, but its invocation served perhaps to reinforce the repugnance of fraud to insurance. What, then, was the role of ‘utmost good faith’ with respect to insurance contract law during the first age of its existence? A key question is the basis and rationale for the central duty associated with the concept of utmost good faith, namely the assured’s pre-formation disclosure obligation. Recognition of this obligation resulted from the fact that an insurer in essence invested in a package of information the content of which it had to take largely on trust. The insurer assessed and priced risk acceptance necessarily in reliance on the description provided by the assured, more rigorous due diligence often being rendered difficult or impossible by the geographically remote location of the insured property, by the lack of tangible manifestation susceptible to inspection of relevant circumstances, such as those pertaining to moral hazard, or by market expectation and practice as to the rapidity of insurers’ decision-making.40 This was, indeed, the foundation for Lord Mansfield’s exposition of principle in Carter v Boehm:41 Insurance is a contract upon speculation. The special facts, upon which the contingent chance is to be computed, lie most commonly in the knowledge of the insured only; the under-writer trusts to his 37 Rayner v Ritson (1865) 6 B & S 888, 890–91; 122 ER 1421, 1421–22 (Cockburn CJ): ‘The exceptional practice in these cases seems to have arisen out of the particular nature of the contract of insurance. The underwriter of a policy of marine insurance who is sued for a constructive total loss of the ship is so much at the mercy of the assured with respect to the circumstances under which the vessel has been abandoned, and there ought to be uberrima fides on the part of the latter, and he ought therefore to lay those circumstances before the underwriter.’ See also China Traders’ Insurance Co Ltd v Royal Exchange Assurance Corp [1898] 2 QB 187 (CA) 193–94; Boulton v Houlder Bros & Co [1904] 1 KB 784 (CA) 791–92; Harding v Bussell [1905] 2 KB 83 (CA) 85–86. 38 Britton v Royal Insurance Co (1866) 4 F & F 905, 176 ER 843. 39 ibid 909; 844. 40 North Star Shipping Ltd v Sphere Drake Insurance plc (No 2) [2005] EWHC 665 (Comm), [2005] 2 Lloyd’s Rep 76 [256]; Versloot Dredging BV v HDI Gerling Industrie Versicherung AG (The DC Merwestone) [2016] UKSC 45, [2017] AC 1 [114]. 41 Carter (n 2) 1909; 1164.

The Three Ages of Utmost Good Faith 69 representation, and proceeds upon confidence that he does not keep back any circumstance in his knowledge, to mislead the under-writer into a belief that the circumstance does not exist, and to induce him to estimate the risque, as if it did not exist.

Moreover, whilst the operative concept was said to be ‘concealment’ and whilst non-disclosure was sometimes described as ‘fraudulent’, any ‘deception’ reflected the erroneous picture in the mind of the insurer rather than any moral culpability attaching to the assured for duplicity or sharp practice in engendering that false understanding:42 such ‘fraud’, therefore, ‘was not actual fraud as known to the common law but a form of mistake of which the other party was not allowed to take advantage’.43 The assured was not allowed to take advantage of a misapprehension on the part of the insurer resulting from information disadvantage. The reasoning, therefore, was that the commercial realities of insurance required a distinctive regime of disclosure obligations, which then attracted the rhetorical label ‘utmost good faith’. English law did not recognise ‘utmost good faith’ as a characteristic of certain contracts, which in turn triggered the applicability of pre-formation disclosure obligations. ‘Utmost good faith’, in short, was little more than a slogan describing a legal response to the transactional dynamic between the parties to an insurance contract.44 The normative underpinnings for the assured’s disclosure obligations lay in that dynamic, not some abstract status of insurance contracts as ‘of the utmost good faith’. Nor do the references to utmost good faith in other contexts indicate a significant normative potency. When the practice of making extended disclosure orders for ship’s papers was challenged in 1809, the practice was justified by Sir James Mansfield CJ by pointing to the practical disadvantages of being compelled to follow the standard procedure of seeking disclosure in equity.45 No reference was made to (marine) insurance contracts being of the utmost good faith. That gloss was added only some decades later. As noted above, the judgment of Willes J in Britton v Royal Insurance Co46 on fraudulent claims is based on sound, pragmatic concerns relating to deterring fraud. Utmost good faith, although rhetorically flourished, contributes nothing of itself. Equally, garnishing the inability to recover for losses caused by the wilful misconduct of the assured with the ‘greatest good faith’ that is required of an assured47 adds nothing to the public policy and construction barriers to recovery.48 Likewise, the decision in Wolff v Horncastle49 is founded on the background to and wording of the relevant legislation. Utmost good faith was, admittedly, invoked in Wolff as at least burnishing the lens through which the legislation should be read, but a less welcoming approach was evidenced by the Privy Council in Beacon Life & Fire Assurance Co v Gibb with respect to policy interpretation.50 A policy form designed for house insurance was used for the insurance of a vessel. The policy excluded liability for any loss occurring at a time when more than 20 lb of gunpowder were ‘on the premises’. At the time of the loss, the vessel’s cargo included more than 20 lb of g unpowder. The Canadian 42 ibid (Lord Mansfield): ‘Although the suppression should happen through mistake, without any fraudulent intention; yet still the under-writer is deceived, and the policy is void; because the risque run is really different from the risque understood and intended to be run, at the time of the agreement.’ 43 The Star Sea (n 14) [42] (Lord Hobhouse). 44 See also A Forte, ‘Good Faith and Utmost Good Faith: Insurance and Cautionary Obligations in Scots Law’ in A Forte (ed), Good Faith in Contract and Property Law (Oxford, Hart Publishing, 1999) 81–86. 45 Goldschmidt (n 36). 46 Britton (n 38). 47 Wainewright (n 25) 486; 167–68 (Lord Abinger). 48 Beresford v Royal Insurance Co Ltd [1938] AC 586 (HL); Porter v Zurich Insurance Co [2009] EWHC 376 (QB), [2009] 2 All ER (Comm) 658. 49 Wolff (n 22). 50 Beacon Life & Fire Assurance Co v Gibb (1862) 1 Moo PC NS 73, 15 ER 630.

70 Howard Bennett appellate court held the exclusion inapplicable given that the policy did not preclude the carriage of gunpowder and that its presence on board did not cause the loss. A denial of indemnification in such circumstances would be contrary to good faith in the absence of more specific wording. According to Badgley J: The law of insurance requires uberrima fides from the insured. The like is or should be imposed upon the insurers, who should have stipulated their refusal to allow gunpowder to be carried as freight. It was in their power to condition their own risk in this respect, and to render the condition plain, precise, and positive, instead of leaving it uncertain and doubtful, and contrary to law, casting the doubt from themselves upon the insured. The rule of law should be held against them. The construction of the contract must be strictly against those for whose benefit the conditions are introduced when they impose burdens on the parties; otherwise, fraud would be paramount, to the exclusion of good faith.51

The Privy Council, however, adopted a different approach and upheld the insurers’ appeal. The exclusion was unambiguous on its wording and applicable. The only way to avoid the exclusion was to import additional words that would change its meaning, for which there was no licence. Neither the exclusion’s wording nor insurance law principle rendered causation of any relevance. Whatever the potential for utmost good faith to influence construction of less clearly worded policies, as to which the Privy Council was silent, a general concept of utmost good faith could not impinge on unambiguous wording. In conclusion, while encapsulating by way of useful shorthand the law’s adoption of especially stringent standards in the context of the negotiation of insurance cover,52 the concept of utmost good faith in its first age lacked normative potency of itself and exerted at most a marginal influence on the development of insurance law doctrine. The idea that insurance contracts are contracts of the utmost good faith certainly seems to have provided welcome support in a variety of contexts, but there is no indication that this idea made a decisive contribution to the courts’ thinking. In short, even without the recognition of a general concept or idea of ‘utmost good faith’, insurance contract law would probably have developed in an identical fashion up to the beginning of the twentieth century.

III. The Second Age: The Marine Insurance Act 1906 The advent of the Marine Insurance Act 1906 heralded a change in the status and prominence of the concept of utmost good faith. In the scheme of Sir Mackenzie Chalmers, the concept was elevated to a fundamental, organising principle of insurance law.53 A group of sections c ollectively headed ‘Disclosure and Representations’ is introduced by section 17, which carries the marginal heading ‘Insurance is uberrimae fidei’ and which provides as follows: A contract of marine insurance is a contract based upon the utmost good faith, and, if the utmost good faith be not observed by either party, the contract may be avoided by the other party. 51 ibid 85–86; 634–35. 52 Seaton v Heath [1899] 1 QB 782 (CA) 792 (Romer LJ: uberrima fides a ‘short and convenient’ expression, denoting the requirement that an assured ‘in setting forth the risk to be insured against, not conceal any material fact affecting the risk known to him’). 53 The relevant provisions of the 1906 Act state for the purposes of marine insurance contracts principles that it is accepted are applicable to all insurance contracts: Joel v Law Union & Crown Insurance Co [1908] 2 KB 863 (CA); Lambert v Co-op Insurance Society Ltd [1975] 2 Lloyd’s Rep 485 (CA); Pan Atlantic Insurance Co Ltd v Pine Top Insurance Co Ltd [1995] 1 AC 501 (HL) 518.

The Three Ages of Utmost Good Faith 71 Utmost good faith in the era of the 1906 Act presents, nevertheless, a structural puzzle, namely the relationship between the utmost good faith of which section 17 speaks and all aspects of utmost good faith as recognised in insurance contract law. There are two possibilities. The first is that section 17 acts as an umbrella provision for the entirety of the doctrine, including those aspects of the doctrine which are addressed explicitly by the 1906 Act and other aspects in respect of which the Act is silent. The second is that the doctrine of utmost good faith is broader than section 17, which acts as an umbrella provision for certain aspects of the doctrine that attach to the negotiation of an insurance contract or its variation but which leaves other aspects free to develop outside the confines of that section. Fundamental to this question is the issue of remedy. Section 17 provides unambiguously that a failure by either party to meet the requirements of utmost good faith affords the other party the right retrospectively to avoid the contract. In the context of insurance contracts, such avoidance is usually invoked by an insurer in the light of circumstances discovered when investigating a claim and will retrospectively deprive the assured of cover at a time when it is too late to obtain alternative insurance. In addition, the insurer will be entitled to repayment of any sums paid in satisfaction of previous claims. The question of the precise role and scope of section 17 is fundamentally about the availability of this remedy that, within the framework of general contract law, responds to a defect in the formation of the contract. It is convenient to consider the structural issues generated by section 17, and thereby the role of utmost good faith in its second age, by distinguishing between utmost good faith before the formation of an insurance contract and utmost good faith during the lifetime of an insurance contract after it has been created. Some conclusions will then be drawn.

A. Pre-formation Utmost Good Faith Utmost good faith in the pre-formation stage relates to the initial presentation of a risk to an insurer prior to the formation of an insurance contract. Section 18 addresses the assured’s pre-formation disclosure obligation, section 19 disclosure by an agent effecting insurance on behalf of the assured and section 20 misrepresentation by either the assured or an agent. These sections, it has been consistently averred, articulate aspects of the broader doctrine of utmost good faith as stated by section 17.54 This analysis of the assured’s obligations of disclosure under section 18 and truth of representation under section 20 is uncontroversial on the wording of the relevant sections, which accurately reflect the earlier case law they codify. As already discussed, utmost good faith became part of the language of insurance contract law primarily, although not exclusively, to highlight the assured’s pre-formation disclosure obligations. Section 19, however, is more difficult. It addresses disclosure where insurance is effected, as is almost invariably the case, through a broker. The circumstances that require disclosure include those which are in fact known by the broker and those which ought to be known by or to have been communicated to the broker, irrespective of whether the broker’s knowledge, actual or constructive, arose in the context of the broker’s

54 Container Transport International Inc v Oceanus Mutual Underwriting Association (Bermuda) Ltd (No 1) [1982] 2 Lloyd’s Rep 178 (Com Ct) 187, [1984] 1 Lloyd’s Rep 476 (CA) 492, 496, 512, 525; Société Anonyme d’Intermediaries Luxembourgeois (SAIL) v Farex Gie [1995] LRLR 116 (CA) 142; PCW Syndicates v PCW Reinsurers [1996] 1 WLR 1136 (CA) 1145; Group Josi Re Co SA v Walbrook Insurance Co Ltd [1996] 1 WLR 1152 (CA) 1170.

72 Howard Bennett agency for the assured. Critically, section 19 is construed as addressing a disclosure obligation owed personally by a placing broker to the insurer. The broker’s knowledge is not attributed to the assured.55 Otherwise, section 19 would be otiose, since the knowledge of the agent would fall within the scope of the assured’s disclosure obligation.56 Section 17 refers, however, to the observance of utmost good faith by the parties to the contract. Section 19 is concerned with the observance of utmost good faith in relation to an insurance contract, but not by a party to that contract. It would be possible to connect non-disclosure by the broker with a failure on the part of the assured to observe the utmost good faith if that doctrine generated an implied undertaking by the assured in an insurance contract that there had been no failure to make due disclosure by either the assured or the broker. While, however, there was some support for such an approach in the nineteenth century,57 it is now settled that the pre-formation duties relating to disclosure and truth of statements arise as a matter of law as legal incidents to the negotiation of an insurance contract, and do not operate by means of implied terms. This is consistent specifically with non-compliance giving rise to a right of avoidance, rather than a remedy for breach of a contractual term,58 and more broadly with general (contract) law.59 The conclusion has to be that section 19, as understood by the case law, does not fit within section 17. The placing broker’s duty of disclosure forms part of the doctrine of utmost good faith, but outside of section 17.60 The disclosure obligations that attach to the initial negotiation of an insurance contract manifest the law’s rejection of a right on the part of the assured to benefit from information advantage. When deciding whether to accept the presented risk and, if so, on what terms, the insurer is entitled to parity of information with the assured and the placing broker. That entitlement applies equally on any subsequent occasion on which the assured asks the insurer to make an underwriting decision, namely a decision about acceptability of additional or altered risk and/or the terms of acceptance of such risk. Consequently, disclosure of material circumstances as well as truth of statements is required if the assured requests the insurer to vary an existing contract of insurance. How does this duty of disclosure fit into the doctrine of utmost good faith and the provisions of the 1906 Act? Section 17 provides by way of remedy for failure to observe the utmost good faith that ‘the contract may be avoided’. The relationship between utmost good faith as attaching to variations of cover and section 17 depends on what is meant by ‘the contract’: in the context of a variation, does it mean the original contract or only the variation?

55 Blackburn Low & Co v Vigors (1887) 12 App Cas 531 (HL) 542–43; El Ajou v Dollar Land Holdings plc (No 1) [1994] 2 All ER 685 (CA) 702; SAIL (n 54) 142–43, 150; PCW Syndicates (n 54) 1145. 56 SAIL (n 54) 150. 57 Blackburn Low & Co (CA) (n 35) 562–63, 578. 58 Banque Keyser Ullmann SA v Skandia (UK) Insurance Co Ltd [1990] 1 QB 665 (CA) 702 (Steyn J), 779–80 (Slade LJ); The Star Sea (n 14) [46], [52]. 59 See, eg the remedial consequences of incorporation of a misrepresentation as a term of the contract. 60 The structural issue raised in the text could perhaps be avoided by a different reading of the legislation. Arguably, s 18 should apply where the assured places the insurance directly with the insurer, while s 19 should apply instead, to the exclusion of s 18, where, as in practice is invariably the case with commercial policies, the insurance is placed by a broker acting on behalf of the assured. In such a case, the disclosure obligation would extend, as stated by s 19, to any material circumstance either (a) within the broker’s actual and constructive knowledge or (b) that would fall to be disclosed by the assured under s 18 if that section applied provided it came to the assured’s knowledge in time to communicate to the broker for disclosure by the broker to the underwriter. Non-disclosure by the broker would be attributable to the assured in accordance with general agency principle and naturally render the policy voidable. On the prevailing analysis, it is puzzling why the breach of a duty owed independently by the broker, and not qua agent of the assured, should prejudice the assured’s contract, especially given the rejection of an implied term analysis. A different approach would also explain why there is no reference in the Act to misrepresentation by the broker. Section 20 refers to misrepresentation ‘by the assured or his agent’, but if the broker owes an independent duty to the insurer in respect of disclosure, it must also owe a parallel independent duty in respect of representations which is not encompassed by a reference to agency.

The Three Ages of Utmost Good Faith 73 One must first clarify the precise consequences of non-disclosure (or misrepresentation) in the context of a policy variation. It is clear that the disclosure obligation attaching to a variation is restricted in scope to circumstances material to the variation and does not extend to circumstances material not to the variation but to the original risk. Disclosure being designed to produce parity of information relating to an underwriting decision, once that decision is taken the obligation to disclose information material to it ceases. The scope of any new disclosure obligation subsequently arising in relation to a new underwriting decision is then delineated by what is material to that decision.61 Similarly, the requirement for subjective inducement of the insurer’s decision to contract on the terms agreed, recognised as necessary to render a non-disclosure or misrepresentation actionable in the context of the original formation of an insurance contract,62 applies equally to variations of cover. There can be no remedy unless the non-disclosure or misrepresentation material to the variation in fact induced the insurer into the variation.63 Equally, the remedy for non-disclosure must reflect the context of the wrong. What is vitiated is the underwriting decision in relation to which the insurer was denied the parity of information to which it was entitled, ie the decision that was induced by the denial of (accurate) information material to it. The remedy should be the reversal of that decision.64 In a case of non-disclosure or misrepresentation in relation to a variation, avoidance of not just the variation but of the original policy itself would be both illogical and indefensibly disproportionate and punitive.65 This may be achieved within the terms of section 17 on the basis that a variation constitutes an independent, severable contract that attaches to the original contract but can be detached from it by avoidance.66 In the context of a variation, ‘the contract’ that section 17 states may be avoided in the event of a failure to observe the utmost good faith is the contract to vary. Variations, therefore, give rise to a standard application of the pre-formation doctrine of utmost good faith and of sections 18–20 of the 1906 Act.67 In the above discussion, section 17 itself and the concept of utmost good faith have little substantive impact. The duties whose breach may be asserted by an insurer are articulated by sections 18–20, which also reiterate the remedy of avoidance. Nevertheless, while utmost good faith constitutes a broad, organising principle (albeit that the placing broker’s duties do not fit within the drafting of section 17), in certain respects section 17 fulfils an independent function. First, section 17 articulates the mutual nature of the doctrine: the insurer owes the assured reciprocal obligations of disclosure and truth of statement. The following sections are confined to the obligations owed by the assured and the placing broker. Although not prominent in the case law, insurers were known on occasion to exploit an information advantage in relation to the fact that the vessel or cargo presented for insurance had already arrived safely and to accept premium for, in truth, no assumption of risk.68 Secondly, sections 18–20 confer a remedy in relation to

61 Lishman (n 35) 182; Iron Trades Mutual Insurance Co Ltd v Companhia de Seguros Imperio [1991] 1 Re LR 213 (Com Ct) 224; Black King Shipping Corp v Massie (The Litsion Pride) [1985] 1 Lloyd’s Rep 437 (Com Ct) 511. 62 Pan Atlantic Insurance (n 53). 63 AC Ward & Son Ltd v Catlin (Five) Ltd (No 2) [2009] EWHC 3122 (Comm), [2010] Lloyd’s Rep IR 695 [195]. 64 Container Transport (Com Ct) (n 54) 191–92; AC Ward & Son Ltd (n 63) [233]. 65 K/S Merc-Scandia XXXXII v Lloyd’s Underwriters (The Mercandian Continent) [2001] EWCA Civ 1275, [2001] 2 Lloyd’s Rep 563 [22(2)]. 66 Occidental Worldwide Investment Corp v Skibs A/S Avanti (The Siboen and the Sibotre) [1976] 1 Lloyd’s Rep 293 (Com Ct) 340. 67 See, eg, Fraser Shipping Ltd v Colton [1997] 1 Lloyd’s Rep 586 (Com Ct), in which the assured’s notification of a change of voyage was too late to comply with the terms of a held covered clause (analysed below). The notification therefore constituted a request for a variation of terms. Without analysis, the court, correctly it is suggested, applied s 18 to the assured’s disclosure obligations. 68 Carter (n 2) 1909; 1164.

74 Howard Bennett non-disclosure or misrepresentation of a ‘material’ circumstance. Section 17 contains no reference to materiality and would have to be the basis for any right of avoidance in the event of non-disclosure or misrepresentation not dependent on proof of materiality. It is clear that a fraudulent assured cannot resist avoidance on the basis that the concealed or misstated circumstance lacked materiality. The weight of authority supports the simple technical explanation that materiality is not required as a matter of principle,69 in which case section 17 would have to apply, although it is possible that materiality is notionally required but a fraudulent assured is not permitted to deny the materiality of its fraud,70 in which case sections 18–20 would still apply. Thirdly, where an insurer requests information that does not qualify as material, so that disclosure would not be required in the absence of enquiry, an inaccurate response would fall outside section 20 but render the contract voidable under section 17.71 Fourthly, it has been stated that where the assured is aware that the insurer is proceeding on the basis of a mistaken misapprehension as to a material circumstance, such as a level of losses, or has made a material mathematical error, utmost good faith requires the assured to correct the insurer’s error, with silence again rendering the policy voidable under section 17.72

B. Post-formation Utmost Good Faith Once the contract has been concluded, it becomes especially important to distinguish between, on the one hand, the applicability of the doctrine of utmost good faith and, on the other hand, the availability of the remedy of avoidance of the contract prescribed by section 17. In the pre-formation context, the doctrine is concerned entirely with parity of information in relation to underwriting decisions, with non-observance rendering the contract voidable consistently with the remedial response espoused by general contract law to induced defects in consent to the formation of a contract. In the post-formation context, however, utmost good faith has been invoked in a variety of different contexts, none of which, by definition, are concerned with consent to the formation of a contract. Retrospective avoidance in such contexts would then afford the insurer the right to reclaim payment of any sums previously paid in settlement of genuine claims. The existence and amount of such earlier payments being a matter of happenstance, such financial consequences of avoidance would be arbitrary. They would also be inconsistent with the approach of general contract law, which confines remedies for defects in the performance of a contract to compensation for loss caused by the defective performance and loss of future entitlements. Any adoption by insurance contract law of a stricter remedial regime requires reasoned justification, not merely the brandishing of a rhetorical flourish. One clear example of post-formation utmost good faith arises where the assured does not request a consensual alteration to the terms of the contract but activates a conditional extension of cover as a matter of right in accordance with the original terms of the policy. Thus, a ‘held covered clause’ typically provides that, in circumstances in which cover would otherwise prospectively cease (typically by virtue of an alteration of risk or breach of warranty),73 cover will instead continue, provided the assured notifies the insurer of the relevant circumstances 69 The Bedouin [1894] P 1 (CA) 12; Gordon v Street [1899] 2 QB 641 (CA); Rafsanjan Pistachio Producers Co-operative v Bank Leumi (UK) Ltd [1992] 1 Lloyd’s Rep 513 (Com Ct) 542; Pan Atlantic Insurance (n 53) 533. 70 The DC Merwestone (n 40) [31]. 71 Haywood v Rodgers (1804) 4 East 590, 102 ER 957; The Bedouin (n 69) 12; Container Transport (CA) (n 54) 512. 72 Container Transport (CA) (n 54) 512, 522, 525. 73 Under the 1906 Act, such cessation was permanent unless the contract stipulated to the contrary, while the default rule for breach of warranty is changed by the Insurance Act 2015, s 10 to suspension of cover pending cure of breach.

The Three Ages of Utmost Good Faith 75 and agrees to such additional premium and other change of terms as the insurer might reasonably require.74 The insurer has no right to decline this extension of cover, provided the assured complies with the stipulated conditions relating to notification and agreement to altered terms.75 A held covered clause therefore constitutes an option granted to the assured under the terms of the contract which the assured has the right to trigger by fulfilling the stipulated conditions regarding notification and agreement to alteration of terms. There is no new contract to which the insurer’s consent is required. Nevertheless, the setting of appropriate new terms in accordance with the clause constitutes an underwriting decision attracting the right to parity and accuracy of information material to that decision encapsulated by the doctrine of utmost good faith.76 In the event of failure to provide full and accurate information, the assured should be denied the benefit of the clause. The failure to observe the utmost good faith does not, however, impugn a decision by the insurer to conclude a new contract, since, as stated, the insurer is already committed to the extension under the terms of the original policy.77 Rather, it constitutes a failure to satisfy the agreed requirements for the extension of cover. The logical result, therefore, is that no extension comes into effect. This does not fit section 17, which contemplates instead the vitiating of a decision to contract rendering the resulting contract voidable, not the denial of existing contractual rights.78 Beyond underwriting decisions, the key test for the potential post-formation applicability of section 17 is the fraudulent claim. Submitting a fraudulent claim constitutes the most nefarious sin an assured can commit during the lifetime of an insurance contract. If dishonesty in such a core aspect of an insurance contract fails to attract the remedial severity of avoidance, as prescribed by section 17, then it is difficult to see in what circumstances such a remedial response can ever be justified. Some parallels can be drawn between the making of a claim and the negotiation of an insurance contract. As at the pre-formation stage, the assured is asking the insurer to make a decision in circumstances where the assured may have an information advantage. Not only may the assured have direct knowledge of the circumstances surrounding the loss, but thorough and independent investigation by the insurer may not be possible. And extensive verification by the insurer is inimical to the expeditious settlement of claims that assureds like to receive and insurers’ marketing departments like to proclaim.79 The demands placed on the assured are not, however, the same. In the claims context, the law does not require disclosure of all circumstances material to assessment of the claim, but instead concerns itself only with fraud, conferring an entitlement on the insurer not to parity of information but only to honesty.

74 The reasonableness limitation is stated in relation to premium by the Marine Insurance Act 1906, s 31(1). 75 Although the clause cannot operate if its terms cannot accommodate the extended risk. Thus, an extension that would command a reasonable additional premium at least equal to the measure of indemnity for any probable ensuing claim would be self-defeating: Greenock Steamship Co v Maritime Insurance Co Ltd [1903] 1 KB 367 (KB). Likewise, an extension that is uninsurable without an alteration of terms as well as additional premium cannot be accommodated by a clause that provides for additional premium only: Liberian Insurance Agency Inc v Mosse [1977] 2 Lloyd’s Rep 560 (QB). 76 Overseas Commodities Ltd v Style [1958] 1 Lloyd’s Rep 546 (Com Ct) 559; Liberian Insurance Agency (n 75) 568. 77 If a held covered clause gave rise to a distinct contract of variation, the clause itself would be a mere agreement to agree, conferring no rights on the assured. 78 It follows that waiver would have to work by way of estoppel rather than affirmation as in the case of breach of promissory warranty, as to which see Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Ltd (The Good Luck) [1992] 1 AC 233 (HL) 263; Kirkaldy (J) & Sons Ltd v Walker [1999] EWHC 839 (Comm), [1999] Lloyd’s Rep IR 410, 422; HIH Casualty & General Insurance Ltd v Axa Corporate Solutions [2002] EWCA Civ 1253, [2003] Lloyd’s Rep IR 1. 79 The DC Merwestone (n 40) [9], [55], and see [111] (Lord Mance): ‘Insurance is about the assessment of risk and the settlement of claims. Both processes depend on good faith and fair information, and both are normally consensual.’ See also Orakpo v Barclays Insurance Services Co Ltd [1995] LRLR 443 (CA) 451.

76 Howard Bennett With respect to remedy, the potential vulnerability of the insurer in relation to dishonest claims, coupled with the law’s repugnance of fraud,80 generated a rule that fraud in a claim forfeits81 all entitlements under the policy that the assured might otherwise have derived from the circumstances giving rise to the claim. Under an indemnity policy, the assured’s entitlement to an indemnity arises as soon as an insured loss occurs. Fraud on the part of the assured in generating a casualty will prevent any loss being insured in the first place.82 Fraud that subsequently taints a claim in relation to an otherwise insured loss retrospectively removes the assured’s rights in relation to that loss.83 Two specific points need to be noted. First, the fraudulent exaggeration of a genuine claim entitles the insurer to reject the entirety of the claim, resulting in forfeiture of the genuine part of the claim: ‘The logic is simple. The fraudulent insured must not be allowed to think: if the fraud is successful, then I will gain; if it is unsuccessful, I will lose nothing.’84 Secondly, the insurer is entitled to recover any sums already paid towards satisfaction of a claim before that claim becomes tainted by fraud:85 ‘There is no obvious reason why the consequences of making a fraudulent claim should depend upon the timing of any payment in respect of any genuine part of the claim.’86 This forfeiture rule is a ‘special common law rule’87 attaching to the making of claims under insurance contracts which operates independently of any discharge of the contract, whether prospective or retrospective. Forfeiture is not a by-product of discharge for repudiatory breach or avoidance for failure to observe the utmost good faith. It is an incident of insurance contracts born, as stated, of the insurer’s dependence on information from the assured and the law’s repugnance of fraud. It has, nevertheless, been connected in several judgments to the doctrine of utmost good faith (with reference being made to the judgment of Willes J in Britton v Royal Insurance Co88), at least to the extent of the doctrine providing inspiration and support for the rule’s development.89 And, indeed, insurance contract law departs from the general law of obligations in penalising dishonesty. As several Commercial Court judges observed to the Law Commissions, a personal injury claimant who dishonestly exaggerates the extent of the injuries sustained still recovers damages in respect of the true injuries.90 Utmost good faith, in that it encapsulates enhanced legal expectations born of dependence on the other party for critical information in circumstances where the normal expectation of undertaking one’s own due diligence is unrealistic, explains insurance law’s severity. It has never been authoritatively decided, however, whether a fraudulent claim constitutes a failure to observe the utmost good faith within section 17 of the 1906 Act, so that the insurer will also have a right of retrospective avoidance of the policy. It has, nevertheless, been indicated obiter on several occasions that any further remedial consequence

80 The DC Merwestone (n 40) [9], [94]. 81 ibid [24]. 82 That is not to suggest that the absence of an insured loss means that the law on fraudulent claims does not apply to the ensuing claim. 83 AXA General Insurance Ltd v Gottlieb [2005] EWCA Civ 112, [2005] Lloyd’s Rep IR 369 [26]. 84 The Star Sea (n 14) [62] (Lord Hobhouse). 85 AXA General Insurance (n 83). Any such sums will be recoverable as paid under a fundamental mistake or on a consideration that has totally failed: [27]. 86 ibid [28] (Mance LJ). 87 ibid [31] (Mance LJ). 88 Britton (n 38) (above, text to n 39). 89 The DC Merwestone (n 40) [8], [67] (but cf [97]: ‘related (but distinct)’), [119]. See also Orakpo (n 79) 451. 90 Law Commission of England and Wales and Scottish Law Commission, Insurance Contract Law: Business Disclosure; Warranties; Insurers’ Remedies for Fraudulent Claims; and Late Payment (Law Com No 353, 2014; Scot Law Com No 238, 2014) para 21.16.

The Three Ages of Utmost Good Faith 77 beyond the forfeiture rule ought to arise through the normal principles of general contract law, so that any discharge would be prospective on the basis that the fraudulent claim constituted a repudiatory breach of contract.91 A discussion of the application of section 17 to post-formation utmost good faith requires reference also to the order for ship’s papers. In Sir Mackenzie Chalmers’ own commentary on the 1906 Act, he wrote by way of a note to section 17:92 It follows from the nature of the contract that even in litigation both parties must pay with the cards on the table; hence the full discovery allowed as to ships’ papers and other material documents.

As noted above, utmost good faith does indeed contribute to the courts’ justification for the exceptional disclosure order that is the order for ship’s papers, at least in judgments from the mid-nineteenth century onwards, including after the advent of the 1906 Act.93 It has never been suggested, however, that a failure to comply with an order for ship’s papers might afford the insurer the right to avoid the policy,94 and it would indeed be anomalous and startlingly harsh for non-compliance with one particular form of disclosure order to incur such a draconian penalty.95 Instead, the stay of action accompanying the order will be continued.96 The order for ship’s papers cannot, therefore, fall within section 17.97 There are two possible explanations: either the order for ship’s papers in truth constitutes no part of the doctrine of utmost good faith at all, or, alternatively, it does form part of the doctrine but falls outside of section 17, which may indeed be a possible interpretation of Chalmers’ comment. If the latter is correct, it demonstrates again that section 17 embraces only part of a broader doctrine. The most detailed judicial analysis of the post-formation duty of utmost good faith is to be found in The Star Sea.98 The insurers contended that, by virtue of the doctrine of utmost good faith, an assured that made a claim was under a duty to disclose all matters material to the insurers’ treatment of that claim, that the standard of the duty mirrored that of the assured’s pre-formation disclosure obligations and so was not confined to refraining from fraud,99 that this duty endured even beyond the instigation of legal proceedings in relation to the claim, and that this duty fell within section 17 so that breach afforded the insurers the right retrospectively

91 Fargnoli v GA Bonus plc [1997] CLC 653 (OH) 670; The Star Sea (n 14) [50]–[52]; Agapitos v Agnew (The Aegeon) (No 1) [2002] EWCA Civ 247, [2003] QB 556 [45(d)]; AXA General Insurance (n 83) [22]; The DC Merwestone (n 40) [8], [67], [119]. In Orakpo (n 79) 452, Sir Roger Parker considered that fraud in a claim could not attract a lesser sanction than innocent non-disclosure in the presentation of a risk, but Hoffmann LJ (451) adopted language indicative of repudiatory breach of contract. See also the text to n 155 for the approach adopted under the Insurance Act 2015. It may be noted also that, in Banque Keyser Ullmann (n 58) 781, one ground on which a damages remedy was denied for breach by the insurer of its pre-formation duty of disclosure was that the duty fell within s 17, s 17 addressed remedies for breach, and in so doing it was to be considered as exhaustive. If that reasoning is correct (a better view might be that the reference to avoidance is directed at clarifying that a failure to observe the utmost good faith renders a contract voidable not void, contrary to several previous dicta, eg Carter (n 2) 1909; 1164), the availability of forfeiture of the claim as an independent remedy would be inconsistent with the making of a fraudulent claim constituting a failure to observe utmost good faith within s 17. 92 M Chalmers and D Owen, The Marine Insurance Act 1906, 2nd edn (London, Butterworth & Co, 1913) 27. 93 Graham Joint Stock Shipping Co Ltd v Motor Union Insurance Co Ltd [1922] 1 KB 563 (CA) 580; Leon (n 36) 579–80. 94 Manifest Shipping Co Ltd v Uni-Polaris Insurance Co Ltd (The Star Sea) [1997] 1 Lloyd’s Rep 360 (CA) 371. 95 Also inconsistent with the amendment of rules of court in 1936 in response to the order having become (according to Greer LJ in Leon (n 36) 588) ‘an unfair and unjust weapon in the hands of the insurer’. See The Sageorge (n 36) 641, 646 on the 1936 amendments, and generally on constraining inappropriate use of the order. 96 London & Provincial Marine & General Insurance Co Ltd v Chambers (1900) 5 Com Cas 241. 97 The Star Sea (n 14) [60]. 98 ibid. 99 In oral argument, this contention was modified in favour of a test of unconscionable non-disclosure: ibid [106].

78 Howard Bennett to avoid the entire policy. The House of Lords rejected this argument on the basis that the duty attaching to the making of claims was confined to fraud and that once litigation commenced disclosure was governed by the rules of procedure. In the course of a wide-ranging survey of the post-formation doctrine of utmost good faith, Lord Hobhouse addressed ‘the problematic character of s 17’.100 He contrasted the prospective consequences in general contract law of a contract being discharged for a repudiatory breach with the retrospectivity of avoidance pursuant to section 17. In the post-formation context, such retrospectivity, with its attendant forfeiture of payments previously and properly received by the assured from the insurer, was ‘effectively penal’ and ‘cannot be reconciled with principle’.101 He proposed the following approach: A coherent scheme can be achieved by distinguishing a lack of good faith which is material to the making of the contract itself (or some variation of it) and a lack of good faith during the performance of the contract which may prejudice the other party or cause him loss or destroy the continuing contractual relationship. The former derives from requirements of the law which pre-exist the contract and are not created by it although they only become material because a contract has been entered into. The remedy is the right to elect to avoid the contract. The latter can derive from express or implied terms of the contract; it would be a contractual obligation arising from the contract and the remedies are the contractual remedies provided by the law of contract.102

Fundamentally, the doctrine of utmost good faith should not be transformed into an instrument of oppression affording an insurer inappropriate access to an extremely severe remedy.103 Had the parties intended the retrospective remedy of avoidance devised by general contract law to respond to problems in the formation of contracts to apply to a problem of performance, they could have so provided in the contract. In the absence of such an extension by the parties, Lord Hobhouse considered that any suggestion of such an extension as a matter of law should be carefully examined to determine whether it was compatible with ‘the over-riding needs of justice’.104 He continued as follows: Where the application of the proposed principle would simply serve the interests of one party and do so in a disproportionate fashion, it is right to question whether the principle has been correctly formulated or is being correctly applied and it is right to question whether the codifying statute from which the right contended for is said to be drawn is being correctly construed.105

In the conclusion to his judgment, Lord Hobhouse returned to the theme of appropriate limits for the doctrine of utmost good faith: [T]he courts should be on their guard against the use of the principle of good faith to achieve results which are only questionably capable of being reconciled with the mutual character of the obligation to observe good faith.106

In The DC Merwestone,107 Lord Sumption, Lord Clarke and Lord Hughes all cited in full the first of these two latter passages and espoused the concern there articulated about proportionality.

100 ibid [41]. 101 ibid [51]. 102 ibid [52]. 103 ‘The courts have consistently set their face against allowing the assured’s duty of good faith to be used by the insurer as an instrument for enabling the insurer himself to act in bad faith’: ibid [57] (Lord Hobhouse). 104 ibid [61]. 105 ibid. 106 ibid [79]. 107 The DC Merwestone (n 40) [27], [45], [97].

The Three Ages of Utmost Good Faith 79 The question was whether the fraudulent claims rule with its forfeiture penalty should extend to instances of fraudulent statements made in support of a legitimate claim (known as ‘fraudulent devices’). The Supreme Court held that such an extension would result in unduly harsh penalties. An assured that garnishes a genuine claim for no more than is genuinely due with a collateral untruth that does not affect the substance of the claim should not be assimilated to an assured that dishonestly seeks to deceive an insurer into making a payment to the assured to which the assured is not entitled.108 Proportionality was invoked in The DC Merwestone in relation to the scope of a common law rule within the doctrine of utmost good faith. On a similar basis, it has been suggested that, in rare circumstances, the doctrine of utmost good faith might deny an insurer a remedy to which it is in principle entitled. In Drake Insurance plc v Provident Insurance plc,109 the assured’s nondisclosure of a material circumstance did not induce the contract. Obiter, however, the Court of Appeal considered that the doctrine of utmost good faith could sustain a ‘concept of proportionality implicit in fair dealing’110 that would deny an insurer the remedy of avoidance where, at the time it purported to avoid, the insurer knew of circumstances that would have countered the materiality of the relevant circumstance if existing at the time of conclusion of the contract.111 Indeed, it was possible that an insurer that did not know of such circumstances but had notice of the possibility that they might exist should advise the assured that it was contemplating avoidance and afford the assured an opportunity to present any further information relevant to the question of materiality.112 The potential for the doctrine of utmost good faith to influence the construction of an insurance contract, or to facilitate the implication of a term into an insurance contract, has been more nebulous. It is clear that the express conferral of a contractual right on one party does not, of itself, attract a right of disclosure by the other of all information material to its exercise. The doctrine of utmost good faith does not, for example, intervene to embellish and strengthen an insurer’s contractual cancellation right by implying a continuing post-formation obligation to disclose to the insurer all circumstances material to the possible exercise of that right.113 To require communication of each and every fresh material circumstance as it came to the knowledge of the assured throughout the duration of the risk ‘would turn what is an indispensable shield for the Underwriter into an engine of oppression against the insured’.114 Any such obligation must be imposed expressly.115 Equally, a contractual discretion is not inherently subject to an implied fetter requiring its exercise to be consistent with the utmost good faith. An assured under a nonproportional reinsurance treaty enjoys an unfettered discretion in deciding whether or not to cede a particular risk.116

108 For invocations of proportionality in The DC Merwestone (n 40) other than the quoted passage from The Star Sea (n 14) [61], see [47], [99], [103]. See also [109] (Lord Toulson): ‘what is just and appropriate’. 109 Drake Insurance plc v Provident Insurance plc [2003] EWCA Civ 1834, [2004] QB 601. 110 ibid [89] (Rix LJ). 111 ibid [91], [144]. Likewise, Brotherton v Aseguradora Colseguros SA (No 2) [2003] EWCA Civ 705, [2003] Lloyd’s Rep IR 746 [34]. 112 This was accepted by Pill LJ (in Drake Insurance (n 109) [177]), but noted as requiring further consideration by Rix and Clarke LJJ (in Drake Insurance (n 109) [92], [145]). 113 Commercial Union Assurance Co v Niger Co (1922) 13 Ll L Rep 75 (HL); New Hampshire Insurance Co v MGN Ltd [1997] LRLR 24 (CA) 60–61. 114 Commercial Union Assurance (n 113) 82 (Lord Sumner). 115 Commercial Union Assurance Co v Niger Co (1921) 6 Ll L Rep 235 (CA) 245. 116 Aneco Reinsurance Underwriting Ltd v Johnson & Higgins Ltd [2001] UKHL 51, [2002] 1 Lloyd’s Rep 157, 192. For the underlying absence of any fetter as a matter of general contract law, see Glasgow Assurance Corp v Symondson & Co (1911) 16 Com Cas 109 (KB) 122; Bonner v Cox [2005] EWCA Civ 1512, [2005] 2 Lloyd’s Rep 152.

80 Howard Bennett Utmost good faith has, nevertheless, been invoked in relation to the implication of contractual terms. Where a term was implied in accordance with the normal principles of general contract law affording an insurer in the Lloyd’s market access to placement and claims documents held in accordance with market practice by the broker, it was said that ‘the insurance context, where good faith operates, supports that conclusion’.117 Similarly, the standard contractual right of a liability insurer to conduct and control proceedings brought against the assured that could give rise to a claim is subject to an implied constraint. The insurer must not abuse the right but exercise it in good faith (not a reference to utmost good faith, but meaning for the purpose for which it is granted) and with proper regard to the interests of the assured, as the defendant in the proceedings, as well as its own, as contractually liable to indemnify the assured in respect of such liability as established in the proceedings.118 This restriction again appears to be implicit in the grant of the power of control, and sustainable as a matter of general contract law,119 but it has also been associated with the doctrine of utmost good faith.120 A supporting role where an implication arises as a matter of general contract law amounts, however, to nothing. The key question is whether utmost good faith could ever support the implication of a term in circumstances where general contract law would deny any implication. In Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (Nos 2 & 3),121 a claims cooperation clause in a reinsurance contract forbade any settlement, compromise or admission of liability without the reinsurers’ prior consent. Mance LJ, with whose judgment Latham LJ agreed, held that the reinsurers’ right to withhold consent was fettered by the context and purpose of the right and that any withholding of consent ‘should take place in good faith after consideration of and on the basis of the facts giving rise to the particular claim and not with reference to considerations wholly extraneous to the subject-matter of the particular reinsurance’.122 Mance LJ stated, however, that this implication derived not from any principles special to insurance law, but rather ‘from the nature and purpose of the relevant contractual provisions’.123 Subsequently, nevertheless, in Eagle Star Insurance Co Ltd v Cresswell,124 Rix LJ, noting that the Court of Appeal had not been unanimous in Gan v Tai Ping, commented of the majority’s implied fetter on the reinsurers’ right to withhold consent that ‘this protection may not depend only on a term to be implied “for business efficacy”, but may be inherent as a matter of law in the very essence of the reinsurers’ mutual obligation of good faith’.125 It remains unclear whether and, if so, to what extent the doctrine of utmost good faith is in truth relevant to questions of construction of insurance contracts and implication of terms into

117 Goshawk Dedicated Ltd v Tyser & Co Ltd [2006] EWCA Civ 54, [2006] 1 Lloyd’s Rep 566 [53] (Rix LJ). 118 Groom v Croker [1939] 1 KB 194 (CA). 119 Zurich Australian Insurance Ltd v Metals & Minerals Insurance Pte Ltd [2007] WASC 62, (2007) 209 FLR 247 [264]. 120 Cox v Bankside Members Agency Ltd [1995] CLC 671 (CA) 693 (Saville LJ). Sir Thomas Bingham MR rejected any suggestion that breach by the insurer of the constraint ‘gives the assured no right other than rescission’ (ibid 680), but it is unclear whether this is based upon the implied fetter not deriving from the doctrine of utmost good faith at all or constitutes an affirmation of remedial flexibility in the post-formation doctrine of utmost good faith. It is not credible that rescission is available on any analysis. 121 Gan Insurance Co Ltd v Tai Ping Insurance Co Ltd (Nos 2 & 3) [2001] EWCA Civ 1047, [2001] Lloyd’s Rep IR 667. 122 ibid [67] (Mance LJ, Latham LJ concurring, Sir Christopher Staughton dissenting). 123 ibid [68]. 124 Eagle Star Insurance Co Ltd v Cresswell [2004] EWCA Civ 602, [2004] 1 CLC 926 [54]. 125 Likewise, Phoenix General Insurance Co of Greece SA v Halvanon Insurance Co Ltd [1985] 2 Lloyd’s Rep 599 (Com Ct) 613–14 (Hobhouse J), accepting as a matter of common law an implied term as to maintenance of and access to records ‘which would probably be imported anyway by the duty of good faith’. However, it is the contractual analysis that has recently been reiterated: Equitas Insurance Ltd v Municipal Mutual Insurance Ltd [2019] EWCA Civ 718, [2019] Lloyd’s Rep IR 359 [109] (implied term that reinsurance claims relating to mesothelioma must be presented in a manner that is not arbitrary, irrational or capricious).

The Three Ages of Utmost Good Faith 81 insurance contracts. There appears to be no example of the doctrine having changed the result achieved by the normal principles of general contract law.

C. Conclusions in Relation to Section 17 In the light of the preceding discussion, some simple propositions emerge about the a rchitecture of utmost good faith in its second age and the role of section 17 of the Marine Insurance Act 1906. It is worth reiterating that the debate about the role of section 17 is driven by concerns about remedy. The section is unequivocal: a failure by one party to observe the utmost good faith to which the section speaks, namely the utmost good faith on which an insurance contract is ‘based’, entitles the other party to avoid the contract. Conceptually, such retrospective avoidance responds in the logic of general contract law to the vitiation of the consent that lies at the heart of a contract’s formation. It is not a recognised response to issues that arise during the lifetime of a contract. That is not to assert that it would be technically impossible for avoidance to constitute the law’s response to a particularly serious issue under a particular type of contract, such as the making of a fraudulent claim under an insurance contract, were it considered appropriate. However, recognising the potentially excessive and arbitrary consequences of retrospective avoidance, the courts have indicated that the remedy would not be an appropriate response even to egregious dishonesty in the operation of a concluded policy. The ineluctable conclusion is that section 17 and its remedy of avoidance must be confined to situations of vitiated consent in the formation of insurance contracts. Logically, there are two ways to achieve this: either all other situations that are candidates for the application of the doctrine of utmost good faith must be analysed as not in fact attracting the application of that doctrine, or the doctrine must be recognised as broader than section 17, with different, and appropriate, remedies attaching to aspects of the doctrine falling outside section 17. ‘Held covered’ clauses support the second option. As discussed above,126 since they involve the making of an underwriting decision, they clearly attract the doctrine of utmost good faith, but, since they do not trigger the formation of a new contract, they do not fit within the pre-formation doctrine, and, since their purpose is to trigger an extension to cover, retrospective avoidance of the entire policy in response to a potentially innocent failure to disclose a circumstance material to the extension would be grossly disproportionate. Held covered clauses should therefore be accepted as falling within the postformation doctrine of utmost good faith but outside of section 17, so that the remedial response to non-disclosure or untruth of statement in relation to circumstances material to the underwriting decision to be made under the held covered clause can be confined to loss of the extension of cover under the clause. The fraudulent claims jurisdiction and the order for ship’s papers submit to a similar analysis. In neither case does the doctrine of utmost good faith apply as a matter of irrefutable logic; the law should have been capable of generating such developments without any assistance from a broader doctrine. In each case, however, the case law is clear that the doctrine was invoked as a basis for the legal response to the relevant problem, but not so as to constrain that response within a remedial straitjacket. The sensible approach is, therefore, again to acknowledge the fraudulent claims jurisdiction and the order for ship’s papers as aspects of the doctrine of utmost good faith but as falling outside of section 17, so that there is no conflict between the remedies acknowledged by the courts for making a fraudulent claim or for non-compliance with an order for ship’s papers and the remedy of avoidance as granted by section 17.

126 See

text to nn 73–78.

82 Howard Bennett The same is true of other possible instances of utmost good faith. If the doctrine influences the construction of a contractual or statutory provision, or provides the (or a) basis for implying a term into an insurance contract, acting contrary to the relevant provision so construed or implied may perhaps be characterised as failing to observe the utmost good faith. It does not, however, follow that avoidance is an appropriate, or even a logically possible, remedial response. If the Court of Appeal in Drake Insurance was correct as regards the suggested restriction on the right of avoidance,127 the consequence must be that in appropriate circumstances the insurer does not have the right to avoid. And there is no justification for insurance contract law departing from general contract law in its response to breach of a contract term.128 Finally, recognising that the doctrine of utmost good faith extends beyond section 17 accommodates the independent duties of disclosure and truth of statements in relation to material circumstances owed by placing brokers, given that section 17 is concerned with the observing of utmost good faith by the parties to the contract. The net result is that section 17 articulates only part of the second age doctrine of utmost good faith, namely that which applies to the parties to a putative insurance contract at the stage of contract negotiation. It is in this sense that insurance contracts are ‘based’ on the utmost good faith to which the section speaks. Section 18 is correctly regarded as elaborating one aspect of the section 17 doctrine. The same is true of section 20 to the extent that it applies to the assured. Section 17 also supports aspects of pre-formation utmost good faith required of the parties that are not addressed in subsequent sections of the 1906 Act, as discussed above.129 This analysis is supported by the seminal judgment of Lord Hobhouse in The Star Sea, a central theme of which is an acknowledgement of the remedial problems attached to section 17 and of the need to ensure that various manifestations of utmost good faith are enforced by appropriate remedies. This approach is reinforced by the decision of the Supreme Court in The DC Merwestone,130 in which the question of whether a fraudulent device fell within the fraudulent claims jurisdiction was determined by considering whether the remedies attaching to a fraudulent claim were appropriate for a fraudulent device. Contrast a second, short judgment in The Star Sea, that of Lord Clyde. He recognised that post-formation duties arising from the doctrine of utmost good faith would vary with respect to their content from the pre-formation duties and from one another. He refused, however, to overcome any argument that section 17 mandated a uniform standard and scope for all postformation duties by confining section 17 to pre-formation duties. That approach he declared with no elaboration to be ‘past praying for’.131 Instead, the requisite flexibility was to be found in the phrase ‘utmost good faith’, which, in particular, extended to innocent non-disclosure and misrepresentation at the pre-formation stage but was confined to fraud in the context of claims. Lord Clyde wholly failed, however, to engage with the question of remedy, which is fundamental to the scope of section 17.132 127 See text to n 109. 128 Goshawk Dedicated (n 117) [53] (Rix LJ): the issue is not that certain conduct constitutes a failure to observe the utmost good faith, but rather that ‘the duty informs the content of the contractual obligation’. 129 See text to nn 68–72. 130 Discussed in the text to n 107. 131 The Star Sea (n 14) [6], quoted in The Mercandian Continent (n 65) [34]. 132 The other reasoned judgment in The Star Sea (n 14), that of Lord Scott, contains elements of confusion. He was adamant (at [81]) that s 17 could not be confined to pre-formation aspects of utmost good faith, but simply because the doctrine was established as having a post-formation dimension (relying in that respect, inter alia, on The Litsion Pride (n 61), which was overruled in The Star Sea as advancing an illegitimate extension of the doctrine: ibid [71]). He therefore assumed that all aspects of the doctrine of utmost good faith fell within s 17, which he carried through to the question of remedy, asserting that a fraudulent claim triggered a right of avoidance (although in another passage (at [110]) he

The Three Ages of Utmost Good Faith 83 With respect, this remedial omission is fatal to the peremptory dismissal of the approach to section 17 advanced in this chapter and supported by the extended and compelling reasoning of Lord Hobhouse. For the doctrine of utmost good faith to apply in appropriately tailored ways to selected post-formation aspects of insurance contracts, flexibility is imperative not only with respect to the duties imposed, but also with respect to the remedies for non-compliance. Section 17 denies the requisite flexibility of remedy. It is therefore imperative, in the framework of the statutory provisions applicable to utmost good faith in its second age, to recognise the post-formation dimension of the doctrine of utmost good faith as independent of section 17.133

IV. The Third Age: The Insurance Act 2015 The third age of utmost good faith commenced with the entry into force of the Insurance Act 2015 on 12 August 2016. The new utmost good faith regime introduced by the 2015 Act applies to insurance contracts concluded on or after that date and to variations concluded on or after that date to contracts concluded before it.134 Insurance contracts and variations concluded before that date remain subject to the second age rules. The 2015 Act, which enacts proposals of the English and Scottish Law Commissions,135 implements major structural and substantive reforms to the doctrine of utmost good faith. The Act introduces the concept of a ‘duty of fair presentation of the risk’ owed by the assured to the insurer whenever negotiating a new insurance contract or variation to an existing insurance contract.136 This duty encompasses obligations of disclosure and truth of statement. No independent duty of the placing broker is recognised, but the knowledge of any individual, whether or not employed by the assured, who participates on behalf of the assured in the process of procuring the assured’s insurance is attributed to the assured.137 And the duty to make a fair presentation of the risk is clearly non-delegable, so that any failure by a placing broker to effect the disclosure or meet the truth of statement required to comply with the duty is a failure of the assured. An insurer seeking a remedy for an alleged breach of the duty of fair presentation of the risk must always prove that the non-disclosure or misrepresentation satisfies uniform tests for materiality and inducement,138 even if the breach is fraudulent. To that extent, the Act is,

considered debatable whether a fraudulent claim constituted a breach of s 17 entitling avoidance), but not c onsidering how s 17 would work remedially in the context of held covered clauses, to which he also referred. Lord Scott failed, therefore, to distinguish between the scope of s 17 and the scope of the doctrine of utmost good faith, but nevertheless also agreed with the reasoning of Lord Hobhouse. Unhelpfully, Lord Hoffmann agreed with both Lord Hobhouse and Lord Scott, as did Lord Steyn, who agreed with Lord Clyde as well. 133 Remedial inflexibility will, moreover, render utmost good faith an unappealing instrument for resolving diverse problematic issues. In Equitas Insurance (n 125) [104], post-formation utmost good faith was rejected by Males LJ as a basis for an implied requirement as to the allocation of losses between reinsurance contracts since it ‘gives rise to a remedy of avoidance of the contract rather than a constraint on the exercise of prima facie contractual rights’. 134 Insurance Act 2015, s 22(1). 135 Law Com No 353 (n 90). 136 Insurance Act 2015, s 3. 137 ibid ss 4(3)(b), (8)(b). 138 The latter under the rubric of ‘qualifying breach’: Insurance Act 2015, s 8. This requires proof that the breach of duty in some way altered the underwriter’s decision, although technically not proof of specifically in what respect that decision was altered (although in practice it may be difficult to convince a court of the former without proof of the latter). Proof of the precise alteration is then required in respect of non-fraudulent breaches in order to apply the statutory scheme of proportionate remedies.

84 Howard Bennett surprisingly, more indulgent to fraud than the earlier ages of utmost good faith which follow general contract law.139 An insurer that establishes that the duty of fair presentation of the risk has been broken by a material non-disclosure or a material misrepresentation, and that the breach induced the insurer to conclude the contract at all, or on the terms agreed, is entitled to the appropriate remedy from a scheme of proportionate remedies. Within that scheme, avoidance is confined to cases of fraud and cases where, but for the breach of duty, the insurer would not have entered into the contract at all. Otherwise, where the insurer would have concluded the contract but on different terms, the insurer’s remedy lies in retrospective rectification of the policy to the terms that would have been agreed, and where the insurer would have concluded the contract but charged a higher premium, the remedy consists of a proportionate reduction in the measure of indemnity.140 Having established the duty of fair presentation of the risk, the 2015 Act then abolishes ‘any rule of law permitting a party to a contract of insurance to avoid the contract on the ground that the utmost good faith has not been observed by the other party’.141 In consequence of these reforms, the 2015 Act then repeals sections 18–20 of the Marine Insurance Act 1906,142 and amends section 17 of the 1906 Act so that it provides simply that ‘A contract of marine insurance is a contract based upon the utmost good faith’.143 The explicit reference to the mutuality of the doctrine and the statement of the right to avoid in the event of failure to observe the utmost good faith are removed. As discussed above, the structural problems afflicting the doctrine of utmost good faith in its second age flowed from the attachment of the remedy of avoidance to any failure to comply with the injunction of section 17. This is addressed, first, by amending section 17 to leave the provision silent as to the consequences of a failure to observe the utmost good faith, and, secondly, by the abolition of avoidance as a remedy for any failure to observe the utmost good faith. Avoidance nevertheless remains the remedy in certain instances of breach of the duty of fair presentation of the risk. A clear structural point follows. Since the duty of fair presentation of the risk can attract a remedy that a failure to observe the utmost good faith in its third age can never attract, it follows that the duty of fair presentation of the risk does not form part of the third age doctrine of utmost good faith. In its third age, therefore, the doctrine of utmost good faith is shorn of the pre-formation disclosure obligations that were primarily responsible for the doctrine’s recognition and which constituted its most significant manifestation during its first two ages. The duty of fair presentation of the risk attaches not only to the formation of new insurance contracts, but also to the variation of existing insurance contracts, with the potential for avoidance reflecting the conclusion of a separate contract involved in a variation,144 as discussed

139 On the stricter approach to fraud and materiality, see above, text to nn 69–70. With respect to inducement, in cases of fraud ‘the court does not allow an examination into the relative importance of contributory causes’, so there is no need to show any particular level of impact of the fraud on the mind of the deceived party: Barton v Armstrong [1976] AC 104 (PC) 118 (Lord Cross). It suffices that the fraudulent statement (or the matter in relation to which there is fraudulent non-disclosure) was ‘actively present to’ the insurer’s mind and it is irrelevant that the insurer was also actuated by other contributing inducements: Edgington v Fitzmaurice (1885) 29 Ch D 459 (CA) 483 (Bowen LJ). See also BV Nederlandse Industrie van Eiprodukten v Rembrandt Enterprises Inc [2019] EWCA Civ 596, [2019] 1 Lloyd’s Rep 491. This appears to have been overlooked by the Law Commissions, which erroneously considered the test for a qualifying breach to replicate the existing law on inducement: Law Com No 353 (n 90) para 11.11. 140 Insurance Act 2015, s 8, sch 1, pt 1. 141 Insurance Act 2015, s 14(1). 142 ibid s 21(2). Nor is there any argument that an underlying common law rule might revive in the light of repeal of the statutory provisions: s 22(3). 143 ibid s 14(3)(a). 144 Insurance Act 2015, s 2(2), sch 1, pt 2.

The Three Ages of Utmost Good Faith 85 previously.145 The duty does not, however, encompass all aspects of the second age pre-formation doctrine. Since no other ground for avoidance of the policy on the basis of a failure to observe the utmost good faith survives into the third age, it follows that aspects of the second age pre- formation doctrine falling outside the duty of fair presentation of the risk but triggering a right to avoid the policy under the second age section 17 have been abolished, retained within the doctrine of utmost good faith but with a different remedy, or relocated to the province of a different doctrine. Specifically, first, the reciprocal duties of disclosure and truth of statement owed by the insurer to the assured have not survived. The view of the Law Commissions was that matters such as the selling of worthless insurance are better addressed through industry regulation.146 Secondly, confinement of the duty of fair presentation of the risk to material circumstances eliminates any scope for pre-formation utmost good faith to attach as a matter of law to non-material circumstances about which the particular underwriter happens to be concerned. The Law Commissions, indeed, were ‘doubtful’ as a matter of principle whether non-disclosure or misrepresentation of a non-material circumstance ‘should attract any remedy at all’,147 and certainly not avoidance. Under the 2015 Act, therefore, any such matters would have to be made the subject of express contractual provision, either by widening the definition of materiality by using the contractingout facility within the Act148 or by including an express stipulation. Thirdly, the Law Commissions considered that the possibility of realisation by one party of a mistake being made by the other should be left to general contract law.149 In contrast to this contraction of utmost good faith, the liberation of section 17 from remedial constraints allows the general doctrine of utmost good faith doctrine as articulated by the amended section to provide a foundation for other duties arising from utmost good faith, namely the post-formation duties. These can be fashioned so that they respond in the most appropriate manner to the relevant issue, adopting, in accordance with the analysis proposed by Lord Hobhouse in The Star Sea,150 the apparatus and remedies of general contract law, albeit with bespoke modifications where appropriate. Within the amended section 17 fall, therefore, the disclosure and truth of statement obligations that attach to the triggering of held covered clauses, which, as discussed above,151 involve the exercise of pre-existing rights rather than a distinct contract to vary. While the duties are directly analogous to those that arise in the pre-formation context, they apply by means of implication into the held covered clause, with non-compliance constituting the non-fulfilment of a condition precedent to the clause’s operation, a consequence that does not involve avoidance.152 A second example of third age section 17 utmost good faith is the fraudulent claims jurisdiction. The 2015 Act sets the remedial side of the jurisdiction on a statutory basis,153 in order to provide clarity to an area of the law that had seen significant recent litigation and to ‘remove the 145 Above, text following n 60. 146 Law Com No 353 (n 90) para 30.34. 147 ibid para 30.30. 148 See Insurance Act 2015, ss 16–17. 149 Law Com No 353 (n 90) para 30.30, also noting that the failure to inform the insurer of the mistake might constitute a breach of the duty of fair presentation. The examples cited in the case law (above, text to n 72) would not, however, constitute mistakes as to contract terms within the rule in Smith (n 1). 150 Above, text to n 102. 151 Above, text to nn 73–77. 152 The Law Commissions considered that held-covered clauses would be treated as variations for the purpose of the duty of fair presentation of the risk: Law Com No 353 (n 90) para 30.41. However, no analysis is provided to support this expectation and, as discussed above, an exercise of existing rights is not a variation. 153 Insurance Act 2015, s 12.

86 Howard Bennett problems created by section 17 of the Marine Insurance Act 1906’.154 For the most part, the 2015 Act codifies the outcome of that case law. Consistently with a breach of contract approach to post-formation utmost good faith, however, the 2015 Act resolves the undecided question of the impact of a fraudulent claim beyond the claim on the contract as a whole in favour of regarding a fraudulent claim as a repudiatory breach triggering, at the insurer’s election, a prospective discharge, albeit with bespoke consequences in that an insurer’s election to treat the contract as discharged takes effect as from the time of the fraud, not the time of the insurer’s acceptance of the breach.155 Questions regarding the scope of the jurisdiction remain, however, for the common law and should be analysed through the lens of utmost good faith.156 As discussed in previous sections, the doctrine of utmost good faith is not necessarily confined to imposing duties of disclosure and honesty in various contexts. The Law Commissions envisaged the retained doctrine functioning as a ‘general interpretative principle’,157 and suggested three ways in which it might serve a useful purpose.158 The first is to provide assistance in interpreting the duty of fair presentation of the risk: Both parties are expected to act in good faith in exchanging information. For example, if a court were to find that an insured had intentionally disclosed only the bare minimum of information, hoping that the insurer would fail to make further enquiries to reveal the full picture, the insured would not have acted in good faith and would therefore be in breach of the duty of fair presentation.159

The duty of fair presentation of the risk introduces the idea that the assured might fail to disclose all material circumstances, but might, nevertheless, make it clear to a prudent insurer that further material information was available if requested.160 Disclosure of a bare minimum with no signalling that more information is available will by definition not satisfy the duty. The point being made by the Law Commissions seems to be that any finding of deliberate intention to produce the result that the insurer does not learn of certain material information will militate strongly against, if not automatically preclude, a finding of partial but adequate disclosure. Absent any such intention, however, it is unclear whether the principle of utmost good faith will be of any assistance in partial disclosure cases. While it may be argued that utmost good faith should tip the balance against an assured that fails to make full up-front disclosure, equally it may be argued to the contrary that utmost good faith now requires an insurer to be astute to the possibility of the need for supplementary enquiries. Whatever the merits of the specific example offered by the Law Commissions, however, it was recognised even in the first age of the doctrine that the interpretation of legislation relating to insurance contracts might be influenced by the fundamental idea that such contracts are based on the utmost good faith.161 The second role for utmost good faith envisaged by the Law Commissions is: To inform the need to imply contractual terms into the policy under the traditional ‘business efficacy’ test. Good faith provides a background when considering whether it is necessary to imply a particular term.162 154 Law Com No 353 (n 90) para 21.1. 155 Indeed, this is a consequence of abolishing avoidance as a remedy for failure to observe the utmost good faith (Law Com No 353 (n 90) para 22.3), showing, in turn, that the Law Commissions envisage the fraudulent claims jurisdiction as continuing to be part of the doctrine of utmost good faith. 156 Perhaps influencing, for example, the tolerance to be allowed for deliberate exaggeration in the making of a claim in order to afford room for negotiation yet still recover the true loss. 157 Law Com No 353 (n 90) para 30.5. 158 ibid para 30.23. 159 ibid. 160 Insurance Act 2015, s 3(4)(b). 161 See text to n 22. 162 Law Com No 353 (n 90) para 30.23.

The Three Ages of Utmost Good Faith 87 That an insurance contract is based on the utmost good faith should, accordingly, favour the implication of duties of cooperation and transparency. As already noted, reference has been made to utmost good faith in the context of implying duties regarding the provision of access to records and other documentation.163 Although the courts would have upheld the same implications even in the absence of the doctrine of utmost good faith, the presence of that doctrine may support such a conclusion and perhaps ease the way towards gaining acceptance in the judicial mind. The Law Commissions, however, go further and suggest that the retained principle is capable of permitting the implication of terms in insurance contracts that would not pass the general contract law test for implication: ‘Identifying good faith as a component of business efficacy recognises the unique nature of insurance contracts, and allows the courts to imply terms which would not be considered necessary outside the field of insurance.’164 Equally, one might add that when faced with an issue of construction, the nature of the transaction as of utmost good faith may assist, or at least be taken into account, in the search for the meaning to be ascribed to express policy wording. Such an idea again can be traced back to the first age of the doctrine.165 The final role for utmost good faith envisaged by the Law Commissions is to provide a juridical basis for the exercise of judicial flexibility. They explain as follows: It is possible that the principle of a mutual duty of good faith could provide a solution to an especially hard case or emergent difficulty. Although we think such cases would be extremely rare, it is possible that the courts could develop the concept to prevent an insurer from relying on a right to deny a claim where it would be manifestly unfair to do so.166

This clearly references the suggestion in Drake Insurance167 that avoidance may be denied where post-formation but pre-avoidance evidence reveals that the materiality of a non-disclosed or misrepresented circumstance was in fact based on inaccurate or incomplete information and that accurate or complete information, if available at the time of conclusion of the contract, would have denied materiality. To what extent any of these suggestions bears fruit remains to be seen, but the amending of section 17 makes it easier for the statutory pronouncement of utmost good faith as the basis of insurance contracts to sustain and encompass any such developments free from any inappropriate remedial repercussions.168

V. Conclusion The evolution of utmost good faith can be divided into three ages. The first age corresponds to the period before 1 January 1907, the date when the Marine Insurance Act 1906 entered into force, during which the idea emerged of insurance contracts commanding an enhanced standard of good faith to which the label ‘utmost good faith’ was increasingly attached. The second age spans the subsequent period of nearly 110 years, during which utmost good faith was the subject

163 See text to n 117. 164 Law Com No 353 (n 90) para 30.60. 165 See text to nn 50–51. 166 Law Com No 353 (n 90) para 30.23. 167 See text to n 109. 168 For a sceptical view, see M Hemsworth, ‘The Fate of “Good Faith” in Insurance Contracts’ [2018] Lloyd’s Maritime and Commercial Law Quarterly 143; for a view that s 17 could prove more productive, not least by analogy with developments in Australia, see B Soyer and A Tettenborn, ‘Mapping (Utmost) Good Faith in Insurance Law – Future Conditional?’ (2016) 132 Law Quarterly Review 618.

88 Howard Bennett of sections 17–20 of the 1906 Act. Considerable problems of structure and substance arose by reason of the drafting of section 17, and specifically the attaching of the remedy of avoidance of the contract to any and every failure to observe the utmost good faith as contemplated by that section. The solution, as argued in this chapter, is to recognise that section 17 encompasses only part of the second age doctrine, namely the reciprocal pre-formation duties of disclosure and truth of statement owed by the assured and insurer to each other. The third, and current, age sees the doctrine lose its signature pre-formation duties in favour of the new duty of fair presentation of the risk, but also the liberation of section 17 through the removal of any reference to a remedy. Throughout its history, the doctrine of utmost good faith has not dictated the duties that have arisen under its auspices. Instead, the phrase was applied to insurance contracts as a pithy means of encapsulating the attachment to such contracts of distinctive legal rules and attitudes requiring, in comparison with contracts generally, an enhanced level of openness, candour and fair dealing. Looking to the future, the Insurance Act 2015 robs the doctrine of utmost good faith of much of its glamour. Gone are the days of the doctrine’s functioning, principally by reason of the extensive demands of pre-formation disclosure and unyielding remedy of avoidance in the event of non-compliance, as probably the key substantive doctrine of insurance contract law. Instead, it has been moved from centre stage to the wings, relegated to providing theoretical underpinnings for post-formation duties of disclosure and honesty, potentially oiling the wheels of implication of contract terms and of both contractual and statutory interpretation, and possibly restraining rights in exceptional cases where their exercise would be grossly unreasonable. It can, however, rest content: its reformed progeny, the duty of fair presentation of the risk, will assume the central role and continue to reflect, through its historical origins, ethos and strictures, albeit not its technical affiliation, the fact that insurance contracts remain based on the utmost good faith.

5 English Jurisdiction Clauses in Marine Cargo Insurance and Direct Actions against Cargo Liability Insurers YVONNE BAATZ

I. Introduction The issue considered in this chapter is whether an insurer can rely on an exclusive English jurisdiction clause in a marine cargo insurance policy, whether the claim is made by the insured or by an assignee of the policy, or by a cargo claimant or subrogated insurer who brings a claim in a direct action against the carrier’s liability insurers, usually a protection and indemnity club (P&I club). Can the cargo insurer, or the insurers of the carrier’s liability for damage to the cargo, rely on any jurisdiction clause in their insurance contract, or does the claimant, particularly in a direct action, have a choice where it sues? The answer is of great significance to the liability insurer. If the cargo claimant is not bound by the jurisdiction clause in the insurance contract, but can sue wherever the damage has been suffered, or in any discharge port to which the cargo has been delivered short or damaged, or should have been delivered but has not been delivered at all, then the insurer may find itself sued anywhere that the ship trades to, or if that is different, where the cargo claimant is based. This could present the liability insurer with a very different risk from that which it thought it had undertaken when it agreed to English law and exclusive English jurisdiction in its contract. Indeed, if the ship trades all over the world the liability insurer could be exposed to suit in any jurisdiction to which the ship trades and to multiple proceedings in many different jurisdictions. It is to avoid that risk that the insurer will insert an exclusive jurisdiction clause in its insurance policy; but is it effective? This chapter will discuss the different attitudes of states to the concept of party autonomy. It will then consider direct action against a liability insurer and the challenges such an action poses for party autonomy. In particular, the right of direct action under English law will be addressed. The jurisdictional advantages for a liability insurer where such an action is determined in the English courts will be examined. These advantages provide a reason for forum shopping. Three approaches to party autonomy in marine cargo insurance are then examined. The first of these is the global position. There is not yet a well-established worldwide approach to jurisdiction clauses in marine insurance contracts, unlike arbitration clauses, where there is much more consensus, and certainly no international harmony on whether a jurisdiction agreement would be binding in a direct action. Secondly, it turns to the system developed by the European Union. Here, the principle of party autonomy is given great weight, but a jurisdiction clause in

90 Yvonne Baatz a liability insurance contract has recently been held not to be binding in a direct action under Council Regulation (EC) 44/2001 of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters.1 Thirdly, and finally, the approach of the English common law to jurisdiction agreements will be critically examined. Again, party autonomy is an important principle. Indeed, it will be argued that under that law a jurisdiction agreement may be binding in a direct action depending on the construction of the statute giving such a right. This illustrates the diversity of approach between different states.

II. Party Autonomy A jurisdiction clause is an extremely significant provision in any commercial contract, and particularly when there is an international element. It is part of the bargain between the parties that if a dispute arises it should be resolved in a particular court or arbitration tribunal.2 It may be argued that a jurisdiction clause should be upheld because it is part of the parties’ contractual bargain and because it serves a useful purpose in marine cargo insurance to define one court to determine all disputes where there are numerous cargo claims arising out of one incident by many different cargo claimants domiciled in different states, rather than having proceedings in numerous different courts,3 with the risk of having to call complex evidence in several different courts and the consequent risk of conflicting judgments and unnecessary legal costs. One could argue that when an insurer agrees to cover a shipowner’s liabilities it takes the risk that it is subject to the jurisdiction of the courts of any place that the ship goes to. However, it is then very difficult for the liability insurer to calculate its exposure, and it seeks to avoid such a situation by inserting an exclusive jurisdiction clause in its insurance contract. It is to the benefit of all parties that all proceedings can be heard in a single court so as to avoid duplicating the costs of litigation. The cargo interests already have the advantage that they can arrest the carrier’s ship or ships4 in order to obtain security for their claims, usually by means of a P&I club letter of undertaking. Thus, enforcement of any judgment subsequently obtained is a matter of complying with the trigger for payment in the letter of undertaking, eg presentation of an English court judgment in favour of the cargo interests. This would avoid the necessity of a direct action and would operate even if the shipowner had become insolvent. Ultimately, the market will decide which jurisdiction commercial parties wish to choose. Material factors for parties when deciding to provide for a particular jurisdiction in their contract include: whether a particular state recognises jurisdiction agreements; the efficiency and speed of the court system; the impartiality and expertise of the judges; the cost; the sophistication of the law of the state, both substantive and procedural, whose jurisdiction has been chosen; and ease of recognition and enforcement of judgments. Whether the courts of a state can satisfy these factors is often largely a question of whether the state observes the rule of law and whether its

1 See n 57 below. 2 L Collins et al, Dicey Morris & Collins on the Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012) with Fifth Supplement; A Briggs, Private International Law in English Courts (Oxford, Oxford University Press, 2014); R Fentiman, International Commercial Litigation, 2nd edn (Oxford, Oxford University Press, 2015); A Mills, Party Autonomy in Private International Law (Cambridge, Cambridge University Press, 2018). 3 eg in the context of bills of lading: Compania Sud Americana De Vapores SA v Hin-Pro International Logistics Ltd [2014] EWHC 3632 (Comm), [2015] 1 Lloyd’s Rep 301; [2015] EWCA Civ 401, [2015] 2 Lloyd’s Rep 1. 4 See M Tsimplis, ‘Procedures for Enforcement’ in Y Baatz (ed), Maritime Law, 4th edn (Abingdon, Informa Law from Routledge, 2018).

English Jurisdiction Clauses in Marine Cargo Insurance 91 legal system is properly resourced. It is a fact of life that legal systems in different states are not on a level playing field. The choice of English court jurisdiction is popular in marine insurance contracts,5 for England is a highly regarded venue for the resolution of such disputes. Furthermore, the choice of English court jurisdiction will usually go hand in hand with an express choice of English law, as it is sophisticated and there is great expertise among the English judiciary in relation to marine insurance. One consideration for states is whether they wish to invest the resources in their court system and to attract foreign business in the arena of dispute resolution. Some dispute resolution centres, such as Singapore and Dubai, have been very active in recent years in pouring resources into their courts and arbitration in order to attract such business, as it brings economic benefits to the state concerned in providing work for their lawyers, arbitrators, expert witnesses, etc. For e xample, Singapore set up its International Commercial Court in 2015.6 It is unsurprising, therefore, that Singapore has ratified the 2005 Hague Convention on Choice of Court Agreements (the Choice of Court Convention).7 Other states may, however, have a wholly different attitude. They may not recognise the principle of party choice when it comes to court jurisdiction but may, for example, apply a principle of exercising jurisdiction where the dispute has some connection with their state but not otherwise.8 Quite apart from principle, a state may not have the resources to invest or, indeed, may not wish to attract foreign parties to use their court system.

III. Direct Action Marine insurance contracts introduce additional complexity as they involve third party issues. Where an insured cargo owner claims from its cargo insurer, the question arises whether it is bound by the jurisdiction agreement in the insurance contract. If the cargo insurer then pays the cargo claimant and can exercise subrogated rights and sue the carrier on the bill of lading,9 but the carrier is insolvent, the subrogated cargo insurer may in turn be able to bring a direct action against the carrier’s liability insurer, usually a P&I club.10 The question will then be whether the claimant is bound by the jurisdiction agreement in the liability insurer’s marine insurance contract in the direct action or whether the claimant can bring proceedings wherever it may be based, and/or where the damage has been suffered and/or where the ship carrying the insured cargo has traded to. A cargo claimant or its subrogated insurer is not a party to the carrier’s liability insurance contract. There is therefore no privity of contract between the cargo claimant and the liability insurer. To solve the privity problem, most states provide for a statutory right to bring p roceedings 5 Marine insurance contracts on the MAR forms, the standard marine insurance forms in the London market, contain an exclusive English court jurisdiction clause, as do the rules of some P&I clubs. The rules of some other P&I clubs provide for London arbitration. 6 www.sicc.gov.sg/. 7 On the signatories to the Choice of Court Convention, see n 35 below. 8 This is currently the case eg in China. 9 For the issues in relation to court jurisdiction clauses in bills of lading, see Y Baatz, ‘Choice of Jurisdiction in Bills of Lading and Cargo Insurance’ in J Chuah (ed), Research Handbook in Maritime Law (Cheltenham, Edward Elgar, 2019); on arbitration clauses in bills of lading, see Y Baatz, ‘Should Third Parties Be Bound by Arbitration Clauses in Bills of Lading?’ [2015] Lloyd’s Maritime and Commercial Law Quarterly 85. 10 See S Hazelwood and D Semark (eds), P&I Clubs: Law and Practice, 4th edn (Abingdon, Informa Law from Routledge, 2010); J Hjalmarsson, ‘Marine Insurance’ in Baatz (n 4) 482–92.

92 Yvonne Baatz direct against the liability insurer, often where the carrier has become insolvent, thereby earmarking the insurance proceeds for the liability insured. States may take diverging views as to how to characterise a direct action: whether it is contractual, in which case the jurisdiction clause may be binding, or non-contractual, in which case it may not be.11 Under English law, the Third Parties (Rights against Insurers) Act 2010,12 which applies to liability insurance generally and not just to marine insurance, has the effect that the rights of the carrier who has incurred a liability, and who is insured for that liability and is insolvent,13 are transferred and vest in the persons to whom the liability was incurred (the ‘third party’).14 The third party may bring a direct action to enforce the rights against the insurer.15 It may do so by establishing the carrier’s liability by a court judgment, arbitration award or settlement agreement; alternatively, where it has not yet established such liability, the third party may seek a declaration as to the carrier’s liability to it and a declaration as to the insurer’s liability to it, as third party.16 If the third party seeks a declaration that the carrier is liable to it in proceedings against the insurer, then it may also make the insured a defendant to those proceedings.17 Where the insurance contract provides for arbitration, the third party can apply for both declarations in the arbitration proceedings,18 and the arbitration tribunal may make the appropriate award against the insurer.19 The insurer may rely on any defence on which the insured carrier could rely if those proceedings had been brought against the insured carrier in respect of its liability to the third party.20 The P&I club rules are likely to provide that the P&I club is only liable to pay its member if the member has already paid for its liability, since the insurance is indemnity insurance. The insured has only suffered a loss once it has paid the cargo claimant in respect of its claim, and until it has suffered a loss, there is nothing in respect of which the P&I club must indemnify. This presents a difficulty where the carrier is insolvent and cannot pay the cargo claimant. The P&I club will therefore have a defence against the cargo claimant or its subrogated insurer, as its obligation to pay under the P&I club rules will not have been triggered. This was held to be the case in litigation under the predecessor of the 2010 Act, the Third Parties (Rights against Insurers) Act 1930.21 The position remains the same under the 2010 Act except in relation to death or personal injury.22 This represents a significant advantage for the liability insurer, and P&I clubs will therefore seek to rely on the exclusive English court jurisdiction or London arbitration agreements and English governing law clauses in the P&I club rules, to make sure that this is the position and that they accordingly have no liability.

11 See, eg Case C-463/06 FBTO Schadeverzekeringen NV v Odenbreit [2007] ECR I-11321, [2008] 2 All ER (Comm) 733; J Hjalmarsson, ‘Direct Claims against Marine Insurers in the English Legal System’ (2010) 18 Asia Pacific Law Review 269; V Ulfbeck, ‘Direct Actions against the Insurer in a Maritime Setting: The European Perspective’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 293. 12 The 2010 Act came into force on 1 August 2016, replacing the Third Parties (Rights against Insurers) Act 1930. 13 Third Parties (Rights against Insurers) Act 2010, ss 6 and 6A. 14 ibid s 1(2). 15 ibid s 1(3). 16 ibid ss 1(4) and 2(2). 17 ibid s 2(9). 18 ibid s 2(7). 19 ibid ss 2(6) and (8). 20 ibid s 2(4). This is subject to s 12(1), which deals with the time limit. 21 Firma C-Trade SA v Newcastle Protection & Indemnity Assoc (The Fanti); Socony Mobil Oil Inc v West of England Shipowners Mutual Insurance Assoc (London) Ltd (No 2) (The Padre Island) [1991] 2 AC 1 (HL). See also Hjalmarsson (n 10) 487. 22 Third Parties (Rights against Insurers) Act 2010, s 9(5) and (6).

English Jurisdiction Clauses in Marine Cargo Insurance 93 In a trio of cases,23 the issue arose whether a claimant who brought a direct action under foreign legislation (under a Finnish statute, a Spanish statute and a Turkish statute, respectively) was bound by the London arbitration clause in the insurance contract with the shipowner’s P&I club when sued by the subrogated cargo insurers, the states of France and Spain (whose coastline was polluted as a result of the sinking of the Prestige) and the charterers of a vessel, respectively. In all three cases, it was held that the claimants were so bound, as on the construction of the foreign statute, the right of direct action was based on the insurance contract. The issue t herefore arises whether, by analogy with an arbitration agreement, an exclusive English jurisdiction agreement would also bind a claimant in a direct action? The position may be contrasted with the approach of the international conventions dealing with pollution and carriage of passengers, which do not permit party autonomy as to jurisdiction. The 1992 International Convention on Civil Liability for Oil Pollution Damage (the 1992 CLC),24 the 2001 International Convention on Civil Liability for Bunker Oil Pollution Damage (the 2001 BOPC)25 and the 2002 Athens Convention,26 which all impose strict but limited liability, with some exceptions, on the shipowner, provide for compulsory insurance27 and direct action against the insurer.28 The shipowner or the insurer may set up a limitation fund.29 The insurer will not be liable if the shipowner’s liability is excluded by the relevant convention or if it can show that the pollution damage resulted from the wilful misconduct of the insurer. However, the insurer cannot rely on any other contractual defence on which it might have been able to rely in proceedings brought by the shipowner against it, such as breach of the duty of fair presentation of breach of warranty.30 The compulsory insurance contract is expressly excluded from the provisions of the Third Parties (Rights against Insurers) Act 2010.31 Under the two oil pollution conventions, the courts of contracting states which have suffered damage in their territory, territorial seas or exclusive economic zone have jurisdiction to hear any claim against the shipowner or the insurer. In contrast, the 2002 Athens Convention gives the passenger a choice of jurisdiction,32 such as the place of residence or principal place of business of the defendant, the contractual place of departure or destination, the claimant’s place of domicile or permanent residency or the place where the contract was made if the defendant has a place of business there and is subject to the jurisdiction of that state.

23 Through Transport Mutual Insurance Assoc (Eurasia) Ltd v New India Assurance Co Ltd (The Hari Bhum) [2004] EWCA (Civ) 1598, [2005] 1 Lloyd’s Rep 67; London Steam Ship Owners Mutual Insurance Assoc Ltd v Spain (The Prestige) [2015] EWCA Civ 333, [2015] 2 Lloyd’s Rep 33; Shipowners’ Mutual Protection and Indemnity Association (Luxembourg) v Containerships Denizcilik Nakliyat ve Ticaret AS (The Yusuf Cepnioglu) [2016] EWCA Civ 386, [2016] Bus LR 755. 24 This convention came into force on 30 May 1996, and is in force in 139 contracting states, including the UK (see Merchant Shipping Act 1995 (MSA 1995), ch III, ss 173–81), as at 8 August 2019. It is a Protocol to the 1969 International Convention on Civil Liability for Oil Pollution Damage. See M Tsimplis, ‘Marine Pollution from Shipping Activities’ in Baatz (n 4) 385–97, esp 392. 25 This convention came into force on 21 November 2008, and is in force in 93 contracting states, including the UK (see the Merchant Shipping (Oil Pollution) (Bunkers Convention) Regulations 2006, SI 2006/1244), as at 8 August 2019. See Tsimplis (n 24) 403–09, esp 408. 26 See M Tsimplis, ‘Carriage of Passengers’ in Baatz (n 4) 216–31, esp 226. 27 2001 BOPC, Art 7; MSA 1995, s 163A(2); 2002 Athens Convention, Art 4bis (1). 28 1992 CLC, Art VII.8; 2001 BOPC, Art 7.10; MSA 1995, s 165 (1A). Similar principles can be found in the 2010 Hazardous and Noxious Substances Protocol (HNS Convention 2010), adopted in May 2010, but this had not come into force as at 8 August 2019. See Tsimplis (n 24) 414–25. 29 1992 CLC, Art V. 30 For non-consumer contracts entered into after 12 August 2016, see the Insurance Act 2015, ss 3, 10 and 11. 31 MSA 1995, s 165(5). 32 2002 Athens Convention, Art 17(1).

94 Yvonne Baatz These international conventions involve very important policy considerations which demand a different approach from commercial contracts between marine cargo insurers and their insured shipowners, and claims by cargo claimants or their subrogated insurers against the carrier’s liability insurers for damage to cargo. Oil pollution concerns potentially enormous claims for damage to the environment caused by cargo or bunkers. Passengers will usually be consumers and therefore need protection, particularly in relation to life and limb.33

IV. The Global Position As we have already seen, there is no one approach to court jurisdiction clauses worldwide. The New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards 1958, recognising party autonomy for arbitration, has been widely accepted worldwide.34 However, it is only recently that a similar convention has come into force recognising party autonomy for jurisdiction clauses – the 2005 Hague Convention on Choice of Court Agreements – and that convention has so far been ratified by a comparatively small number of states.35 This is perhaps surprising, as arbitration involves the courts of a state giving up jurisdiction in favour of an arbitration tribunal chosen by the parties, and some states may be reluctant to do so. Some states are more pro-arbitration than others. They may consider that party choice is a good thing in itself. In addition, a thriving arbitration centre takes pressure off the court system of a state, which is expensive to provide, and generates income for that state. If party choice is a recognised good in itself, then it might appear that more states would be more enthusiastic about accepting court jurisdiction where their courts have been chosen by the parties. It is not surprising that the Choice of Court Convention has been ratified by the EU, as the approach of the convention shares much in common with that of the provisions on choice of jurisdiction in the EU Regulations.36 It remains to be seen whether the convention can attract support from other major economic powers, such as the USA and China. If it does, it is more likely to attract further support. There is some indication that this may happen. China signed the convention on 12 September 2017, although it has not yet ratified it. This shift in attitude to party

33 This is also reflected in the Third Parties (Rights against Insurers) Act 2010, s 9(5) and (6): see n 22. 34 Hereafter ‘the New York Convention’. It has been ratified by 160 states as at 29 July 2019, including the USA, China and all EU Member States, including the UK. 35 Hereafter ‘the Choice of Court Convention’. It entered into force on 1 October 2015 in Mexico and the EU, thus binding 27 of the EU Member States but not Denmark; on 1 October 2016 in Singapore; on 1 August 2018 in Montenegro; and on 1 September 2018 in Denmark. As the Choice of Court Convention was approved by the EU and not the UK, and as the UK wishes to ensure continuity of application of the convention even if it ceases to be a Member of the EU, it has acceded to the convention, but its accession was suspended until 1 November 2019 as the European Council agreed to extend the period of withdrawal of the UK from the EU until 31 October 2019 (on 28 October 2019 it agreed to a further extension until 31 January 2020). The UK Government stated in para 22 of its paper ‘Providing a Cross-Border Civil Judicial Cooperation Framework’ (22 August 2017) that its ‘intention to continue to be a leading member in the Hague Conference and to participate in those Hague Conventions to which we are already a party and those which we currently participate in by virtue of our membership of the EU’: www.gov.uk/government/publications/ providing-a-cross-border-civil-judicial-cooperation-framework-a-future-partnership-paper. See also the UK Government’s paper ‘Handling Civil Legal Cases that Involve EU Countries if There Is No Brexit Deal’ (13 September 2018) www.gov.uk/government/publications/handling-civil-legal-cases-that-involve-eu-countries-if-theres-no-brexit-deal/. See most recently the Convention of 2 July 2019 on the Recognition and Enforcement of Foreign Judgments in Civil and Commercial Matters, an international treaty concluded within the Hague Conference on Private International Law. It is not in force as at 8 August 2019. 36 See section V below.

English Jurisdiction Clauses in Marine Cargo Insurance 95 autonomy may be as a result of China’s One Belt, One Road Initiative, which involves a number of EU Member States which are already party to the convention. Other states, such as Australia, may also ratify.37 If more states ratify the convention, it will also be important to see what, if any, declarations those states make limiting its application. For example, states may declare that its courts may refuse to determine disputes to which an exclusive choice of court agreement applies if, except for the location of the chosen court, there is no connection between that state and the parties or the dispute.38 Or a state may declare that its courts may refuse to recognise or enforce a judgment given by a court of another contracting state if the parties were resident in the requested state, and the relationship of the parties and all other elements relevant to the dispute, other than the location of the chosen court, were connected only with the requested state.39 Party autonomy results in three basic principles in the Choice of Court Convention: (i) if parties to a contract choose an exclusive court jurisdiction agreement,40 that choice shall be given effect to unless the agreement is null and void under the law of that state;41 (ii) no other court should take jurisdiction;42 and (iii) any court judgment43 resulting from such choice should be recognised and enforced. Each of these principles is subject to exceptions. The Choice of Court Convention does not apply to some maritime contracts,44 but it does apply to contracts of insurance and reinsurance that relate to maritime matters, including cargo insurance.45 Moreover, while the EU and Denmark have both made declarations that the convention does not apply to some insurance contracts, those declarations state that it does apply to: reinsurance contracts;46 insurance contracts where the jurisdiction agreement is entered into

37 Parliament of Australia, Joint Standing Committee on Treaties, Report 166 – Australia’s Accession to the Convention on Choice of Court Agreements (The Hague, 30 June 2005) (23 November 2016) has recommended accession. See also A Mills, ‘The Hague Choice of Court Convention and Cross-Border Commercial Dispute Resolution in Australia and the Asia-Pacific’ (2017) 18 Melbourne Journal of International Law 1; BA Marshall and M Keyes, ‘Australia’s Accession to the Hague Convention on Choice of Court Agreements’ (2017) 41 Melbourne University Law Review 1; M Douglas, ‘Will Australia Accede to the Hague Convention on Choice of Court Agreements?’ (2017) 17 Macquarie Law Journal 148. 38 Choice of Court Convention, Art 19. Other declarations limiting the scope of the convention may be made under Arts 20 and 21, and the EU has made a declaration in relation to insurance contracts which is discussed in this section below. Singapore, Mexico and Montenegro have not made any declarations. 39 Choice of Court Convention, Art 20. 40 The Choice of Court Convention applies to an exclusive jurisdiction agreement (Art 3) which has been entered into after the convention has entered into force for the state of the court chosen (Art 16). Art 3(a) defines an ‘exclusive choice of court agreement’ as ‘an agreement concluded by two or more parties that meets the requirements of paragraph (c) and designates, for the purpose of deciding disputes which have arisen or may arise in connection with a particular legal relationship, the courts of one Contracting State or one or more specific courts of one Contracting State to the exclusion of the jurisdiction of any other courts’. An agreement is deemed to be exclusive if it designates the courts of one contracting state or one or more specific courts of one contracting state unless the parties have expressly provided otherwise (Art 3(b)). Art 3(c) requires an exclusive choice of court agreement to be concluded or documented: (i) in writing; or (ii) by any other means of communication which renders information accessible so as to be usable for subsequent reference. The court chosen has no discretion to decline jurisdiction (Art 5). However, states may extend its application to non-exclusive jurisdiction clauses by making a declaration (Art 22). 41 Choice of Court Convention, Art 5(1). 42 ibid Art 6. 43 ibid Art 8. 44 Carriage of passengers and goods (Art 2(2)(f)), marine pollution, limitation of liability for maritime claims, general average, and emergency towage and salvage (Art 2(2)(g)). The reason for these exclusions is that most of them are already subject to specific international conventions. The convention would apply to other maritime contracts, such as ship building contracts, ship sale and purchase contracts, international sale contracts and probably time charter parties, but not voyage charter parties. However, many of these contracts usually provide for arbitration. 45 Choice of Court Convention, Art 17. 46 Art 2(a) of the Declaration by the EU to the Choice of Court Convention dated 11 June 2015, the declaration by Denmark dated 30 May 2018 and the extension of the Choice of Court Convention by the UK to Gibraltar on 31 July 2019, which are in similar terms.

96 Yvonne Baatz after the dispute arises;47 and insurance of ‘large risks’, including ‘(ii) any loss of or damage to goods in transit or baggage other than passengers’ baggage, irrespective of the form of transport’,48 ‘(iv) any liability, other than for bodily injury to passengers or loss of or damage to their baggage, for loss or damage caused by goods in transit or baggage as referred to in point (ii)’,49 any risk or interest connected with any of these risks50 or ‘(viii) any other risks where the policy holder carries on a business of a size which exceeds the limits of at least two of the following criteria: (a) a balance-sheet total of EUR 6.2 million; (b) a net turnover of EUR 12.8 million; (c) an average number of 250 employees during the financial year’. These declarations reflect the special rules on jurisdiction ‘in matters relating to insurance’,51 which apply to protect a weaker party who seeks to sue an insurer or its branch, agency or other establishment in an EU Member State in the Recast Brussels Regulation and its predecessors.52 There are no provisions in the Choice of Court Convention which provide whether a subrogated insurer is bound by any jurisdiction clause in the contract on which it sues. In any event, if that contract is a bill of lading, the convention does not apply.53 Nor does the convention provide for whether a jurisdiction agreement in a marine insurance contract will bind a cargo claimant or its subrogated insurer where either or both of them bring a direct action against a liability insurer such as a P&I club which has covered the carrier’s liability in relation to carriage of cargo. If there is a right of direct action against the liability insurer of the carrier, usually pursuant to statute,54 and sometimes where the carrier has become insolvent, this question is left to be decided by national law, and the national law of different states may vary significantly.55 In summary, the Choice of Court Convention provides some international harmony concerning marine insurance contracts, but there is no international harmony as to whether a cargo claimant or its subrogated insurer is bound by an exclusive English jurisdiction clause in a marine insurance contract if it brings a direct action against the liability insurer of the carrier, and this may lead to conflicts.

V. The European Position The European Union has developed a sophisticated system of rules on jurisdiction and applicable law to ensure the ‘proper functioning of the internal market’.56 At the heart of these rules is the principle of party autonomy.

47 ibid Art 2(b). 48 ibid Art 2(d)(ii). 49 ibid Art 2(d)(iv). 50 ibid Art 2(d)(vi). 51 See s 3 of ch II of the Recast Brussels Regulation (ie Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2012] OJ L351/1). 52 Discussed below at section V. 53 See n 44 above. 54 For the position under English law, see section III above. 55 For the positions under the Brussels Recast Regulation and under English law, see sections V and VI, respectively. On the conflict of laws which can result, see A Mills, The Confluence of Public and Private International Law, Justice, Pluralism and Subsidiarity in the International Constitutional Ordering of Private Law (Cambridge, Cambridge University Press, 2009) 16–17. 56 Recast Brussels Regulation, Recital 12. The UK Government has also recognised the need for such a system even if the UK is no longer a member of the EU. See HM Government, ‘Providing a Cross-Border Civil Judicial Cooperation Framework: A Future Partnership Paper’ paras 14–15, assets.publishing.service.gov.uk/government/uploads/system/ uploads/attachment_data/file/639271/Providing_a_cross-border_civil_judicial_cooperation_framework.pdf.

English Jurisdiction Clauses in Marine Cargo Insurance 97 The Recast Brussels Regulation57 applies to legal proceedings instituted on or after 10 January 2015.58 The regulation seeks to determine the international jurisdiction of the courts of the EU Member States so that all EU Member States are bound by the same rules and will therefore recognise and enforce each other’s judgments speedily. It provides for certain and predictable rules with little discretion.59 The Recast Brussels Regulation applies to ‘civil and commercial matters’,60 with some exceptions. Those exceptions that are particularly relevant to maritime disputes are where there are proceedings relating to the winding up of insolvent companies61 or arbitration.62 The basic rule is that a defendant domiciled in an EU Member State must be sued in that state.63 It does not matter where the claimant is domiciled.64 There are a number of important exceptions to this basic rule. Most notable for present purposes are where the parties have chosen the court of an EU Member State to have exclusive jurisdiction and the jurisdiction agreement satisfies the formalities of Article 25: the defendant has submitted to the jurisdiction of the court of an EU Member State without contesting its jurisdiction;65 or an international convention must conflict with the Recast Brussels Regulation.66 If none of the exceptions apply, the claimant may have some additional choices under Article 7, 8 or 9.

A. Jurisdiction Agreements Party autonomy is recognised as an important principle, although it is limited in relation to insurance (but not marine insurance),67 consumer and employment contracts, because the insured, the consumer and the employee, respectively, are given consumer protection.68 57 The Recast Brussels Regulation applies to all EU Member States, including Denmark according to the Agreement between the European Community and the Kingdom of Denmark on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (OJ L79/4, 21 March 2013). It replaced the Brussels I Regulation, ie Council Regulation (EC) 44/2001 of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters. See also the revised Lugano Convention which entered into force in Denmark (1 January 2010), Norway (1 January 2010), Switzerland (1 January 2011), the EU (1 January 2010) and Iceland (1 May 2011); this replaced the EFTA Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters 1988. 58 Recast Brussels Regulation, Art 66. It is assumed for the purposes of this chapter that any proceedings are commenced after that date. Much of the case law referred to is decided on the provisions of the Brussels I Regulation or its predecessors, but where the wording of the provisions has not changed, the case law is equally applicable to the Recast Brussels Regulation: Case C-543/10 Refcomp SpA v Axa Corporate Solutions Assurance SA [2013] 1 All ER (Comm) 1201; Case C-302/13 flyLAL-Lithuanian Airlines [2014] 5 CMLR 27 [25]. 59 See, eg Case C-281/02 Owusu v Jackson [2005] QB 801; Case C-386/05 Color Drack GmbH v Lexx International Vertriebs GmbH [2007] ECR I-3699, [2010] Bus LR 1044 [19], [20], [32]–[33]; but the Recast Brussels Regulation, Arts 33 and 34 expressly provide for discretion. 60 Recast Brussels Regulation, Art 1(1). 61 Recast Brussels Regulation, Art 1(2)(b). The reason for this exclusion is that the Recast Insolvency Regulation (ie Regulation (EU) 2015/848 of the European Parliament and of the Council of 20 May 2015 on Insolvency Proceedings) applies to insolvency proceedings opened after 26 June 2017 (Art 84(1)) and Council Regulation (EC) No 1346/2000 of 29 May 2000 on insolvency proceedings applies to insolvency proceedings opened before that date (Art 84(2) of the Recast Insolvency Regulation). 62 Recast Brussels Regulation, Art 1(2)(d). 63 Recast Brussels Regulation, Art 4. 64 Case C-412/98 Universal General Insurance Co v Group Josi Reinsurance Co SA [2000] ECR I-3699, [2001] QB 68. 65 Recast Brussels Regulation, Art 26. See, eg Case C-150/80 Elefanten Schuh v Jacqmain [1981] ECR 1671; Case C-111/09 Ceská Podnikatelská Pojiìtovna as, Vienna Insurance Group v Bilas [2010] Lloyd’s Rep IR 734; Case C-1/13 Cartier Parfums-Lunettes SAS v Ziegler France SA [2014] ILPr 25; Deutsche Bank AG London Branch v Petromena ASA [2015] EWCA Civ 226, [2015] 1 WLR 4225; Case C-175/15 Taser International Inc v SC Gate 4 Business SRL [2016] QB 887. 66 Recast Brussels Regulation, Art 71. See this section below. 67 Recast Brussels Regulation, Art 15(5). See J Dunt, Marine Cargo Insurance, 2nd edn (London, Informa Law 2015) ch 2. See further this section below. 68 See Recast Brussels Regulation, Recital 14 and ss 3, 4 and 5.

98 Yvonne Baatz Autonomy in this context relates to a jurisdiction agreement. It is irrelevant that the parties have chosen the law applicable to the agreement, as this does not found jurisdiction under the Recast Brussels Regulation.69 Therefore, if the parties have provided that their contract shall be governed by English law, this will not give the English court jurisdiction under the Recast Brussels Regulation. Article 25 provides that where the parties have chosen the courts of an EU Member State, no matter where the parties to the jurisdiction agreement are domiciled, the court chosen shall have jurisdiction. Therefore, if both parties are domiciled in non-EU Member States, but have chosen a neutral venue to determine their disputes in an EU Member State, that choice will be upheld. The court chosen has no discretion whether to exercise jurisdiction on forum non-conveniens70 or other grounds where there is a valid jurisdiction clause in its favour, whether the jurisdiction agreement is exclusive71 or non-exclusive.72 The parties will only have made a choice within Article 25(1) of the Recast Brussels Regulation if the jurisdiction agreement satisfies certain formalities. Those formalities are intended to ensure legal certainty and that the parties have consented to the jurisdiction agreement.73 The standard of proof is that the claimant must establish that they have the better of the arguments on the material presently available. The question is one of relative plausibility and does not impose a balance of probabilities standard.74 The Recast Brussels Regulation strengthens the principle of party autonomy by permitting the court chosen to determine the substantive validity of a jurisdiction clause in accordance with the law of the EU Member State chosen in the jurisdiction agreement.75 It is for the national court to interpret the jurisdiction clause invoked before it in order to determine which disputes fall within its scope.76 Another important respect in which the Recast Brussels Regulation strengthens party autonomy is that its lis pendens provisions77 are subject to Article 31(2),78 which provides that 69 cf the position at common law under para 3.1(6)(c) of Practice Direction 6B supplementing Part 6 of the Civil Procedure Rules 1998 (‘CPR’). See section VI below. 70 See section VI below. 71 Art 25(1) of the Recast Brussels Regulation provides that the jurisdiction of an EU Member State will be exclusive unless the parties have agreed otherwise. Therefore, if the agreement simply provides for the jurisdiction of the English High Court, that court will have exclusive jurisdiction: Case C-322/14 El Majdoub v CarsOnTheWeb.Deutschland GmbH [2015] 1 WLR 3986 [24]. If the clause provides for the ‘exclusive and final jurisdiction of the courts of Paris’, this will be sufficiently clear that it provides for the exclusive jurisdiction of the courts of France: Case C-222/15 Höszig Kft v Alstom Power Thermal Services [2016] ILPr 36. 72 See IMS SA v Capital Oil and Gas Industries Ltd [2016] EWHC 1956 (Comm), [2016] 4 WLR 163, where there was an exclusive English jurisdiction clause. The English court has also held that where the jurisdiction agreement is nonexclusive, it does not have a discretion to decline jurisdiction – on grounds of forum non conveniens or lis alibi pendens: UCP plc v Nectrus Ltd [2018] EWHC 380 (Comm), [2018] 1 WLR 3409. Compare the position under the Choice of Court Convention as stated in n 40 above. 73 El Majdoub (n 71). 74 Bols Distilleries BV (t/a Bols Royal Distilleries) v Superior Yacht Services Ltd [2006] UKPC 45, [2007] 1 WLR 12; Joint Stock Co ‘Aeroflot-Russian Airlines’ v Berezovsky [2013] EWCA Civ 784, [2013] 2 Lloyd’s Rep 242; Quinn Building Products Ltd v P&S Civil Works Ltd [2013] NIQB 142; Mar-Train Heavy Haulage Ltd v Shipping.DK Chartering A/S [2014] EWHC 355 (Comm); Goldman Sachs International v Novo Banco SA [2018] UKSC 34, [2018] 1 WLR 3683; IMS (n 72) [45]–[52] (where Popplewell J did not think that either party had the better of the arguments); Deutsche Bank AG v Comune di Savona [2017] EWHC 1013 (Comm), [2018] 1 BCLC 358; Aspen Underwriting Ltd v Credit Europe Bank NV [2018] EWCA Civ 2590, [2019] 1 Lloyd’s Rep 221 [26]–[33]; Kaefer Aislamientos SA de CV v AMS Drilling Mexico SA de CV [2019] EWCA Civ 10, [2019] 1 WLR 3514; Pan Ocean Co Ltd v China-Base Group Co Ltd [2019] EWHC 982 (Comm). 75 Recast Brussels Regulation, Art 25(1). 76 Höszig (n 72). 77 Recast Brussels Regulation, Art 29. 78 Perella Weinberg Partners UK LLP v Codere SA [2016] EWHC 1182 (Comm) (non-exclusive jurisdiction agreement); Commerzbank Aktiengesellschaft v Liquimar Tankers Management Inc [2017] EWHC 161 (Comm), [2017] 1 WLR 3497

English Jurisdiction Clauses in Marine Cargo Insurance 99 where a court has been chosen in accordance with Article 25, any court of another EU Member State shall stay its proceedings until such time as the court seised on the basis of the agreement declares that it has no jurisdiction under the agreement.79 Where the court chosen has established that it has jurisdiction, any court of another EU Member State shall decline jurisdiction in favour of that court.80 Recital 22 clarifies that the designated court ‘has priority to decide on the validity of the agreement and on the extent to which the agreement applies to the dispute pending before it’,81 even if it is second seised and irrespective of whether the other court has already decided on the stay of proceedings. Where, however, there is a conflict as to whether both courts have been chosen, then the court first seised will determine the validity of the jurisdiction clause.82 Special rules on jurisdiction ‘in matters relating to insurance’ apply to protect a weaker party who seeks to sue an insurer or its branch, agency or other establishment in an EU Member State.83 In such matters, instead of the general rule that the claimant must sue the defendant where it is domiciled if it is domiciled in an EU Member State,84 the insured has the choice of suing, inter alia, where the insurer is domiciled85 or where the insured itself is domiciled.86 Jurisdiction agreements87 are only permitted in limited circumstances. Those circumstances include where the insurance covers a large risk, such as marine or aviation risks,88 as parties who insure against such risks do not need protection. Where a contract of insurance, entered by an insurer and a policy holder, and stipulated by the latter to be for his benefit and to enure for the benefit of third parties to such a contract, contains a clause conferring jurisdiction relating to proceedings which might be brought by such third parties, the latter, even if they have not expressly signed the said clause, may rely upon it. However, this is subject to the provisos that, as between the insurer and the policy holder, the condition as to writing laid down by what is now Article 25 of the Recast Brussels Regulation has been satisfied, and that the consent of the insurer in that respect has been clearly manifested.89 (where Cranston J held that asymmetric jurisdiction clauses in loan agreements were exclusive jurisdiction agreements). In Dexia Crediop SpA v Provincia Brescia [2016] EWHC 3261 (Comm) it was held that there was not the same cause of action in the Italian and English proceedings but, had there been, Art 31(2) would have applied. 79 See Recital 22 and Arts 29(1) and 31(2). This solution is similar to that adopted by Art 6 of the Choice of Court Convention – see n 42. These provisions reverse the decision in Case C-116/02 Erich Gasser GmbH v MISAT SRL [2003] ECR I-14693, [2005] QB 1 (and see also Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG (The Alexandros T) [2013] UKSC 70, [2014] Bus LR 873). They are a major improvement. 80 Recast Brussels Regulation, Art 31(3). 81 See also Case C-352/13 Cartel Damage Claims (CDC) Hydrogen Peroxide SA v Akzo Nobel NV [2015] QB 906. 82 Recast Brussels Regulation, Recital 22 para 2. 83 See s 3 of ch II of the Recast Brussels Regulation and s 3 of ch II of the Brussels I Regulation. 84 Recast Brussels Regulation, Art 4. 85 Recast Brussels Regulation, Art 11(1)(a). 86 Recast Brussels Regulation, Art 11(1)(b). There are other choices; for example, where the insurer (whether or not domiciled in an EU Member State) has a branch, agency or other establishment in an EU Member State and the dispute arises out of the operations of that branch, agency or other establishment, then the claimant may bring proceedings in the courts where the branch, agency or other establishment is situated (Recast Brussels Regulation, Arts 7(5) and 11(2)). Where there are co-insurers, see Recast Brussels Regulation, Art 11(1)(c). Where there is liability insurance or insurance of immovable property, see Recast Brussels Regulation, Art 12. 87 Recast Brussels Regulation, Arts 15–16. 88 Recast Brussels Regulation, Arts 15(5) and 16. See the declaration made by the EU when it approved the Choice of Court Convention (see nn 46–52), which is similar to this provision. 89 Case C-201/82 Gerling Konzern Speziale Kreditversicherungs AG v Amministrazione del Tesoro dello Stato [1983] ECR 2503. See also Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG (The Alexandros T) [2014] EWHC 3068 (Comm), [2015] 2 All ER (Comm) 747; Aspen Underwriting Ltd v Kairos Shipping Ltd (The Atlantik Confidence) [2017] EWHC 1904 (Comm), [2017] 2 Lloyd’s Rep 295; [2018] EWCA Civ 2590, [2019] 1 Lloyd’s Rep 221 (under appeal). In The Atlantik Confidence, the CA upheld the decision of Teare J that a mortgagee bank was not bound by the English jurisdiction clause in an insurance policy or in a settlement agreement on that policy. See also J Dunt, Marine Cargo Insurance, 2nd edn (London, Informa Law at Routledge, 2015) ch 2.

100 Yvonne Baatz There are provisions on direct action90 which enable an injured party to choose where to bring an action directly against the insurer where such direct actions are permitted91 in a range of different courts.92 Unless the policy holder or the insured may be joined as a party to the direct action in accordance with the law governing that action (in which case the same court will have jurisdiction over both the direct action and the claim against the policy holder or the insured93), an insurer may only bring proceedings in the courts of the Member State in which the defendant is domiciled, irrespective of whether the defendant is the policy holder, the insured or a beneficiary.94 It was not made explicit whether the provisions on direct action applied even where the insurance contract provided for a particular court to have jurisdiction. The Court of Justice of the European Union has recently considered whether an exclusive jurisdiction agreement in a marine insurance contract is binding on a claimant bringing a direct action against a liability insurer. Although the case was not about damage to cargo but concerned damage done by a ship to a quay, the decision would apply by analogy to a claim by a cargo claimant. In Assens Havn v Navigators Management (UK) Ltd (The Sea Endeavour I),95 Skane Entreprenad Service AB (SES), a Swedish company, chartered the tug Sea Endeavour I to carry sugar beet from Assens, a port in Denmark, to another port in Denmark, Nakskov. SES took out liability insurance with Navigators. The insurance policy contained an English law and exclusive English court jurisdiction clause. The Sea Endeavour I damaged the quay installations at Assens Havn. Danish law gives the right to bring a direct action against a company providing insurance cover for the party that has caused the harm where, for example, the policy holder has gone into liquidation.96 SES had gone into liquidation and therefore Assens Havn brought a direct action before the Maritime and Commercial Court in Denmark against Navigators as the liability insurer of the party that allegedly caused the harm. The Danish Supreme Court referred the issue of whether the Danish court was bound to stay its proceedings as a result of the provisions of the Brussels I Regulation to the Court of Justice of the European Union. The equivalent provisions of the Recast Brussels Regulation are Article 15(5), which provides that the special provisions in Section 3 of Chapter 2 relating to insurance can only be departed from by a jurisdiction agreement if that agreement satisfies one of the conditions set out in that article, including if it relates to a contract of insurance insofar as it covers one or more of the risks set out in Article 16 of the Recast Brussels Regulation. This contract did cover such a risk, ie any liability arising out of the use or operation of seagoing ships arising from perils that relate to their use for commercial purposes.97 Was such an agreement, which was validly concluded between the insurer and the policy holder, binding on the injured party, who is permitted under national law to bring an action directly against the insurer?

90 Recast Brussels Regulation, Art 13. 91 See Art 18 of the Rome II Regulation (ie Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations [2007] OJ L199/40). Rome II applies to events giving rise to damage occurring after 11 January 2009: Case C-412/10 Homawoo v GMF Assurance SA & Ors [2011] ECR I-11603. Events giving rise to damage which occurred before 11 January 2009 are governed by the Private International Law (Miscellaneous Provisions) Act 1995, ss 9–15 (which entered into force on 1 May 1996): Mapfre Mutualidad Cia de Seguros Y Reaseguros SA v Keefe [2015] EWCA Civ 598, [2016] 1 WLR 905. 92 Outlined in the Recast Brussels Regulation, Arts 10–12. 93 Recast Brussels Regulation, Art 13(3). See Mapfre (n 91). 94 Recast Brussels Regulation, Art 14(1). 95 Case C-368/16 Assens Havn v Navigators Management (UK) Ltd (The Sea Endeavour I) [2018] QB 463. 96 Forsikringsaftaleloven [= Law on Insurance Contracts], Art 95. Compare the position under English law, where the Third Parties (Rights against Insurers) Act 2010 also gives such a right of direct action. However, the liability insurer is entitled to rely on any pay to be paid provision in the liability insurance: see section III above. 97 Recast Brussels Regulation, Art 16(2)(a).

English Jurisdiction Clauses in Marine Cargo Insurance 101 The Court of Justice of the European Union referred to the scheme of the provisions in Section 3 of Chapter II on insurance and their underlying objectives. Section 3 establishes an autonomous system for the conferral of jurisdiction in matters of insurance as determined in Société Financière et Industrielle du Peloux v Axa Belgium.98 As with consumers and employees, there is an ‘imbalance between the parties’,99 and the rules seek to address this by giving the weaker party rules of jurisdiction that are more favourable to its interests than the general rules.100 One example of this is that the victim of insured damage can sue the liability insurer before the courts of the place where the harmful event occurred, provided that the national law permits such a direct action as a result of the Recast Brussels Regulation, Articles 12 and 13(2). It is possible to derogate from the provisions of Section 3 by a jurisdiction agreement, in particular under Article 15(5), where the insurance covers specified risks, as it did in this case. However, the Court of Justice of the European Union concluded that Article 13(2) does not refer to Articles 15 and 16 and jurisdiction agreements. The court had previously noted in Peloux that in matters of insurance, choice of jurisdiction should be strictly limited in order to protect the economically weaker party.101 Article 15 exhaustively lists the cases in which the parties may derogate from the rules laid down in Section 3. The court also referred to the equivalent of Article 25(4) of the Recast Brussels Regulation, which provides that an agreement shall have no legal force if it is contrary to Article 15. As regards the third party element, the court stated: Moreover it is clear that the situation of a third party victim of insured damage is even farther removed from the contractual relationship involving an agreement on jurisdiction than an insured beneficiary who did not expressly consent to that agreement, as referred to in [Peloux].102

The court concluded that the victim of insured damage who brings a direct action against the liability insurer is not bound by a jurisdiction agreement between the insured and its insurer, but may bring a direct action against the insurer where the harmful event occurred or where the victim is domiciled, a possibility accepted by the European Court of Justice in FBTO Schadeverzekeringen NV v Odenbreit.103 If jurisdiction agreements were to bind the claimant in a direct action, this could compromise the objective of protecting the economically and legally weaker party. The decision in The Sea Endeavour I is very wide, and is a worrying development for liability insurers in the marine market and of large risks, such as P&I clubs. It is arguably wrong, but, even

98 Case C-112/03 Société Financière et Industrielle du Peloux v Axa Belgium [2005] ECR I-3707, [2006] QB 251 [29]. 99 Case C-77/04 Groupement d’Interet Economique (GIE) Reunion Europeenne v Zurich Espana [2005] ECR I-4509, [2006] 1 All ER (Comm) 488. 100 Case C-347/08 Vorarlberger Gebietskrankenkasse v WGV-Schwabische Allgemeine Versicherungs [2009] ECR I-8661, [2010] Lloyd’s Rep IR 77 [40]; Case C-340/16 Landeskrankenanstalten-Betriebsgesellschaft – KABEG v Mutuelles du Mans Assurances – MMA IARD SA [2018] Lloyd’s Rep IR 556; Case C-106/17 Hofsoe v LVM Landwirtschaftlicher Versicherungsverein Munster AG [2018] Lloyd’s Rep IR 608; The Atlantik Confidence (n 89) (where the decision of Teare J was given before the judgments of the CJEU in KABEG and Hofsoe). See further Y Baatz, ‘Matters Relating to Insurance and Protecting the Weaker Party’ [2018] Lloyd’s Maritime and Commercial Law Quarterly 1 and SE Tan, ‘Restitution and Insurance in the Brussels I Regulation Recast’ [2018] Lloyd’s Maritime and Commercial Law Quarterly 9, both published after the judgment in KABEG but before the judgment in Hofsoe, and arguing that Teare J was wrong to hold that a mortgagee bank was not a weaker party in the light of the decision in KABEG. The CA then upheld Teare J’s decision. The appeal to the UK Supreme Court is fixed for November 2019. This author remains of the view that the bank was a weaker party and that to hold the bank to be an insurance professional was to stretch this class too wide; also, the bank did not take part in the insurance claim. See too J Gilman et al, Arnould: Law of Marine Insurance and Average, 19th edn (London, Sweet & Maxwell, 2018) and First Supplement (2019) para 5.06. 101 The Sea Endeavour I (n 95) [31]. 102 ibid [39]. 103 FBTO Schadeverzekeringen (n 11) [31].

102 Yvonne Baatz if it is right, it should be limited to its facts. Article 15 begins, ‘The provisions of this Section may be departed from only by an agreement’. This wording appears to apply Article 15 to all of the provisions of Section 3, including those in Article 13 on direct action. Furthermore, the underlying objectives of the section do not justify the decision. It is clear from Recital 18 that ‘in relation to insurance … the weaker party should be protected by rules of jurisdiction more favourable to his interests than the general rules’. Yet Recital 19 recognises that limited party autonomy is allowed in relation to insurance. That limited autonomy is in relation to insurance covering certain types of risk where, in effect, the insured is not a weaker party and is not therefore in need of any consumer protection. Such risks include, for example, marine risks and large risks.104 If the insurance covers such risks, a jurisdiction agreement is permitted. It is arguable that any third party also claiming on that insurance policy is not in need of protection because of the type of risk. Even if Assens Havn was a weaker party on the facts of The Sea Endeavour I, in accordance with the decision in KABEG,105 the third party claiming on a direct action would often be a subrogated insurer who was not a ‘weaker party’ in accordance with Vorarlberger106 and Hofsoe.107 Such an insurer should not be protected and should be bound by the jurisdiction agreement.108 For example, if A’s goods are damaged as a result of a breach of the bill of lading by the carrier, A can claim against its cargo insurer. The latter would, in turn, exercise its subrogated rights to sue the carrier, and if the latter had become insolvent would bring a direct action against the liability insurers of the carrier, the latter’s P&I club. The combined effect of KABEG and The Sea Endeavour I is to provide very wide protection to claimants bringing a direct action against insurers in the courts of an EU Member State. Liability insurers may find themselves sued in courts which they had not agreed to. The combination of the uncertainty as to the extent of the decision in The Sea Endeavour I, together with the uncertainties of Brexit,109 is likely to drive liability insurers to arbitration, so that their choice of tribunal will be upheld by the English courts.110

104 Recast Brussels Regulation, Art 16; Brussels I Regulation, Art 14. 105 In KABEG (n 100), the CJEU did not distinguish between assureds that are weaker parties and those which are economically strong enough not to be such a party. This is odd, as the definition of large risks (in respect of which a jurisdiction agreement is permitted) does distinguish assureds. See also the declaration by the EU to the Choice of Court Convention (on which see section IV), which includes in ‘large risks’ both marine risks and also ‘(viii) any other risks where the policy holder carries on a business of a size which exceeds the limits of at least two of the following criteria: (a) a balance-sheet total of EUR 6,2 million; (b) a net turnover of EUR 12,8 million; and (c) an average number of 250 employees during the financial year’. See the discussion in Aspen Underwriting v Credit Europe Bank (n 74) [80]–[122]. 106 Vorarlberger (n 100). 107 Hofsoe (n 100). 108 Contrast The Sea Endeavour I (n 95), where there was a jurisdiction agreement, with the position under English law, where there is an arbitration agreement in the liability insurance – see n 23. Also compare the position under the Rome II Regulation, Art 18, which provides that a claim against a liability insurer may be brought by the person having suffered the damage if the law applicable to the non-contractual obligation or the law applicable to the insurance contract so provides. 109 On the possible consequences of the UK leaving the EU, see S Masters and B McRae, ‘What Does Brexit Mean for the Brussels Regime?’ [2016] Journal of International Arbitration 483; A Dickinson, ‘Back to the Future: the UK’s Exit and the Conflict of Laws’ [2016] Journal of International Private Law 195; R Aikens and A Dinsmore, ‘Jurisdiction, Enforcement and the Conflict of Laws in Cross-Border Commercial Disputes: What Are the Legal Consequences of Brexit?’ [2016] European Business Law Review 903. See also HM Government (n 56) para 22: ‘we will seek to continue to participate in the Lugano Convention that, by virtue of our membership of the EU, forms the basis for the UK’s civil judicial cooperation with Norway, Iceland and Switzerland’. 110 See text to n 23.

English Jurisdiction Clauses in Marine Cargo Insurance 103

B. Other International Conventions Article 71 of the Recast Brussels Regulation111 provides that the regulation shall not affect any conventions to which the EU Member States are parties and which govern jurisdiction in relation to particular matters. The Choice of Court Convention will be important in relation to jurisdiction agreements.112 So, if there is a jurisdiction agreement in favour of Singapore and proceedings are commenced in England, the English court would be obliged to stay its proceedings in favour of Singapore pursuant to Article 6 of the Choice of Court Convention, which would prevail over Articles 33 and 34 of the Recast Brussels Regulation as a result of Article 71. The 1992 International Convention on Civil Liability for Oil Pollution Damage, the 2001 International Convention on Civil Liability for Bunker Oil Pollution Damage and the 2002 Athens Convention113 all contain provisions on jurisdiction which again would prevail over the Recast Brussels Regulation.

VI. The English Common Law A. Applicability of the English Common Law Rules At present, if an English jurisdiction agreement in a marine insurance contract satisfies the requirements of Article 25 of the Recast Brussels Regulation and the proceedings are commenced on or after 10 January 2015, the English court has jurisdiction by virtue of that article, no matter where the parties are domiciled, and the English court has no discretion.114 However, the English common law rules on English jurisdiction agreements may still apply in certain situations. For example, where: 1. Proceedings were commenced before 10 January 2015, because, unlike Article 25(1) of the Recast Brussels Regulation, Articles 4 and 23(3) of its predecessor, the Brussels I Regulation, have the effect that the common law rules apply where neither party to the English jurisdiction clause is domiciled in an EU Member State. If, for example, an insured domiciled in Hong Kong wished to rely on an English jurisdiction clause in a marine insurance contract with an insurer domiciled in Australia, the English court would then apply the common law rules under which it has a discretion.115 2. Articles 4 and 23(3) of the Revised Lugano Convention, which applies currently between the Lugano contracting states, provide that the common law rules apply where neither party to the English jurisdiction clause is domiciled in a Lugano contracting state. 3. If the UK were to leave the EU without an agreement on jurisdiction,116 Article 25 of the Recast Brussels Regulation and Article 23 of the Revised Lugano Convention would no 111 See Y Baatz, ‘The Conflict of Laws’, in Baatz (n 4) 40–42; Y Baatz, ‘Forum Selection in Contracts for the Carriage of Goods by Sea: The European Dimension’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 208. 112 See section IV. 113 See section III. 114 Although Art 6 of the Recast Brussels Regulation provides that if the defendant is not domiciled in an EU Member State, the jurisdiction of the courts of each Member State shall be determined by the law of that Member State, Art 6 is subject to other provisions, including Art 25. 115 Akai Pty Ltd v People’s Insurance Co Ltd [1998] 1 Lloyd’s Rep 90 (Com Ct). 116 As to which possibility, see n 109.

104 Yvonne Baatz longer apply. The UK would then apply the Choice of Court Convention as between the contracting states to that convention, including those of the EU,117 but would otherwise apply the common law. The English common law rules also apply where the English court has jurisdiction over a party not domiciled in an EU Member State or a Lugano contracting state but there is a jurisdiction clause in favour of a court not in such a state.

B. The English Common Law Rules on Choice of Jurisdiction The English common law rules are to be found in the Civil Procedure Rules118 and case law. Party autonomy plays an important role in these rules. Jurisdiction is founded as of right for in personam proceedings where a defendant is served with proceedings within the jurisdiction in accordance with Part 6 of CPR. A stay will only be granted on the ground of forum non-conveniens where the defendant can show that England is not the natural or a ppropriate forum for the trial and that there is another available forum which is clearly or distinctly more appropriate than England: ie the case may be tried more suitably for the interests of all the parties and the ends of justice.119 A stay would be very unlikely where the parties have chosen the English court. Where the defendant is not within the jurisdiction, the claimant requires the permission of the court to serve proceedings on the defendant outside the jurisdiction. The claimant must satisfy the court that it has a good arguable case that the claim against the foreign defendant falls within one or more of the grounds in paragraph 3.1 of Practice Direction 6B supplementing Part 6 of CPR. Party autonomy is very important, as such grounds include a claim in respect of a marine insurance contract120 where the contract provides for English law121 and/or English court jurisdiction122 or a claim for a declaration that no marine insurance contract exists, for example, where an insurer alleges that the contract has been avoided due to misrepresentation and/or nondisclosure, where, if it was found to exist, it would provide for English law and/or English court jurisdiction.123 Permission to serve the claim form out of the jurisdiction cannot be granted unless the court is ‘satisfied that England and Wales is the proper place in which to bring the claim’.124 In determining this issue, the English court will uphold a jurisdiction clause, for example by granting a stay of its proceedings if there is a foreign jurisdiction clause, or granting an anti-suit injunction125 to restrain the other party from commencing or pursuing foreign proceedings in breach of an 117 As to which, see the text to n 35. 118 The Civil Procedure Rules 1998 (CPR) replaced the Rules of the Supreme Court in April 1999. They will be referred to here except where case law is discussed which was decided under the Rules of the Supreme Court. 119 Spiliada Maritime Corp v Cansulex Ltd (The Spiliada) [1987] AC 460. 120 Global 5000 Ltd v Wadhawan [2012] EWCA Civ 13, [2012] 1 Lloyd’s Rep 239. See also A Dickinson, ‘Service Out of the Jurisdiction in Contract Cases: Straightening Out the Deck Chairs’ [2012] Lloyd’s Maritime and Commercial Law Quarterly 181. 121 CPR Practice Direction 6B, para 3.1(6)(c). This differs from the Recast Brussels Regulation, which does not confer jurisdiction on the basis of the governing law of the contract. 122 CPR Practice Direction 6B, para 3.1(6)(d). 123 CPR Practice Direction 6B, para 3.1(8). 124 CPR, r 6.37(3). See also The Spiliada (n 119). 125 The English court may grant an anti-suit injunction to restrain a party from commencing or continuing proceedings in a foreign court, other than the courts of an EU Member State (see Case C-159/02 Turner v Grovit [2004] ECR I-3565, [2005] 1 AC 101) or a Lugano contracting state, if it would be just and convenient to do so under the Senior Courts Act 1981, s 37. There are sanctions for breach: Compania Sud Americana (n 3).

English Jurisdiction Clauses in Marine Cargo Insurance 105 English jurisdiction clause, unless there are strong reasons not to. In Donohue v Armco Inc, Lord Bingham stated that the general rule is clear: where parties have bound themselves by an exclusive jurisdiction clause effect should ordinarily be given to that obligation in the absence of strong reasons for departing from it. Whether a party can show strong reasons, sufficient to displace the other party’s prima facie entitlement to enforce the contractual bargain, will depend on all the facts and circumstances of the particular case.126

However, in that case, an anti-suit injunction was not granted to enforce an exclusive English jurisdiction clause as there were already proceedings between multiple parties in New York, some of whom were not bound by the English jurisdiction agreement. As part of its policy of enforcing party autonomy, the English court may also award damages for breach of an English jurisdiction clause.127 This may be combined with a freezing injunction.128 An exclusive English jurisdiction clause has been upheld in insurance cases, even in the face of legislation in the other jurisdiction which would make the English jurisdiction clause invalid.129 Where the jurisdiction clause does not expressly state that the jurisdiction of the English court is exclusive, it will be necessary to construe whether the clause is an exclusive choice of jurisdiction or whether it merely gives the parties an option to sue in England. An agreement to English law and jurisdiction is generally an agreement to the exclusive jurisdiction of the English court, unless there is some indication that non-exclusive jurisdiction was intended, and is not simply an option for either party to commence proceedings in England but an obligation on both parties to commence proceedings there.130 This is in line with the approach of the Choice of Court Convention131 and that of the EU Regulations.132 Nevertheless, where parties are drafting a contract, it would be wise to state that the jurisdiction is exclusive if that is what is required. There is a clear trend in the English courts towards a wide interpretation of English jurisdiction clauses. This is evident in the recent case law on issues such as the scope of the disputes covered by the clause and the persons who may be bound by or benefit from such a clause.133 126 Donohue v Armco Inc [2001] UKHL 64, [2002] 1 Lloyd’s Rep 425 [24]. 127 Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG (The Alexandros T) [2014] EWCA Civ 1010, [2014] 2 Lloyd’s Rep 544; noted in A Dickinson, ‘Once Bitten – Mutual Distrust in European Private International Law’ [2015] LQR 186. Damages for the tort of procuring a breach of contract may also be available: Horn Linie GmbH & Co v Panamericana Formas E Impresos SA (The Hornbay) [2006] EWHC 373 (Comm), [2006] 2 Lloyd’s Rep 44 [26]; AMT Futures Ltd v Marzillier [2017] UKSC 13, [2018] AC 439. A party may recover indemnity costs where the other party has acted in breach of a jurisdiction clause, an arbitration clause or an anti-suit injunction. See, eg A/S D/S Svendborg v Akar [2003] EWHC 797 (Comm), where Flaux J made a declaration that the claimants were entitled to an indemnity in respect of any future costs and expenses incurred as a consequence of the proceedings commenced by the defendants in Guinea and Hong Kong in breach of an English jurisdiction clause. See also Kyrgyz Mobil Tel Ltd v Fellowes International Holdings Ltd (No 2) [2005] EWHC 1314 (Comm) (London arbitration agreement); Travelers Casualty and Surety Co of Canada v Sun Life Assurance Co of Canada (UK) Ltd [2006] EWHC 2885 (Comm); A v B (No 2) [2007] EWHC 54 (Comm), [2007] 1 Lloyd’s Rep 358 (Swiss arbitration subject to the jurisdiction of the Swiss courts); National Westminster Bank v Rabobank Nederland RV [2007] EWHC 1742 (Comm) (breach of jurisdiction clause and anti-suit clause). 128 Compania Sud Americana (n 3). 129 Akai (n 115); applied in OT Africa Line Ltd v Magic Sportswear Corp [2005] EWCA Civ 710, [2005] 2 Lloyd’s Rep 170 noted in Y Baatz, ‘An English Jurisdiction Clause Does Battle with Canadian Legislation Similar to the Hamburg Rules’ [2006] Lloyd’s Maritime and Commercial Law Quarterly 143. 130 The Alexandros T [2011] EWHC 3381 (Comm), [2012] 1 Lloyd’s Rep 162 [19]–[23] concerned insurance policies which expressly provided for exclusive English jurisdiction and settlement agreements, one of which provided for English law and exclusive English jurisdiction but the other of which provided for English law and English jurisdiction. Burton J held that the latter settlement agreement should be construed as providing for exclusive English jurisdiction. This part of his judgment was not subject to appeal. For detailed discussion in a bill of lading case, see Compania Sud Americana (n 3). 131 See section IV. 132 See section V. 133 The Alexandros T (Comm) (n 89); The Alexandros T (CA) (n 127).

106 Yvonne Baatz As seen above,134 where a third party brings a direct action against a liability insurer, the English courts have held that the claimant is bound by a London arbitration clause in the insurance contract. There is no reason to think that their approach would be any different in relation to an English court jurisdiction clause in the insurance contract if the EU Regulations did not apply and the English court was not bound by The Sea Endeavour I.135

VII. Conclusion The benefits of an exclusive jurisdiction clause are certainty for the parties and consolidation of proceedings, so that there are not multiple proceedings in numerous jurisdictions. It is sensible that all proceedings should be consolidated in a single court, where they arise on the same contractual terms, rather than being scattered throughout the different courts of one state or many states, with potentially complex evidence having to be called in numerous courts, with the consequent risk of conflicting decisions and additional legal costs. This is especially so for a liability insurer which may otherwise face litigation anywhere that the insured shipowner trades to. Jurisdiction clauses in marine insurance contracts may raise complex issues, such as whether the clause is exclusive, what disputes it covers and which parties it binds. Despite the need for contracting parties to have certainty as to whether their jurisdiction clause will be upheld and any resulting judgment be enforced, there is no global solution to these issues. The Choice of Court Convention needs to attract many more ratifications. Furthermore, although it applies to marine insurance contracts, there are no specific provisions about subrogation and direct action. There are therefore gaps where national law will apply, which results in a lack of harmony. The EU system provides for certainty where the parties have chosen the court of an EU Member State in a marine insurance contract. However, the Court of Justice of the European Union has held that a jurisdiction agreement is not binding in a direct action against a liability insurer. Under the English common law, an English jurisdiction clause in a marine insurance contract will be upheld unless there are exceptional circumstances. Furthermore, in contrast to the position under the EU Regulations, it is likely that a party bringing a direct action against a liability insurer would also be bound by such a clause if the statutory right to bring a direct action is based on the insurance contract. Should the UK leave the EU without an agreement on jurisdiction, the Choice of Court Convention will apply, and this will give protection to jurisdiction agreements in marine insurance contracts as between the UK and the contracting states. The UK would not be bound by the decision in The Sea Endeavour I and would be likely to hold that an English court jurisdiction clause is binding in a direct action. The contractual bargain as to jurisdiction has a very important role to play in promoting certainty and consolidating proceedings in relation to one carriage in one court. There is not a level playing field as to the court system in all states and the expertise of the judiciary in relation to the complex issues that may arise under insurance contracts. Commercial parties should be allowed to choose where their disputes in these matters will be resolved and that choice should extend to a direct action against a liability insurer for cargo damage.

134 See 135 See

n 23. n 95.

6 Choice of Law in Determining the Ownership of Ships MARTIN DAVIES

I. Introduction Rule 133 of Dicey, Morris and Collins on the Conflict of Laws states the basic proposition that the validity of a transfer of tangible movable property and its effect on the proprietary rights of the parties thereto is governed by the lex situs,1 and Rule 129(3) states that ‘a chattel is situated where it is at any given time’.2 Although the latter rule is expressed to be subject to Exception 1, which is that ‘A merchant ship may at some times be deemed to be situate at her port of registry’,3 the commentary to that exception states that the reasons for ascribing this artificial situs are ‘not compelling’ when the ship is in ‘territorial or national waters’, where the artificial situs of the port of registry should be displaced by the actual situs.4 In this chapter, it will be argued that the law of the ship’s actual situs should always be used when considering questions of ownership, except in the context of ship arrest. In the context of ship arrest, it has been suggested that the possible choices are: the law of the forum (lex fori), the law of the flag as an artificial lex situs, the law of the ship’s actual situs when it is in territorial waters and the law of the domicile of the registered owner or operator.5 It will be argued here that it is appropriate to use only the lex fori when considering ownership for purposes of ship arrest. Section III provides arguments against using the law of the flag as an artificial lex situs. Section IV provides arguments for using the law of the ship’s actual situs when considering questions of title in the abstract. Section V provides arguments for using the lex fori when considering beneficial ownership for purposes of ship arrest. First, however, section II disposes briefly with the idea that the law of the underlying sale transaction can be used to determine ownership of ships.

1 L Collins (gen ed), Dicey, Morris and Collins on The Conflict of Laws, 15th edn (London, Sweet & Maxwell, 2012) vol 2, 1336. 2 ibid 1286. 3 ibid 1300. 4 ibid 1301. 5 Tisand Pty Ltd v Owners of the Ship MV Cape Moreton (ex Freya) [2005] FCAFC 68, (2005) 143 FCR 43 [147] (Ryan and Allsop JJ).

108 Martin Davies

II. The Law of the Underlying Transaction When it comes to the question of applicable law, it is generally accepted that the validity and effect of a contract for sale of goods must be distinguished from the validity and effect of the transfer of title to those goods, which may affect the rights of parties other than the parties to the sale contract.6 The only possible reason for making an exception to that rule in relation to ownership of ships would be an attempt to find a law that is not dependent on the chance location of a trading ship somewhere in the world when the ship is sold. That is not reason enough in itself, however, because in many cases it would produce the result that the governing law, although constant and predictable, might have no connection whatever with the ship’s nationality or its trading pattern. The most widely used standard form contract for the sale of second-hand ships, BIMCO’s Norwegian sale form, provides three alternatives for choice of governing law: English law, US law or the law of some selected place, with English law being the default if no affirmative choice is made.7 Similarly, its main competitor, the Singapore Maritime Foundation’s sale form SSF 2011, provides for a choice between Singapore law, English law or the law of some selected place, with Singapore law being the default if no affirmative choice is made.8 As a result, it is likely that most contracts for the sale of second-hand ships are governed by either English law or Singapore law, even though those laws may have no connection whatever with the nationality of the ship or its location at the time of sale. Consistency may be a virtue in the conflict of laws, but so is objective connection with the events about which a governing law must be chosen. In any event, practical considerations militate against use of the governing law of the sale contract, for the reasons given by Moore-Bick J of the High Court of England and Wales in Glencore International AG v Metro Trading International Inc: Practical control over movables can ultimately only be regulated and protected by the state in which they are situated and the adoption of the lex situs rule in relation to the passing of property is in part a recognition of that fact. That is just as much true in relation to the passing of property between the parties to the transaction as it is in relation to the passing of property between one or other of them and a third party.9

Because of this practical need to refer to the lex situs, the choice is then between the law of the ship’s actual situs when the transaction is completed and its artificial situs, the place where it is registered.

III. The Law of the Flag It is sometimes contended that questions concerning the ownership of a ship should be determined by reference to the law of the ship’s flag, on the basis that the country of the ship’s registry

6 M Davies, A Bell, P Brereton and M Douglas, Nygh’s Conflict of Laws in Australia, 10th edn (Sydney, LexisNexis Butterworths, 2019) paras 33.5–33.13; J Carruthers, The Transfer of Property in the Conflict of Laws (Oxford, Oxford University Press, 2005) ch 4. 7 BIMCO, Saleform 2012: Norwegian Shipbroker’s Association’s Memorandum of Agreement for Sale and Purchase of Ships, cl 16 (Law and arbitration); see www.bimco.org/-/media/BIMCO/Contracts-and-Clauses/Contracts/Samplecopies/Sample-copy-SALEFORM-2012.ashx. 8 Singapore Maritime Foundation, Singapore Ship Sale Form SSF 2011, cl 15 (Arbitration and governing law); see www.singforms.com/wp-content/uploads/2018/12/32-the-ssf-singapore-ship-sale-form-english.pdf. 9 Glencore International AG v Metro Trading International Inc (No 2) [2001] 1 Lloyd’s Rep 284 (Com Ct) [31].

Choice of Law in Determining the Ownership of Ships 109 should be regarded as its constant situs for the purposes of the usual rule that the lex causae governing questions of ownership of moveable property is the lex situs. This preference for the law of the flag as the artificial lex situs is supported by the fact that, unlike most movable chattels, a ship has a recognisable nationality.10 For example, in Tisand Pty Ltd v Owners of the Ship MV Cape Moreton (ex Freya),11 the Full Court of the Federal Court of Australia held that the existence, nature and extent of any ownership rights in a ship should be determined at least in part12 by the lex causae, the law indicated by the Australian choice-of-law rules, rather than by Australian law as the lex fori,13 and that there were ‘powerful reasons for giving effect to the law of the country of register as the lex situs in relation to questions of title, property and assignment’,14 because ‘[t]he chance location of a working merchant ship in a port within its range of sailing or on the high seas appears to introduce an element of arbitrariness to the legal analysis’.15 Although there is some superficial attractiveness to deciding ownership questions by reference to the law of the flag, which remains constant as the ship travels from country to country, there are powerful objections to that approach, both in legal principle and in practical reality. In most countries, registration of a ship reflects legal title; it does not confer it. For example, in Mentink v Registrar of Australian Register of Ships,16 a UK citizen alleged that he had bought an Australian-registered yacht in Indonesia, removed it from the Australian register and then sold it to an American buyer, who registered it in the USA. The case did not turn on the question of who owned the yacht as a matter of law, but rather on whether anyone other than the original registered owner, an Australian, was entitled to have the yacht removed from the Australian register of ships. Nevertheless, the facts of the case highlight that ownership and registration are two quite different things, as the Full Court of the Federal Court of Australia observed: ‘Ownership precedes and, indeed, is a requirement for, registration’.17 The registered Australian owner denied that he had sold the yacht to the UK citizen who purported to be the buyer, which raised the question of whether the UK citizen could validly pass title in the yacht to the American buyer, or whether the principle nemo dat quod non habet meant that the yacht still belonged, as a matter of law, to the original Australian owner. To which legal system should one look to decide whether the American bona fide purchaser in good faith could acquire good title from the UK citizen? Should it be Indonesian law, Indonesia being the place where the yacht was when it was sold by the UK citizen to the American buyer? Should it be American law, the law of the place where the yacht ended up, and which was sufficiently satisfied that the American buyer was the owner that it registered him as owner in the US register of ships? Whatever the answer to that question, there seems to be little reason to think that the answer should be found in Australian law, simply because that was where the yacht was registered. None of the transactions that purportedly transferred title to the yacht occurred in Australia, and the yacht was not in Australia at the time that those transactions occurred. The court’s decision was that only the Australian registered owner

10 Dornoch Ltd v Westminster International BV (The WD Fairway) [2009] EWHC 889 (Admlty), [2009] 2 Lloyd’s Rep 191 [92] (Tomlinson J). 11 Tisand (n 5) [140] (Ryan and Allsop JJ). 12 The full analysis suggested by the Tisand court is considered in section V below. 13 Tisand (n 5) [146] (Ryan and Allsop JJ). See also Korea Shipping Corp v Lord Energy SA [2018] FCAFC 201, (2018) 363 ALR 312 [13] (Allsop CJ, Besanko and McKerracher JJ). 14 Tisand (n 5) [146] (Ryan and Allsop JJ). 15 ibid. 16 Mentink v Registrar of Australian Register of Ships [2015] FCAFC 150, (2015) 234 FCR 458. 17 ibid [53] (Rares, Logan and McKerracher JJ).

110 Martin Davies was entitled to remove the ship from the Australian register and that the registered owner could not demand rectification of the register without giving the UK citizen and the American buyer the right to be heard; but none of that said anything about the question of whether the Australian registered owner was still the true owner of the yacht as a matter of law, which was a completely separate question from that of registration. In many countries, the distinction between the fact of registration and the law governing ownership is starkly obvious. Despite the requirement in the UN Convention on the Law of the Sea that there be a ‘genuine link’ between the owner of a ship and the country of its r egistration,18 many countries operate ‘open registers’ that do not require shipowners to be citizens of the country of registry, the only ‘genuine link’ with that country being a desire to have one’s ship registered there. In ‘open register’ countries, all that is necessary to achieve registration is proof of ownership by a country’s laws, not necessarily ownership according to the law of the country of registration. The three largest ship registers in the world – Panama, the Marshall Islands and Liberia19 – are ‘open registers’. Of the 7,914 ships registered in Panama in 2018,20 only 75, or 0.94 per cent, were owned by Panamanian owners.21 Of the 3,419 ships registered in the Marshall Islands in 2018,22 only 10, or 0.03 per cent, were owned by Marshall Islands owners.23 Of the 3,321 ships registered in Liberia in 2018,24 only two, or 0.06 per cent, were owned by Liberian owners.25 In other words, over 99 per cent of the ships on the world’s three largest ‘open registers’ (14,567 out of 14,654 ships) are owned by nationals of a country other than the country of the ship’s registry. To expect a connection between the law of the flag and the law governing ownership of the ship is therefore optimistic, to put it mildly. On the other side of the same coin, it can be seen that Greek owners own the largest percentage of the world’s fleet of merchant ships by tonnage, but only 19.7 per cent of G reek-owned ships are registered in Greece.26 Japanese owners own the second-largest percentage of ships by tonnage, but only 17 per cent of Japanese-owned ships are registered in Japan.27 Singaporean owners own the fifth-largest percentage of the world’s fleet by tonnage, but only 2.2 per cent of those ships are registered in Singapore.28 In summary, as UNCTAD observes: ‘Most commercial ships are registered under a flag that differs from the country of ownership’.29 It follows that in the large majority of cases, it is inaccurate – wildly inaccurate – to assume that the law of a ship’s flag has anything to do with the ship’s ownership. For that very practical reason, reference to the law of the flag is not reliable when considering the question of ship ownership.

18 UN Convention on the Law of the Sea (UNCLOS), 10 December 1982, 1833 UNTS 3163, Art 91(1). 19 Panama is the world’s largest register by deadweight tonnage, with 17.46% of the world total in 2018; the second largest register by deadweight tonnage is the Marshall Islands, with 12.36%; Liberia is third, with 11.63%: see UNCTAD, Review of Maritime Transport 2018, 35, Table 2.8; see unctad.org/en/PublicationsLibrary/rmt2018_en.pdf. 20 UNCTADStat, ‘Merchant Fleet by Flag of Registration and by Type of Ship, Annual, 1980–2018’; see unctadstat. unctad.org/wds/TableViewer/tableView.aspx?ReportId=93. 21 UNCTADStat, ‘Merchant Fleet by Country of Beneficial Ownership, Annual, 2014–2018’; see unctadstat.unctad.org/ wds/TableViewer/tableView.aspx?ReportId=80100. 22 Merchant Fleet by Flag of Registration (n 20). 23 Merchant Fleet by Country of Beneficial Ownership (n 21). 24 Merchant Fleet by Flag of Registration (n 20). 25 Merchant Fleet by Country of Beneficial Ownership (n 21). 26 Review of Maritime Transport 2018 (n 19) 30, Table 2.3. 27 ibid. 28 ibid. 29 ibid 34.

Choice of Law in Determining the Ownership of Ships 111

IV. The Law of the Ship’s Actual Situs Dicey’s preference for the law of the actual situs when the ship is in territorial waters30 is at odds with the position taken by the Full Court of the Federal Court of Australia in Tisand, which was that the law of the flag should be applied ‘not merely when the ship is on the high seas’, but also when it is ‘in a port within its range of sailing’.31 The position preferred by Dicey was adopted by Tomlinson J of the High Court of England and Wales in The WD Fairway.32 In that case, a badly damaged Dutch-flagged dredger was towed to Thailand. The dredger’s hull insurer indemnified the assured shipowner in full for a constructive total loss and also paid the considerable costs of salvage/wreck removal. The dredger was deregistered and was later sold by the assured to an associated company in order to prevent the hull insurer from realising the open market value of the vessel, because the assured did not want its competitors to have access to the vessel for inspection. The particular question considered by Tomlinson J was whether the hull insurer had acquired some form of proprietary interest in the dredger before deregistration, such as an equitable lien or beneficial interest under a trust. That, in turn, raised the question of what law governed the acquisition of proprietary rights in the dredger: Dutch law (the law of the flag, as the artificial situs at the time the hull insurer made the relevant payments) or Thai law (the law of the actual situs at that time)?33 Tomlinson J held that Thai law governed, explicitly rejecting the proposition adopted in Tisand, namely that the artificial lex situs of the ship’s flag should apply at all times, even when the ship is in territorial waters. However, Tomlinson J made an important distinction between legal title and proprietary rights short of ownership: It seems to me likely that, so far as concerns questions of legal title, most systems of law would inevitably, in consequence of their own conflict rule, look to the law of the place of registration. Since however the register would not ordinarily be concerned with interests falling short of legal ownership, I can see no compelling reason for regarding the law determining the incidence of such rights as being the law of the place of registration as opposed to the physical situs of the vessel at the time when such rights are created.34

Similarly, when dealing with the analogous situation of proprietary rights in aircraft in Blue Sky One Ltd v Mahan Air,35 Beatson J of the High Court of England and Wales held that the validity of aircraft mortgages should be determined by the law of the place where the aircraft were when the mortgages were executed, rather than by the law of the place of the aircraft’s registration (lex registrii), saying that the argument for the application of the lex registrii ‘finds virtually no support in English cases or commentaries’.36 Strictly speaking, neither The WD Fairway nor Blue Sky One is inconsistent with Tisand, which was concerned with the law governing ownership of a ship – ‘questions of title, property and assignment’, in the Tisand court’s words37 – rather than proprietary rights short of ownership,

30 Dicey, Morris and Collins on The Conflict of Laws (n 1) 1301. 31 Tisand (n 5) [146]–[147] (Ryan and Allsop JJ). 32 The WD Fairway (n 10). 33 ibid [101]: ‘The question I have to decide is whether the lex situs is here to be regarded as Dutch law, the law of the place of registry, for the purpose of determining the incidence of proprietary interests in the vessel prior to deregistration.’ 34 ibid [103]. 35 Blue Sky One Ltd v Mahan Air [2010] EWHC 631 (Comm). 36 ibid [154]. 37 Tisand (n 5) [146] (Ryan and Allsop JJ).

112 Martin Davies such as the mortgages in Blue Sky One or the purported equitable rights in The WD Fairway. Indeed, the passage quoted above from Tomlinson J’s judgment in The WD Fairway lends some support to the proposition that the question of ownership of a ship must ‘inevitably’ be decided by the law of the place of registration. That might hold true if registration conferred title in the country in question,38 but it is not necessarily true in the majority of countries, where registration merely records the fact that the registered owner has title by the law of a country, not necessarily that country. Indeed, as the statistics in section III show, most commercial ships are registered under a flag that differs from the country of ownership. In those countries, registration merely records the fact of ownership established under the laws of another country. For example, all that is required for provisional ship registration in Panama is some form of document, such as a bill of sale, showing transfer of title in the ship to the applicant, who need not be Panamanian; the bill of sale or other document must be notarised by a Panamanian consul, but it need not record the fact of ownership under Panamanian law.39 In contrast, a ship can only be registered on the US register by a US citizen or a US corporation, association, trust, joint venture or partnership in which each member, partner, chief executive officer or director is a US citizen,40 and only upon proof of a complete chain of title from the time of building to the present owner.41 Even in the USA, though, there is no requirement that the underlying transaction by which title is transferred to the registered owner must be governed by US law. Provided that the applicant satisfies the citizenship requirements, he/she/it can register a ship that has previously been flagged in a foreign country, to which title was transferred by the laws of that foreign country.42 Thus, even in a country with such stringent ownership and citizenship requirements as the USA, the law of the flag is not necessarily identical to the law governing ownership of the registered ship. In ‘open register’ countries, the law of the flag is almost certainly different from the law by which the registered owner acquired title to the ship. Despite this absence of connection between the law of registration and the law of ownership, a forum court tasked with determining the ownership of a ship could nevertheless choose to regard the fact of registration on a particular country’s register as conclusive evidence that the registered owner is the true legal owner. Unfortunately, that assumption would be questionable, at best. Some countries, such as the USA, specifically provide that registration does not constitute conclusive evidence of ownership in a proceeding in which ownership is at issue.43 Others, such as Australia and New Zealand, provide that the certificate of registration constitutes only prima facie evidence of ownership.44 Furthermore, although a ship is regarded as stateless if it claims 38 The UK legislation comes close to this by providing that the registered owner of a ship has the power absolutely to dispose of the ship (Merchant Shipping Act 1995 (UK), sch 1, reg 1(1)), but it also provides that contractual or equitable rights can subsist in relation to a registered ship and may be enforced in the same manner as in respect of any other personal property (Merchant Shipping Act 1995 (UK), sch 1, reg 1(2)). 39 Autoridad Marítima de Panamá (Panama Maritime Authority), ‘Régistro Publico – Titulos de Propriedad de Naves (Public Registry – Ship Ownership Titles)’; see amp.gob.pa/servicios/registro-publico-de-naves/titulos-depropiedad-de-naves/. 40 46 USC § 12103. 41 46 CFR § 67.53(c), (d). 42 Although flagged in the USA after registration, such a ship could nevertheless not engage in coastwise trade from one US port to another, which is the main (indeed, almost the only) reason why US owners flag their ships on the US register. To engage in coastwise trade, a ship must have been built in the USA, and be owned and crewed by US citizens: 46 USC §§ 12112, 55102. A vessel procured outside the USA that satisfies the ownership requirements may be flagged on the US register, but may only engage in foreign trade or trade with Guam, American Samoa, Wake, Midway or Kingman Reef: 46 USC § 12115(b). 43 46 USC § 12134(3). 44 Shipping Registration Act 1981 (Cth), s 77(1); Shipping Registration Regulations 2019 (Cth), reg 37; Ship Registration Act 1992 (NZ), s 80(1). The legislation in Singapore provides no more than that the certificate is admissible in evidence: Merchant Shipping Act c 179 (Sing), s 186(1), (2).

Choice of Law in Determining the Ownership of Ships 113 more than one nationality, switching back and forth between flags, ‘using them according to convenience’,45 it is possible for a ship to have more than one nationality, by being registered, at least provisionally, on more than one national register. That was the case, for example, in Tisand, where, at the material time when the ship was arrested, it had been provisionally registered in Hong Kong, but remained registered on the Liberian register,46 and it was also the case in Mentink, where the yacht was registered in both Australia and the USA, with different registered owners in each country. The author has worked on a case where a ship was registered in Country A, sold in Country B and provisionally registered anew in Countries C and D. The original owner (the government of Country A) contested the validity of the sale in Country B, and the ship was not removed from the register in Country A. Thus, the ship was registered in three countries (A, C and D), but it remained in Country B. However, the confused question of the ship’s nationality was essentially different from the question of who was the true owner. Only the law of Country B could adequately answer that question, for the reasons stated above in section II: ‘practical considerations of control over movables … can ultimately only be regulated and protected by the state in which they are situated’. For those reasons, only Country B could effectively determine the validity and effect of the contested sale that took place in that country. There are other good reasons for focusing on the law of the actual situs in relation to the transfer of ownership in a ship, as well as in relation to the creation of rights short of ownership. Under the UK’s sale of goods legislation,47 and in jurisdictions with sale of goods legislation modelled on the UK’s, property in the ship as ‘specific or ascertained goods’ is transferred at such time as the parties to the contract of sale intend it to be transferred, having regard to the terms of the contract, the conduct of the parties and the circumstances of the case.48 In practice, if a second-hand ship is sold, property usually does not pass until delivery is made at an agreed place. Standard form ship sale agreements generally require the seller to deliver the ship to the buyer at an agreed port or place,49 and they make it clear that the sale is not complete, and title does not pass, until delivery is made. The sale contract may be made while the ship is on the high seas, and the documents relating to the ship, including the bill of sale, may be handed over to the buyer in return for the purchase price at a closing transaction at some other designated place,50 but the sale is complete only upon physical delivery of the ship, which may (and often does) occur at a different place from that of the closing transaction.51 If the ship is lost before delivery at that agreed place, the agreement becomes ‘null and void’;52 conversely, property in all spare parts, whether on board or ashore, is expressed to pass upon delivery.53 Although the ship’s flag, and thus its nationality, may change when the ship is sold, that has no necessary connection to the actual situs where the parties have agreed that delivery should be effected. Although the actual situs may be

45 UNCLOS (n 18) Art 92(2). 46 Tisand (n 5) [43]–[46] (Ryan and Allsop JJ). 47 The sale of goods legislation applies to the sale of ships: Behnke v Bede Shipping Co Ltd [1927] 1 KB 649 (KB); Lloyd del Pacifico v Board of Trade (No 1) (1929) 35 Ll L Rep 217 (CA). 48 Sale of Goods Act 1979 (UK), s 17; Sale of Goods Act c 393 (Sing), s 17; Sale of Goods Act 1954 (ACT), s 22; Sale of Goods Act 1972 (NT), s 22; Sale of Goods Act 1923 (NSW), s 22; Sale of Goods Act 1896 (Qld), s 20; Sale of Goods Act 1895 (SA), s 17; Sale of Goods Act 1896 (Tas), s 22; Goods Act 1958 (Vic), s 22; Sale of Goods Act 1895 (WA), s 17; Contract and Commercial Law Act 2017 (NZ), s 144; Sale of Goods Act (Ont) RSO 1990, c S 1, s 18; Sale of Goods Act (Novia Scotia) RS, c 408, s 20; Sale of Goods Act (BC) RSBC 1996, c 410, s 22. 49 Saleform 2012 (n 7) cl 5 (Time and place of delivery and notices); SSF 2011 (n 8) Box 10 (Delivery place) and cl 4 (Condition on delivery). 50 Saleform 2012 (n 7) cl 8; SSF 2011 (n 8) Box 8(iii), cl 8. 51 I Goldrein, Ship Sale and Purchase, 6th edn (London, Informa Law at Routledge, 2012) para 5.4.2. 52 Saleform 2012 (n 7) cl 5(e); SSF 2011 (n 8) cl 5(c). 53 Saleform 2012 (n 7) cl 7; SSF 2011 (n 8) cl 7.

114 Martin Davies anywhere in the world, it is seldom purely random or accidental. There is good reason for choosing the governing law in relation to ownership to be the law of the place selected by the parties as the place where the sale should be consummated.

V. Ship Arrest – Lex Fori However compelling may be the case for applying the law of the actual situs when considering whether and when title to a ship has been transferred, the position is – or should be – very different when it comes to considering who owns a ship for the purposes of determining whether that ship can be arrested. In the context of ship arrest, the question is not who is the owner in an abstract or absolute sense, but rather whether the ‘relevant person’ is the ‘beneficial owner’ for purposes of the statute of the forum giving the claimant the right to proceed in rem against the ship. On the face of it, that is a purely jurisdictional question that ought to be resolved by applying the lex fori (the law of the forum). In The Andres Bonifacio,54 three actions were brought in rem in the High Court of Singapore against the Philippines-registered ship Andres Bonifacio, claiming damages for breach of a series of charter parties.55 The Andres Bonifacio was arrested, then released upon provision of security in the form of bail bonds. The registered owner (PNOC), a corporation owned by the government of the Philippines, moved for the in rem writs to be set aside and the security to be released on the ground that it was not the beneficial owner of all the shares in the Andres Bonifacio when the writs were issued, because an associated corporation (POCI) operated the ship, incurred all its expenses, assumed the liabilities for its acquisition and had effectively financed the purchase of the ship. The Singapore Court of Appeal affirmed the trial judge’s refusal to consider the question of beneficial ownership by reference to the law of the Philippines. Delivering the judgment of the court, Lai Kew Chai J said: the learned judge had in our view correctly treated the question of jurisdiction as one governed by Singapore law as the lex fori … In our view the determination by the court of its own jurisdiction in rem depended on Singapore law as the lex fori … [I]n the present case, whether there was beneficial ownership and whether there was a link to a foreign jurisdiction would depend on the lex fori, Singapore law. This was only right since the appellants had come to arrest a ship here and it was surely right that the court would not allow it to do so unless it was beneficially owned as understood under Singapore law. It is not inconceivable that some jurisdictions might not have a concept of beneficial ownership similar to the Singapore concept.56

The Singapore Court of Appeal later confirmed its adherence to the lex fori in The Jarguh Sawit,57 where M Karthigesu JA said (delivering the judgment of the court): ‘Clearly, the scope of judicial power must be determined by the law of the state which confers the power, ie the lex fori, and

54 The Andres Bonifacio [1993] SGCA 70, [1993] 3 SLR(R) 71. 55 The claimants (plaintiffs in Singaporean terminology) were owners of two other ships, Feoso Ambassador and Feoso Ambassador 2. They sought damages for breach of charterparty contracts by the owner of Andres Bonifacio as charterer of Feoso Ambassador and Feoso Ambassador 2. The ‘relevant person’, the owner of Andres Bonifacio, was charterer of Feoso Ambassador and Feoso Ambassador 2 when the claim arose and the owner of Andres Bonifacio when the action was brought, thereby permitting surrogate ship arrest of Andres Bonifacio by the literal terms of High Court (Admiralty Jurisdiction) Act c 123 (Sing), s 4(4)(ii). The Singapore Court of Appeal approved surrogate ship arrest in such circumstances in The Permina 108 [1976] SGCA 12, [1974–76] SLR(R) 850. 56 The Andres Bonifacio (n 54) [35]–[36]. 57 The Jarguh Sawit [1997] SGCA 59, [1997] 3 SLR(R) 829.

Choice of Law in Determining the Ownership of Ships 115 cannot be determined by reference to the laws of another state’.58 There is much to be said for that approach as a matter of principle, but courts in England, New Zealand and Australia have all found a role for foreign law in considering the question of beneficial ownership of an arrested ship. In The Nazym Khikmet,59 the Court of Appeal of England and Wales adopted a modified lex fori approach,60 considering whether the defendant was ‘under the law to which it is subject what in our law would be regarded as the beneficial owner as respects all shares in the vessel’.61 It is not clear whether the court’s reference to ‘the law to which [the defendant] is subject’ was intended to be a reference to the law of the defendant’s domicile, a possibility alluded to by the court in Tisand,62 or the law of the flag of the ship. On the facts of the case, that choice made no difference, as both pointed to Ukrainian law, but subsequent commentators, both academic63 and judicial,64 have interpreted the Nazym Khikmet court’s approach as requiring consideration of the law of the flag, as the artificial situs of the ship, rather than the law of the defendant’s domicile. If that interpretation is correct, the relevant question according to the Nazym Khikmet court is whether the law of the flag confers rights that would be regarded as equivalent to beneficial ownership under the lex fori. A similar approach was taken (or apparently taken) by the New Zealand Court of Appeal in Vostok Shipping Co Ltd v Confederation Ltd,65 where the majority (Richardson P and Blanchard J) expressly accepted both The Andres Bonifacio and The Nazym Khikmet, apparently in the alternative,66 despite the fact that those two cases took different approaches to the choice of law question.67 Although the majority stated that the right to invoke the court’s jurisdiction is ‘a matter of procedure and as such governed by the lex fori’,68 they went on to say that New Zealand law ‘necessarily has to look to the Russian register as a means of beginning the process of determining the ownership of the ship’.69 A hybrid position of this kind now seems to be settled in Australia, although it is more clearly articulated in Australia than it was in The Nazym Khikmet and Vostok Shipping. The Full Court of the Federal Court of Australia has stated that the proper approach is that: the law of Australia … govern[s] the characterisation of the rights of the parties as they are derived from the relevant foreign transaction. The existence, nature and extent of such rights created by the relevant transaction [are] governed by the law that Australian rules of private international law regarded as relevant.70

The Full Court of the Federal Court of Australia recently applied an ostensibly similar analysis when considering recognition of foreign maritime liens, holding that the court’s first step should 58 ibid [31]. 59 The Nazym Khikmet [1996] 2 Lloyd’s Rep 362 (CA). 60 P Myburgh, ‘Arresting the Right Ship: Procedural Theory, the In Personam Link and Conflict of Laws’ in M Davies (ed), Jurisdiction and Forum Selection in International Maritime Law: Essays in Honor of Robert Force (The Hague, Kluwer Law International, 2004) 298. 61 The Nazym Khikmet (n 59) 371 (Sir Thomas Bingham MR). 62 Tisand (n 5) [147] (Ryan and Allsop JJ). 63 Myburgh (n 60) 298–99. 64 Tisand (n 5) [137] (Ryan and Allsop JJ). 65 Vostok Shipping Co Ltd v Confederation Ltd [1999] NZCA 220, [2000] 1 NZLR 37. 66 ibid [26] (Richardson P and Blanchard J). 67 Myburgh (n 60) 300–01. 68 Vostok Shipping (n 65) [26] (Richardson P and Blanchard J). 69 ibid. 70 Korea Shipping (n 13) [13] (Allsop CJ, Besanko and McKerracher JJ), referring to Tisand (n 5) [140] (Ryan and Allsop JJ).

116 Martin Davies be to apply the lex causae to identify the relevant foreign law right, with the second to be to characterise the identified foreign right by reference to the lex fori, Australian law, to determine whether it is, or is sufficiently analogous to, a maritime lien recognised by Australian law.71 However, despite the apparent similarity between the two situations,72 there is a fundamental difference between recognition of foreign maritime liens and consideration of beneficial ownership for purposes of ship arrest. The former involves consideration of a legal right conferred by a foreign lex causae to see whether it should be enforced by the forum court; the latter involves consideration of facts occurring in a foreign country to determine whether they satisfy the jurisdictional requirements of a statute of the forum. The latter question does not obviously require consideration of foreign law at all, much less the law of the flag, for the reasons already given. In jurisdictions with legislation similar to that in the UK, two very different questions arise when the court’s in rem jurisdiction is invoked. First, it is necessary to identify what the legislation calls ‘the relevant person’, the person who would be liable in personam. In relation to this question, it suffices that the plaintiff merely alleges a claim of a kind that would fall within the statutory list of admiralty claims if all the facts were as alleged by the plaintiff. That is because the relevant legislation in UK-based systems defines the ‘relevant person’ as the person who would be liable in personam, not the person who is liable in personam.73 Thus, the plaintiff need not prove that it has a prima facie case for its alleged claim, but merely that it has alleged a claim of the right kind.74 In contrast, the legislation states that an action in rem is permitted only if the ‘relevant person’ was the owner, charterer, or in possession or control of the ship when the cause of action arose and is the beneficial owner (or demise charterer) when the action is commenced.75 That requires the claimant to prove the ‘jurisdictional fact’ of the requisite ownership connection on the balance of probabilities.76 The inquiry raised by the definition of ‘relevant person’ is a conditional one, based on hypothetical liability, whereas the inquiry raised by the question of the ownership connection is an unconditional one, concerned with facts that need to be established.77 The second, unconditional, question is one of jurisdictional fact, not foreign law. In The Shin Kobe Maru, the High Court of Australia pointed out that: Where jurisdiction depends on particular facts or a particular state of affairs, a challenge to jurisdiction can only be resisted by establishing the facts on which it depends. And, of course, they must be established on the balance of probabilities in the light of all the evidence advanced in the proceedings held to determine whether there is jurisdiction.78

71 Ship ‘Sam Hawk’ v Reiter Petroleum Inc [2016] FCAFC 26, (2016) 246 FCR 337 [103]–[104] (Allsop CJ and Edelman J). 72 In the sentence immediately after the passage in Korea Shipping (n 13) quoted above, the Full Court said: ‘See also the approach to characterisation of maritime liens in Ship “Sam Hawk” v Reiter Petroleum Inc.’ 73 Senior Courts Act 1981 (UK), s 21(4)(b); High Court (Admiralty Jurisdiction) Act c 123 (Sing), s 4(4)(b); High Court Ordinance (HK), s 12B(4)(b); Admiralty Act 1988 (Cth), s 3 (definition of ‘relevant person’); Admiralty Act 1973 (NZ), s 5(2)(b). 74 Schwarz & Co (Grain) Ltd v St Elefterio ex Arion (Owners) (The St Elefterio) [1957] P 179 (PDAD); The Moschanthy [1971] 1 Lloyd’s Rep 37 (QB (Admlty)); Owners of the Motor Vessel ‘Iran Amanat’ v KMP Coastal Oil Pte Ltd (The Iran Amanat) [1999] HCA 11, (1999) 196 CLR 130; Kingstar Shipping Ltd v Owners of the Ship ‘Rolita’ (The Rolita) [1989] 1 HKLR 394; Equatorial Marine Fuel Management Services Pte Ltd v The Bunga Melati 5 [2012] SGCA 46, [2012] 4 SLR 546. 75 Senior Courts Act 1981 (UK), s 21(4)(b) and (i); High Court (Admiralty Jurisdiction) Act c 123 (Sing), s 4(4)(b) and (i); High Court Ordinance (HK), s 12B(4)(b) and (i); Admiralty Act 1988 (Cth), s 17(a), (b); Admiralty Act 1973 (NZ), s 5(2)(b) and (i). 76 Owners of Shin Kobe Maru v Empire Shipping Inc [1994] HCA 54, (1994) 181 CLR 404, 426 (Mason CJ, Brennan, Deane, Dawson, Toohey, Gaudron and McHugh JJ). 77 The Iran Amanat (n 74) [18] (Gleeson CJ, McHugh, Gummow, Kirby and Hayne JJ). See also The Vinalines Pioneer [2015] SGHCR 1. 78 Shin Kobe Maru (n 76) 426 (Mason CJ, Brennan, Deane, Dawson, Toohey, Gaudron and McHugh JJ).

Choice of Law in Determining the Ownership of Ships 117 There is ample authority for the proposition that the beneficial owner of a ship is the person who has the right to sell, dispose of or alienate all the shares in that ship,79 also described as the person who has the power to have and dispose of dominion, possession and enjoyment of the ship.80 The facts relevant to those questions may have occurred in another country, but it does not follow that they should be interpreted according to the law of that country. For example, in Vostok Shipping, the key question was whether the registered owner of the ship Kapitan Lomaev was still the beneficial owner at the time the ship was arrested in New Zealand, because the ship had been sold to a new buyer and provisionally registered on the Belizean register while it was on the high seas on the voyage to New Zealand. At the time of the arrest, the ship’s Russian registration had not been cancelled. Under New Zealand law, the buyer would have been regarded as the equitable or beneficial owner of the ship, but the expert evidence suggested that Russian law does not recognise the concept of beneficial ownership, and would have regarded the ship’s registration as conclusive of legal ownership.81 A thoroughgoing consideration of Russian law as the law of the flag would therefore have concluded that the new buyer was not the beneficial owner, but a consideration of the facts occurring in Russia, characterised according to New Zealand law as lex fori, would have concluded that the new buyer was the beneficial owner for the purposes of the New Zealand statute conferring in rem jurisdiction on the forum court.82 Although the majority of the Vostok Shipping court said, rather confusingly, that it made no difference whether one applied Russian or New Zealand law on the facts of the case,83 the result was consistent with application of the lex fori, New Zealand law, to the jurisdictional facts of the transactions that had taken place in Russia: the arrest of Kapitan Lomaev was set aside on the ground that the registered owner was no longer the beneficial owner when the arrest was made. The facts of Vostok Shipping add force to the observation of the Singapore Court of Appeal in The Andres Bonifacio that it makes little sense to ask whether the ‘relevant person’ is the beneficial owner by the law of a country that does not recognise the concept of beneficial ownership.84 The very phrase used in the legislation of the UK-based systems is one peculiar to their law as the lex fori; it only makes sense to interpret it according to the concepts employed by the lex fori.85 In relation to the first, conditional, question, namely whether the claim is one that falls within the court’s admiralty jurisdiction at all, there is never any doubt that it falls to be considered according to the lex fori as a simple matter of interpretation of the relevant jurisdictional statute. For example, in The Andres Bonifacio, the parties accepted without argument that the plaintiffs’ claims for breach of charterparty contracts were of the kind that fell within the admiralty jurisdiction.86 Nevertheless, if there is any doubt about whether the first requirement for the availability of an action in rem exists – ie whether the claim falls within the statutory list – courts 79 The Pangkalan Susu/Permina 3001 [1977] SGCA 5, [1977–78] SLR(R) 105 [9]; Pacific Navigation Co Pte Ltd v Owners of and All Other Persons Interested in the Ship or Vessel ‘Ohm Mariana ex Peony’ (The Ohm Mariana, ex Peony) [1993] SGCA 43, [1993] 2 SLR(R) 698 [35]; Consorcio MGT v Owners and/or Demise Charterer of the Vessel ‘Min Rui’ (The Min Rui) [2016] SGHC 183 [52] (Belinda Ang Saw Ean J); Kent v Vessel SS ‘Maria Luisa’ (No 2) [2003] FCAFC 93, (2003) 130 FCR 12 [61] (Tamberlin and Hely JJ); Ship ‘Gem of Safaga’ v Euroceanica (UK) Ltd [2010] FCAFC 14, (2010) 182 FCR 27 [15] (Besanko J); Shagang Shipping Co Ltd v Ship ‘Bulk Peace’ [2014] FCAFC 48, (2014) 314 ALR 230 [20] (Allsop CJ). 80 Tisand (n 5) [119] (Ryan and Allsop JJ). 81 Vostok Shipping (n 65) [24] (Richardson P and Blanchard J). 82 Myburgh (n 60) 304. 83 Vostok Shipping (n 65) [26] (Richardson P and Blanchard J). 84 The Andres Bonifacio (n 54) [35]–[36]. 85 Ironically, given the number of cases that have considered the issue in Australia, the phrase ‘beneficial owner’ is not used in the Australian legislation, which refers only to ‘the owner of the ship’: see Admiralty Act 1988 (Cth), ss 17(b), 19(b). Nevertheless, those words have been interpreted to refer to a beneficial owner: see the Australian cases referred to in n 79 above. 86 The Andres Bonifacio (n 54) [4].

118 Martin Davies apply the lex fori without hesitation and without even acknowledging that they are doing so. For example, in Heilbrunn v Lightwood plc,87 the Federal Court of Australia had to consider whether a claim for damage done to a vintage car while loading it into a shipping container in inland England fell within the admiralty jurisdiction of the Australian court, on the ground that it was a claim ‘arising out of an agreement that relates to the carriage of goods or persons by a ship or to the use or hire of a ship, whether by charterparty or otherwise’, for purposes of the Admiralty Act 1988 (Cth), section 4(3)(f). Allsop J (as he then was), one of the two judges88 who declared the law of the flag to be relevant to the question of beneficial ownership in Tisand, held that the claim did fall within the admiralty jurisdiction. Allsop J’s reasons gave a typically careful explanation of why the answer should be determined by Australian law as a jurisdictional question for the lex fori: The jurisdiction of the court is conferred by laws of the Commonwealth Parliament. The court only has jurisdiction here if the plaintiff ’s claim is a general maritime claim under the Act, such that the claim for damages in bailment and tort can be seen as a claim in federal jurisdiction, in the exercise of the court’s admiralty and maritime jurisdiction … The fact that the events took place in the United Kingdom does not affect the question of jurisdiction. The Act confers jurisdiction irrespective of the place where events took place: s 5 of the Act, ‘all maritime claims, wherever arising’. The relevant question to address is whether the assertions of right or interest recognised by law and carrying an entitlement to relief made by the plaintiff in the application and statement of claim (being the claim of the plaintiff) have (and has) the legal character of that which is set out in s 4(3)(f) … This task is to be undertaken with the meaning and content of s 4(3)(f) understood.89

In other words, the ‘legal character’ of the plaintiff ’s claim as a ‘general maritime claim’ must necessarily be understood by reference to the ‘meaning and content’ of the words as used in the context of the Australian statute. As Allsop J himself pointed out, it was irrelevant that the plaintiff ’s claim alleged a tort committed in England, which an Australian court would consider to be governed by English law.90 The question was simply whether the facts alleged by the plaintiff fell within the statutory phrase used in section 4(3)(f) of the Australian jurisdictional legislation. Although a very similar phrase (‘arising out of any agreement relating to the carriage of goods in a ship or to the use or hire of a ship’) also appears in the UK legislation,91 it would have made no sense at all to ask whether the plaintiff ’s claim would be considered as falling within the scope of that phrase under English law. The relevant facts occurred in England, but the jurisdiction of the court was conferred by Australian law and its scope was obviously to be determined by reference to Australian law. Although the burden of proof is different in relation to the second, unconditional, question about ownership in relation to ship arrest, the conflict of laws analysis should be exactly the same, an analysis of facts occurring elsewhere according to the language of the forum’s jurisdictional statute. A comparison with the equivalent questions in US law is instructive in highlighting when a choice of law analysis is relevant and when it is not. Unlike most countries, which have a statutory

87 Heilbrunn v Lightwood plc [2007] FCA 1518, (2007) 164 FCR 1. 88 In Tisand (n 5), the Full Court comprised only two members, Ryan and Allsop JJ, because the third member of the court, Cooper J, died after the hearing had commenced but before judgment was given. The parties consented to the matter being dealt with by Ryan and Allsop JJ as a Full Court pursuant to the Federal Court of Australia Act 1976 (Cth), s 14(3). 89 Heilbrunn (n 87) [6], [21]–[23]. 90 The Australian choice of law test for international torts applies the law of the place of the wrong (lex loci delicti) without any flexible exception: Regie Nationale des Usines Renault SA v Zhang [2002] HCA 10, (2002) 210 CLR 491. 91 Senior Courts Act 1981 (UK) s 20(2)(h).

Choice of Law in Determining the Ownership of Ships 119 list of the kinds of claims that fall within the admiralty jurisdiction,92 based more or less closely on the list in the Arrest Convention 1952,93 there is no such list in the USA. The US jurisdictional statute confines itself to the Delphic statement that federal courts have subject-matter jurisdiction over ‘[a]ny civil case of admiralty or maritime jurisdiction’,94 leaving it to the courts to decide which claims have access to the special admiralty procedures of arrest and attachment. As a result, admiralty cases in the USA are often called on explicitly to consider the prior question of whether the plaintiff ’s claim falls within the admiralty jurisdiction at all, which is the equivalent to what the High Court of Australia in The Iran Amanat called the first, conditional, question.95 In the USA, the second question of who owns an arrested ship is irrelevant because of the American doctrine of personification, which treats the ship itself as a legal person liable for claims against it in contract or tort, with the result that the ship may be held liable in rem in circumstances where its owner is not liable in personam.96 However, ownership of assets is centrally important in relation to the related procedure of maritime attachment, which is done pursuant to an in personam action in the admiralty jurisdiction. Under Supplemental Admiralty Rule B of the Federal Rules of Civil Procedure (FRCP), a plaintiff may attach any of the in personam defendant’s ‘tangible or intangible personal property’ that is found within the jurisdiction, provided that the defendant itself is ‘not found within the district’ (meaning the federal district in which the court is located).97 Arrest of a ship is done under Supplemental Admiralty Rule C without reference to the ship’s ownership;98 a ship may be attached under Supplemental Rule B, like any other of the defendant’s assets, but only if the plaintiff can show that the ship is ‘tangible … personal property’ of the in personam defendant – ie that the in personam defendant owns the ship in question. Because Rule B attachment is peculiar to the admiralty jurisdiction, it is (obviously) available only if the plaintiff ’s claim is of a kind that falls within that jurisdiction. As noted above, there is no statutory list of claims falling within the admiralty jurisdiction of US courts, so courts typically answer the preliminary question of subject-matter jurisdiction simply by applying the principles of federal maritime law stated by the US Supreme Court.99 As in the case of the first, conditional, question under UK-based systems, it is sufficient that the plaintiff alleges a claim of the right kind.100 Like their Commonwealth counterparts, US courts seldom acknowledge that their consideration of this preliminary jurisdictional question amounts to application of the lex fori, but occasionally, the choice of law question is addressed explicitly, as it was in Blue Whale Corp v Grand China Shipping Development.101 In Blue Whale, a Liberian-flagged ship was chartered to carry a cargo of iron ore from Brazil to China. The charterer refused (or, rather, allegedly refused) to pay freight as agreed in

92 See, eg, Senior Courts Act 1981 (UK), s 20(2); High Court (Admiralty Jurisdiction) Act c 123 (Sing), s 3(1); High Court Ordinance (HK), s 12A(2); Admiralty Act 1988 (Cth), ss 4(2), (3); Admiralty Act 1973 (NZ), s 4(1). 93 International Convention Relating to the Arrest of Seagoing Ships, 10 May 1952, 439 UNTS 6330 (1962) Art 1(1). 94 28 USC § 1333(1). 95 See above, n 77. 96 M Davies, ‘In Defense of Unpopular Virtues: Personification and Ratification’ (2000) 75 Tulane Law Review 337. 97 FRCP Supp Admlty r B(1)(a). 98 FRCP Supp Admlty r C. 99 See, eg Norfolk Southern Railway Co v James N Kirby Pty Ltd 543 US 14, 125 S Ct 385 (2004) (what constitutes a maritime contract); Jerome B Grubart Inc v Great Lakes Dredge & Dock Co 513 US 527, 115 S Ct 1043 (1995) (what constitutes a maritime tort). 100 Bell v Hood 327 US 678, 682, 66 S Ct 773, 776 (1946): ‘Jurisdiction, therefore, is not defeated … by the possibility that the averments might fail to state a cause of action on which petitioners could actually recover. For it is well settled that the failure to state a proper cause of action calls for a judgment on the merits and not for a dismissal for want of jurisdiction.’ 101 Blue Whale Corp v Grand China Shipping Development 722 F 3d 488, 2014 AMC 145 (2d Cir 2013).

120 Martin Davies the charterparty, so the shipowner withheld delivery of the cargo, incurring over $1 million in damages as a result of the delay. The shipowner claimed that sum in arbitration in London, but it also proceeded in the US District Court for the Southern District of New York against a company associated with the charterer. The shipowner alleged that the associated company was the alter ego of the charterer, with the result that the associated company’s property could be attached under Rule B as property of the charterer.102 On appeal, the US Court of Appeals for the Second Circuit reversed the district court’s order vacating the Rule B attachment. The court held that the Rule B claim was procedural in part and substantive in part. The preliminary question of whether the plaintiff ’s claim sounded in admiralty was a procedural question governed by federal maritime law (the lex fori); the question whether the claim had sufficient prima facie validity to allow attachment to proceed was a substantive question governed by the relevant substantive law (the lex causae). Although the plaintiff ’s claim was for breach of a charterparty governed by English law, the court held that the choice-of-law clause in the charterparty did not control the question of whether the company that owned the attached assets was the alter ego of the charterer.103 Federal maritime law did not apply automatically as lex fori to the question of the corporate identity of the in personam defendant. The true corporate identity of the defendant was to be determined by applying federal maritime choice of law rules. The applicable multifactor choice of law test104 pointed to the lex fori, the forum being the strongest relevant point of contact. The court said: This particular case arose from a charterparty between a Chinese company, [the charterer], and another foreign company, [the shipowner], to ship iron ore from Brazil to China on a Liberian vessel. This n arrative yields several potential sources of law; none have a particularly strong connection to the transaction … Importantly, however, the relevant ‘transaction’ in this case is not [the charterer’s] alleged failure to comply with the charterparty – it is [the shipowner’s] claim to pierce the corporate veil. The district court in this Rule B action is charged only with determining whether [the shipowner] stated a prima facie valid alter-ego claim against [the associated company] in furtherance of its motion to attach [the associated company’s] property in New York. Accordingly, United States law has the strongest ‘points of contact’ with this claim by virtue of the location of [the associated company’s] property, [the shipowner’s] corresponding choice of forum and the unavailability of an alternative forum, and the absence of a dominant foreign choice of law.105

The substantive question considered by the court in Blue Whale arose because the Federal Rules of Civil Procedure provide that when property has been arrested or attached, the owner is entitled to ‘a prompt hearing at which the plaintiff [is] required to show why the arrest or attachment should not be vacated’.106 The prompt post-seizure hearing does not take the form of a full adversary hearing constituting a mini-trial of the merits; it is intended only to make a preliminary

102 Arrest or attachment for the purpose of obtaining security for arbitration occurring elsewhere is permitted by 9 USC § 8. 103 Blue Whale (n 101) F 3d 496, AMC 154. 104 The test stated in Lauritzen v Larsen 345 US 571, 73 S Ct 921 (1953) and in Hellenic Lines Ltd v Rhoditis 398 US 306, 90 S Ct 1731 (1970) has eight factors: (1) the place of the wrongful act; (2) the law of the ship’s flag; (3) the domicile of the injured party; (4) the domicile of the shipowner; (5) the place of the contract; (6) inaccessibility of the foreign forum; (7) the law of the forum; and (8) the shipowner’s base of operations. The US Court of Appeals for the Second Circuit has observed that the fifth, sixth and seventh factors have no application beyond the particular facts of Lauritzen (a case involving personal injury to a seaman) and should be ignored, and the eighth factor should be expanded by considering the base of operations of all parties: see Carbotrade SpA v Bureau Veritas 99 F 3d 86, 1997 AMC 98 (2d Cir 1996). 105 Blue Whale (n 101) F 3d 499–500, AMC 160. See also Richardson Stevedoring & Logistics Services Inc v Daebo International Shipping Co Ltd, 2015 WL 1781712 (ED La 2015), applying a similar analysis with a similar result: alter ego status was determined by US maritime law, although the underlying claim was governed by Korean law. 106 FRCP Supp Admlty r E(4)(f).

Choice of Law in Determining the Ownership of Ships 121 determination that there were reasonable grounds, or ‘probable cause’, for requiring seizure of the defendant’s property.107 Nevertheless, the Rule E(4)(f) hearing (as it is known) does require at least some consideration of the strength of the plaintiff ’s case on the merits. That is the crucial difference from UK-based systems, a difference that highlights when choice of law is relevant and when it is not. An American court conducting a Rule E(4)(f) hearing in relation to a claim arising in another country must necessarily consider the (possibly foreign) lex causae, because it must consider whether the attachment should be allowed to continue. That is not a jurisdictional question; it is an exercise of the court’s admiralty jurisdiction established at the first stage of the court’s inquiry. In contrast, the question of ship ownership under UK-based systems is a jurisdictional question, not a substantive question that requires any consideration of the merits of the case. The Singapore Court of Appeal was quite right to observe in The Jarguh Sawit that the scope of judicial power in such cases must be determined by the law of the state which confers the power, ie the lex fori, and cannot be determined by reference to the laws of another state.108

VI. Conclusion Choice of law in relation to the ownership of ships is one of the more stubbornly difficult questions in the conflict of laws, largely because of the unavoidable fact that trading ships sail the world, spending much of their time on the high seas beyond the geographical territory of any legal jurisdiction and then calling at ports in countries that have no necessary connection with the ship or its owner. Nevertheless, reliance on the law of the ship’s flag as a supposedly constant referent is a chimera, both in practical terms and as a matter of principle. The question of who owns a ship can only be determined by the law of the place where the ship is when the relevant transaction is completed, which has the advantage of the fact that it is usually a place chosen by the parties. In contrast, the question of beneficial ownership of a ship for purposes of ship arrest is properly a jurisdictional question for the lex fori, with no obvious reason to consider foreign law, as opposed to jurisdictional facts that may have occurred in a foreign country.

107 Salazar v The Atlantic Sun 881 F 2d 73, 79–80, 1989 AMC 2594, 2602–03 (3d Cir 1989); World Fuel Services Singapore Pte Ltd v MV AS Varesia 727 Fed Appx 811, 814, 2018 AMC 686, 690–91 (5th Cir 2018); 20th Century Fox Film Corp v MV Ship Agencies Inc 992 F Supp 1429, 1434, 1998 AMC 2514, 2518 (MD Fla 1997). 108 See above n 58.

122

7 Private International Law and the Privy Council ADRIAN BRIGGS

I. Introduction Before lawmakers in Europe found that they had a free hand to replace our laws, the received wisdom was that the common law’s elaboration of private international law, occasionally after a stern word from Dr Morris, was the envy of the world. The common law, in this field, was a comfortably uniform thing; it was ‘the’ common law. English courts were able to take advantage of decisions which were technically foreign but were delivered by the Privy Council, and the law was duly enriched. As to whether benefits flowed the other way, it may have been argued that access to the intellectual resources of the common law offered something of intangible value, and a perspective which adjudication before an inexperienced court might not always achieve. But as the world changes it is timely for a private international lawyer to take a sideways look at big issues which arrive in London from the smaller jurisdictions of the common law world. The relationship between the Privy Council and the laws of foreign states and societies has become controversial in particular in dealing with penal laws of the Caribbean, but its recent contributions to private international law suggest that it is time for a broader reflection. A problem for the smaller overseas jurisdictions of the common law world is that they are sometimes very small indeed. Their written laws and published law reports barely fill a shelf in the library. Their private international law is for all practical purposes unwritten, and yet their courts increasingly have to engage with intricate issues arising from cross-border finance and cross-border insolvency. Answers to questions of private international law have to be found somewhere, but where, and by what means? The Privy Council may be called upon to give the answers, but where does it find them? The courts of some of these jurisdictions will often find themselves dealing with complex cases with overseas facts. Their economies and legal systems need to cooperate as much as they also compete with each other; and their appreciation of how they should go about it is, one imagines, illuminated by the individual and peculiar concerns of each individual jurisdiction. When a case or matter is big and important enough to make its way to the Privy Council, it is fair to say that it will be adjudicated far from home and by a judiciary not rooted in local soil or cut from local cloth.1 The Privy Council is likely to be dealing with a set, or non-set, of laws which differ, at several levels, from what would be presented to the Supreme Court in an English appeal. This presents challenges which are not always recognised or sympathetically managed. 1 This is certainly not to say that the Privy Council is insensitive to such concerns, which would plainly be wrong. The true issue is whether its perspective on them is the most appropriate one.

124 Adrian Briggs

II. The Effect of Statutes There are two questions which may face the court in a small jurisdiction. How should it read its statutes? And if the conclusion is that these statutes do not tell the court to do what it wishes to do, can they yet help it to its preferred answer? First, the interpretation of statutes. The question for the Bermuda courts in PricewaterhouseCoopers v Saad Investments Co Ltd2 was whether they could make an order to wind up a Cayman company which was also being wound up in its home jurisdiction. The practical reason for doing so was to allow the Bermuda court to grant certain relief consequential on a winding up order.3 The Bermuda courts held that the order could be made, on the basis that the relevant provisions of the Bermuda Companies Act 1981, when read together, allowed the court to wind up this overseas company, and that it was just and equitable – and purposeful – to make the order applied for. But the Privy Council shortly rejected this reading of the Act. It held that the courts had no power to wind up a company outside the terms of the Act, and that these did not extend to an overseas company which did not carry on business in Bermuda.4 The local courts had read the legislation differently, and they were told that they were wrong. It is far from clear that this interpretation served the interests of the courts, law, and public policy of Bermuda; and if those interests were to be immaterial to the interpretation of a Bermuda statute, it is far from clear that the Privy Council was the body to say so. The result was that the winding up order was set aside, leaving the Bermuda courts powerless to act in a case in which they had found it to be appropriate to act, and in circumstances in which the Bermuda legislature had not explicitly said that the courts could not act. It is not difficult to see when this might, in Bermuda, cause real and deep problems; given the views of the courts in Bermuda, it was not a compelling way to read a Bermuda statute.5 But the Privy Council simply said that the statute did not permit the court to wind up the Cayman company, even if6 the Cayman court would have raised no objection; and that was that. The Privy Council proceeded as though the Bermuda legislature had measured out with precision the powers it was prepared to entrust to its judiciary, conferring the winding up power as a ‘not unless’ power – as saying that, ‘except as provided by this section’, a company may not be wound up. This reading of the statute appears to be founded on an assumption about the capacity, intentions and ability to see into the future of the local law-making process which is at least debatable. The 1981 Act set out when the court might wind up a company, but did not expressly say (although it may have assumed) that it was forbidden to act in other cases. The Privy Council was therefore discerning the presence of words which were not there, or if there, were written in invisible ink. It was applying to Bermuda legislation a rule of statutory construction which made

2 PricewaterhouseCoopers v Saad Investments Co Ltd [2014] UKPC 35, [2014] 1 WLR 4482. 3 The proposition that a court can grant relief which is ancillary only if it is seised with a principal issue has bedevilled the law of insolvency and winding up just as much as it vexed the matrimonial courts which could grant financial relief only if they granted the divorce. In the matrimonial context, the legislative solution (eg the Matrimonial and Family Proceedings Act 1984) was to allow relief even though another court dissolved the marriage. In the insolvency context, the solution was to allow an ancillary (or ‘non-main’) liquidation to run alongside a principal (or ‘main’) proceeding. 4 And on this issue was significantly different from the English Act. 5 Moreover, as the Companies Act 1981 (Bermuda), s 184 gave the court a power to stay winding up proceedings, it would have been possible to hold that the court had power to act but should stay what had been started. 6 The report does not disclose the attitude of the Cayman court, as the application was brought by the liquidators acting in their own right. But it is reasonable to suppose that the Cayman court did not object to the application being made, any more than the Bermuda court did to the granting of it.

Private International Law and the Privy Council 125 sense in England, as though there were no material difference to be observed. It may be said to be a truth in England that ‘a statute applying to dogs cannot be applied to cats’.7 However, it is patently not so: if a court says it can be, it does. When this kind of question arises, the real issue is where, and by whom, the answer should be given. In Saad Investments, the applicants had based their application on the statutory winding up power in the Bermuda Companies Act 1981.8 However, as the Act did not give the Bermuda courts the power to wind up the company as an overseas company, might it still allow the court to make orders which would assist the foreign court or the liquidator appointed and recognised in Bermuda as such? If that were so, the pieces of the cross-border jigsaw would fit together fairly well: the home court of the overseas company would do the winding up; the Bermuda court would assist, always presuming that it wished to assist, by exercising the various ancillary powers that the statute gave it. As long as the Bermuda court recognised the foreign proceedings as properly brought, and recognised the status of the officers appointed by the foreign court to represent the company, it would be rational9 for the Bermuda Companies Act 1981 to direct the court to assume a supporting role. In Singularis Holdings Ltd v PricewaterhouseCoopers,10 the Cayman liquidators, believing that the Cayman company’s auditors, a Bermuda partnership, held papers which would assist them in fulfilling their mandate, applied to the Bermuda court for their disclosure,11 pointing to section 195 of the Bermuda Companies Act 1981. The High Court and Court of Appeal accepted that the section could not be used directly to order the relief applied for, as the company was not being wound up in Bermuda; the Privy Council held that this conclusion was correct. Desirability of outcome (as to which no conclusion was expressed) could not overcome the construction of the statute. However, the Chief Justice had foreseen this possibility in Re Saad Investments and Re Singularis (the two matters being joined before the Bermuda High Court for the purpose of the assistance and disclosure applications). He reasoned that section 195 could be ‘applied by analogy’ to justify the granting the relief sought: there was, as he saw it, nothing in the statute to prevent it, no rule of common law to oppose it and good reason to do it.12 To someone trained in the common law tradition, it might be said that he had derived a rule of public policy13 from his legislation; to someone trained in another legal tradition, it might be said that he accorded ‘reflexive effect’ to the statute.14 The Bermuda Court of Appeal was dubious, and found distinct grounds to refuse

7 A Burrows, Thinking About Statutes: Interpretation, Interaction, Improvement (The Hamlyn Lectures) (Cambridge, Cambridge University Press, 2018) 43. Another approach would recall that the beatitude ‘blessed are the cheesemakers’ is not meant to be taken literally, but applies to any manufacturers of dairy products. 8 A second argument (it is unclear that it was relied on below), that the External Companies (Jurisdiction in Actions) Act 1885 (Bermuda) allowed the Cayman company, doing business in Bermuda, to be sued in Bermuda, went nowhere: there was no cause of action arising in Bermuda alleged against the company, and it was not being sued. 9 Though, plainly, it is not the only rational view. 10 Singularis Holdings Ltd v PricewaterhouseCoopers [2014] UKPC 36, [2015] AC 1675. 11 They had taken the view that as a matter of Cayman law, the Cayman court had no power to order disclosure, on the basis that the information did not belong to the company. Lord Sumption doubted the soundness of this conclusion: ibid [30]. 12 Re Saad Investments Co Ltd [2013] SC (Bda) 28 [36]–[77], esp [47], [53], [71], [73]. 13 It is true that the usual role of public policy in the English system is to prevent the application of a foreign law rather than to justify the application of a local one. But if the results are different, the underlying justifications do overlap. 14 In the context of the Recast Brussels Regulation (ie Regulation (EU) 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters [2012] OJ L351/1) and its predecessor instruments, it is generally understood that certain of its provisions may be applied ‘with reflexive effect’, or by analogy, to a case in which the regulation does not directly apply but the logic of the regulation applies just as obviously.

126 Adrian Briggs the relief.15 But in the Privy Council in Singularis, Lord Collins’s indignation was extraordinary. Not only would he have dismissed the appeal, but he would also have advised Her Majesty: that to have allowed it on the basis of the liquidators’ primary argument would have involved Her Majesty’s judges in a development of the law and their law-making powers which would have been wholly inconsistent with established principles governing the relationship between the judiciary and the legislature and therefore profoundly unconstitutional.16

No other member of the Privy Council used such language, but none expressed any obvious support for the approach which the Chief Justice had favoured. Some may see in that reaction a good opportunity squandered: squandered because, as we shall see, Lords Sumption and Collins did not read section 195 as preventing the ordering of relief under the common law if the common law could otherwise support it. In other words, section 195 was neither a ‘not unless’ or an ‘except as provided by’ statutory provision. If the Bermuda Companies Act 1981 did not forbid the granting of the relief sought, the question whether the powers of the court to assist should be shaped by the reflection of Bermuda statute law as made in section 195, or decoded by reference to the opaque and frankly baffling voices of the English common law,17 did not obviously require the answer which the Privy Council favoured. Some may consider that the development of local extra-statutory law by reference to a local statute is more coherent than to try and make sense, and to determine the trajectory,18 of the common law of England. I will return to this point more than once. But the Privy Council was having none of it.

III. Developing the Common Law One therefore turns to the common law. A frank admission will set the scene. ‘The extent of the extra-statutory powers of a common law court to assist foreign liquidators,’ said Lord Neuberger, ‘is a very tricky topic on which the Board, the House of Lords and the Supreme Court have not been conspicuously successful in giving clear or consistent guidance.’19 One may note in passing the reference to ‘a common law court’ as though the guidance should be the same across the jurisdictions of the common law. The point about the assistance being extra-statutory arises with particular importance where these small jurisdictions are concerned and there is less statutemaking, because if the courts are driven back to the common law more frequently than would be the case in England, the law is all the more liable to be unclear and inconsistent. Almost before we have begun, the case for looking to a statute for clarity and consistency becomes apparent. Three cases, all of which ended up before the Privy Council, frame the analysis which follows. In Al Sabah v Grupo Torras SA,20 a Cayman Islands case, the courts of The Bahamas issued, at the request of a Bahamian trustee in bankruptcy, a letter of request to the Cayman court, asking 15 PricewaterhouseCoopers v Saad Investments Co Ltd [2013] CA (Bda) 7 Civ. Compare the judgment of Bell AJA (with whom Zacca P agreed) at [50]–[61] with the judgment of Auld JA at [53]–[61] (the paragraph numbering of each judgment starts at [1]). The conclusive reason was that even if the court had power to make the order, it should not do so, as it was relief which the originating Cayman court could not itself have made, prompting the response that if the only time an order should be made is when the foreign court could itself have made it, there is very little point. 16 Singularis (n 10) [108]. One wonders whether Her Majesty has ever received advice from the Board in that form. It was an astonishing way to treat a perfectly respectable opinion from the Chief Justice of Bermuda. 17 For justification of this assessment, see immediately below. 18 In the elegant idiom of Beatson J in Blue Sky One Ltd v Mahan Air [2010] EWHC 631 (Comm) [88]. 19 Singularis (n 10) [154]. 20 Al Sabah v Grupo Torras SA [2005] UKPC 1, [2005] 2 AC 333.

Private International Law and the Privy Council 127 it to set aside a trust established by the bankrupt. The Cayman courts considered that they should and could act, under either of two Cayman statutes,21 failing which there was an inherent power to assist. The Privy Council held that though the Cayman Bankruptcy Law did not apply to the facts, the Bankruptcy Act 1914, which had been extended to the Cayman Islands, still applied. It doubted, in passing, the proposition that even if the statute had not authorised the relief, the Cayman courts still had an inherent power to grant it, saying that the inherent powers of the Cayman court could not be broader than the corresponding powers of an English court in the light of English legislation; and it held that, in any event, any such powers could not achieve a result which the Cayman Bankruptcy Law failed to produce. Every part of that reasoning is questionable. In Cambridge Gas Transportation Corp v Official Committee of Unsecured Creditors of Navigator Holdings plc,22 a case from the Isle of Man, a US court had adopted a scheme for the financial reconstruction of a Bahamian company, part of which involved, or would require, the share register of an associated Manx company to be altered to substitute the committee as holder of shares in that Manx company. The appellate division of the Manx High Court considered that its Companies Act 1931 allowed it to grant the relief applied for. A unanimous Privy Council rejected this reading of the Act but still saw good reason to dismiss the appeal of the former shareholders and make the order on the basis, as it appears, of the inherent power of the Manx court to assist a foreign bankruptcy court in doing its job. The third case is Singularis Holdings Ltd v PricewaterhouseCoopers.23 Having rejected the direct and indirect application of the Bermuda Companies Act 1981, a majority of the Privy Council accepted that a common law power to assist an overseas liquidator could be found to exist independently of statute, and could in principle be used in a case of this kind, because there was no statutory bar to its application. But they held that it did not justify the relief applied for, for that would involve the Bermuda court making an order which the Cayman court, supervising the liquidation, could not have made under its own law. There is something puzzling about the proposition that the Bermuda court should only make the order if the Cayman court could have made it. If the Cayman court could have made the order and had made it, there would be no obvious need to apply to the Bermuda court; if the Cayman court could have made the order but had not done so, it would be wrong for the Bermuda court, in effect, to contradict it;24 and if that is right, it makes the common law little more than a mirage. No wonder Lord Neuberger said what he said. What is striking about these three cases is the way in which the local judges arrived at the conclusion that, in light or spite of the wording of their statutes, they had power to give the assistance which was sought. On the footing that they were experienced local judges who understood what answer made the most sense for their own laws, and looked for the most sensible route to it, it is disconcerting to read the way in which the Privy Council told them that they were wrong. In all three cases, the local courts were assessing their laws in conditions very different from those

21 Cayman Bankruptcy Law (1997 revision), ss 107 and 156, or the Bankruptcy Act 1914 (UK), s 122 (which had not been repealed by the Insolvency Act 1985 (UK) so far as the Cayman Islands were concerned). It is clear from the judgment of the Privy Council that following the trail of non-Cayman legislation into present-day Cayman law may require almost superhuman power. 22 Cambridge Gas Transportation Corp v Official Committee of Unsecured Creditors of Navigator Holdings plc [2006] UKPC 26, [2007] 1 AC 508. 23 Singularis (n 10). 24 cf Crédit Suisse Fides Trust SA v Cuoghi [1998] QB 818 (CA), where the analysis is different: if the Swiss court could not have made the order, the English court was free to.

128 Adrian Briggs applicable to an English judge, who has to work in a very different legal landscape;25 to put it only a little crudely, the local judges were best placed to know, and were well able to decide, how their law worked. It seems fair to say that there was a broadly common thread running through the decisions of the courts in these small jurisdictions, which is clearly in favour of finding ways to assist overseas liquidators who seek their help. So far as the local courts drew indirect support from their statutes, it was the Privy Council, on the basis of what appeared to be a statement of English constitutional principle, that held that they were wrong to see the issues as they did. The proposition adopted by the Board appears to be that if an answer is to be derived from a statute, it must be derived directly; and if it cannot be, the courts must wait for the local legislator to amend the statutes. This is surely unrealistic. The resources needed to think through and make the statutory changes which have been made to English (including, where relevant, European) company and insolvency legislation are not always available in these small jurisdictions, where the duty of keeping the law in good working order is, faute d’alternative, the responsibility of the local judiciary. In this respect, it makes little difference whether the appeal was from the courts of a colony or former colony,26 or from the Channel Islands or the Isle of Man. It is at least arguable – it is certainly not revolutionary or unconstitutional to think – that in a country whose capacity for legislative law-making is not equivalent to that of the UK, the law should assign to their judges a rather greater responsibility to fill in the gaps, and refurbish the laws, than is accepted in England. Is recourse to the common law a sufficient answer, at least ad interim? The main trouble is that for any such argument to work one must find a pre-existing rule of the common law – which must surely mean the common law of the jurisdiction in question – with which to work. Sometimes there is none, meaning there is nothing to develop and nothing to be done. At that point, the courts are paralysed, and just have to wait. To take an easy example, in the case of service of process out of the jurisdiction, the whole of the law is statutory. The common law has no rule or authority which allows process to be served out of the jurisdiction on a defendant who is not within the territorial jurisdiction of the court; there is therefore no common law rule to develop to address any shortcoming in the statute law. If it is also correct that there is no power at common law to wind up a company, the only power being statutory and set out in legislation – such as the Bermuda Companies Act 1981 – there is no common law to develop to allow a court to act when the statute does not.27 By contrast, it was accepted in Singularis28 that there was a deeply obscure29 power at common law to order the disclosure of information, and to make certain orders to make more effective the administration of a recognised overseas insolvency; and what the common law had allowed, the common law could build upon and develop. The distinction between the two classes of case appears to be based on little more than English historical accident. It does not even reflect a distinction – which would at least be intelligible – between substantive and procedural law, for the two examples just given are procedural in nature. It does not appear to ask any questions

25 Lord Sumption alone acknowledged the point in Singularis (n 10) [9]–[11], but it seems that little practical account was taken of it. 26 Or ‘British Overseas Territory’, as colonial territories are now designated. 27 It is correct to say that the Companies Act 1981 (Bermuda) is silent. It would be strange to characterise the law on winding up as stating that a company has a right not to be wound up unless that right is overwritten by a statute, but that, in effect, is what is being said by those who read the statute as a ‘not unless’ provision. 28 Accepted rather more cautiously in Al Sabah (n 20). 29 The divergence of detailed analysis in the judgments of Lords Sumption and Collins, who were the proponents of this common law power, to say nothing of the dissent of Lord Mance, speaks for itself.

Private International Law and the Privy Council 129 about the legal structure of Bermuda or any other small jurisdiction; nor does it reflect any other rational principle, either. It is a mess.

IV. The Limits of the Common Law, and Cambridge Gas At least part of the reason for the peculiar force of the judgments in Singularis was the perception that the Bermuda courts had been distracted by the decision of the Privy Council, on appeal from the Isle of Man, in Cambridge Gas. In order to complete the analysis of what the common law can and cannot do, it is appropriate to now say a little more about this oddly vexing Manx decision, the essential facts of which were given above. The disquiet over Cambridge Gas may have three elements. One objection to it is entirely orthodox: its effect was to treat a US judgment or order as having a conclusive effect in Manx law, even though it did not comply with the Manx law on foreign judgments. Lord Hoffmann’s assertion that it was doing no such thing cannot survive the duck test.30 A second problem would be the way in which the Privy Council appeared to encourage the Manx court to look to its statute law and to reason that even if the Manx Companies Act 1931 did not directly justify the relief, as no proceedings had been brought under the Act, the legal policy for which it stood could be used – by analogy, if you like – to justify the relief asked for. It may have encouraged the Chief Justice in Singularis, though he was perfectly able to reach his preferred conclusion without it. The third sticking point may have been the view that a belief in ‘modified universalism’, no matter how fervently held, was not a talisman which allowed a court to do whatever it felt right, or to grant relief which did not rest on a sustainable application of a pleadable cause of action. This principle requires that English courts should, so far as is consistent with comity, justice and the public policy of the UK, cooperate with the courts in the country of the principal liquidation to ensure that all the company’s assets are distributed to its creditors under a single system of distribution. However, while it may be a reason for doing things which can be done, it is not a basis for doing things which cannot otherwise be done. For present purposes, no issue is taken with the first or third of these; they are objections which are understood by any lawyer. The second needs more care. The first judicial blow against Cambridge Gas was struck in Rubin v Eurofinance SA:31 an English case, entirely concerned with English law. A US court exercising insolvency jurisdiction had requested the cooperation of the English courts, specifically in asking that a US judgment be used as the basis for an English order requiring a respondent to pay (back) money into the bankrupt estate. The Supreme Court refused to grant the relief, on the footing that for it to make the order would, in effect, contradict the English common law, which required the (non-)recognition of this US judgment as res judicata. If this is indeed what English law said, the Supreme Court was bound to say it. But the court in Rubin was applying English private international law, and that makes the statement that Cambridge Gas was wrong seem inapt, given that Cambridge Gas was a case on the laws of the Isle of Man.32 So far as Singularis was concerned, it was the Board itself which had raised the possibility that there was, separately from the statutory and quasi-statutory arguments, an inherent power in the law of Bermuda to assist the Cayman court. The liquidators took the bait and developed the argument, based on Cambridge Gas. They may have been

30 If

it looks like a duck, swims like a duck and quacks like a duck, then it probably is a duck. v Eurofinance SA [2012] UKSC 46, [2013] 1 AC 236. two members of the court declined to associate themselves with it.

31 Rubin 32 And

130 Adrian Briggs rather surprised by the ferocity of the response. Standing back, it may seem rather a contrived observation that ‘the parties are entitled to have the views of the Board on the argument [based on Cambridge Gas] which was actually put before it’.33 If a common law court has power to grant relief outside the statute, how far can it go? If it would involve the court requested in doing something which is forbidden by a ‘not unless’ rule of its own statute law, it will be unable to act. For the English court in Rubin, to accede to the application made by or on behalf of the US court would have required the English court to recognise as res judicata a foreign judgment which did not qualify for recognition under the applicable rules of the common law. The Supreme Court concluded that its common law rules on the recognition of foreign judgments were to be understood as though they were statutory and made in ‘not unless’ form, for otherwise it would have been open to the court to ‘develop’ the relevant rule of common law.34 This might also explain Cambridge Gas: for the Manx court to grant relief would be for it either to recognise as res judicata a foreign judgment which did not satisfy Manx law on res judicata or to order the alteration of the share register without the procedural and other requirements of the Manx Companies Act having been followed. If those provisions of common law and statute respectively are interpreted as being mandatory and exclusive in a ‘not unless’ sense, as Lord Collins35 appears to have regarded them, then they cannot be bypassed by purported discovery of a common law power to assist another court. But, as Lord Sumption had pointed out in Singularis,36 section 195 of the Bermuda Companies Act 1981 did not preclude the recourse to the common law. The observations about the rightness or wrongness of Cambridge Gas were a distraction from what should have been the court’s primary task, namely dealing with the true state of the laws of Bermuda. In Singularis, Lord Sumption considered that to make the order applied for would not in principle undermine section 195, but to have wound up the Cayman company would, it appears, have undermined section 161. How could we tell that? How can we make such judgement calls in the future? On the section 195 point, Lords Collins and Clarke agreed without giving substantial reasons. Lords Mance and Neuberger disagreed, though whether this was based on the exclusivity of section 195 or scepticism that it was sufficient to say that the common law may develop, or both, is unclear. When Lord Neuberger said that he would expect a power to order the production of documents to depend on whether it had been bestowed by the legislature, he drew attention, if accidentally, to the point made earlier: that such an observation, perfectly attuned to the English context, did not naturally apply to a state with plenty of law and finance to regulate but an entire national population equivalent to the population of Salisbury.37 One would think that it made perfect sense, in the context of a small jurisdiction with a lot of business to look after, to say that unless the legislature had enacted laws in ‘not unless’ form, the default position was that the court had power to go beyond the statute where the court considered it proper, or just and equitable, or consistent with the policy of the law as the courts of that jurisdiction understand it, to do so. 33 Singularis (n 10) [35] (Lord Collins). Lord Collins had also delivered the principal judgment in Rubin (n 31). However, in defence of this aspect of Singularis, it is only right to observe that trying to make sense of Cambridge Gas and its status had plainly taken up a lot of time and effort, in particular in the Court of Appeal, and sorting it out was liable to be for the benefit of other courts that would be saved the trouble. 34 It appears that we must be able to distinguish between common law rules, clear of any statute, which can be developed, and common law rules, clear of any statute, which cannot. How this works in England is completely unexplained; how it makes sense for a small overseas jurisdiction is quite incomprehensible. 35 Lord Collins was not party to the decision in Cambridge Gas. 36 Singularis (n 10) [28]. 37 That is, one thousandth of the population of the UK. The point is even more acute in the case of jurisdictions like Sark (population 600; one hundredth of the population of Bermuda), of which Lord Hodge said: ‘It has its own legislature, the Chief Pleas, but it has limited resources to draft and enact legislation’: A v R [2018] UKPC 4.

Private International Law and the Privy Council 131 That does not tell us how to identify a statute as in substance a ‘not unless’ one, but it would put rational distance between the position which is taken in English law and that which applies in common law jurisdictions which are materially very different. It is not hard to argue that if a judgement of this kind has to be made, the court best placed to make it is one embedded in the system whose law this is. We may all be common lawyers, but the Privy Council was not called upon to take and apply the principles of English common law, or even of a delocalised common law. It was required to think in terms of Bermuda common law, Cayman common law or Manx common law, and to find its answers within that framework. These decisions may only be the tip of the small jurisdiction iceberg, but they prompt the broader question: how should a court proceed when a statute provides that it may act in certain cases, but does not say in express terms that it may not act in others? It is also necessary to ask whether a single answer is available for the whole of the common law world. There is no shortage of material available to those who wish to understand the common law rules on the construction of statutes as they are understood in England; the recent Hamlyn lectures are a wonderfully lucid way in.38 No doubt the starting point is that the court may not act by reference to the statute unless the statute positively says so; that ‘if ’ and ‘when’ generally mean ‘not unless’. For example, service of process out of the jurisdiction may be authorised to be made when the conditions set out in the legislation are satisfied, but not unless they are. It would seem strange to argue that these statutory provisions could be extrapolated, if that is the right word for it, to a case in which their requirements were not met.39 But whatever else these are, they are statements of English interpretation law, applied to English statutes. The real question is whether the rules of statutory interpretation, and the relationship between common law and statute law, is uniform across the common law world, or is particular to each common law jurisdiction. The view taken here is that it is the latter, not the former; that the English common law rules which explain how one approaches an English statute are geographically, culturally and legally confined to England, English scholarship and English law. The principal objection to the broad approach of the Privy Council cases is that they approach appeals from these small jurisdictions as though they are something which they simply are not.

V. The Derivation of Rules from Statutes It is time to return to the proposition that a statute which does not apply directly may yet allow a decision to be reached by applying it ‘by analogy’. Lord Collins disparaged this technique as treating the statute ‘as if ’ it had been worded in a way in which it was not, or as invoking the ‘equity of the statute’ as a source of law.40 As was said above, the application of statutes ‘with reflexive effect’ is an idea which has become familiar to some private international lawyers.41 The case for it lies in the fact that if the national lawmaker has formulated a legal policy and then given it effect in legislation which is not framed in ‘not unless’ form, that policy may point the court to a justifiable answer where the legislation does not speak and no other positive law prevents its acting.

38 Burrows (n 7). 39 Having said that, the courts allow service to be made where there is a ‘good arguable case’ that the jurisdictional rule is satisfied. Despite the lack of clarity in the law over the precise meaning of this expression, it can be said to produce the result that service out is allowed in cases which do not, on a plain reading, satisfy the statutory or sub-statutory rule. 40 Singularis (n 10) [38], [64], [78], [82], [83] and [93] (Lord Collins). Lord Sumption agreed (at [18]), as did Lord Mance (at [143]). 41 See n 14.

132 Adrian Briggs It appears to be a more reliable method than that of reaching back into a version of the common law which has not evolved from the decisions of local courts, and asking whether it should now ‘develop’. If one asks whether either of these techniques for judicial problem-solving is optimal, the answer is of course not. But the law was made for man, not man for the law; and no sane person would deny the need for the law to be mobile. If there is a need, the only useful question is how that need should be supplied. It can be from the creative (re)interpretation of the words of the legislation, for example interpreting ‘harm’ in a Victorian statute as extending to emotional or psychological trauma of a kind, the existence of which would never have crossed anyone’s mind at the time of enacting. It can be from the creative alteration of the common law: where would we be without Donoghue v Stevenson or Hedley Byrne v Heller? And it can, it is submitted, be done just as well by deriving a rule from a statute in circumstances which are the equivalent of the development of the common law. The perception that there is a problem is not new. Professor Calabresi exposed it, brilliantly, almost 40 years ago,42 but perhaps he was before his time. He was certainly not focused on the small jurisdictions of the common law. The immediate objection to the technique of deriving a general rule from the specific and non-general words of a statute is to the uncertainty which this may inject into settled law, though in the context of Singularis it was surprising to see an objection to ‘statutory development’ not being applied in equal measure to the alteration, sub nom ‘development’, which may be ‘incremental’, of the common law. All of these alter received law, and it is a fantasy to say otherwise. Might it, then, result that a court could decide to allow service to be made out of the jurisdiction in a case which did not fall within one of the heads of Order 11 or gateways set out in the Practice Direction? Yes, it might: when it arises, as it has recently done in The Bahamas,43 it seems right and proper that the local court should decide for itself whether service out should be allowed because the matter can actually fit within one of the heads, or should be allowed because, even though it does not fall within the rule, a provision of those rules should apply by analogy to justify it. For example, suppose a claim is not contractual but quasi-contractual, or is founded not on a tort but on a purely equitable wrong, or is founded on a statutory right to claim a payment which is sui generis,44 or is not founded on property which is no longer within the jurisdiction only because it has been wrongfully removed from the jurisdiction; and that in all these cases the legislation does not authorise service to be made but (equally) does not forbid it. The question is whether a court should have power to allow service out by treating a head of jurisdiction as applicable ‘by analogy’; it is a proper question, and there is no reason to object if a court examines it with an open mind. The material test should be whether any statutory or other rule of law prohibits the extension of the law; if there is not, it should make no difference whether the law being extended into territory into which it has not previously stepped is a rule of common law or a rule of statute law.

VI. The Other Side of the Coin: Trusts and Trustees If the cases discussed above illustrate some of the difficulties which arise when the Privy Council does not appear to share the insight and outlook of the local court, it would be a mistake

42 G Calabresi, A Common Law for the Age of Statutes (Cambridge, MA, Harvard University Press, 1982). 43 AWH Fund Ltd v ZCM Asset Holding Co (Bermuda) Ltd (Bahamas CA, 16 February 2017), appeal dismissed [2019] UKPC 37. 44 Most provisions for service out also allow service out (and often without leave) if a statute gives the court jurisdiction.

Private International Law and the Privy Council 133 to think that there is only one side to the story. Investec Trust (Guernsey) Ltd v Glenalla Properties Ltd45 is the other side of the coin, and it teaches a very different lesson. The context first. In Jersey and Guernsey, legislation to regulate trusts and trustees in some detail, and by way of departure from the principles of English equity, has plainly taken the law in a new direction. Here, as well as in certain Caribbean jurisdictions, the ‘trusts industry’ makes a major contribution to the economic welfare of the state. Some observers view with distaste the whole idea of wealth concealment and liability avoidance which these trusts are designed to encourage. Some consider that these trusts – trusts without beneficiaries but with ‘enforcers’, purpose trusts set up for uncharitable and private purposes, trusts in which the settlor is able to exercise de facto influence over the use of funds while pretending to have no legal interest in or power over them – are too far from the traditional model to count as trusts, though if one were to ask for particulars of the traditional model or irreducible core, the answer is not so easy to predict. Unsurprisingly, trusts legislation46 in these jurisdictions concentrates on the position of trusts established under local law, and on the enforcement of judgments which seek to penetrate these trusts and prise them open to get at the assets. The Jersey and Guernsey statutes both contain a rule, expressly applying to trusts governed by their own law, that serves in effect47 to limit or exclude the personal liability of a trustee for contractual and other debts resulting from the acts of that person when acting or dealing, or known to be acting or dealing, as a trustee.48 This is in sharp contrast to the position taken in English law, according to which a person who acts because he is, and intends to act as, trustee still acts as a person, and still incurs liability personally in accordance with and to the extent provided for by, and only by, the terms of the contract and the lex contractus.49 To put it another way, English law acknowledges the existence and effect of separate corporate personality, but it has never accepted anything like separate trustee personality save where a contract creates and provides for something resembling it. In Investec, a Guernsey court was confronted by a claim to enforce an English law50 contract against the trustee of a Jersey law trust. The trustee pleaded and sought to rely on the statutory provision which limited the financial exposure of the trustee contained in Article 32 of the Trusts (Jersey) Law and applicable, according to that statute, to Jersey trusts. The basis on which it failed before the Deputy Bailiff was opaque, but it succeeded before the Guernsey Court of Appeal. When the matter came before the Privy Council, the Privy Council observed that Jersey trusts law applied this rule to the trustee of a Jersey trust without regard to the nationality of the trustee; and it observed further that the Trusts (Guernsey) Law both contained a similar rule for Guernsey law

45 Investec Trust (Guernsey) Ltd v Glenalla Properties Ltd [2018] UKPC 7, [2019] AC 271. 46 And it really has to be legislation, for trusts were imported from English law, having no ancestry in customary law. 47 In effect, though not necessarily in law. The Trusts (Jersey) Law, Art 32 (probably also the Trusts (Guernsey) Law, s 42) restricts the property against which enforcement may be obtained to the assets of the trust; it will not extend to the personal estate of the trustee. This has to be seen as a cap on liability (notwithstanding Investec (n 45) [61](ii) and First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637 [31]), for the Guernsey court applied it as a rule of Jersey law. Had it instead simply been an issue of (or a rule about) the enforcement and execution of judgments, it is not at all clear that a Guernsey court could have applied a Jersey law rule to that, as execution of a Guernsey judgment is governed by the lex fori, Guernsey law. The reasoning of the Privy Council is therefore to treat Art 32 of the Trusts (Jersey) Law as a law imposing a cap on the liability of a trustee where that liability arises from his being party to a matter affecting the trust. 48 Trusts (Jersey) Law, Art 32; Trusts (Guernsey) Law, s 42. 49 Muir v City of Glasgow Bank (1879) 4 App Cas 337 (HL). 50 It was not clear whether the contract was governed by English law, but it was clear that it was not governed by the laws of Guernsey or Jersey.

134 Adrian Briggs trusts and provided51 in very general terms that a non-Guernsey trust was governed by its proper law. The Privy Council put the pieces together and deduced that Guernsey private international law would treat – would characterise, if that is preferred – the particular issue of liability which had arisen before the Guernsey court as one of trusts law, to which it would apply the material provision of Jersey law, the lex fiduciae, in accordance with Guernsey private international law. The result was that the trustee was able to take advantage of the law under which it was a trustee to shield it from the claim based on the law of the contract to which it was a party. In effect, a majority of the Privy Council derived a specific rule of Guernsey private international law from the terms of the Trusts (Guernsey) Law, which was almost entirely concerned with the domestic trusts law of Guernsey; in doing so, they departed from the pattern of reasoning which English common law would have pointed to. As well they might. The decision of the Privy Council was a nice example of an old, if irregular, desideratum: that there should be ‘internal consistency’, as Professor Kahn-Freund put it,52 in a system’s rules of domestic and private international law, for they all partake of the one law. A conflicts rule which reflected the organisation and structure of domestic law was from this point of view justified and justifiable. The case was not (Heaven forfend) one in which the Guernsey statute was interpreted ‘as if ’ it applied also to a non-Guernsey law trust. The answer was found in the application of the Jersey statute to a Jersey trust, not the Guernsey statute. Had it stopped there, the reasoning of the Privy Council would have been unassailable. It was right on Guernsey law – not only because its view aligned with the assessment of the Guernsey Court of Appeal, whose business this really was, but also because the process of discerning the rules of private international law in a system whose legislative infrastructure is very different from that of England, is not a matter for English legal science or for English assumptions.53 In Investec, a rule of private international law, which led to the application of foreign law, was deduced from the scheme and apparent assumptions of the Trusts (Guernsey) Law, adopting a style of reasoning which was not of the type denounced in Singularis. The court did not apply the Guernsey rule as if it applied to a Jersey trust; it applied to a trust governed by Jersey law the Jersey rule which limited the personal liability of a Jersey trustee. It is really very difficult to see anything to object to in that. (As an aside: although it is no part of the current argument, it is worth reflecting on the pattern of reasoning which emerges from Investec. The English common law had faced analogous issues in the past, and had resolved them in much the same way. It was held to be, for example, a good answer to a claim framed in tort for the defendant to point to an exemption clause in a contract by which the parties had bound themselves to each other. The claimant’s assertion, which was entirely sound, that it was his right to frame his claim as he chose and to sue in tort was answered by the defendant’s assertion, which was equally sound, that it was his right to rely on a contractual promise which was on its true construction applicable to the circumstances of the case; and the putting of these together was the task of the conflict of laws.54 From that perspective, the shape of the argument in Investec can hardly be said to be novel: the claim was framed as one based on a contract, but the question of the extent of the defendant trustee’s liability was a matter

51 Trusts (Guernsey) Law, s 65. 52 O Kahn-Freund, General Problems of Private International Law (Leiden, Sijthoff, 1976) 223. 53 In this respect, the judicial statement in Investec (n 45) [64], [70], that it was common ground that the relevant conflicts rules of Guernsey law were those of the common law save where the statute law of Guernsey provided a different answer, is of no use until the statute has been properly evaluated. 54 For an explanation of how complicated this could be in the days in which the choice of law rule for tort claims required (more or less) double actionability and at least a flirting with the ‘incidental question’, see PM North, ‘Contract as a Tort Defence in the Conflict of Laws’ (1977) 26 ICLQ 914.

Private International Law and the Privy Council 135 of trusts law; and putting these together was the task of the Guernsey conflict of laws. What made the conclusion of the majority notable but not exportable was the way in which it derived support for its analysis from the framework of domestic law, but structurally was firmly rooted in traditional common law foundations.) So far, so excellent: in deducing the detail of Guernsey private international law from Guernsey legislation which said rather little about it, the reasoning is coherent and sound. But the majority of the Privy Council went further and expressed the view that, as a matter of ‘the common law’, the law governing the trust governed the ‘status’ of the trustee, including any issue of limitation on his power to act and liability to assume burdens, and this view of status was not confined to trustees.55 It was accompanied by the judicial suggestion that this interpretation of ‘the status of a trustee’ would now also apply as part of English law (and, presumably, to other common law jurisdictions whose legislation did not preclude it). This appears to have provoked the indignant opposition of Lord Mance, who called for the majority view to be reconsidered and overruled at the first available opportunity, as well as the withering disapproval of Professor Dickinson.56 The conclusions which resulted from the Privy Council’s reading of Guernsey statute law and which make perfect Guernsey legal sense have, in the light of what has been said above, no obvious relevance to English law, for what is sauce for the goose must be sauce for the gander. For the Privy Council to derive a rule of common law private international law for England57 from the construction of a Guernsey statute was bound to be seen as provocative and, one must say, rather ill-judged. All the important questions in the case were Guernsey law questions, and only Guernsey law questions. Guernsey had enacted its limitation of liability law to offer a measure of financial protection to trustees of Guernsey trusts. Insofar as it made sense to ask why Guernsey law would allow the trustees of a foreign trust to bring with them and rely on the same kind of benefit if the lex fiduciae provided it, the answer presumably was that persons established in Guernsey may become trustees of trusts not governed by Guernsey law; but whatever the answer was, it was one for Guernsey law alone to give and live with. For a minority in the Privy Council to asperse the majority for finding Guernsey law to be different from the settled view of English law was quite wrong. But equally, for the majority to propose their conclusions about Guernsey law for reception into English law was just as wrong.

VII. Common Laws, Private International Laws and the Privy Council In private international law, and perhaps not only there, it makes less sense than it once did to think that there is such a thing as ‘the’ common law, for today it is much more apparent that there is English common law, Australian common law, Bermudan common law, Cayman common law, Guernsey common-ish58 law, and so on. Within each jurisdiction, these common laws determine how local statutes are to be interpreted. Insofar as the rules which tell lawyers how to read statutes are rules of common law, they are liable to diverge; to grow up and grow away from the

55 Investec (n 45) [83]–[89]. The reference to ‘the common law’ is at [88]. 56 A Dickinson, ‘Status Check: Trustee Liability to Third Parties in the Conflict of Laws’ [2018] Lloyd’s Maritime and Commercial Law Quarterly 456. 57 Or, indeed, for common law jurisdictions other than England. For a suggestion that the Court of Appeal viewed this with more than a little discomfort, see First Tower Trustees (n 47). 58 Because a part of the law is also customary; for a short summary of the significance of this, see A v R (n 37).

136 Adrian Briggs original root. Some think that English common law has been bent out of shape by legislation from Westminster, as well as legislation and case law from Brussels and Strasbourg. Common law jurisdictions, even those which are not independent, have their own concerns; their courts have their own business to attend to. And anyone who wants an English book which states and explains the common law principles of private international law from an English perspective will need to scour the secondhand bookshops for an edition of Dicey which is 30 years old: it is sometimes hard for a private international lawyer to believe in the common law. What does this all mean? If one accepts that the jurisdictions of the common law world have grown up and grown apart, several ideas are worth thinking about. Four are mentioned here. First, the smaller jurisdictions of the common law, whose current59 capacity for law-making by legislation is modest, would benefit from settling for themselves, and then enacting, a statute to put on a formal basis the interpretation and application of their statute laws, spelling out in suitable terms: (i) the general principles for the interpretation of local legislation; (ii) the circumstances in which a local court is required to treat the literal wording of the statute as exhaustive; and (iii) whether a court may proceed by analogy to apply a statutory provision to a case to which it does not apply (whether expressing its conclusion as a matter of common law or of public policy, or on the ground that it is in accordance with justice, equity and good conscience for it to do so). It could also address the question whether the answers to these three issues should draw a distinction between substantive and procedural law; but these jurisdictions should do it for themselves. Secondly, a statute should authorise a court to decide in accordance with the rules and principles of English law, where to do so is in the opinion of the court consistent with the written and unwritten laws of the state, and should authorise a court to depart from the rules and principles of English law which are otherwise applicable where it is in the opinion of the local court appropriate to do so. Thirdly, if one accepts that the general approach of the majority in Investec is right, and the courts of a small jurisdiction should develop their rules of private international law in a way which is coherent with their domestic laws, a statute should provide that the court has authority to derive a rule of its private international law, which it considers necessary for the decision in a matter before it, by applying domestic law (written or unwritten) by analogy, where there is no other written law and it is, in the opinion of the court, appropriate to do so. A short statute in those terms would have provided the sure foundation for what the local courts were seeking to achieve in practically all the cases mentioned. It would reset private international law away from the default application of English rules, which are increasingly distant, in favour of a more internally coherent formulation of the state’s own laws. Fourthly, very serious consideration should be given to ending appeals to the Privy Council. It sometimes seems that several jurisdictions are on the verge of taking this step, or trying to take it; and once they are gone, they will be gone for good or ill. In place of appeals, there could be, and perhaps should be, a mechanism rather closer in effect to the procedure for obtaining an advisory or interpretive ruling on a point of law.60 Despite the arguments made here, that there are as many common laws as there are common law jurisdictions, it may still be useful for the courts

59 No position is taken on whether this would also have been the case when colonial law-making had a larger Westminster involvement. 60 Various models exist. The most obvious inspiration is the procedure by which the Court of Justice of the European Union may be asked to make rulings on the meaning and interpretation of rules of European law, leaving it to the referring court to deal with the rest of the case in the light of the ruling. Another model is that of stating a case for the opinion of another (superior) court on a question of law.

Private International Law and the Privy Council 137 in a small jurisdiction to be able to refer a question of legal interpretation to the Privy Council for ruling or answer. The laws of the state in question may be, as they were in Al Sabah v Grupo Torras SA, complicated (being in large part made externally, and simply landing or washing up on the islands) and hard to understand. It may be that the court wishes to know how the Privy Council would interpret the current state of common law: if that is what it wants to know, why should it not ask and be told? It may be possible to dispense with the need for an oral hearing before the Privy Council, not least because the final decision, on the application of the law to the facts, should be taken by the national court and not in London. There would be both negative and positive advantages to this. The negative advantages would be a reduction in the cost and inconvenience of a second or even third appeal,61 and the alleviation of the sense that the debate before the Privy Council in Westminster is just too remote from the jurisdiction from which the question comes. The positive advantage would be that the Privy Council would assume the role for which it is uniquely qualified, namely the answering of purely legal questions which a national court considers necessary to allow it to do its job. In 2015, Lady Hale said:62 It is not for the courts of England and Wales to interpret the law of the Channel Islands or decide what the law is there. In so far as that task rests with the courts, it rests with the Island courts, culminating ultimately in the Judicial Committee of the Privy Council.

That makes perfect sense today;63 the question for tomorrow is whether ‘culminating ultimately’ is good enough or should be replaced by ‘with, if they ask for it, the advice and assistance of the Judicial Committee of the Privy Council’. After all, if something like this is not considered, more of these jurisdictions will simply abolish appeals to the Privy Council and the cord will be cut, never to be re-tied; it is hard to see the gain, but easy to see the loss, if this were to happen.64 Time will tell; but if the English legal system is to rediscover its common law roots, as it puts its European years behind it, it makes sense to ask, with a particular focus on these smaller common law jurisdictions, what the role of English law, and of the Privy Council, should be today. The submission made here is that the legal expertise of the Privy Council should be available to the courts of these jurisdictions, but by entertaining and answering requests for a ruling or interpretation sent up by the courts of these jurisdictions; and that it should generally leave to the local judges the ultimate issue of how the facts fit its answer and its answer fits the facts. This would reflect the true reality that the scattered jurisdictions of the common law are related to each other, but are now masters in their own houses. The greatest service which the Privy Council can offer them, and those who litigate in them, is to be on hand to answer their questions rather than to overrule their decisions; and if this is not taken in hand soon, it may be too late.

61 Which, in some jurisdictions, lies to the Privy Council as of right even though the sum at stake may be as little as £500: see Court of Appeal (Guernsey) Law 1961, s 16, as interpreted in A v R (n 37). That this should be the law makes no sense at all. 62 R (Barclay) v Lord Chancellor & Secretary of State for Justice (No 2) [2014] UKSC 54, [2015] AC 276. 63 Although after Investec one might also say that it does not rest with the Island courts, culminating ultimately with the Judicial Committee of the Privy Council, to interpret the law of England and Wales or decide what the law is there. 64 Whether jurisdictions which had abolished appeals to the Privy Council might reinstate this degree of association is an interesting question.

138

8 Bulk Goods and Title Transfer MICHAEL BRIDGE

I. Introduction This chapter is concerned with one of the exceptions to the rule of nemo dat quod non habet, namely, the power of a seller, left in possession of goods or documents of title to goods, to t ransmit a good title to a second buyer notwithstanding the passing of property to the first buyer.1 It does not seek to give a full account of the seller in possession exception,2 but its essential character will be presented before attention is turned to the way that the exception might apply in connection with reforming legislation enacted in 1995.3 This legislation, as will be demonstrated, has a part to play where a seller has disposed of an undivided share in bulk goods to a first buyer before dealing with the same undivided share, or a part thereof, in favour of a second buyer or pledgee. Prior to the 1995 amendments to the Sale of Goods Act 1979 (SGA), section 16 of the Act, first enacted in 1893, stated without qualification that the property in unascertained goods could not pass to the buyer unless and until the goods were ascertained. This bar constituted an exception to the well-entrenched rule that otherwise the passing of property in the goods was a matter for the joint intention of the contracting parties.4 As a result of the 1995 amendments, this rule in section 16 is now a qualified one. It is subject to section 20A of the SGA, which provides a means of acquiring a property interest in goods by way of an undivided share, even whilst the goods remain unascertained. A lingering attachment to the notion of ascertainment nevertheless continues in so far as the unascertained goods in question must exist as part of a larger ascertained whole, namely an identified bulk. As will be seen, a question of critical importance is whether the property right to an undivided share created under section 20A is a ‘property’ in the goods for the purpose of the SGA. The SGA does not provide a comprehensive definition of property; rather, it asserts that property means the ‘general property’ and excludes from

1 Sale of Goods Act (SGA) 1979, s 24; Factors Act 1889, s 8. 2 For example, it does not deal with the question whether a delivery of goods under a bailment with a licence to consume (see PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2016] UKSC 23, [2016] AC 1034) amounts to a ‘disposition’ under s 24. 3 Sale of Goods (Amendment) Act 1995, adding ss 20A–20B to the SGA 1979 and making other consequential amendments to the Act. 4 SGA 1979, ss 17–19.

140 Michael Bridge the statutory meaning of property the ‘special property’, which is a limited interest of the type granted to a bailee.5 The general property is understood to mean ownership, and recent consumer legislation explicitly provides that to be so.6 The bar to the acquisition of a general property in unascertained goods, exemplified by section 16, had for decades proved to be commercially inconvenient when it came to dealings in quantities of goods that were part of a bulk – even an identified bulk. The same could not be said for the acquisition of shares in goods of the sort that arise in the case of ships and racehorses. These have been classified as items of intangible property,7 and in this respect fall in the same category as shares in a company, which give shareholders no proprietary interest in the physical assets owned by the company.8 As shares in goods are not themselves goods but instead items of intangible property, dealings in them, prior to the 1995 amendments, subject to one exception, fell outside the SGA. That exception concerned an extension of the SGA, in one minor respect, to contracts for the transfer of one co-owner’s interest to another, which was treated as a contract of sale for the purposes of the Act.9 This provision did not concern the proprietary aspect of that transaction, nor did it speak in general terms of co-ownership, whether by way of joint tenancy or tenancy in common. Prior to the 1995 legislation, an unavailing attempt was made to invoke the aid of equity to invest a buyer making payment for goods with an interest in the nature of an equitable lien.10 Later attempts, equally unsuccessful, were made to treat the buyer as the beneficiary of a trust.11 It had long been recognised that the SGA did not preclude the creation of an express trust,12 though such was unlikely to represent the intention of buyer and seller in an ordinary commercial contract of sale. These attempts foundered because of their failure to satisfy two critical requirements of a trust, namely, the intention to settle property on a trust13 and the certainty of subject matter needed for a trust to be properly constituted. In this latter respect, equity’s insistence on certainty mirrored the requirement in section 16 of the SGA that the goods be ascertained before the general property could pass. The later judicial recognition, in the case of fungible assets,14 that a trust might be created over a portion of the whole even in the absence of segregation15 – which, of course, corresponds to ascertainment – was the trusts equivalent of the statutory reforms of 1995 in the case of sale.16 At about the same time as an attempt was made in Re Wait to show that interests imposed by equitable lien survived the first codification of sales law in 1893, an attempt was made in another case to show that a legal property interest by way of tenancy in common could arise between

5 SGA 1979, s 61(1). 6 Consumer Rights Act 2015, s 4(1). 7 Re Sugar Properties (Derisley Wood) Ltd [1988] BCLC 146 (Ch). 8 Prest v Petrodel Resources Ltd [2012] EWCA Civ 1395, [2013] 2 WLR 557 [71], [94]–[97]; [2013] UKSC 34, [2013] 2 AC 415 [8]. 9 SGA 1979, s 2(2). 10 Re Wait [1927] 1 Ch 606 (CA). Expressing it as an equitable lien is more accurate than a charge because a charge is a consensual creation and the point about the interest claimed on behalf of the buyer was that it was an interest arising by operation of law. 11 Re London Wine Co (Shippers) Ltd [1986] PCC 121 (Ch); Re Goldcorp Exchange Ltd [1995] 1 AC 74 (PC). 12 Re Wait (n 10). 13 Re Goldcorp Exchange (n 11). 14 On the nature of fungibility, see RM Goode, ‘Are Intangible Assets Fungible?’ [2003] Lloyd’s Maritime and C ommercial Law Quarterly 379. 15 Hunter v Moss [1993] 1 WLR 934 (Ch), [1994] 1 WLR 452 (CA). 16 See MG Bridge, ‘Certainty, Identification and Intention in Personal Property Law’ in PS Davies and J Penner (eds), Equity, Trusts and Commerce (Oxford, Hart Publishing, 2017).

Bulk Goods and Title Transfer 141 seller and buyer in the case of goods stored with a third party warehouse. This attempt failed, though for reasons that were not fully expressed.17 One plausible reason was that an attempt to pass the property in unascertained goods, barred by section 16, ought not to be recharacterised as an attempt to pass a share of the ownership of bulk goods. Arguments for a tenancy in common returned again before the courts just before the enactment of the 1995 reforms, but this was in a case where the buyer of goods was considered to have acquired the general property in them prior to their loss of identity when mixed with goods of the same kind in a larger whole.18 It was not a case of co-ownership rights preceding the acquisition of the general property. Although the expression tenancy in common is not used in the 1995 reforms, it is precisely this type of legal interest, as opposed to an equitable interest, that the buyer acquires in the circumstances provided for by the reforms.

II. The Seller in Possession Before we turn to the title transfer issues presented by the 1995 reforms, a preliminary treatment of the rules of title transfer, particularly the so-called seller in possession exception to the rule of nemo dat quod non habet, is in order. As a further preliminary, the distinction between property and title should be observed. ‘Property’ is what passes between seller and buyer, and amounts to the divestiture by the seller of his ownership rights in the goods – though in some cases the seller may retain a possessory interest, either under the terms of a bailment, as occurs with a sale and leaseback, or in the exercise of a lien against payment of the price. ‘Title’ is an expression that measures the relative strength of two or more competing ownership interests. The buyer’s failure of consideration claim in the well-known case of Rowland v Divall19 was made, not because of the seller’s failure to pass the property in the goods, but because the seller’s title was defective, being inferior to that of the true owner. A seller may have, as against the true owner, an imperfect title; but such seller may nevertheless, in some circumstances, when selling the goods and investing in the buyer the property in them, pass on to the buyer a superior title to that of the earlier, true owner.20 It is common in American writing to refer to the property in the goods as title, which may be due to the general expulsion from Article 2 of the Uniform Commercial Code of the passing of property between seller and buyer. The same loose usage, encouraged by the widespread use of so-called retention of title clauses, sometimes creeps into English case law.21 The common law, whilst acknowledging the fundamental clash between the need to protect private property and the need to encourage commerce,22 has consistently favoured the former over the latter. It has never accepted a general exception to the rule that a non-owner is unable to transfer a title superior to his own just because a transferee for value is acting in good faith.23 Instead, it accepted that a transferee could defeat the claims of an owner, both duped by the dishonesty of a rogue, in those cases where the owner represented by words or conduct that the

17 Laurie & Morewood v Dudin & Sons [1926] 1 KB 223 (CA). 18 Re Stapylton Fletcher Ltd [1994] 1 WLR 1181 (Ch). 19 Rowland v Divall [1923] 2 KB 500 (CA). 20 SGA 1979, s 24; Factors Act 1889, s 8. 21 The Res Cogitans (n 2). 22 Bishopsgate Motor Finance Corp v Transport Brakes Ltd [1949] 1 KB 322 (CA) 336–37. 23 The closest the law came was an exception for sales in market overt – now abolished by the Sale of Goods (Amendment) Act 1994 – where the circumstances of the sale were tightly defined as to both time and place.

142 Michael Bridge rogue was either the owner’s authorised agent (apparent authority)24 or was instead the true owner (apparent ownership).25 In both instances, the common law rules had, in the case of mercantile agency, been extended by factors legislation so as to remove any true requirement of appearance of ownership or agency. Transferees dealing with a mercantile agent acting in the ordinary course of business of a mercantile agent could acquire a title at the expense of the true owner where the owner entrusted goods or documents of title to a person with the status of a mercantile agent. This was regardless of whether the transferee knew or ought to have known that he was dealing with a mercantile agent,26 and regardless of whether anything had been done by the true owner over and above the mere transfer of possession. In 1877, the shortcomings of factors legislation were exposed in a case, Johnson v Credit Lyonnais Co,27 concerning a merchant selling tobacco who was allowed by the buyer to remain in possession of warehouse warrants. The buyer, by his own account unaware of the practice of issuing warrants, left the tobacco in bond in the name of the seller so as to delay the payment of excise duty. The seller then fraudulently pledged the warrants in return for advances. Between the first instance and Court of Appeal judgments, the legislature bestirred itself to pass what later became section 8 of the Factors Act 1889,28 a provision for the most part replicated in section 24 of the SGA. This, of course, did not affect the outcome in the Johnson case, which concluded in favour of the buyer. The seller had not been entrusted by the buyer with the goods or documents of title representing them in order to deal as an agent, nor could it be said that the buyer’s conduct amounted to a representation that the seller was the owner of the tobacco. As regards the first of these two points, such agency capacity as the seller possessed was merely to hold the goods in custody.29 Bramwell LJ carried out an extensive analysis of the various Factors Acts in order to reach the conclusion that the entrustment of goods or documents of title to a mercantile agent had to be an entrustment in his capacity as a mercantile agent. This, of course, would be invisible to the pledgee or anyone else dealing with the seller in possession. It was the second point, concerning the buyer’s conduct, that caused the court the greater concern: when the question arises as to which of [the buyer and the pledgee] the loss should fall upon, in reason and justice the loss ought to fall on [the buyer] who might have prevented, and as a matter of common prudence ought to have prevented, the possibility of the fraud, is what I cannot bring myself to doubt.30

This assertion echoes the celebrated dictum (and badge of a losing cause) of Ashhurst J in Lickbarrow v Mason – a case not cited to the court or referred to in the judgments – that, of two innocent persons, the one who enabled the loss should bear it.31 The court, nevertheless and with some regret, found against the pledgee despite the buyer’s failure to take earlier steps to prevent the seller’s fraud. There was no representation of ownership to be found in the passive behaviour of the buyer in leaving the seller’s name as owner in the books of the bonded warehouse. It may have been market practice for the buyer here to behave as it did, but this did not absolve the buyer from the criticism that he had behaved negligently. Nevertheless, though the buyer may have

24 Very

difficult to establish: eg Farquharson Bros & Co v King & Co [1902] AC 325 (HL). Distributors Ltd v Goldring [1957] 2 QB 600 (CA). 26 Oppenheimer v Attenborough & Son [1908] 1 KB 221 (CA). 27 Johnson v Credit Lyonnais Co (1877) 3 CPD 32 (CA). 28 The antecedent s 3 of the Factors Act 1877 was limited to documents of title. 29 Johnson (n 27) 50 (Brett LJ). 30 ibid 36 (Cockburn CJ). 31 Lickbarrow v Mason (1787) 2 TR 63, 70; 100 ER 35, 39. 25 Eastern

Bulk Goods and Title Transfer 143 been negligent, that negligence did not count against him in the absence of a duty owed to the pledgee or members of the general public. The seller in possession exception to the rule of nemo dat was therefore enacted and grafted on to a body of common law and statute in order to reverse the first instance result in the Johnson case. It did not rely upon the greater negligence of buyers in suffering sellers to remain in possession than the negligence of persons lending goods to or storing goods with others, or upon the more potent appearance of ownership or authority to act of someone who has sold the goods as opposed to someone to whom goods have been bailed. It was enough that the seller, for whatever reason, remained in possession of the goods or documents of title to the goods after the sale to the first buyer. A most striking feature of the legislation is that it makes no reference to the seller continuing in possession with the consent of the buyer. Consequently, a seller who in breach of contract fails to deliver the goods, even one who is pressed by the buyer to make delivery, has the power to transmit a good title to a second buyer.32 It is not as if the buyer can be faulted for failing to take the necessary steps to separate the seller from possession, thus contributing to the deception practised by the seller on the second buyer. Yet, the merits of someone dealing with a seller in possession are no greater in the absence of special circumstances than those who deal with a mere bailee or even a thief. Nevertheless, the adoption of the seller in possession exception was said by Lord Pearce in Pacific Motor Auctions Pty Ltd v Motor Credits (Hire Finance) Pty Ltd to be ‘intended as a protection to innocent purchasers in cases where estoppel gave insufficient protection’.33 This may have been so, but there are numerous other cases where estoppel continues to give insufficient protection to innocent purchasers. All of this merely tells us that we have what we have. We have started with the rule of nemo dat and have grafted on to it a limited common law rule of estoppel. The negligence of owners did not serve to ground a duty to innocent purchasers: the Johnson case was decided even before the premature attempt by Brett MR to articulate in Heaven v Pender34 a general basis for duties of care in the years preceding Donoghue v Stevenson.35 The Johnson case, along with the later restrictions imposed in modern times on liability in negligence for economic loss,36 as well as the abolition by statute of the rule of market overt, has meant that the law protecting innocent purchasers lacks rational coherence and can be explained only as piecemeal responses to particular cases, for which the law has been rightly criticised.37 Good faith purchasers are not protected because they have put their faith in appearances, but rather because they are dealing with a seller who has the status of a mercantile agent, or of a seller or buyer in possession, a fact of which the good faith

32 See Worcester Works Finance Ltd v Cooden Engineering Co Ltd [1972] 1 QB 210 (CA) 217: ‘It does not matter what private arrangements may be made by the seller with the customer – such as whether the seller remains bailee or trespasser or whether he is lawfully in possession or not.’ 33 Pacific Motor Auctions Pty Ltd v Motor Credits (Hire Finance) Ltd [1965] AC 867 (PC) 883. The language of the section is defective in giving rise to the argument that all it does is to render lawful the transfer of possession of the goods or documents of title by the seller (see L Rutherford and I Todd, ‘Section 25(1) of the Sale of Goods Act 1893: The Reluctance to Create a Mercantile Agency’ [1979] CLJ 346), but this very limited measure does not even begin to respond to the problem the legislation was designed to redress or explain why the legislature acted as quickly as it did in response to the first instance decision in Johnson. 34 Heaven v Pender (t/a West India Graving Dock Co) (1883) 11 QBD 503 (CA) 509: ‘The proposition which these recognised cases suggest, and which is, therefore, to be deduced from them, is that whenever one person is by circumstances placed in such a position with regard to another that every one of ordinary sense who did think would at once recognise that if he did not use ordinary care and skill in his own conduct with regard to those circumstances he would cause danger of injury to the person or property of the other, a duty arises to use ordinary care and skill to avoid such danger.’ 35 Donoghue v Stevenson [1932] AC 562 (HL). 36 eg Moorgate Mercantile Credit Co Ltd v Twitchings [1977] AC 890 (HL). 37 A Tettenborn, ‘Transfer of Chattels by Non-owners: Still an Open Problem’ [2018] CLJ 151.

144 Michael Bridge purchaser may be wholly ignorant. The character of the law thus revealed, which sometimes leads to capricious outcomes,38 bears upon the way we apply the seller in possession exception to cases involving bulk goods, to which we now turn.

III. Bulk Goods and Oversales Beginning with a simple case, let us take the example of a seller who is the owner of a 30,000 gallon bulk and who contracts to sell 15,000 gallons to each of A and B. The seller receives payment for the same and the bulk has been identified by agreement in each of the contracts or at a subsequent date. By virtue of section 20A of the SGA 1979, both A and B will acquire rights in common to the 30,000 gallon bulk and will divide that bulk between them in equal shares. Suppose now that the seller, who may well be in possession of a bill of lading, a warehouse receipt or a warrant in respect of that bulk, contracts to sell 15,000 gallons to C, the bulk again being identified and payment being made. Let us now take section 20A(4) and consider its literal application to the case in hand. It provides: Where the aggregate of the undivided shares of buyers in a bulk … would at any time exceed the whole of the bulk at that time, the undivided share in the bulk of each buyer shall be reduced proportionately so that the aggregate of the undivided share is equal to the whole of the bulk.

On one reading, A, B and C should be treated as owners of the bulk and, since their entitlements are quantitatively equal, they should each receive 10,000 gallons after a pro rata reduction. The correctness or not of that conclusion may be assessed by working through a number of illustrative cases. Suppose first that C had purchased the whole of that bulk in circumstances compliant with section 24 of the SGA.39 C, purchasing from a seller in possession, would not share the bulk with A and B, but would acquire a good title to the whole. The same result in favour of C would be reached if A and B were replaced by the compound person AB, who purchases the entire bulk whilst leaving the seller in possession of the bill of lading or other document representing the goods. The next case, which is the case mentioned above, is where C purchases only a 15,000 gallon part of the 30,000 bulk. This might occur in two different sets of circumstances: (i) where A and B have each purchased 15,000 gallons and have acquired part interests in the bulk; and (ii) where AB has purchased the entire 30,000 gallon bulk. The question now is whether C acquires a good title in this case, not just by virtue of section 24 of the SGA, but also by virtue of section 20A, so as to become a rateable sharer with A and B or AB. Some light on this issue can be cast by reviewing two buyer in possession decisions preceding section 20A that dealt with the application of exceptions to the rule of nemo dat in favour of purchasers and pledgees of undivided shares in bulk goods. In Ant Jurgens Margarinefabrieken v Louis Dreyfus & Co,40 a buyer in possession case, the seller of 2,600 bags of mowra seed, out of a larger quantity of 6,400 bags on board a named vessel,

38 The 39 For

40 Ant

best example is Worcester Works Finance Ltd v Cooden Engineering Co Ltd (n 32). ease of exposition, the parallel provision of the Factors Act 1889, s 8 will not be mentioned hereinafter. Jurgens Margarinefabrieken v Louis Dreyfus & Co [1914] 3 KB 40 (KB).

Bulk Goods and Title Transfer 145 gave delivery orders adding up to the contract quantity addressed to the seller’s Hamburg establishment41 to the buyer,42 who paid by means of a cheque that was later dishonoured. The buyer indorsed the delivery orders in favour of a sub-buyer, but the seller put a stop on delivery of the goods. The decision in favour of the sub-buyer was given under section 47 of the SGA, which provides that an unpaid seller’s lien or right of retention may not be asserted against a good faith purchaser when a document of title has been transferred to that person. The delivery orders here were documents of title further to the extended definition found in the Factors Act 1889 and incorporated by reference in the SGA. The effect of the decision that there was no lien was that the seller was bound to honour the delivery order, which meant that the seller was required to see to it that the sub-buyer’s goods were separated from bulk and delivered to it. The same result should have been reached if the alternative, buyer in possession route in section 25 of the SGA had been adopted instead, even though the decision does not as such say that, prior to ascertainment, the sub-buyer had a proprietary interest in the goods. This feature of the case was not developed in the judgment, which was largely concerned with concluding that a document of title created by the seller could be said to be ‘transferred’ to the buyer, an issue that is not germane to the matter under discussion here. In Capital & Counties Bank Ltd v Warriner,43 section 24 of the SGA, or rather section 9 of the Factors Act 1889, was more directly engaged. A warehouse, faced with claims by an unpaid seller and a bank to which warehouse warrants had been pledged, interpleaded. The buyer of a quantity of Azima wheat on credit – ex Duke of Cornwall and stored onshore as part of a larger bulk of 2,918 quarters – had failed before it paid the seller. The seller had issued a delivery order to the warehouse for the contract quantity and the warehouse had first attorned to the buyer before issuing three warrants, amounting to 1,400 quarters, in the buyer’s favour. The buyer had then pledged the warrants with the bank as security for an overdraft. Despite the seller’s argument that the pledged quantity had never been separated from the larger bulk, the court found in favour of the bank. It was sufficient in the circumstances that the buyer had bought or agreed to buy the contract quantity. Had it been necessary to do so, the court would have found in favour of the bank on the ground of an estoppel: the seller must have known that the warrants would have been issued and used in the ordinary course of business. These two judgments are both brief and both at first instance, but they have stood for more than a hundred years. It should follow a fortiori that good faith purchasers claiming the protection of the seller or buyer in possession exceptions to the rule of nemo dat should be in a stronger position if they have an undivided share in the goods as a result of section 20A. The position here is not complicated by the existence of parallel provisions in section 24 of the SGA and section 8 of the Factors Act 1889. Taking the language of section 24, suppose that a buyer has acquired a share in a bulk under section 20A but leaves the seller either in possession of the bulk or in possession of a document of title, such as a warehouse receipt or warrant, representing either the whole bulk or the contract quantity. The seller now enters into a transaction that amounts to an ‘oversale’ in that the seller has either no, or no sufficient, proprietary interest in the bulk to carry out lawfully the second transaction. Section 24 applies in the case of someone who has ‘sold’ goods and who then enters into a ‘sale, pledge or other disposition’ of the goods.

41 The goods had been carried on ex ship Hamburg terms. 42 The question whether the buyer was acting as a principal in relation to the seller and sub-buyer or as an agent establishing privity of contract between them was treated as immaterial. 43 Capital & Counties Bank Ltd v Warriner-Bretherton Ford & Co (1896) 21 Com Cas 314, (1896) 12 TLR 216 (QB). This account of the case draws from both reports.

146 Michael Bridge Given the breadth of the word ‘disposition’, the transfer of an undivided share to the second buyer should certainly fall within the terms of section 24,44 but has the seller in this case ‘sold’ the goods to the first buyer or buyers? If the 30,000 gallons in our example have been sold to the compound person AB, then the bulk has indeed been sold to that person. More difficult is the case where there are two initial buyers, A and B, each acquiring an undivided share. Here, the transactions in which they take part have to be sales and not merely dispositions. According to section 2(1) of the SGA, a contract of sale of goods is one in which the seller ‘transfers or agrees to transfer the property in goods’ to the buyer. The property, as we have seen, is laconically defined in the SGA as the general property. This excludes a special property but does not explicitly exclude an undivided share, which is more than a mere possessory interest. A definition like this is more significant for what it excludes rather than for what it might include. Yet it is certainly arguable that an interpretation of ‘sale’ to include the transfer of an undivided share would collapse the distinction between a contract for the sale of goods and a contract for the sale of a share in goods, and thus amount to a collapsing of the distinction between the tangible (the goods themselves) and the intangible (shares in the goods). The SGA, as amended in 1995, does indeed go some of the way towards collapsing these distinctions in that it now defines ‘goods’ so as to include ‘an undivided share in goods’.45 The legislation nevertheless continues to make the distinction between a contract for the sale of a quantity of goods in a bulk and one for the sale of a share of the bulk.46 All the legislation does is subsequently recharacterise the subject matter of the contract of sale of goods once the conditions laid down by section 20A have been met. Moreover, the statutory definition of ‘sale’ has not been amended and the Act continues unhelpfully to define sale in terms of the archaic distinction between the two common counts of goods bargained and sold and goods sold and delivered. The definition was thus designed to serve a procedural purpose only in identifying those instances where a cause of action for the recovery of a debt was permitted. It does not address the question whether the general property and a part interest may both be the subject of a sale. Yet, if ‘sale’ were to be given a narrow interpretation under section 2(1), this would then jar against section 2(2), which provides, admittedly in non-proprietary terms, that there can be a contract of sale between one part owner and another. Section 2(2) does not explain how the purchasing part owner acquires ownership of the whole, or how one of more than two part owners acquires an enhanced share from one of the remaining co-owners. Taking the latter of these two instances, where the enhanced undivided share may continue indefinitely without partition, it all adds up to a strong case for a liberal interpretation of ‘sale’ so as to include the transfer of an undivided share in goods. The above problem concerning the meaning of ‘sale’ does not arise in a case concerning the buyer in possession exception to the rule of nemo dat in section 25 of the SGA. Suppose that an unpaid seller has armed the buyer with a document of title relating to a quantity of goods in an identified bulk but the buyer is not to acquire a proprietary interest in the goods until payment has been made. The buyer now concludes a contract with a sub-buyer on terms that comply with section 20A. It does not matter, as we have seen, that for present purposes the buyer lacks a good ‘title’, for the buyer is still able to transfer a ‘property’ right in the form of an undivided share under section 20A. Moreover, the contract between seller and buyer contemplates the eventual sale of the contract goods to the buyer so that the buyer qualifies as one who has ‘bought or

44 See Worcester Works Finance (n 32) 218, especially Lord Denning MR’s somewhat extravagant interpretation of that word, which is supported by P4 Ltd v Unite Integrated Solutions plc [2006] EWHC 2640 (TCC) [115]. 45 SGA 1979, s 61(1). See also Law Commission, Sale of Goods Forming Part of a Bulk (Law Com No 215, 1993) para 5.3. 46 SGA 1979, s 20A(1). See also Law Com No 215 (n 45) para 5.2.

Bulk Goods and Title Transfer 147 agreed to buy the goods’ under section 25(1). The transaction concluded by the buyer in possession with the sub-buyer, the effect of which is to transfer an undivided share to the sub-buyer, certainly amounts to a ‘disposition’ under section 25(1),47 so the sub-buyer should acquire that interest at the expense of the seller. The title acquired by an oversale buyer purchasing an undivided interest in a bulk was considered cursorily by the Law Commission, which concluded that a good title to an undivided share could be acquired by means of section 24.48 The Law Commission was of the view that the oversale buyer could not acquire a good title by virtue of section 20B, which deals with deemed consent to removals from bulk by the owners of undivided shares in that bulk,49 but no reasons were given for this view. Section 20B provides that those who acquire a share of the bulk under section 20A are deemed to have consented to a delivery out of the bulk to any other owner in common of ‘goods which are due to him under his contract’. Does an oversale buyer become an owner in common by virtue of section 20A, and more accurately, by section 20A alone? In terms of ‘property’ passing between seller and buyer, the oversale buyer does become an owner in common under section 20A. As a matter of ‘title’, however, the oversale buyer does not, because section 20A is not in that part of the SGA that deals with title transfer. Even if section 20B did apply in favour of the oversale buyer, that buyer would acquire only an immunity from suit. The other co-owners give their consent to the delivery, not to the acquisition of a good title,50 and for that reason have no cause of action against the oversale buyer.51 That buyer is not in the position of a person rendered immune from suit under the Limitation Act 1980 who, after the limitation period has expired, also acquires a good title to the goods.52 Consequently, that buyer would be in a precarious position if seeking to sell the goods on after delivery to a subsequent buyer. The prospect of liability under section 12(1) of the SGA for lack of a right to sell to that subsequent buyer would bulk large. In respect of the title acquired by the oversale buyer, the Law Commission was content simply to say: ‘The policy of protecting the good faith buyer under section 24 is of long standing and we see no reason why it should not apply in the new situation.’53 It should not be supposed, however, that what the Law Commission would like to see and what ensuing legislation provides for are the same thing.54 The Law Commission, having articulated the effect of section 24 on the transfer of undivided shares, then immediately undermined that conclusion by asserting that the oversale buyer’s share of the bulk should be pro rata reduced alongside the reduction meted out to the existing co-owners as a result of the oversale.55 This conclusion further sits uncomfortably with the view that section 20B may not be employed to validate the oversale buyer’s removal of the contract goods from the bulk. If an oversale buyer’s share, at the point of acquisition, is simultaneously diminished by the oversale, this must be by virtue of section 20A, at least in part.

47 It is tempting to say that if what happened in Worcester Works Finance (n 32) was a disposition, then almost anything could be. 48 Law Com No 215 (n 45) para 4.18. 49 ibid. 50 SGA 1979, s 20B(1)(a). 51 SGA 1979, s 20B(2), (3)(a). If the oversale buyer in turn sold on to a sub-buyer, the immunity from suit that appears to enure in favour of the sub-buyer (SGA 1979, s 20B(1)(b)) would not preclude an action against the oversale buyer under the SGA 1979, s 12(1) (right to sell). 52 SGA 1979, s 3(2). 53 Law Com No 215 (n 45) para 4.18. 54 See MG Bridge, ‘Risk, Property and Bulk Goods in International Sales’ [2019] Lloyd’s Maritime and Commercial Law Quarterly 57. 55 Law Com No 215 (n 45) para 4.18.

148 Michael Bridge It seems that, if the Law Commission’s view is correct, the oversale buyer acquires rights by virtue of both section 20A and section 24. So, is the Law Commission right, in light of the 1995 amendments to the SGA, to conclude that the oversale buyer’s share is simultaneously reduced? There is no warrant for this in the legislation. The notion of an overriding title being diminished at the point of its acquisition calls for critical evaluation. Either section 24 applies or it does not. There is no more room for a halfway house than there is for a general splitting of the loss between owner and innocent transferee advocated by Devlin LJ nearly 60 years ago.56 A counter-argument to this view of section 24’s effect is that the notion of a property interest diminished at the point of acquisition is recognised elsewhere in the law. This can be seen where competing equitable interests, arising at different intervals in a dynamic securities account, have to be reconciled when the fund is distributed. The orthodox view is that the competing entitlements are computed at the end of the day when the distribution takes place.57 If the fund is already in deficit when an investor or other claimant acquires an interest in its contents, this deficit is charged against that person at the point the interest is acquired and is then carried forward, increasing or diminishing, as the fund is managed before the distribution date. An alternative approach, based on the notion of a so-called ‘rolling charge’, would, by calculating entitlements from day to day, mean that there is no initial diminution at the point of acquisition since only subsequent losses would be charged against a particular investor or claimant. Now, an approach deemed appropriate for dealing with equitable interests is not appropriate for legal interests regulated, as in the case of section 24, by statute. Section 24 provides no warrant for such an approach. Moreover, even outside the statute, it has been rejected in a case where a warehouse contained a shortfall of goods in which various banks had competing trust receipt interests of a legal nature58 arising out of transactions conducted at different times. The Singapore court, in that case,59 opted for the rolling charge approach, so that each bank’s entitlement was calculated from day to day and no bank suffered the effect of a prior deficit. Applying the favoured view stated above, as a result of the oversale, and at that point and not before, the oversale buyer joins the community of co-owners and should only be vulnerable in its turn to subsequent oversales. If there is an objection to the section 24 approach that I propose, it has to be found in section 20A, particularly in sub-section (4), which provides for a pro rata reduction of entitlements ‘[w]here the aggregate of undivided shares in a bulk … would at any time exceed the whole of the bulk at that time’. I have criticised elsewhere the vagueness of this temporal reference in an article dealing with the rights of a seller with a retained share and buyers, and shall not repeat the criticism here.60 The point I want to make now is that there is no qualifying language in section 24 that would reduce the authority deemed to be given by the co-owning buyers to the seller’s oversale. Even if section 20A could somehow be blended with section 24 in an impressionistic way, let us not forget that there is also the seller in possession provision in section 8 of the Factors Act 1889. As much as courts have advocated treating the title transfer provisions of the Factors Act 1889 and the SGA as a complete code,61 it is too much of a 56 Ingram v Little [1961] 1 QB 31 (CA) 73–74, a suggestion that was roundly rejected by the Law Reform Committee in its Twelfth Report (Transfer of Title to Chattels) (Cmnd 2958, 1966) para 15. 57 Barlow Clowes International Ltd v Vaughan [1992] 4 All ER 22 (CA). 58 The goods in question were held by the warehouse under a deemed continuing pledge (a legal form of security) granted in favour of the banks. 59 Pars Ram Brothers (Pte) Ltd v Australian & New Zealand Banking Group Ltd [2018] SGHC 60, [2018] 4 SLR 1404. 60 See Bridge (n 54) 69–72. 61 Worcester Works Finance (n 32) 220; Forsyth International (UK) Ltd v Silver Shipping Co Ltd (The Saetta) [1994] 1 WLR 1334 (QB) 1344.

Bulk Goods and Title Transfer 149 stretch to blend a provision in an 1889 Act and 1995 statutory amendments. It might make some practical sense for this to be done, but we have the legislation that we have. The problem raised here, which may well happen only very infrequently in practice, is the result of a statutory intervention that implanted new provisions in the SGA without any systematic account being taken of the overall property and title picture (to which may also be added the allocation of risk).62 Selective statutory reform, especially of a statute of long duration, is dangerous statutory reform. At this point, it is useful to provide a summative example. The seller-cum-owner of a 30,000 gallon bulk grants undivided shares of 15,000 gallons to each of A and B, before subsequently granting in an oversale a 15,000 gallon share to C. That should leave A and B entitled each to 7,500 gallons and C to 15,000 gallons. Should there be any subsequent shrinkage in the bulk after C, the first oversale buyer, has intervened, then it should distributed among the co-owners in the usual pro rata way.63 That would be the solution provided by ordinary tenancy in common principles. It is also explicitly provided for by section 20A(4), which should apply to the relations of all co-owning buyers, regardless of whether they need to rely upon section 24 as against earlier buyers. Hence, if there is a second oversale, by which the seller then grants 12,000 gallons to D, A’s and B’s shares would be reduced to 4,500 gallons and C’s to 9,000 gallons, leaving D with its 12,000 gallons. Turning now to shrinkage and the relations of co-owners, whether they have acquired an interest with the aid of section 24 or not, there is section 20B to consider. The allocation of shrinkage under section 20A(4) is substantially offset by section 20B, according to which a person who has become an owner in common of the bulk ‘by virtue of section 20A’ is deemed to consent to removals by other co-owners in accordance with their contractual entitlements. Suppose now that a buyer with a rateable share of the bulk contracts to sell its contractual quantity to a subbuyer in circumstances compliant with section 20A, so that the sub-contract becomes a sub-sale. Does the consent by other co-owners to removals by sub-buyers of the full contractual amount in the sub-contract apply when the shrinkage in the bulk has already occurred by the time of the sub-sale? Consider the case where the contractual entitlement of the buyer is 10,000 gallons but where, as a result of shrinkage, its proprietary pro rata entitlement is 8,000 gallons. The subject matter of the sub-contract is 10,000 gallons. Suppose, too, that the buyer does not hold a document of title, as defined in the Factors Act 1889, at the time of the sub-sale, so that section 24 of the SGA is not engaged.64 It must be the case that, when it comes to applying section 20B to this subsale, the sub-buyer may not rely without restriction on the quantity of goods provided for in the sub-contract, or else the quantity might be expressed in whatever amount the parties provided, even an amount consuming the whole of the bulk. Assuming that between buyer and sub-buyer the requirements of section 20A are met, a sensible interpretation of section 20B would define the sub-buyer’s contractual entitlement in that section by reference to the buyer’s pro rata 8,000 gallon entitlement. The delivery referred to in section 20A(1)(a) includes an appropriation of the goods to the contract between buyer and sub-buyer, which crystallises the sub-buyer’s entitlement. Section 20A(1)(b), an alternative to section 20(1)(a), does not lend itself to the expanded assessment of the sub-buyer’s rights, dismissed above, since it would clearly limit the effect as

62 See Bridge (n 54). 63 Indian Oil Corp Ltd v Greenstone Shipping Co SA (Panama) (The Ypatianna) [1988] QB 345 (Com Ct). 64 It might be argued that a buyer acquiring an undivided share in the goods is in constructive possession of them by virtue of an attornment by the seller (or from a remote seller routed through a chain of sellers). There is nothing in the legislation to encourage this view and it is doubted in L Gullifer, ‘Constructive Possession after the Sale of Goods (Amendment) Act 1995’ [1999] Lloyd’s Maritime and Commercial Law Quarterly 93.

150 Michael Bridge against other co-owners of the delivery (or other dealing with the goods) to the buyer’s proprietary entitlement of 8,000 gallons. The final question that needs now to be considered concerns the effect of section 20B deliveries out of bulk as between the first buyers and the oversale buyer. May the first buyers, by taking delivery of their full contractual entitlements from the bulk, in effect reverse or reduce the application of section 24 in favour of the oversale buyer? The oversale buyer has become an owner in common of the bulk with the aid of section 24, but the question is whether that buyer also has need of section 20A. At first sight, the answer is yes, but this ignores earlier case law, the effect of which was to apply exceptions to the nemo dat rule in favour of buyers out of bulk.65 If oversale buyers did not need section 20A to acquire co-ownership rights, should it not follow that they are free from the risk of excessive deliveries out of the bulk to other co-owning buyers? On the other hand, if it can be said that section 20A has overtaken the earlier case law’s approach, then oversale buyers are exposed to these deliveries being made. On balance, the better view is that section 20B should be applicable at the expense of oversale buyers and that the rights of oversale buyers laid down in the Ant Jurgens66 and Capital & Counties Bank67 cases have been diminished. The reasons of convenience that lie behind section 20B,68 which include the avoidance of complex accounting exercises and the alleviation of burdens resting on warehouse keepers, carriers and company liquidators, are just as much relevant to a community of co-owners that includes an oversale buyer. That said, pro rata clauses in standard form sale contracts may provide solutions that may not accord with section 20B. A typical though somewhat obscurely expressed example is clause 22 of GAFTA 100, which in paragraph (b) makes provision for market rate adjustments between receivers of goods when one obtains more than his pro rata share.69

IV. Conclusion This has been a limited foray into the difficulties arising out of the 1995 changes to the SGA 1979. Detailed legislation of the type favoured by the Westminster draftsman may expose a problem that a common-sense interpretation of broad legislation might suppress. There is much to be said for the view that a simple statement of the buyer’s rights as a co-owner would have been preferable, along the lines of Article 2-105(4) of the Uniform Commercial Code,70 with the courts being left to determine the finer incidents of that co-ownership. This would in effect be putting off the day of resolution, but the difficulties that I have here and elsewhere identified with the legislation may never come to light. The 1995 changes have not been the subject yet of litigation in this country and may not become so in the foreseeable future.

65 Ant Jurgens (n 40); Capital & Counties Bank (n 43). 66 Ant Jurgens (n 40). 67 Capital & Counties Bank (n 43). 68 Law Com No 215 (n 45) para 4.17. 69 GAFTA 100, cl 22: ‘In the event of this [a receiver] receiving more or less than his pro-rata share or apportionment, he shall settle with the other(s) on a pro-rata basis in cash at the market price.’ It is not clear whether this provision applies to shrunken bulks in general or just to cases where a non-shrunken bulk has been inaccurately divided. GAFTA 100 is a standard form CIF contract for feeding stuffs in bulk that contemplates shipments destined for a number of different buyers. 70 UCC, Art 2-105(4): ‘An undivided share in an identified bulk of fungible goods is sufficiently identified to be sold although the quantity of the bulk is not determined. Any agreed proportion of such a bulk or any quantity thereof agreed upon by number, weight or other measure may to the extent of the seller’s interest in the bulk be sold to the buyer who then becomes an owner in common.’

9 The Duty to Take Delivery of Goods ANDREW TETTENBORN

I. Introduction ‘It is the duty of the seller,’ says section 27 of the Sale of Goods Act (SGA) 1979, ‘to deliver the goods.’ So far so good. But is there a corresponding duty in the buyer to take them, and if there is, what does it entail? That is a surprisingly awkward question; and it is the subject of this chapter. Other legal systems and instruments regard the point as obvious. Article 60 of the UN Convention on Contracts for the International Sale of Goods obliges a buyer not only physically to take over the goods, but goes further, requiring him to do ‘all the acts which could reasonably be expected of him in order to enable the seller to make delivery’.1 German law is equally blunt: the buyer ‘is obliged to pay the agreed price and to take the thing he has bought’.2 The International Chamber of Commerce’s Incoterms 2010 goes even further. In all of its 11 standard international sale contracts it does not simply oblige the buyer to take delivery, but in addition requires him, where appropriate, to collect the goods from the carrier who transports them to him, having taken them over from the seller.3 Curiously enough, however, the Sale of Goods Act (SGA) 1979 has no equivalent provision. The nearest it gets is one isolated sub-section, section 37(1), which runs as follows: When the seller is ready and willing to deliver the goods, and requests the buyer to take delivery, and the buyer does not within a reasonable time after such request take delivery of the goods, he is liable to the seller for any loss occasioned by his neglect or refusal to take delivery, and also for a reasonable charge for the care and custody of the goods.

This, it will be noticed, is highly restricted. It only deals with putting a foot-dragging buyer on notice, and not with the case where the contract provides a fixed date and time for d elivery. Moreover, it merely gives a power to award compensation and/or a warehousing charge. It does not even say in so many words that the buyer is guilty of a breach of contract (though it might be argued that this is implicit). Further, although it preserves whatever rights the seller

1 UNCITRAL, United Nations Convention on Contracts for the International Sale of Goods (CISG) (Vienna, 1980) Art 60. 2 BGB, § 433: ‘Der Käufer ist verpflichtet, dem Verkäufer den vereinbarten Kaufpreis zu zahlen und die gekaufte Sache abzunehmen’. 3 See Term B4 attached to each term in turn: ‘The buyer must take delivery of the goods when they have been delivered … and receive them from the carrier at the named port of destination.’ We return to the second part of this provision at section IV.D(ii) below.

152 Andrew Tettenborn otherwise has to treat non-take-up as a repudiation of the contract,4 it says nothing about any other consequences. It might be thought that the answer to the puzzle of the ‘take-up’ duty (or lack of it) in England lay in the second part of section 27 of the SGA 1979. This obliges a buyer of goods, as a corollary of the seller’s duty to deliver them, to ‘accept and pay for them, in accordance with the terms of the contract of sale’. Might this be read as incorporating a take-up duty?5 Take-up and acceptance are, after all, closely related: for example, if a buyer is not bound to accept goods (for instance, because they are defective), it stands to reason that he can equally have no duty to take delivery of them. But it is suggested that matters are not as simple as this. The converse does not apply: the presence of an obligation to accept does not automatically imply a take-up duty.6 True, some take-up failures will be breaches of section 27: if the seller of a car tenders it to the buyer in good time and condition, but the buyer declines the keys and documents and tells the seller to take the vehicle away, this must amount to non-acceptance. But not all will; it is, it is suggested, perfectly possible to accept goods without taking them over. Take another homely example. Imagine a buyer who chooses a satellite dish in a shop, pays for it, becomes owner and arranges for the dealer to hold onto it and install it in a week’s time. He later tells the seller not to bother because his wife disapproves, and adds that he is quite happy for the seller to keep both the price and the dish. He may be foolish or even feckless; but having paid for the dish on the shelf and become owner of it he cannot, it is suggested, be guilty of non-acceptance.

II. Is there a Free-Standing Take-up Duty to Accept Delivery? The question therefore remains. Outside the specific circumstances of section 37(1) of the SGA 1979, does the buyer have an implied duty to take up goods at the time and place specified for delivery, so that if he does this late or not at all he is in breach of contract? Two plausible lines of reasoning might actually suggest that the answer is ‘No’. One is a simple point of statutory interpretation: if the draftsman of the SGA 1979 imposed a restricted take-up duty under section 37(1), the courts should not second-guess this decision by conjuring up any more general duty free of the restrictions affecting that section. But this is unconvincing. Despite occasional suggestions to the contrary,7 the Sale of Goods Acts 1893 and 1979 are not so much codes as a compilation of instances taken from those pre-1893 cases that happened to attract Sir Mackenzie Chalmers’ notice; often, indeed, they look like little more than a hyper-organised student’s lecture notes. There are plenty of cases where courts have essentially recognised this point. They have, for instance, extended specific performance from its section 52 boundaries

4 See SGA 1979, s 37(2) (‘Nothing in this section affects the rights of the seller where the neglect or refusal of the buyer to take delivery amounts to a repudiation of the contract’). 5 As suggested obiter by Burnyeat J the British Columbia Supreme Court in Re Redfern Resources Ltd 2011 BCSC 771, (2011) 88 BLR (4th) 29 [53] (‘“Acceptance” means both the physical receipt of the goods by a buyer and the legal duty of the buyer to accept goods properly delivered by the seller – a duty the breach of which the seller may sue for damages for non-acceptance’). 6 A point made in M Bridge (gen ed), Benjamin’s Sale of Goods, 10th edn (London, Sweet & Maxwell, 2017) para 9.003. It is submitted that the seemingly contrary suggestion in M Bridge, Sale of Goods, 3rd edn (Oxford, Oxford University Press, 2014) para 6.05 cannot be supported. 7 ‘It is well known that the Factors Acts of 1899 and 1890 and the Sale of Goods Act 1893, must for many purposes be treated as one code’: Worcester Works Finance Ltd v Cooden Engineering Co Ltd [1972] 1 QB 210 (CA) 220 (Megaw LJ). See also Re Wait [1927] 1 Ch 606 (CA) 635–36 (Atkin LJ); Michael Gerson (Leasing) Ltd v Wilkinson [2001] QB 514 (CA) [35] (Clarke LJ).

The Duty to Take Delivery of Goods 153 (only for the buyer, and only for specific or ascertained goods)8 to cover unascertained goods,9 and also claims by sellers.10 Again, more recently, the courts have accepted that section 49(1), under which the buyer’s right to sue for the price of goods accrues when property passes, does not preclude such claims in other cases.11 There is thus nothing implausible about the idea that section 37(1) simply reproduces what exiguous nineteenth-century authority there was on the duty to accept delivery,12 rather than listing exhaustively the cases where the duty arises. The second line of reasoning relies on the well-established principle that a contractor can always renounce the benefit of a term inserted solely for his advantage.13 For example, I am not, it is suggested, in breach of contract if I tell my regular cleaner that she has a free day but I will still pay her for not working;14 or if I sell you goods, deliver them and then refuse point-blank to take any payment. Why not regard the buyer’s right to delivery in the same light? After all, if you can deliver goods and then refuse payment, it might seem a little odd that you cannot deliver payment and then decline the goods. A closer look, however, shows that this argument equally will not work. Whatever the status of a promise to pay – where it is indeed hard to infer an obligation to take money one does not want – the duty to deliver goods cannot, it is submitted, be regarded as inserted solely for the buyer’s benefit. A seller has a clear interest not only in being paid, but also in being disembarrassed of the goods he is selling, an interest that sounds both in convenience and very often in hard cash.15 Indeed, it is not difficult to think of cases where the price, or even the willingness of the seller to sell at all, might vary according to the willingness of the buyer to take the goods away. Examples would be where a seller of trucks is short of showroom space, or the owner of land encumbered with a number of derelict cars wishes a timely disposal in order to develop the site. With these obstructions cleared away, what about indications in favour of a take-up duty? Direct authority is, admittedly, almost entirely lacking. The only instance that comes to mind

8 ‘In any action for breach of contract to deliver specific or ascertained goods the court may, if it thinks fit, on the plaintiff ’s application, by its judgment or decree direct that the contract shall be performed specifically’ (emphasis added). See the comments in Re Wait (n 7) 635–36 (Atkin LJ); International Finance Corp v DSNL Offshore Ltd [2005] EWHC 1844 (Comm), [2007] 2 All ER (Comm) 305 [50]; Timmerman v Nervina Industries (International) Pty Ltd [1983] 1 Qd R 1, 8. 9 eg Thames Valley Power Ltd v Total Gas & Power Ltd [2005] EWHC 2208 (Comm), [2006] 1 Lloyd’s Rep 441 [63]. See also G Treitel, ‘Specific Performance in the Sale of Goods’ [1966] Journal of Business Law 211, 216–17. 10 See Astro Exito Navegacion SA v Southland Enterprise Co (The Messiniaki Tolmi) (No 2) [1983] 2 AC 787 (HL), esp 797 (Lord Roskill) (ship); Record v Bell [1991] 1 WLR 853 (Ch) (chattels where vendor of house obtained relief); Shell-Mex Ltd v Elton Cop Dyeing Co Ltd (1928) 34 Com Cas 39 (KB) 46 (Wright J). 11 PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2016] UKSC 23, [2016] AC 1034 [40]–[58] (Lord Mance), overruling FG Wilson (Engineering) Ltd v John Holt & Co (Liverpool) Ltd [2013] EWCA Civ 1232, [2014] 1 WLR 2365 on this point. 12 And the support was indeed remarkably thin. The traditional authority for s 37(1) is nothing more than an obiter dictum by Lord Ellenborough in Greaves v Ashlin (1813) 3 Camp 426, 427; 170 ER 1433, 1434. 13 Quilibet potest renunciare juri pro se introducto. See eg National Westminster Bank Ltd v Halesowen Presswork & Assemblies Ltd [1972] AC 785 (HL) 808. Lord Simon in the Halesowen case, however, preferred the definition in the 10th edition of Broom, Legal Maxims (London, Sweet & Maxwell, 1939): ‘Anyone may at his pleasure, renounce the benefit of a stipulation or other right introduced entirely in his own favour.’ 14 ‘Provided I pay my cook her wages regularly she cannot complain if I choose to take any or all of my meals out’: Collier v Sunday Referee Publishing Co Ltd [1940] 2 KB 647 (KB) 650 (Asquith J). Although this statement now has to be heavily qualified in the case of employment (eg William Hill Organisation Ltd v Tucker [1999] ICR 291 (CA), 297–301 (Morritt LJ)), there is no suggestion that this qualification affects independent contractors. Compare Cohen & Co v Ockerby & Co Ltd [1917] HCA 58, (1917) 24 CLR 288, 299 (Isaacs J): ‘If I contract to pay a certain sum to carry my goods two miles, I may dispense with the carriage of them after a mile, provided I pay the agreed price and occasion no burden or inconvenience to the carrier.’ 15 eg storage expenses; and possibly for other reasons too: eg if the seller faces a tax or other charge based on the goods being under his control.

154 Andrew Tettenborn is an old case proceeding on the assumption that there would be a claim for demurrage where a buyer free on board (FOB) failed to ship the goods on time from a barge alongside, shipment in such a case being equivalent to delivery.16 It is not entirely clear why authority should be so sparse, but one possible explanation is that in contracts where delayed take-up is likely to cause serious loss, such as commodity sales, provision is often expressly made for demurrage payments or carrying charges on the buyer’s part.17 Nevertheless, there are several indirect arguments in favour of a take-up duty. The first, and perhaps the most obvious, comes from the decision in Penarth Dock Co Ltd v Pounds.18 Lord Denning MR, sitting unusually at first instance, held that where a buyer had accepted and paid for a floating pontoon moored in the sellers’ dock, he was under an implied contractual duty physically to take it away within a reasonable time.19 It may be that this case is not strictly a case about failure to take physical delivery, since it is entirely possible that at the time the action was brought the buyer was already technically in possession of the pontoon.20 Nevertheless it is a suggestive decision, especially in combination with the fact that the law has had no compunction in imposing contractual obligations on contractual bailors other than buyers to physically remove their goods within a given time.21 If this is so, there seems no reason not to impose a take-up duty directly on buyers under the contract of sale itself. Secondly, and less obviously, there is the decision in Demby Hamilton v Barden22 concerning risk. Section 20(2) of the SGA 1979 supplements the rule in section 20(1) that risk presumptively passes with property by saying that ‘where delivery has been delayed through the fault of either buyer or seller the goods are at the risk of the party at fault as regards any loss which might not have occurred but for such fault’. Demby Hamilton was a textbook application of this sub-section. A buyer of fruit juice delayed taking delivery for some weeks because of problems further down his own supply chain; the juice having gone bad meanwhile, it was held to be at his risk, so that he had to pay for it. The significance of this case lies in the word ‘fault’. In this connection, it is submitted that this can only mean breach of contract. If the sub-section is there to prevent a party invoking the rules about transfer of risk if by so doing he is relying on his own wrong, it can hardly penalise a buyer who is not in breach in delaying delivery. And if that is so, then the only way the result in the Demby Hamilton case can be justified is that the buyer was regarded as having a contractual duty to take up the goods at the time stipulated for delivery. Thirdly, there is a point about damages for non-acceptance. There is no doubt that these can, where appropriate, include an element for subsequent warehousing and similar costs incurred by the seller.23 For instance, in Vitol SA v Conoil plc,24 where buyers of oil failed to accept the necessary ship-to-ship transfers and thus repudiated the contract, damages included sums in

16 See All Russian Co-op Society Ltd v Smith & Sons (1923) 14 Ll L Rep 351 (CA) (the claim failed only because the incurring of the expense was actually due to the seller’s own breach). 17 For coverage of such clauses, see Benjamin’s Sale of Goods (n 6) para 19.089. 18 Penarth Dock Co Ltd v Pounds [1963] 1 Lloyd’s Rep 359 (QB). 19 ibid 361. 20 On the basis that there is no indication that the sellers, who had been paid, did not allow him access to it. Compare the case of a small yacht sold while moored in a marina: it must be possible for the seller to deliver it without untying it, by simply allowing the buyer to go on board and take charge. 21 eg carriers. See eg Metall Market OOO v Vitorio Shipping Co Ltd [2013] EWCA Civ 650, [2014] QB 760 [70] (Sir Bernard Rix); P&O Nedlloyd BV v Arab Metals Co (The UB Tiger) (No 2) [2006] EWCA Civ 1717, [2007] 1 WLR 2288. 22 Demby Hamilton & Co Ltd v Barden [1949] 1 All ER 435 (KB). 23 Instances are Harlow & Jones Ltd v Panex (International) Ltd [1967] 2 Lloyd’s Rep 509 (Com Ct) and Vitol SA v Conoil plc [2009] EWHC 1144 (Comm), [2009] 2 Lloyd’s Rep 466. 24 Vitol (n 23).

The Duty to Take Delivery of Goods 155 respect of demurrage incurred by the seller to the owners of the transferring vessel, together with the cost of storing what had become essentially unsaleable oil. An award under this head seems difficult to justify unless there is actually a duty on the buyer physically to take the goods off the seller’s hands; if there is not, then it amounts to making the buyer liable in damages for failing to do what he was not bound to do in the first place, which is normally impermissible.25 Lastly, there is a point about specific performance. If, as seems to be the case, a contract to buy chattels can be specifically enforced at the seller’s suit,26 it is difficult to see what such an order requires the buyer to do, unless it is in some way physically to take over the goods.27

III. The Legal Basis of a Take-up Duty These arguments in favour of a take-up duty are, it is suggested, convincing. If we accept them, what might be the legal basis for it? There are, it is suggested, two possibilities. The first is straightforward: a sale contract that excluded any duty to take delivery of goods would make little commercial sense,28 and there is therefore a good reason to imply such a duty on the simple Moorcock29 ground of business efficacy. Quite apart from this, however, there is a compelling argument for imposing such a duty on the basis of the general contractual duty to cooperate in achieving performance. Put simply, this is the rule stated by Lord Blackburn in Mackay v Dick:30 where … it appears that both parties have agreed that something shall be done, which cannot effectually be done unless both concur in doing it, the construction of the contract is that each agrees to do all that is necessary to be done on his part for the carrying out of that thing, though there may be no express words to that effect.31

Lord Atkin forcefully confirmed this rule in 1940: there was, he said, ‘a positive rule of the law of contract that conduct of either promiser or promisee which can be said to amount to himself “of his own motion” bringing about the impossibility of performance is in itself a breach.’32 Insofar as a seller agrees under section 27 of the SGA 1979 to deliver, and since delivery, in the sense of ‘voluntary transfer of possession from one person to another’,33 cannot take place without the buyer’s aid (or at least acquiescence), it must follow that the buyer is implicitly bound to give that aid, and accordingly is in breach and liable in damages if he fails to do so.

25 ‘[A] defendant is not liable in damages for not doing that which he is not bound to do’: Abrahams v Herbert Reiach Ltd [1922] 1 KB 477 (CA) 482 (Scrutton LJ). 26 Such contracts were specifically enforced against buyers in Record v Bell (n 10), and earlier in Timmerman (n 8), both cases where the supposed restrictions contained in s 52 of the Sale of Goods Act 1893 (or, in the latter case, s 53 of the Queensland Sale of Goods Act 1896, which was word-for-word the same) did not apply. See also The Messiniaki Tolmi (No 2) (n 10) esp 797 (Lord Roskill); Shell-Mex v Elton Cop Dyeing (n 10) 46. 27 The matter can be tested in this way: if no delivery is required, and there are no other formal requirements, as in the case of investment wine stored in a third party cellar, there would, it seems, be nothing whatever to enforce specifically. As soon as ownership passed under s 17, the seller could simply sue for the price, with no need for any further order. 28 It would, it is suggested, fall within the words of Lord Salmon in Liverpool City Council v Irwin [1977] AC 239 (HL) 262. It at once becomes clear that the inclusion of the proviso renders this part of the contract ‘inefficacious, futile and absurd’. 29 See The Moorcock (1889) 14 PD 64 (CA). 30 Mackay v Dick (1881) 6 App Cas 251 (HL). 31 ibid 263 (Lord Blackburn). See also Swallowfalls Ltd v Monaco Yachting & Technologies SAM [2014] EWCA Civ 186, [2014] 2 All ER (Comm) 185 [32]–[33] (Longmore LJ). 32 Southern Foundries (1926) Ltd v Shirlaw [1940] AC 701 (HL) 717. 33 See SGA 1979, s 61; Benjamin’s Sale of Goods (n 6) para 9.003.

156 Andrew Tettenborn Such a solution has the advantage of embracing the special case of section 37(1) and yet going beyond it, for example, to cases where there is a fixed date for delivery. It also has a further advantage, in that it can also help to explain the limits on any take-up duty. Some sale contracts, it is worth remembering, do not contemplate physical delivery. An example, discussed further below, is a sale of specific wine or gold bars held in the hands of some third party X. Here delivery is required, but only in the form of attornment by X;34 conversely, this must exclude a physical take-up duty on the buyer (though possibly it leaves a duty not actively to refuse to accept X’s attornment). Another example is sale and leaseback, where mere agreement between seller and buyer is sufficient to create a relationship of bailee and bailor that is regarded as equivalent to delivery.35 It is also worth noting that at least one type of contract in regular use specifically ousts the duty to take physical delivery: the ‘take-or-pay’ contract commonplace in energy sales.36

IV. A Take-up Duty: Fault-Based or Strict? The seller’s duty to deliver goods, and to do it on time, is presumptively a strict liability obligation to see to it that the goods are there. Even if he entrusts the job of delivery to a third party, he still remains liable even though he is personally blameless for the latter’s acts and omissions.37 In principle, there is no reason not to apply the same reasoning to the buyer’s take-up duty. True, insofar as this duty is based on reasonable cooperation, lack of fault may incidentally excuse the buyer (for instance, if he is not given adequate information about a delivery to enable him to take it up). But exceptional cases like that aside, his duty must be to see to it that the required cooperation is supplied. So, if he fails to collect goods at the stipulated time, he cannot escape liability to compensate the seller by arguing that the van he sent was involved in an unavoidable accident en route. Even if the van is owned by an independent contractor, the result should be the same. A seller agreeing to deliver to a buyer cannot excuse himself on the basis of the failure of a third party to whom he delegates the job; the buyer agreeing to collect from the seller should be in the same position. Only if there is an agreement for an exclusive mode of collection that later becomes impossible should there be any exception – for example, if a given firm of truckers is stipulated in the contract and a drivers’ strike prevents it from collecting the goods.38 The question of liability for third parties could become particularly important in the case of sub-sales. Imagine a sale of a truck or caterpillar tractor by A to B, with B immediately sub-selling to C before delivery is due and A being happy to deliver direct to C. There is no doubt that such

34 See SGA 1979, s 29(4): ‘Where the goods at the time of sale are in the possession of a third person, there is no delivery by seller to buyer unless and until the third person acknowledges to the buyer that he holds the goods on his behalf.’ 35 As in the decision in Michael Gerson (Leasing) (n 7). See also the fairly similar Australian case of Gamer’s Motor Centre (Newcastle) Pty Ltd v Natwest Wholesale Australia Pty Ltd [1987] HCA 30, (1987) 163 CLR 236. 36 See, eg B Holland, ‘Enforceability of Take-or-Pay Provisions in English Law Contracts – Resolved’ (2016) 34 Journal of Energy and Natural Resources Law 443. Specific discussion of such contracts in English law is largely limited to whether they amount to penalties (eg M&J Polymers Ltd v Imerys Minerals Ltd [2008] EWHC 344 (Comm), [2008] 1 Lloyd’s Rep 541). That they do not is virtually certain since Cavendish Square Holdings BV v Makdessi [2015] UKSC 67, [2016] AC 1172. 37 Including the case where the third party is his own seller. Thus, each seller in a string commodity contract guarantees to his buyer that delivery will be timeously and duly made, and that the goods have been shipped on time, even though he has no control over the matter. See Profindo Pte Ltd v Abani Trading Pte Ltd (The MV Athens) [2013] SGHC 10 [40]; cf Great Elephant Corp v Trafigura Beheer BV [2012] EWHC 1745 (Comm), [2013] 1 All ER (Comm) 415 [113] (Teare J) (appeal allowed on other grounds at [2013] EWCA Civ 905, [2013] 2 All ER (Comm) 992). 38 cf Nickoll & Knight v Ashton Edridge & Co [1901] 2 KB 126 (CA) (analogous position as regards seller’s duty to deliver: contract frustrated when stipulated ship unavailable).

The Duty to Take Delivery of Goods 157 direct delivery is regarded as amounting to delivery under both contracts: by delivering to C on B’s order, A delivers to B, and B is taken to deliver to C through the agency of A.39 But what happens if C delays in taking delivery, causing A to incur storage costs? C, having no contract with A, cannot be liable for these, but what about B? B might think that once he had paid A and sub-sold to C he was free of responsibility, but he would (it is suggested) be wrong. In actuality, B is in the converse position to that of the seller in a string contract who guarantees proper performance to his buyer: his duty is, it is submitted, not simply to take over the goods, but also to ensure that they are taken over by anyone downstream of him to whom they are delivered at his request. If authority is needed, it can come from the law on express clauses binding on a buyer, such as provisions in a CIF (cost, insurance and freight) contract stipulating the time for unloading the goods and requiring payment of demurrage in default.40 There is no doubt that such clauses apply just as much against a buyer who has resold, and who thus has no part to play in the discharge process, as to a buyer who actually takes delivery.41

V. The Take-up Duty: Its Application to Specific Cases A. Straightforward Scenarios: Seller Handing Goods Over to Buyer In simple cases, the take-up duty should raise few difficulties. Insofar as goods are deliverable at the seller’s place of business (the default position under section 29(2) of the SGA 1979), the buyer must presumably take steps to collect them at the stipulated time42 or, in the absence of such a stipulation, on reasonable notice.43 If a direct handover is arranged for somewhere else, for example in the street, then, again, the buyer must be on hand to take over the goods. If the seller undertakes to deliver to the buyer’s premises, either using his own transport or using the services of a firm employed as his agent,44 it is suggested that the duty will consist in the buyer ensuring that the goods are accepted on arrival, or, where this is not necessary, then at least acquiescing in their being left there and ensuring that it is physically possible to do so (for instance, by unlocking any security gates, or providing the seller with a key or other means of access).

B. Delivery Followed by Later Collection It may be that goods are to be delivered, not at the buyer’s or seller’s premises, but at those of a third party intermediary X, with the buyer later collecting them from X. A straightforward commercial example is a contract on DAT45 terms calling for delivery at, and later pickup from, a container terminal; a more domestic example is a contract requiring delivery of a necklace at a jeweller, or a shotgun at a firearms dealer, for later retrieval. 39 See Four Point Garage Ltd v Carter [1985] 3 All ER 12 (QB) esp 15 (Simon Brown J). 40 Benjamin’s Sale of Goods (n 6) para 19.089. 41 See eg the facts of Gill & Duffus SA v Rionda Futures Ltd [1994] 2 Lloyd’s Rep 67 (Com Ct). 42 cf Demby Hamilton (n 22) above. 43 Which would be the case under SGA 1979, s 37(1) anyway. 44 On which, see the fraud cases of Galbraith & Grant Ltd v Block [1922] 2 KB 155 (KB) and Computer 2000 Distribution Ltd v ICM Computer Solutions plc [2004] EWCA Civ 1634, [2005] Info TLR 147. In both it was held that sellers to someone with apparent authority at the buyer’s premises fulfilled their obligations. 45 ‘Delivery At Terminal’: see Incoterms 2010 (available as a booklet: Incoterms 2010: ICC Rules for the Use of Domestic and International Trade Terms (ICC, 2010).

158 Andrew Tettenborn At one level, cases of this sort raise essentially the same issue as with delivery to a third party sub-buyer. The seller performs his duty to deliver, in the sense of making a voluntary transfer of possession from one person to another, by handing the goods to intermediary X. It must follow that in such a case the buyer, having delegated the take-up function to X, must for his part see to it that the goods are not left on the seller’s hands. However, there is an extra complication. Here the parties contemplate not one step in delivery but two: a handover by the seller to X, then another handover by X to the buyer. It has been suggested that the buyer guarantees a take-up at the first stage, but what about the second? The point may be significant: X, for example, may have insisted that the seller accept liability for storage or security charges pending ultimate collection. The point is a difficult one. Logically, however, it would seem that the buyer’s duty ought presumptively to be limited to the first stage; as regards the second, it is better regarded as a matter between the buyer and X alone. The reasoning runs thus. First, it is clear that section 37(1) – even if otherwise in point – cannot apply here, since the duty it imposes is to take delivery from the seller on demand, and the buyer, through X, has already done precisely this. Secondly, as regards any more general take-up duty, the underlying seller’s obligation to deliver has already been satisfied by delivery to X, and the converse buyer’s duty has been satisfied by acceptance on X’s part. This solution seems borne out by what authority there is. In the analogous case of goods in a carrier’s hands, it seems clear that no duty exists to collect the goods on arrival unless expressly stipulated, even if as a result the seller suffers loss;46 and there seems no reason why the same should not apply to any other intermediary. All this, however, leaves a further possibility. What if parties stipulate delivery for later pickup at no one’s premises? This is unlikely, but by no means impossible; imagine, for example, an owner of woodland arranging to leave cut logs at the side of the road for the buyer to collect in a truck the next day. Oddly enough, it is suggested that here the buyer is obliged actually to pick up the goods, and will be in breach if he does not. The logic runs like this. The buyer’s take-up duty depends on the seller’s delivery duty; leaving goods to be collected, in contrast to actually handing them over to an intermediary, is not delivery, since there is no transfer of possession; delivery therefore takes place when, and only when, goods are actually collected;47 the buyer is therefore bound to collect.48

C. Goods Remaining in the Hands of Bailees What of goods in the hands of bailees such as warehousemen? (We can leave carriers on one side here, since they raise particular problems that are considered separately in the next sub-section). Two situations need to be distinguished. The first case is where the contract contemplates physical delivery to the buyer by the bailee on the seller’s behalf. Here the buyer’s duty must be the same as if the seller himself were making the delivery, and we need say no more. The second case is less easy. What happens where parties contemplate the goods simply remaining passive in the hands of the bailee? Although one might think there could be no take-up

46 See below, text at section IV.D(ii). 47 See Thomas v Times Book Co Ltd [1966] 1 WLR 911 (Ch) 919 (Plowman J). 48 Which may matter if, for example, charges are levied by authorities for materials left at the roadside, or questions arise of the liability of the goods owners to third persons.

The Duty to Take Delivery of Goods 159 duty in this situation, this would be an oversimplification. Delivery is, perhaps s uprisingly, still required,49 though in the form of attornment: that is, by the bailee acknowledging that he holds on behalf of the buyer.50 Some such attornment must be shown,51 even if in practice it is very readily inferred.52 However, even here an important point arises: on principle, attornment requires the consent of both bailee and bailor, since a person cannot be made a bailor against his will.53 It follows that the buyer is not entirely exonerated. But if so, what is his duty? If, as seems likely, assent in attornment cases can be inferred from acquiescence,54 then the best way to express it is negatively, in the form of a duty not to render ineffectual any attornment on the part of the bailee by positively refusing the attornment. Even this, however, might not be an adequate solution in all cases. Suppose that under the contract a document such as a warehouse receipt or delivery order had to be tendered by the seller. The handing over of such a document, even if accepted by the buyer, does not amount per se to a making of delivery;55 there must be some input from the bailee as well.56 If so, it is suggested that the buyer would be obliged to provide cooperation by at least presenting the document to the bailee with a view to obtaining an attornment from him. The point, moreover, could matter. Although the seller’s contract with the bailee governs the extent of his duty to pay storage charges, and how far this duty survives a change of bailor, it is perfectly conceivable that a late taking over by the buyer, or failure to accept liability for such charges, may prolong the seller’s duty and cause loss to the seller. Indeed, the importance of the point goes further. Other duties attach to bailors for reward apart from payment: for example, to take the goods away on request,57 or to pay for emergency measures to preserve them.58 These are liabilities the seller will clearly wish to get rid of (or at least have accepted by the buyer as well). Insofar as the buyer fails to take what steps he can to effectuate the substitution, there is scope for substantial liability in damages.

D. Goods in the Hands of Carriers Where independent carriers intervene between seller and buyer, matters unfortunately get somewhat complex. Here we need to distinguish two cases: (i) those involving bills of lading and 49 Because of a combination of SGA 1979, s 27 (requiring delivery in all cases) and SGA 1979, s 29(4) (defining delivery as attornment in the case of goods in the hands of a bailee). 50 SGA 1979, s 29(4). 51 See Benjamin’s Sale of Goods (n 6) para 8.013. Not sufficient is eg receipt of a delivery order by the bailee from the buyer without comment (Laurie & Morewood v Dudin & Sons [1926] 1 KB 223 (CA)); nor such an order sent by the seller to the buyer alone (Alicia Hosiery Ltd v Brown Shipley & Co Ltd [1970] 1 QB 195 (QB)). 52 ‘Very little will suffice to create an attornment’: Laurie & Morewood (n 51) 237 (Scrutton LJ). See also Woodley v Coventry (1863) 2 H&C 164, 159 ER 68 (receipt of order by bailee coupled with intimation that order ‘all right’ suffices). 53 See Sale of Goods and Supply of Services, Halsbury’s Laws of England 91 (London, LexisNexis, 2012) para 261. See also cases such as Whitehead v Anderson (1842) 9 M&W 518, 152 ER 219 (no carrier’s attornment accepted by buyer for purposes of stoppage in transit); Bolton v Lancashire & Yorkshire Rly Co (1865–66) LR 1 CP 431 (refusal of attornment by buyer). 54 This seems to be the result of Swanwick v Sothern (1839) 9 Ad & E 895, 112 ER 1453. 55 Because in the absence of attornment no document other than a bill of lading can in general transfer legal possession of goods in a third party’s hands: see Inglis v Robertson [1898] AC 616 (HL); Mercuria Energy Trading Pte Ltd v Citibank NA [2015] EWHC 1481 (Comm), [2015] 1 CLC 999. 56 See Mercuria Energy (n 55) (duty to redeliver metal under repo arrangement not satisfied by mere sending of delivery order addressed to warehouseman). 57 See eg P&O Nedlloyd (n 21) (a case of contract, but equally, it is suggested, applicable to bailments). 58 N Palmer (gen ed), Palmer on Bailment, 3rd edn (London, Sweet & Maxwell, 2009) paras 10.47–10.51. See also China-Pacific SA v Food Corp of India (The Winson) [1982] AC 939 (HL); ENE Kos 1 Ltd v Petroleo Brasileiro SA (The Kos) (No 2) [2012] UKSC 17, [2012] 2 AC 164.

160 Andrew Tettenborn other maritime documents of title, in particular CIF and FOB contracts; and (ii) others. I will deal first with the latter – that is, cases involving carriage by a non-sea carrier who is neither employed by, nor contracted to obey exclusively the instructions of, either party.59

(i) Carriers Other than Sea Carriers Typically, this situation involves goods carried by an independent trucker from the seller’s premises to the buyer’s; but it could equally involve rail, air or inland waterway transport, or any combination thereof. The difficulty here is the same as that already touched on as regards other cases of delivery through an intermediary: the process of delivery is split, in this case between handing to and handover by the independent carrier. At what point, if at all, does the buyer’s duty to take over the goods bite? At first sight, the answer looks obvious: it must be the earlier point of handing to the independent carrier. Under section 32(1) of the SGA 1979, the seller prima facie satisfies his duty to deliver by handing the goods to the carrier; if so, the buyer’s correlative obligation to cooperate must arise simultaneously. It is the same as any other nomination of a third party to take delivery. The carrier here is the nominee – indeed, section 32 is traditionally explained on the basis that he acts as the buyer’s agent.60 From this, it must follow that the buyer guarantees take-up by the carrier. Unfortunately, this reasoning cannot be taken at face value. True, it works where the carrier is engaged by the buyer. But section 32 applies even where the carrier is engaged by the seller himself.61 There is, to put it mildly, something queer about the idea that if a buyer orders goods and the seller engages and pays a trucker to transport them to him, the buyer should then be liable if the trucker refuses to take them over. It cannot be right. The explanation, it is suggested, lies in the slightly peculiar nature of section 32(1). Although expressed in terms of delivery, it does not say that delivery to a carrier is delivery to the buyer (which it may not be),62 but merely that in certain cases it may be deemed to be. In fact, it is suggested that section 32(1) is concerned with little more than the rules on risk and property: its classic expressions are the rule that, having shipped goods, the seller can recover the price whether or not the buyer ever gets them,63 and that any inconveniences or expenses arising from events taking place after that time are prima facie for the buyer’s account alone.64 Furthermore, the view that section 32(1) is based on deemed agency, however venerable, is misleading. The carrier may of course be the buyer’s agent in fact; indeed, this will be a strong inference if the carrier is appointed and paid by the buyer, and acknowledges no duty to anyone else.65 But there

59 If the carrier is merely the agent of either party, the case counts as one of direct delivery: see the fraud cases of Galbraith & Grant (n 44); Computer 2000 Distribution (n 44). 60 See Sale of Goods and Supply of Services (n 53) para 186. See also cases at common law such as Vale v Bayle (1775) 1 Cowp 294, 98 ER 1094; Dawes v Peck (1799) 8 TR 330, 101 ER 1417; Dunlop v Lambert (1839) 6 Cl & F 600, 620; 7 ER 824, 831 (Lord Cottenham); Wait v Baker (1848) 2 Exch 1, 7; 154 ER 380, 383 (Parke B). 61 See SGA 1979, s 32(1): ‘Where … the seller is authorised or required to send the goods to the buyer, delivery of the goods to a carrier (whether named by the buyer or not) … is prima facie deemed to be a delivery of the goods to the buyer.’ 62 eg delivery to the buyer destroys the seller’s lien: but despite SGA 1979, s 32(1), the seller retains his lien if he delivers to a carrier and retains ownership. See SGA 1979, s 43(1). 63 An aspect of the rules relating to risk, but sometimes going further: cf DL Electrical Supplies (Mitcham) Ltd v GL Group Ltd 1987 SLT (Sh Ct) 36. 64 To that extent, SGA 1979, s 32(1) is the other side of the coin to s 18, r 5(2) and the prima facie passing of risk at the time of despatch. 65 The qualification is important. In Wait v Baker (n 60) it was held that a sea carrier who acknowledged no authority except that of the seller was not the agent of the buyer; hence, when the buyer refused the cargo, the seller was able to give good title to a third party.

The Duty to Take Delivery of Goods 161 is no need to use agency to explain the section; rather, it simply reflects the fact that a seller who is bound to make delivery to the buyer satisfies his obligation by delivery to the carrier as stipulated; once the seller has made such delivery, he is presumptively under no further liability.66 If this is right, then the extent of the buyer’s take-up duty in these carriage cases becomes a good deal clearer. The relevant time is indeed the handover to the carrier; but, unlike the case of delivery to a named third party, this is not an instance of delegation. The buyer need merely cooperate in the handover to the carrier insofar as these matters are under his control, for instance by providing any necessary information about where goods are to be sent, and not actively hindering their dispatch. Provided he does this, there is no reason why, in the absence of an actual relationship of agency between him and the carrier, he should bear any further liability. So much for the buyer’s duty at the time of delivery to the carrier. Is there any scope for an additional obligation attaching at the point of arrival? The better view must be against any such duty, for the same reason as in the case of delivery to an intermediary.

(ii) Carriers by Sea If carriers raise difficulties, they raise more intractable ones in connection with documentary sales on CIF or FOB terms, or for that matter any sales involving multimodal carriage with a large maritime element, where bills of lading are active in the process of documentary transfer. A bill of lading, it has to be remembered, symbolises possession of the goods it represents, and unlike other so-called documents of title, its physical transfer can produce the same legal effects as would have resulted from delivery of the goods themselves.67 Because of this, the application of section 32(1) of the SGA 1979 – whereby delivery is presumptively effected by delivery to the carrier – is problematical in such cases. The better view is that in all such cases the presumption contained in it is implicitly rebutted; if the seller cannot perform his obligation to deliver without handing over the bill of lading, it must follow that he does not satisfy it merely by shipping the goods.68 Instead, the seller’s duty to deliver is governed by the common law, and is bifurcated. He must ship the goods as required under the contract69 (or, more accurately, warrant that they have been so shipped);70 but he must also complete his performance by delivering the bill of lading.71 (There is a third stage of delivery, namely final collection by the buyer;72 but this is not relevant to the seller’s duty,73 since he notoriously 66 Save for a limited duty not actively to prevent the buyer getting hold of the goods: eg Empresa Exportadora de Azucar v Industria Azucarera Nacional SA (The Playa Larga and Marble Islands) [1983] 2 Lloyd’s Rep 171 (CA). 67 ‘[W]hen the vessel is at sea and the cargo has not yet arrived, the parting with the bill of lading is parting with that which is the symbol of property, and which, for the purpose of conveying a right and interest in the property, is the property itself ’: Meyerstein v Barber (1869–70) LR 4 HL 317, 326 (Lord Hatherley). See also E Clemens Horst Co v Biddell Bros [1912] AC 18 (HL) 22–23 (Lord Loreburn). 68 See eg Scottish & Newcastle International Ltd v Othon Ghalanos Ltd [2008] UKHL 11, [2008] 2 All ER 768 [20] (Lord Bingham). There is also another technical point, which is that in contracts CIF and FOB delivery consists in putting goods on board a vessel and not simply handing them over to the carrier: see Benjamin’s Sale of Goods (n 6) para 8.014. 69 Hindley & Co Ltd v East Indian Produce Co Ltd [1973] 2 Lloyd’s Rep 515 (Com Ct) (seller liable to buyer in damages when goods covered by bill of lading never in fact shipped). 70 As happened in Kwei Tek Chao (t/a Zung Fu Co) v British Traders & Shippers Ltd [1954] 2 QB 459 (QB). This is standard in the case of commodity trading, where each intermediate seller guarantees to his buyer the correctness of the date of shipment stated in the bill of lading. 71 See eg Hansson v Hamel & Horley Ltd [1922] 2 AC 36 (HL) (shipment of goods but no proper bill of lading: no proper delivery). 72 For descriptions of delivery CIF as involving three stages (shipment, documentary transfer and collection), see E Clemens Horst Co v Biddell Bros [1911] 1 KB 934 (CA) 955–59 (Kennedy LJ); Schmoll Fils & Co Inc v Scriven Bros & Co (1924) 19 Ll L Rep 118 (KB) 119 (Roche J). 73 Scottish & Newcastle International Ltd v Othon Ghalanos Ltd [2006] EWCA Civ 1750, [2007] 1 All ER (Comm) 1027 [43] (Rix LJ).

162 Andrew Tettenborn makes no promise to make the goods available at their destination at any particular time74 – or, indeed, at all.)75 If that is right, then it would seem that in this context the buyer’s take-up duty is correspondingly bifurcated: he must do what is necessary to cooperate with the seller at the stages of both shipment and documentary transfer. In practice, however, any such duty is likely to be fairly undemanding. As regards shipment, this can in most cases be effected without any input from the buyer anyway. Moreover, when it comes to take-up of documents, the practical significance is again likely to be low, since refusal to take up a document of title will normally amount to non-acceptance,76 and its late physical acceptance will in most cases cause the seller no loss. Nevertheless, this will not always be the case. For example, since delivery under a contract FOB is prima facie effected by shipping goods on a ship provided by the buyer,77 this clearly requires timeous provision of a vessel to load the goods, and where delay in making shipping space available causes the seller loss, this will be recoverable by him.78 Again, late take-up of shipping documents might cause loss. One example might arise out of an indorsee’s statutory liability under a bill of lading based on his having demanded delivery of the goods it represents.79 This liability disappears if the indorsee subsequently transfers the bill to someone else;80 in which case, a late take-up could plausibly cause considerable loss to the seller. It is submitted, however, that the buyer’s duty is limited to the stages of shipment and documentary transfer. There is no reason to extend it to collection of the goods from the carrier, since there is no corresponding duty on the seller to deliver at that point. And, indeed, this seems to be the law:81 if a seller wishes to have an indemnity against demurrage charges or some other provision in the nature of a ‘demurrage clause’, then he must stipulate for it.82 It does need to be noted, however, that insofar as the contract expressly incorporates Incoterms, then the buyer is under a duty to collect from the carrier. Even though the seller makes no promises that the goods will arrive, if they do, the buyer is bound to ‘receive them from the carrier at the named port of destination’.83

VI. Effect of Breach of the Take-up Duty Assuming there is a free-standing take-up duty implied at common law, clearly any breach of it by the buyer, whether by refusal to take up goods or by delay in doing so, will make him liable to the seller in damages. Indeed, this is likely to be its most important consequence: in most cases, the disappointed seller will simply be claiming extra storage and other expenses incurred by the seller

74 See eg Vitol SA v Esso Australia Ltd (The Wise) [1989] 2 Lloyd’s Rep 451 (CA) 454 (Mustill LJ); ERG Petroli SpA v Vitol SA (The Ballenita and The BP Energy) [1992] 2 Lloyd’s Rep 455 (Com Ct) 464. 75 See Manbre Saccharine Co Ltd v Corn Products Co Ltd [1919] 1 KB 198 (Com Ct) 202 (McCardie J); Bowden Bros & Co Ltd v Little [1907] HCA 14, (1907) 4 CLR 1364, 1377 (Griffith CJ). 76 Gill & Duffus SA v Berger & Co Inc [1984] AC 382 (HL). Such non-acceptance will on principle give a right to damages, but more importantly exonerates the seller from any further duty to perform: ibid 391. 77 See Benjamin’s Sale of Goods (n 6) paras 8.014, 20.015. 78 All Russian Co-op (n 16) (where, however, the seller failed because he was to blame). 79 On which, see the Carriage of Goods by Sea Act 1992, s 3(1). 80 See Borealis AB v Stargas Ltd (The Berge Sisar) [2001] UKHL 17, [2002] 2 AC 205 [40]–[46] (Lord Hobhouse, with whom all other members of the House agreed). 81 See C Debattista, ‘Laytime and Demurrage Clauses in Contracts of Sale’ [2003] Lloyd’s Maritime and Commercial Law Quarterly 508, 511–12. 82 See Benjamin’s Sale of Goods (n 6) para 19.089. 83 See Incoterms 2010, Rule B4 (stated to be applicable to contracts CIF, FOB, CIP, etc).

The Duty to Take Delivery of Goods 163 as a result of a buyer’s failure to take over the goods, together with financing expenses in cases where the price is payable only on the taking of delivery. But what are the other effects of breach? First, there is no reason why the take-up duty should not be specifically enforceable (assuming the remedy of specific performance is indeed available at the suit of the seller in a sale of goods contract).84 Damages might well be regarded as an inadequate remedy if, for example, a buyer of scrap metal littering a building site declined to remove it as promised and there was no other person who could carry out a clear-up within a reasonable time. Again, the same might apply where goods were otherwise difficult to dispose of, as, for example, in the case of slightly hazardous waste. In P&O Nedlloyd BV v Arab Metals Co (No 2),85 a carrier was left with a mildly but inconveniently radioactive cargo whose owner wrongfully declined to collect it: the court regarded as distinctly arguable a claim for a specific order telling the owner to do so. By analogy, it is suggested that the same result might well follow in the case of a contract of sale. Secondly, what might be the effect of breach of the buyer’s take-up duty on the seller’s own obligations? One point is obvious. Just as under section 28 of the SGA 1979 the duty to deliver is suspended unless and until the buyer is ready to pay the price, clearly the same thing must go for delivery and take-up: unless the buyer is ready and willing to take the goods over, he cannot complain of failure to deliver or to make the goods ready on time. More important, however, is the question whether the seller faced with breach can go further and actually terminate the contract. In other words, assuming there is a presumptive take-up duty, what is its status on the condition/warranty/innominate term scale?86 Classification as a condition can be swiftly ruled out.87 It would be curious, for example, if a refusal to take delivery of part of goods sold, accompanied by a willingness to pay for the whole, was automatically regarded as a repudiatory matter going to the root of the contract and allowing the seller in response to refuse to supply any goods at all. Indeed, there is clear authority that this is true of failure to accept part of the subject matter of a contract of sale,88 and if this is true of failure to accept the goods, it must a fortiori be true of a failure to take physical delivery of them. So too with late take-up. It would be equally odd if the seller of a high-value car or truck, although paid, was entitled to return the money and cancel the sale on the simple basis that the buyer had wrongfully delayed collecting the goods for a few days.89 If we are left with a choice between a warranty and an innominate term, at first sight it might be thought difficult to justify anything more than a warranty. Cases are two-a-penny where refusal by a buyer to take and pay for goods has been held to be a breach justifying the seller in withdrawing from the contract. But these are really instances of repudiatory non-acceptance, which is something different. The same goes for the special case of section 48(3) of the SGA 1979, allowing

84 See n 9 above. 85 P&O Nedlloyd (n 21) (contract to take delivery of mildly but inconveniently radioactive cargo: no strikeout of specific performance claim). This case involved a claim by a carrier against a consignee, but its reasoning applies equally to a seller and buyer. 86 The point is explicitly left open in cases where SGA 1979, s 37 applies: see s 37(2) (‘Nothing in this section affects the rights of the seller where the neglect or refusal of the buyer to take delivery amounts to a repudiation of the contract’). See also Francis v Lyon [1907] HCA 12, (1907) 4 CLR 1023, 1034 (Griffith CJ). 87 It is true that in commercial contracts the time of delivery is readily regarded as a condition (eg Macpherson Train & Co Ltd v Howard Ross & Co Ltd [1955] 1 WLR 640 (QB)). But the analogy with collection is not close. In the nature of commerce, the obtaining of goods is likely to be more important to a buyer than the getting rid of them is to the seller. 88 See Warinco AG v Samor SpA [1977] 2 Lloyd’s Rep 582 (Com Ct) 588 (Donaldson J) (reversed on other grounds, [1979] 1 Lloyd’s Rep 450 (CA)). See also the careful discussion in the Australian decision in Francis v Lyon (n 86) 1033–37 (Griffith CJ); Benjamin’s Sale of Goods (n 6) para 9.010. 89 cf SGA 1979, s 10(2), stating that the status of stipulations as to time depends on the parties’ intention; the sub-section draws no distinction between sellers’ and buyers’ obligations.

164 Andrew Tettenborn an unpaid seller to resell and cancel the contract if the buyer fails to come up with the money on reasonable notice;90 failure not only to take up goods but also to pay for them will almost invariably be repudiatory. Where the only complaint by the seller is tardiness in collection, or refusal to take goods coupled with an expressed willingness to pay the price, this is much less likely to amount to a failure of the ‘substance and foundation of the adventure’,91 such that damages would not amply vindicate the seller’s interests. Less likely indeed, but not impossible. Take, for example, the facts of Penarth Dock Engineering Co v Pounds.92 Although the emphasis in that decision was on the measure of damages, the background to the case was, as the defendant well knew, that the claimants had sold him the pontoon with a view to getting rid of it quickly and thus allowing redevelopment of the site. In such a situation, it is entirely plausible to say that serious foot-dragging by the buyer would indeed deprive the seller of a major part of the consideration it had bargained for, and would justify the seller in finding someone else prepared to remove the offending dock instantly, returning the original buyer’s money and closing with the new buyer there and then. It is not hard to think of other examples. One such might be a dealer in fashion items who sells last year’s contents at a knockdown price in order to clear space in a very cramped warehouse for the new season: if the buyer fails to take the goods away, even after having paid, there is no reason why the seller should not return the price and sell to someone else. If so, the obvious solution is to classify the obligation to take delivery as an innominate term. In the same way as failure to deliver on time may be, or may become, repudiatory depending on the parties’ intentions, the same should go for failure to collect: insofar as circumstances show that this deprived the seller of a substantial part of the consideration he had contracted for, the failure to collect should allow the seller to cancel the contract.

VII. Conclusion The conclusions of this chapter can be simply stated. Quite apart from the specific provision contained in section 37(1) of the SGA 1979, a buyer of goods presumptively owes a general duty in English law to take over physical possession of the goods. This duty, which exists in addition to the duty to accept the goods, arises at common law and is based on the implied duty of cooperation. It is closely tied to the seller’s duty to make delivery, and obliges the buyer to do everything necessary to allow the seller to perform that duty. Where goods are to be conveyed from seller to buyer by independent carrier, the duty applies at the time of dispatch, and in the case of contracts involving transfer of a bill of lading, also at the time of the transfer of documents; however, there is no presumptive duty to collect the goods physically on arrival in the absence of specific provision. As regards status, the take-up duty is best regarded as an innominate term of the sale contract. In practice, most breaches of it are unlikely to be repudiatory. Nevertheless, in some cases, the seller may indeed be entitled to escape from his own obligations where the buyer refuses to have anything to do with collecting what he has bought.

90 That such a resale cancels the contract on the basis that notice has made time of the essence is clear from RV Ward Ltd v Bignall [1967] 1 QB 534 (CA). 91 Lord Esher’s phrase in Bentsen v Taylor, Sons & Co (No 2) [1893] 2 QB 274 (CA) 281. 92 Penarth Dock (n 18).

10 The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century LOUISE GULLIFER

I. Introduction This chapter considers the relevance of the Sale of Goods Act 1979 (SGA 1979) to modern commerce. As a result of changing financing practices, of developing technology and of legal developments, the scope of application of this once all-embracing Act has been severely reduced, so that, in many situations, it operates more as a source of inspiration and analogy than as hard law. Moreover, many aspects of consumer sales have been taken out of the SGA 1979 altogether and included in the more comprehensive Consumer Rights Act 2015. This chapter explores the ramifications of this conclusion and considers whether the resulting fragmentation can be characterised as beneficial common law flexibility or criticised as a failing to treat functionally similar cases alike. The chapter starts with some historical context, pointing out that even in Roman times, the line between sales and other methods of acquisition of goods was porous. Three requirements of the definition of a contract for sale of goods under section 2 of the SGA 1979 are then considered to illustrate how one or more are not met in many modern contracts of acquisition. First, the requirement that property in goods is transferred is considered in the contexts of equipment, inventory and hybrid contracts. Secondly, the requirement that what is sold is ‘goods’ is examined in the light of the huge rise in contracts for the supply of digital content. Thirdly, the requirement that the consideration is a price of money is considered in the light of acquisition in return for cryptocurrencies, of barter and of free acquisition. Given the number of acquisition contracts which fall outside the scope of the SGA 1979, the next section examines what is left to fall within it. The final section considers whether any of this matters, and concludes that there are convincing arguments that it does, with the result that there should be new legislation, to ensure that like cases are treated alike and to attempt to future-proof the law, at least for the next 40 years,1 if not for the next 126.2

1 The 2 The

period since the Sale of Goods Act 1979 (SGA 1979). period since the Sale of Goods Act 1893.

166 Louise Gullifer

II. Some Brief Historical Comments English law was not the first system to treat sale contracts as a special category of contracts. In Roman law, the contract of sale was one of the four nominate consensual contracts,3 and provided actions for both buyer (the actio ex empto) and seller (the actio ex venditio).4 Specific rules developed regulating important aspects of the contract, such as when the contract of sale was concluded,5 the passing of risk,6 and the duties of seller and buyer.7 The Roman jurists were acutely aware of the difficulties of categorising contracts as contracts of sale, and there is considerable discussion in the texts of the boundaries of the category. For example, there was clearly a fierce debate as to whether the ‘price’ should consist of a sum of money or whether an ‘exchange of things’ (that is, a barter) could fall within the contract of sale. This debate was decided, at least by the late classical period,8 but probably by the time of Gaius,9 in favour of the former view, on the grounds that otherwise it would not be possible to distinguish which party was the buyer and which the seller. This was important, as their duties were quite different. Another boundary that was explored was that between sale and hire. Sale involved the transfer of ownership, whilst hire of a thing only involved the hirer having detention of the thing, with the owner remaining the owner. The line was largely clear, but Gaius discusses the problem of the perpetual lease of municipal land: although this was functionally the same as a sale, Gaius concludes that it is a contract of hire.10 Another problem identified by Gaius was that of the provision of gladiators, which obviously involved a heavy risk of damage to or destruction of the subject matter of the contract. The market (apparently) was such that a much higher price11 was payable if the gladiator was maimed or killed than if he was returned unhurt,12 but the question of legal categorisation was more difficult. The jurists developed the idea of a conditional sale: ownership passed if the gladiator was killed or maimed (so that the contract became one of sale), but not if the gladiator was returned unhurt (so that the contract was one of hire).13 The boundary between contracts of sale and contracts for services was also explored by discussion of the case of the goldsmith who agrees to make some gold rings of a specified weight and form. The preferred solution was that this was a contract of sale, unless the gold was provided by the client, in which case it was a hire of services.14 This brief digression into Roman law demonstrates that the boundaries of sale were always contentious, and that many of the issues confronted by modern law are not at all new. Of course, the correct characterisation mattered a great deal in Roman law, since it defined the action which could be brought and therefore the formula which was put before the judge. Even more critically,

3 Institutes of Gaius, 3.135. 4 For other actions giving specific remedies developed during the Roman law period, see WW Buckland, A Textbook of Roman Law from Augustus to Justinian (Cambridge, Cambridge University Press, 1921) 485–91. 5 Institutes of Gaius, 3.139; Institutes of Justinian, 3.23.pr. 6 Institutes of Justinian, 3.23.3. 7 See JAC Thomas, Textbook of Roman Law (Amsterdam, North-Holland Publishing Co, 1976) 300–05. 8 Digest, 18.1.1.1 and 19.4.1, both texts of Paulus. 9 Institutes of Gaius, 3.141. 10 Institutes of Gaius, 3.145. 11 1,000 silver pieces. 12 20 silver pieces. 13 Institutes of Gaius, 3.146. 14 Institutes of Gaius, 3.147; Institutes of Justinian, 3.24.4.

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 167 if an agreement did not fall within the nominate contracts, there was the possibility that an action could not be brought at all.15 In English law, sales law is a subset of contract law, and the general law can fill in any gaps, as is the case with a contract of barter, for example.16 The flexibility of the common law, plus sensible legislative intervention, can avoid most undesirable outcomes. However, categorisation still matters, because some different statutory consequences apply depending on the applicable statute or provisions, and because if too many matters are left up to the common law, this is productive of uncertainty in an area of law where predictability is important.

III. The Sale of Goods Act 1979 The current iteration of the SGA 1979 is a much-amended version of the original codifying statute of 1893. The provision primarily setting out its scope is section 2(1), which provides that ‘A contract of sale of goods is a contract by which the seller transfers or agrees to transfer the property in goods to the buyer for a money consideration, called the price’ (emphasis added).17 The added italics highlight the aspects of definition which are discussed in this chapter. While not entirely comprehensive,18 the SGA covers most of the contractual and proprietary issues that commonly arise in relation to a sale of goods: the formation of the contract,19 implied terms delineating the level of protection of the buyer,20 the passing of property and risk,21 the nemo dat rule and exceptions to it,22 performance on the parts of buyer and seller (including provisions about the operation of the self-help right of rejection),23 provisions protecting an unpaid seller before delivery24 and remedies for breach.25 All of these provisions apply only to contracts for the sale of goods and actual sales.26 In considering the application of the SGA 1979 to today’s world, we must ask what people (that is, consumers and businesses, acting both as suppliers and acquirers) acquire these days and how they acquire such things?

IV. ‘Transfer the Property’ In Roman times, there was conceptually a huge line between owning something and not owning it.27 Even then, the Romans realised that, practically, having the use and ‘detention’28 15 This situation was mitigated by praetorial intervention, giving rise to ‘innominate’ contracts to cover such situations. These contracts were, however, limited as they were only enforceable on part performance. See Thomas (n 7) 327. 16 MG Bridge (ed), Benjamin’s Sale of Goods, 10th edn (London, Sweet & Maxwell, 2017) paras 1.034–1.038. 17 The scope is further defined by the SGA 1979, s 61, which concerns the interpretation of words and phrases in the Act, and s 62, which deals with its interaction with related statutes and areas of law. 18 See MG Bridge, The Sale of Goods, 3rd edn (Oxford, Oxford University Press, 2014) para 1.11. 19 SGA 1979, ss 2–9. 20 ibid ss 10–15A. 21 ibid ss 16–20B. 22 ibid ss 21–26. 23 ibid ss 27–37. 24 ibid ss 38–48. 25 ibid ss 49–54. 26 A contract of sale becomes a ‘sale’ once property passes: ibid s 2(4). 27 B Nicholas, An Introduction to Roman Law (Oxford, Clarendon Press, 1962) 153–57. 28 Under Roman law, the term ‘possession’ was reserved for those who were entitled to the protection of the possessory interdicts: Thomas (n 7) 146. There was no generic Roman law term for others who had actual control of a thing, but the modern term is ‘detention’: Nicholas (n 27) 112.

168 Louise Gullifer of something was worth having, hence the similarity (in many respects) between the contract of sale and the contract of hire. In modern law, use and ‘detention’ can be achieved by having possession of goods, and the need for ownership often pales into insignificance compared to possession, at least in practical terms. Building on this reality, English law, in common with that of many other jurisdictions, has enabled the legal concept of ownership to be so decoupled from use and possession that the acquisition of ownership is no longer necessarily what an acquirer wants. Given this situation, it is not surprising that suppliers of goods on credit have manipulated the concept of passing of property in order to protect themselves against non-payment. The combination of these two phenomena has resulted in actual sales of goods becoming an increasingly rare species. This section will examine these propositions considering different types of goods. As mentioned above, although a large portion of the law in relation to consumer acquisitions has been taken out of the SGA 1979, the rules governing the passing of property are still to be found there.29

A. Equipment In this chapter, the word ‘equipment’ is given a relatively narrow meaning: that is, objects which are acquired for long-term use and which are not destroyed in the short term by that use. Of course, all equipment will become unusable in the end, either through physical wear or through obsolescence; thus, this meaning is not intended to be a hard-edged definition, but merely to indicate the kind of goods discussed in this section. These goods have two important characteristics: first, they are relatively expensive, and secondly, they last a relatively long period of time. These two characteristics combine to make their acquisition on credit both necessary (if the acquirer has limited cash flow) and possible, in that the provider can obtain a proprietary interest in the goods in order to ‘secure’ the provision of credit. Even where the acquirer could afford to pay for the goods outright, it is often beneficial to use a credit transaction to acquire them. First, for companies, there are likely to be tax advantages in that the repayments are tax deductible, and the value of the goods can be written down at approximately the same rate as the amounts expended.30 Secondly, it gives flexibility to the acquirer, who can usually terminate the arrangement, or upgrade or update the equipment, during the life of the equipment without undue cost. There are four main legal structures for this type of finance (often called ‘asset finance’).31 The first, conditional sale, consists of a sale agreement under which the price is paid in instalments and property does not pass until all the instalments are paid. This transaction would be a contract of sale under the SGA 1979, and it is for this reason that it is little used in commercial acquisitions of equipment. This is because section 25 of that Act would apply, and a sub-sale by the conditional buyer before property passed would enable a good faith sub-purchaser to obtain good title. Instead, a hire purchase agreement (under which the acquirer hires the goods with an option to purchase for a (usually) nominal sum) is often used. Section 25 does not apply in this case, as a person acquiring goods under a hire purchase agreement is not a ‘person … [who has] agreed to

29 See below section VII.A. 30 See, eg HMRC Business Leasing Manual BLM00710, published online at www.gov.uk/hmrc-internal-manuals/ business-leasing-manual/blm00710. 31 For a fairly typical account of the use of the term, see www.fundingoptions.com/knowledge/asset-finance/.

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 169 buy goods’ under that section.32 In consumer acquisitions, this difference has been eliminated, and so both transaction types are used.33 The other two structures that are used are both forms of leasing: finance leases and operating leases.34 Under a finance lease, the period of the lease is typically the useful life of the goods, and the instalments are the sum of the price of the goods plus a finance charge. At the end of this period, the lessee often has the right to continue hiring the goods at a very low rate, or the goods are sold and the bulk of the sale proceeds are transferred to the lessee.35 The important point is that the lessee has substantially all the risks and rewards of ownership without actually being the owner.36 An operating lease is a short-term lease, which gives maximum flexibility to the lessee since it can be terminated at any point and allows easy upgrading of equipment.37 Operating leases are not only popular with businesses, but are increasingly being used by consumers as an alternative to acquiring a vehicle on conditional sale or hire purchase terms.38 Given this situation, it is not surprising that statute has intervened to provide a contractual level of acquirer protection in relation to hire purchase and leasing contracts which is equivalent to that in the SGA 1979.39 However, many of the rules on the transfer of title to non-owners are not the same: no exceptions to the nemo dat rule apply to leasing and, apart from the acquisition of vehicles for which there are private registers, none apply to hire purchase. All other proprietary issues are left up to the common law. Much consumer acquisition on credit is governed by the Consumer Credit Act 1974, which also covers some non-corporate business acquisition. It should be acknowledged that there are some low value goods which would count as ‘equipment’ under the loose definition above but which are not usually acquired on credit40 and in relation to which property passes at least at the point of delivery. Examples include books, household goods and clothes, but even these are not necessarily the subject of sale any more. Books are increasingly acquired in digital form,41 and there is a rental market for household goods,42 and even clothes.43 32 Helby v Matthews [1895] AC 471 (HL). The Finance and Leasing Association statistics for the 12 months leading up to May 2019 for asset finance entered into by their members (FLA 2019 statistics) do not even include conditional sale as a ‘product’ within their data categories: see www.fla.org.uk/business-information/documents/asset-finance-summary/. 33 In agreements which are consumer credit agreements under the Consumer Credit Act 1974, a buyer under a conditional sale agreement is also not a person who has agreed to buy goods: SGA 1979, s 25(2). Moreover, a large majority of consumer credit acquisitions are vehicles. The Hire Purchase Act 1964, s 27 enables a good faith acquirer of a vehicle from a person who has possession either under a conditional sale agreement or a hire purchase agreement to take free of the right of the owner, thereby equating the two forms of transaction, although, from the point of view of the supplier, in the direction of a decrease in protection. 34 The FLA 2019 statistics (n 32) show that a total of £33.8bn of asset finance was undertaken with their members in the 12 months leading up to May 2019, of which £4.3bn was finance leasing, £6.1m was operating leasing and £19bn was hire purchase. 35 H Beale et al, The Law of Security and Title-Based Financing, 3rd edn (Oxford, Oxford University Press, 2018) para 7.47. 36 This has up to now been the test for accounting purposes applied to determine whether a lease is a finance lease or an accounting lease, but the new IFRS 16 standard requires both types of leases to appear on the company’s balance sheet: see www.ifrs.org/issued-standards/list-of-standards/ifrs-16-leases/. 37 See, eg www.barclayscorporate.com/solutions/corporate-banking-solutions/financing/asset-finance/. 38 See www.fleetnews.co.uk/news/fleet-industry-news/2019/02/08/steady-growth-in-leasing-broker-market-reportsbvrla and www.moneyadviceservice.org.uk/en/articles/leasing-a-car#the-big-difference-between-pcp-and-pch-buyingand-owning-the-vehicle. 39 Supply of Goods (Implied Terms) Act 1973, ss 8–11; Supply of Goods and Services Act 1982, ss 7–10; the Consumer Rights Act 2015 (CRA 2015), pt 1, ch 2 applies to all contracts for the ‘supply’ of goods, including sales contracts, hire-purchase contracts and contracts of hire (leases): CRA 2015, s 3(2). 40 Or credit is supplied more generally, eg by a credit card company or by a general bank loan. 41 See section V below. 42 See, eg www.forbes-rentals.co.uk/index.asp. 43 See www.bbc.co.uk/news/business-45630395.

170 Louise Gullifer

B. Inventory Inventory here refers to goods which are acquired in order to be consumed, destroyed, made into something else or resold. In the consumer sphere, most of these are bought outright for cash (or using credit cards). Given the short-term nature of the acquisition, issues relating to consumers usually involve their contractual rights, which are governed by the Consumer Rights Act 201544 and not the SGA 1979. It is very common, however, for commercial parties to acquire such goods on credit and for the acquisition agreement to include a term stating that property will not pass until the price is paid. In many situations, recent authority suggests that this will not be a sale of goods contract. First, it could be a ‘sui generis contract’. The case of PST Energy 7 Shipping LLC v OW Bunker Malta Ltd45 concerned a contract for the sale of bunker oil. The Supreme Court held that because it was envisaged that the bunkers would be consumed before the end of the credit period, so that in relation to all or most of the bunkers property would never pass, it was not a contract for the sale of goods within section 2(1) of the SGA 1979, but a sui generis contract ‘to permit consumption prior to any payment … and … if and so far as bunkers remained unconsumed, to transfer the property in the bunkers so remaining to the Owners in return for the Owners paying the price’.46 The background to, and reasoning in, the case has been discussed in detail elsewhere.47 It is, however, strongly arguable that the analysis applied by the Supreme Court to the contract for the supply of bunkers will also apply to many other types of contract for the supply of goods on credit. First, it appears directly applicable to credit supply contracts for goods which will be consumed or destroyed before the end of the credit period, such as food or wine to restaurants, cleaning fluid to offices or factories, and fuel to many types of businesses. Secondly, the reasoning can apply to goods to be used before the end of the credit period in the manufacture of a new product, whereby the goods lose their identity and are subsumed into a new thing.48 Thirdly, it can, on the basis of similar reasoning, apply to goods incorporated into real estate, which lose their identity and become part of the land, such as paint on a wall.49 Fourthly, and perhaps more contentiously, the reasoning could apply to goods which are sold to a sub-buyer, but only if the first acquirer sells as agent and not as principal: in the latter case, a situation discussed immediately below, property would pass to the first acquirer immediately before the sub-sale. The likelihood that consumption, manufacturing and so on will take place before property passes – so that the contract is a sui generis contract – is increased very significantly if the retention of title clause is an ‘all monies’ clause, whereby property does not pass until all sums due from the acquirer to the supplier are paid.50 The second possible situation where goods are acquired on credit on terms that property will not pass until the price is paid, and where there is not a sale of goods, is where the acquirer

44 CRA 2015, ss 9–30. 45 PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2016] UKSC 23, [2016] AC 1034. 46 ibid [28]. 47 L Gullifer, ‘“Sales” on Retention of Title Terms: Is the English Law Analysis Broken?’ (2017) 133 LQR 244; K Low and K Loi, ‘Bunkers in Wonderland: A Tale of How the Growth of Romalpa Clauses Shrank the English Law of Sales’ (2018) 3 Journal of Business Law 229; D Saidov, ‘Sales Law Post-Res Cogitans’ (2019) 1 Journal of Business Law 1. 48 For a discussion of this process and its legal effect, see Borden (UK) Ltd v Scottish Timber Products Ltd [1981] Ch 25 (CA) 41. This is an example of the Roman law doctrine of specificatio: Thomas (n 7) 146, 186–88. 49 This is an example of the Roman law doctrine of accessio: Thomas (n 7) 146, 180–88. 50 See, eg Armour v Thyssen Edelstahlwerke AG [1991] 2 AC 339 (HL).

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 171 merely acts as agent for the supplier, in possessing and then selling the goods. This analysis was applied by a majority of the Court of Appeal in Caterpillar (NI) Ltd (formerly known as FG Wilson (Engineering) Ltd) v John Holt & Co (Liverpool) Ltd51 to a contract which on its face was a sale of goods contract. The contract contained a term stating ‘Nothing herein contained shall be deemed to create an agency … between the parties’. Further, the terms ‘sale’, ‘seller’ and ‘buyer’ were used throughout the contract. Despite this, it was held that the retention of title term, which stated: Buyer shall hold the products as Seller’s fiduciary agent … Prior to title passing Buyer shall be entitled to resell or use the products in the ordinary course of business and shall account to the Seller for the proceeds of sale

meant that the acquirer acted throughout as the supplier’s agent, and not as a buyer. The result of this interpretation was that property never passed to the acquirer but instead passed straight to the person to whom the acquirer sold; accordingly, the supplier could not sue the acquirer for the price, since neither section 49(1) of the SGA 1979 (which requires that the property in the goods has passed to the buyer for an action to be available) nor section 49(2) (which requires that the price is payable on a day certain irrespective of delivery) were satisfied. It is, of course, perfectly possible for parties to enter into an agency distribution agreement if that is what they want to do.52 Under such an agreement, the distributor must account for all proceeds of any sale, owes the provider of the goods the duties of an agent and is usually paid some sort of fee or commission. The provider takes the risk of whether the goods are sold or not, and will have a direct relationship with the buyer. There are many reasons why a normal supply on credit is not such an agreement even if the goods are going to be resold. The buyer takes the risk of whether, and at what price, the goods are resold, and is entitled to all profit. It will not want to owe the duties of an agent to the seller, nor does the seller want a direct relationship with any sub-buyers.53 Section 49 is worth consideration as it is a provision which is only found in the SGA 1979, and therefore only applies to sale contracts. Section 49 provides that the seller may maintain an action for the price where property in the goods has passed to the buyer or where the price is payable on a day certain irrespective of delivery. It does not say in terms that these are the only two situations in which the seller can sue for the price, but this was the conclusion reached by the Court of Appeal in the Caterpillar case. This conclusion provoked the argument in PST Energy that the contract was not one of sale, so that, despite the fact that property had not passed, the party providing the bunkers had a claim in debt for the sum described in the contract as ‘the price’. In PST Energy, Lord Mance said, obiter, that section 49 did not set out the only situations in which the price could be sued for, and that other situations in which this could happen included where risk passes to the buyer before property passes. Since risk passes in this way in virtually all sales on retention of title terms, it would seem that the exception is likely to swallow up the rule. If this dictum had, instead, been the ratio of the case, the scope of the SGA 1979 might have remained within its previous contours. As it is, its application to credit sales of inventory appears to be very limited.54

51 Caterpillar (NI) Ltd (formerly known as FG Wilson (Engineering) Ltd) v John Holt & Co (Liverpool) Ltd [2013] EWCA Civ 1232, [2014] 1 WLR 2365. 52 See, eg Angove Pty Ltd v Bailey (Re D&D Wines International Ltd) [2013] EWHC 215 (Ch), where an agreement was characterised in precisely such terms. The case was appealed, but not in relation to the judge’s finding that the company acted as agent: [2014] EWCA Civ 215, [2015] 1 All ER (Comm) 36 [16]; [2016] UKSC 47, [2016] 1 WLR 3179 [5]. 53 L Gullifer, ‘The Interpretation of Retention of Title Clauses: Some Difficulties’ [2014] Lloyd’s Maritime and Commercial Law Quarterly 564, 568–71. 54 Its application is also subject to considerable uncertainty: see section VII.B below.

172 Louise Gullifer

C. Hybrid Contracts In addition to the contracts considered above, which concern the supply of just goods, there are also increasing numbers of contracts where goods are supplied as part of another type of contract. There have always been contracts for the provision of one-off services which also entail the supply of goods, such as building contracts and repair contracts. Here, it is likely that property in the goods will pass, normally, when they accede to the building or the item being repaired. Increasingly, however, goods are also supplied as part of ongoing services, such as a mobile phone supplied as part of a contract for the use of mobile phone services, a box of equipment supplied as part of a contract for cable television or a smart meter supplied as part of a contract for electricity. Often, property will not pass in such a contract, although the position will vary. As with the various methods of credit supply of equipment, statute has stepped in to provide very similar protection for buyers as under the SGA 1979,55 although in this situation it is necessary to identify which parts of the contract relate to the supply of goods and which relate to services,56 since liability for the former is strict, whereas for the latter there is a standard of reasonable care and skill.57 As before, proprietary issues are left up to general law, with Consumer Credit Act 1974 regulation applying to credit agreements.

V. ‘Goods’ Unsurprisingly, the Sale of Goods Act 1979 covers sales of ‘goods’. The question of what amounted to ‘goods’ has historically caused a few, but limited, problems – mainly related to crops58 and other tangible assets which are attached to land.59 The SGA 1979 deals overtly with these in the definition of goods in section 63: ‘goods’ includes all personal chattels other than things in action and money … and in particular ‘goods’ includes emblements, industrial growing crops, and things attached to or forming part of the land which are agreed to be severed before sale or under the contract of sale

However, in the light of the technological advances in the last few decades, much of what people (including businesses) want to acquire in the twenty-first century is not tangible at all, but is ‘digital content’. Digital content used to be held on a physical medium, which could be transferred and therefore sold. In St Albans City and District Council v International Computers Ltd,60 the Court of Appeal decided that software delivered in the form of floppy disks would potentially count as ‘goods’, and that the supply contract would fall within the SGA 1979 if property in the disks was to be transferred (which did not happen on the facts of the St Albans case). Glidewell LJ adopted a passage from the first instance judgment which now seems rather quaint: The really important part of the system is the software. Programs are the instructions or commands that tell the hardware what to do. The program itself is an algorithm or formula. It is of necessity contained 55 Supply of Goods and Services Act 1982, ss 1–5; Supply of Goods (Implied Terms) Act 1973, ss 8–11; CRA 2015, s 5(2) (a wider definition of ‘sale’ contract than under SGA 1979, s 2, which probably covers some contracts for work and materials: see Bridge (n 16) para 14.057) and s 8. Any consumer contracts in the hybrid category that do not fall within the categories governed by the CRA 2015 (see section VII.A below) will continue to fall within the Supply of Goods and Services Act 1982: see s 1(1) of that Act. 56 Thus mirroring the problem Gaius considered in relation to the goldsmith: see text to n 14. 57 Supply of Goods and Services Act 1982, s 13; CRA 2015, s 49. 58 See Bridge (n 18) paras 2.04–2.11. 59 ibid paras 2.12–2.15. 60 St Albans City and District Council v International Computers Ltd [1996] 4 All ER 481 (CA).

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 173 in a physical medium. A program in machine readable form must be contained on a machine readable medium, such as paper cards, magnetic tapes, discs, drums or magnetic bubbles.61

Twenty-three years later, it is very rare indeed for digital content to be delivered in physical form: it is downloaded directly to the user’s computer, tablet or phone. Moreover, the amount and variety of digital content has massively increased, so that much information, entertainment and instruction that used to be delivered in physical form – such as books, CDs, manuals and so on – is now chiefly obtained in downloaded digital form. This has two effects. First, what is being obtained is not goods. Secondly, the method of acquisition has changed: what the acquirer typically obtains is not ownership of anything, but a licence to use the software. Each of these effects will now be discussed. The existing case law, while not determinative, appears to indicate that supply of digital content will not fall under the SGA 1979 as currently constituted and that, apart from cases of consumer acquisition,62 it is presently governed by the common law.

A. Is Software ‘Goods’? There are a number of authorities which have considered whether downloaded digital content is ‘goods’, although none of them have been in the context of the applicability of the SGA 1979. All of them have either come to the conclusion that downloaded digital content is not ‘goods’ or have found reasons on the facts not to decide the issue. The contexts have been varied. In Thunder Air Ltd v Hilmarsson,63 it was common ground that documents held in electronic form were not ‘goods’ and therefore could not be the subject of a claim in conversion under the Torts (Interference with Goods) Act 1977. This concession appears to have been approved of by Moore-Bick LJ in Your Response Ltd v Datateam Business Media Ltd,64 a case concerning whether a possessory lien could be taken over an electronic database. A third context concerns whether an agent is a commercial agent within Regulation 2(1) of the Commercial Agents (Council Directive) Regulations 1993,65 namely ‘a self-employed intermediary who has continuing authority to negotiate the sale or purchase of goods on behalf of another person’. In considering whether a person who negotiated the sale of digital content was such an agent, one first instance judge avoided the difficult question altogether by finding that the software was supplied via tangible material.66 Thus, the argument that this was a supply of goods within the regulations raised a serious question to be tried and so permission to serve the claim out of the jurisdiction could be granted. This route was not available to the court in Computer Associates UK Ltd v Software Incubator Ltd,67 and, in fact, both the judge at first instance and Gloster LJ in the Court of Appeal commented that it was illogical for the status of software as goods to depend on the nature of the medium by which it was delivered.68 The judge at first instance, working from first principles and the

61 ibid 492. 62 See CRA 2015, pt 1, ch 3. 63 Thunder Air Ltd v Hilmarsson [2008] EWHC 355 (Ch). 64 Your Response Ltd v Datateam Business Media Ltd [2014] EWCA Civ 281, [2015] QB 41 [20]. 65 Commercial Agents (Council Directive) Regulations 1993, SI 1993/3053. 66 Fern Computer Consultancy Ltd v Intergraph Cadworx & Analysis Solutions Inc [2014] EWHC 2908 (Ch), [2014] Bus LR 1397 [86]. 67 Computer Associates UK Ltd v Software Incubator Ltd [2016] EWHC 1587 (QB), [2017] Bus LR 245; on appeal: [2018] EWCA Civ 518, [2019] Bus LR 522. 68 Computer Associates (HC) (n 67) [39], [52]; Computer Associates (CA) (n 67) [51].

174 Louise Gullifer ‘goods-like’ way in which the relevant product was treated and perceived by the parties, held that the supply of software was, at least for the purpose of the regulations, a sale of goods.69 This decision was overturned by the Court of Appeal, which, despite acknowledging that the distinction between tangible and intangible ‘products’ was arbitrary, decided that the weight of precedent compelled the conclusion that software that was not delivered through a tangible medium was not goods. If sale of these products was to be included in statutory regimes, this was a matter for Parliament, as had already been done in the Consumer Rights Act 2015,70 and in New Zealand and Australian sales legislation.71 This conclusion has also been reached by the New South Wales Supreme Court in Gammasonics Institute for Medical Research Pty Ltd v Comrad Medical Systems Pty Ltd.72

B. Is the Method of Acquisition a ‘Sale’? As discussed above, the SGA 1979 concerns only sales and not other forms of acquisitions. Software is commonly acquired under a licence (often a perpetual licence), rather than by an outright transfer, so that even were it to fall within the term ‘goods’, most acquisitions would not be sales of goods. In Southwark LBC v IBM UK Ltd,73 Akenhead J held that when software was supplied under a perpetual licence, this was not a sale within the SGA 1979, since property in the software remained in the supplier.74 It is, of course, possible for the supply of software to be an outright transfer of property in that software, but this is rarely done as suppliers usually want to prevent unauthorised duplication of the material.75

VI. Money Consideration Several methods of acquisition of goods that are gaining popularity in the modern world arguably fall outside section 2 of the SGA 1979 as they are not transfers of property in goods in exchange for a money consideration. The first is the possibility of ‘payment’ in cryptocurrencies, the second is barter and the third is acquisition for no consideration at all. 69 Computer Associates (HC) (n 67) [68]. 70 CRA 2015, pt 1, ch 3 covers the ‘supply’ of ‘digital content’, defined in s 2 as ‘data which are produced and supplied in digital form’. This is a wider concept than that of ‘computer software’, which appears in New Zealand and Australian legislation, and arguably avoids the problem identified by Edelman J in Australian Competition and Consumer Commission v Valve Corp (No 3) [2016] FCA 196, (2016) 337 ALR 647 [138], [156], that ‘computer software’ includes executable data but not non-executable data. 71 Consumer Guarantees Act 1993 (NZ), s 2 (definition of ‘goods’ ‘to avoid doubt [includes] … computer software’); Australian Consumer Law (Competition and Consumer Act 2010 (Cth), sch 2), s 2(1) (definition of ‘goods’). 72 Gammasonics Institute for Medical Research Pty Ltd v Comrad Medical Systems Pty Ltd [2010] NSWSC 267, (2010) 77 NSWLR 479 [42]–[47]. 73 Southwark LBC v IBM UK Ltd [2011] EWHC 549 (TCC), (2011) 135 Con LR 136. 74 ibid [95]. See also Fern Computer Consultancy (n 66) [96], where the judge held that there was a serious issue to be tried as to whether supply by means of a licence fell within the Commercial Agents (Council Directive) Regulations 1993; Computer Associates (CA) (n 67) [69], where the court did not consider this question given its view that computer software was not ‘goods’, although the judge at first instance had taken the view that a perpetual licence could be viewed as a sale (Computer Associates (HC) (n 67) [69]); Australian Competition and Consumer Commission (n 70) [145]–[157], where the provision of a licence to use computer software was held to be a ‘supply of goods’ under the Australian Consumer Law. 75 See S Green, ‘Sales Law and Digitised Material’ in D Saidov (ed), Research Handbook on International and Comparative Sale of Goods Law (Cheltenham, Edward Elgar Publishing, 2019).

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 175

A. Cryptocurrencies As mentioned in section II, even in Roman times the ‘price’ for a sale of goods had to be in money. The Romans did not trouble themselves too much as to what ‘money’ was: it was clearly distinguished from ‘goods’, so that it was easy to tell who was the buyer and who was the seller. This was not the case where goods were bartered for other goods, and so the contract was not one of sale. The Sale of Goods Act 1893 took the same approach,76 and was followed in the SGA 1979, whereas the Supply of Goods and Services Act 1982 adopts a different, broader approach, which does not require money consideration.77 ‘Money’ is not defined in the SGA 1979, but it is reasonably arguable that it has the same meaning as under the common law, as opposed to its meaning under any particular statute (for example, the Value Added Tax Act 199478 or the Proceeds of Crime Act 2002). ‘Money’ is notoriously difficult to define: there are a number of theories as to what it is, none of which are entirely satisfactory.79 However, what seems reasonably clear in relation to a sale of goods is that the price can comprise any form of payment which is denominated in ‘money’ (including foreign currency).80 What is not at all clear is whether a price denominated or paid81 in ‘cryptocurrency’, such as bitcoin, can fall within the definition in section 2 of the SGA 1979. Money has often been described as something which is a medium of exchange, a store of value and a unit of account.82 The theories of money, however, require something more before something can be classed is money. For example, the state theory of money83 requires that money is issued by a state, which is clearly not the case in relation to cryptocurrencies such as bitcoin. The institutional theory of money requires money to be supported by an institutional framework,84 which, again, does not apply to cryptocurrencies, at least in the way that institutions are usually known. The societal theory of money requires that money is generally used in the community in the three ways set out in the first sentence of this paragraph. While this theory can be criticised as being too vague to be of use, it is also not (yet) fulfilled for cryptocurrencies, even bitcoin, in that, while they may be used as a store of value, they are rarely used as a medium of exchange or as a unit of account.85 It is, therefore, reasonably likely that, at the moment, a contract to

76 For discussion of the history of the restriction in the 1893 Act to sales for money consideration, see S Gleeson, The Legal Concept of Money (Oxford, Oxford University Press, 2018) para 7.87. Note that the definition of ‘goods’ in the SGA 1979, s 61, excludes ‘money’. 77 The Supply of Goods and Services Act 1982, s 1(3) provides that a contract is a relevant contract for the transfer of goods ‘whatever is the nature of the consideration for the transfer or agreement to transfer’. 78 In Case C-264/14 Skatteverket v Hedqvist [2016] STC 372 the CJEU held that contracts for the exchange of bitcoin and fiat currency were contracts for the supply of services falling within Art 135(e) of the Council Directive 2006/112/EC of 11 December 2006 on the common system of value added tax [2006] OJ L347/1 under which Member States were obliged to exempt ‘transactions, including negotiation, concerning currency, bank notes and coins used as legal tender’. However, the context, and the wide interpretation given to these exemptions, means that this is not a reliable guide to whether bitcoin is ‘money’ or ‘currency’ under English law. 79 See, eg M Bridge et al, The Law of Personal Property, 2nd edn (London, Sweet & Maxwell, 2017) paras 7.008–7.011. 80 Bridge (n 16) para 1.034, citing Daewoo Australia Pty Ltd v Suncorp-Metway Ltd [2000] NSWSC 35, (2000) 155 FLR 54. 81 The distinction between the two is discussed in Gleeson (n 76) paras 7.93–7.94. 82 Bridge et al (n 79) para 7.002. 83 C Proctor, Mann on the Legal Aspects of Money, 7th edn (Oxford, Oxford University Press, 2012) para 1.17. 84 A Sáinz de Vicuňa, ‘An Institutional Theory of Money’ in M Giovanoli and D Devos (eds), International Monetary and Financial Law: The Global Crisis (Oxford, Oxford University Press, 2010) paras 25.17–25.23. 85 M Carney, ‘The Future of Money’ (2 March 2018), www.bankofengland.co.uk/-/media/boe/files/speech/2018/thefuture-of-money-speech-by-mark-carney.pdf?la=en&hash=A51E1C8E90BDD3D071A8D6B4F8C1566E7AC91418.

176 Louise Gullifer transfer the property in goods in exchange for the transfer of a certain amount of bitcoin or other cryptocurrency is not a contract for the sale of goods,86 and therefore is not governed by the SGA 1979.87

B. Barter It is very clear from the discussion above that a barter is not a sale.88 This had caused few problems during the last 100 years, since barter was relatively rare. It was difficult to find someone who was willing to swap something they had that a person wanted for whatever it was that that person wanted to swap. All this has changed with the advent of online platforms, which put together not only buyers and sellers, but also potential parties to a swap. While it is, perhaps, not yet a mainstream way of acquiring things, swapping through swap sites on the internet is becoming more widespread,89 as are shops offering swaps as well as or instead of sales.90

C. Free Acquisition While there has rarely been much chance of acquiring goods absolutely free (‘free’ gifts are usually given as an added incentive to purchase other goods), a considerable amount of digital content is offered free of charge. This is particularly true of ‘apps’: computer applications often used on a mobile phone.91 The suppliers of apps usually make their money elsewhere, for example, by advertising or by use of the data generated by the app. It is hard to see this as a ‘money consideration called the price’.

VII. What is Left for the Sale of Goods Act 1979? In light of the analysis above, it might be asked whether there is anything left in the modern world to be governed by the terms of the SGA 1979. The answer is yes, but it is worth mapping the contours, in order to consider the ramifications of its limited scope, and possible responses.

A. Some Aspects of Consumer Sales Many aspects of consumer sales in the SGA 1979 have been superseded. The Consumer Rights Act 2015 covers the supply of goods to a consumer by a trader: ‘supply’, here, includes not

86 See D Fox and S Green (eds), Cryptocurrencies in Public and Private Law (Oxford, Oxford University Press, 2019) para 2.42. 87 It is, however, governed by the CRA 2015: s 8 defines a contract for the transfer of goods as a contract where the trader agrees to transfer the ownership of goods to the consumer and the consumer agrees to provide consideration otherwise than by paying a price. 88 Barter is expressly covered in the CRA 2015, s 8; see n 87 above. 89 See, eg www.swapz.co.uk; www.gumtree.com/swap-shop; www.carswappers.co.uk. 90 See, eg www.independent.co.uk/news/business/indyventure/stock-exchange-govanhill-swap-market-glasgow-ailierutherford-a8536851.html. 91 See, eg www.microsoft.com/en-gb/store/top-free/apps/pc.

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 177 only sale,92 but also hire, hire-purchase and a ‘contract for the transfer of goods’, the definition of which is wide.93 It also covers the supply of digital content, as discussed above, and the supply of services. The 2015 Act provisions set out rights for the consumer, as well as remedies if those rights are breached. It also provides for the passing of risk,94 and certain rules about the delivery of goods.95 However, there are a number of aspects of sale contracts that it does not cover, which are, if applicable, left to be governed by the SGA 1979.96 The most important of these relate to the time at which ownership passes under a contract of sale,97 the rules governing sales by a non-owner98 and the seller’s rights against the buyer – that is, the action for the price and damages for non-acceptance.99 Moreover, the 2015 Act only applies to supply by a ‘trader’, that is, a person acting for purposes relating to that person’s trade, business, craft or profession. A sale of goods by an individual not acting in the course of business will continue to be covered by the SGA 1979 – although of the implied terms set out in that Act, only the terms in section 12 (that the seller has a right to sell) and section 13 (that the goods will comply with the description) will apply to a sale made outside the course of business. With the development of online auction sites, such as eBay, there are now large numbers of consumer-to-consumer (C2C) sales. Though disputes are often in practice governed by the resolution dispute mechanism run by the platform, these C2C sales are governed at least in theory by the rules of the SGA 1979.

B. Sales to Businesses within the Sale of Goods Act 1979 As pointed out earlier, many credit sales are now likely to be classified as sui generis contracts and not sale of goods contracts. Whether there are any credit sales left within the scope of the SGA 1979 will be considered below. This section first considers the situations in which there are likely to be non-credit sales to businesses, which will continue to be governed by the SGA 1979, in order to assess the ongoing significance of the Act. As mentioned above, much equipment is acquired by businesses and by consumers on hire-purchase or lease terms. The counterparty to the hire-purchase or lease is rarely a manufacturer of, or a dealer in, the equipment. Instead, it is likely to be a financier, probably a company specialising in this type of finance. It may be affiliated to the manufacturer or dealer,100 it may be affiliated to a bank101 or it may be a free-standing financial company.102 In each case, the finance

92 The definition of which is wide (see CRA 2015, s 5) and so probably covers some ‘hybrid’ contracts, as discussed in section IV.C above. 93 CRA 2015, s 8, which seems to cover not only barter, but also other ‘hybrid’ contracts. 94 CRA 2015, s 29. 95 CRA 2015, s 28. 96 See Bridge (n 16) 14.047 for a list of these. 97 CRA 2015, s 4, referring specifically to SGA 1979, ss 16–20B (except s 20). The definition of ‘contract of sale’ in the CRA 2015 is based on the definition in European Parliament and Council Directive 1999/44/EC of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees [1999] OJ L171/12, Art 1(4). This is wider than that in the SGA 1979. Presumably the passing of ownership in contracts which fall outside the SGA 1979 is governed by the contract and ordinary property law (as it would be in the case of eg barter). 98 SGA 1979, ss 21–26. 99 SGA 1979, ss 49–50. 100 See, eg Siemens Financial Services Ltd: new.siemens.com/uk/en/products/financing/equipment-and-technologyfinance/financial-products.html. 101 See, eg BNP Paribas Leasing Solutions Ltd: leasingsolutions.bnpparibas.co.uk/our-solutions/. 102 See, eg Lombard North Central plc: www.lombard.co.uk/lombard.html.

178 Louise Gullifer company will have to buy the goods from the manufacturer or dealer, and this will be a contract of sale. The finance company is unlikely to buy on credit: they will fund the credit they give by other means, such as using invoice discounting or securitisation to raise money backed by the receivables under the hire-purchase or leasing contracts. The sale to the finance company will, then, be governed by the SGA 1979. Some commercial buyers still pay immediately on delivery, or even pre-pay before delivery. This is particularly likely when the goods are in transit and the buyer can raise money against the transit documents, such as a bill of lading. In these circumstances, there clearly appears to be a sale falling under the SGA 1979. Sometimes the credit period is short, so that payment is definitely due before the goods will be destroyed, used on manufacture or onsold. This is increasingly unlikely as the use of ‘just-in-time’ supply chains grows.103 There are also contracts which could either be within the SGA 1979 or could be sui generis contracts. These are supply chain contracts where it is unclear whether the goods will be destroyed or used in manufacture or resold before payment is made. The result of the PST Energy case is that it is extremely uncertain whether such contracts fall within the SGA 1979. The ‘test’ as to the nature of the contract appears to be that set out in the judgment of Moore-Bick LJ in the Court of Appeal. He said: Whenever the contract provides that property is to pass on payment and that anything more than a minimal part of the goods may be consumed before payment is due, the parties necessarily contemplate that part of the goods may not exist at that time. In truth, therefore, it is not an agreement to sell the nominal contract quantity, but an agreement to sell whatever remains at the time of payment. It may be that in some circumstances, for example, where the parties contemplate that the overwhelming majority of the goods will continue to exist at the date when property is to pass, that the ability to transfer property in the remainder will be of fundamental importance and an inability to do so will amount to a total failure of consideration or a breach which goes to the root of the contract, but that is not this case.104

It is trite law that a contract is to be interpreted at the time when it is entered.105 Thus, at that point, the nature of the contract depends on whether it was envisaged that more than a ‘minimal’ amount of the goods would be consumed before payment or whether the ‘overwhelming majority’ of the goods would remain at the time of payment. Here, we have two imponderables: when will the goods be consumed and when will the buyer pay? The parties may not have a fixed intention about either of these at the time of contracting. Even if they do, what is the test for these intentions? Do they have to be reasonable or realistic? What if the parties, agreed on every other point, have different intentions as to these two matters? The parties may, moreover, change their minds (or events may intervene). Does this amount to a variation of the contract (changing its nature)? The Supreme Court in PST Energy were adamant that the contract could only be of one nature, in that it could not be a sui generis contract in relation to goods consumed before payment and a contract of sale in relation to goods remaining at the time of payment.106 The characterisation, presumably based on the parties’ intentions at the time of contracting, is therefore determinative. To make matters worse, goods are often sold in a chain of contracts. Where the credit periods and terms are different at different stages of the chain, there is a real possibility that some contracts may be contracts for the sale of goods and others may be ‘sui generis’ contracts. One possibility,

103 For an account of their growth, see www.ups-scs.com/solutions/white_papers/wp_JIT.pdf. 104 PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2015] EWCA Civ 1058, [2016] 2 WLR 1072 [19]. 105 Whitworth Street Estates (Manchester) Ltd v James Miller & Partners Ltd [1970] AC 583 (HL) 603, 611, 614. 106 PST Energy (SC) (n 45) [31], rejecting the analysis in PST Energy (CA) (n 104) [33].

The Vanishing Scope of the Sale of Goods Act 1979 in the Twenty-first Century 179 which could both overcome the uncertainty and restore at least some of the reach of the SGA 1979, is that parties will provide expressly in their contract that property in the goods shall pass at the time of payment or when the goods are destroyed, used in manufacturing or onsold. The first instance judge in PST Energy rejected the argument that there was an implied term that ‘property in the bunkers passed to the [shipowners] at the moment of consumption’107 on the grounds that this was inconsistent with the express term that property would pass on payment when consumption before payment was envisaged. This rejection appears to be approved by the Supreme Court.108 However, an express term to the effect suggested here would not fail the test of consistency.

VIII. Does it Matter? The benefit of the common law system is that, despite all the matters considered in this chapter, there will not be any black holes.109 The contracts discussed above which do not fall squarely within the SGA 1979 are all governed by law in one way or another. Some are governed by one or more statutes,110 while others are governed by the common law. It is likely that terms similar to those implied (or imposed) by the SGA 1979 will be implied into those governed by the common law,111 but the position is not definitively determined, and not all the implied terms are appropriate as they stand.112 Other matters, such as the available remedies, would depend on the general law of contract. While the SGA 1979 can be a guide in these circumstances, some of the remedies available in relation to sales will not be appropriate,113 and the details, such as prima facie quantification of damages, may be different. It could be said, though, that these consequences merely result from untidiness of the law, rather than any failure of content. Since every situation is covered, a restatement of the law on the acquisition of ‘things’ (including digital content) would be a ‘nice-to-have’ tidying exercise, but is not imperative. However, there are some good arguments to the contrary. First, the present patchwork of statutory provision and common law gives rise to the need for characterisation in a similar way to Roman law, leaving scope for debate and legal uncertainty.114 Secondly, the content, at least of the common law, is far from certain in areas not yet decided. Thirdly, there is a real danger that like cases will not be treated alike, leading at best to fragmentation and, at worst, to unfairness. Moreover, although this is a debate for another place, the manipulation of the passing of property in credit sales has arguably meant that the financing aspect of these transactions has not been given sufficient weight, and that many of these transactions should be treated as sales plus secured transactions.115

107 PST Energy 7 Shipping LLC v OW Bunker Malta Ltd (The Res Cogitans) [2015] EWHC 2022 (Comm), [2015] 2 Lloyd’s Rep 563 [61]. 108 PST Energy (SC) (n 45) [28]. 109 Unlike, at least potentially, the position under Roman law. 110 See sections IV.A, IV.B and VII.A above. 111 This was the assumption of the court in PST Energy (SC) (n 45) [31]. 112 For example, in relation to sui generis contracts, see Bridge (n 16) 2nd supplement, paras 4.001, 4.002, 4.009. 113 For example, a right to reject may not be appropriate in a contract for supply of digital contract; there is no short-term or final right to reject in relation to such a contract under the CRA 2015. 114 See, eg the discussion in section VII.B. 115 See Gullifer (n 47) 264–68; Law Commission, Company Security Interests: A Consultative Report (Law Com CP No 176, 2004) paras 2.89–2.101.

180 Louise Gullifer

IX. Conclusion It is argued in this chapter that, over the 126 years since the first Sale of Goods Act, societal and technological changes, legislative reform and the manipulation of the concept of the passing of property, together with some less than desirable court decisions, have reduced the practical scope of the current Sale of Goods Act to near vanishing point. The main power of the SGA 1979, currently, is as a starting point for analogous application, either in other legislation or by the courts. This position, though not disastrous, is far from satisfactory. There is a need for new legislation, particularly, though not entirely, in the commercial sphere, to ensure that the way people and businesses acquire what they want and need is covered by a coherent and accessible body of law in which like cases are treated alike. It would be too much to hope for a comprehensive statute which lasts as well over the next 126 years as the 1893 Act (at least in outline) has lasted. A more achievable target is to devise a structure and content which would govern the law of acquisition for the next 40 years.

11 Are there Any General Principles of Commercial Law? GERARD McMEEL

I. Introduction In this chapter, I address the question of whether there are any general principles of commercial law. The subject has long been celebrated by its practitioners and adherents – from Mr Justice Buller’s fulsome tribute to Lord Mansfield in the Age of Enlightenment1 to the twenty-first century.2 Its virtues are trumpeted as being myriad, embracing certainty, predictability, flexibility, amenity to innovation and, foremost, a facilitative demeanour. The pride in English achievements often sails dangerously close to meriting the rebuke that the English, like the Romans, were ‘much too convinced of the superiority of their legal and political system to pay much attention to foreign laws’.3 The domestic rhetoric often involves hostility to notions such as good faith. So, Lord Falconer boasted: ‘English commercial law provides predictability of outcome, legal certainty, and fairness. It is clear and built upon well-founded principles, such as the ability to require exact performance and the absence of any general duty of good faith.’4 This represents a widely held view as to the key strengths of English commercial law. From spelling out the standard incidents of mercantile documents essential to international trade, such as marine policies and bills of exchange,5 to providing the default regime for twenty-first century international commercial and financial transactions, English law has often been a regime of choice. In purely

1 Lickbarrow v Mason (1787) 2 TR 63, 73; 100 ER 35, 41: ‘From that time we all know the great study has been to find some certain general principles which shall be known to all mankind, not only to rule the particular case then under consideration, but to serve as a guide for the future. Most of us have heard these principles stated and reasoned upon, enlarged, and explained, till we have been lost in admiration at the strength and stretch of human understanding. I should be very sorry to find myself under a necessity of differing from any case on this subject which has been decided by Lord Mansfield, who may truly be said to be the founder of the commercial law of this country.’ 2 For a recent (if controversial) example, see Lord Thomas CJ, ‘Developing Commercial Law through the Courts: Rebalancing the Relationship between the Courts and Arbitration’ (BAILLI Lecture, 9 March 2016). 3 K Zweigert and H Kotz (trans T Weir), An Introduction to Comparative Law, 3rd edn (Oxford, Oxford University Press, 1998) 49, quoting Cicero’s observation that all non-Roman law was ‘confused and quite absurd’. 4 Lord Falconer of Thoroton, Constitutional Affairs Secretary and Lord Chancellor, ‘The Commercial Bar Association Lecture’ (18 October 2005). 5 Reflected in the importance of standard forms: ‘In a sense, we do not have one commercial code but several, to be found in all the standard forms so widely used throughout the world – the Lloyds form of standard marine policy, the various forms of charter party and commodity trade contract, and so on’: Lord Goff of Chieveley, ‘“Opening Address”, Second Annual JCL Conference in London on 11 September 1991’ (1992) 5 Journal of Contract Law 1, 3.

182 Gerard McMeel material terms, English professional legal and associated services, including both public and private commercial dispute resolution, are a major source of invisible earnings. But are virtues the same as a coherent philosophy, and do they yield identifiable general principles which make a difference either in the development of doctrine or in the resolution of individual disputes? In the course of this discussion, I will consider the tensions between the suggested general principles, policies and techniques of English commercial law.

II. Is there Such a Thing as ‘Commercial Law’? The logically prior question to my titular one is whether or not there is any such phenomenon as ‘commercial law’. Whilst there have been many textbooks on commercial or mercantile law over the last two centuries, few have had any staying power equivalent to Chitty on Contracts6 or other leading private law texts. Furthermore, there has been little consensus on commercial law’s content, boundaries and the topics to be covered. Indeed, many academic and other selfstyled commercial lawyers suffer existential angst about whether the subject they teach, research or practise in has any autonomous or meaningful existence. Until Professor Harry Street of the University of Manchester commissioned Professor Roy Goode of Queen Mary College in the early 1970s to produce a single tome aimed at both lawyers and business people, the subject lacked a truly definitive treatise identifying its content and scope.7 Previously, some textbook writers had given up any attempt to define commercial law. Amongst those admitting defeat was Disney: ‘“Commercial law” is an expression incapable of strict definition, but which is used to comprehend all that portion of the law of England which is more especially concerned with commerce, trade and business.’8 Others persisted. So Gutteridge usefully observed: The object of commerce is to deal in merchandise and, if we adopt this criterion, commercial law can be defined as the special rules which apply to contracts for the sale of goods and to such contracts as are ancillary thereto, namely, contracts for the carriage and insurance of goods and contracts the main purpose of which is to finance the carrying out of contracts of sale.9

That captures the traditional reach of mercantile law, but perhaps focuses excessively on tangible property and international trade. The late Professor Peter Birks usefully posited the distinction between (i) conceptual categories of law, such as the law of obligations, property law, and unjust enrichment, defined at a level of abstraction and typically found in most developed legal systems, and (ii) contextual categories of law, grouped around some identifiable sphere of activity, such as shipping law or banking law.10 Where does commercial law belong within this divide?

6 The second volume of Chitty, with its alphabetical treatment of particular species of contract, represents a substantial repository of English commercial law at the heart of the Common Law Library series, but is obviously unsystematic and not concerned with general principle. Current edition: H Beale (gen ed), Chitty on Contracts, 33rd edn (London, Sweet & Maxwell, 2018). 7 The first edition was published in 1982. It is now edited by Professor Ewan McKendrick: see E McKendrick, Goode on Commercial Law, 5th edn (London, Penguin, 2016). 8 HW Disney, The Elements of Commercial Law, 4th edn (London, Macdonald & Evans, 1931) 1. 9 HC Gutteridge, ‘Contract and Commercial Law’ (1935) 51 LQR 91. 10 For the distinction between conceptual categories as opposed to contextual categories of law, see P Birks, An Introduction to the Law of Restitution, rev edn (Oxford, Oxford University Press, 1989) 73–74.

Are there Any General Principles of Commercial Law? 183 Goode’s textbook boldly essays a comprehensive and, to some extent, conceptual definition, being concerned with ‘that branch of the law which is concerned with rights and duties arising from the supply of goods and services by way of trade’.11 The conceptual unity lends the subject some intellectual dignity. Less ambitious and more workaday is Goode’s explicitly contextual definition in his Hamlyn Lectures: ‘the totality of the law’s response to mercantile disputes [and] the private law rights and obligations of parties to commercial transactions’.12 That to my mind captures the reality that commercial law is a contextual and essentially pragmatic subject, concerned with the application (with necessary modifications) of the English law rules concerning contracts, torts (and equitable wrongs), unjust enrichment, and personal property, to commercial and financial disputes. The public policy dimension, derived from statute and regulatory rules – especially prominent in the financial sphere – is highly significant. The types of claim which are the concern of the Commercial Court in London represent, I suggest, a reliable checklist of the concerns of commercial law.13 Goode has also been the leading proponent of a modern codification of English commercial law. If we consider what the Victorians did for us,14 then, in addition to radical and necessary restructuring of civil justice and the courts (including ultimately the emergence of the Commercial Court15), we must identify as important steps their cross-border Mercantile Law Amendment Acts (reciprocally adopting the best of both English and Scottish doctrines) and famous minicodifications.16 Yet, despite Goode’s persuasive advocacy,17 codification of commercial law has not gained any traction in the UK, or more narrowly in England and Wales. In this context, the analogy of the USA’s Uniform Commercial Code (UCC) breaks down in the absence of any significant federal–state divide and the dominance of the English and Welsh jurisdiction. The UCC was first launched in 1952, largely the creation of Professor Karl Llewellyn.18 It is a model statute, which requires implementation by each of the individual American states. Its express purpose is ‘to simplify, clarify and modernize the law governing commercial transactions; to permit the continued expansion of commercial transactions through custom, usage and the agreement of the parties; to make uniform the laws among the various jurisdictions’.19 The first two sets of aims are laudable, and applicable here. And,

11 Goode on Commercial Law (n 7) para 1.10. The text goes on to say: ‘Its scope is not clearly defined, and no two textbooks seem to adopt the same approach to the spheres of commercial activity that ought properly to be included in a work on the subject.’ 12 R Goode, Commercial Law in the Next Millennium (London, Sweet & Maxwell, 1997) 8–9. 13 Civil Procedure Rules 1998, SI 1998/3132 (CPR), r 58.1(2). See too the remit of the Circuit Commercial Courts (in London, Manchester, Cardiff, Bristol, Birmingham, Leeds and Mold(!)) in CPR, r 59. Both are now part of the (marketing) umbrella of the Business and Property Courts of England and Wales (nomenclature adopted in October 2017). 14 See A Rodger, ‘The Codification of Commercial Law in Victorian Britain’ (1992) 109 LQR 570. 15 V Veeder, ‘Mr Justice Lawrance: the “True Begetter” of the English Commercial Court’ (1994) 110 LQR 292. 16 The Bills of Exchange Act 1882, the Sale of Goods Act 1893 and (just into the Edwardian era) the Marine Insurance Act 1906, the last dealt with at length in FD Rose, Marine Insurance: Law and Practice, 2nd edn (London, Informa Law at Routledge, 2012). 17 R Goode, ‘The Codification of Commercial Law’ (1988) 14 Monash University Law Review 135. See also R Goode, ‘Insularity or Leadership? The Role of the United Kingdom in the Harmonisation of Commercial Law’ (2000) 50 ICLQ 751, 761–63 (arguing for a British Commercial Code); R Goode, ‘Removing the Obstacles to Commercial Law Reform’ (2007) 123 LQR 602. 18 For an English perspective on the history and philosophy of the UCC, see W Twining, Karl Llewellyn and the Realist Movement 2nd edn (Cambridge, Cambridge University Press, 2012) chs 11 and 12 (a work first published in 1973). The American literature is enormous. 19 The UCC comprises, after its introductory sections: sales (Art 2); leases (Art 2A); negotiable instruments (Art 3); bank deposits and collections (Art 4); wholesale wire transfers (Art 4A); letters of credit (rev Art 5); bulk transfers and bulk sales (rev Art 6); warehouse receipts, bills of lading and other documents of title (Art 7); investment securities (rev Art 8); and secured transactions (Art 9).

184 Gerard McMeel given the future position of the UK, the time has come for further consideration of whether its commercial law should be codified or restated, or at least whether the legislative underpinnings require modernisation. At least then it may be easier to provide an answer to the question, where can English commercial law be found?

III. Modifying Private Law Dogmas One of the most scintillating aspects of teaching commercial law at undergraduate level is the necessary discarding of private law shibboleths to which the attentive student had only been introduced in the preceding year or two. These often involve the architectonic doctrines of consideration, requiring a bargain or exchange before a promise or agreement is enforceable, and of privity, limiting the range of parties who can enforce. Applied mechanically or without exception, these doctrines could frustrate commercial activities and render transactions unenforceable, or not enforceable by the appropriate person. Whilst the common law was suspicious of third parties having or acquiring contractual rights, the imperatives of commerce favoured easy transferability of commercial instruments. I will consider the concept of negotiability further below, but will first address circumventions of the doctrine of consideration in business contexts. In commercial law courses which include the law of agency, it is essential to learn quickly that one highly significant commercial phenomenon – namely, a principal conferring actual authority upon another person to negotiate and trade on its behalf – is a species of deal which requires no consideration, but is nevertheless given full legal effect. If the law of personal property features in the commercial law course, or at least that portion of it concerned with dealings in goods, then the undermining of any incipient ‘consideration fundamentalism’ is continued by surveying the common law’s peculiar fascination with bailments – another widespread consensual arrangement which attracts legal incidents without the need for any bargain or exchange. In Volcafe Ltd v Cia Sud Americana de Vapores SA,20 the Supreme Court recently revisited the seminal bailment case of Coggs v Barnard.21 Lord Sumption described it as a ‘landmark case’. It should not be forgotten that the occasion for Holt CJ’s taxonomy of bailments was to determine the appropriate standard of care for a gratuitous bailee. Lord Holt’s reasoning posited a spectrum of possibilities, from the strict liability of the common carrier to liability for gross negligence in the case of a bailment exclusively for the benefit of the bailor. This specific discussion has not stood the test of time. The Supreme Court,22 despite some comparatively recent statements differentiating the standard of care of the bailee for reward from the gratuitous bailee, favoured a unitary standard of care for bailees and ditched the time-honoured concept of the common carrier.23 The court confirmed that ‘a bailee of goods is not an insurer. His duty is limited to taking reasonable care of the goods.’24

20 Volcafe Ltd v Cia Sud Americana de Vapores SA (t/a CSAV) [2018] UKSC 61, [2019] AC 358. 21 Coggs v Barnard (1703) 2 Ld Raym 909, 92 ER 107. 22 Volcafe (n 20) [8] (Lord Sumption). 23 Houghland v RR Low (Luxury Coaches) Ltd [1962] 1 QB 694 (CA); Sutcliffe v Chief Constable of West Yorkshire [1996] RTR 86 (CA). See M Bridge, Personal Property Law, 4th edn (Oxford, Oxford University Press, 2015) 64. 24 Volcafe (n 20) [8].

Are there Any General Principles of Commercial Law? 185 In Volcafe, Lord Sumption restated the evidential burden on bailees, gratutitous or otherwise, and the pragmatic justification which emerged for it: while the rule about the burden of proof in English law developed long before any pragmatic justification was advanced for it, its continued importance in the law of bailment has consistently been supported on the ground that because the bailee is in possession of the goods it may be difficult or impossible for anyone else to account for the loss or damage sustained by them.25

I am less sure that the reversal of the burden of proof was not motivated by the pragmatic approach characteristic of English commercial law. The categories of gratuitous agency and gratuitous bailment demonstrate that consensual commercial arrangements attract well-developed legal incidents irrespective of the supposedly narrow focus that the doctrine of consideration might appear to dictate. If a course on commercial law travels on to consider the financing of international trade, and in particular the documentary credit, then the easy dispensability of the doctrine of consideration – when it stands in the way of commerce and finance, and in particular abstract payment undertakings – is reiterated. The dual issues faced by the Supreme Court in MWB Business Exchange Centres Ltd v Rock Advertising Ltd26 divided audiences. My academic law colleagues eagerly anticipated a definitive ruling on the problems of consideration for contractual renegotiations, which has troubled thoughtful lawyers since Lord Blackburn’s near dissent in Foakes v Beer,27 and rumbled on to the unsatisfactory confession and avoidance of Williams v Roffey Bros & Nicholls (Contractors) Ltd.28 Lord Sumption thought the ‘points of distinction’ between the two cases ‘somewhat forced’.29 But the re-examination of these authorities was put off to another day, to groans from contract law lecturers. My practitioner colleagues and I were, on the other hand, excited about the treatment of restriction of variation clauses. The damp squib observations on consideration in MWB may have caught many legal academics’ eyes in 2018, but for me the brisk observations of the Supreme Court the previous year in Taurus Petroleum Ltd v State Oil Marketing Co of the Ministry of Oil, Republic of Iraq30 were much more significant. For the purposes of that case, it was important whether a promise under a letter of credit created a joint or sole debt. The majority of the Supreme Court held as a matter of construction that there was a sole beneficiary of the letter of credit. In an otherwise turgid speech, tracking too closely the judgments below, Lord Clarke of the majority stated: Given that conclusion on the construction of the letters of credit, Moore-Bick LJ did not have to consider a further submission made by Mr Pollock that, even if CBI was the beneficiary of the bank’s promises to pay under these letters of credit, they were not promises which it could enforce, because they were not supported by consideration. He did not consider this point in any detail but did say that he would be loath to hold, particularly in a commercial context, that a promise which both parties intended should be relied on was unenforceable for want of consideration. So would I.31

This is very striking for any banking or commercial lawyer. One of our leading commercial silks had a fallback argument that letters of credit were not enforceable because unsupported by consideration. The promise of the bank or banks to pay under a letter of credit is unilateral in nature

25 ibid

[10]. Business Exchange Centres Ltd v Rock Advertising Ltd [2018] UKSC 24, [2019] AC 119. v Beer (1884) 9 App Cas 605 (HL). 28 Williams v Roffey Bros & Nicholls (Contractors) Ltd [1991] 1 QB 1 (CA). 29 MWB (SC) (n 26) [18]. 30 Taurus Petroleum Ltd v state Oil Marketing Co of the Ministry of Oil, Republic of Iraq [2017] UKSC 64, [2018] AC 690. 31 ibid [25]. 26 MWB

27 Foakes

186 Gerard McMeel and is said under the international standard terms – the International Chamber of Commerce’s Uniform Customs and Practice for Documentary Credits – to constitute a ‘definite undertaking’ to pay by banks.32 And yet in common law jurisdictions the absence of any consideration from the beneficiary poses a potential problem. It would appear that no bank has ever taken the point, and the submission does not appear to have been pressed in the Supreme Court. Nevertheless, two experienced and conservatively inclined commercial judges, Moore-Bick LJ and Lord Clarke, breezily brushed away a no consideration argument. Contracts, seriously meant and relied upon in a commercial context, should be enforced irrespective of consideration. That would certainly be a desirable state of the law and, one may say, in cases of gratuitous agency, gratuitous bailment and abstract payment undertakings such as letters of credit, is effectively already the law. Doctrinal certainties yield to pragmatic realities. Last under this head, returning to the law of agency, commercial law students who had dutifully learnt that promissory estoppel, and perhaps other species of estoppel, will not create causes of action or new legal rights have to wrestle with the commercially essential topic of ostensible or apparent authority – sometimes described as agency by estoppel. This manifestly does create rights of action against any principal who has held out another as entitled to deal on its behalf for a third party who has relied upon that holding out. As usual, it may be possible to say that the sword changes hand, or some other guff, but the reality is that apparent or ostensible authority is a valuable doctrine which resists classical analysis. Indeed, in what is either a leading case33 or a controversial extension34 of the doctrine, Steyn LJ appeared to rationalise it by reference to the reasonable expectations of honest business people, which, as we shall see below, was his euphemism for good faith.

IV. The Leading Advocate of General Principles of Commercial Law In his article ‘The Codification of Commercial Law’,35 Goode ambitiously set out, in the context of a call for commercial codification, some 17 general features of commercial law. Eight propositions represent aspects of the ‘philosophy’ of commercial law, five are accorded the status of ‘concepts’ of commercial law and four demarcate the scope of commercial law. I will start with the last cluster, namely Goode’s propositions concerning the scope of commercial law. First, commercial law should be transactional, not institutional – concerned with trade, not with the vehicles for trade. Goode’s intention appears to be to exclude company law, partnership and, arguably, representation. (In passing, it is noteworthy that the first edition of Goode’s textbook did not address the law of agency, but that it did so from its second edition.) Secondly, commercial law’s constituency concerns merchants, and not consumers. Obviously, this attempts

32 International Chamber of Commerce, Uniform Customs and Practice for Documentary Credits – UCP 600 (2007 revision) Art 2 definitions of ‘credit’ and ‘confirmation’. 33 First Energy (UK) Ltd v Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194 (CA); see also Kelly v Fraser [2012] UKPC 25, [2013] 1 AC 450. For my view, see G McMeel, ‘Agency and the Retail Distribution of Financial Products’ in D Busch, L Macgregor and P Watts (eds), Agency Law and Commercial Practice (Oxford, Oxford University Press, 2016) 177. 34 See, eg P Watts, ‘Some Wear and Tear on Armagas v Mundogas: The Tension between Having and Wanting in the Law of Agency’ [2015] Lloyd’s Maritime and Commercial Law Quarterly 36. 35 Goode, ‘The Codification of Commercial Law’ (n 17).

Are there Any General Principles of Commercial Law? 187 to exclude business-to-consumer dealings, where public policy interventions are more significant. Thirdly, commercial law is centred on contracts and market usages. That says much about significant sources of the subject, or more particularly the source of detailed obligations and incidents for particular transactions. Fourthly, commercial law is concerned with repeat transactions, but, from the perspective of contract theory, is generally concerned with discrete and not relational contracts. That appears to be an indirect way of acknowledging the significant role of standard form contracts and standard terms, and their judicial exegesis, in the field. Goode’s most intriguing general principles are his ‘philosophy of commercial law’, embracing: (a) party autonomy; (b) predictability; (c) flexibility; (d) good faith; (e) the encouragement of self-help; (f) the facilitation of security interests; (g) the protection of vested interests; and (h) the protection of third parties. These eight mini-discussions have a high degree of abstraction. The first thing to note, which would probably cause no surprise to a civil lawyer, is that many of them are really general concepts or concerns of mainstream private law. From the perspective of a system whose starting point is its existing civil code, with an accompanying commercial code, the latter assumes the content of the former and provides points of departure for standard mercantile transactions. From an English perspective, Goode’s list may be said to be merely importing familiar concerns from private law. This borrowing is also reflected in the multiple pairs thrown up by the list. Some of these pairs will provide a main concern of later sections of this chapter. At this point, I shall briefly address the last two: the tension between the protection of vested interests and the protection of third parties can be said simply to replicate the central concern of property law. In the context of dealings with goods, this aspect of personal property law is already largely codified at the heart of the Sale of Goods Act 1979, representing predominantly nineteenth-century accommodations between the significant weight attributed to the rights of owners and the desire to recognise the legitimate interests of those dealing in good faith with representatives of the owners, or others entrusted with possession of the goods.36 Allocating the risk of fraud looms large. Save that personal property law is an essential building block of commercial law, there is not much that is distinctive here.37 Many of the cases both prior and subsequent to the various legislative interventions concern dealings between merchants, but the cases are as likely to feature individual buyers who are the victims of fraud, typically involving cars, an inherently mobile repository of wealth.38

36 There are two important recent accounts: first, L Gullifer in M Bridge et al (eds), The Law of Personal Property, 2nd edn (London, Sweet & Maxwell, 2017) ch 30, which places the exceptions to nemo dat quod habet in their proper context; and secondly, A Tettenborn, ‘Transfer of Chattels by Non-owners: Still an Open Problem’ [2018] CLJ 151, on the need for further reform. 37 Contrast Goode on Commercial Law (n 7) para 2.19: ‘land is governed by primarily by property law concepts, goods and choses in action by commercial law concepts’. That treats personal property as a conceptual element of commercial law, divorced from property law in respect of land. The better view is that personal property is a property law subject: see Bridge et al (n 36) para 1.057. 38 Indeed, the leading modern case of Shogun Finance Ltd v Hudson [2003] UKHL 62, [2004] 1 AC 919 is such, and concerned with the principal 20th-century statutory intervention: Hire-Purchase Act 1964, pt III.

188 Gerard McMeel A number of other questions arise from Goode’s list for this chapter. What are the limits of party autonomy? The tension between party autonomy and good faith lies at the heart of the general law of contract, and therefore unavoidably resurfaces in the lex specialis of commercial transactions. We can also note the tension between predictability and good faith, which has been prominent in recent judicial jousting. Again, the lack of distinctiveness from ordinary private law concerns may be noted. There is also the obvious potential tension between the needs of predictability and flexibility which is endemic. Flexibility also ties in to the facilitative philosophies of encouraging self-help and the willing recognition of security interests. Turning finally to what Goode labels the ‘concepts’ of commercial law, these are fivefold: (a) (b) (c) (d) (e)

the concept of a market; the importance of customs and usages; the importance of a course of dealing; the concept of negotiability; and the enforceability of abstract payment undertakings.

This is a very short list of distinctive characteristics for a major subject. The first three ‘concepts’ are again important sources for the general law of contract. Thus, the marketplace39 is part of the admissible background for the construction of contracts generally,40 and it will be relevant to determining expectation-based damages, as explicitly recognised in the Sale of Goods Act 1979, sections 50(3) and 51(3). Similarly, both a course of dealing, as a means of incorporation of standard terms, and the reception of custom and usage, to aid the interpretation of contracts or as a source of implied terms, are relevant features of general contract law. Here, Goode’s discussion may be said to be an exhortation to give these sources of obligation even more emphasis in the commercial arena. The fourth concept – negotiability – is different. If one were seeking to identify the core distinctiveness of English commercial law as it developed in the seventeenth and eighteenth centuries, then it would be the ready recognition of the transferability of commercial paper. Indeed, in any history of our mercantile law, this is the critical innovation.41 Whether achieved through some reception of the law merchant or by judicial innovation, such dealings were permitted despite the common law of contract’s refusal to recognise the assignability of choses in action and its restrictive approach to third party rights. The negotiability of commercial instruments is the paradigm instance of what I earlier called the modifying of private law dogma. It embraced both negotiable instruments embodying a right to the payment of money, such as bills of exchange or promissory notes, and documents of title to negotiable securities. Transferability also applied documents of title to goods, such as the bill of lading.42 Such documents are capable of being

39 For Professor Goode’s further development, see R Goode, ‘The Concept of a Market in Commercial Law’ [1991] Lloyd’s Maritime and Commercial Law Quarterly 177; Goode on Commercial Law (n 7) paras 4.24–4.44. The latter discussion also reflects the almost universal condemnation of the decision of the House of Lords in Hazell v Hammersmith & Fulham LBC [1992] 2 AC 1 (HL), including by E McKendrick, ‘Local Authorities and Swaps: Undermining the Market?’ in R Cranston (ed), Making Commercial Law: Essays in Honour of Roy Goode (Oxford, Oxford University Press, 1997). Given that interest rate swaps are a comparatively recent invention only dating back to the 1980s, and given the dangers posed by their speculative use, the better view is that public law trumping commercial law in that context was both inevitable and sound. 40 Reardon-Smith Ltd v Hansen-Tangen (The Diana Prosperity) [1976] 1 WLR 989 (HL) 995–97. 41 The best account is JS Rogers, The Early History of the Law of Bills and Notes – A Study of the Origins of A nglo-American Commercial Law (Cambridge, Cambridge University Press, 1995). 42 Statutory intervention was required to facilitate the practice of transfer, conferring rights of suit on certain transferees, originally by the Bills of Lading Act 1855 and now the Carriage of Goods by Sea Act 1992.

Are there Any General Principles of Commercial Law? 189 bailed and, in particular, pledged. So the ability to pledge the bill of lading whilst the goods are at sea is a significant feature of commerce in facilitating and financing export sales, with an express clause agreeing to pledge the bill and other shipping documents a common feature of banks’ standard terms for documentary credits. Some caution is required here, as ‘negotiability’ is used in two senses. First, negotiability may mean only that the documentary intangible is transferable, such as the bill of lading.43 Secondly, in the context of negotiable instruments, negotiability means both that the document is transferable and that the transferee may acquire rights greater than those of the transferor. In respect of the latter, the good faith purchaser of the principal species of negotiable instrument, the bill of exchange, generally holds it free of any defects of title of prior parties.44 But the crucial point is that ready transferability of commercial paper was recognised. The fifth concept also deserves stronger emphasis – the recognition of what Goode labels abstract payment undertakings, such as documentary credits (or letters of credits) and performance guarantees.45 This is another example of the lex specialis of commercial or banking law departing from the private law dogma of consideration. As stated above, the autonomy of such undertakings receives a fillip from recent dicta of the Supreme Court in Taurus.46

V. Freedom of Contract versus Sanctity of Contract Modern written contracts contain various provisions aimed at protecting the integrity of the instrument, by attempting to provide against it being contradicted, supplemented or easily varied by arrangements outside the four corners of the document. Entire agreement clauses, no reliance clauses, and even no representation and no advice clauses have featured prominently in recent commercial and financial cases.47 Until recently, ‘no oral modification clauses’, or, as I prefer to call them, ‘restriction of variation clauses’ – ie clauses which purport to restrict variation or waiver, such as requiring variation to be in writing and signed – have not attracted the same attention. Nevertheless, they were routinely inserted by drafters, who probably assumed that if push came to shove they would probably not work. The reason was encapsulated by the great Cardozo J in the New York Court of Appeals in Alfred C Beatty v Guggenheim Exploration Co: Those who make a contract, may unmake it. The clause which forbids a change, may be changed like any other. The prohibition of oral waiver, may itself be waived … What is excluded by one act, is restored by another. You may put it out by the door, it is back through the window. Whenever two men contract, no limitation self-imposed can destroy their power to contract again.48

43 ‘This form of transfer has also traditionally, but idiosyncratically, been referred to as ‘negotiability’: JI MacWilliam Co Inc v Mediterranean Shipping Co SA (The Rafaela S) [2003] EWCA Civ 556, [2004] QB 702 [1] (Rix LJ). Contrast Rogers (n 41), contending that negotiability originally meant merely transferability, and that it is the dual sense – involving the coupling with a bona fide purchase defence in respect of negotiable instruments – which came later. 44 Bills of Exchange Act 1882, s 27. 45 For further development, see R Goode, ‘Abstract Payment Undertakings’ in P Cane and J Stapleton (eds), Essays for Patrick Atiyah (Oxford, Oxford University Press, 1991); Goode on Commercial Law (n 7) paras 18.17 ff and 35.52–35.57. 46 Taurus (n 30) [25]. 47 G McMeel, McMeel on the Construction of Contracts, 3rd edn (Oxford, Oxford University Press, 2017) ch 26; especially in relation to their use by banks, see G McMeel, ‘The Impact of Exemption Clauses and Disclaimers: Construction, Contractual Estoppel and Public Policy’ in A Dyson, J Goudkamp and F Wilmot-Smith (eds), Defences in Contract (Oxford, Hart Publishing, 2017) 239. 48 Alfred C Beatty v Guggenheim Exploration Co 225 NY 380, 387–88 (1919).

190 Gerard McMeel The premiss is that freedom of contract or party autonomy entails that parties are always free to vary their contracts, or waive the other’s obligations or breaches, whether in writing, orally or by conduct. Prior to 2016, English law yielded inconsistent, unreported Court of Appeal authorities.49 That year, the Court of Appeal on two occasions adopted the party autonomy rationale for refusing to give effect to restriction of variation clauses, first in Globe Motors, Inc v TRW Lucas Varity Electric Steering Ltd,50 then in MWB Business Exchange Centres Ltd v Rock Advertising.51 However, the latter decision was reversed on appeal, and the restriction of variation clause was given full effect. It is rare that I am taken by surprise by a decision of the Supreme Court in a contract law case,52 but MWB was, as the partial dissentient Lord Briggs noted, something of a ‘clean break’ from the dominant common law approach.53 One cannot but admire Lord Sumption’s brisk and muscular judgment in MWB. However, it erects a wholly unrealistic hurdle for potentially deserving parties who seek to escape from what the Supreme Court saw to be the virtues of such clauses. The context in which such disputes will be played out will be applications for accelerated justice, by way of striking out claims, or more typically counterclaims, and for summary judgment. MWB itself concerned a commercial licence of office premises, a build-up of arrears and a claim that there had been a negotiated rescheduling of payments by telephone conversation with a relatively junior employee, resulting in payments somewhat shy of the licensor’s original full contractual entitlement. The judge at first instance held that the extended practical benefit test for consideration was satisfied, but that the licensee lost because of a restriction of variation provision. The Court of Appeal upheld the judgment on consideration, but reversed it on the effect of the clause. The Supreme Court determined the case solely on the basis of the restriction of variation clause, to which it gave full effect, overruling Globe Motors. Lord Sumption’s reasoning in MWB moved in four stages. First, English law recognised no formalities for contracts in general, but did impose formalities for particular species, such as dealings in land. Secondly, despite mixed authority on the topic, English law does and should enforce such clauses. Thirdly, his Lordship tackled head-on the ‘party autonomy’ argument, which found favour in the Court of Appeal:54 Party autonomy operates up to the point when the contract is made, but thereafter only to the extent that the contract allows. Nearly all contracts bind the parties to some course of action, and to that extent restrict their autonomy. The real offence against party autonomy is the suggestion that they cannot bind themselves as to the form of any variation, even if that is what they have agreed.55

No authority is cited by Lord Sumption for this fallacy concerning the reach of party autonomy. It echoes the careful discussion by Professor Charles Fried in his highly influential

49 World Online Telecom UK Ltd (formerly Localtel Ltd) v I-Way Ltd [2002] EWCA Civ 413; United Bank Ltd v Asif (CA, 11 February 2000). See also Energy Venture Partners Ltd v Malabu Oil and Gas Ltd [2013] EWHC 2118 (Comm) [273] (Gloster LJ). 50 Globe Motors, Inc v TRW Lucas Varity Electric Steering Ltd [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601 [95]–[104] (Beatson LJ). See also [117] (Underhill LJ) and [120] (Moore-Bick LJ). 51 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2016] EWCA Civ 553, [2017] QB 604 [19]–[36] (Kitchin LJ). 52 The last occasion was the manifestly instrumental outcome in the ‘bank charges’ litigation: Office of Fair Trading v Abbey National plc [2009] UKSC 6, [2010] 1 AC 696. 53 MWB (SC) (n 26) [32]. 54 MWB (CA) (n 51) [34] (Kitchin LJ). 55 MWB (SC) (n 26) [11].

Are there Any General Principles of Commercial Law? 191 liberal p hilosophical account of the law of contract in Contract as Promise.56 It in the nature of promising, in order to give meaningful effect to the freedom to choose, that one must have the ability to restrict one’s future conduct to enable mutual cooperative behaviour. Freedom of contract necessitates respect for the arrangements once concluded: pacta sunt servanda. However, it is clear that Lord Sumption pushes his fallacy argument too far. On his logic, contracts that bind in perpetuity would be inescapable. There is an unusual tension engendered by these clauses between freedom of contract and sanctity of contract, and the common law instinct has been to give effect to proven informal deals bypassing such clauses. Fourthly, Lord Sumption identified three legitimate commercial reasons for including such a clause: it prevents abusive attempts to raise defences to frustrate claims to summary judgment; it obviates disputes about whether a variation took place and its content; and it assists corporations in policing the conferral of authority for particular transactions. No public policy required the recognition of such clauses to be restricted. Lord Sumption employed a pleasing diversity of source material to support his preferred approach. First, he noted academic comment favourable to such clauses.57 Secondly, non-English legislative material provided some effect for such clauses, including the USA’s Uniform Commercial Code, section 2-209, the UNIDROIT Principles of International Commercial Contracts,58 Article 2.1.18 and the Vienna Convention on Contracts for the International Sale of Goods 1980, Article 29(2). For example, the UNIDROIT Principles provide: A contract in writing which contains a clause requiring any modification or termination by agreement to be in a particular form may not be otherwise modified or terminated. However, a party may be precluded by its conduct from asserting such a clause to the extent that the other party has reasonably acted in reliance on that conduct.

Thirdly, in favour of the general common law approach, he cited Australian and Canadian cases, together with German academic commentary. Lastly, he recited a long list of cases giving effect to ‘entire agreement’ clauses.59 He seemed to recognise that, as a matter of construction, entire agreement clauses are of limited effect, merely excluding collateral warranties (and not claims for misrepresentation), and possibly not even genuine collateral agreements supported by their own consideration. But he failed to cite Court of Appeal authority establishing that such clauses are exemption clauses and subject to review under the Unfair Contract Terms Act 1977.60 As we shall see, non-reliance clauses and the like are undoubtedly subject to both section 3(1) of the Misrepresentation Act 1967 and the Unfair Contract Terms Act 1977, because, as a matter of substance, they seek to exclude or restrict claims in misrepresentation which might otherwise succeed.

56 C Fried, Contract as Promise – A Theory of Contractual Obligation, 2nd edn (Oxford, Oxford University Press, 2015) (a work first published in 1981) 14. 57 J Morgan, ‘Contracting for Self-Denial: On Enforcing “No Oral Modification” Clauses’ (2017) 76 CLJ 589; E McKendrick, ‘The Legal Effect of an Anti-oral Variation Clause’ (2017) 32 Journal of International Banking Law and Regulation 439; J O’Sullivan, ‘Unconsidered Modifications’ (2017) 133 LQR 191. 58 UNIDROIT, UNIDROIT Principles of International Commercial Contracts 2016, 4th edn (Rome, UNIDROIT, 2016). There is more than a touch of irony in Lord Sumption deploying UNIDROIT Principles to support his approach when it is difficult to see him with equal enthusiasm endorsing that instrument’s provisions on recognising a general duty of good faith. 59 From Brikom Investments Ltd v Carr [1979] QB 467 (CA), casting doubt on Lord Denning MR’s dismissal of such a clause, via Inntrepreneur Pub Co Ltd v East Crown Ltd [2000] 2 Lloyd’s Rep 611 (Ch) and Deepak Fertilisers & Petrochemical Corp Ltd v ICI Chemicals & Polymers Ltd [1998] 2 Lloyd’s Rep 139 (Com Ct) 168 (Rix J) and (1999) 1 Lloyd’s Rep 387 (CA) [34], to more recent first instance authority, including First Tower Trustees Ltd v CDS (Superstores International) Ltd [2017] EWHC 891 (Ch), [2017] 4 WLR 73 (which might miss the point of that case). 60 AXA Sun Life Services plc v Campbell Martin Ltd [2011] EWCA Civ 133, [2011] 2 Lloyd’s Rep 1.

192 Gerard McMeel The main problem in practice with MWB is not its giving prima facie effect to such restriction of variation clauses, but rather the almost impossible bar the majority set for qualifying such recognition by an argument based on estoppel. On one reading, the majority appear to require both an agreement to vary the contract and an explicit agreement to depart from the restriction of variation clause. It is difficult to envisage such a two-pronged agreement in real life. The crucial passage in Lord Sumption’s judgment is: The enforcement of No Oral Modification clauses carries with it the risk that a party may act on the contract as varied, for example by performing it, and then find itself unable to enforce it. It will be recalled that both the Vienna Convention and the UNIDROIT model code qualify the principle that effect is given to No Oral Modification clauses, by stating that a party may be precluded by his conduct from relying on such a provision to the extent that the other party has relied (or reasonably relied) on that conduct. In some legal systems this result would follow from the concepts of contractual good faith or abuse of rights. In England, the safeguard against injustice lies in the various doctrines of estoppel. This is not the place to explore the circumstances in which a person can be estopped from relying on a contractual provision laying down conditions for the formal validity of a variation. The courts below rightly held that the minimal steps taken by Rock Advertising were not enough to support any estoppel defences. I would merely point out that the scope of estoppel cannot be so broad as to destroy the whole advantage of certainty for which the parties stipulated when they agreed upon terms including the No Oral Modification clause. At the very least, (i) there would have to be some words or conduct unequivocally representing that the variation was valid notwithstanding its informality; and (ii) something more would be required for this purpose than the informal promise itself.61

The one crumb of comfort is that this discussion, tentative as it is, is explicitly obiter dicta (and recognised as such by the Appeal Cases law reporter). Neither the UNIDROIT Principles nor the Vienna Convention, in their qualifications of the prima facie recognition of the effect of such clauses, restrict those qualifications to such improbable variations, perhaps forcing business people into perjured evidence that the clause had been discussed. Lord Briggs dissented from the majority’s approach and would have given effect to any renegotiation which either expressly or by necessary implication departed from the clause. It will be left to judges wrestling with summary judgment and strike out applications (which are likely to become more numerous) to work out how high a hurdle should be required of the party alleging an estoppel. It is to be hoped that the first thoughts of the Supreme Court are watered down in practice. The question is also likely to arise whether restriction of variation clauses now fall within the Unfair Contract Terms Act 1977.

VI. The Limits of Party Autonomy In respect of legislative control, 2018 witnessed the most important decision in the law of contract for at least a decade, a decision which has been immediately appreciated as having reversed disturbing trends outside of its particular context of commercial leases. Without being too reductionist, First Tower Trustees Ltd v CDS (Superstores International) Ltd62 makes the first-year law student point that public legislation is a superior source of law to private contract.

61 MWB (SC) (n 26) [16], citing Actionstrength Ltd (t/a Vital Resources) v International Glass Engineering In Gl En SpA [2003] UKHL 17, [2003] 2 AC 541 [9] (Lord Bingham), [51] (Lord Walker). 62 First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396, [2019] 1 WLR 637. On 10 January 2019, the Supreme Court refused the landlords permission to appeal: [2019] 1 WLR 1399.

Are there Any General Principles of Commercial Law? 193 To u nderstand why such an obvious proposition necessitated such careful analysis by two leading contract lawyers, one has to recall what English judges had managed to persuade themselves the law was in the wake of the global financial crisis, especially where investors brought claims against banks. Entire agreement, no advice, no fiduciary relationship and no reliance clauses were not exemption clauses, but were duty-defining or ‘basis clauses’, both destroying any arguments of poor advice and misrepresentation by way of causes of action, and also insulating themselves and the contract from review under the Unfair Contract Terms Act 1977 and the Misrepresentation Act 1967. The nadir was reached in the swaps mis-selling cases of Crestsign Ltd v National Westminster Bank63 and especially Thornbridge Ltd v Barclays Bank plc.64 I have subjected these authorities to detailed criticism elsewhere.65 Fortunately, the Court of Appeal has since joined in, and expressly disapproved of the analysis in Thornbridge.66 First Tower featured warehouse premises seriously impaired by asbestos issues, initially unbeknownst to the landlords. The normal conveyancing process of pre-contractual enquiries was followed. During negotiations, but after responses to those enquiries, a third party report identified this concern, falsifying the landlords’ answer to enquiries that they were unaware of any environmental problems. An express clause required answers to be corrected if the landlord became aware of any issue that rendered an answer incorrect. The agreement for the lease had a clause stating that the tenant acknowledged that it had not relied upon any statements or representations save those in written responses to the inquiries. In contrast, the lease starkly stated in clause 5.8 that the tenant had not relied upon any statements or representations made by or on behalf of the landlord. The landlords’ counsel insisted this was a ‘basis clause’ and not an exemption clause. The judgments of both Lewison and Leggatt LJJ repay careful study, and are likely to be cited at length over coming years. A few highlights will have to suffice. Lewison LJ acknowledged that ‘contractual estoppel’ reasoning was ‘now firmly established at this level in the judicial hierarchy’, but that ‘the position at common law is not the end of the enquiry’.67 Whether a clause was an exemption clause was a question of substance. The courts were bound to give effect to the clear statutory policy of section 3 of the Misrepresentation Act 1967 to prevent parties escaping liability for misrepresentation unless it is reasonable to do so.68 Applying the test, Lewison LJ agreed with the judge that, whilst the balanced clause in the agreement for the lease was reasonable, clause 5.8 in the lease would render the important process of pre-contractual inquiries pointless. Leggatt LJ was most scathing about the ‘so-called “basis clause”’ analysis put forward for the landlord.69 Clever drafting should not subvert the application of the statutory reasonableness standard.70 Parliament deliberately extended section 3(1) of the Misrepresentation Act 1967 to all

63 Crestsign Ltd v National Westminster Bank plc [2014] EWHC 3043 (Ch), [2015] 2 All ER (Comm) 133. 64 Thornbridge Ltd v Barclays Bank plc [2015] EWHC 3430 (QB). 65 McMeel, ‘The Impact of Exemption Clauses and Disclaimers’ (n 47) 239, 259–71. See also G McMeel, ‘Documentary Fundamentalism in the Senior Courts: The Myth of Contractual Estoppel’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 185. 66 First Tower (n 62) [44], [49]–[50], [65]–[66] (Lewison LJ), [104], [110] (Leggatt LJ). 67 ibid [47], [49]. Sir Kim Lewison has been more critical of contractual estoppel reasoning extrajudicially: K Lewison, ‘How Far Can You Go? The Limits of Contractual Estoppel’ (COMBAR Lecture, 12 February 2019). 68 One area where Lewison LJ can be criticised is his view, repeated on a number of occasions, that ‘the aspiration of certainty’ is relevant to the statutory test of reasonableness: First Tower (n 62) [69]–[70]. Commercial certainty is not a relevant factor, and not mandated by the Act. It would always potentially tip the balance in favour of reasonableness. Clauses which exclude the minimum decencies, like cl 5.8 here, are very certain, but extremely unfair. 69 First Tower (n 62) [90]. 70 ibid [99].

194 Gerard McMeel contracts, making it the most intrusive limit on exemption clauses in English law. Leggatt LJ then made a very important statement of principle: The importance which English law attaches to the freedom of parties to contract on whatever terms they choose depends crucially on the assumption that their consent to the terms of the contract has been obtained fairly. That is not the case where one party’s consent has been induced by a misrepresentation made by the other contracting party. Misrepresentation is a paradigm ‘vitiating factor’ which undermines the validity of a contract. This does not mean that a party cannot choose to give up the right to complain that its consent to the terms of the contract was obtained by misrepresentation. But insofar as a contract term is said to have removed that right, a control mechanism is needed to ensure that this term was a fair and reasonable one to include. That, at all events, is the policy which Parliament has thought it right to adopt. It is the duty of the courts to uphold and not to subvert that policy choice.71

The rebuke to judges who have ignored the will of Parliament, of whom Moulder J in Thornbridge is only the most extreme example, is more powerful for its restraint. Statute trumps contract. It should not have needed to be said at this level. Party autonomy must always yield to overriding public policy.

VII. Party Autonomy, Certainty and Good Faith There are two oppositions here. First, party autonomy versus good faith. Secondly, certainty versus good faith. These are necessarily linked. Here, I can only touch upon these long-running debates.72 There are potentially three broad tenable positions as to the status of good faith in English contract and commercial law: (i) no doctrine of good faith; (ii) the functionally equivalent doctrines approach – an analysis grounded in the techniques of comparative law – where it is said that good faith is implemented through other doctrines; or (iii) a more general principle of some sort. The orthodox position is that English law does not recognise any overarching obligation to act in good faith in the negotiation or performance of contracts, or in their construction.73 These three positions are reflected in three well-known judicial perspectives. The first is the hostile or sceptical approach, represented by Walford v Miles, where Lord Ackner stated that a duty to negotiate in good faith was ‘inherently repugnant’ to the adversarial position of negotiating parties and ‘unworkable in practice’.74 The second is the functional equivalence approach advocated by two leading English judges, Lord Bingham and Lord Steyn. In Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd,75 Bingham LJ, who had a long-standing interest in comparative law, identified ‘piecemeal solutions’ developed at common law, in equity and by legislation which played an analogous role to good faith in civil law systems. Similarly, Lord Steyn, writing extrajudicially, observed that English law tends to use discrete doctrinal tools, rather than

71 ibid [104]. 72 Having apparently assumed that good faith was a non-negotiable aspect of the philosophy of commercial law in his ‘The Codification of Commercial Law’ (n 17), Goode later appeared to beat a tactical retreat in his Hamlyn Lectures: Goode (n 12) 18–20. 73 Professor Burrows’s Restatement proceeds on the basis that there is no ‘free-standing’ duty to perform in good faith at common law: A Burrows, A Restatement of the English Law of Contract (Oxford, Oxford University Press, 2016) 93. 74 Walford v Miles [1992] 2 AC 128 (HL) 138 (Lord Ackner). 75 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433 (CA). See also T Bingham, ‘The Law as the Handmaid of Commerce’ (Sultan Azlan Shah Lecture 2001) in T Bingham, Lives of the Law – Selected Essays and Speeches 2000–2010 (Oxford, Oxford University Press, 2011).

Are there Any General Principles of Commercial Law? 195 an overarching principle: ‘Thus the continental systems recognize an overarching duty of good faith in the performance of contracts. The common law achieves similar results by a resort to implied terms, rectification and estoppel.’76 In a number of cases, Lord Steyn invoked the euphemism ‘reasonable expectations of honest men’ as a proxy for good faith.77 The third approach is a bolder synthesis advocated by Sir George Leggatt. He has been the most prominent judicial proponent of greater recognition of the role of good faith in English commercial contract law in a number of first instance cases, starting with his decision in Yam Seng Pte Ltd v International Trade Corp Ltd,78 where he concluded that there was nothing ‘novel or foreign’79 in the concept of good faith and fair dealing, and that the supposed traditional English hostility is ‘misplaced’.80 Similarly, in MSC Mediterranean Shipping Co SA v Cottonex Anstalt,81 in the context of contractual discretions, Leggatt J referred to the ‘increasing recognition in the common law world of the need for good faith in contractual dealings’.82 In stark contrast, on appeal, Moore-Bick LJ restated traditional English hostility: ‘There is in my view a real danger that if a general principle of good faith were established it would be invoked as often to undermine as to support the terms in which the parties have reached agreement.’83 This is very strong language, unconsciously echoing the ‘dangerous supplements’ of the half-forgotten critical legal studies movement, and comes close to Gunther Tuebner’s ‘legal irritants’84 or Alan Watson’s ‘legal transplants’. Good faith here is once again being erected as the opponent of legal certainty, as also intimated by Lord Falconer, cited above.85 The conclusion I have reached on this debate is that only the latter two approaches are available – ie the functionally equivalent doctrines approach or some commitment to a general principle – since everyone accepts as a minimum that there are at least some discrete doctrines or rules occupying the space filled by civil codes’ commitments to good faith.86 Even an academic good faith sceptic such as Professor Michael Bridge accepts that other doctrines are doing much of the work in English law, although he prefers the surgical precision of those more defined rules – for Bridge, good faith ‘might amount to little more than a moral compass guiding the development of the law’.87 I would welcome a morally grounded explanation of our system’s restraints on unbridled freedom and sanctity of contract. Promise-keeping is itself grounded in social morality, and its limits must have some connection with the same source. In those circumstances, the debate must now be conducted in terms of the proper role of good

76 Lord Steyn, ‘Interpretation: Legal Texts and their Landscape’ in BS Markesinis (ed), The Clifford Chance Millennium Lectures – The Coming Together of the Common Law and the Civil Law (Oxford, Hart Publishing, 2000) 79. See also Sir J Steyn, ‘The Role of Good Faith and Fair Dealing in Contract: A Hair-Shirt Philosophy?’ [1991] Denning Law Journal 131, 133. 77 eg in the context of formation: G Percy Trentham Ltd v Archital Luxfer Ltd [1993] 1 Lloyd’s Rep 25 (CA) 27. 78 Yam Seng Pte Ltd v International Trade Corp Ltd [2013] EWHC 111 (QB), [2013] 1 All ER (Comm) 1321. See also Sir G Leggatt, ‘Contractual Duties of Good Faith’ (COMBAR Lecture, 18 October 2016); Al Nehayan v Kent [2018] EWHC 333 (Comm). 79 Yam Seng (n 78) [145]. 80 ibid [153]. 81 MSC Mediterranean Shipping Co SA v Cottonex Anstalt [2015] EWHC 283 (Comm), [2015] 2 All ER (Comm) 614 [97]. 82 ibid [97]. 83 MSC Mediterranean Shipping Co SA v Cottonex Anstalt [2016] EWCA Civ 789, [2017] 1 All ER (Comm) 483 [45]. 84 G Teubner, ‘Legal Irritants: Good Faith in British Law or How Unifying Law Ends Up in New Divergences’ (1998) 61 MLR 11. 85 See the text to n 2 above. 86 G McMeel, ‘Foucault’s Pendulum: Text, Context and Good Faith in Contract Law’ [2017] Current Legal Problems 365, 390–97. 87 M Bridge, ‘Doubting Good Faith’ (2005) 11 New Zealand Business Law Quarterly 430, 431.

196 Gerard McMeel faith: (i) as merely a c omparative law grouping of discrete rules and doctrines; (ii) as an organising concept for those doctrines, providing a grounding for future development or as a cross-check for difficult cases; or (iii) as a general principle of good faith. If a general principle is preferred, then further questions arise as to the principle’s application. Should it apply (i) only in the context of negotiation and formation, such as duties of disclosure or liability for wasted expenditure where negotiations are broken off; (ii) in the context of performance, such as by expressly supplementing techniques of interpretation and implication; (iii) in the context of enforcement, such as limiting abusive use of termination rights; or (iv) all of the foregoing? On examination, the position in English law is not very different from that embodied in the fudge in the Vienna Convention on Contracts for the International Sale of Goods 1980, Article 7. Having rejected a general principle, the Diplomatic Conference instead made the ‘observance of good faith in international trade’ an interpretive tool for the Convention itself, and therefore inevitably one of its general principles, at the very least for gap-plugging purposes. In the result, good faith has a rather centrifugal impact, with myriad interpretations of its status in the Convention possible, often perhaps influenced by the background of the commentator. All of the foregoing illustrate that there is no easy answer to the inevitable tensions between certainty and good faith, and party autonomy and good faith.

VIII. Facilitation, Intangible Assets and Cryptoassets I wish finally to take one current topic to explore briefly the facilitative philosophy which proponents of the genius of English commercial law often flaunt. This is an area where the flexibility of personal property law is critical. As I like to say, personal property law is the law of property with all the accumulated learning about interests in land taken out. As Andrew Bell noted, it is personal property that is charged with accepting new forms of asset into the law’s protective grip: ‘the list of personal property is an open-ended one: any novel phenomenon that is recognised as property will in practice be classified as personal property’.88 In recent times, judges have grappled with the proprietary status of milk quotas,89 human body products90 and carbon credits.91 An old debate may prove significant here. Blackstone treated the division of personal property into things in possession and things in action as apparently exhaustive.92 More stridently, in Colonial Bank v Whinney,93 Fry LJ stated: ‘all personal things are either in possession or in action. The law knows no tertium quid between the two.’94 A different view is that there are two categories of intangible property, namely things in action and ‘other intangible property’. In support of this alternative view, one might note the Patents Act 1977, which provides that neither a patent nor a patent application is a thing in action, even though both are classified as personal property.95 To similar effect, in A-G of Hong Kong v Nai-Keung,96 the Privy Council held, when interpreting 88 A Bell, Modern Law of Personal Property in England and Ireland (London, Butterworths, 1989) 1. 89 Swift v Dairywise Farms Ltd [2000] 1 WLR 1177 (Ch); affirmed [2001] EWCA Civ 145, [2003] 1 WLR 1606 (Note). 90 Yearworth v North Bristol NHS Trust [2009] EWCA Civ 37, [2010] QB 1. 91 Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2013] Ch 156. 92 Sir William Blackstone: W Blackstone, Commentaries on the Laws of England (1st edn, 1765–69; University of Chicago reprint edn, 1979) vol II, 389. 93 Colonial Bank v Whinney (1885) 30 Ch D 261 (CA). 94 ibid 285. The House of Lords endorsed Fry LJ’s judgment: (1886) 11 App Cas 426 (HL). 95 Patents Act 1977, s 30(1). 96 Attorney-General of Hong Kong v Nai-Keung [1987] 1 WLR 1339 (PC).

Are there Any General Principles of Commercial Law? 197 Hong Kong legislation, that textile export quotas were not things in action, but did constitute ‘other intangible property’; and in England, carbon trading units were held to be probably ‘other intangible property’.97 In contrast, in Your Response Ltd v Datateam Business Media Ltd,98 the Court of Appeal preferred the first and narrower view, in the course of deciding that an electronic database was not capable of being possessed and therefore subject to a common law lien. Database rights were deemed to be property rights by secondary legislation, implementing EU law.99 Rejecting a submission that the database constituted intangible property, but not a chose in action, Moore-Bick LJ considered that Colonial Bank v Whinney made it ‘very difficult to accept that the common law recognises the existence of intangible property other than choses in action (apart from patents, which are subject to statutory classification)’.100 However, his Lordship was content to rest his conclusion on House of Lords authority that the tort of conversion did not extend to intangible property.101 This reluctance to accept a broader category of intangible property, not tied to the traditional concept of things in action, is likely to come under increasing pressure with the proliferation of new forms of wealth, and is not in line with a facilitative philosophy. Unworldly attempts to force software and databases into the tangible category of goods have now been rejected. Not long ago, when software was stored on a physical disk or data stick, an analogy with sale of goods on occasion proved persuasive. However, developments in technology whereby software is now almost invariably downloaded in purely digital format means that any debate about whether software corresponds to goods or is purely intangible has now been resolved in favour of software as an intangible species of property. In Computer Associates UK Ltd v Software Incubator Ltd,102 the Court of Appeal ruled that software generally sold as downloads did not constitute goods for the purposes of the Commercial Agents regime,103 either as a matter of EU or UK law. The distinction between goods and intangible property was well recognised in the authorities and the textbooks in that field.104 Gloster LJ made reference to authority in a wide number of contexts and concluded: ‘This court cannot simply ignore the weight of judicial authority that supports maintaining the tangible/intangible distinction.’105 The court observed that the development of a bespoke regime for the supply of digital content to consumers in the Consumer Rights Act 2015 demonstrated that these matters involved policy considerations which were a matter for Parliament.106 In Your Response, Moore-Bick LJ opined that a database ‘if it constitutes property at all, does not constitute property of a kind that is susceptible of possession or of being the subject of the tort of conversion’.107 Similarly, information or documents held in electronic form are not amenable to the jurisdiction to grant delivery up of goods under the Torts (Interference with Goods) Act 1977.108 The better view is that these assets are, as common sense suggests, intangible in nature; that they analogous to, but not identical to, things in action;

97 Armstrong (n 91) [61]. 98 Your Response Ltd v Datateam Business Media Ltd [2014] EWCA Civ 281, [2015] QB 41. 99 Copyright and Rights in Databases Regulations 1997, SI 1997/3032, regs 13 and 16. 100 Your Response (n 98) [26]. 101 OBG Ltd v Allan [2007] UKHL 21, [2008] AC 1. 102 Computer Associates UK Ltd v Software Incubator Ltd [2018] EWCA Civ 518, [2019] Bus LR 522. 103 Commercial Agents (Council Directive) Regulations 1993, SI 1993/3053. 104 Computer Associates (n 102) [25]. 105 ibid [55]. 106 ibid [56], [67]. 107 Your Response (n 98) [17]. That remark is wrong about the proprietary status in light of Copyright and Rights in Databases Regulations 1997, SI 1997/3032, reg 13. 108 Thunder Air Ltd v Hilmarsson [2008] EWHC 355 (Ch) [29] (Patten J); Environment Agency v Churngold Recycling Ltd [2014] EWCA Civ 909, [2015] Env LR 13.

198 Gerard McMeel and that they should be seen as a category within intangible property. In terms of vindication of rights, the possessory remedies are inapt, and again the better view is that the appropriate cause of action is restitutionary in form, albeit not grounded on unjust enrichment.109 Cryptocurrencies,110 or more broadly cryptoassets, are a fashionable topic. Personally, I have encountered too many Ponzi schemes in the course of my practice to be an enthusiast. Nevertheless, they raise significant issues of public policy111 and, in this context, the question whether such interests will be afforded the status of property as a matter of law. The government has set up a UK Jurisdiction Taskforce (UKJT),112 which published a consultation on cryptoassets in May 2019.113 The UKJT, chaired by Sir Geoffrey Vos, the Chancellor of the High Court, has the objective of demonstrating that English law and English jurisdiction can provide an appropriate legal foundation for these innovations. This is very much part of the tradition of cheerleading for England as a centre for commercial dispute resolution. Sir Geoffrey Vos has himself recently addressed the question whether cryptoassets can be regarded as property under English law in an extrajudicial lecture.114 He identified as a potential problem that, whilst intangible assets were recognised in English law, a cryptoasset did not generate a right against another person like a thing in action or a bank account. The UKJT therefore consulted on whether a cryptoasset is capable of being recognised as property, together with related questions as to whether it is personal property, a thing in action or some other species of personal property. Oddly, the questions identified included whether a cryptoasset can be the subject of a bailment.115 The better view is that cryptoassets are a species of intangible personal property, albeit not a thing in action, reinvigorating the debate about whether the categories of things in possession and things in action are exhaustive. This solution should be within the grasp of judicial development, given the case law on carbon credits and other intangibles. Nevertheless, as with database rights, the safer course may be express legislative recognition. One possible recommendation was that a short Act of Parliament clarifying and settling the proprietary status of cryptoassets be adopted.116 However the UKJT has now endorsed what is clearly intended to be a definitive statement that as a matter of English law cryptoassets have all the indicia of property, are intangible in nature, and are not capable of being possessed or bailed. The statement appears to accept that such interests are ‘other’ intangible property rather than things in action.117 The recent flurry of reports from policymakers domestically and internationally on new repositories of wealth is set to test the

109 The High Court decision in Armstrong (n 91) may demonstrate the way forward. See Bridge et al (n 36) para 1.042. 110 D Fox and S Green (eds), Cryptocurrencies in Public and Private Law (Oxford, Oxford University Press, 2019); R Cranston et al, Principles of Banking Law, 3rd edn (Oxford, Oxford University Press, 2017), 369–71; KFK Low and EGS Teo, ‘Bitcoins and Other Cryptocurrencies as Property?’ (2017) 9 Law, Innovation and Technology 235. 111 Bank of England, Financial Conduct Authority and HM Treasury, Cryptoassets Taskforce: Final Report (October 2018). 112 Following the publication of the Cryptoassets Taskforce: Final Report (n 111), the UK Government launched a FinTech Delivery Panel, including a LawTech Delivery Panel, which includes the UK Jurisdiction Taskforce, consisting of members of the judiciary, the Law Commission for England and Wales, and technology and legal professionals, together with the Financial Conduct Authority as a consultant. 113 UK Jurisdiction Taskforce of the LawTech Delivery Panel, ‘Public Consultation – The Status of Cryptoassets, Distributed Ledger Technology and Smart Contracts Under English Private Law’ (May 2019). 114 G Vos, ‘Cryptoassets as Property: How Can English Law Boost the Confidence of Would-Be Parties to Smart Legal Contracts?’ (Joint Northern Chancery Bar Association and University of Liverpool Lecture, 3 May 2019), especially paras 23–36; see www.judiciary.uk/announcements/speech-by-sir-geoffrey-vos-chancellor-of-the-high-court-cryptoassetsas-property/. 115 UK Jurisdiction Taskforce (n 113) annex 1. 116 Vos, ‘Cryptoassets as Property’ (n 114) paras 47, 50–55. 117 The LawTech Delivery Panel, Legal statement on cryptoassets and smart contracts [-] UK Jurisdiction Taskforce (November 2019), 15–17, 66–84.

Are there Any General Principles of Commercial Law? 199 willingness of this jurisdiction and others to embrace new commercial practices. The message seems to be that the UK is open for business so far as new intangible sources of wealth are concerned.

IX. A Conclusion of Sorts There are probably no general principles of commercial law that are not legal principles, policies or techniques which are already present in law in general, and private law in particular. If one accepts the characterisation of commercial law posited in this chapter – namely, that it is a highly significant context in which concepts of private law (and public law and public policy) play out – it is inevitable that those principles coexist in both private and commercial law. The more important analysis becomes how prominent particular principles or policies should be in commercial law. It has been part of the rhetoric of commercial lawyers since at least the time of Lord Mansfield that certainty and predictability are of utmost importance, but it would be difficult to assert that those values were more significant in commercial law than in, say, land law or intellectual property. So, too, will those who are resistant to, or sceptical of, doctrines of good faith, in whatever guise, wish to minimise recourse to such measures in commercial fields of activity. Conversely, the advocacy of commercial law’s receptiveness to market practices marks out its facilitative philosophy. But, again, it is to be hoped that flexibility and a recognition of changing social practices are a feature of the application of laws in all its forms. I hope I have demonstrated that commercial law has as meaningful an existence as any other contextual field of law, and that it involves the sensitive application and, where appropriate, modification of private law doctrine to commercial and financial activities. That has always yielded, and continues to yield, greater willingness to recognise the existence of intangible interests, and their ready transferability, and the enforcement of autonomous payment undertakings which underpin and facilitate commercial activity.

200

12 Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 CHARLES MITCHELL*

I. Introduction This chapter discusses the circumstances in which eighteenth- and nineteenth-century English courts would admit evidence of mercantile usage as an aid to resolving disputes which turned on the construction of contracts and the implication of terms. The task of construing commercial contracts in the eighteenth century was often made difficult by the minimalist drafting techniques used by the parties. When they were written down at all, commercial contracts tended to be short and therefore silent about issues which were later disputed. Their brevity also rendered their meaning ambiguous or obscure. To deal with these problems, the courts often admitted evidence of market usage. They justified this on the basis that parties who chose to deal in a market with well-established usages, and who chose not to record all the details of their bargain in a written document, must have meant not to create a bespoke agreement but to create an agreement that reflected the typical expectations and practices of participants in that market. Eighteenth-century judges faced with a contract which said nothing about a disputed matter might alternatively imply terms into the contract which would align the contract with market practices and expectations. The justification offered for such gap-filling exercises was, once again, that this reflected the parties’ intention that their relationship should be governed by the usages of the market in which they had dealt. Once again, this intention was said to be evidenced by the parties’ decision to trade in a market with well-established usages and to write down little or nothing more than the bare bones of their agreement. Between 1750 and 1850, commercial drafting practices began to change. Written contracts became more detailed and standard form contracts started to be used, developments which would continue, and indeed become more pronounced, in the later 1800s. One reason for these changes was that trading activity grew in volume, sophistication and variety. Another was that court decisions to imply terms into particular types of contract prompted merchants who did not wish them to form part of their bargain to write in clauses to exclude them. As time passed, the increasing length and complexity of contracts also made it harder for parties to argue that

* I thank Helena Ratcliff for her accomplished research assistance.

202 Charles Mitchell a dditional clauses should be read into them, because the judges held that terms were more likely to have been omitted deliberately from a contract with many terms than from a contract with few. New rules also emerged which held that terms could not be implied into contracts which would subject the parties to market usages that were inconsistent with the terms of their agreement. Fine distinctions then came to be drawn between cases where implication was permissible and cases where it was not.

II. Commercial Drafting Practices and Mercantile Usage In two cases of 1750 and 1751, Lord Hardwicke LC made some observations about commercial drafting practices and the relevance of mercantile usage to the resolution of disputes, which provide an apt starting point for the discussion in this chapter. In Baker v Paine, he said that: On mercantile contracts relating to insurances, &c. Courts of Law examine and hear witnesses of what is the usage and understanding of merchants conversant therein: for they have a stile peculiar to themselves, which is short, yet is understood by them; and must be the rule of construction.1

In Blunt v Cumyns, he then said that: ‘In all mercantile contracts or adventures the Articles are commonly extremely short; and where a doubt arises about them, the usage and understanding of merchants is read thereto; and is constantly so at Guildhall.’2 Here are plain statements of the reasons for the courts’ approach to construction and the implication of terms in eighteenth-century commercial cases. When they bothered to write their contracts down,3 merchants tended to be very brief. They traded in small markets where participants typically knew one another, and where market customs were typically well known by the participants. Many of the details of their contracts were not recorded because the parties assumed that if something went wrong then the problem would be solved in accordance with these market usages. If a dispute found its way into the court, the task of the judge and jury was therefore to find out what these usages were and resolve the dispute accordingly. I shall first say something about the ways in which such information was accessed and then consider how commercial drafting practices developed in the period running from 1750 to 1850.

A. Mercantile Usage Mercantile usage came to the courts’ notice in different ways, some formal and some informal. Writing in the 1670s, Sir Matthew Hale summarised these as follows: The court use to ascertain themselves by Speech with Merchants or Civilians, or else it comes in question upon the general Issue; and then if it be a question touching the Customes of Merchants, Merchants are usually Jurors at the request of either party, and Merchants produced on either side are produced to

1 Baker v Paine (1750) 1 Ves Sen 456, 459; 27 ER 1140, 1141–42. 2 Blunt v Cumyns (1751) 2 Ves Sen 331, 331; 28 ER 213, 213. 3 Outside London, in the 18th and early 19th centuries, ‘corn was traded at market entirely through verbal agreements’: M Lobban, ‘Contract’ in WR Cornish et al, The Oxford History of the Laws of England, vol XII: 1820–1914: Private Law (Oxford, Oxford University Press, 2010) 476, citing Report from the Select Committee on the Sale of Corn, PP 1834 (517), vii, 1, 165.

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 203 ascertaine the Court and Jury touching the Custome of Merchants; and if it be a question touching the Marine Law, Merchants, Mariners and Civilians are heard.4

The practice of hearing cases with a special jury of merchants was one of long standing. It was well established by the time that Sir John Holt adopted it in the early eighteenth century, and it was often used thereafter, most famously, but far from uniquely, by Lord Mansfield. By this means, and through less formal conversations with merchants,5 the judges acquired an expertise in commercial matters;6 the increase in commercial litigation in the late 1700s and early 1800s also meant that judges came to be appointed for whom this had been a significant practice area during their time at the Bar.7 Lord Ellenborough and Lord Tenterden are prominent examples. The practice of permitting litigants to call witnesses with knowledge and experience of mercantile usage also had a long history.8 An example from our period is Camden v Cowley,9 where two insurance policies were issued on a ship, the first ‘from London to Jamaica generally’, the second ‘at and from Jamaica to London’. The ship was lost off the Jamaican coast, after she had touched for some days at one port on the island but before she had delivered all of her outwardbound cargo at other ports. The question was whether the outward-bound risk had ended, and the homeward-bound risk begun, at the time when the ship was lost. Lord Mansfield held that ‘insurance-brokers and others might be examined, as to the general opinion and understanding of the persons concerned in the trade’.10 In the event, the witnesses knew of no relevant precedent or usage, but a special jury found that the outward-bound risk had come to an end once the ship had been moored for 24 hours at her first port of call. The judges might also allow parties to lead evidence of written rules governing markets in which they had dealt, where these existed. In Sutton v Tatham,11 for example, the defendant mistakenly instructed the plaintiff broker to sell 250 shares on his behalf on the London Stock Exchange. The defendant had wanted to sell only 50 shares, but by the time his mistake had been discovered, the plaintiff had already contracted with another broker to sell him 250 shares and had transferred 209 shares to him. On being told of the mistake, the second broker refused to cancel the bargain and bought the remaining shares elsewhere at a loss, which he charged to the first broker, invoking the printed rules of the Stock Exchange, which provided that ‘if the

4 MJ Prichard and DEC Yale, Hale and Fleetwood on Admiralty Jurisdiction (London, Selden Society, 1992) 57, quoted in W Swain, The Law of Contract, 1670–1870 (Cambridge, Cambridge University Press, 2015) 47. Swain also observes at 47–48 that a body of literature on mercantile law and practice emerged in the 18th century, references to which were rare prior to 1750 but increased thereafter. Examples are Dawkes v De Lorane (1771) 3 Wils KB 207, 212; 95 ER 1015, 1018; Power v Whitmore (1815) 4 M & S 141, 149; 105 ER 787, 791; Busk v Royal Exchange Assurance Co (1818) 2 B & Ald 73, 80; 106 ER 294, 296. 5 Mutford v Walcot (1700) 1 Ld Raym 574, 575; 91 ER 1283, 1283: ‘a question was made, whether [a plaintiff could declare upon a bill that was negotiated after the day of payment, and Holt CJ] said, that he had all the eminent merchants in London with him at his chambers at Serjeant’s-Inn in the long vacation about two years ago, and they all held it to be very common, and usual, and a very good practice.’ See also Wilson v Smith (1764) 3 Burr 1550, 1556; 97 ER 975, 978; Salvador v Hopkins (1765) 3 Burr 1707, 1714; 97 ER 1057, 1061. 6 JS Oldham, English Common Law in the Age of Mansfield (Chapel Hill, NC, University of North Carolina Press, 2004) 20–27; JS Oldham, The Varied Life of the Self-Informing Jury (London, Selden Society, 2005) 23–31. 7 Swain (n 4) 128. 8 See eg Dehers v Harriot (1691) 1 Show KB 163, 164; 89 ER 513, 513 (‘twenty merchants attested their customs’); Tassell v Lewis (1695) 1 Ld Raym 743, 743; 91 ER 1397, 1397 (‘Merchants in evidence at a trial at Guildhall, before Holt Chief Justice, swore the custom of merchants to be such’). 9 Camden v Cowley (1762) 1 W Bl 417, 96 ER 237. See also Lilly v Ewer (1779) 1 Doug KB 72, 73; 99 ER 50, 51 (noting the testimony of ‘Mr Gorman, an eminent merchant’); Caine v Horsefall (1847) 2 Car & K 349, 350; 175 ER 144, 145 (testimony of ‘a witness … well acquainted with the dealings of the African trade’). 10 ibid 417; 237. 11 Sutton v Tatham (1839) 10 Ad & El 27, 113 ER 11.

204 Charles Mitchell selling broker is not prepared to fulfil his contract, the purchaser may buy in shares to make up the deficiency, and charge the selling broker with any loss by difference of price’. The defendant was held liable to reimburse the plaintiff. He objected to reception of the Stock Exchange rules in evidence, but this objection was overruled. The court also held that it did not matter whether the defendant was aware of the rules; it sufficed that he had impliedly authorised the plaintiff to be bound by the rules of any market which the plaintiff had to enter to carry out the defendant’s instructions.

B. Commercial Drafting Practices In Greaves v Ashlin,12 a contract for the sale of oats was entered by the plaintiff buyer and the agent of the defendant seller. The plaintiff failed to collect the oats for a week and so the defendant gave notice that they should be collected immediately. However, the plaintiff still failed to collect them. The defendant therefore resold them (at a slightly higher price) to a third party. The plaintiff sued for non-delivery and the defendant said that he had been entitled to resell following the plaintiff ’s failure to collect the oats after notice had been given. However, the parties’ written contract said nothing about this, and Lord Ellenborough said that a witness could not be asked whether the usage of the corn market was to insert a delivery date into the written terms of a contract where delivery was not to be immediate, ‘as that was only an indirect method of giving parol evidence to vary the written contract’.13 The contract was a sold note, which read as follows: ‘Sold to John Greaves, 50 quarters of oats at 45s 6d per quarter, out of 175 quarters. J STEVENSON for J ASHLIN.’ Obviously, some contracts for the sale of goods were longer than this,14 but even so, the brevity of the parties’ written contract in Greaves is very striking. A similar contract today would contain many more terms, covering quality, title and risk, delivery and non-delivery, exclusion of warranties implied by law, time limits on buyer’s claims respecting shortage or damage incurred in transit, limitation of seller’s liability, payment date, termination, force majeure and applicable law, all rounded off with an entire agreement clause. By the end of the eighteenth century, some other types of contract had become much more sophisticated documents than the simple contract in Greaves. An obvious example is the bill of exchange, a negotiable instrument which had been very widely used since the early 1700s, operating as an essential adjunct to a commercial economy where cash was in short supply. By the late 1700s, marine insurance policies had also become increasingly elaborate as a variety of express warranties had been added to them, a reflection of the developing expertise, knowledge and business practices of underwriters.15 12 Greaves v Ashlin (1813) 3 Camp 426, 170 ER 1433. 13 ibid 427; 1433. 14 But see also Syers v Jonas (1848) 2 Ex 111, 154 ER 426, where the parties’ contract was another sold note: ‘I have this day sold you on account Mr WHL Syers, fifty-one bales tobacco ex “Lucretia/ La Guayra”, at 11d per lb in bond: customary allowances: payment two and two months.’ 15 Swain (n 4) 88, citing Woolmer v Muilman (1763) 3 Burr 1419, 97 ER 905; Bean v Stupart (1778) 1 Doug KB 11, 99 ER 9; De Hahn v Hartley (1786) 1 TR 343, 99 ER 1130. See also Vallejo v Wheeler (1774) 1 Cowp 143, 145; 98 ER 1012, 1013: ‘so very liberal are the underwriters in these days of their possessions in policies of what they do insure against, that some writers have thought it next to impossible that where a loss does happen a doubt should remain. Molloy, B. II. cap. 7, sect. 7, says, “Almost all those curious questions that former ages and the civilians according to the marine law, nay and the common lawyers too, have controverted, are now out of debate. Scarce any misfortune that can happen, or provision to be made, but the same is provided for in the policies that are now used: for they insure against heaven and earth, stress of weather, or whatsoever detriment shall happen or come to the thing insured.”’

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 205 In the early years of the nineteenth century, trade expanded further, more commercial contracts were entered, of a more diverse kind, and as a result more commercial litigation found its way into court.16 According to Lord Campbell, looking back on this period in his autobiography, another cause of the increase in commercial litigation was the war with France, as a result of which ‘more new questions arose between underwriters and merchants, between shipowners and shippers of goods, between foreign consigners and English factors, in a year than in a century of peace or regular warfare’.17 He added that ‘the perpetual fluctuation in the price of commodities caused innumerable controversies respecting the fulfillment of contracts’ and that ‘the growing depreciation of paper currency … occasioned a mass of bankruptcy litigation before unknown’. A similar account of this time was given by William Townsend in his biographical study of Lord Ellenborough: The Court of King’s Bench was … filled by men of singularly vigorous understandings, and profound professional knowledge, while the great variety and difficulty of the cases arising from extended trade, respecting questions of insurance, of charter parties, of principal and agent, of factors and brokers, of del credere commissions, of insurers and underwriters, the rights of aliens, neutrals, and belligerents, questions of set-off, of stoppage in transitu, of bankruptcy, of compositions with creditors, of agreements with the fourth and seventeenth section of the Statute of Frauds, enabled those judges to follow out the general principles of the law of contracts, which had been settled by Lords Mansfield and Kenyon – to apply them to the more complicated facts, and novel combination of circumstances, which then arose.18

The effect of decisions which construed contracts in particular ways, or implied p articular terms into them, was to create pressure on parties who did not wish their contracts to be open to such construction or contain such terms. To escape from this, they wrote terms to the contrary into their contracts, a development regretted by Lord Mansfield in Lilly v Ewer, where he observed that ‘wherever you render additional words necessary, and multiply them, you also multiply doubts and criticisms’; also, that it ‘may be hard, because words have been added in some instances, to force a construction in … [other cases], from the omission of them’.19 The drafting of contracts for the sale of goods is a well-known example of these changes. The law governing such contracts was rooted in the caveat emptor principle,20 although an action on the case lay for deceit for breach of warranty where the seller had given ‘either an express warranty or an affirmation known to be false at the time of the sale’.21 Starting in the early 1800s, however, the courts began to imply warranties of title, quality and/or quantity into contracts if they were satisfied that these reflected mercantile usages,22 and these warranties were enforceable

16 Reflected in Lord Ellenborough’s caseload at Guildhall and Westminster Hall between 1802 and 1818. Contract cases accounted for 46% of combined Guildhall and Westminster Hall business, negotiable instrument cases for 26% of Guildhall business and 18% of Westminster Hall business: J Oldham, ‘Law-Making at Nisi Prius in the Early 1800s’ (2004) 25 Journal of Legal History 221, 229–31. 17 MS Hardcastle (ed), Life of John, Lord Campbell, Lord High Chancellor of Great Britain (London, John Murray, 1881) vol 1, 214. 18 WC Townsend, The Lives of Twelve Eminent Judges of the Last and of the Present Century (London, Longman, Brown, Green & Longmans, 1846) vol 1, 384. These quoted passages all come from Oldham (n 16). 19 Lilly (n 9) 74; 52. See also Anderson v Pitcher (1800) 2 Bos & Pul 164, 168; 126 ER 1216, 1218 (Lord Eldon): ‘Perhaps it is to be lamented, that in policies of insurance, parties should not be left to express their own meaning by the terms of the instrument.’ 20 For which, see eg Deering v Farrington (1673) 3 Keble 303, 84 ER 734. 21 AWB Simpson, A History of the Common Law of Contract: The Rise of the Action of Assumpsit (Oxford, Oxford University Press, 1975) 536. See eg Stuart v Wilkins (1778) 1 Doug KB 18, 99 ER 15. 22 Morley v Attenborough (1849) 3 Ex 500, 512–13; 154 ER 943, 948 (Parke B).

206 Charles Mitchell whether or not the seller had knowledge of a relevant defect: his liability was strict. The courts would not always imply a warranty of title, but they would do so where the seller gave the buyer to understand that he was the owner.23 And although they would not imply a general quality obligation,24 they would imply warranties that goods were of merchantable quality in various special cases:25 where the goods were manufactured;26 where the buyer could not inspect them for himself;27 and where the description of the goods contained an implied undertaking that they were merchantable.28 Again, where goods were bought for a disclosed purpose, the courts would imply a warranty that they were fit for this purpose;29 and where goods were sold by sample, the courts would imply a warranty that the quality of the bulk of goods sold corresponded to the quality of the sample.30 In the course of time, these findings made their way into the Sale of Goods Act 1893, where implied terms were enumerated concerning title, quiet possession, sale by description and sale by sample; and, where the seller traded in goods, implied conditions of fitness for purpose and merchantable quality.31 These terms could be excluded, however, and in sale contracts they often were, particularly once standard form contracts had been agreed between the different interest groups operating in particular markets, in a process driven forward in the second half of the nineteenth century by trade associations.32 ‘As trades became more organized,’ standard form contracts of sale became ‘more elaborate and more specialized, according to route and product’;33 they also tended to include terms that disputes were to be settled by arbitration, which might itself be provided by the relevant trade association.34

III. Construction of Contracts Let us return to the eighteenth century and look at some cases where the courts admitted evidence of mercantile usage as an aid to construction of contracts. One was Chaurand v Angerstein,35

23 Early v Garrett (1829) 9 B & C 928, 109 ER 345; Eichholz v Bannister (1864) 17 CB (NS) 708, 144 ER 284. 24 Parkinson v Lee (1802) 2 East 314, 102 ER 389; Bluett v Osbourne (1816) 1 Stark 384, 171 ER 504. 25 SJ Stoljar, ‘Conditions, Warranties and Descriptions of Quality in Sale of Goods’ (1952) 15 MLR 425 and (1953) 16 MLR 174; P Mitchell, ‘The Development of Quality Obligations in Sale of Goods’ (2001) 117 LQR 645; Lobban (n 3) 475–85. 26 Jones v Bright (1829) 5 Bing 533, 130 ER 1167. 27 Gardiner v Gray (1815) 4 Camp 144, 171 ER 46; Jones v Just (1868) LR 3 QB 197 (QB). 28 Wieler v Schilizzi (1856) 17 CB 619, 139 ER 1219; Josling v Kingsford (1863) 13 CB (NS) 447, 143 ER 177; Randall v Newson (1877) 2 QBD 102 (CA). 29 Gray v Cox (1825) 4 B & C 108, 107 ER 999; Brown v Edgington (1841) 2 M & G 279, 133 ER 751; Bigge v Parkinson (1862) 7 H & N 955, 158 ER 758. 30 Syers v Jonas (n 14). 31 Sale of Goods Act 1893, ss 12–15. On the right to reject for short delivery, see s 30, discussed in S Thomas, ‘The Development of the Implied Terms on Quantity in the Law of Sale of Goods’ (2014) 35 Journal of Legal History 281. 32 As described by Lord Diplock in A Schroeder Music Publishing Co Ltd v Macaulay [1974] 1 WLR 1308 (HL) 1316. See also R Cranston, ‘The Rise and Rise of Standard Form Contracts: International Commodity Sales 1800–1970’ in R Cranston, J Ramberg and J Zeigler (eds), Commercial Law Challenges in the Twenty-First Century (Stockholm, Iustus Forlag, 2007). 33 Lobban (n 3) 477. 34 H Barty-King, Food for Man and Beast: The Story of the London Corn Trade Association, the London Cattle Food Trade Association and the Grain and Feed Trade Association 1878–1978 (London, Hutchinson, 1978); AWB Simpson, ‘The Origins of Futures Trading in the Liverpool Cotton Market’ in J Stapleton and P Cane (eds), Essays for Patrick Atiyah (Oxford, Oxford University Press, 1991). 35 Chaurand v Angerstein (1791) Peake 61, 170 ER 79.

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 207 which concerned an insurance policy issued by the defendant underwriter on a ship sailing from San Domingo to Nantes ‘in the month of October’. The ship set sail on 11 October, but the defendant was held to be entitled to repudiate liability after the court admitted the evidence of ‘several merchants and commercial men’, who said that the expression was ‘well understood amongst men used to commercial affairs to signify some time between the 25th of that month, and the 1st or 2d of the succeeding month’; and that the difference in dates would have affected the premium charged by 15 per cent. Lord Kenyon observed that The evidence of underwriters is good evidence on this subject. In questions on the arts and sciences the evidence of persons versed in those arts is daily admitted. Foreign laws are also matters of evidence, and yet all these are only the opinions of the witnesses.36

Speaking on the same theme soon afterwards, in Brough v Whitmore, the same judge said that the rule of law is to be given (not by merchants) but by the Court; though, when a question arises on the construction of the words of an instrument, which are in themselves ambiguous, it is a matter fairly within the province of those who alone act upon such instruments to declare the meaning of them.37

Another case illustrating the application of this principle was Studdy v Sanders.38 The plaintiff was an apple farmer in Devon, where the word ‘cyder’ was commonly used by apple farmers to describe the juice pressed from apples. The defendants were ‘cyder merchants’ in Bristol whose business was to sell the alcoholic drink manufactured from apple juice, which was also named ‘cyder’. The parties entered a contract under which the defendants would buy the plaintiff ’s ‘cyder’. They also agreed that juice pressed from the plaintiff ’s apples would be put into casks and used to manufacture the alcoholic drink, under the direction of a third party employed by the defendants, but on the plaintiff ’s farm. Before the alcoholic drink had been taken away by the defendants, the farm was raided by excise officers who seized the casks, asserting that the defendants had failed to pay excise duty for which they were liable. The defendants then refused to pay the plaintiff under the contract, arguing that they had never taken delivery of any ‘cyder’, meaning the alcoholic drink. The court found for the plaintiff. The defendant had relevantly taken delivery of the apple juice when this had come under the control of the defendant’s employee and because the word ‘cyder’ used in the parties’ written contract was ambiguous, ‘parol evidence was admissible to explain it; and from the evidence given … here it meant the apple juice’.39 The parol evidence rule provided that extrinsic oral evidence of the parties’ intentions could not generally be admitted where the words of a written instrument were unambiguous. However, there were exceptions to this principle, and these included cases where the meaning of a document was unclear because it was written in a foreign language or employed technical or peculiar terminology; and an analogy was drawn between contracts in a foreign language and mercantile contracts which used mysterious terms which were understood by the parties in a particular way. So, Thomas Starkie wrote in his treatise on evidence that ‘the Courts have long allowed mercantile instruments to be expounded, according to the usage and custom of merchants, who have a stile and language peculiar to themselves, of which, usage and custom are the legitimate interpreters’.40

36 ibid 62; 79. 37 Brough v Whitmore (1791) 4 TR 206, 208; 100 ER 976, 977. In the same case (at 210; 978), Buller J accepted the same principle, though he was less happy about it: ‘a policy of assurance has at all times been considered in Courts of Law as an absurd and incoherent instrument: but it is founded on usage, and must be governed and construed by usage’. 38 Studdy v Sanders (1826) 5 B & C 628, 108 ER 234. 39 ibid 639; 238. 40 T Starkie, A Practical Treatise on the Law of Evidence (London, J and WT Clarke, 1824) 1033.

208 Charles Mitchell He added that the ‘witnesses for this purpose may be considered to be the sworn interpreters of the mercantile language in which the contract is written’.41 These principles were applied, for example, in many cases that turned on the wording of bills of exchange.42 These documents were drafted in language which self-evidently possessed a technical meaning that had emerged as a result of many decades of evolving practice and judicial decision-making. Evidence of this technical meaning, as it was understood by the bankers and merchants who commonly used such bills, was therefore admitted as an aid to construing them when the parties did not agree what they meant. For instance, in Rowe v Young,43 an action was brought by the holder and indorsee of a bill against the acceptor, who lived in Torpoint, in Cornwall, although the terms of the bill provided that it was ‘payable at Sir John Perring and Co.’s bankers, London’. The question arose whether a bill worded in this way rendered the acceptor generally liable to pay on presentment, with an option given to the holder to present the bill to the bankers as one means of receiving payment, or whether the parties had intended that the holder should only be entitled to present the bill to the bankers and be paid by them. The House of Lords consulted the common law judges, a majority of whom said that on their understanding of mercantile usage, bills drafted in this way exposed the acceptor to a qualified liability only, ie they thought that the holder could only present the bill for payment to the bankers. The House of Lords consequently made a finding to this effect, although this was later reversed by legislation when commercial groups complained that in fact general mercantile usage was exactly the opposite.44 A trickier question was how to interpret ‘plain and ordinary terms and expressions’ when these were used by merchants whose understanding of such phrases differed from the general understanding. Starkie argued that in cases of this sort, the everyday sense and meaning of the words ‘ought not to be altered by evidence of a mercantile understanding and usage to the contrary’.45 Otherwise, he thought, ‘the written contract would become a dead letter; the question would not be, what is the actual contract, but what is the usage; and the very same terms would denote different contracts as often as mercantile fashions varied’.46 A case in which the court stuck to the general meaning of words was Syers v Bridge.47 The plaintiff shipowner secured a letter of marque, ie an Admiralty commission to capture enemy merchant ships as a private profit-making enterprise: following capture, they could be brought before the Admiralty courts for condemnation and sale. The plaintiff bought an insurance policy for a voyage from Liverpool to Antigua, ‘with liberty to cruize six weeks’, after which his ship would return with such prizes as she could capture. In the event, however, the ship was herself taken by an American privateer, and the defendant underwriter refused to pay on the ground that the captain had not chased possible prizes during a single continuous six-week period, but instead for several shorter periods, albeit that these had cumulatively added up to a period of less than six weeks. Lord Mansfield found for the defendant, reasoning that ‘cruise is a well known expression for a connected portion of time’ and that ‘six weeks is a continuation, a congregate denomination of time’.48 It followed that on the ordinary meaning of the words used in the policy, coverage had only been provided for a single continuous period spent in the pursuit of prizes and not for several shorter periods between each of which additional time had elapsed.

41 ibid

1036. also Gibbon v Young (1818) 8 Taunt 254, 129 ER 381 (charterparty; ‘freight measurement’). 43 Rowe v Young (1820) 2 Bli 391, 4 ER 372. 44 Bills of Exchange Act 1821, later re-enacted as the Bills of Exchange Act 1882, s 19(2)(c). 45 Starkie (n 40) 1036. 46 ibid 1037. 47 Syers v Bridge (1780) 2 Doug KB 526, 99 ER 335. 48 ibid 530–31; 337–38. 42 See

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 209 The everyday meaning of words was also favoured by the court in Cross v Eglin.49 There the plaintiffs agreed to buy ‘about 300 quarters, more or less’ of foreign rye from the defendants. The defendants shipped 350 quarters and refused to deliver any part of the shipment unless the plaintiffs would accept the whole. The plaintiffs treated this refusal as a repudiatory breach, which they accepted, and sued to recover their advance payment. At the York Assizes, evidence was offered that where the words ‘more or less’ were used in a contract for the sale of grain, it was contrary to market usage to require the buyer to accept so large an excess of grain above the amount specified in the contract. Parke J admitted this evidence and the jury found for the plaintiff. The case then went to the Court of King’s Bench, where the plaintiff won again, but the judges doubted that the evidence of market usage was admissible. This did not affect the outcome, because if the evidence was admissible, then the jury’s verdict was correct, and if it was not, then the jury’s verdict was still correct, either because the words ‘about’ and ‘more or less’ could not include 50 quarters more than 300 quarters on their ordinary meaning (per Lord Tenterden CJ and Littledale J) or because the ambiguity of the words made it impossible for the defendants to prove otherwise (per Parke and Patteson JJ). On the admissibility of the evidence, however, Littledale J had this to say: ‘“about”, and “more or less”, seem to be words of general import, and I should have much difficulty in saying that evidence ought to be received to ascertain their meaning’.50 These authorities were consistent with Starkie’s description of the law: where terms were used that had a well-known everyday meaning, evidence was not admissible to establish that these terms were understood by traders in a particular market to mean something different. This is also what Lord Lyndhurst CB meant when he said in Blackett v Royal Exchange Assurance Co that ‘Usage may be admissible to explain what is doubtful, it is never admissible to contradict what is plain’.51 In many other cases of this kind, however, the courts did admit such evidence and did then hold that contracts should be construed in accordance with market usage. A well-known example, and one that was much cited in later cases, was Smith v Wilson.52 The plaintiff leased agricultural land from the defendant which included a rabbit warren. The terms of the lease provided that the defendant would buy ‘ten thousand’ rabbits from the plaintiff when the lease expired. On expiry of the lease, however, the plaintiff supplied the defendant with 19,200 rabbits, and the plaintiff refused to pay for all of these. The court admitted evidence that in the locality where the parties had made their bargain, the word ‘thousand’ meant ‘twelve hundred’ when applied to the sale of rabbits, and it followed that the defendant was liable to pay for more rabbits than the general meaning of the word ‘thousand’ might have appeared to suggest. According to Taunton J: ‘Words denoting weight, or measure, or number, must undoubtedly be understood in their ordinary sense, unless some specific meaning be … given by custom.’53 Many other cases were decided in the same way: the parties’ contract included words that had an everyday meaning but these words were construed by the court to mean something else because this was the meaning attached to them by traders in the relevant market.54 It has to be 49 Cross v Eglin (1831) 2 B & Ad 106, 109 ER 1083. 50 ibid 110; 1084. 51 Blackett v Royal Exchange Assurance Co (1832) 2 Cr & J 244, 249–50; 149 ER 106, 108. See also Caine (n 9) 350; 145. 52 Smith v Wilson (1832) 3 B & Ad 728, 110 ER 266. 53 ibid 734; 268. 54 See eg Lethulier’s Case (1692) 2 Salk 443, 91 ER 384 (marine policy; departure on voyage ‘with convoy’); Burnet v Kensington (1795) Peake Add Cas 71, 170 ER 198 (marine policy; ‘stranding’); Uhde v Walters (1811) 3 Camp 16, 170 ER 1291 (marine policy; ‘port in the Baltic’); Taylor v Briggs (1827) 2 Car & P 525, 172 ER 238 (charterparty; ‘cotton in bales’); Hutchinson v Bowker (1839) 5 M & W 535, 151 ER 227 (contract of sale; ‘good barley’ and ‘fine barley’).

210 Charles Mitchell said, therefore, that the cases on this general issue were rather muddled. No doubt the results of the cases tended to turn on their facts. But the most one can say about the principles which they disclosed is that the courts would allow a departure from the parol evidence rule only if the terms of a written contract were ‘ambiguous’, that sometimes it was clear enough that an ‘ambiguity’ was present, but that sometimes a mismatch between the everyday meaning of terms and the special meaning attached to them by traders might itself be enough to lead the courts to discover an ‘ambiguity’, and it was unpredictable when the courts would be willing to make a finding to this effect.

IV. Implication of Terms In Hutton v Warren, a case which itself concerned an agricultural lease rather than a contract between commercial parties, Parke B said that: It has long been settled, that, in commercial transactions, extrinsic evidence of custom and usage is admissible to annex incidents to written contracts, in matters with respect to which they are silent … [This] has been done upon the principle of presumption that, in such transactions, the parties did not mean to express in writing the whole of the contract by which they intended to be bound, but a contract with reference to those known usages.55

Similarly, in Browne v Byrne, Coleridge J said that in all contracts, as to the subject matter of which known usages prevail, parties are found to proceed with the tacit assumption of these usages; they commonly reduce into writing the special particulars of their agreement, but omit to specify these known usages, which are included however, as of course, by mutual understanding: evidence therefore of such incidents is receivable. The contract in truth is partly express and in writing, partly implied or understood and unwritten.56

One case of this sort was Jones v Bowden.57 The defendant brokers bought some bags of seadamaged pimento, repackaged them and sold them to the plaintiffs at an auction by candle. Prior to the auction, a sale catalogue had been circulated, in which the goods had been simply described as ‘187 bags of pimento, bonded’. The plaintiffs sued for deceit, alleging breach of an implied warranty that the goods were free from damage. Evidence was admitted that although pimento was sometimes sold by auction with an express warranty of soundness, when damaged pimento was offered for sale by auction, it was usual in the trade to state that it was damaged, and that if nothing was added with respect to its quality, it was usual for buyers to assume that the pimento was sound. The jury found for the plaintiffs, and Lord Mansfield later held that it would be too much to deprive the Plaintiffs of the benefit of this verdict. Since it is usual to mention the fact if pimento is sea-damaged, when this is not mentioned as such, how would any one understand the catalogue, having simply the word pimento, but not particularized as being sea-damaged?58

55 Hutton v Warren (1836) 1 M & W 466, 475–76; 150 ER 517, 521. See also Starkie (n 40) 1038. 56 Brown v Byrne (1854) 3 El & Bl 703, 715; 118 ER 1304, 1309. See also Gibson v Small (1853) 4 HLC 353, 375; 10 ER 499, 508 (Talfourd J). 57 Jones v Bowden (1813) 4 Taunt 847, 128 ER 565. 58 ibid 852; 567.

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 211 Another case in point was Allan v Sundius.59 This concerned an agreement between two ship brokers: Lamont, who was based in Liverpool, and Sundius, who acted for the French Government. They agreed that Lamont would introduce Sundius to shipowners from whom the French Government could charter steamers for use as transports to carry troops to the Crimean War. They also agreed that Lamont would be paid a share of Sundius’s commission on a charter entered between the French Government and a shipowner. However, their contract was silent as to whether Lamont should also receive a commission in the event that a charter was renewed. Persuaded by evidence of a custom among ship brokers that an ‘introducing broker’ should receive renewed commission on every renewal of a charter effected through him, Bramwell B (with whom Martin B agreed) found for Lamont, stating that a ‘custom may be annexed to documents with which it is not inconsistent’ and holding on the facts that the payment of additional commission on renewal would not be inconsistent with anything in the parties’ contract.60 Pollock CB agreed with the general principle stated by the other two judges, but thought that this applied only in the usual course of business, and (unlike the others) thought that the parties’ agreement in the case was outside the ordinary course of business, with the result that the rule should not apply.61 Two additional points emerge from these decisions, about which a little more should be said. First, the logic of the courts’ reasoning suggested that a term should be implied into a contract to import a market usage only if the parties were aware of the usage, or at least had good reason to think that it existed even if they were unfamiliar with its content. This followed from the justification advanced by the courts for their rule of evidence: that this gave effect to the parties’ own intentions regarding the content of their agreement. Secondly, terms should not be implied into a contract to import market usages that were repugnant to the parties’ intentions because they were inconsistent with the express terms of the contract. A rather absolute statement of the first of these ideas can be found in Gabay v Lloyd.62 This concerned an insurance policy on goods, including three horses, which were shipped by the plaintiffs from Liverpool to Jamaica. During the voyage, the ship ran into a storm and the excessive rolling of the vessel caused the slings by which the horses were suspended to break and the partitions by which they were separated to collapse. The horses were then so battered by the ship’s rolling and by their own kicking that they all died. The defendant underwriter declined to pay for the loss of the horses, leading evidence that the practice of underwriters at Lloyd’s Coffee-House when insuring livestock shipped by sea was either to include the risk of death by an express term or to say nothing about it, in which case liability for such loss was impliedly excluded. The court found for the plaintiffs because it was not found that the plaintiffs were in the habit of effecting policies at that place. There was nothing, therefore, to shew that the plaintiffs had any knowledge of such a usage. Now the principle upon which usage is received as evidence to explain the ambiguous language of a policy is, that the parties to it are supposed to contract with a knowledge of that usage. Here it does not appear that the plaintiffs had any knowledge of the usage, and they cannot be supposed to have contracted with reference to it.63

In other cases, however, the courts were willing to hold that constructive knowledge of market usages would suffice to support a finding that a contracting party meant to be governed by them.

59 Allan

v Sundius (1862) 1 H & C 123, 158 ER 827. 129; 830. 61 ibid 132–33; 831. 62 Gabay v Lloyd (1825) 3 B & C 793, 107 ER 927. 63 ibid 797; 929. 60 ibid

212 Charles Mitchell For example, Coltman J held in Bayley v Wilkins that a ‘person who goes into the Stock-Exchange to buy shares, must be supposed to have a knowledge of the usual course of business there, and of the law applicable to it’.64 Many dicta to this effect can also be found in cases where a principal employed an agent to trade in a market on his behalf. So, for example, Alderson B said in Bayliffe v Butterworth that A person who deals in a particular market must be taken to deal according to the custom of that market, and he who directs another to make a contract at a particular place, must be taken as intending that the contract may be made according to the usage of that place.65

There was, however, an exception to this rule, namely that a market usage of which a contracting party had only constructive knowledge would not be implied into the contract if the usage was unreasonable (although an unreasonable usage would be implied into a contract if the contracting parties had actual knowledge of it).66 Turning to the situation where an inconsistency existed between a contract’s written terms and the usage of the market in which the parties had dealt, we can begin with Parke B’s statement in Syers v Jonas that in mercantile transactions ‘evidence of established usage is admissible … to annex customary incidents [but only] when it is not expressly or impliedly excluded by the tenor of the written instrument’.67 In cases where a contract was clearly silent on a relevant issue, this was not a problem. However, more difficult cases could arise in which it was not immediately obvious whether or not the wording of a written contract was consistent with the mercantile usage upon which one of the parties sought to rely. For instance, in Brown v Byrne,68 a firm of cotton traders based in New Orleans shipped 110 bales of cotton on board the plaintiff ’s vessel which was bound for Liverpool. The master signed a bill of lading which was endorsed and forwarded to the defendants who were a firm of Liverpool merchants. On the ship’s arrival at Liverpool, the defendants claimed and received the cotton as the holders and indorsees of the bill of lading. However, they refused to pay the full amount owed under the written terms of the bill, invoking a custom of merchants and shipowners in Liverpool, that three months’ discount was deductible from freights payable under bills of lading, on goods coming from ports in the southern states of America, including New Orleans. In the ensuing litigation, the Court of Queen’s Bench found for the defendants. Lord Campbell CJ accepted that evidence of mercantile usage ‘must not be of a particular which is repugnant to, or inconsistent with, the written contract’, but concluded on the facts that evidence of the usage could not be excluded on this ground: The written contract expressly settles the rate of payment: the custom does not set this aside; indeed it adopts it, as that upon which it is to act, by establishing a claim for allowance of discount upon freight to be paid after that rate. The consignee undertakes to pay freight on delivery after that rate; the shipowner undertakes to allow three months’ discount on freight paid after that rate; the latter contract is dependent on the former, but is not repugnant to it. If the bill of lading had expressed, or if, from the language of it, the intention of the parties could have been collected, that the freight at the specified

64 Bayley v Wilkins (1849) 7 CB 886, 902; 137 ER 351, 358. 65 Bayliffe v Butterworth (1847) 1 Ex 425, 429; 154 ER 181, 183. See also Sutton (n 11); Pollock v Stables (1848) 12 QB 765, 116 ER 1057; Cropper v Cook (1868) LR 3 CP 194, 200. 66 Robinson v Mollett (1875) LR 7 HL 802. 67 Syers v Jonas (n 14) 116; 429. See also Starkie (n 40) 1039–40; SM Phillipps, A Treatise on the Law of Evidence, 9th edn (London, Saunders & Benning, 1843) vol 2, 343–44. 68 Brown (n 56).

Mercantile Usage, Construction of Contracts and the Implication of Terms, 1750–1850 213 rate should be paid, free from all deductions, customary or otherwise, then it would have been repugnant to it to set up the custom, and the case would have been brought within the restriction mentioned above.69

The same point arose in Humfrey v Dale,70 another case decided by Lord Campbell CJ a few years later. There he gave the following ‘familiar instance by way of illustration’: on the face of a bill of exchange at three months after date the acceptor would be taken to bind himself to the payment precisely at the end of the three months; but, by the custom, he is only bound to do so at the end of the days of grace, which vary, according to the country in which the bill is made payable, from three up to fifteen. The truth is that the principle on which the evidence is admissible is that the parties have not set down on paper the whole of their contract in all its terms, but those only which were necessary to be determined in the particular case by specific agreement, and which of course might vary infinitely, leaving to implication and tacit understanding all those general and unvarying incidents which a uniform usage would annex, and according to which they must in reason be understood to contract unless they expressly exclude them. To fall within the exception, therefore, of repugnancy, the incident must be such as if expressed in the written contract would make it insensible or inconsistent.71

In Humfrey, the defendant brokers signed a sold note relating to the purchase of linseed oil. It was stated in the note that they were acting for principals, although the principals’ names were not declared. As disclosed agents, the defendants would not themselves ordinarily have been regarded as being personally liable on the contract in these circumstances. However, there was evidence of a market usage that a stated agent should be personally liable in addition to the principal, if the principal’s name were not declared within a reasonable period. These being the facts of the case, the defendants were held liable, the court holding that such a term could be implied into the contract without creating an inconsistency with the express terms of the written agreement. These cases can be contrasted with Phillipps v Briard,72 which concerned a charterparty between the defendant shipowner and the plaintiff charterers that the ship should load cargo in London and proceed to Hong Kong, where the cargo should be unloaded and the ship should be consigned to the plaintiffs’ agents ‘free of commission on this charter’. The point of this clause was to exclude the operation of a custom of London merchants by which the consignees were entitled to procure a charter or cargo for a return voyage (and thereby earn a commission), and which required the shipowner to reimburse them for the loss of such a commission if the shipowner procured a charter or cargo for the return voyage himself. The defendant procured a cargo for the return voyage and refused to pay a commission to the consignees, and the plaintiffs had to pay the consignees for their loss. The plaintiffs’ action to recover the amount of this payment was rejected, Pollock CB stating that I should have thought the custom unreasonable; but, at all events, there is no authority for saying that a written instrument can be varied by parol evidence of such a custom. This is not explaining a contract by evidence of something incidental to it, but introducing another and a different contract.73

69 ibid

716–17; 1309. v Dale (1857) 7 El & Bl 266, 119 ER 1246. 71 ibid 274–75; 1249. 72 Phillipps v Briard (1856) 1 Hurl & N 21, 156 ER 1101. 73 ibid 27; 1104. 70 Humphrey

214 Charles Mitchell

V. Concluding Remarks The world of cider and rabbits and horses was smaller than our world, and fewer people lived in it. But it would be a serious mistake to think that they lacked sophistication or were unable to identify and solve complex economic and legal problems in creative and imaginative ways. By 1700, Britain was already an advanced commercial economy; in the following 150 years, overseas trade significantly expanded and the country industrialised. Between 1850 and 1875, it was the world’s leading exporter of industrial products and manufactured goods (before it was overtaken in the late 1800s by the USA and Germany). By 1850, the activities of British industrialists, manufacturers, merchants, bankers, insurers and shipowners were all supported by a variety of legal instruments, some of which had already become quite lengthy and sophisticated documents, and most of which would become still more so. It has been suggested by some scholars of ‘new institutional economics’ that the development of legal forms and the evolution of legal rules played a significant causative role in British economic growth during the eighteenth and nineteenth centuries. Historians of the period have expressed scepticism about such claims, however, objecting that they entail too linear and too simplistic a causal explanation of the relation between legal and economic change, and ignore the likelihood that that causality might run in the other direction, ie that economic growth might cause new legal forms and new legal rules to emerge.74 That the two were connected in some way appears indisputable, however, and this chapter has considered one part of this story by discussing the different roles played by traders, legal draftsmen and the courts in the evolution of contractual forms and contract law. Traders developed practices which suited them and with which they assumed their trading partners would comply. These did not immediately find their way into written commercial contracts, however, and so the courts had to develop rules of construction and rules governing the implication of terms by the application of which trading practices were effectively read into commercial contracts. Traders who wished to deal on different terms – to enter a bespoke rather than a more typical form of agreement – were then forced to modify their written contracts by the use of exclusion clauses and other drafting devices. This caused contracts to grow in length and sophistication, and this obliged the courts in turn to develop the rules of contract law to meet the new circumstances created by these evolving drafting practices. So it was that trading practices, drafting practices and the law governing the construction of contracts and the implication of terms developed hand in hand.

74 For general discussion of this controversy, see W Cornish et al, Law and Society in England 1750–1950, 2nd edn (Oxford, Hart Publishing, 2019) 6–10.

13 ‘The Obscure, the Implied and the Illegal’: English and French Approaches to the Interpretation of Written Contracts, Implication of Terms and Contracts Affected by Illegality RICHARD AIKENS*

I. Introduction The purpose of comparative law studies is to analyse how different systems deal with common problems. All systems of law have to deal with agreements between persons, whether natural or legal. What English law calls a ‘contract’ and the French Code Civil calls ‘un contrat’ will characteristically create rights and obligations on two parties.1 Contracts, wherever made, give rise to the same type of problems. Characteristically, one party alleges that the other has not carried out a promise according to the agreement. The solution to the dispute frequently turns on how the contract is to be interpreted. All systems of laws therefore have to develop rules for determining what terms a contract contains and their meaning. It is only once issues concerning the meaning of a contract or the precise extent of its terms have been determined that disputes, such as whether there has been a breach of the agreement, can be tackled. Another question that can arise is: have the parties apparently agreed to do something that is contrary to some law? Whether there is ‘illegality’ may depend on what, exactly, the parties agreed to do, and the relationship between that obligation and other laws of the country where the agreement was made or is to be

* I am very grateful for assistance in research from Alfred Lewis BA (Oxon), LLM and Manel Chibane BA (Paris-Nanterre), LLM (Paris II), avocat du barreau de Paris. All descriptions, conclusions and any errors are, of course, mine alone. 1 French law distinguishes between une convention and un contrat. Convention is a generic term for an agreement. Contrat signifies an agreement that creates obligations, either on one or both parties. Art 1106 of the new French Code Civil, in force since 2016, defines one of several types of contrat as being ‘synallagmatic where the parties undertake reciprocal obligations in favour of each other’; in the French: ‘Le contrat est synallagmatique lorsque les contractants s’obligent réciproquement les uns envers les autres.’ The term ‘synallagmatic contract’, in the same sense, was famously adopted by Diplock LJ in Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd (The Hong Kong Fir) [1962] 2 QB 26 (CA) 65.

216 Richard Aikens performed. The consequences of ‘illegality’ may also depend on the proper interpretation of what the parties agreed to do. English and French law have similar concepts of interpretation of a contract to find out its true meaning, and of implied terms to fill in contractual gaps. Both systems have a public policy concept that if the object or performance of a contract is at odds with the interests of the state as set out in laws (or perhaps accepted standards of morality), then there are consequences in terms of either the validity of the contract or the rights of the parties to it. However, the ways these concepts are worked out are different. Under the law of England and Wales,2 the common law rather than statute determines issues of interpretation of a contract; whether a term of contract can be implied as a matter of fact;3 and whether a contract is ‘illegal’ and, if so, what the consequences will be. In his dissenting judgment in Patel v Mirza,4 Lord Sumption stated that the common law ‘has a greater inherent flexibility and capacity to develop independently of legislation than codified systems do’. But, as he recognised, this flexibility comes at a price ‘in terms of certainty and accessibility to those who are not professional lawyers’. Under the codified law of France, the Code Civil provides the framework of the law affecting these issues. There have been recent developments in the laws of both countries in the three areas of interpretation, implied terms and ‘illegality’. It thus seems appropriate to compare the approach of the two systems to these common issues. In England, the House of Lords and the Supreme Court have pronounced at least eight times in the last 23 years on the interpretation of contracts.5 Since 2009, the Privy Council and Supreme Court have twice considered the process a court must go through before it can ‘imply’ a term in a contract.6 In 2016, after some tergiversations in previous Supreme Court decisions,7 the majority of a nine-member Supreme Court in Patel8 decided to revolutionise, at least in part, the common law on the effect of illegality on a contract. The French Civil Code (which I will refer to as the Code, identifying other particular codes if necessary) was enacted in 1804. Until 2016, this venerable and venerated Code had remained largely untouched insofar as it concerned contract law. Then, on 1 October 2016, a ministerial decree brought into force a reformed Code.9 The parts of the Code dealing with obligations were 2 For brevity, I will refer only to the law of England or English law from now on. 3 As opposed to implication as a matter of law, as in the case of terms implied by custom and usage, or by legislation, eg Sale of Goods Act 1979, ss 12–15. The distinction was recently recognised again in Geys v Société Générale, London Branch [2012] UKSC 63, [2013] 1 AC 523 [55] (Lady Hale); Marks & Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72, [2016] AC 742 [15] (Lord Neuberger). I am concerned only with terms implied ‘by fact’. 4 Patel v Mirza [2016] UKSC 42, [2017] AC 467 [226]. 5 Charter Reinsurance Co v Fagan [1997] AC 313 (HL); Mannai Investment Co Ltd v Eagle Star Life Assurance Co Ltd [1997] AC 749 (HL); Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (HL) (hereafter, ICS); Sirius International Insurance Co (Publ) v FAI General Insurance Co Ltd [2004] UKHL 54, [2004] 1 WLR 3251; Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] AC 1101; Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900; Arnold v Britton [2015] UKSC 36, [2015] AC 1619; Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173. 6 A-G of Belize v Belize Telecom Ltd [2009] UKPC 10; [2009] 1 WLR 1988; Marks & Spencer (n 3). Although the Belize case itself concerned the construction of the articles of association of a company, the advice of the Privy Council given by Lord Hoffmann stated that it applied to the implication of terms in contracts also. 7 Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] AC 430; Bilta (UK) Ltd (in liq) v Nazir (No 2) [2015] UKSC 23, [2016] AC 1. In the latter case, Lord Neuberger stated at [15] that the sharp division of approaches between the members of the Supreme Court which had been exhibited in the various judgments in those two cases had to be addressed as soon as appropriately possible. It did not take long to find a suitable case for treatment. 8 Patel (n 4). As argued in the text at n 109, this decision did little to clarify the law on when a contract is to be regarded as ‘illegal’; it dealt only with what the consequences might be if a court holds that it is illegal. 9 The Ordinance was promulgated in February 2016: Ordonnance no 2016-131 du 10 février 2016 portant réforme du droit des contrats, du régime général et de la preuve des obligations, JUSC1522466R. The decree was ratified by the French Parliament in 2018: loi de ratification no 2018-287 du 20 avril 2018.

‘The Obscure, the Implied and the Illegal’ 217 ‘fundamentally reformed’.10 The sections of the Code on contract law, under Book III, Title III, were rewritten. The aim of the reforms was to modernise, simplify and clarify French contract law and make it more accessible.11 The original Code had been much developed by judicial decisions in the course of 200 years, and this encrustation on the Code had created much obscurity in the law and even, in some cases, minimised the connection between the terms of the Code and the law as actually applied by the courts.12 Thus, one main reason for the reforms was to put into the revised Code the established case law: une codification à droit constant. But there was a more important reason for the reforms, which is referred to specifically in the Report to the President of the Republic on the proposed changes.13 This was the fact that the World Bank report on ‘doing business’14 had unfavourably compared French contract law to that of the common law, and of England and Wales and the USA in particular. The World Bank reports portrayed French law as being complex, unpredictable and unattractive. One expressed aim of the reformed text was to make it more attractive to foreign investors and commercial enterprises in order to encourage them to make their contracts subject to French law.15 It would take more than a chapter to elaborate fully just the English law, let alone the French law, on issues of interpretation, implied terms and illegality in connection with contracts. Instead, I have two aims. The first is to distil the approaches of the two systems of law to the common problems that they face. In doing so, I will attempt to gauge whether either system is coherent and, to use a word currently in vogue in English judgments, ‘principled’. Secondly, I will attempt to discern whether the two laws would produce a different result, giving an example in each category.

II. Interpretation of Written Contracts A. The English Approach Parties can agree the terms of a contract in one of three ways. They can do so orally and not set them down in writing at all; they can agree the terms, perhaps orally to start with or by using a standard form of contract wording that is modified in discussion, and then put all the terms of the agreement down in writing; or they can agree the terms of a contract orally and put only some of the terms in writing.16 In all cases, obviously, the only means that the parties have of recording what they have agreed is by using language in the form of words and figures. Although issues of

10 J Smits and C Calomme, ‘The Reform of the French Law of Obligations: Les Jeux Sont Faits’ (2016) 23 Maastricht Journal of European and Comparative Law 1040. The reforms were the result of many reports and discussions, which need not be discussed in detail here. 11 Introduction to Rapport au Président de la République relatif à l’ordonnance no 2016-131 du 10 février 2016 portant réforme du droit des contrats, du régime général et de la preuve des obligations: NOR: JUSC 1522466P. 12 S Rowan, ‘The New French Law of Contract’ (2017) 66 ICLQ 805. 13 See n 11 above. 14 World Bank, ‘Doing Business in 2004: Understanding Regulations’, www.doing business.org/-/media/WBG/ DoingBusiness/Documents/Annual-Reports/English/DB04-Full Report.p.d.f. 15 Whether this objective has been achieved remains to be seen; Smits and Calomme (n 10) express doubts. 16 For an example of the last type, which gave rise to considerable problems about what had been agreed, how to interpret the contract and whether to imply terms, see Crema v Cenkos Securities plc [2010] EWCA Civ 1444, [2011] 1 WLR 2066.

218 Richard Aikens interpretation, possible implied terms and illegality can arise with all three types of contract,17 I will concentrate on wholly written contracts, and in particular commercial contracts. In a commercial context, problems often arise because the document has been slovenly drafted,18 or the parties may have used a ‘shorthand’ which may (or may not be) understood in the commercial context in which it is used or may have used deliberately vague wording in a clause because they might not have been able to agree on anything more precise.19 It is often said that the approach of English law to the interpretation of contract terms was ‘revolutionised’ in the latter half of the twentieth century. This so-called revolution was characterised in striking terms in 1998 by Lord Hoffmann in giving the leading speech in Investors Compensation Scheme Ltd v West Bromwich Building Society.20 He said that there had been a ‘fundamental change in this branch of the law’ such that ‘[a]lmost all the old intellectual baggage of “legal” interpretation’ had, by the date of that case, ‘been discarded’.21 The result was, he said (and subject to one important exception),22 ‘to assimilate the way in which documents are interpreted by judges to the common sense principles by which any serious utterance would be interpreted in ordinary life’. In Sirius International Insurance Co (Publ) v FAI General Insurance Ltd,23 Lord Steyn referred to the major element of this ‘baggage’ as being ‘literalism’. He implied that in the past judges had even been in the habit of perversely applying a meaning to words or a contract that was contrary to what the parties intended them to mean.24 In fact, it is doubtful whether there had ever been a ‘literalist’ school of contract interpretation within the English judiciary, even in the nineteenth century.25 Lord Wilberforce pointed out in Prenn v Simmonds26 that

17 As noted above, Crema (n 16) concerned the interpretation of and the implication of terms in a part written, part oral contract. The contract between Patel and Mirza appears to have been oral, although this is not explicitly stated in the facts of the report, nor in Lord Toulson’s leading judgment. 18 The complaint of Lord Lloyd in ICS (n 5) 899. 19 As suggested by Lord Wilberforce in Prenn v Simmonds [1971] 1 WLR 1381 (HL) 1385. 20 ICS (n 5). 21 ibid 912. 22 The exception is not identified. It must be the fact that whereas in everyday life people can look at what was said or written in the course of negotiations when arguing about what was agreed and also what happened afterwards, a court will not do this when interpreting a contract, unless there is an action for ‘rectification’. The reason for this, Lord Hoffmann explained, was ‘practical policy’: ibid 913. In fact, as Lord Wilberforce made clear in Prenn (n 19) 1384, evidence of negotiations is positively ‘unhelpful’ because the parties’ positions will be changing and only the final agreement will record the consensus. 23 Sirius International Insurance Co (Publ) v FAI General Insurance Ltd [2004] UKHL 54, [2004] 1 WLR 3251 [19]. 24 Lord Steyn suggests that this doctrine of ‘literalism’ espoused by 19th-century judges was partly the result of statements by W Paley in The Principles of Moral and Political Philosophy (London, R Faulder, 1785), which, Lord Steyn says, greatly influenced English contract lawyers in the 19th century. Lord Steyn refers to the example related in Book III, Chapter 5, part II of Paley’s work, headed ‘In what sense promises are to be interpreted’. Paley recounts that the tyrant Temures had promised the garrison of Sebastia that if they surrendered ‘no blood should be shed’. They surrendered and he buried them alive. Lord Steyn gives this as an example of ‘literalism’. But Paley did not say that was a correct interpretation of Temures’ promise. Paley explained that although Temures fulfilled his promise in one sense, this ‘was not the sense in which the garrison of Sebastia actually received it: which last sense, according to our rule, was the sense in which he was in conscience bound to have performed it’. Accordingly, contrary to Lord Steyn’s inference, Paley was not advocating ‘literalism’ at all, but an objective approach to the interpretation of a promise. He advocated that if a promise admitted of more sense than one, then the promise was to be performed ‘in the sense in which the promiser apprehended at the time that the promisee received it’. Paley’s example is referred to with approval in J Chitty, A Practical Treatise on the Law of Contracts not under Seal, 2nd edn (London, S Sweet, 1834) 63, a work which may have had more influence on English judges than Paley’s original. 25 Save perhaps in the case of the interpretation of wills, where there were special rules; but even there, ‘literalist’ interpretation was not uniform. 26 Prenn (n 19) 1384.

‘The Obscure, the Implied and the Illegal’ 219 in 1877, Lord Blackburn had stated the principles for construing ‘instruments in writing’ in a way that was recognisably ‘modern’. Lord Blackburn said that in all cases the object is to see what is the intention expressed in the words used. But, from the imperfection of language, it is impossible to know what that intention is without inquiring further, and seeing what the circumstances were with reference to which the words were used and what was the object, appearing from those circumstances which the persons using them had in view; for the meaning of words varies according to the circumstances with respect to which they are used.27

The idea that in interpreting a written contract the court has to ascertain the intention of the parties goes back further. Its roots can be traced back at least to the judgment of Baron Parke in the Exchequer Chamber in Ford v Beech28 in 1848. The objective approach to ascertaining the ‘intention of the parties’ or the ‘aim or object or commercial purpose of a contract’ was re-emphasised by Lord Wilberforce in 1976 in Reardon Smith Line Ltd v Hansen-Tangen (The Diana Prosperity).29 It meant finding the intention ‘which reasonable people would have had if placed in the position of the parties’. This element of ‘objectivity’, rather than the subjective beliefs of the parties, was reiterated by Lord Hoffmann in ICS.30 It is the keystone to the English approach to interpretation. Lord Sumption, speaking extrajudicially in 2017 in his widely noted Harris Lecture,31 argued that Lord Hoffmann’s speech in the ICS case marked a legal offensive by the highest court in the land (which he dubbed ‘the Hoffmann offensive’), whose target was to ‘free the construction of contracts from the shackles of language and replace them with some broader notion of intention’. In support of this argument, Lord Sumption particularly emphasised two of the five principles of construction that Lord Hoffmann had set out in the ICS case, viz numbers (4) and (5).32 In the first of these, Lord Hoffmann had contrasted the meaning of a document with the meaning of its words. He had said that ‘the meaning of words is a matter of dictionaries and grammars; the meaning of a document is what the parties using those words against the relevant background would reasonably have understood them to mean’.33 In some cases, according to Lord Hoffmann, it might even be concluded that the parties had ‘used the wrong words or syntax’. In the other principle, Lord Hoffmann put the same point in a different way, quoting and adopting the celebrated statement of Lord Diplock in Antaios Compania Naviera SA v Salen Rederierna AB that ‘if detailed semantic and syntactical analysis of words in a commercial contract is going to lead to a conclusion that flouts business common sense, it must be made to yield to business common sense’.34 Lord Sumption’s thesis in his Harris Lecture was that ‘the Hoffmann offensive’ encouraged judges to abandon the task of elucidating what the words of a contract meant in

27 River Wear Commissioners v Adamson (1877) 2 App Cas 743 (HL) 763. 28 Ford v Beech (1848) 11 QB 852, 866; 116 ER 693, 698 (Parke B): ‘In adjudicating upon the construction and effect in law of this agreement, the common and universal principle ought to be applied: namely, that it ought to receive that construction which its language will admit, and which will best effectuate the intention of the parties, to be collected from the whole of the agreement, and that greater regard is to be had to the clear intent of the parties than to any particular words which they may have used in the expression of their intent.’ 29 Reardon Smith Line Ltd v Hansen-Tangen (The Diana Prosperity) [1976] 1 WLR 989 (HL) esp 995–97 (Lord Wilberforce). It was in this case (at 997) that Lord Wilberforce coined the now celebrated phrase ‘factual matrix’. 30 In particular in his propositions (1), (3) and (4), where he refers to the meaning of a document as it would appear to the ‘reasonable man’, the judges’ anthropomorphism for objectivity in approach. 31 Lord Sumption, ‘A Question of Taste: The Supreme Court and the Interpretation of Contracts’ in D Clarry (ed), The UK Supreme Court Yearbook, Volume 8: 2016–2017 (Cambridge, Appellate Press, 2019). 32 ICS (n 5) 912–13. 33 ibid 913. 34 Antaios Compania Naviera SA v Salen Rederierna AB [1985] AC 191 (HL) 201.

220 Richard Aikens favour of trying to ascertain what would be a reasonable result of the interpretive process, even if that meant modifying or contradicting the words used. For Lord Sumption, ‘the Hoffmann offensive’ reached its farthest point of advance in the Supreme Court’s decision in Rainy Sky SA v Kookmin Bank.35 Kookmin Bank had given ‘advance payment bonds’ to buyers of ships that were to be built. The buyers paid instalments of the purchase price. The yard got into financial difficulty and the buyers (through an assignee of their rights) made demands under the bonds for repayment of instalments paid. The bank refused, saying the bonds’ terms (which were in two different paragraphs) did not cover the factual situation that had arisen. Lord Clarke of Stone-cum-Ebony gave the leading judgment with which the other Justices agreed.36 The issue was how a court should interpret a commercial contract which was capable of two meanings, neither of which ‘flouted common sense’. After a full discussion of the cases, Lord Clarke gave his conclusion: ‘where a term of a contract is open to more than one interpretation, it is generally appropriate to adopt the interpretation which is most consistent with business common sense’.37 On the facts of the case, he concluded that because the bank could give no credible reason for a narrow reading of the bonds, the buyers’ wider construction was to be preferred. It was ‘consistent with the commercial purpose of the bond in a way in which the bank’s construction was not’.38 Some regarded this conclusion as placing too much emphasis on the interpretive role of ‘business common sense’ at the expense of an examination of the plain meaning of the words in the context of the contract as a whole and its factual background. In his Harris Lecture, Lord Sumption went so far as to suggest that Lord Clarke’s approach would tempt a judge to uncover two or more possible meanings of words so as then to impose a meaning which seemed more ‘reasonable’ to the judge in accordance with his or her view of what accorded with ‘business common sense’. As Lord Sumption rightly points out, the difficulty with this approach is that the judge has not – indeed, cannot have – any evidence of what the parties actually intended at the time because extrinsic evidence of their views is inadmissible for the purposes of interpreting a document. The judge can only take an educated guess at what the parties ‘intended’ from a reading of the contract words and from the ‘factual matrix’, which rarely leads to one clear construction over another. As it turned out, the ‘Hoffmann offensive’, if that is what it was, got no further than the salient of Rainy Sky. As Lord Sumption might have put it, the forces of orthodoxy mounted a counterattack in Arnold v Britton39 and Krys v KBC Partners LP.40 The former decision of the Supreme Court is the more important of the two. It concerned the interpretation of a clause in 99-year 35 Rainy Sky (n 5). 36 Lord Phillips, Lord Mance, Lord Kerr and Lord Wilson. 37 Rainy Sky (n 5) [34]. 38 Meaning that the bank’s construction was one that would not have covered a liability to make repayments to the buyer of the ship on the bond in the event of the builder’s insolvency. Lord Clarke agreed with the trial judge, Simon J, and the dissenting judge in the Court of Appeal, Sir Simon Tuckey, that this was the situation for which an advance payment bond was most likely to be needed. 39 Arnold (n 5). Lord Carnwath dissented. He regarded the construction of the lease terms that the majority adopted as a ‘result which is commercial nonsense’: ibid [158]. 40 Krys v KBC Partners LP [2015] UKPC 46. Lord Sumption gave the advice of the majority and relied for his conclusion on ‘the clear language of the articles’, even though it was argued for the unsuccessful appellants that the apparent result of that construction was extraordinary: ibid [15]. The majority held the result was not. Lord Mance dissented, saying that the majority advice had, as in Re Sigma Finance Corp [2009] UKSC 2, [2010] 1 All ER 571 [12], attached ‘too much to what the courts perceived as the natural meaning of the words … and too little weight to the context … and to the scheme as a whole’. He concluded that the interpretation of the majority did not give effect to ‘any intention that the [parties to the partnership deed] are likely to have had or did have when agreeing these articles’. In other words, his conclusion was based on deducing what he thought was the intention of the parties from a detailed examination of the clauses overall.

‘The Obscure, the Implied and the Illegal’ 221 leases for holiday chalets, which related to the service charges payable by the lessees. The leases had been granted when inflation was high. They provided for annual 10 per cent compound increases in the charges. This was much higher than earlier leases for the same type of chalet, which provided for 10 per cent compound increases every three years. The lessees of the later leases challenged the landlord’s right to the fixed but increasing charges. If the ‘natural meaning’ of the service charge clause in the later leases was applied, it had the effect that, after a time and with a decrease in the rate of inflation, the landlord received far more than the cost of the services he provided. It was argued that this result contradicted ‘business common sense’. This was therefore an ideal case for determining the extent to which ‘business common sense’ should govern a court’s approach to the interpretation of a contentious clause. Lord Neuberger, giving the majority judgment, restated the English courts’ approach to the interpretation of written contracts. He said that the basic task of a court was to identify the intention of the parties by reference to ‘what a reasonable person having all the background knowledge which would have been available to the parties would have understood them to be using the language in the contract to mean’.41

However, he also emphasised seven factors. Their essence is that the use of ‘commercial common sense’ as a guide to the intention of the parties cannot displace the importance of the language used in the clause in issue, viewed in the context of the contract as a whole and of what the parties knew at the time of the conclusion of the contract. As Lord Neuberger pithily put it, ‘the mere fact that a court may be pretty confident that the subsequent effect or consequences of a p articular interpretation was not intended by the parties does not justify rejecting that interpretation’.42 In the latest case in this series, Wood v Capita Insurance Services Ltd,43 the judgment of the Supreme Court was given by Lord Hodge. The issue was the interpretation of an indemnity clause in a share purchase agreement. Lord Hodge refused to be drawn on whether there was a difference of approach between Rainy Sky and Arnold. He said that it was ‘not appropriate in this case to reformulate the guidance given in the Rainy Sky and Arnold cases’.44 He did not accept that the latter ‘involved a recalibration of the approach summarised’ in the former. Had there been any change in the approach of the English law to the interpretation of contracts over the last half century or so? Lord Hodge thought not. He said, perhaps boldly, that ‘the recent history of the common law of contractual interpretation is one of continuity rather than change’.45 He commended this ‘stability and continuity’ in contractual interpretation as being one of the attractions of English law as a legal system of choice in commercial matters. The evolutionary thesis in this area of the law is not shared by Lord Hoffmann, as he stated in an article responding to Lord Sumption’s Harris Lecture.46 Lord Hoffmann argues that, at least in the case of the construction of wills by nineteenth (and even twentieth) century judges, there were some ‘irrebuttable presumptions’ of construction which forbade the introduction of any extrinsic evidence to rebut a ‘strict and primary meaning’ of words used in the will.47 That may

41 ibid [15]. The quotation within this passage is from Lord Hoffmann’s speech in Chartbrook (n 5) [14]. 42 ibid [28]. 43 Wood (n 5). The court consisted of Lord Neuberger, Lord Mance, Lord Clarke and Lord Sumption: the Justices who had given the judgment of the court in Rainy Sky (n 5) and the majority judgment in Arnold (n 5). 44 Wood (n 5) [9]. 45 ibid [15]. 46 Lord Hoffmann, ‘Language and Lawyers’ (2018) 134 LQR 553. 47 ibid 566.

222 Richard Aikens have been so, but Lord Hoffmann is not able to refute the fact that, from the mid-nineteenth century onwards, when a court has considered the interpretation of a written contract, it has looked at what we now call ‘the factual matrix’ in order better to put the words of the contract in their context when interpreting the language. Lord Hoffmann also suggests48 that Lord S umption, in his Harris Lecture, is arguing for the return of the old rules of irrebuttable presumptions and for not taking account of the ‘factual matrix’ or trying to ascertain what the parties intended. That is going too far. All Lord Sumption is doing is arguing for the primacy of the language of the contract in ascertaining the intention of the parties when interpreting what the contract means. A very recent and, I respectfully suggest, accurate summary of the current law was given by Popplewell J in Lukoil Asia Pacific PTE Ltd v Ocean Tankers (PTE) Ltd (The Ocean Neptune).49 I have divided what he said into eight numbered sections: The court’s task is [1] to ascertain the objective meaning of the language which the parties have chosen in which to express their agreement. The court must [2] consider the language used and ascertain what a reasonable person, that is a person who has all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract, would have understood the parties to have meant. The court [3] must consider the contract as a whole and, depending on the nature, formality and quality of drafting of the contract, give more or less weight to elements of the wider context in reaching its view as to the objective meaning of the language used. [4] If there are two possible constructions, the court is entitled to prefer the construction which is consistent with business common sense and to reject the other. [5] Interpretation is a unitary exercise; [6] in striking a balance between the indications given by the language and the implications of the competing constructions, the court must consider the quality of drafting of the clause and it must also be alive to the possibility that one side may have agreed to something which with hindsight did not serve his interest; similarly, the court must not lose sight of the possibility that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms. [7] This unitary exercise involves an iterative process by which each suggested interpretation is checked against the provisions of the contract and its commercial consequences are investigated. [8] It does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each.

Popplewell J’s summary sets out the key elements of the English law approach to the interpretation of written contracts. The approach is principled because it is based on concentrating on the language of the contract in the first place. Objectivity rather than subjectivity in construing the words of the contract is fundamental to this approach. Either could be adopted, but English law has nearly always preferred an objective approach. It is logical (and is emphasised) that, when considering the intention of the parties, the court must place the contract in its context, and not read it in isolation. It is principled to take account of ‘business sense’ in commercial contracts, but not at the expense of the clear meaning of the language of the contract. If Popplewell J’s summary of the English law on interpretation were expressed as a code, it would work well. Is it different from the French law approach?

48 ibid 568. 49 Lukoil Asia Pacific PTE Ltd v Ocean Tankers (PTE) Ltd (The Ocean Neptune) [2018] EWHC 163 (Comm), [2018] 2 All ER (Comm) 108 [8]. Popplewell J refers to all the cases discussed above, starting with ICS (n 5) and finishing with Wood (n 5). The numbers in square brackets in the quoted passage have been inserted by the author.

‘The Obscure, the Implied and the Illegal’ 223

B. The French Approach In conformity with the aim of the reformed Code50 to make the law on contract more accessible and intelligible (as well as updating it), the structure of the Code on contract law has been considerably revised. Within Book Three, Title III, Chapter I contains a set of ‘Introductory Provisions on Contract’. These set out general principles, of which three are of particular importance for the present purposes: the freedom to contract; the binding force between the parties of contracts which have been ‘lawfully formed’; and the principle that contracts must be negotiated, formed and performed in good faith. The last provision is stated to be a ‘matter of public policy’.51 Chapter II deals with the formation of contracts. Article 1128, which sets out the elements that are necessary for the validity of a contract, is particularly important in relation to ‘illegality’.52 Chapter III is headed ‘Contractual Interpretation’. There are five short articles, Articles 1188–1192, which replace nine in the old Code.53 The general principle is stated in Article 1188. The second part of that article is new. The whole article reads as follows: A contract is to be interpreted according to the common intention of the parties rather than stopping at the literal meaning of its terms. Where this intention cannot be discerned, a contract is to be interpreted in the sense in which a reasonable person placed in the same situation would give to it.

Article 1188 is followed by more particular rules, which broadly follow those of the old Code. Article 1192 codifies a decision of the Cour de Cassation of 1872,54 which established the doctrine of ‘dénaturation’; that is, where a term is clear and unambiguous, it is not to be subject to interpretation, ‘as doing so risks [its] distortion’.55 Under the old Code, the articles relating to interpretation were regarded not as rules of law, but, in the celebrated words of Jean Carbonnier, as ‘un petit guide-âne’ – a donkey to guide you, or, here, to guide the judge. That is also the intention of the new Code. But what is meant in Article 1188 by the ‘common intention of the parties’? Contracting parties can, of course, make their intention known expressly by technical methods in the contract itself. They can use a prefatory clause which indicates certain intentions or guidance; or a definitions clause; or a clause indicating which term has priority over another in the case of a clash. But if those express

50 I have used the English translation of the new Code prepared by J Cartwright, B Fauvarque-Cosson and S Whittaker; this was commissioned by the Direction des affaires civiles et du sceau, Ministère de la Justice, République française, and is published online at: www.textes.justice.gouv.fr/art_pix/THE-LAW-OF-CONTRACT-2-5-16.pdf. Other English translations are my own. 51 Arts 1102, 1103 and 1104. The second part of Art 1104, making the requirement of ‘good faith’ a matter of public policy, is new. It means that parties cannot contract out of that requirement. 52 The three elements (replacing the former four) necessary for the validity of a contract are: the consent of the parties, their capacity to contract and ‘content which is lawful and certain’. The old Art 1108 provided also that a contract was valid only if it had a cause which was lawful. The ambit of the requirement of ‘cause’ gave rise to much case law and doctrinal debate. Views differ on the effect of its deletion from the Code as a requirement for the creation of a valid contract: for discussion, see Rowan (n 12) esp 816–17. 53 Old Arts 1156–64. 54 Cass Civ 15 avr 1872, Veuve Foucauld. The principle is based on the Latin maxim interpretio cessat in claris. The need for Art 1192 resulted from the fact that, over time, despite the Cour de Cassation’s decision in 1872, many courts had disregarded the first part of what is now Art 1188 and had asserted that further interpretation was rendered necessary for one reason or another. Successful appeals on the grounds that the judge at first instance (juge du fond) had overstepped his interpretive powers and engaged in dénaturation du contrat were few. See, eg Cass Civ 3ème, 6 fév 2002 n° 00-12.675; Cass Com, 22 mai 2002, n° 99-11.052; Cass Civ 3ème, 20 av 2017, n° 16-13.462. 55 The French Code, Art 1192 states that ‘clear and unambiguous terms are not subject to interpretation as doing so risks their distortion’ (the French text is ‘On ne peut pas interpréter les clauses claires et précises à peine de dénaturation’).

224 Richard Aikens means of interpretation do not exist, or are not helpful, it will be the judge who has to interpret the disputed term. It is well established that the judge at first instance (juge du fond) has ‘le pouvoir souverain’ – literally, the sovereign power – to interpret a contract in accordance with Article 1188.56 But what tools can he or she use consistently with Article 1188? The judge has to search for the ‘real’ intention of the parties, without being constrained by the wording of the contract. Judges are thus entitled to look at negotiations,57 parties’ subjective intentions58 and their post-contract actions.59 But if the common intention of the parties cannot be discerned, the judge will be in the position of the ‘reasonable person’ who is ‘placed in the same situation’, and must give the interpretation that such a person would give it. The phrase is not precise, and there has been discussion as to who constitutes the ‘reasonable person’ and what is meant by ‘the same situation’. As judges are supposedly reasonable people, the judge can effectively give the contract the interpretation that he or she would give it, subject always to the requirement that the judge must not subject the contract to dénaturation.60 The articles that follow Article 1188 are therefore just examples of the tools that a judge can use in the quest for the intention of the parties and in order to interpret the contract. Article 118961 reflects existing case law.62 Article 1190 gives guidance to the interpreting judge who encounters ambiguous terms in two particular types of contract: a ‘bespoke contract’ and a ‘standard form contract’.63 It encapsulates a contra proferentem principle. In the first case, the ambiguity is interpreted ‘against the creditor’;64 in the second case, it is interpreted ‘against the person who put it forward’ (the proferor). Article 1191 states that ‘where a contract term is capable of bearing two meanings, the one which gives it some effect is to be preferred to the one which makes it produce no effect’.65 This guideline must assume that there is an ambiguity in the contract term, with the result that it is not subject to Article 1192. Article 1192 reflects well-established case law, which held that a judge must not dénaturer – that is, change the sense of a clause or a contract (or refuse to apply it) – where the meaning is clear. However, there still remains the issue of when a term is not clear or is a mbiguous. That is

56 Established by the Cour de Cassation as early as 1808: Cass, 2 fév 1808. 57 Cass Civ 1ère, 18 fév 1986, n° 84-12.347. 58 G Chantepie, Répertoire de Droit Civil (Dalloz, janvier 2018) para 29. See also C Valcke, ‘Contractual Interpretation at Common Law and Civil Law: An Exercise in Comparative Legal Rhetoric’ in JW Neyers et al (eds), Exploring Contract Law (Oxford, Hart Publishing, 2009) 86. Valcke argues that French jurists have ‘long considered that the subjective conception of contractual intention is logically entailed by the moral ideal of the “autonomy of the will” which traces the source of contractual obligation to the will of the individual as self-governing agent’. 59 Cass Civ 1ère, 13 déc 1986, n° 19.068. 60 This is the subject of much doctrinal writing. It means effectively that the court cannot give a meaning it likes if the wording is clear – as codified in Art 1192 of the present Code. 61 ‘All the terms of a contract are to be interpreted in relation to each other, giving to each the meaning which respects the consistency of the contract as a whole. Where according to the common intention of the parties, several contracts contribute to one and the same operation, they are to be interpreted by reference to this operation.’ 62 See, eg Cass Civ 1ère, 28 octobre 2015, n° 14.11498, on the ability to look at related contracts to interpret one contract in the group concerning one operation. 63 These terms are the translations used by Cartwright et al (n 50) for contrat de gré à gré and contrat d’adhésion. They characterise the first as a contract ‘in which the parties come together in an amicable agreement’ and, I would add, where they have negotiated the terms as more or less equals. It is to be contrasted with a ‘standard form contract’, or contrat d’adhésion, where, almost by definition, the parties are not of equal standing and one party has to adhere to the terms set out in the standard form of the other. 64 As Cartwright et al (n 50) point out, in the Code the word ‘creditor’ or ‘créancier’ is used in a general sense to denote a person having a ‘right’ corresponding to an obligation; it is not confined to money transactions generally as understood in English law terminology. 65 This reflects the old Art 1157.

‘The Obscure, the Implied and the Illegal’ 225 for the juge du fond to decide in the first place. However, if the judge resorts to a principle of law to arrive at a particular term’s interpretation that can be subject to correction, it will ultimately be decided by the Cour de Cassation.66 The new Civil Code on interpretation of contracts is coherent. It is the result of much debate in projects for the reform of the Civil Code on contracts. It is principled. But it is too soon to say whether there will be any difficulties with the application of its terms.

C. Comparison of the Two Approaches A person who knows the English law on interpretation of written contracts might conclude that the Civil Code provisions on interpretation bear a striking similarity to the English law principles. Some things are the same: finding the true intention of the parties; looking at the contract in its context; construing it as a whole; and not ‘interpreting’ when the wording is clear and unambiguous. But there are vital differences. The chief distinction is that the French first instance judge is, subject to the dénaturation doctrine, the master of interpretation. The judge has a wide scope to interpret the terms. Unless he or she errs on some principle of law, the interpretation cannot be challenged in a higher court. The second distinction is the fact that the French juge du fond can look at many more aspects than the English judge to ascertain the parties’ intention: their subjective views; the negotiations and matters which occurred after the contract was concluded; and other contracts (even if not part of a group of which the instant contract is one). As such, the argument of the appellants in the House of Lords in Prenn, in which they sought without success to introduce evidence of the negotiations, would have succeeded before a French juge du fond. Moreover, the view of the juge du fond would not be appealable, subject to the doctrine of dénaturation. Would the two approaches produce different results to a similar problem? If a French judge were to consider the facts of Arnold, the judge’s ability to look at the subjective intention of the parties, at their negotiations and at their acts post-contract might lead the judge to the same result as that favoured by Lord Carnwath in his minority judgment. The French judge might say that the parties contemplated high inflation and an increase in prices, that this is what had happened for some time after the contract had been made and that the parties assumed that this would continue. So the contract should be construed to take account of changing circumstances. However, it is possible that such a construction would be regarded as an impermissible dénaturation.

III. Implication of Terms A. The English Approach Why should a court ‘imply’ a term in a contract, especially when the contract has been written and is detailed and apparently drafted carefully? In most cases, the court will not do so. But, as Sir Thomas Bingham MR pointed out in Philips Electronique Grand Public SA v British Sky

66 F

Terré, P Simler and Y Lequette, Droit Civil, Les Obligations, 11th edn (Paris, Dalloz, 2013) n° 459.

226 Richard Aikens Broadcasting Ltd,67 the question of whether a term should be ‘implied’ in a contract and, if so, what its content should be ‘almost inevitably arises after a crisis has been reached in the performance of the contract’ and there are no express terms in the contract to deal with what is to happen in that event. One party then suggests that a term should be ‘implied’ in the contract which will magically deal with the contentious issue at hand. As stated earlier, English law distinguishes between terms that are to be implied ‘by fact’, ‘by law’ and ‘by custom and usage’.68 This chapter is concerned (at least from the perspective of English law) with terms ‘implied in fact’. Many instances of ‘implied terms in fact’ appeared in the law reports before the celebrated case of The Moorcock in 1888.69 But that case is regarded as the source of the modern law of England on when and why a term can be implied in a contract to deal with a factual situation which is not covered by an express term. In The Moorcock, the defendant wharfingers entered a contract with the owners of the steamer Moorcock which provided that the vessel could unload cargo into their wharf.70 This required the ship to moor alongside the adjacent jetty, which was not owned or controlled by the wharfingers. It was understood that the vessel, when laden, would have to take the ground when alongside, as the tide ebbed. The riverbed was uneven, so that as the ship took the ground she broke her back. The shipowners sued the wharfingers in contract for damages. One of their arguments was that there was an implied term in the contract that the wharfingers would take reasonable care to ascertain that the riverbed at the jetty was in such a condition as not to endanger the vessel using their premises in the ordinary way. The shipowners succeeded on this argument, before Butt J and in the Court of Appeal. In the Court of Appeal, Lord Esher MR did not consider the legal basis for implying a term in a contract when there was no express term to cover the factual situation which had given rise to the claim for breach of contract. However, Bowen LJ did so. He said that:71 an implied warranty, or, as it so called, a covenant at law, as distinguished from an express contract or warranty, really is in all cases founded on the presumed intention of the parties, and upon reason … I believe if one were to take all the cases, and there are many, of implied warranties or covenants at law, it will be found that in all of them the law is raising an implication from the presumed intention of the parties with the object of giving to the transaction such efficacy as both parties must have intended that at all events it should have.

Subsequently, in two decisions of the Court of Appeal,72 two particular ‘tests’ were elaborated, which the court said had to be fulfilled before it would be prepared to imply a term in a contract as a matter of ‘fact’. First, the term to be implied has to be ‘necessary in the business sense to give efficacy to the contract’;73 secondly, the term has to be ‘so obvious that it goes without saying’. Three further ‘tests’ were also developed in the cases: the proposed implied term must be reasonable and equitable; it must be capable of clear expression; and it must not contradict the express

67 Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 (CA) 482. 68 See n 3 and accompanying text above. 69 The Moorcock (1889) 14 PD 64 (CA). 70 It is unclear from the law reports whether the contract was oral or in writing. Either way, it is clear that it had few express terms. 71 The Moorcock (n 69) 68. 72 Reigate v Union Manufacturing Co (Ramsbotton) Ltd [1918] 1 KB 592 (CA); Southern Foundries (1926) Ltd v Shirlaw [1939] 2 KB 206 (CA). 73 The requirement of ‘necessity’ had been emphasised by the House of Lords in French and Co Ltd v Leeston Shipping Co Ltd [1922] 1 AC 451 (HL) 454–55 (Lord Buckmaster).

‘The Obscure, the Implied and the Illegal’ 227 terms of the contract. These five ‘tests’ were set out by Lord Simon in BP Refinery (Westernport) Pty Ltd v Shire of Hastings.74 However, in 1973, in a little remarked upon decision of the House of Lords, Trollope & Colls Ltd v North West Metropolitan Regional Hospital Board,75 Lord Pearson had had to consider whether a term was to be implied in a building contract. He said that: [an] unexpressed term can be implied if and only if the court finds that the parties must have intended that term to form a part of their contract … a term which, though tacit, formed part of the contract which the parties made for themselves.76

He held there was no implied term in that case. Then in Equitable Life Assurance Society v Hyman,77 Lord Steyn referred to a term being implied ‘in fact’ from the language of the articles of association of the Equitable Life Assurance Society ‘read in its particular commercial setting’. It was Lord Pearson’s and Lord Steyn’s analyses in these two cases which Lord Hoffmann said supported his reasoning in Attorney-General of Belize v Belize Telecom Ltd.78 Lord Sumption, in his Harris Lecture, said that Lord Hoffmann’s approach in Belize came close to ‘abolishing the implication of terms as a distinct legal concept’, where the implication was one ‘in fact’ rather than ‘in law’.79 What was so apparently revolutionary in Lord Hoffmann’s approach to ‘implication of terms’ in Belize? The Privy Council decision, on appeal from the Court of Appeal of Belize, concerned the terms of the articles of association of Belize Telecommunications Ltd. The articles of association did not deal with the factual situation which had arisen concerning the appointment and removal of directors. The issue was whether directors who had been appointed in accordance with the articles of association when fact situation A existed could be removed when fact situation B pertained and the articles of association did not deal with what was to happen in those circumstances. One party said the directors were irremovable; the other said that the articles of association must be construed so as to provide, by implication, that in fact situation B the relevant directors would cease to hold office. Lord Hoffmann, in giving the advice of the Board, made some ‘general observations about the process of implication’, whether it be in relation to a contract, a statute or articles of association.80 He started by restating what he had said in the ICS case: the meaning of an instrument is what it conveys to a reasonable person having all the background knowledge which would reasonably be available to the audience to whom the instrument was addressed. It was that ‘objective meaning’ which is conventionally called the ‘intention’ of ‘the parties’, Parliament or the author of the instrument.81 Lord Hoffmann then considered the case where the instrument does not expressly provide for what is to happen when ‘some event occurs’. In most cases, nothing is to happen, and if there is loss as a result, it ‘lies where it falls’. But there can be a case where the reasonable addressee would understand that the only meaning ‘consistent with the other provisions of the instrument, read against the relevant background, is that something is to happen’. In such a case, the court ‘implies a term as to what will happen if the event in question occurs’. So far, so apparently orthodox.

74 BP

Refinery (Westernport) Pty Ltd v Shire of Hastings [1977] UKPC 13, (1977) 180 CLR 266, 282–83. & Colls Ltd v North West Metropolitan Regional Hospital Board [1973] 1 WLR 601 (HL). 76 ibid 609 (emphasis added). I think these words were crucial to Lord Hoffmann’s reasoning in Belize (n 6). 77 Equitable Life Assurance Society v Hyman [2002] 1 AC 408 (HL) 458–59. 78 Belize (n 6) [19]–[21]. 79 Sumption (n 31) 85. 80 Belize (n 6) [16] ff. 81 ibid [16]. 75 Trollope

228 Richard Aikens It is the next two sentences which sent academic writers scurrying to their computers to comment,82 although judges did not immediately appear to fret about where the law was going.83 Lord Hoffmann said: ‘But the implication of the term is not an addition to the instrument. It only spells out what the instrument means.’ This followed, he said, as a matter of logic and also as a matter of authority, quoting the speech of Lord Pearson in Trollope,84 noted above. He added that the ‘tests’ for when a term was to be implied that had been adumbrated in previous cases were not ‘different or additional tests’. Instead, ‘there is only one question: is that what the instrument, read as a whole against the relevant background, would reasonably be understood to mean?’ Lord Simon’s list of ‘tests’ in the BP Refinery case85 was to be regarded as a ‘collection of different ways in which judges [had] tried to express the central idea that the proposed implied term must spell out what the contract actually means’ or why it cannot mean what one of the parties contends. Two aspects of Lord Hoffmann’s approach in Belize caused debate. The first was his analysis that the term which a court says is ‘implied’ in the contract is always there, although not revealed to the world until the court pulls back the curtain. Lord Hoffmann insisted that it must always have existed because that was what the parties intended should be the response in the circumstances which have arisen. Secondly, Lord Hoffmann seemed, at least to some academic writers, to abolish or weaken the traditional ‘necessity’ test by saying that there was a danger in detaching the phrase ‘necessary to give business efficacy’ from ‘the basic process of construction of the instrument’.86 The heated academic debate that followed Belize has been doused, to some extent, by the Supreme Court’s decision in Marks & Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd.87 The case concerned four leases and the issue was whether a term could be implied in them to entitle the tenant (M&S) to a return of rent, which had been paid in advance for a period when M&S had not occupied the properties. The judge had found for the tenant, holding that such a term could be implied. The Court of Appeal allowed the landlord’s appeal. In the Supreme Court, Lord Neuberger gave the judgment of the majority, dismissing the tenant’s appeal and holding that the tenant’s claim failed. Lord Carnwath agreed that the appeal should be dismissed, but disagreed with some of Lord Neuberger’s comments about Belize. Three comments of Lord Neuberger on Lord Hoffmann’s statements in Belize are important in the present context. First, he acknowledged as acceptable88 the notion ‘that a term will be implied if a reasonable reader of the contract, knowing all its provisions and the surrounding circumstances, would understand it to be implied’. This confirms that the test of whether a term is to be implied is an objective one. But the use of the words ‘would understand it to be implied’

82 Academic commentary on Lord Hoffmann’s approach is listed in H Beale (gen ed), Chitty on Contracts, 33rd edn (London, Sweet & Maxwell, 2018) para 14.003, fn 6; see also Y Goh, ‘Lost But Found Again: The Traditional Tests for Implied Terms in Fact’ [2016] Journal of Business Law 231, 231, fn 3. 83 See, eg Mediterranean Salvage & Towage Ltd v Seamar Trading & Commerce Inc (The Reborn) [2009] EWCA Civ 531, [2010] 1 All ER (Comm) 1, in the judgment of Lord Clarke MR passim, although his prediction, at [8], that Lord Hoffmann’s analysis would ‘soon be as much referred to as his approach to the construction of contracts in ICS’ may not now be correct. See also Stena Line Ltd v Merchant Navy Ratings Pension Fund Trustees Ltd [2011] EWCA Civ 543, [2011] PLR 223 [36]; Wuhan Ocean Economic & Technical Cooperation Co Ltd v Shiffahrts-Gesellschaft ‘Hansa Murcia’ MBH & Co KG [2012] EWHC 3104 (Comm), [2013] 1 All ER (Comm) 1277 [15]. But courts in other jurisdictions were not so sure: Goh (n 82) 231, fn 6. 84 Trollope (n 75) 609. Lord Hoffmann also referred to a statement of Lord Steyn in Equitable Life (n 76) 459: ‘If a term is to be implied, it could only be a term implied from the language of [the instrument] read in its commercial setting.’ 85 BP Refinery (n 74) 26. 86 Belize (n 6) [23]. 87 Marks & Spencer (n 3). 88 ibid [23].

‘The Obscure, the Implied and the Illegal’ 229 is interesting, because that suggests that the term is always there even if not visible until revealed by the court. Secondly, Lord Neuberger accepted that Belize did not ‘dilute’ the requirement that a term will not be implied unless it is ‘necessary for business efficacy’, contrary to the views of a number of academic writers.89 His third comment, on ‘the suggestion that the process of implying a term is part of the exercise of interpretation’, is the most important. Lord Neuberger said that ‘construing words used and implying additional words are different processes governed by different rules’. He added that When one is implying a term or a phrase, one is not construing words, as the words to be implied are ex hypothesi not there to be construed; and to speak of construing the contract as a whole, including the implied terms, is not helpful, not least because it begs the question as to what construction actually means in this context.90

Indeed it does, but Lord Neuberger does not elaborate on what he understood construction to mean ‘in this context’. Lord Hoffmann, however, has elaborated on what he regards ‘construction’ to mean in this context in his article, ‘Language and Lawyers’,91 although to my mind, that article does no more than expand what he had expressed in terser language in his speech in ICS. In both that speech and his article, Lord Hoffmann draws a distinction between the meaning of words and the meaning of the document as a whole – hence his distinction (unfairly decried by Lord Sumption) between the meaning of words, which is undoubtedly a matter of dictionaries and grammars, and the objective meaning of a document. A document’s ‘meaning’ must be what the parties would reasonably have understood it to mean, having used the words they did against the relevant background. Many cases have emphasised that the court cannot make or remake a contract for the parties when asked to interpret the contract or to decide whether there is an implied term in it. Logically, therefore, if a court holds that a certain term is to be implied because, in Lord Neuberger’s words, the ‘reasonable reader … would understand it to be implied’, that term must always have been there. Otherwise the court would be engaging in the forbidden process of making a contract for the parties, by adding terms to it. If you were to ask a judge who has ruled that a term is to be implied in a contract ‘are you therefore adding a new term to the contract?’, the answer would be ‘no’. So the process that the court engages in when asked to decide whether there is an implied term in a contract is one of determining the objective meaning of the document as a whole, as reasonably understood by the parties. Lord Hoffmann’s approach is logical, principled and coherent. It is consistent with the fundamental tenet of English contract law that the parties are free to make their contract and that it is not the court that does it for them, save in the very limited circumstances where the law implies a term, as mentioned above. Hence the insistence that any term implied ‘in fact’ must be necessary to give the contract ‘business efficacy’ rather than the court holding that it is ‘reasonable’. Overall, however, the striking fact about the English law governing the implication of terms is how rarely an argument to imply a term succeeds. It may do so when there are few terms in the contract, such as in The Moorcock; but it will not when the parties have agreed an elaborate contract, as in Marks & Spencer.

89 ibid 90 ibid

[23] and [24]. Lord Carnwath’s judgment also emphasises this point. [27]. (n 46) 572.

91 Hoffmann

230 Richard Aikens

B. The French Approach Neither the old Code nor the new Code deals specifically with ‘implied terms’ or ‘implied obligations’ in contracts.92 Yet there must be the same problem of ‘gaps’ in the terms of contracts governed by French law as there are in those governed by English law. French judges have fashioned means of providing ‘ad hoc gap fillers’.93 There is no general French law doctrine of when an ‘obligation implicite’ can be found or created. Nor has French law specifically differentiated between terms that are implied ‘in law’ and those that are implied ‘in fact’. However, three classes of ‘implied term’ can perhaps be discerned, which are similar to those found in English law: terms implied on the basis of custom or usage; terms implied ‘by law’; and terms implied ‘in fact’. For all of these three, the judge can invoke Article 1194 of the new Code, with its reference to obligations being created by the ‘consequences’ of ‘equity, usage or legislation’.94 In general, there has been a reluctance by French courts to permit implied terms except in certain defined situations.95 This reluctance is probably due to the very strong influence of the writings of Pothier in the eighteenth century and of the distinguished jurist Josserand in the twentieth century. Pothier was insistent that only terms which were necessary to the nature of the contract could be implied.96 Josserand was a critic of judges and legislators who interfered with freedom of contract. He coined the phrase ‘forçage du contrat’, meaning any attempt to create contractual terms which were not there and which should not be invented.97 The chief areas of the law where the judges have been most inventive of implied terms are in transport contracts, service contracts and other contractual situations where individual safety is involved. This type of implied term has been given the sobriquet ‘l’obligation de sécurité’. The origin of this ‘obligation to maintain a contracting party safe and sound’ is a decision of the Cour de Cassation in 1911. A passenger was injured by a falling barrel in the course of a voyage from Algiers to another port in Algeria, then called Bône, now Annaba. The contract terms on the ticket contained an exclusive jurisdiction clause in favour of the Tribunal de Commerce of Marseilles. The passenger brought a claim for damages for personal injury in the court of Algiers, and the Court of Appeal of Algiers refused to enforce the exclusive jurisdiction clause. The Cour

92 An early draft of the new Code, prepared by the Ministry of Justice, had proposed a distinction between express and implied obligations in contracts. In that draft, whether a term was to be implied would have depended on the intention of the parties; the parties’ relationship (eg whether there was a ‘course of dealing’); the nature of the contract; all to be considered in ‘the light of the law, custom and equity’, thus reflecting to some extent what has become Art 1194. However, this idea was dropped in subsequent drafts: Y Picod, ‘Obligations’ in Répertoire de Droit Civil (Paris, Dalloz, June 2017) s 2, Classifications, Art 1 §3, paras 50–51. 93 Lord Steyn’s phrase in Equitable Life (n 77) 459. 94 Replacing old Art 1135. The two articles are in almost identical terms. New Art 1194 states, in full, in English: ‘Contracts create obligations not merely in relation to what they expressly provide; but also to all the consequences which are given to them by equity, usage or legislation.’ (in French: ‘Les contrats obligent non seulement à ce qui y est exprimé, mais encore à toutes les suites que leur donnent l’équité, l’usage ou la loi.’). 95 A classic statement of the courts’ reluctance to find additional terms is found in Cass Civ 6 mars 1876, D 76.1193: ‘Dans aucun cas, il n’appartient aux tribunaux, quelque équitable que puisse leur paraître leur décision, de prendre en considération le temps et les circonstances pour modifier les conventions des parties et substituer des clauses nouvelles à celles qui ont été librement acceptés pas les contractants.’ (in English: ‘in no circumstances, even if it would seem equitable to do so, should the court take into account the time and circumstances in order to modify the parties’ contract and to substitute new terms for those that were freely agreed by the parties.’). 96 ‘Les choses qui sont seulement de la nature du contrat sont celles qui, sans être de l ’essence du contrat, font partie du contrat, quoique les parties contractantes ne s’en soient point expliquées, étant de la nature du contrat que ces choses y soient renfermées et sous-entendues’: Traité des obligations 1ère partie 1761 (in English: ‘Matters which are of the nature of the contract include those are which, although not of its essence, are part of the contract even though the parties have not expressed it; being of the nature of the contract they are included in it and are implied’). 97 L Josserand, Cours de Droit Civil Positif Français (Paris, Sirey, 1929–30) vol 2, para 491.

‘The Obscure, the Implied and the Illegal’ 231 de Cassation overturned that decision.98 It held that the terms of the ticket, including the jurisdiction clause, were a part of the contract of carriage and were valid. In doing so, the court held that the performance of the contract of carriage by the carrier included an obligation to carry the passenger safely to his destination. The short judgment refers laconically to Article 1134 of the old Code.99 Whilst not stating explicitly that this obligation of ‘safety’ was part of the requirement to perform the contract ‘in good faith’, the court nevertheless implied the safe carriage obligation, with the consequence that if that obligation were broken, the claimant would be entitled to damages and interest for breach of contract. Josserand was very critical of this judicial intervention, which, he said, had nothing to do with the intention of the parties or their agreement. It was nothing more than ‘socialisme juridique, du dirigisme contractuel’.100 Nonetheless, this ‘obligation of safety’ has subsequently been held to apply in contracts for all types of transport, even ski lifts. It has been applied to contracts concerning sports and leisure activities, and, indeed, all activities where the user might be exposed to personal injury. It has been extended into the sphere of medical supplies and hospital infection,101 and to restaurants.102 But, as Hocquet-Berg has observed, it is arguable that the ‘obligation de sécurité’ has now passed its zenith, given the increasing legislation for consumer protection on all fronts.103 There are two other principal types of obligation that the French courts have ‘implied’ in contracts: first, an obligation of ‘information and advice’; and secondly, a ‘security obligation’.104 To found such obligations, the courts have invoked the old Article 1135 (now Article 1194) as well as the obligation to perform a contract in ‘good faith’, as stated in the old Article 1134 and now in Article 1104. There is considerable debate among French jurists about the basis for judges ‘implying’ terms in contracts where there is no express obligation. Thus, it has been stated that courts have used the old Articles 1134 and 1135105 ‘to create a breach through which courts have charged to insert implied obligations in the contract of which the parties had never dreamt’.106 In particular, it is argued that courts have used the Code’s requirement that a contract be performed in ‘good faith’ to remodel contracts, on the basis of what the parties might reasonably have wished, even if not stated expressly. It could be said that this is discovering the true intention of the parties.107 However, it has also been argued that it is a valid use of the judge’s creative power to ‘imply’ terms using the requirement that the parties perform their contract ‘in good faith’.108 This use of the judge’s creative power has been defended, on the one hand, because the judge is the guardian of the parties’ duty to obey the contract and because the contract is an instrument of ‘solidarity’

98 In 1911 Algeria was, of course, still part of metropolitan France. 99 This provided, in part, that contracts must be performed ‘in good faith’. Art 1104 of the present Code provides that ‘Contracts must be negotiated, formed and performed in good faith. This provision is a matter of public policy.’ 100 L Josserand, Le Contrat Dirigé (DH 1933, chron 89), cited in S Hocquet-Berg, ‘Le Fabuleux Destin de l’Obligation de Sécurité’ [2019] Résponsabilité Civile et Assurances 13, 14, fn 4. 101 Hocquet-Berg (n 100) 14 and fns 9–14. 102 Cass Civ 1ère, 15 déc 2011, n° 10-23.528. 103 Hocquet-Berg (n 100) 15. 104 Respectively ‘l’obligation d’information’ and ‘l’obligation de garantie’. See P de Tourneau and M Poumarède, Répertoire de Droit Civile: Bonne Foi (Paris, Dalloz, 2017) ch 3, section 1, Art 1, §2. See also A Vey, ‘Assessing the Content of Contracts: Implied Terms from a Comparative Perspective’ (2011) 22 European Business Law Review 501. 105 Now respectively Arts 1103 and 1104 (for former Art 1134) and 1194 (for former Art 1135). 106 L Aynes, ‘Le Contrat, Loi des Parties’ (2005) 17 Cahiers du Conseil Constitutionnel (Dossier: Loi et Contrat). 107 ibid; Picod (n 92); de Tourneau and Poumarède (n 104) paras 72–74. 108 Picod (n 92) para 54.

232 Richard Aikens between the parties; on the other hand, and more specifically, it has been defended as being part of the doctrine of ‘solidarisme contractuel’.109

C. Comparison of the Two Approaches Clearly, both systems of law make use of the notion of implied terms to fill in ‘gaps’ in the terms of a contract. In both systems there are conflicting arguments about the theoretical basis for a judge being able to imply a term, at least where it is one not founded on custom and usage or implied by virtue of legislation. Although the UK Supreme Court has partially suffocated the theory that terms implied by fact are a part of the interpretive process, it adheres to the idea that an implied term is there because it is the true intention of the parties and is a necessary adjunct of the express terms of the contract. That logic seems to be the basis for the French courts’ willingness to imply terms ‘in fact’. However, the French courts appear to be less rigid in relation to the requirement of ‘necessity’ than an English court would be. French law, however, requires that contracts be negotiated and performed in good faith. Perhaps that is why French courts are more willing to imply terms in contracts because it is ‘reasonable’ to do so. Thus, I would have expected a French court to imply a similar term to that implied in The Moorcock. Equally, an English court would be bound to imply a term in a contract of carriage of a passenger about his safety. But the English court would put it as a contractual duty to take reasonable care. It would not be an absolute obligation on the carrier.

IV. Illegality A. The English Approach English contract law only concerns itself with questions of illegality once it is established that, prima facie, there is a contract between the parties.110 ‘Illegality’ has always been seen as a defence to claims for performance or for damages for non-performance of a contractual obligation. From the mid-eighteenth century, if not earlier, until 2016, the English courts struggled to apply consistently a legal principle or rule of public policy, or at least a maxim, about the effect of illegality on a claim, whether in contract or tort, or to negate a transfer of property or goods.111 The relevant statement (to avoid having to characterise its legal nature for the moment) was famously enunciated by Lord Mansfield CJ in 1775, in Holman v Johnson:112 ‘if, from the plaintiff ’s own stating or

109 D Mazeaud, ‘Le Nouvel Ordre Contractuel’ (2003) 22 Revue des Contrats 295. 110 It is probably because of the analysis that an illegal contract exists that the English case law prior to Patel (n 4) was so inconsistent about characterising the effect of illegality on a contract: courts have variously said the contract is ‘void’, ‘void ab initio’, ‘voidable’ or ‘unenforceable’. 111 In Gray v Thames Trains Ltd [2009] UKHL 33, [2009] 1 AC 1339 [30], Lord Hoffmann said that the Latin maxim ex turpi causa expressed ‘not so much a principle as a policy’. He did not elaborate on the difference between ‘principle’ and ‘policy’, save to say that ‘that policy is not based on a single justification but on a group of reasons, which vary in different situations’. In Patel (n 4), all of the Supreme Court Justices accepted that the ‘illegality’ defence was based on public policy rather than considerations of justice, with the implication that it could therefore work injustice on occasion. For a different view, see NJ McBride, ‘Not a Principle of Justice?’ in S Green and A Bogg (eds), Illegality after Patel v Mirza (Oxford, Hart Publishing, 2018). 112 Holman v Johnson (1775) 1 Cowp 341, 343; 98 ER 1120, 1121.

‘The Obscure, the Implied and the Illegal’ 233 otherwise, the cause of action appears to arise ex turpi causa,113 or the transgression of a positive law of this country, there the court says he has no right to be assisted’. Yet, in that very case, Lord Mansfield held that a seller of goods (in Dunkirk) who knew that the buyer would smuggle the goods into England was nonetheless entitled to the price of the goods sold. The reason given was that the seller was not himself involved in the smuggling. Decades later, in Pearce v Brooks,114 the claim of a coachbuilder who sued a prostitute for the hire of an ‘ornamental brougham’ failed once the jury found as facts that the brougham was hired for the purpose of prostitution and the coachbuilder knew that. Knowledge itself was enough, even though there was no suggestion the coachbuilder was involved in procuring clients for the hirer. In Parkinson v College of Ambulance Ltd,115 the claimant sued unsuccessfully for the return of a gift of £3,000 to a charity. He had given the money after the charity had illegally (and fraudulently) promised that it could obtain a knighthood for him. Lush J said the court could not adjudicate on such a claim with ‘propriety or decency’.116 But in St John Shipping Corp v Joseph Rank Ltd,117 a shipowner was held entitled to the freight owed for the carriage of goods across the Atlantic, despite the fact that the vessel had become overloaded during the voyage and the Master had thereby committed an offence under the Merchant Shipping Acts in the course of carrying out the contract. A list of apparently inconsistent decisions, all allegedly applying Lord Mansfield’s statement, would fill pages. In some cases, the reason given for not applying Lord Mansfield’s statement was that the relevant act was not ‘illegal’. In others, the reason was that there was an insufficient connection between the act and the contract. In yet others, it was that penalising the claimant for the ‘illegality’ would be ‘disproportionate’. As Lord Sumption has said, the statements made in this ‘large body of inconsistent authority … rarely [rise] to the level of general principle’.118 The problem arises in broadly three types of case. There are those, like the St John Shipping case, where the claimant asserts that it is contractually entitled to a sum either in debt or in damages. Then there are others, like Parkinson and Patel itself,119 where the claimant seeks the return of money paid because a contract has not been carried out for some reason or other. Finally, there are cases, like Bowmakers Ltd v Barnet Instruments Ltd,120 where the claimant sues for the return of goods, and others, such as Tinsley v Milligan,121 in which the claimant asserts an interest in real property. In all the cases, the defendant pleaded that some aspect of the transaction connected with the claim was ‘tainted’ with illegality and that this debarred the claimant from succeeding. There are three issues that English courts have been concerned with in the cases about the effect of ‘illegality’ on a contract. First, what type of conduct is to be characterised as ‘illegal’

113 Literally, ‘from an immoral cause’; hence, presumably, the contrast with the following phrase about the ‘transgression of a positive law of this country’. 114 Pearce v Brooks (1866) LR 1 Ex 213 (Exch). 115 Parkinson v College of Ambulance Ltd [1925] 2 KB 1 (KB). In Patel (n 4) [150], Lord Neuberger said that the case should be overruled. At [117]–[118], Lord Toulson appeared to doubt the result was correct, but did not say it should be overruled. Neither did others in the majority. Lord Sumption did not comment on its correctness: see [228]. 116 ibid 13. 117 St John Shipping Corp v Joseph Rank Ltd [1957] 1 QB 267 (QB) esp 288 (Devlin J). 118 Laboratoires Servier (n 7) [14]. 119 Patel had given £620,000 to Mirza pursuant to an agreement whereby Mirza would buy shares on the basis of insider information. The information never materialised and so the purchase was not made, but Mirza refused to return the money, asserting that Patel had no right to it as the agreement was illegal, constituting a conspiracy to commit an offence under the Criminal Justice Act 1993, s 52. On this point, see Patel (n 4) [266] (Lord Sumption). 120 Bowmakers Ltd v Barnet Instruments Ltd [1945] KB 65 (CA). 121 Tinsley v Milligan [1994] 1 AC 340 (HL).

234 Richard Aikens for the purposes of possibly invoking the ‘illegality’ defence? Secondly, in what circumstances will a court hold that there is a sufficient connection between the contract and the ‘illegal’ act or conduct, such as to lead the court to say that the contract is ‘tainted’ with illegality? Thirdly, what are the consequences if it reaches the conclusion that the contract is ‘tainted’ by illegality? On the first issue, English law was clarified by the Supreme Court’s decision in Les Laboratoires Servier v Apotex Inc.122 For present purposes, the relevant issue to be decided by the Supreme Court was whether the fact that Apotex had produced and exported a drug from Canada to the UK in disregard of a valid Canadian patent held by Servier was an act that was sufficiently ‘illegal’ to prevent Apotex from making a claim under a cross-undertaking for damages. This had been given by Servier in English proceedings in return for an injunction to prevent Apotex from importing the drug into the UK. Subsequently, the English court held that the injunction Servier had obtained had been wrongly granted. The judgment of the majority of the Supreme Court was given by Lord Sumption. He said that the types of act which can trigger the ‘illegality’ defence are criminal acts and ‘immoral or illegal’ acts. Put more generally, for the purpose of the ‘illegality’ defence, relevant acts are those which ‘engage the interests of the state, or as we would put it today, the public interest’. The most straightforward examples are criminal acts.123 A different category is where a law prohibits a particular type of contract, even if not making it a criminal offence to enter into it. In that case, the contract is void ‘and any right derived from it is non-existent’.124 But, beyond criminal acts, the dividing line between legality and illegality is more difficult to discern and almost impossible to define in general terms. Apart from criminal acts, Lord Sumption labelled the type of acts which are to be regarded as ‘illegal’ for the purposes of the ‘illegality’ defence as being ‘acts which are contrary to the public law of the state and engage the public interest’, or what he calls ‘quasi-criminal acts’. It is easier to give examples. He listed: dishonesty or corruption (even in civil disputes), some ‘anomalous categories of misconduct, such as prostitution’, and ‘the infringement of statutory rules enacted for the protection of the public interest and attracting civil sanctions of a penal character’.125 But torts (other than where dishonesty is an essential element), breaches of contract, statutory wrongs and other civil wrongs are not to be characterised as ‘illegal’ for these purposes. This is because they offend against private, not public, interests. There are cases at the margins which might not engage the public interest, even though a criminal act is involved.126 Therefore, on this analysis, an infringement of a patent, being a tort, does not engage the public interest and cannot trigger the ‘illegality’ defence.127 The second and third issues were the subject of the Supreme Court’s decision in Patel. Hitherto, the accepted view had been that the ‘illegality’ defence could be invoked if a claimant had to rely on an illegal act to prove his claim. This had become known as the ‘reliance principle’.128 The Law Commission had started a project to consider illegality after Tinsley v Milligan and had

122 Laboratoires Servier (n 7). 123 Although some criminal acts may be too trivial to constitute ‘turpitude’ for the purposes of the ‘illegality’ defence: Gray (n 111) [83] (Lord Rodger). 124 Laboratoires Servier (n 7) [23]. 125 ibid [25]. In the last category Lord Sumption placed breaches of competition law, although whether that could be successfully invoked depended on the circumstances in which the breach was raised as an ‘illegality’ defence: Safeway Stores Ltd v Twigger [2010] EWCA Civ 1472, [2011] Bus LR 1629. 126 ibid [29]. See also J Goudkamp, ‘The Law of Illegality: Identifying the Issues’ in Green and Bogg (n 111) 52–54. 127 ibid [30]. The other Justices agreed. This issue was not debated in Patel (n 4). 128 See Tinsley (n 121), as explained by Lord Sumption in Patel (n 4) [234]. As Lord Toulson pointed out in Patel (n 4) at [20], the reasoning in Tinsley had not been directly called into question between 1994 and 2016.

‘The Obscure, the Implied and the Illegal’ 235 produced three reports.129 Ultimately, it had concluded that legislative change was not needed outside the area of trusts law.130 Courts had then commented on the unsatisfactory way English law dealt with a claim in relation to which there was some ‘illegal’ act, in the sense clarified in Laboratoires Servier.131 The majority’s decision in Patel swept away the ‘reliance rule’ and sought, instead, to find a more sophisticated test for when the ‘illegality’ defence could be invoked by the court. Lord Toulson, giving the leading judgment of the majority, summed up the new principle.132 I would paraphrase and enumerate this principle as follows: (1) the rationale for the ‘illegality’ defence is the protection of the public interest in the integrity of the legal system;133 (2) the question is whether enforcement of the particular claim would be harmful to that integrity; (3) the framework within which to examine each case requires consideration of (a) the underlying purpose of the prohibition which has been transgressed and whether that purpose will be enhanced by denial of the claim, (b) any other relevant public policy on which the denial of the claim may have an impact and (c) whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts; and (4) in considering those framework issues, ‘various factors’ may be relevant, but the court cannot decide a case ‘in an undisciplined way’; rather, it must do so by a ‘principled and transparent’ assessment of the considerations identified. Lord Toulson drew on the relevant part of Andrew Burrows’s Restatement of the English Law of Contract as an inspiration for his analysis,134 and although the majority did not adopt the Restatement’s approach in its entirety,135 Burrows appears to be content with the court’s analysis.136 What of the third issue: the consequences of holding that the ‘illegality’ defence is a pplicable? This was not fully worked out in Patel. The case concerned a claim in unjust enrichment to recover money which had been paid by the claimant to the defendant. The defendant sought to escape restitutionary liability by arguing that the money had been paid to buy shares in accordance with a plan which would have resulted in illegal insider trading if it had been carried out. All the Justices thought that restitution was the correct result. But, curiously, none of the majority expressly decided the case by an application of Lord Toulson’s principles and factors. It seems plain, however, that, save in exceptional circumstances, if there is a claim for restitution of money on the basis of unjust enrichment, then it will be granted if it is otherwise just to do so, despite any ‘illegality’.137 What of other contractual claims or claims to interests in property? Patel does

129 Illegal Transactions: The Effect of Illegality on Contracts and Trusts (Law Com CP No 154, 1999); The Illegality Defence: A Consultative Report (Law Com CP No 189, 2009); The Illegality Defence (Law Com No 320, 2010). 130 Law Com No 320 (n 129) para 3.40. The history of the Law Commission’s work on illegality and the relationship between this work and judicial developments is discussed in J Lee, ‘Illegality, Familiarity and the Law Commission’ in Green and Bogg (n 111). 131 For the final rounds prior to Patel, see the debate between Lords Sumption, Toulson and Hodge in Bilta (n 7), following on from the judgment of Lord Wilson in Hounga v Allen [2014] UKSC 47, [2014] 1 WLR 2889 and the differing views of Lords Sumption and Toulson in Laboratoires Servier (n 7). 132 Patel (n 4) [120]. 133 Lord Toulson also mentioned protection of ‘certain aspects of public morality’, but did not elaborate as that was not in issue in that case. Just as well, perhaps, as public morality is a very dangerous area for judges to venture into. 134 Patel (n 4) [82]–[94], citing A Burrows, A Restatement of the English Law of Contract (Oxford, Oxford University Press, 2016). 135 ibid [108], where Lord Toulson rejected the approach proposed by Burrows (n 134) 229–30, that there should be a ‘range of factors approach’ guided by eight identified factors. 136 A Burrows, ‘A New Dawn for the Law of Illegality’ in Green and Bogg (n 111). 137 Patel (n 4) [116] (Lord Toulson). Practical examples of when an exception would apply are difficult to find. As Lord Toulson pointed out, a person demanding the return of money paid for a contract killing that was not done is unlikely to sue. The fate of a present-day claimant like Parkinson, the unsuccessful briber, remains undecided.

236 Richard Aikens not deal with them specifically. Lord Neuberger pointed to decisions in the eighteenth and nineteenth centuries where the courts had held that in the case of money being paid for an ‘unlawful purpose’ or ‘illegal transaction’, the parties should be put in the position which they had occupied before the transaction was begun.138 Lord Sumption said that this rule had applied only to ‘executory contracts’ and not to ‘executed contracts’. He suggested (and, by implication, Lord Clarke agreed) that the ‘rational rule’ (which he boldly said he would ‘hold to be the law’) is that ‘restitution is available so long as mutual restitution of benefits remains possible’.139 This, amongst many other details, will be for the courts to work out as new cases appear.

B. The French Approach In the old Code Civil there were three principal articles which impinged on contracts and illegality.140 The first was Article 6, which stipulates that ‘it is not possible by virtue of a specific agreement, to derogate from laws concerning public policy or public morals’.141 This article applied generally, not just to contracts. The second was Article 1131, which dealt with both the concept and the consequences of a ‘cause illicite’. Article 1131 stipulated, in part, that a contract with a ‘cause illicite’ (‘illegal cause’) had no effect.142 The third article, Article 1133, defined when a ‘cause’ was ‘illicite’. This was said to be when something was prohibited by law or was contrary to good morals. Solène Rowan, in common with many other commentators, has said that the concept of ‘cause’ is ‘one of the most difficult [notions] for a common lawyer to grasp’.143 Rowan has explained that French lawyers traditionally divided the notion of ‘cause’ into ‘objective’ and ‘subjective’ notions. The first, the objective notion, constituted the abstract goal of the contract. It determined whether the contract existed or not. The second, the subjective notion, constituted the subjective reasons why the parties had entered into the contract. Traditionally, an examination of the parties’ intentions was used by courts to determine the legality of the ‘cause’.144 Thus, in a case in 1996, the Cour de Cassation held that a bank loan was illegal and contravened Article 1131 because the loan was to be used, as the parties knew, for the sale of the private data of patients of a dental practice. Such a sale was illegal.145 But, in a later decision of the Cour de Cassation, it was held that a contract could be nullified for illegality even if one of the parties was unaware that the contract was illegal. Former spouses had concluded a contract (after their divorce) whereby the ex-husband would repay a loan made to him by his ex-wife, but he would do so in the form of an

138 Neville v Wilkinson (1782) 1 Bro CC 543, 547–48; 28 ER 1289, 1291 (Lord Thurlow LC); Taylor v Bowers (1876) 1 QBD 291 (CA) 300 (Mellish LJ); possibly Tribe v Tribe [1996] Ch 107 (CA). Lord Sumption added Petherpermal Chetty v Muniandi Servai [1908] UKPC 15, (1908) LR 35 Ind App 98, 103 (Lord Atkinson). 139 Patel (n 4) [253]. 140 Under the old Code, there were four requirements for a contract to be valid: the consent of the parties; their capacity to contract, an objet or content of the contract which was certain and a cause which was legal: old Code Art 1108. 141 My translation. The French text is the same in both the old and new Codes: ‘On ne peut pas déroger, par des c onventions particulières, aux lois qui intéressent l’ordre public et les bonnes mœurs.’ 142 The article also stipulated that with contracts without a cause or based on a false cause, as well as contracts with an illegal cause, had no effect. In French: ‘L’obligation sans cause, ou sur une fausse cause, ou sur une cause illicite, ne peut avoir aucun effet.’ 143 Rowan (n 12) 814, fn 56 and text. See too B Häcker, ‘The Impact of Illegality and Immorality on Contract and Restitution from a Civilian Angle’ in Green and Bogg (n 111) 346, fn 92, noting that the famous French jurist André Rouast is reputed to have said ‘if you have understood la cause, that’s because it has been badly explained to you’. 144 ibid 814–15. See also J Rochfeld, Répertoire de droit civil on ‘Cause’ at ‘Généralités’ para 4. 145 Cass Civ 1ère, 1 octobre 1996, n° 94-18.876.

‘The Obscure, the Implied and the Illegal’ 237 increased alimony payment. This resulted in an illegal tax deduction. The Cour de Cassation held that a contract could be held illegal or immoral and thus could be nullified even if the parties did not appreciate the illegal or immoral nature of the contract. The fact that the object of the contract was to enable the ex-husband to make tax deductions which were themselves illegal was sufficient to make the contract illegal and so result in it being a nullity.146 The concept of ‘cause’, as a requirement for the validity of a contract, has been removed from the new Code.147 The new Article 1128 requires only three things to make a contract valid: first, the consent of the parties; secondly, their capacity to contract; and thirdly, that the contract has ‘content which is lawful and certain’.148 Article 1128 does not state expressly what the consequence will be if the content is not lawful. However, there are four groups of articles in the new Code which impinge on the effect of the content of a contract being unlawful. First, there is Article 6, which, as noted already, remains unchanged. Secondly, there is Article 1162, which repeats the same idea with particular reference to contracts. It states that ‘a contract cannot derogate from public policy either by its stipulations or by its purpose, whether or not this purpose was known to the parties’.149 It has been suggested that the use of the notion of ‘purpose’ of the contract indirectly reintroduces into the Code the concept of ‘cause’.150 Whether it does will have to await more case law. The third and fourth groups of articles of the new Code are Articles 1178–85 and 1352–59. To understand these new provisions, it is necessary to go back to the old Code. Under the old Code, the consequences of a contract being declared ‘illegal’ under Article 1131 were not always the same. In some cases, the consequence was that the contract was declared to be a ‘nullité absolue’. This meant that the contract was nullified not only so far as the parties were concerned, but also with respect to third parties who would otherwise have been affected by it. All involved had to be put in a position as if there had never been any contract. In other cases, the contract was declared a ‘nullité relative’. In general, ‘nullité absolue’ resulted from an action which infringed ‘l’ordre public’, whereas cases of ‘nullité relative’ involved only the infringement of a private interest.151 But even in cases of ‘nullité absolue’, where the parties were put back into the position they would have been in had there been no contract, the court could order that the party that had benefited from the services of the other (albeit under an illegal contract) must recompense the latter for the value of its services. The remedy was non-contractual in form, a kind of restitution to reverse unjust enrichment.152 It has been said that, under the old Code, the general approach of French law was that if the contract was illegal, the court would attempt to put the parties back in the

146 Cass Civ 1ère, 7 octobre 1998, n° 96-14.359. 147 As Professor Philippe Simler put it at para 28 of his commentary on the reformed Code, ‘Réforme du droit des obligations’, published 1 October 2018: ‘le concept de cause, qui se trouve banni du droit français du contrat’ (in English: ‘the concept of cause thus finds itself outlawed from the French law of contract’). There ensued a great debate between those in favour and those against the old concept on whether there had really been a banishment and if so whether that was a good thing. 148 The third requirement is stated in French as ‘un contenu licite et certain’. 149 In French: ‘Le contrat ne peut déroger à l’ordre public ni par ses stipulations; ni par son but, que ce dernier ait été connu ou non par toutes les parties.’ 150 See, eg Simler (n 147) para 38. 151 ibid para 49. 152 Cass Com, 26 juin 2019, n° 18-13689. This case was dealt with under the old Code. A company engaged another to give it legal advice about getting a tax refund, but the advising company was not authorised to give such advice. The client obtained two tax refunds, but refused to pay the fees of the advisor, saying that the contract was illegal and so Art 1131 applied. The Cour de Cassation upheld that conclusion, but, contrary to the Cour d’Appel de Paris, declared that the advisors were entitled to some remuneration, albeit non-contractual.

238 Richard Aikens osition as if no contract had been made. This was so even if the contract was judged to be not p only illegal, but also ‘immoral’.153 The new Code has a section on the effect of ‘nullity’ of a contract comprising Articles 1178–85.154 A contract which does not satisfy the conditions required for its validity is a nullity. It is deemed never to have existed. Acts of performance which have been carried out will give rise to restitution in accordance with Articles 1352–59 of the new Code.155 The distinction between ‘nullité absolue’ and ‘nullité relative’ is maintained and defined in the new Code.156 The general rule remains that if a contract is ‘nul’ because of illegality, then the sanction is ‘nullity’, or ‘nullité’. In consequence, the parties must be put back into the position they would have been in had the contract had never existed. However, whether this is done may depend on whether the contract is a nullity because it is simply ‘illegal’ or whether, in addition, it is also ‘immoral’.157 Two recent decisions of the Cour de Cassation, albeit decided under the old Code, may illustrate this point. In the first,158 one party, B, had given the second party, X, an exclusive option to buy a house (via an estate agent) for €4.1 million. There was a clause in the contract whereby if the sale did not take place, the second party would pay the first a sum equal to 10 per cent of the purchase price.159 X said that she did not at that moment have funds to lodge the indemnité d’immobilisation with the notaire. The estate agent therefore agreed to lend €205,000 to X and it paid that sum to the notaire, on X’s behalf. However, the estate agent did not have any written authorisation from X to pay any sums to the notaire on her behalf.160 The option to purchase was then assigned to a third party, Z. The option was not exercised. The notaire paid the sum of €205,000 over to B as part of the indemnité d’immobilisation. The estate agent sought to recoup the €205,000 from X and Z. The Cour d’Appel of Paris rejected the claim, on the ground that the loan of the money to X by the estate agent was illegal. The Cour de Cassation overturned that decision. It held that the estate agent’s action of lending money to X was itself not illegal. However, its remittance of funds to the notaire was an illegal act because it was made without the written authorisation of X, as required by law. Nevertheless, despite this illegality, the action was not of an ‘immoral character’. Therefore, the estate agent was entitled to restitution of the money lent to X and paid to the notaire, by virtue of the provisions of Article 1902 of the old Code.161 The decision of the Cour de Cassation does not (and probably cannot, given Article 5 of the Code Civil)162 set out any test for when a contract will be both ‘illegal’ and ‘immoral’. To use an English judicial phrase, the distinctions must be

153 N Enonchong, ‘Effects of Illegality: A Comparative Study in French and English Law’ (1995) 44 ICLQ 196. 154 In Book III Title III, ch 2, s IV. 155 Art 1178. It also provides that ‘nullity must be declared by a court, unless the parties establish it by mutual consent’. The second possibility is a novelty, whose utility and effect are debatable. 156 Arts 1179–81. 157 In older cases under the old Code, the court would invoke the Latin maxim nemo auditor propriam turpitudinem allegans (no one can be heard to invoke his own immorality) (shades of Holman and Tinsley). The extent to which this remained the law under the old Code after the 19th century has been doubted by Enonchong (n 153). 158 Cass Civ 1er, 26 septembre 2018, n° 16-25.184. 159 In French law, this 10% sum is called an indemnité d’immobilisation. It has been characterised by the Cour de Cassation as the price of the exclusive right to exercise the option to buy. 160 Required under Art 6 of Law No 70-9 of 2 January 1970 and regulations made thereunder. 161 In a characteristically laconic judgment of less than a page, the court simply states that the action was illegal and not ‘immoral’, and so the lower court had misapplied Art 1902 of the old Code, which stipulates that a borrower is obliged to repay whatever has been lent, in the same quantity and quality, at the time agreed. 162 Art 5 forbids judges to pronounce general principles applicable to other cases.

‘The Obscure, the Implied and the Illegal’ 239 decided on a ‘case by case basis’. It could, however, be argued that the ‘nemo auditor’ principle is out of favour with courts, at least for the present. In another case,163 an osteopath had ordered some publicity space from a specialist company in order to publicise her practice. The company began performance of the contract, but the osteopath then cancelled it. She did not pay the company for the work it had done and the company sued for payments due. The judge rejected the claim. The Cour de Cassation upheld that decision, although on a different basis of law. It reasoned that, because the professional code of osteopaths forbade direct or indirect publicity between a professional and the patient, the contract between the parties had an illegal object and was therefore a nullity.164 Thus, the Cour de Cassation held that the breach of a professional code, as opposed to a law, can be enough to make a contract illegal and so be nullified. This appears to elevate the breach of a professional code to an act which is contrary to ‘l’ordre public’. It is arguable that the same result would come about under Article 1162 of the new Code. The purpose of the contract could be said to be to do something which would ‘derogate from public policy’ because it had the effect of breaching the professional code of conduct. A contract which derogates from public policy is unlawful and so is not valid (Article 1128), and is therefore a nullity (Article 1178). The new Code creates a framework for the legal analysis of illegality and its consequences that is, broadly speaking, clear. If the ‘content’ of a contract is unlawful, then there is no valid contract: Article 1128. A contract also cannot derogate from ‘public policy’ (ordre public) either by its terms or by its purpose: Article 1162 and Article 6. If it does, the contract will be a nullity: Article 1178. The parties will still often be restored to the position they were in before the ‘illegal contract’ was ‘made’, but that will not be permitted if the terms or purpose are both ‘illegal’ and ‘immoral’. Yet there is clearly scope for judges to decide three issues on a case-by-case basis: first, whether the terms or purpose of a contract are contrary to ‘public policy’; secondly, whether the content is ‘unlawful’, which might mean contrary to some law or, seemingly, contrary to something like a professional code; and thirdly, even if a contract is declared ‘illegal’, that does not automatically stop restitution – it depends if the contract or its purpose was both ‘illegal’ and ‘immoral’.

C. Comparison of the Two Approaches Both systems of law start from the public policy proposition that if there has been illegality in connection with a contract, then that can have consequences. In English law, Laboratoires Servier has, broadly, settled the principles according to which a contract will be treated as ‘illegal’. But the effect of Patel is that (even if there were some before) there are no fixed rules as to what the consequences of this illegality will be. English law’s development in this area will depend on how judges work out the ‘principles’ and ‘factors’ identified by the Supreme Court majority, as applied to different facts.165 The result in any given case will lie somewhere between two extremes. A court

163 Cass Civ 1er, 6 fév 2019, n° 17-20463; this case was also decided under the old Code. 164 The single phrase used to explain the decision is ‘qu’un tel contrat est nul en raison du caractère illicite de son objet’ (in English: ‘such a contract is a nullity by reason of the illegal nature of its subject matter’. 165 See, eg Singularis Holdings Ltd (in liq) v Daiwa Capital Markets Europe Ltd [2018] EWCA Civ 84, [2018] 1 WLR 2777 [65]–[67], where the Court of Appeal considered the test for when an appellate court could interfere with the judge’s assessment of the ‘factors’ in deciding whether the illegality defence should prevent any recovery; Gujra v Roath [2018] EWHC 854 (QB), [2018] 1 WLR 3208, where a claimant who had agreed with the defendants to set fire to their cars so they could recover from insurers had his claim for damages for malicious prosecution and breach of duty of care struck out, on the grounds that both the nature of the act and the proximity of it to the claim meant that the illegality defence must succeed.

240 Richard Aikens could conclude that the nature and effect of the ‘illegality’ is such that the court will give no aid to either party, and so the losses must lie where they fall, as when the ‘old rule’ was strictly applied. But in other cases, presumably, the court will say the ‘illegality’ was of little importance so that the contract can be enforced; or at least a right to damages for breach will be. In between, a court might find the contract to be ‘illegal’ but will, if possible, put the parties into the same position as if the contract had never been made. English law in this area may have been revolutionised and made more flexible, but it has also been made less certain. With French law, the terms of the new Code are clear and certain, but, again, the way they are to be worked out in different factual situations may be difficult to predict. Judges may have different views on whether a particular contravention is something that affects ‘public policy’ or makes a contract unlawful in its content, or whether an infraction is ‘immoral’ as well as ‘illegal’. This last element may be particularly hard to predict. It is thus difficult to say whether the results of cases would be different under the two systems. For example, the contract in Patel would seem to be illegal in its content under Article 1128 of the Code. It would also appear, by its ‘purpose’ if not its stipulations, to be contrary to ‘public policy’, whether or not either Mirza or Patel knew that was so. Accordingly, it would also be contrary to Article 1162. The key question might be whether Mirza and Patel’s ‘contract’ was not only ‘illegal’ but also ‘immoral’, with the result that Mirza could not recover his £620,000 in a French court. The effect of the majority judgment in Patel may be that ‘immorality’ in the form of ‘value judgments’ is something that now has to be considered under English law, just as it does in French law. A meeting somewhere in mid-manche?

V. Conclusions The judicial processes of interpretation of a contract and, I would say, of determining whether any terms are to be implied in it ‘in fact’ are both means to an end. The ‘end’ is to decide what obligations the parties owe to one another and then to judge whether there has been a failure to fulfil one or other obligation. The series of Supreme Court decisions addressing the interpretation of written commercial contracts has effectively created an English law code on the interpretation of contracts. The summary of Popplewell J in the Lukoil case demonstrates this well. French judges have more tools available to them to discern the ‘intention of the parties’. But I doubt whether, ultimately, the results of the exercise of interpretation would differ substantially. On implied terms, I would say that French judges have more scope to ‘fill the gaps’ than their English counterparts. This flexibility must owe a lot to the requirement, by Article 1104 of the Code, that contracts must be ‘formed’ in good faith. The role of ‘good faith’ in the negotiation and formation of contracts in English law is, as yet, very uncertain.166 Whether its development is regarded as an opportunity or a danger is debatable. But if English law were to create a principle that contracts must be formed in good faith, this would be likely to lead to the ‘necessity’ rule for implied terms being weakened in favour of what is ‘reasonable’ in the light of the obligation of good faith. 166 The excitement amongst some commentators following the seminal judgment of Leggatt J (as he then was) in Yam Seng Pte Ltd v International Trade Corp [2013] EWHC 111 (QB), [2013] 1 All ER (Comm) 1321, hailing a new dawn for a general principle of good faith in English contract law, has had to be tempered. See, eg the comments in MSC Mediterranean Shipping Co SA v Cottonex Anstalt [2016] EWCA Civ 789, [2017] 1 All ER (Comm) 483 [45] (Moore-Bick LJ).

‘The Obscure, the Implied and the Illegal’ 241 As for ‘illegality’ in relation to contracts, the public policy objective of both systems is the same: to discourage parties from entering into contracts whose object or effect is contrary to the interests of the state (as opposed to private interests). French law seems more settled than English law after Patel. Yet there is a clear common element to both approaches. In French law, the consequences of ‘illegality’ depend on a value judgment of whether the acts that are ‘illegal’ are also ‘immoral’. In English law, the consequences of ‘illegality’ depend on an evaluation of the acts and the contract, using the ‘principles’ and ‘factors’ set out in Lord Toulson’s judgment in Patel. Both methods give the judge much flexibility in assessment. Lord Goff, one of the finest judges and legal scholars of the twentieth century, once said that in his view the major differences between the common law and civil law are to be found ‘rather in the form than in the substance of our law’.167 That assessment certainly is right when the approaches of English and French law to common concepts of interpretation, implied terms and illegality are compared. Lawyers and judges in different jurisdictions can benefit from a comparison of the concepts, the substance, the form and the procedure of other countries’ laws. For, in today’s world, no system of law is an island, entire of itself.

167 Lord

Goff, ‘The Future of the Common Law’ (1997) 46 ICLQ 745.

242

14 The Interpretation of Written Contracts JAMES EDELMAN

‘Interpretation’ is used in different ways. Sometimes it is used, interchangeably with construction, to describe at least two interrelated phenomena: the meaning of words and sentences, and their legal effect when applied to facts.1 Although in straightforward cases the application of meaning to facts will be simple, this will not always be the case. On some occasions, ‘interpretation’ is used to describe only the latter of those phenomena.2 More commonly today, and in the approach taken in this chapter, it is used to describe only the former, namely the meaning of words in legal instruments. This chapter explores the interpretation of written contracts. It begins by illustrating the history of interpretation, which Wigmore described as ‘the history of a progress from a stiff and superstitious formalism to a flexible rationalism’.3 A remnant of that history is a debate about the role of ambiguity as a gateway to the admissibility of extrinsic evidence of context. The chapter then turns to the underlying principle without which that issue cannot be resolved, namely the role of the ‘intention of the parties’, and considers two further modern debates that require consideration of that underlying principle: implication of terms and rectification.

I. Interpretation as Initially a Type of Translation In ordinary discourse, such as conversation, interpretation involves an attempt to understand the intended meaning of the speaker. This involves a flexible rationalism that depends upon both words and context. When parties seek to bind themselves to written contracts they use the same essential techniques of communication as ordinary discourse. The core principles of interpretation of their meaning ought, therefore, not to be essentially different. But, historically, the exercise of interpreting the words of a written contract was often different. In many cases, the interpretation of contractual words was primarily a matter of law, in an exercise akin to translating the words into a different language.

1 Chatenay v Brazilian Submarine Telegraph Co Ltd [1891] 1 QB 79 (CA) 85. 2 Life Insurance Co of Australia Ltd v Phillips [1925] HCA 18, (1925) 36 CLR 60, 78. 3 JH Wigmore, A Treatise on the System of Evidence in Trials at Common Law (Boston, Little, Brown, & Co, 1905) vol 4, 3476, §2462.

244 James Edelman A thought experiment by John Searle illustrates the difference between interpretation and what might be described as translation.4 Searle supposed that he was locked in a room with a batch of paper with Chinese writing on it. Pieces of paper were passed into the room with different Chinese writing on them which, unknown to him, contained questions. Although Searle did not know a single Chinese character, he imagined that he possessed an Englishlanguage set of rules (a ‘program’) that he could manually follow, inputting the characters he received and following instructions about the paper to pass back through the door containing the answers. He imagined that he could eventually follow the program so accurately that his answers to the questions were indistinguishable from those of a native Chinese speaker. Searle was engaging in translation, but not in interpretation. He had no ability to ask himself what the person writing the message intended and therefore no concern for the context in which the words were used. The historical approach to interpretation of many words of written contracts attempted to confine interpretation, as far as possible, to a type of legal translation. Conveyancers knew that many words had defined legal meanings that applied irrespective of context and independently of any meaning that those words might reasonably have been understood to have by the parties. Legal rules would define the meaning of those words. The best conveyancers and drafters attempted to confine themselves to those words. There would be considerable certainty of application of those words because their meaning would not be determined by context or pragmatics. As an exercise akin to translation, the meaning of these words was a matter of law and a matter for lawyers. The legal monopoly on interpretation was justified by pointing to an asserted need for certainty and a need to preserve the legal rules. In 1555, in Throckmerton v Tracy,5 Brook CJ said that ‘if a man should bend the law to the intent of the party, rather than the intent of the party to the law, this would be the way to introduce barbarousness and ignorance, and to destroy all learning and diligence’. In 1702, Holt CJ said that ‘if we once travel into the affairs of the testator and leave the will we shall not know the mind of the testator by his words, but by his circumstances; so that if you go to a lawyer he shall not know how to expound it’.6 In 1756, Gilbert wrote that ‘the Sense and Signification of the Words must be expounded by the Law’ and determined by ‘the Rules of Law’ otherwise ‘no Man could be certain of any Property’.7 This was echoed in 1821 by Lord Redesdale, who said that, if a different approach were taken, ‘all property must be in hazard’.8

II. Relaxation of the Translation Approach Subject to an Ambiguity Constraint The translation type approach to interpretation was not, and could never have been, universal. Many words had no defined legal meaning and therefore their meaning had to be their ordinary,

4 JR Searle, ‘Minds, Brains and Programs’ (1980) 3 The Behavioral and Brain Sciences 417, 417–24. 5 Throckmerton v Tracy (1555) 1 Plow 145, 162; 75 ER 222, 251. 6 Cole v Rawlinson (1702) 1 Salk 234, 235; 91 ER 207, 208. 7 G Gilbert, The Law of Evidence (London, H Lintot, 1756) 81. See also Countess of Rutland’s Case (1604) 5 Co Rep 25b, 26a–b; 77 ER 89, 90; WM Best, A Treatise on the Principles of Evidence and Practice as to Proofs in Courts of Common Law (London, S Sweet, 1849) 246, § 204; R Cross, Evidence (London, Butterworth & Co, 1958) 495. 8 Smith v Doe d Jersey (1821) 2 Brod & B 473, 612; 129 ER 1048, 1102.

The Interpretation of Written Contracts 245 non-legal meaning. But there might be several options for that meaning. Where this was so, there needed to be a technique for identifying the ordinary meaning that would be attributed to those words. In these cases of ‘ambiguity’, the technique was to identify the ordinary meaning by reference to what would reasonably be understood to have been intended. That reasonable understanding required reference to be had to context which, in turn, included circumstances extrinsic to the contract. In the sixteenth century, Bacon explained that ambiguities could be either latent or patent.9 An example of a patent ambiguity is a transfer ‘to James and Paul and heirs’ without clarity about whose heirs are being described. An example of a latent ambiguity is a transfer to ‘my aunt Molly’, where the transferor has two aunts called Molly. Unfortunately, as Phipson observed, a misunderstanding of Bacon’s discussion led to extrinsic evidence being seen to be admissible for latent, but (subject to exceptions) not for patent, ambiguities.10 The exceptions included partial blanks in the written instrument, a legatee being referred to by a term of endearment, and an instrument beginning ‘I, A’ and signed ‘B’.11 Despite this misunderstanding, both Phillipps and Thayer observed that patent ambiguity should not, and often in practice was not, treated any differently from a latent ambiguity.12 Nevertheless, there remained the difficult question of when words would be ambiguous so as to permit the reception of extrinsic evidence. A powerful attempt at definitive explication came in Shore v Wilson,13 where the House of Lords received the advice of the common law judges in 1842 concerning the meaning of ‘godly preachers of Christ’s holy Gospel’ where a woman who executed a deed containing those words was a member of a religious group that used them in a particular way. In a famous passage, the Lord Chief Justice, Tindal CJ, advised that words should be given their common meaning without resort to evidence of context outside the agreement other than in various categories of ambiguity: The general rule I take to be, that where the words of any written instrument are free from ambiguity in themselves, and where external circumstances do not create any doubt or difficulty as to the proper application of those words to claimants under the instrument, or the subject-matter to which the instrument relates, such instrument is always to be construed according to the strict, plain, common meaning of the words themselves; and that in such case evidence dehors the instrument, for the purpose of explaining it according to the surmised or alleged intention of the parties to the instrument, is utterly inadmissible. If it were otherwise, no lawyer would be safe in advising upon the construction of a written instrument, nor any party in taking under it … The true interpretation, however, of every instrument being manifestly that which will make the instrument speak the intention of the party at the time it was made, it has always been considered as an exception, or perhaps, to speak more precisely, not so much an exception from, as a corollary to, the general rule above stated, that where any doubt arises upon the true sense and meaning of the words themselves, or any difficulty as to their application under the surrounding circumstances, the sense and meaning of the language may be investigated and ascertained by evidence dehors the instrument

9 F Bacon, A Collection of Some Principal Rules and Maximes of the Common Lawes of England (London, J More, 1639) 82–84. 10 SL Phipson, The Law of Evidence, 5th edn (London, Stevens & Haynes, 1911) 579. See, eg Colpoys v Colpoys (1822) Jac 451, 463; 37 ER 921, 925; Doe d Gord v Needs (1836) 2 M & W 129, 139–40; 150 ER 698, 702–03. 11 Phipson (n 10) 579. See, eg Doe d Gord (n 10) 139–40; 702–03. 12 SM Phillipps and TJ March, A Treatise on the Law of Evidence, 10th edn (London, William Benning & Co, 1852) vol 2, 389–90; JB Thayer, A Preliminary Treatise on Evidence at the Common Law (Boston, Little, Brown, & Co, 1898) 424–25. 13 Shore v Wilson (1842) 9 Cl & Fin 355, 8 ER 450.

246 James Edelman itself; for both reason and common sense agree that by no other means can the language of the instrument be made to speak the real mind of the party. Such investigation does of necessity take place in the interpretation of instruments written in a foreign language; in the case of ancient instruments, where, by the lapse of time and change of manners, the words have acquired in the present age a different meaning from that which they bore when originally employed; in cases where terms of art or science occur; in mercantile contracts, which in many instances use a peculiar language employed by those only who are conversant in trade and commerce; and in other instances in which the words, besides their general common meaning, have acquired, by custom or otherwise, a well-known peculiar idiomatic meaning in the particular country in which the party using them was dwelling, or in the particular society of which he formed a member, and in which he passed his life. … I conceive the exception to be strictly limited to cases of the description above given, and to evidence of the nature above detailed.14

In this exposition, Tindal CJ made two important points which were in tension with each other. One point was that the meaning of the contractual words was that which gave effect to the reasonably understood intention of the parties at the time the agreement was made. Ordinary techniques of language would lead to the expectation that all context would be relevant to determining that intention. But the second point was that, in order to increase certainty, context outside the agreement was inadmissible to ascertain that reasonably understood intention other than in the various specified categories where ‘doubt arises upon the true sense and meaning of the words themselves, or any difficulty as to their application under the surrounding circumstances’. It was not long before there was pressure upon the categories. Eventually, the categories became generalised, so that resort to extrinsic evidence began to be seen as legitimate in every case of ambiguity.15 With this process of liberalisation, the general ambiguity threshold itself came under pressure. In 1905, Wigmore described the ambiguity gateway as a relic of ‘the old formalism, namely, the rule that you cannot disturb a plain meaning’.16

III. Relaxation of the Ambiguity Constraint The existence of an ambiguity gateway to extrinsic evidence in interpretation is today subject to serious doubt. In 2002, Lord Steyn said that it is ‘crystal clear that an ambiguity need not be established before the surrounding circumstances may be taken into account’.17 In Australia, it has been held, at least in some intermediate appellate courts, that no ambiguity gateway exists.18 But there remains controversy about whether the ‘true rule’ in Australia is that extrinsic evidence

14 ibid 565–67; 532–33. 15 E Cockle, Leading Cases and Statutes on the Law of Evidence, 2nd edn (London, Sweet & Maxwell, 1911) 300; Great Western Rly and Midland Rly v Bristol Corp (1918) 87 LJ Ch 414 (HL) 418–19, 424–25. 16 Wigmore (n 3) vol 4, 3479, § 2462. 17 R (Westminster City Council) v National Asylum Support Services [2002] UKHL 38, [2002] 1 WLR 2956 [5]; approved in Oceanbulk Shipping & Trading SA v TMT Asia Ltd [2010] UKSC 44, [2011] 1 AC 662 [36]. 18 Franklins Pty Ltd v Metcash Trading Ltd [2009] NSWCA 407, (2009) 76 NSWLR 603 [14]–[18]; Mainteck Services Pty Ltd v Stein Heurtey SA [2014] NSWCA 184, (2014) 89 NSWLR 633 [71]–[86]; Stratton Finance Pty Ltd v Webb [2014] FCAFC 110, (2014) 314 ALR 166 [36]–[41]. Compare Western Export Services Inc v Jireh International Pty Ltd [2011] HCA 45, (2011) 282 ALR 604; McCourt v Cranston [2012] WASCA 60, [2012] ANZ Conv R 12-0006 [23]; Cape Lambert Resources Ltd v MCC Australia Sanjin Mining Pty Ltd [2013] WASCA 66, (2013) 298 ALR 666 [107].

The Interpretation of Written Contracts 247 of surrounding circumstances is only admissible ‘if the language is ambiguous or susceptible of more than one meaning’.19 Those who support the ambiguity gateway might, however, point to its existence in other areas of interpretation of legal instruments. First, in relation to interpretation of wills, in the UK and in every Australian state and territory except South Australia, legislation allows the admission of extrinsic evidence to construe a will where the language makes the will or any part of it meaningless, ambiguous on its face or ambiguous in light of surrounding circumstances,20 or, in the case of the Australian Capital Territory and Victoria, if the will is uncertain.21 Secondly, in relation to interpretation of statutes, in the UK extrinsic material can only be admitted to construe a legislative instrument when it is ambiguous, obscure or leads to an absurdity.22 In Australia, as late as 1977, the High Court held that at common law parliamentary materials were inadmissible as an aid to interpretation,23 although an exception developed where the materials could show the purpose of the statute.24 Only Murphy J dissented from this approach, arguing that regard could generally be had to it in cases of ambiguity; he reasoned that a reasonable reader of a statute should be able to accept unambiguous words without reference to parliamentary materials, but that such reference could be made in cases of ambiguity.25 Ultimately, legislative change in Australia permitted recourse to these materials generally, although in some circumstances there are remnants of the ambiguity gateway first opened by Murphy J.26 One might have thought that the same approach would apply to ascertaining the purpose of a legislative provision since the purpose is not independent of the words used. Curiously, however, neither common law nor legislation requires any ambiguity before reference can be made to extrinsic materials to ascertain the purpose of the legislation.27 Ultimately, in order to decide whether, or the extent to which, the ambiguity gateway should be relaxed, it is necessary to ask what the exercise of interpretation involves and what principles underlie the exercise. Only then would it be possible to consider the extent to which functional concerns of certainty should be permitted to intrude upon those underlying principles. The ambiguity gateway is not the only issue affected by the underlying approach that is taken to interpretation. Others include the following. Is extrinsic evidence admissible for the purpose of interpretation if it was not reasonably available to both parties? Is extrinsic evidence admissible for the purpose of interpretation if it post-dates entry into the contract? What rules should govern the implication of terms? And what rules should govern the rectification of terms?

19 Codelfa Construction Pty Ltd v State Rail Authority of NSW [1982] HCA 24, (1982) 149 CLR 337, 352. See the comprehensive discussion in JD Heydon, Heydon on Contract (Sydney, Lawbook Co, 2019) 329–425. 20 Wills Act 1968 (ACT), s 12B; Wills Act 1970 (WA), s 28A; Succession Act 1981 (Qld), s 33C; Administration of Justice Act 1982 (UK), s 21; Wills Act 1997 (Vic), s 36; Succession Act 2006 (NSW), s 32; Wills Act 2008 (Tas), s 46; Wills Act 2000 (NT), s 31. See also B Häcker, ‘Thy Will Be Done’ (2014) 130 LQR 360, 363. 21 Wills Act 1968 (ACT), s 12B(b)–(c); Wills Act 1997 (Vic), s 36(1)(b)–(c). 22 Pepper (Inspector of Taxes) v Hart [1993] AC 593 (HL) 620, 640. 23 Commissioner for Prices and Consumer Affairs (SA) v Charles Moore (Aust) Ltd [1977] HCA 38, (1977) 139 CLR 449, 461–62, 470, 476–78. See also Victoria v The Commonwealth [1975] HCA 39, (1975) 134 CLR 81, 152. 24 Federal Commissioner of Taxation v Whitfords Beach Pty Ltd [1982] HCA 8, (1982) 150 CLR 355, 373–75; Gerhardy v Brown [1985] HCA 11, (1985) 159 CLR 70, 104, 111; Hoare v The Queen [1989] HCA 33, (1989) 167 CLR 348, 360. 25 Commissioner for Prices and Consumer Affairs (n 23) 480. 26 Acts Interpretation Act 1901 (Cth), ss 15AA–15AB; Acts Interpretation Act 1931 (Tas), ss 8A–8B; Acts Interpretation Act 1954 (Qld), ss 14A–14B; Interpretation of Legislation Act 1984 (Vic), ss 35(a)–35(b); Interpretation Act 1984 (WA), ss 18–19; Interpretation Act 1987 (NSW), ss 33–34; Interpretation Act 1987 (NT), ss 62A–62B; Legislation Act 2001 (ACT), ss 139, 141–43. Compare South Australia: Australian Education Union v Department of Education and Children’s Services [2012] HCA 3, (2012) 248 CLR 1 [33]. 27 CIC Insurance Ltd v Bankstown Football Club Ltd [1997] HCA 2, (1997) 187 CLR 384, 408; Newcastle City Council v GIO General Ltd [1997] HCA 53, (1997) 191 CLR 85, 99, 112–13.

248 James Edelman

IV. ‘The Intention of the Parties’ as the Underlying Interpretative Principle In the process of interpretation, words will sometimes have a range of different potential meanings, including, in some cases, meanings which are not within the range of literal meanings of the words. How, then, should the best meaning be chosen? It is often said that the chosen meaning, in a commercial contract, should be found by considerations of commercial convenience or commercial sense.28 In other contexts, it is said that the chosen meaning, whether construing a statute or a contract, should be determined by reference to common sense.29 But there must be a yardstick against which the choice, including considerations of ‘convenience’ and ‘common sense’, can be made. After all, one party’s commercial convenience or commercial sense may often be the other’s commercial inconvenience and commercial nonsense. The answer, given on so many occasions it does not bear citation, is that the touchstone is the ‘intention of the parties’. In hindsight, this shorthand expression, with its connotations of subjectivity, may have been better avoided in English law. The concern is not with the subjective intention of the parties, or the subjective intention of either of them. The ‘intention of the parties’ is a shorthand description of an objective approach that is concerned with what a reasonable person would understand to have been intended by the words if written by a notional reasonable person in the position of both of the parties. This is usually described as the objective theory of contract. There are some who dispute the objective theory. They believe that the underlying principle in contract law is a concern to enforce subjective intentions. On that view, an objective approach to the ‘intention of the parties’ is very difficult to justify. Why should the law be concerned with the intention of a notional reasonable person in the position of both of the parties when the underlying focus is really on the extent of subjective agreement of the two actual parties? An example

28 Miramar Maritime Corp v Holborn Oil Trading Ltd (The Miramar) [1984] AC 676 (HL) 682; Antaios Compania Naviera SA v Salen Rederierna AB (The Antaios) [1985] AC 191 (HL) 201; Mannai Investment Co Ltd v Eagle Star Life Assurance Co Ltd [1997] AC 749 (HL) 770–71; Society of Lloyd’s v Robinson [1999] 1 WLR 756 (HL) 763; Zhu v Treasurer of New South Wales [2004] HCA 56, (2004) 218 CLR 530 [82]; Barclays Bank plc v HHY Luxembourg SARL [2010] EWCA Civ 1248, [2011] 1 BCLC 336 [25]–[26]; Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900 [21]–[30], [39]–[40]; Electricity Generation Corp v Woodside Energy Ltd [2014] HCA 7, (2014) 251 CLR 640 [35], citing Re Golden Key Ltd [2009] EWCA Civ 636 [28]; Arnold v Britton [2015] UKSC 36, [2015] AC 1619 [15]; Mount Bruce Mining Pty Ltd v Wright Prospecting Pty Ltd [2015] HCA 37, (2015) 256 CLR 104 [51]; Simic v New South Wales Land and Housing Corp [2016] HCA 47, (2016) 260 CLR 85 [78]; Ecosse Property Holdings Pty Ltd v Gee Dee Nominees Pty Ltd [2017] HCA 12, (2017) 261 CLR 544 [17]; Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173 [11]. See also Hydarnes Steamship Co v Indemnity Mutual Marine Assurance Co [1895] 1 QB 500 (CA) 504; McCann v Switzerland Insurance Australia Ltd [2000] HCA 65, (2000) 203 CLR 579 [22]; Fiona Trust & Holding Corp v Privalov [2007] UKHL 40, [2007] Bus LR 1719 [8]; International Air Transport Assoc v Ansett Australia Holdings Ltd [2008] HCA 3, (2008) 234 CLR 151 [8]. cf Mitsubishi Heavy Industries Ltd v Gulf Bank KSC [1997] 1 Lloyd’s Rep 343 (CA) 350; Cargill International SA v Bangladesh Sugar and Food Industries Corp [1998] 1 WLR 461 (CA) 468; Skanska Rashleigh Weatherfoil Ltd v Somerfield Stores Ltd [2006] EWCA Civ 1732 [22]. 29 Barnes v Jarvis [1953] 1 WLR 649 (Div Ct) 652; Richardson v Austin [1911] HCA 28, (1911) 12 CLR 463, 477–78, citing Dyke v Elliott (The Gauntlet) (1874) LR 4 PC 184, 191; Cooper Brookes (Wollongong) Pty Ltd v Federal Commissioner of Taxation [1981] HCA 26, (1981) 147 CLR 297, 320; Rayware Ltd v Transport and General Workers’ Union [1989] 1 WLR 675 (CA) 681–82; Collector of Customs v Agfa-Gevaert Ltd [1996] HCA 36; (1996) 186 CLR 389, 400; Charter Reinsurance Co Ltd v Fagan [1997] AC 313 (HL) 356; Mannai Investment Co (n 28) 782; Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (HL) 912; Cadogan Estates Ltd v McMahon [2001] 1 AC 378 (HL) 388; DC Pearce and RS Geddes, Statutory Interpretation in Australia, 8th edn (Sydney, LexisNexis Butterworths, 2014) para 4.1.

The Interpretation of Written Contracts 249 of a subjective approach to contract law can be seen in French law. Ordonnance nº 2016-131 of 10 February 2016, Article 1188 provides: A contract is to be interpreted according to the common intention of the parties rather than stopping at the literal meaning of its terms. Where this intention cannot be discerned, a contract is to be interpreted in the sense which a reasonable person placed in the same situation would give to it.30

In French law, this reference to the common intention of the parties appears to be a reference to their subjective intentions. Only where their subjective intentions are not clear does the objective rule apply. One driver of modern French law was Pothier’s treatise on the law of obligations.31 His first rule of contractual interpretation was to ‘examine what was the common intention of the contracting parties rather than the grammatical sense of the terms’.32 Pothier’s will theory of contract had powerful supporters. A version of it was adopted by von Savigny, strongly influenced by Kant. But although Pothier’s treatise was translated by Sir William Evans in 1806, the subjective French theory was never accepted by the English. Bentham was highly critical of the English for this reason. He described the refusal to admit oral evidence of intention ‘an enormity, an act of barefaced injustice, unknown everywhere but in English jurisprudence’.33 English and Australian law do not accept the subjective theory of contract. The entrenchment of the objective approach is clearly seen in Australian law. In 1983,34 Mason ACJ, Murphy and Deane JJ said that the objective approach had ‘command of the field’. Since then, the High Court of Australia has reiterated that ‘[t]he legal rights and obligations of the parties turn upon what their words and conduct would be reasonably understood to convey, not upon actual beliefs or intentions’.35 As Gummow and Hayne JJ observed in Byrnes v Kendle,36 ‘the “objective theory” of contract formation … is concerned not with “the real intentions of the parties, but with the outward manifestations of those intentions”’. Similarly, Heydon and Crennan JJ, quoting from Oliver Wendell Holmes, said that ‘parties may be bound by a contract to things which neither of them intended’.37 It is no coincidence that this is the same approach that people use every day to understand ordinary discourse. The objective theory recognises that contracts are an attempt by people to bind themselves and others to obligations using the same techniques of everyday language. Suppose that a parent promises a child that he will attend her swimming competition on the weekend but that the parent arrives in the final 10 minutes, after all the child’s races have been swum. It will not avail the parent to say that he only intended to arrive for some part of the event. The words of the promise meant what the child would reasonably understand them to mean that the parent would arrive in time to watch the child.

30 J Cartwright, B Fauvarque-Cosson and S Whittaker (trans), The Law of Contract, The General Regime of Obligations, and Proof of Obligations: The New Provisions of the Code Civil Created by Ordonnance nº 2016-131 of 10 February 2016, 17. (This officially authorised English translation of the text is published online at www.textes.justice.gouv.fr/art_pix/THELAW-OF-CONTRACT-2-5-16.pdf.). 31 RJ Pothier, Traité des Obligations (Paris, Debure, 1761). 32 RJ Pothier, A Treatise on the Law of Obligations, or Contracts, WD Evans (trans) (London, J Butterworth, 1806) vol 1, para 91. 33 J Bentham, ‘Rationale of Judicial Evidence’ in J Bowring (ed), The Works of Jeremy Bentham (Edinburgh, W Tait, 1843) vol 7, 556–57. See also S Comyn, A Treatise of the Law Relative to Contracts and Agreements Not Under Seal (London, J Butterworth, 1807) vol 2, 533–34. 34 Taylor v Johnson [1983] HCA 5, (1983) 151 CLR 422, 429. 35 Equuscorp Pty Ltd v Glengallan Investments Pty Ltd [2004] HCA 55, (2004) 218 CLR 471 [34]. 36 Byrnes v Kendle [2011] HCA 26, (2011) 243 CLR 253 [59]. 37 ibid [100], quoting OW Holmes, ‘The Path of the Law’ (1897) 10 Harvard Law Review 457, 463.

250 James Edelman As in ordinary discourse, the ‘reasonably understood intention’ approach also has the effect that when we interpret the meaning of a written contract, we are not limited to the range of literal meanings of the words used. The range of meaning of words is important, but it is not conclusive. In Fitzgerald v Masters,38 the concluding clause in a written contract purported to embody a set of conditions for sale ‘so far as they are inconsistent’. The word ‘inconsistent’ might have been capable of a number of different literal meanings, but one literal meaning that it did not have was ‘consistent’. Nevertheless, as Dixon CJ and Fullagar J said, ‘consistent’ was the meaning that the parties ‘must clearly have intended’.39 No reasonable person in the position of the parties would understand the parties to have set out carefully all of the terms of their contract only then to provide for incorporation of other terms only so far as they are inconsistent. Examples of words bearing the opposite of their literal meaning illustrate what Lord Hoffmann said in Chartbrook Ltd v Persimmon Homes Ltd: [T]here is not, so to speak, a limit to the amount of red ink or verbal rearrangement or correction which the court is allowed. All that is required is that it should be clear that something has gone wrong with the language and that it should be clear what a reasonable person would have understood the parties to have meant.40

Although the process of interpretation contains no limit on the departure from the literal meaning of words, the more that an interpretation departs from the range of the literal meanings of the words, the more the departure will need to be justified by the context. One matter of context is if the written contract was drafted by professionals who might be expected to choose their words with care, albeit that this is a context which is an extrinsic circumstance usually considered even if the words are not ambiguous. An example of the role of that context is Wood v Capita Insurance Services Ltd.41 In that case, the UK Supreme Court considered a clause by which a seller agreed to indemnify a buyer against losses ‘following and arising out of claims or complaints registered with the FSA [Financial Services Authority]’. The court held that the indemnity was limited to losses arising out of claims or complaints. It did not extend to losses arising from a compensation scheme for misled customers that was set up after consultation with the FSA. Lord Hodge, giving the judgment of the court, emphasised that the contract was ‘detailed and professionally drafted’,42 and that the parties were ‘commercially sophisticated’.43 His Lordship said: ‘Some agreements may be successfully interpreted principally by textual analysis, for example because of their sophistication and complexity and because they have been negotiated and prepared with the assistance of skilled professionals.’44 It might seem strange that the clause in Wood meant that a scheme set up after an FSA complaint from a single customer would be covered by the indemnity, but that one set up after voluntary reporting or notice from a whistleblower to the FSA would not. This, counsel for the buyer had argued, was especially odd because an interpretation that covered both circumstances was possible with only slight modifications to the grammar of the clause. But one key point of the case is that the further that one departs from the range of literal or interpretative meanings of the words, the more justification is required, particularly in professionally drafted contracts.

38 Fitzgerald 39 ibid

427.

v Masters [1956] HCA 53, (1956) 95 CLR 420.

40 Chartbrook

Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1 AC 1101 [25]. (n 28) [18]. 42 ibid [16]. 43 ibid [28]. 44 ibid [13]. 41 Wood

The Interpretation of Written Contracts 251 As an underlying interpretative principle, the intention of the parties, when understood in this objective manner, provides the means by which these long-standing issues of interpretation can be resolved. In relation to the admissibility of extrinsic evidence as to context, this principle suggests that almost all evidence of context should be admissible in the interpretation of a contract. The qualification ‘almost’ includes that context, and therefore extrinsic evidence, which is not reasonably available at the time of contracting might not be imputed to the reasonable person in the position of both of the parties, just as matters known to one party but not reasonably available to the other would not be imputed to the notional reasonable person.45

V. ‘Intention of the Parties’ and Ambiguity The constraint of an ambiguity gateway cannot be justified by the underlying interpretative principle. However, one justification given for the ambiguity gateway is an assertion that it would lead to greater certainty in the process of interpretation of written contracts. As we have seen, such assertions of a need for certainty were also used to justify an approach akin to translation, in addition to the limited categories of ambiguity approach. But, long after the translation approach was abandoned, certainty retained a grip in relation to extrinsic circumstances. In the mid-twentieth century, Rupert Cross even argued that the need for certainty was a ‘basic reason’ for the existence of a law of interpretation and one that warranted ‘a literal standard of interpretation’.46 Arguments of certainty need, however, to be more nuanced. For instance, if extrinsic materials are admissible for the purpose of ascertaining whether an ambiguity exists, including in cases of latent ambiguity, then how far would certainty be advanced by an ambiguity gateway? Moreover, ‘[t]he term “ambiguity” is itself not inflexible’:47 it is capable of both broader and narrower interpretations. As the most powerful defender of the ambiguity gateway acknowledges, if ‘ambiguity’ is given a broad meaning, the controversy becomes ‘empty and illusory’.48

VI. ‘Intention of the Parties’ and Rectification If the view of those who believe that the underlying principle in contract law is to enforce subjective intention were correct, the law of contract would not be a law of agreement; it would be a law of common subjective intentions. Suppose two parties enter into a contract. The agreement records the term as one year. In prior negotiations, one party had mentioned the possibility of a ten-year term. Both parties subjectively thought that this had been accepted but, in fact, the other party had not replied and the formal agreement had not been amended. There is an agreement in form providing for a term of one year, which both parties had thought that they agreed to be for 10 years, but no agreement had been communicated. Towards the end of the year, one party, who wishes to terminate, notices the written one-year term and refuses to renew. English and Australian law, following the objective approach, would treat the agreement as being one for a year, not 10 years. The objectivity of the underlying principle of interpretation means that an

45 Maggbury

Pty Ltd v Hafele Australia Pty Ltd [2001] HCA 70, (2002) 210 CLR 181, 188. (n 7) 495. 47 Life Insurance Co of Australia (n 2) 78. 48 Heydon (n 19) para 9.680. 46 Cross

252 James Edelman a greement is something separate from the subjective intentions of either or both parties. Hence, in Re Atkinson’s Will Trusts,49 Megarry V-C said that if a testator, when signing a will, said ‘“[w]here the will say X, I intend X to mean Y”, I cannot see that evidence of this statement would be admissible’. One of the most significant challenges for the objective approach to agreement in contract law are the rules of rectification. Although an understanding of rectification depends upon the fundamental principle of the intention of the parties, the role of rectification and its relationship with interpretation is deeply unsettled. In recent years, English judges have been writing prolifically on the subject.50 I put to one side in this chapter the different rules of so-called unilateral mistake rectification. In unilateral mistake rectification, ‘one party to a transaction knows that the instrument contains a mistake in his favour but does nothing to correct it’.51 The unilateral mistake known to the other party may be grounds for the mistaken party to rescind the contract.52 However, in some English cases, it has been held sufficient to rectify the agreement even without any prior agreement or accord between the parties.53 Allowing rectification in these cases, rather than rescission, is an exceptional principle perhaps justifiable only by a principle that the unilateral mistake that is known to the other party removes the objective basis for the other party to retain the benefit of those contractual rights that would otherwise exist without rectification. In Canada, the principle has been confined to cases of ‘fraud or the equivalent of fraud’.54 The more common area of rectification is what is described as common mistake rectification. Those who support the view of contract as a law of common subjective intention argue that it is ‘quite unjust to allow a party to hold the other party to a contract that the evidence irrefutably establishes that neither party intended’.55 The notion that the terms of a contract can be rectified where the parties are both subjectively mistaken about the meaning initially seems to provide substantial support for this view. The references in many cases to common ‘intention’, without clarification that the common intention is to be objectively ascertained as an intention of a person in the position of both contracting parties, provides some support to the view that common mistake rectification is concerned with enforcing common subjective intentions.

49 Re Atkinson’s Will Trusts [1978] 1 WLR 586 (Ch) 590. 50 K Lewison, ‘If It Ain’t Broke Don’t Fix It: Rectification and the Boundaries of Interpretation’ in K Lewison, The Interpretation of Contracts: First Supplement to the Fourth Edition (London, Sweet & Maxwell, 2010) 127; R Buxton, ‘“Construction” and Rectification after Chartbrook’ (2010) 69 CLJ 253; C Nugee, ‘Rectification after Chartbrook v Persimmon: Where Are We Now?’ (2012) 26 Trust Law International 76; P Morgan, ‘Rectification: Is it Broken? Common Mistake After Daventry’ (2013) 21 Restitution Law Review 1; N Patten, ‘Does the Law Need to be Rectified? Chartbrook Revisited’ (Chancery Bar Association Annual Lecture, 29 April 2013), www.chba.org.uk/for-members/library/annuallectures/does-the-law-need-to-be-rectified-chartbrook-revisited; R Toulson, ‘Does Rectification Require Rectifying?’ (TECBAR Annual Lecture, 31 October 2013), www.supremecourt.uk/docs/speech-131031.pdf; T Etherton, ‘Contract Formation and the Fog of Rectification’ (2015) 68 Current Legal Problems 367; L Hoffmann, ‘Rectification and other Mistakes’ (Commercial Bar Lecture, 3 November 2015) https://app.pelorous.com/media_manager/public/260/Lord%20 Hoffmann%20Lecture%203.11.15.pdf. 51 Maralinga Pty Ltd v Major Enterprises Pty Ltd [1973] HCA 23, (1973) 128 CLR 336, 351, citing Whiteley v Delaney [1914] AC 132 (HL); Monaghan CC v Vaughan [1948] IR 306 (HC); George Cohen, Sons & Co Ltd v Docks and Inland Waterways Executive (1950) 84 Ll L Rep 97 (CA). 52 Taylor (n 34) 432. 53 See A Roberts & Co Ltd v Leicestershire CC [1961] Ch 555 (Ch) 570; Riverlate Properties Ltd v Paul [1975] Ch 133 (CA) 145; Thomas Bates and Son Ltd v Wyndham’s (Lingerie) Ltd [1981] 1 WLR 505 (CA) 514–16. 54 Canada (Attorney-General) v Fairmont Hotels Inc [2016] SCC 56, [2016] 2 SCR 720 [15]. 55 D McLauchlan, ‘The Many Versions of Rectification for Common Mistake’ in S Degeling, J Edelman and J Goudkamp (eds), Contract in Commercial Law (Sydney, Lawbook Co, 2016) 203.

The Interpretation of Written Contracts 253 There is a strong argument against understanding intention in this subjective manner, and hence permitting a subjective approach to rectification. Contracts can change the duties that people owe. If the duties of people, inter se, are to change then they should only do so by communication. The power of your unspoken thoughts should not be sufficient to put me under a duty to you that I did not otherwise owe. That is why an intention to create legal relations is not a subjective intention but is only one that a reasonable person would ascertain from the way in which the right-holder has behaved. Hence, in that area, ‘intention’ describes ‘what it is that would objectively be conveyed by what was said or done’.56 The approach to the existence of the agreement should also apply to the contents of the agreement. There is a view of common mistake rectification that is wholly consistent with the objective theory of contract. On this view, the doctrine applies where the meaning of the words used to a reasonable person in the position of the parties is different from the meaning which existed in a different consensus between the parties. The extrinsic evidence of the latter is not sufficient for a reasonable person in the position of the parties to conclude that the words of the written agreement have a different meaning. Rectification has the effect that the words of the written contract have their meaning, as interpreted, but there is an equitable power, subject to the usual bars to relief, for either party to apply for the court to amend the words of the document so that they will have a different, but previously agreed, meaning. The court’s power to amend the words of the document, and therefore to give them a different meaning, arises to make the written document conform with the earlier agreement. On this traditional approach, rectification is simply equity’s response to a common mistake by correcting the words, and therefore the meaning, of the later written agreement to make it conform with the earlier agreement. This is so even if the earlier agreement in its entirety was not itself a contract that would have been enforced, for instance, because it was expressed to be subject to contract.57 But rectification might loosely be seen as specific performance of an implied undertaking in the earlier agreement by the parties that any later written version of the agreement would take a form that reflected the meaning of their earlier agreement. This view of common mistake rectification was clearly explained by Lord Hoffmann in a judgment with which the other Justices agreed in the Hong Kong Final Court of Appeal in Kowloon Development Finance Ltd v Pendex Industries Ltd.58 Lord Hoffmann explained that common mistake rectification arises where the parties have a common mistake ‘about whether a written document correctly reflects what the parties had, on an objective assessment, agreed it should contain’.59 Referring to his judgment in the House of Lords in Chartbrook Ltd v Persimmon Homes Ltd,60 Lord Hoffmann said: it is true to say that the concept of rectification for common mistake involves carrying into effect what the parties appear to have actually agreed that the document should say. And in deciding what the

56 Ermogenous v Greek Orthodox Community of SA Inc [2002] HCA 8, (2002) 209 CLR 95 [25]. See also, generally, Masters v Cameron [1954] HCA 72, (1954) 91 CLR 353, 362; Gissing v Gissing [1971] AC 886 (HL) 906; Australian Broadcasting Corp v XIVth Commonwealth Games Ltd (1988) 18 NSWLR 540, 549; Deutsche Genossenschaftsbank v Burnhope [1995] 1 WLR 1580 (HL) 1587; Pacific Carriers Ltd v BNP Paribas [2004] HCA 35, (2004) 218 CLR 451 [22]; Toll (FGCT) Pty Ltd v Alphapharm Pty Ltd [2004] HCA 52, (2004) 219 CLR 165 [38], [40]; Equuscorp (n 35) [34]; Sirius International Insurance Co (Publ) v FAI General Insurance Ltd [2004] UKHL 54, [2004] 1 WLR 3251 [18]; Automotive, Food, Metals, Engineering, Printing and Kindred Industries Union v ALS Industrial Australia Pty Ltd [2015] FCAFC 123, (2015) 235 FCR 305 [103]. 57 Joscelyne v Nissen [1970] 2 QB 86 (CA). See also Simic (n 28) [117]. 58 Kowloon Development Finance Ltd v Pendex Industries Ltd [2013] HKCFA 35, (2013) 16 HKCFAR 336. 59 ibid [19]. 60 Chartbrook (n 40).

254 James Edelman parties have agreed, the common law adopts its usual objective stance, looking at what a reasonable observer would have understood the parties to mean and not concerning itself with their uncommunicated states of mind.61

In Chartbrook,62 Lord Hoffmann approved the approach of Denning LJ in Frederick E Rose (London) Ltd v William H Pim Jnr & Co Ltd.63 That case involved the sale of Moroccan horsebeans. Both the buyer and the seller had mistakenly believed that all horsebeans were feveroles. They had discussed this. But the Court of Appeal did not accept that a reasonable person in the position of the parties would have understood the word ‘horsebeans’ to mean ‘feveroles’. The buyer specifically needed feveroles for a resale. The horsebeans that were supplied were not feveroles. The buyer sought to have the contract rectified. He failed. Denning LJ said: Rectification is concerned with contracts and documents, not with intentions. In order to get rectification it is necessary to show that the parties were in complete agreement on the terms of their contract, but by an error wrote them down wrongly; and in this regard, in order to ascertain the terms of their contract, you do not look into the inner minds of the parties – into their intentions – any more than you do in the formation of any other contract. You look at their outward acts, that is, at what they said or wrote to one another in coming to their agreement, and then compare it with the document which they have signed. If you can predicate with certainty what their contract was, and that it is, by a common mistake, wrongly expressed in the document, then you rectify the document; but nothing less will suffice.64

The most recent consideration of rectification in England came in the comprehensive judgment of the Court of Appeal in FSHC Group Holdings Ltd v GLAS Trust Corp Ltd.65 That case was a useful test of the objective approach to common mistake rectification. Two deeds of security were subjectively intended by both parties to include only the provision of an additional security which had previously been omitted from a complex transaction. However, on the remarkable facts of the case, including the failure of the solicitors involved to review the terms of the deeds, the terms of the agreement included additional onerous obligations. The trial judge found, and the Court of Appeal accepted, that there had been no prior objective consensus between the parties that the deeds of security would be limited only to the provision of additional security. After a cogent and exhaustive treatment of the difficult history of rectification in English law, the Court of Appeal concluded that, contrary to Lord Hoffmann’s approach in Chartbrook and the approach of Denning LJ in Frederick E Rose (London) Ltd, it was not necessary to show a prior objective accord between the parties in order for common mistake rectification of a written agreement.66 However, earlier in the joint reasons, the Court of Appeal affirmed as ‘sound in principle’ the requirement of ‘an outwardly expressed accord of minds’: a test ‘based on mutual assent which the parties have manifested to each other and not on uncommunicated intentions which happen, without the parties knowing it, to coincide’.67 Ultimately, the different approaches led to no different result because there was both a common subjective intention as well as a prior objective accord that the written agreement would contain only the additional security.68

61 Kowloon

Development Finance (n 58) [19]. (n 40). 63 Frederick E Rose (London) Ltd v William H Pim Jnr & Co Ltd [1953] 2 QB 450 (CA) 461. 64 ibid 461. 65 FSHC Group Holdings Ltd v GLAS Trust Corp Ltd [2019] EWCA Civ 1361. 66 ibid [155]–[156]. 67 ibid [76]–[77]. 68 ibid [183]. 62 Chartbrook

The Interpretation of Written Contracts 255 One decision that the Court of Appeal cited in support of the subjective approach was the decision of the High Court of Australia in Simic v New South Wales Land and Housing Corp.69 But that case does not obviously depart from the objective approach requiring a prior accord. In the joint judgment of Gageler, Nettle and Gordon JJ, their Honours said that the purpose of rectification was to make the ‘written instrument conform to the “true agreement” of the parties’.70 They said that the true agreement between the parties required a common intention. Although their Honours said that there was no requirement for communication of the common intention by an express statement between the parties, they also said that there was a requirement that the common intention be viewed objectively from the parties’ words or actions.71 The need for a prior objective accord was also emphasised in the Supreme Court of Canada in an impeccable passage by Brown J, giving reasons with which McLachlin CJ and Cromwell, Moldaver, Karakatsanis, Wagner and Gascon JJ concurred. Importantly, his Honour spoke of the parties’ true agreement, not of the parties’ true subjective intentions:72 If by mistake a legal instrument does not accord with the true agreement it was intended to record – because a term has been omitted, an unwanted term included, or a term incorrectly expresses the parties’ agreement – a court may exercise its equitable jurisdiction to rectify the instrument so as to make it accord with the parties’ true agreement. Alternatively put, rectification allows a court to achieve correspondence between the parties’ agreement and the substance of a legal instrument intended to record that agreement, when there is a discrepancy between the two.

There remains the question of why a common mistake, continuing to the date of the written contract,73 is needed in order to rectify. One answer, given by Professor Stevens, is that the common mistake permits partial rescission of an express or implied term of the contract that the written terms of the agreement are the entirety of the contractual obligations.74 As Longmore LJ said in Barclays Bank plc v Unicredit Bank AG,75 an entire agreement clause ‘is intended to exclude any evidence or argument to the effect that the terms of the contract are to include any mutual understanding that is not recorded in the contract’. A common mistake of the parties effectively operates to permit recourse to evidence of a term beyond the ‘entire written agreement’ by permitting rescission of an express or implied entire agreement clause. The need for a common mistake should not apply to rectification of a unilateral instrument, such as a will. Unlike a contract, implementing the intention of a testator does not impose duties upon other parties. It is a unilateral act that confers rights upon them. Hence, unlike a contract, where some prior consensus is required to allow rectification, a will should be capable of rectification without any prior act consensus. Although it had been assumed for many years that equity does not have the power to rectify a will,76 this approach might have been based only upon a concern not to undermine the legislative formality in section 9 of the Wills Act 1837. Independently of these statutory formality concerns, there may have been a power for equity to rectify

69 ibid [169]. See Simic (n 28) [103]. 70 ibid [103]. 71 ibid [104]. See also JIS (1974) Ltd v MCP Investment Nominees I Ltd [2003] EWCA Civ 721 [33]–[34]; FSHC Group Holdings (n 65). 72 Canada (Attorney-General) (n 54) [12]. 73 Fowler v Fowler (1859) 4 De G & J 250, 265; 45 ER 97, 103. 74 R Stevens, ‘The Meaning of Words and the Intentions of Persons’ in Degeling et al (n 55) 182. 75 Barclays Bank plc v Unicredit Bank AG [2014] EWCA Civ 302, [2014] 2 All ER (Comm) 115 [27]. See also NHS Commissioning Board v Vasant [2019] EWCA Civ 1245 [47]. 76 Marley v Rawlings [2014] UKSC 2, [2015] AC 129 [27]–[28].

256 James Edelman a will to make it conform with the unilateral objective intention of the testator before death, as revealed by extrinsic materials.77

VII. ‘Intention of the Parties’ and Implication of Terms Another issue to which the underlying principle of the intention of the parties is relevant is in the implication of terms. An unexpressed term that is sought to be drawn from the words of a written contract must be based upon the meaning which is reasonably thought to be intended by the parties. In Attorney-General of Belize v Belize Telecom Ltd,78 Lord Hoffmann (delivering the judgment of the Board comprising Lord Rodger, Lady Hale, Lord Carswell and Lord Brown) said that implication is an exercise of interpretation. He gave several reasons in logic and authority for this. First, as a matter of logic, implication must be an exercise in interpretation ‘since a court has no power to alter what the instrument means’. Secondly, as a matter of authority, in Trollope & Colls Ltd v North West Metropolitan Regional Hospital Board,79 Lord Pearson, with whom Lord Guest and Lord Diplock agreed, had said that ‘[a]n unexpressed term can be implied if and only if the court finds that the parties must have intended that term to form part of their contract’. Lord Hoffmann therefore concluded that in every case in which it is said that some provision ought to be implied in an instrument, the question for the court is whether such a provision would spell out in express words what the instrument, read against the relevant background, would reasonably be understood to mean.80

He emphasised that the five ‘criteria’ enunciated by the Privy Council in BP Refinery (Westernport) Pty Ltd v Shire of Hastings81 should not detract from the ultimate question, which is what the instrument is reasonably understood to mean.82 In the High Court of Australia in 2014, in Commonwealth Bank of Australia v Barker,83 French CJ, Bell and Keane JJ referred to Lord Hoffmann’s approach in Belize and reiterated the words of Mason J84 (with whom Stephen J85 and Wilson J86 had agreed), who had said that implication is not ‘an orthodox exercise in the interpretation of the language of a contract, that is, assigning a meaning to a particular provision’; it is an ‘exercise in interpretation, though not an orthodox instance’. More recently, doubt has been cast upon this approach in England. In 2015, in Marks and Spencer plc v BNP Paribas Securities Services Trust Company (Jersey) Ltd,87 the Supreme Court 77 See the comprehensive discussion in B Häcker, ‘What’s in a Will – Examining the Modern Approach Towards the Interpretation of Testamentary Instruments’ in B Häcker and C Mitchell (eds), Current Issues in Succession Law (Oxford, Hart Publishing, 2016) 158–61. 78 A-G of Belize v Belize Telecom Ltd [2009] UKPC 10, [2009] 1 WLR 1988 [19]. 79 Trollope & Colls Ltd v North West Metropolitan Regional Hospital Board [1973] 1 WLR 601 (HL) 609. 80 A-G of Belize (n 78) [21]. 81 BP Refinery (Westernport) Pty Ltd v Shire of Hastings [1977] UKPC 13, (1977) 180 CLR 266, 283: for a term to be implied, the following conditions (which may overlap) must be satisfied: (i) it must be reasonable and equitable; (ii) it must be necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it; (iii) it must be so obvious that ‘it goes without saying’; (iv) it must be capable of clear expression; and (v) it must not contradict any express term of the contract. 82 A-G of Belize (n 78) [27]. 83 Commonwealth Bank of Australia v Barker [2014] HCA 32, (2014) 253 CLR 169 [22]. 84 Codelfa Construction (n 19) 345. 85 ibid 344. 86 ibid 392. 87 Marks & Spencer plc v BNP Paribas Securities Services Trust Co (Jersey) Ltd [2015] UKSC 72, [2016] AC 742.

The Interpretation of Written Contracts 257 considered the issue in the context of a lease that had been granted with rent payable quarterly in advance. The tenant exercised its right under a break clause to terminate the lease. The tenant had paid the full quarter’s rent in advance the month earlier, and so it argued that there was an implied term that it could recover from the landlords the advance payment of rent for the remainder of the quarter. The Supreme Court unanimously held that it could not. One reason was that non-apportionability of such rent had been long and clearly established by authority, and the lease was a very full and carefully considered contract, which includes express obligations of the same nature as the proposed implied term, namely, financial liabilities in connection with the tenant’s right to break, and that term would lie somewhat uneasily with some of those provisions.88

That approach, with respect, is an entirely orthodox approach to interpretation which gives significant weight to the literal meaning of the words used in an instrument drafted by professionals. The broader significance of the decision came in the discussion by Lord Neuberger (with whom Lords Sumption and Hodge agreed) concerning implied terms and the Belize decision. Lord Neuberger disagreed with the approach of the Privy Council in Belize. His Lordship said that ‘construing the words used and implying additional words are different processes governed by different rules’.89 Quoting from Sir Thomas Bingham,90 he described interpretation as concerned with resolving ambiguities or reconciling apparent inconsistencies to attribute the true meaning to the language used by the parties. In contrast, implication was said to ‘deal with matters for which, ex hypothesi, the parties themselves have made no provision’.91 Lord Neuberger then refined the five BP criteria effectively to three: (i) it must be necessary to give business efficacy to the contract, so that no term will be implied if the contract is effective without it; (ii) it must be capable of clear expression; and (iii) it must not contradict any express term of the contract. It is not entirely clear what Lord Neuberger meant by his remark that implication is not an exercise in construction. It can be immediately accepted that implication arises where the parties have not made express provision concerning that subject matter. But this does not mean that implication is unconcerned with ascertaining the meaning that a reasonable person would understand the parties to have intended. It might best be understood as an insistence that, since implication involves a clear departure from the formal words of the contract, the meaning reasonably intended by the parties should only contain an implication where the implication is clear and necessary to make the contract effective.

VIII. Conclusion The historical development of the interpretation of written contracts has not been rapid. The reason for this is that strict and literal approaches were thought to carry greater certainty. A remnant of the historical approach, to the extent to which it still exists, is the ambiguity gateway to admissibility of extrinsic materials for interpretation of written contracts. Whether or not that

88 ibid [49]. 89 ibid [26]. 90 ibid [29], quoting from Philips Electronique Grand Public SA v British Sky Broadcasting Ltd [1995] EMLR 472 (CA) 481. 91 Philips Electronique (n 90) 481.

258 James Edelman remnant can be justified by reference to concerns of certainty or other arguments, it is essential to consider those concerns and arguments in light of a proper understanding of the guiding principle of the intention of the parties. As the interpretation of written contracts progressed from a stiff and superstitious form of literalism to a flexible rationalism, it did so by reference to the guiding principle of the intention of the parties. Fundamental to an understanding of the flexible rationalism of interpretation, and to the resolution of disputed issues, such as the approach to be taken to an ambiguity gateway, rectification or implication, is an appreciation that references to the ‘intention of the parties’ describe the intention of a notional reasonable person in the position of both of the parties. Interpretation of written contracts follows the same pattern as ordinary discourse in the rest of life. There remain, of course, further and related issues concerning how interpreted meaning should be applied to the facts of a case. But, even in a chapter for Frank Rose, those issues must be left for another day.

15 The Boilerplate and the Bespoke: Should Differences in the Quality of Consent Influence the Construction and Application of Commercial Contracts? DAVID FOXTON

I. Introduction Lawyers use the expression ‘boilerplate’ to describe clauses in contractual forms to which the parties do not address their minds,1 pre-packaged language which is not negotiated and, in all probability, is not up for negotiation. It is its ostensible non-negotiability which gives this form of drafting its name, boilerplate being the ‘thick, tough steel sheets used to build boilers’, which came to describe plates of unamendable syndicated text distributed for publication by newspapers across the USA.2 That metallurgical metaphor has expanded, to reflect a drafting process whereby ‘bits of legal boilerplate are bolted together so that it is the words that are allowed to shape the concept instead of the other way round’.3 The American contract scholar Margaret Jane Radin has argued that the mass use of boilerplate, particularly in digital form, is leading to ‘normative degradation’, expanding the territory of what lawyers and courts are prepared to label as contracts to the point where ‘consent is degraded to assent, then to fictional or constructive or hypothetical assent and then further to mere notice … until finally we are left with only a fictional or constructive notice of terms’.4 While Radin’s principal focus is boilerplate in commercial interactions between consumers and large corporations, the challenges posed by boilerplate to the quality of contractual consent are not limited to the consumer context. One general counsel of a large multi-national contractor has noted that ‘the reality is that not all contractual provisions are created equal, and there are factors that will impede a complete review, including time constraints and budgetary

1 Motortrak Ltd v FCA Australia Pty Ltd [2018] EWHC 990 (Comm) [122] (Moulder J). 2 MJ Radin, Boilerplate: The Fine Print, Vanishing Rights and the Rule of Law (Princeton, Princeton University Press, 2014) xvi–xvii. 3 Oxonica Energy Ltd v Neuftec Ltd [2008] EWHC 2127 (Pat) [2] (Peter Prescott QC). 4 Radin (n 2) 30. For responses, see O Ben-Shahar, ‘Regulation through Boilerplate: An Apologia’ (2014) 112 Michigan Law Review 883; NS Kim, ‘Boilerplate and Consent’ (2014) 17 Green Bag 2d 293.

260 David Foxton concerns’, with the negotiating process focusing on ‘the commercial and technical aspects’ and ‘a few select legal issues, such as warranty, limitation of liability, termination, dispute resolution and the governing law provisions of the contract’.5 He further noted that ‘the large majority of contracts placed by companies are automatically generated and performed without the parties ever reaching a final written agreement or complying with the formal legal requirements’.6 For these reasons, similar concerns have been articulated by English commentators about the corrosive effect of pre-printed and non-negotiable packages of contract terms in commercial contracts, including by Gerard McMeel, who has criticised judges for preferring ‘the virtual reality of the boilerplate clauses to the more complicated reality of commercial negotiations’,7 and for according a significance to boilerplate provisions which bears ‘no relationship to reality’.8 These and other commentators point to a crisis of normative legitimacy in contract law, in which ‘[t]he contracting still imagined by courts and law teachers as typical, in which both parties participate in choosing the language of their entire agreement, is no longer of much more than historical importance’.9 There have been legislative responses to the phenomenon of boilerplate and its propensity to generate contractual liabilities or immunities without meaningful consent. This has taken the form of international conventions applicable in particular trades such as the Hague Rules, and domestic or implementing legislation such as the Unfair Contract Terms Act 1977 and the Consumer Rights Act 2015.10 This chapter considers how far the common law is able to reflect differences in the quality of consent between terms which are negotiated for the specific contract in contemplation – the ‘bespoke’ – and legal boilerplate in construing and applying commercial contracts. It begins with a brief history of boilerplate, and the legislative responses to it. It then addresses various contexts in which courts have had to consider what significance to attach to the peripheral role of boilerplate terms during the period when contractual intention is formed and communicated, considering in turn issues of incorporation, interpretation, integration, variation and finally when the fact of an agreement is relied on as evidence. It concludes that the common law of contract does on occasion calibrate its normative response to variations in the quality of contractual consent, but does so inconsistently, and in some respects insufficiently.

II. Boilerplate – A Short History In A Schroeder Music Publishing Co Ltd v Macaulay,11 Lord Diplock traced the phenomenon of standard form contracts back to contracts ‘of very ancient origin’, which ‘set out the terms upon which mercantile transactions of common occurrence are to be carried out’. He gave as e xamples 5 D Echenberg, ‘Negotiating International Contracts: Does the Process Invite a Review of Standard Contracts from the Point of View of National Legal Requirements?’ in G Cordero-Moss (ed), Boilerplate Clauses, International Commercial Contracts and the Applicable Law (Cambridge, Cambridge University Press, 2011) 11–12. 6 ibid 17. 7 G McMeel, ‘Documentary Fundamentalism in the Senior Courts: The Myth of Contractual Estoppel’ [2011] Lloyd’s Maritime and Commercial Law Quarterly 185, 188. See also G McMeel, ‘The Impact of Exemption Clauses and Disclaimers: Construction, Contractual Estoppel and Public Policy’ in A Dyson, J Goudkamp and F Wilmot-Smith (eds), Defences in Contract (Oxford, Hart Publishing, 2017) 249: ‘the virtual world of the contractual documentation trumps the actual reality of complex negotiations and deal-making’. 8 McMeel (2011) 207. 9 WD Slawson, ‘Standard Form Contracts and Democratic Control Over the Lawmaking Power’ (1971) 84 Harvard Law Review 529. 10 See text to nn 16–18 below. 11 A Schroeder Music Publishing Co Ltd v Macaulay (formerly Instone) [1974] 1 WLR 1308 (HL) 1316.

The Boilerplate and the Bespoke 261 bills of lading, charter parties, policies of insurance and contracts of sale in the commodity markets. He suggested that standard or pre-packaged clauses in this context were relatively benign because ‘the standard clauses in these contracts have been settled over the years by negotiation by representatives of the commercial interests involved and have been widely adopted because experience has shown that they facilitate the conduct of trade’. This description is over-idealised. Even in this context, the use of boilerplate provisions presented on a ‘take it or leave it’ basis proved problematic, the most noteworthy example being bills of lading. Judge Anthony Diamond QC noted that the absence of statutory control had the result that ‘shipowners had been able to introduce into the small print on the reverse of their bills of lading exceptions clauses relieving them from an ever increasing range of liabilities’.12 Francis Rose has traced through the judicial and legislative response to this blue-water boilerplate,13 culminating in 1924 in the Hague Rules.14 Alongside these categories came what Lord Diplock described as standard forms which resulted from ‘the concentration of particular kinds of business in relatively few hands’, of which he gave railway ticket cases in the nineteenth century as the first examples, followed by cloakroom and warehouse receipts.15 As late as 1953, Harry Sales, in the Modern Law Review, described the use of standard form contracts outside of the traditional categories of shipping and insurance as ‘a new phenomenon’, and the rate of increase in their use in the twentieth century as ‘phenomenal’.16 There were a number of targeted legislative responses to boilerplate of this kind – the Railway and Canal Traffic Act 1854, the Bills of Sale Act 1878 and the Moneylenders Act 1927 among them. A more generalised response came in the form of the Unfair Contract Terms Act 1977 (UCTA 1977), with section 3 providing some form of protection to one contracting party dealing ‘on the other’s written standard terms of business’, and the Unfair Terms in Consumer Contracts Regulations 1999 (UTCCR 1999), which distinguished between standard terms and ‘terms which have been individually negotiated’.17 The identification of which terms fell into which category for the purpose of these statutory controls was not without its difficulties,18 and the ‘terms individually negotiated’ exception was not maintained in the Consumer Rights Act 2015.

III. Incorporation A. The Notice Cases The context in which the issue of boilerplate received its first and most extended treatment by the common law is in what are usually referred to as the ‘ticket’ cases – cases where what

12 Transworld Oil (USA) Inc v Minos Compania Naviera SAL (The Leni) [1992] 2 Lloyd’s Rep 48 (Com Ct) 51–52. 13 FD Rose, ‘Carriage of Goods by Sea’ in A Burrows (ed), Principles of English Commercial Law (Oxford, Oxford University Press, 2015) paras 3.26–3.28. 14 See generally MF Sturley, ‘The History of COGSA and the Hague Rules’ (1991) 22 Journal of Maritime Law and Commerce 1. 15 A Schroeder Music Publishing (n 11) 1316. 16 HB Sales, ‘Standard Form Contracts’ (1953) 16 MLR 318, 319, 333. 17 UTCCR 1999, SI 1999/2083, reg 5(1). 18 On UCTA 1977, s 3, see H Beale (gen ed), Chitty on Contracts, 33rd edn (London, Sweet & Maxwell, 2018) vol 1, para 15.084; African Export–Import Bank v Shebah Exploration and Production Co Ltd [2017] EWCA Civ 845, [2018] 1 WLR 487 [21]; St Albans City and District Council v International Computers Ltd [1996] 4 All ER 481 (CA) 490–91; Salvage Association v Cap Financial Services [1995] FSR 654 (QB); Pegler Ltd v Wang (UK) Ltd (No 1) [2000] BLR 218 (TCC); Commercial Management (Investments) Ltd v Mitchell Design & Construct Ltd [2016] EWHC 76 (TCC), 164 Con LR 139 [61]–[73]. On the UTCCR 1999, reg 5(1), see Chitty on Contracts (n 18) vol 2, paras 38.243–38.244; UK Housing

262 David Foxton profess to be contractual terms appear on an unsigned document produced, or at least made available, by one contracting party to the other at or around the time of contracting. In considering the legal treatment of such documents, it is helpful to note Margaret Radin’s distinction between ‘procedural’ and ‘substantive’ responses to the issue of boilerplate: Procedural unconscionability refers to the notion that the way in which the contract came into being looks somehow improper … It refers to defects in bargaining … not defects in the bargain. By contrast, substantive unconscionability refers to defects in the bargain itself.19

The judicial response to the ticket cases was initially ‘procedural’, formulating a test which would determine whether the document was accorded contractual status and offering a binary outcome as to the application of all its terms in the light of the court’s answer to that question.20 This emphasis on the bargaining rather than the bargain reflected the traditional hostility of the common law to any requirement that the court should itself assess the substantive fairness of a particular contract term.21 However, the law evolved to accept that there were circumstances in which the contractual status of the document was not itself conclusive of the contractual status of all terms in the document, at least where the term in issue was particularly onerous or unusual, or took away a right given by statute.22 In what remains one of the highwater marks of the common law’s response to the problem of boilerplate, a two judge Court of Appeal in Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd23 held that a provision on a receipt imposing a very substantial payment obligation for the overdue hire of transparencies was unenforceable. One judge, Dillon LJ, did so on conventional ‘non-incorporation’ grounds, and the other, Bingham LJ, on what is capable of being interpreted as a wider ‘fairness’ analysis, taking into account the unusual or onerous character of the term, the degree of notice which the counterparty had of it and the nature of the parties’ relationship.24 However, if that seed was sown, it failed to sprout, and the courts have remained wary of any suggestion that the lack of negotiation of a particular contractual term is relevant to its contractual status. In Goodlife Foods Ltd v Halt Fire Production,25 Coulson LJ thought that ‘it cannot be right in principle that a party who did not negotiate over every proposed clause but instead accepted what was being offered is in a better position to argue subsequently that the clause in question is unreasonable’. Similarly, mere non-negotiability cannot itself be determinative. As Daniel Markovits has noted, ‘it is no defence against a conventional bespoke contract that the offeror refused to adjust to, or even to entertain, counteroffers, but insisted on his terms or none’.26 It remains possible to analyse this judicial foray into substantive fairness as merely a more sophisticated procedural assessment of whether there has been a sufficient level of consent to

Alliance (North West) Ltd v Francis [2010] EWCA Civ 117, [2010] Bus LR 1034; Khurana v Webster Construction Ltd [2015] EWHC 758 (TCC), [2016] 1 All ER (Comm) 466 [52]; Harrison v Shepherd Homes Ltd [2011] EWHC 1811 (TCC), (2011) 27 Construction Law Journal 709 [103]–[105]. 19 Radin (n 2) 85. 20 In the line of authorities beginning with Parker v South Eastern Rly Co (1877) 2 CPD 416 (CA). 21 eg the rule that the court will not assess the adequacy of consideration: Chitty on Contracts (n 18) vol 1, para 4.014. 22 eg Thornton v Shoe Lane Parking Ltd [1971] 2 QB 163 (CA). 23 Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433 (CA) 439 (Dillon LJ), 439, 445 (Bingham LJ). 24 For conflicting views as to the basis of Bingham LJ’s holding, see Amiri Flight Authority v BAE Systems plc [2002] EWHC 2481 (Comm), [2003] 1 All ER (Comm) 1; [2003] EWCA Civ 1447, [2004] 1 All ER (Comm) 385, in which Tomlinson J said at first instance at [14] that the judgment involved more than simply a decision on incorporation, and in which Mance LJ said the opposite in the Court of Appeal at [15]. 25 Goodlife Foods Ltd v Halt Fire Production Ltd [2018] EWCA Civ 1371, [2018] BLR 491 [91]. 26 D Markovits, ‘Good Faith Bargaining in the Shadow of a Form’ in Dyson et al (n 7) 49.

The Boilerplate and the Bespoke 263 justify giving the term contractual status. It can be argued that a party given notice that a document contains a set of putative contractual terms can fairly be treated as assenting to terms within that document of a kind that the party ought reasonably to have expected the document to contain, but with an enhanced test of consent – or ‘strict scrutiny’ – applying to terms falling outside that reasonable contemplation. That analysis is not without its difficulties. If the issue is what a party should reasonably expect as a matter of fairness and entitlement, rather than what experience ought to lead it to expect the terms to contain as a matter of fact – what Radin terms normative rather than empirical expectation27 – then it would require the court to engage in an assessment of the fairness of the term. This could involve difficult and sensitive issues of policy, could not sensibly be carried out without considering any quid pro quo provided by other terms of the bargain and would involve judges in a task for which they are not particularly well equipped.28 By contrast, if the test is one of empirical expectation, this could encourage a ‘race to the bottom’ by those in a position to stipulate for their own boilerplate, to a point where it might legitimately be said of any contracting party that ‘surely everyone knows by now that unpleasant clauses lurk in boilerplate’. But there is a difficulty with a purely procedural response to boilerplate, namely that it is much more difficult to apply to signed boilerplate.

B. Unexpected Terms in Signed Contracts Where the boilerplate terms have been signed, there has been strong judicial resistance to the suggestion that the unusual or onerous nature of a particular term can lead to the conclusion that the term does not form part of the contract. Here, the influence of AT Denning’s traumatising success as counsel in L’Estrange v F Graucob Ltd29 still looms large,30 with the signature on a document being treated as virtually conclusive of consent to the terms in the signed document absent fraud, misrepresentation or duress, even where the contract reflects one party’s boilerplate. Rix LJ doubted whether there was any scope for the Interfoto principle to be applied to a signed contract,31 and Moore-Bick LJ in Peekay Intermark Ltd v Australian and New Zealand Banking Group Ltd32 went so far as to suggest that the principle that terms in a signed document had contractual status is ‘an important principle of English law which underpins the whole of commercial life; any erosion of it would have serious repercussions far beyond the business community’. There have been occasional suggestions that signed boilerplate might not be effective to incorporate all clauses in all circumstances, suggestions that have focused not purely on the nature of the term, but also on the extent to which the party faced with the term had a fair opportunity

27 Radin (n 2) 85. 28 cf Lord Bramwell’s comment in Manchester, Sheffield & Lincolnshire Rly Co v Brown (1883) 8 App Cas 703 (HL) 718 on the requirement that the court opine on the reasonableness of clauses under the Railway and Canal Traffic Act 1854, s 7: ‘It seems to me perfectly idle, and I cannot understand how it could have been supposed necessary, that it should be referred to a judge to say whether an agreement between carriers, of whose business he knows nothing, and fishmongers, of whose business he knows nothing, is reasonable or not.’ 29 L’Estrange v F Graucob Ltd [1934] 2 KB 394 (KB) 403. See generally JR Spencer, ‘Signature, Consent and the Rule in L’Estrange v Graucob’ [1973] CLJ 104, which notes that the decision of the Divisional Court in that case reflected the earlier ruling in Blake v Pollard & Morris [1930] 1 KB 628 (CA). 30 For Lord Denning’s account, see HL Deb 23 May 1977, vol 383, cols 1119–20; Lord Denning, The Family Story (London, Butterworths, 1981) 99, 174–75. 31 HIH Casualty & General Insurance Ltd v New Hampshire Insurance Co [2001] EWCA Civ 735, [2001] CLC 1480 [209]. See also Andrew Popplewell QC in Do-Buy 925 Ltd v National Westminster Bank plc [2010] EWHC 2862 (QB) [91]. 32 Peekay Intermark Ltd v Australian & New Zealand Banking Group Ltd [2006] EWCA Civ 386, [2006] 1 CLC 582 [43].

264 David Foxton to consider it before the point of contracting. In Ocean Chemical Transport Inc v Exnor Craggs Ltd,33 Evans LJ, with the support of Henry and Waller LJJ, was prepared to assume that onerous and unusual clauses in a signed contract might not be incorporated ‘in an extreme case where a signature was obtained under pressure of time or other circumstances’. Mance LJ in Amiri Flight Authority v BAE Systems plc34 floated the following graphic examples: The case of a car-owner entering a car-park and being asked to sign a ticket handed to him by a car-parking attendant, or that of a holiday-maker required to sign a long small-print document in order to hire a family car at an airport.35

However, there is no case in which an argument of this kind has carried the day. Those cases in which the argument has failed suggest that the court will require relatively little in the surrounding circumstances for the consent to have the requisite quality to establish incorporation. In Ocean Chemical, the court noted that the term had long been available for perusal by the buyer’s agents. In Do-Buy 925 Ltd v National Westminster Bank plc,36 Andrew Popplewell QC noted that the signatory was provided with an opportunity to read the terms prior to signature, and that a clause recommending that the signer read the terms appeared immediately above the signature box.37 In Barclays Bank plc v Grant Thornton UK LLP,38 Cooke J envisaged very little scope for the argument that a term in signed boilerplate was not incorporated into a contract outside the consumer sphere: Whereas the law has sometimes asked for more when ‘boilerplate’ terms in small print are utilised by corporations seeking to exclude liability to unsophisticated individual consumers who cannot be expected to wade through large volumes of text, the individual who dealt with Grant Thornton in relation to each of the audit engagements undertaken with Barclays itself, if he failed to read the relevant paragraphs, has only himself to blame and must be taken to have taken the risk in failing to do so.

C. Arbitration and Jurisdiction Clauses Margaret Radin identifies as one of the most pernicious consequences of according boilerplate the same status as negotiated contractual terms the fact that a claimant may lose the constitutionally protected right of access to the courts if the boilerplate contains an arbitration or jurisdiction clause.39 In a US litigation context, this can amount to denying a claimant the ability to participate in a class action, or to sue in a court which is physically accessible to them. In English law, by contrast, there are contexts in which courts have required a higher consent threshold for the incorporation of arbitration or jurisdiction clauses than provisions of other kinds, at least where the mechanism whereby they are to be given contractual effect is by seeking to incorporate them from another contract involving a different party or parties.

33 Ocean Chemical Transport Inc v Exnor Craggs Ltd [2000] 1 All ER (Comm) 519 (CA) [48]. 34 Amiri Flight Authority (CA) (n 24) [16]. 35 The reference to the ‘family car’ might suggest external pressure to sign of the ‘are we nearly there?’ kind. 36 Do-Buy 925 (n 31) [92]. 37 Those circumstances were stronger in Dawson v Bell [2016] EWCA Civ 96, [2016] 2 BCLC 59 [103], where a draft had been provided two weeks before the point of contracting. 38 Barclays Bank plc v Grant Thornton UK LLP [2015] EWHC 320 (Comm), [2015] 1 CLC 180 [105]. 39 Radin (n 2) xiii–xv. In the USA, this evokes the ‘Due Process’ protection afforded by the Fifth and Fourteenth Amendments to the US Constitution, but a similar policy is reflected in common law cases concerned with ‘ouster’, discussed in Chitty on Contracts (n 18) vol 1, para 16.072–16.075 and now reflected in Art 6 of the European Convention on Human Rights.

The Boilerplate and the Bespoke 265 The most extended consideration of this issue is to be found in Christopher Clarke J’s judgment in Habaş Sinai Ve Tibbi Gazlar Isthisal Endüstriaş v Sometal SAL.40 He distinguished four categories of case: (i) a contract purporting to incorporate standard terms containing an arbitration or jurisdiction clause; (ii) a contract which seeks to incorporate the terms of a previous contract between the same parties containing such a clause; (iii) a contract which seeks to incorporate the terms of a contract between one of those contracting parties and a third party which include an arbitration or jurisdiction clause; and (iv) a contract seeking to incorporate the terms of a contract between two other parties including such a clause. In relation to categories (iii) and (iv), the judge accepted that there was a heightened test for the incorporation of an arbitration or jurisdiction clause, one reflected in a long line of cases addressing the incorporation of charter party terms into bills of lading, and which generally required distinct or specific reference to the arbitration or jurisdiction clause rather than mere general words of incorporation.41 However, he held that this heightened test did not apply to categories (i) and (ii), suggesting that in those contexts arbitration or jurisdiction clauses were not to be treated any differently from any other clauses so far as the issue of incorporation is concerned. The grounds for distinguishing categories (i) and (ii) from categories (iii) and (iv) are not particularly compelling. In the case of a bill of lading incorporating the terms of a charter party, the bill’s negotiable character would provide a relevant ground for distinction,42 albeit one which implicitly acknowledges that it is the lesser quality of consent inherent in receiving the transfer of a contract originally concluded by someone else which justifies a heightened test for incorporation of clauses of this kind. However, categories (iii) and (iv) have been applied to any contract where the incorporation is of a term in another contract, including incorporation from other contracts in a tower of insurance coverage, incorporation of terms in an insurance contract into a reinsurance contract and issues of incorporation in chains of construction or sale contracts.43 Further, there are statements in some of those cases which would support the argument for a heightened test for incorporation of arbitration or jurisdiction clauses in all four categories because such clauses remove or restrict a right of access to the courts – for example, Sir John Megaw in Aughton Ltd v MF Kent Services Ltd44 and Evans LJ in AIG Europe (UK) Ltd v The Ethniki.45 While some courts and commentators have taken the view that it is the heightened test in categories (iii) and (iv) which is the anomaly requiring justification,46 the treatment of this issue at least reveals a capacity on the part of the common law to require a more exacting test of identification and incorporation for particular types of term to achieve contractual status.

40 Habaş Sinai Ve Tibbi Gazlar Isthisal Endüstri AS v Sometal SAL [2010] EWHC 29 (Comm), [2010] Bus LR 880 [12]–[13]. 41 Starting with TW Thomas & Co Ltd v Portsea Steamship Co Ltd (The Portsmouth) [1912] AC 1 (HL). See generally Y Baatz, ‘Should Third Parties Be Bound by Arbitration Clauses in Bills of Lading?’ [2015] Lloyd’s Maritime and Commercial Law Quarterly 85. 42 See, eg Federal Bulk Carriers Inc v C Itoh & Co Ltd (The Federal Bulker) [1989] 1 Lloyd’s Rep 103 (CA) 105, where Bingham LJ stresses the negotiable character of bills of lading as one reason for the ‘strict test of incorporation’. 43 See D Joseph, Jurisdiction and Arbitration Agreements and their Enforcement, 3rd edn (London, Sweet & Maxwell, 2015) para 5.17. 44 Aughton Ltd (formerly Aughton Group Ltd) v MF Kent Services Ltd (1991) 31 Con LR 60 (CA) 86–88. 45 AIG Europe (UK) Ltd v The Ethniki [2000] 2 All ER 566 (CA) [33]–[34], [37]. 46 For criticism, see Peiniger Joint Venture v Argos Engineering & Heavy Industries Co Ltd [1994] HKCFI, [1994] 3 HKC 328 (Hong Kong HC); International Research Corp plc v Lufthansa Systems Asia-Pacific Pte Ltd [2013] SGCA 55, [2014] 1 SLR 130. Various views are canvased in A Tweeddale and K Tweeddale, ‘Incorporation of Arbitration Clauses Re-visited’ (2010) 76 Arbitration 656; Baatz (n 41).

266 David Foxton

IV. Contractual Interpretation Margaret Radin does not refer to the rules of contractual interpretation when discussing the ‘traditional oversight doctrines’ for boilerplate in contract, referring to coercion or duress, fraud, deception or misrepresentation and invalid contract formation.47 Under English law, however, the principles of interpretation of contracts are an important tool which provide the courts with some scope for channelling contracts into judicially approved norms by requiring higher and more explicit manifestations of consent for deviations from those norms.48 Edwin Peel has suggested that ‘courts will look closely to see precisely what is the bargain that has been reached’ and that ‘interpretation can operate as a control against boilerplate clauses’.49 Certainly, it has historically been the case that clauses which aim to diminish or exclude one party’s basic obligations under a contract or any common law duty arising apart from contract50 have been required to satisfy more exacting standards of explicitness and precision through the application of the contra proferentem rule of construction, although there are signs that the rigour of this doctrine is waning, and the construction of exemption clauses is gradually being assimilated with conventional principles of construction.51 However, those principles of construction were triggered by the content of particular types of clause, rather than the form in which they appeared. There is also another form of the contra proferentem rule which requires ambiguous provisions to be construed against the party who proposed the wording for inclusion, but once again this does not depend on whether the term was proferred in boilerplate or otherwise.52 These instances apart, is the differential quality of consent between bespoke and boilerplate terms reflected in principles of interpretation more generally?

A. Determining the Priority to be Accorded to Conflicting Clauses The suggestion that terms within a contract should be accorded relative significance by considering the extent to which they were the subject of specific consideration finds distinguished support from Karl Llewellyn, who noted that: Instead of thinking about ‘assent’ to boilerplate clauses, we can recognize that so far as concerns the specific, there is no assent at all. What has in fact been assented to, specifically, are the few dickered53 terms, and the more broad type of the transaction, and one thing more. The one thing more is a blanket assent (not a specific assent) to any not unreasonable or indecent terms the seller may have on his form that do not alter or eviscerate the reasonable meaning of the dickered terms. The fine print which has not been read has no business to cut under the reasonable meaning of those dickered terms which constitute the dominant and only real expression of agreement.54 47 Radin (n 2) 123. 48 Carter and Courtney refer to this process as the promotion of ‘standardised solutions’: JW Carter and W Courtney, ‘Unexpressed Intention and Contract Construction’ (2017) 37 OJLS 326, 336. 49 E Peel, ‘The Common Law Tradition: Application of Boilerplate Clauses Under English Law’ in Cordero-Moss (n 5) 133, 178. 50 Youell v Bland Welch & Co Ltd (No 1) [1992] 2 Lloyd’s Rep 127 (Com Ct) 134. 51 For a summary of the history of the various rules applying a strict construction to exemption clauses and recent indications of the relaxation of those rules, see S Tofaris, ‘Commercial Construction of Exemption Clauses’ [2019] Lloyd’s Maritime and Commercial Law Quarterly 270. 52 Youell (n 50). 53 ie negotiated. 54 K Llewellyn, The Common Law Tradition: Deciding Appeals (Boston, Little Brown & Co, 1960) 370. See also K Llewellyn, ‘“The Standardization of Commercial Contracts and Continental Law” by O Prausnitz’ (1939) 52 Harvard Law Review 700 (book review).

The Boilerplate and the Bespoke 267 There has long been strong support in English law for the principle that those terms which were specifically negotiated should be accorded greater weight than boilerplate provisions, whether incorporated or taking the form of printed text in the signed contract. A relatively early case, Glynn v Margetson & Co,55 involved a ‘conflict’ between the typed-in words of a bill of lading (carriage of an identified perishable cargo from A to B) and a printed clause giving the broadest liberty to deviate in the course of any voyage. The House of Lords was not willing to countenance a reading which gave the printed term no scope whatsoever, but it read the term down so as to render it heavily subservient to the ‘bespoke’ terms. The printed term had to give way to ‘the main object and intent’ of the contract, it being ‘legitimate to bear in mind that a portion of the contract is on a printed form applicable to many voyages, and is not specially agreed upon in relation to the particular voyage’. While the case represents a strong instance of privileging the bespoke terms over boilerplate language, it illustrates a tendency in the judicial response to conflicts between the bespoke and the boilerplate of approaching the issue of interpretation as a jigsaw puzzle in which all the pieces must be used, and in which the optimal outcome is an interpretation which gives some effect to all parts. It is often noted that it would be ‘a strong thing’ to disregard the printed terms altogether.56 The tendency is also evident in Langley J’s approach to a similar issue in Bayoil SA v Seawind Tankers Corp: The Courts will … seek to construe a contract as a whole and if a reasonable commercial construction of the whole can reconcile two provisions (whether typed or printed) then such a construction can and in my judgment should be adopted.57

The impetus to give some function to all clauses, whether bespoke or boilerplate, means that the court imposes a fairly demanding test of what constitutes a ‘conflict’ between two clauses. In Alchemy Estates Ltd v Astor, Sales J put the matter in the following terms: It is true that special terms take precedence over standard conditions if there is a clear conflict between them … However, where the parties have adopted both standard conditions and special conditions, and have not indicated that the standard conditions they have adopted are to be displaced in some respect by a special condition and there is no necessary inconsistency between them, then it appears that the parties’ intention is that the standard conditions and the special conditions should be interpreted together as parts of one coherent contractual scheme.58

However, this search for all-embracing coherence risks a false equivalence between the boilerplate and the bespoke. It can legitimately be said that it is only the bespoke terms which were the subject of meaningful consent, and that the reasonable expectation of the contracting parties will ordinarily be that those bespoke provisions – certainly those defining the contract’s characteristic obligations – will operate in accordance with their natural meaning, irrespective of the boilerplate. For those bespoke provisions to be read down, or altered in their meaning, because of an unnoticed and unnegotiated provision in the boilerplate, risks the contractual vice of allowing a party to grant with one clause of a contract what is surreptitiously taken away by another, the ‘derogation from grant’ to which English law has generally been hostile.59

55 Glynn v Margetson & Co [1893] AC 351 (HL) 344–45 (Lord Herschell). 56 Schiffshypothekenbank Zu Luebeck AG v Compton (The Alexion Hope) [1988] 1 Lloyd’s Rep 311 (CA) 316. 57 Bayoil SA v Seawind Tankers Corp (The Leonidas) [2001] 1 Lloyd’s Rep 533 (Com Ct) 536. 58 Alchemy Estates Ltd v Astor [2008] EWHC 2675 (Ch), [2009] 1 WLR 940 [35]. 59 A principle most commonly applied to the construction of contracts for the sale of land or leases, but which reflects a general principle of contractual construction: Esso Petroleum Co Ltd v Addison [2003] EWHC 1730 (Comm) [47]–[49]; Molton Builders Ltd v City of Westminster London BC (1975) 30 P & CR 182 (CA); Johnston & Sons Ltd v Holland [1988] 1 EGLR 264 (CA).

268 David Foxton In a collection of essays in honour of a maritime lawyer of such distinction, it is fitting that English law’s most compelling engagement with the issue of the differing qualities of consent enjoyed by bespoke and boilerplate terms is to be found in a case addressing the terms of a bill of lading, the decision of the House of Lords in The Starsin.60 In that case, the issue arose as to whether the apparently clear identification of the charterer as the contracting party on the face of the bill of lading was qualified by boilerplate language suggesting that the document was signed as agent for the owner of the ship who was really the contractual counterparty. Lord Bingham rejected the ‘jigsaw’ approach, warning that ‘to seek perfect consistency and economy of draftsmanship in a complex form of contract which has evolved over many years is to pursue a chimera’.61 Lord Hoffmann said that where the majority in the Court of Appeal had gone wrong was ‘that they conscientiously set about trying, as lawyers naturally would, to construe the bill of lading as a whole’, whereas ‘in fact the reasonable reader of a bill of lading does not construe it as a whole. For some things he goes no further than what it says on the front’,62 and if that is clear, ‘no attempt at reconciliation is required’.63 A consistent theme of the judgments was that the parties’ expectations were framed by the bespoke terms, and that a construction of the contract which failed to reflect that would ‘frustrate the reasonable expectations of businessmen’.64 Lord Steyn complained about ‘how far removed from the real world of commerce the technical approach advocated by the cargo owners in this case is’.65 The judgments recognise that, for certain purposes, it was not reasonable to expect the parties to look further than the face of the bill and into the boilerplate to determine who they were contracting with. Lord Bingham had great difficulty in accepting that a shipper or transferee of a bill of lading would expect to have to resort to the detailed conditions on the reverse of the bill (and to persevere in trying to read the conditions until reaching conditions 33 and 35) in order to discover who he was contracting with

and ‘even greater difficulty in accepting that he would expect to do so when the bill of lading contains, on its face, an apparently clear and unambiguous statement of who the carrier is’.66 Lord Steyn focused on the ‘reasonable expectations’ of the user, who would give ‘greater weight to words specially chosen’,67 noting that in international trade ‘there is little time for examining the impact of barely legible printed conditions at the time of the issue of the bill of lading’. The majority in the Court of Appeal were criticised for giving ‘preponderant effect to the boilerplate clauses on the back of the bill’ rather than adopting the ‘mercantile’ view. Lord Hoffmann, similarly, said that, in respect of certain issues, ‘if the words [on the front] are reasonably sufficient to communicate the information in question, he does not trouble with the back’.68 The emphasis in The Starsin on the realities of the contracting process, and on the level of cognition and consent which can reasonably be expected of contracting parties, is particularly noteworthy, because it came in an area of law – the carriage of goods by sea – where the view that parties can be expected to look after themselves has particular force, and because the decision was reached notwithstanding a tradition of rather technical and formalistic analyses of the same

60 Homburg

Houtimport BV v Agrosin Ltd (The Starsin) [2003] UKHL 12, [2004] 1 AC 715. [12]. 62 ibid [82]. 63 ibid [85]. 64 ibid [12]. 65 ibid [47]. 66 ibid [15]. 67 ibid [45]–[46]. 68 ibid [82]. 61 ibid

The Boilerplate and the Bespoke 269 or similar issues.69 While it is possible to suggest that the approach in the case reflects the peculiarly fundamental nature of the identification of a contracting party,70 there are indications that the same approach would be justified for other ‘essential commercial provisions’.71 While The Starsin did not herald any significant change in English contract law outside of its immediate bill of lading context, it shows that the common law is not entirely bereft of tools to tackle the challenges posed by boilerplate, if the will is there to use them.

B. Obviously Inapposite Forms There is a greater willingness to ignore boilerplate terms, or to accord them minimal influence in ascertaining the correct construction, when an obviously inappropriate form has been made part of the written contract. In The Alexion Hope,72 a translation of Swedish wording had been incorporated into an insurance policy, expressed in words of which Lloyd LJ observed ‘whether they have a plain and ordinary meaning in Swedish I do not know. But they certainly have no plain or ordinary meaning in English’. He felt able to disregard ‘the printed words of a contract more or less in their entirety’ because ‘at least half the printed words are inapplicable on any view’.73 In AIB Group (UK) plc v Martin,74 Lord Millett suggested that while ordinarily the meaning of a standard form would not be influenced by the particular context in which it was used, a different approach was appropriate when a standard form was used ‘in circumstances for which it was not designed’. Indeed, where an obviously inappropriate form is incorporated, it may be possible to imply a term into a contract even when the term contended for would conflict with a provision in that form. In Aspden v Webbs Poultry & Meat Group (Holdings) Ltd,75 the employer had introduced a generous permanent health insurance scheme, but carried on using a form of employment contract which provided that the contract could be terminated if the employee became ill. Sedley J concluded that it was possible to imply a term that the power of termination would not be exercised once the employee became entitled to the benefits of the scheme, notwithstanding its conflict with the express terms, because the inconsistent terms of the contract were the result of using an inappropriate form without appreciating the consequences of doing so.76

C. The Width to be Accorded to a Boilerplate Provision? Leaving aside conflicts between bespoke and boilerplate language, does the boilerplate nature of a term justify giving the term a narrower interpretation than the language would have if specifically chosen by the parties? There is no authority which supports a special rule of construction

69 As submitted in that case, there had not been a prior English decision in which a bill of lading with a ‘demise clause’ in the boilerplate had been held to be a charterer’s bill of lading: see [2004] 1 AC 715, 728. 70 See in particular ibid [75] (Lord Hoffmann), [175] (Lord Millett). 71 eg ibid [44] (Lord Steyn), [75] (Lord Hoffmann). 72 The Alexion Hope (n 56). 73 ibid 313, 316. See also ibid 322 (Purchas LJ). 74 AIB Group (UK) plc v Martin [2001] UKHL 63, [2002] 1 WLR 94 [7]. 75 Aspden v Webbs Poultry & Meat Group (Holdings) Ltd [1996] IRLR 521 (QB). 76 But cf Lord Millett’s view that rectification of the express term might have been the better course in Reda v Flag Ltd [2002] UKPC 38, [2002] IRLR 747 [51].

270 David Foxton of boilerplate of this type. In Axa Sun Life Services plc v Campbell Martin Ltd,77 Rix LJ favoured a narrower reading of an entire agreement clause in part because ‘it appears to be boilerplate wording’. And in Conductive Inkjet Technology Ltd v Uni-Pixel Displays Inc,78 Roth J accepted a submission that a clause providing that a contract would ‘supersede’ other agreements should not be interpreted as leading to the supersession of two prior specific and formal agreements, in part because it was ‘a boilerplate clause’. However, in Ravennavi SpA v New Century Shipbuilding Co Ltd,79 when addressing a similar clause, Moore-Bick LJ rejected the suggestion that the significance of clauses of this type was ‘diminished if they are found among what are sometimes called the “boilerplate” provisions of a formal contract of this kind’. This appears to be one of those contexts in which the boilerplate nature of a clause can feature as part of the justification of a construction largely arrived at for other reasons, but does not have sufficient force to carry the argument on its own. It is noteworthy that all of these decisions concern entire agreement or similar clauses. Clauses of this kind are almost always creatures of boilerplate, their function being to delineate the perimeter of the bargain. As discussed below, such clauses address issues of commercial certainty in a particularly compelling context, and it is arguable that they can be distinguished in this respect from other boilerplate provisions which might, for example, impose additional obligations on one party or reduce or limit the scope of substantive obligations provided for elsewhere in the contract or otherwise arising as a matter of legal or factual implication. For clauses of this latter type, there may be grounds for approaching the construction of bespoke and boilerplate terms differently, with the latter meriting a presumption of narrow application. Sir Kim Lewison offers some support for such a distinction in suggesting a principle of construction that ‘general words in a standard form of contract are to be limited to the particular object of a particular contract’,80 although he cites no supporting case law. There is something to be said for a cautious interpretative approach of this kind, where the wide language of the boilerplate provisions has been prepared independently of and without regard to the impact of its outer textual reaches on the specific contractual context or the parties’ bespoke provisions. It is when contractual provisions are drafted in the abstract, rather than with a specific bargain in contemplation and a specific potential counterparty who might have to be ‘looked in the eye’, that the drafting habit identified by Hoffmann LJ is most frequently encountered: the determination ‘to make sure that one has obliterated the conceptual target’ and left ‘no loophole for counterattack’.81

D. Admissible Materials If the status of language as bespoke or boilerplate can impact on its interpretation, is material to establish that status admissible? Generally, documents exchanged in the course of the parties’ negotiations are not admissible, a rule largely driven by practical considerations of finality and efficiency rather than from any principled objection.82 Very often, the boilerplate nature of the 77 Axa Sun Life Services plc v Campbell Martin Ltd [2011] EWCA Civ 133, [2011] 2 Lloyd’s Rep 1 [96]. 78 Conductive Inkjet Technology Ltd v Uni-Pixel Displays Inc [2013] EWHC 2968 (Ch), [2014] 1 All ER (Comm) 654 [105]. 79 Ravennavi SpA v New Century Shipbuilding Co Ltd [2007] EWCA Civ 58, [2007] 2 All ER (Comm) 756 [25]. 80 K Lewison, The Interpretation of Contracts, 6th edn (London, Sweet & Maxwell, 2015) para 7.013. 81 Arbuthnott v Fagan [1995] CLC 1396 (CA) 1404. 82 cf Lord Hoffmann in Investors Compensation Scheme Ltd v West Bromwich Building Society [1988] 1 WLR 896 (HL) 913: ‘The law makes this distinction for reasons of practical policy and, in this respect only, legal interpretation differs from the way we would interpret utterances in ordinary life.’

The Boilerplate and the Bespoke 271 language will be apparent from the physical form of the document. If it is not, and evidence to establish it cannot otherwise be adduced, a court faced with a submission based on the language’s alleged boilerplate status might be unwilling to accept that status.83 However, the rules on admissibility of evidence for the purposes of construing contracts have been adapted to some degree in cases of boilerplate language. Where a standard form has been modified, the unmodified form is admissible in evidence for the purposes of interpretation.84 It must follow that the same material would be admissible to show that the contract text was pre-packaged boilerplate. Indeed, reference to the boilerplate in its native form was allowed to correct an error in inaccurately reproduced boilerplate in The Starsin, where a line of text had been omitted in error.85 And while deletions from drafts are generally inadmissible as an aid to interpretation,86 the courts have been willing to look at deletions made to standard form contracts as part of the interpretative exercise, and even at deletions from standard form clauses87 (albeit, as Diplock J noted, there is a ‘pleasant diversity of authority’ on the subject).88 This exception to the parol evidence rule was rationalised by Lord Reid in Timber Shipping Co SA v London & Overseas Freighters Ltd,89 it being said that deletions from a conventional draft are no different from any other form of preliminary drafting suggestion, whereas deletions from a printed form involve removing text which was not included as part of the drafting process. That explanation is not particularly satisfactory,90 but its effect is to provide an occasion when the court privileges the parties’ bespoke contract design (the deletion or amendment), rather than allowing the text of the standard terms that remain to speak for itself.

V. Integration: Entire Agreement Clauses The bespoke and the boilerplate frequently come into conflict when one party alleges that the agreement included an oral term in addition to those recorded in writing, and the other relies on a boilerplate entire agreement clause stipulating that the entirety of the parties’ bargain is to be found in the signed document: clauses referred to by US lawyers as integration clauses, and by English lawyers as entire agreement clauses (EACs). There was a period when the prevailing English orthodoxy was that a specific oral agreement would ‘trump’ the boilerplate EAC wording in this scenario.91 In 1986, the Law Commission summarised the position as follows: But if it were proved that the intention of the parties was to make one contract partly in writing and partly orally, the court would give effect to that contract. The parties might have been aware of the

83 As Moulder J did in Motortrak (n 1) [121]. 84 Fraser v Canterbury Diocesan Board of Finance (No 2) [2003] EWHC 1075 (Ch), [2003] WTLR 1125. 85 The Starsin (n 60) [22]: the error arose as a result of ‘homoeoteleuton, where one sentence contains a word which closed the preceding sentence and the transcriber’s eye has wandered from one to the other, leading to the entire omission of the whole passage lying between them’. 86 Lewison (n 80) para 3.04. 87 Team Services plc v Kier Management and Design Ltd (1993) 63 BLR 76 (CA); Bovis Lend Lease Ltd v Cofely Engineering Services [2009] EWHC 1120 (TCC) [23]. 88 Louis Dreyfus & Cie v Parnaso Cia Naviera SA (The Dominator) [1959] 1 QB 498 (Com Ct) 513. The cases are summarised and analysed by Christopher Clarke J in Mopani Copper Mines plc v Millennium Underwriting Ltd [2008] EWHC 1331 (Comm), [2008] 1 CLC 992 [100]–[123] in the most complete treatment of the issue. 89 Timber Shipping Co SA v London & Overseas Freighters Ltd (The London Explorer) [1972] AC 1 (HL) 15. 90 As Christopher Clarke J noted in Mopani (n 88) [112]. 91 See generally D McLauchlan, ‘The Entire Agreement Clause: Conclusive or a Question of Weight?’ (2012) 128 LQR 521.

272 David Foxton i ntegration clause when they agreed the additional terms but have agreed to ignore it, or they might have forgotten about the clause or never read it. Whatever the reason for there being an integration clause and additional terms, the court will give effect to the intention of the parties as it is proved or admitted to have been.92

Those who argue for a presumptive rather than absolute effect for EACs do so precisely because of the importance of privileging specific agreement over boilerplate. David McLauchlan, for example, notes that the EAC ‘may, for example, be part of the boilerplate that neither party noticed and the actual position may be that an undertaking not included in the written agreement was given and was intended to be legally binding’.93 However, such clauses are now invariably enforced by the courts,94 in what might be regarded as a strong judicial statement that a higher quality of consent (in this context, to a collateral warranty) does not trump the deemed consent involved in signing up to boilerplate. A particularly influential judgment is Lightman J’s in Inntrepreneur Pub Co Ltd v East Crown Ltd, where he identified the purpose of such clauses as being: to preclude a party to a written agreement from thrashing through the undergrowth and finding in the course of negotiations some (chance) remark or statement (often long forgotten or difficult to recall or explain) on which to found a claim such as the present to the existence of a collateral warranty. The entire agreement clause obviates the occasion for any such search and the peril to the contracting parties posed by the need which may arise in its absence to conduct such a search. For such a clause constitutes a binding agreement between the parties that the full contractual terms are to be found in the document containing the clause and not elsewhere, and that accordingly any promises or assurances made in the course of the negotiations (which in the absence of such a clause might have effect as a collateral warranty) shall have no contractual force, save insofar as they are reflected and given effect in that document. The operation of the clause … is to denude what would otherwise constitute a collateral warranty of legal effect.95

However, when a contract is concluded through the signature of a document, the fact that something said orally is not included in the written form may itself be highly significant. McLauchlan notes that one reason why something said orally may not find itself in writing is because ‘negotiations may have been “delicate” and hence it may be feared that an insistence on putting everything in writing will jeopardise the deal’.96 It is not only for lawyers that a signed written agreement has a qualitative significance over and above an oral agreement, as captured in the oft-quoted misattribution to Samuel Goldwyn that ‘a verbal contract is not worth the paper it’s written on’.97 Atiyah also noted the widespread appreciation of the significance of signing a document.98 It is precisely for this reason that courts are often sceptical of claims by business people that something was agreed orally but did not find its way into the document. For example, Mr Justice Langstaff said

92 Law Commission, The Parol Evidence Rule (Law Com No 154, 1986) para 2.15. 93 McLauchlan (n 91) 523. 94 See, eg Axa Sun Life Services (n 77) [34]; Springwell Navigation Corp v JP Morgan Chase Bank (formerly Chase Manhattan Bank) [2010] EWCA Civ 1221, [2010] 2 CLC 705. There are rare exceptions: eg Ryanair Ltd v SR Technics Ireland Ltd [2007] EWHC 3089 (QB) (criticised in Lewison (n 80) para 3.11 and in Mileform Ltd v Interserve Security Ltd [2013] EWHC 3386 (QB) [104]–[105]); 1406 Pub Co Ltd v Hoare [2001] 23 EG 154 (CS) (Ch) (not followed in Inntrepreneur Pub Co (CPC) Ltd v Sweeney [2002] EWHC 1060 (Ch), [2002] EGLR 132). 95 Inntrepreneur Pub Co Ltd v East Crown Ltd [2000] 2 Lloyd’s Rep 611 (Ch) [7]. 96 McLauchlan (n 91) 529. 97 PF Boller and J George, They Never Said It: A Book of Fake Quotes, Misquotes and Misleading Attributions (Oxford, Oxford University Press, 1990) 42. 98 PS Atiyah, ‘Form and Substance in Legal Reasoning: The Case of Contract’ in N MacCormick and PBH Birks (eds), The Legal Mind: Essays for Tony Honoré (Oxford, Oxford University Press, 1986) 34.

The Boilerplate and the Bespoke 273 of a litigant in his case that ‘this is a man who, if he had insisted upon commission terms, and if he is the businessman I think him to be, would have wanted to put it in writing’.99 There is ample sense in Peden and Carter’s observation that there is a ‘general commercial understanding that a negotiated document is executed with the object of crystallising the bargain’.100 It is precisely for this reason that an EAC will not ordinarily defeat a side letter to a written contract.101 In short, the very fact that something agreed orally has not found its way into the written document itself gives the oral agreement a qualitatively inferior status. There is the further difficulty that it can be argued that an EAC clause involves not simply determining the relative weight to be accorded to an oral collateral agreement and the written boilerplate, but the relative weight to be accorded to a consensus reached at a point before the parties were bound and a clause which takes effect at the point of commitment. In addition to weighing the quality of the respective consent, therefore, it is also necessary to take into account the general preference of the law for the later expression of the parties’ consent over an earlier one.102 In SERE Holdings Ltd v Volkswagen Group UK Ltd,103 Christopher Nugee QC said of an EAC that he could see no reason why parties who have in fact reached an agreement in precontractual negotiations that would otherwise constitute a collateral contract should not subsequently agree in their formal written contract that any such collateral agreement should have no legal effect.

Will an EAC always trump specific consent between the parties which does not find its way into the written contract? If the test for rectification of the written instrument is met, the answer appears to be no, even though the presence of such a clause may militate against the continuing existence of the common intention which it is said that the document fails to reflect.104 And in this context, it is likely to matter whether the EAC was specifically considered by the parties or simply formed part of the boilerplate. Christopher Pymont QC observed in Surgicraft Ltd v Paradigm Biodevices Inc: If an entire agreement clause was part of the travelling draft but there is no evidence that the parties themselves actually understood what it meant or let it affect their thinking in any way, it may be difficult to derive anything from it: its existence may simply be part of the mistake in expressing the parties’ intentions. It could be different if the evidence showed that the parties actually considered what this clause meant as part of their negotiation.105

There is an obvious risk here of rectification being used to deprive EACs of all or most of their effect. Perhaps for that reason, in Procter & Gamble Co v Svenska Cellulosa Aktiebolaget SCA,106 Hildyard J held that rectification was available ‘to correct an errant provision’ in a contract containing an EAC, but not ‘to put in an extra term which is presently not there at all’. This may be too prescriptive – the purpose of EACs would be maintained if rectification extended to

99 Forstater v Dweck [2007] EWHC 1991 (QB) [55]. 100 E Peden and JW Carter, ‘Entire Agreement – and Similar – Clauses’ (2006) 22 Journal of Contract Law 1. 101 Cheverney Consulting Ltd v Whithead Mann Ltd [2007] EWHC 3130 (Ch). 102 cf the general rule that a later agreement replaces an earlier one in Patmore v Colburn (1834) 1 Cr M & R 65, 149 ER 996; SW Strange Ltd v Mann [1965] 1 WLR 629 (Ch). 103 SERE Holdings Ltd v Volkswagen Group UK Ltd [2004] EWHC 1551 (Ch) [22]. 104 See, eg I Spry, Equitable Remedies, 9th edn (Pyrmont, NSW, Law Book Co, 2014) 636; Phillips Petroleum Co (UK) Ltd v Snamprogetti Ltd [2001] EWCA Civ 889, (2001) 79 Con LR 80 [32]. 105 Surgicraft Ltd v Paradigm Biodevices Inc [2010] EWHC 1291 (Ch) [75]. 106 Procter & Gamble Co v Svenska Cellulosa Aktiebolaget SCA [2012] EWHC 498 (Ch) [101]. For criticism, see McLauchlan (n 91) 535.

274 David Foxton including terms which the parties had a common intention would appear in that written agreement, but not otherwise. In this way, the distinction which parties might reasonably be expected to draw between the matters discussed orally which did find their way into the written contract and those which did not would be maintained. For essentially similar reasons, an EAC will not trump an estoppel by convention as to the proper effect of the written terms.107

VI. Variation Sometimes, the undocumented agreement which finds itself in conflict with contractual boilerplate is a post-contractual oral agreement to vary the written contract, when the boilerplate provides that the contract can only be varied or supplemented in writing (a no oral modification, or NOM, clause). After initial equivocation,108 the Court of Appeal concluded that a NOM clause did not prevent a subsequent oral modification from taking effect. In Globe Motors Inc v TRW LucasVarity Electric Steering Ltd,109 the Court of Appeal held that, just as it was open to the parties to agree the NOM clause in the first place, it was open to them to agree to dispense with it, and they were no more able to fetter their ability to do so in advance than Parliament was able to bind its successors.110 That decision was followed by a differently constituted Court of Appeal in MWB Business Exchange Centres Ltd v Rock Advertising Ltd.111 In reaching these conclusions, the courts were not influenced by any relative difference in the quality of consent of the two agreements, but rather by an overriding respect for party autonomy which included the autonomy to change their minds. In short, it was the temporal argument which is sometimes employed to support EAC clauses – the later in time agreement or term will prevail – which was seen to negate the operation of the NOM clause. The ‘entirely conceptual’ premise that the parties could unmake whatever agreement they had reached was rejected by the Supreme Court, who held that if the parties wished to attach self-imposed formal requirements to their ability to vary their bargain, they could do so.112 Lord Sumption for the majority saw no reason not to respect the parties’ autonomy to limit their future autonomy.113 It is only Lord Briggs’s dissenting judgment which attaches significance to the comparative qualities of consent, requiring a heightened consent which specifically considered the NOM clause for a subsequent oral agreement to override it: either express reference or necessary implication beyond the mere fact of reaching an oral agreement.114 Those who argued that the Court of Appeal decisions in Globe Motors and MWB were wrong – in particular, Jonathan Morgan in an influential contribution115 and Janet O’Sullivan116 –

107 Dubai Islamic Bank PJSC v PSI Energy Holding Co BSC [2011] EWHC 2718 (Comm). 108 There were initially conflicting statements. In the Court of Appeal, Sedley LJ assumed the efficacy of a NOM clause in United Bank Ltd v Asif (CA, 11 February 2000), whereas Moore-Bick LJ assumed that the NOM clause could itself be varied orally in Westbrook Resources Ltd v Global Metallurgical Inc [2009] EWCA Civ 310, [2009] 2 Lloyd’s Rep 224 [13]. Gloster J, in Energy Venture Partners Ltd v Malabu Oil and Gas Ltd [2013] EWHC 2118 (Comm) [274], and Stuart-Smith J, in Virulite LLC v Virulite Distribution Ltd [2014] EWHC 366 (QB), [2015] 1 All ER (Comm) 204 [55]–[60], thought NOM clauses were ineffective. 109 Globe Motors Inc v TRW LucasVarity Electric Steering Ltd [2016] EWCA Civ 396, [2017] 1 All ER (Comm) 601. 110 ibid [97], [99], [119]. 111 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2016] EWCA Civ 553, [2017] QB 604. 112 MWB Business Exchange Centres Ltd v Rock Advertising Ltd [2018] UKSC 24, [2019] AC 119 [13]. 113 ibid [11]. 114 ibid [22], [26], [30]–[31]. 115 J Morgan, ‘Contracting for Self-Denial: On Enforcing “No Oral Modification Clauses”’ [2017] CLJ 589, 596–603. 116 J O’Sullivan, ‘Unconsidered Modifications’ (2017) 133 LQR 191, 195.

The Boilerplate and the Bespoke 275 laid significant emphasis on the analogy between an EAC and a NOM clause. That argument appealed to the Supreme Court,117 who, having rejected the argument that the contracting parties could not tie their hands as to their future dealings, saw no legitimate basis for distinguishing the two. However, it is suggested that many commercial lawyers would see the two fact patterns differently, and would support the Inntrepeneur118 approach while disagreeing with the Supreme Court decision in MWB. While both situations involve a clash between bespoke oral and written boilerplate terms, they occur in very different contexts. Concluding a legal agreement is a quintessentially legal occasion, one in which the significance of the written document being signed at the same time will be obvious, and the potential significance of terms appearing or not appearing in the signed document will be equally so. It is also an occasion on which the parties are likely to be in receipt of legal advice, or at least alive to the need for it. And it is a ‘threshold’, a moment when the parties know that signing will effect a significant change to their relationship. By contrast, once a contract is signed, it will all too often sit dormant in the files during performance – including misperformance and non-performance – until the emergence of a dispute generates a further legal occasion. While it has been noted that ‘day-to-day contractual negotiations take place in many cases between commercial or technical people and without a lawyer being present’,119 that is even more true of variations in the course of the performance of a contract. And a post-contractual variation will very often merge into the continuum of performance, rather than standing out as a significant event in its own right. For those reasons, it is suggested that MWB is a case where the decision in the Court of Appeal better accords with the realities of the contracting process than the ‘freedom of contract’ fundamentalism of the Supreme Court. It can no doubt be said, as the Supreme Court did say,120 that it will be possible to address through estoppel any cases of genuine hardship which result from an oral variation against the background of a NOM clause. But it is far from clear that MWB will be confined to wholly executory variations.121 Lord Sumption noted that ‘the scope of estoppel cannot be so broad as to destroy the whole advantage of certainty for which the parties stipulated when they agreed upon terms including the No Oral Modification clause’.122 In any event, estoppels are messy. It has been suggested that EACs have the benefit of ‘simplifying the task before the court’ and saving the court from a trawl through pre-contractual exchanges,123 but it seems inevitable that disputed oral modification cases will almost invariably feature an estoppel plea. The very fact that, unlike pre-contractual negotiations, the parties’ dealings occur in the context of an existing legal relationship will make it much easier to plead an estoppel. If the boundaries of contract are drawn in such a way that it becomes a matter of routine that hard cases are catered for through ‘the various doctrines of estoppel’,124 then we are likely to be drawing them in the wrong place.

117 MWB (SC) (n 112) [14]–[15] (Lord Sumption). Lord Sumption refers to the articles by Morgan (n 115) and O’Sullivan (n 116) at [9]. 118 Above n 95. 119 MC Vettese, ‘Multinational Companies and National Contracts’ in Cordero-Moss (n 5) 22. 120 MWB (SC) (n 112) [16] (Lord Sumption). 121 Morgan (n 115) suggests (at 606) that successful estoppel cases will be rare and (at 611) that they should be rare because ‘full acceptance of this suggestion would mean that many, perhaps most, informal modifications would be enforced by way of estoppel’. 122 MWB (SC) (n 112) [16]. 123 Morgan (n 115) 596–97. 124 MWB (SC) (n 112) [16] (Lord Sumption).

276 David Foxton Lord Briggs preferred the analogy of an agreement ‘subject to contract’ (STC) to that of an EAC,125 a view which finds support from Robert Harris, who notes that: whereas NOM clauses seek to prevent parties from varying their agreement in the future, thus ensuring the continued effect of their earlier intentions, entire agreement clauses seek to nullify representations which the parties have made in the past, thus giving effect to their latest intentions.126

However, it is once again suggested that the analogy is inapt. An STC provision is not something imposed without thought or in boilerplate. Its role is to prevent something agreed in the past having effect until a point in the future: to prevent the parties making the important transition into a legally binding relationship without a specific and conscious decision to this effect. By contrast, NOM clauses operate when the parties have not simply crossed the Rubicon, but might be well into Gaul. It is clear that an STC provision can be waived without expressly addressing it, and further that such a waiver is implicit in entering into an otherwise binding agreement. In RTS Flexible Systems Ltd v Molkerei Alois Müller GmbH & Co KG, it fell to the Supreme Court to consider a case in which the parties had embarked on the performance of their agreement, notwithstanding the fact that clause 48 provided that it would ‘not become effective until each party has executed a counterpart and exchanged it with the other’, and this had yet to occur. They asked and answered the following question: Had the parties agreed to be bound by the agreed terms without the necessity of a formal written contract or, put another way, had they agreed to waive that requirement and thus clause 48? We have reached the conclusion that they had.127

If an STC provision can be overcome in this way, without the parties specifically addressing the STC clause, a NOM clause ought to be capable of being overridden by nothing more than a subsequent binding agreement. One final field for the possible application of MWB to contractual boilerplate should be considered. A common boilerplate provision provides that there will be no waiver of a contractual right unless in writing (a ‘no waiver’ provision). Does such a provision prevent conduct by one or other party which would otherwise satisfy the requirement for a waiver from having this effect? Two situations must be considered. The first is where a party orally waives an obligation created for its sole benefit,128 and its counterparty subsequently seeks to suggest for its own reasons that the waiver was invalid because it was not in writing as the boilerplate required. The pre-MWB cases held that the ‘no waiver’ provision could be waived for the same reason (ie that the ‘no waiver’ provision was also intended to benefit the waiving party, and could be waived by that party): Crédit Agricole Indosuez v BB Energy BV129 and RGI International Ltd v Synergy Classic Ltd.130 It is suggested that these decisions remain correct – MWB is concerned with the ability of the parties by contract to limit their future power to contract, not with the ability of one party unilaterally to excuse the other party’s non-compliance or non-performance. The second

125 ibid [28]. 126 R Harris, ‘Modifications, Wrangles and Bypassing: Rock v MWB’ [2018] Lloyd’s Maritime and Commercial Law Quarterly 441, 444(emphasis in original). The same point about the different chronological orderings of NOM clauses and EACs is made by PS Davies, ‘Varying Contracts in the Supreme Court’ [2018] CLJ 464, 466. 127 RTS Flexible Systems Ltd v Molkerei Alois Müller GmbH & Co KG [2010] UKSC 14, [2010] 1 WLR 753 [86]–[87] (emphasis added). 128 Under the principle in Chitty on Contracts (n 18) vol 1, para 2.156. 129 Crédit Agricole Indosuez v BB Energy BV [2004] EWHC 750 (Comm) [32]–[33] (Cooke J). 130 RGI International Ltd v Synergy Classic Ltd [2011] EWHC 3417 (Comm) [51]–[52] (Hamblen J).

The Boilerplate and the Bespoke 277 situation is where a party seeks to rely on a ‘no waiver’ provision to prevent conduct which would otherwise give rise to a waiver from having this effect. It is clear that a clause of this kind will not prevent an election by inconsistent conduct.131

VII. Agreement as Evidence This section of the chapter considers the significance of boilerplate when the fact of agreement is relied upon not for the purpose of determining whether it meets the general requirements for being given normative effect, but as evidence for some other purpose. In applying the doctrine of restraint of trade, and in particular considering whether it has been shown that the restraint is reasonable in the interests of the parties, the courts have frequently relied upon the fact that the restraint was in fact agreed as evidence of its reasonableness for this purpose. Here, however, the courts are astute to distinguish between mere contractual status and ‘real’ agreement. In the Schroeder Music Publishing case,132 Lord Diplock held that a music publisher derived no assistance from the fact that it, and other music publishers, routinely imposed obligations of a particular kind on those with whom they contracted: It is not without significance that on the evidence in the present case music publishers in negotiating with song writers whose success has been already established do not insist upon adhering to a contract in the standard form they offered to the respondent. The fact that the appellants’ bargaining power vis-a-vis the respondent was strong enough to enable them to adopt this take-it-or-leave-it attitude raises no presumption that they used it to drive an unconscionable bargain with him, but in the field of restraint of trade it calls for vigilance on the part of the court to see that they did not.

In the law of penalties, the courts have expressed a strong predisposition to uphold such clauses ‘in the case of commercial contracts freely entered into’.133 Although the law on contractual penalties was significantly remodelled by the Supreme Court in Cavendish Square Holdings BV v Makdessi,134 the fact that the parties to a commercial contract have agreed a clause is itself sufficient to raise a presumption in favour of its enforceability. In Signia Wealth Ltd v Vector Trustees Ltd,135 Marcus Smith J warned that ‘the court should be careful, in a commercial case, where the contract has been negotiated without suggestion of oppression, when applying the doctrine’. These cases raise the question of whether a boilerplate liquidated damages provision which had not been negotiated or the subject of any specific consideration in the course of negotiations would benefit from a similar judicial deference so far as the penalty doctrine is concerned. However, the penalty doctrine is a fast-eroding outcrop of pre-Cambrian detritus, soon to be overwhelmed by the surrounding sedimentary deposits of freedom of contract.136 It is difficult to conceive that anything short of oppression in the bargaining process would prove sufficient to neutralise the presumption of enforceability.

131 Tele2 International Card Co SA v Post Office Ltd [2009] EWCA Civ 9 [56]; CDV Software Entertainment AG v Gamecock Media Europe Ltd [2009] EWHC 2965 (Ch) [91]. 132 A Schroeder Music Publishing (n 11) 1316. 133 Alfred McAlpine Capital Projects Ltd v Tilebox Ltd [2005] EWHC 281 (TCC), [2005] BLR 271 [48] (Jackson J). 134 Cavendish Square Holdings BV v Makdessi [2015] UKSC 67, [2016] AC 1172 [33] (Lord Neuberger). 135 Signia Wealth Ltd v Vector Trustees Ltd [2018] EWHC 1040 (Ch) [653(3)]. 136 cf Aries Tanker Corp v Total Transport Ltd (The Aries) [1977] 1 WLR 185 (HL) 193.

278 David Foxton

VIII. Conclusion Francis Rose observed that ‘the law maintains its vitality by constant reexamination and reassessment’.137 Is the common law’s treatment of boilerplate in need of re-examination and reassessment? Margaret Radin has argued that the pervasive use of boilerplate has led the law of contract to deviate so far from the philosophical premise which underpins that law – that of the voluntary exchange of entitlement through the exercise of free will – that it may be necessary to turn to an expanded law of tort to bring it under control.138 Gerard McMeel has contended that the English judiciary, in the grip of a documentary fundamentalism, has similarly allowed a vast gulf to open up between the reality of commercial negotiations and their reflection in contracts as recognised and enforced in courts,139 an ‘episode in the common law of contract’ which he describes as ‘the most distressing of [his] academic and professional career’.140 It has been noted that boilerplate contracts represent a ‘prominent face of contractual practice’,141 and for this reason ignoring the challenges boilerplate poses to traditional consent models of contract in the development and application of the common law should not be an option. Equally, however, there are narrow limits within which the common law can adapt itself to the practical realities of the bargaining process. Reconstructing that process in court proceedings would be evidentially exacting and consume significant amounts of court time, consequences which are rightly anathema to the ascertainment of the respective rights and obligations of parties to a commercial contract and fundamentally incompatible with the ‘aspiration to certainty’142 which is a legitimate goal of the law of contract, albeit far from its only lode star. This chapter has sought to show that, within these limits, the common law of contract is not wholly insensitive to the differences in the quality of consent between those contracting parties who specifically negotiate contractual terms and those who sign up to boilerplate presented on a ‘take it or leave it’ basis, which passes virtually unnoticed through the contracting process. While falling far short of the more radical intervention that Radin and, to a lesser extent, McMeel might contend for, the common law principles addressing interpretation and incorporation privilege bespoke terms over boilerplate to some degree, and could do so to a greater extent. The significance attached to boilerplate EACs over oral discussions can itself be justified by the commercial significance that failure to document a term discussed orally ought to have for the parties in circumstances in which their legal commitment is marked by the signing of a written contract. By contrast, it is suggested that the decision of the Supreme Court in MWB failed to accord sufficient weight to the higher quality of consent inherent in the oral variation when it came into conflict with the boilerplate NOM clause which preceded it in time. In the common law of contract, the search for consensus ad idem, as Francis Rose entitled a collection of essays in honour of Guenther Treitel,143 is undoubtedly imperfect, but is neither illusory nor in vain.

137 FD Rose (ed), Consensus Ad Idem: Essays on the Law of Contract in Honour of Guenter Treitel (London, Sweet & Maxwell, 1996) x, a preface written, appropriately for such a loyal Magdalen man, on May Day 1996. 138 Radin (n 2) ch 11. 139 McMeel, ‘Documentary Fundamentalism’ (n 7). 140 McMeel, ‘The Impact of Exemption Clauses and Disclaimers’ (n 7) 241. 141 Markovits (n 26) 43. 142 Lewison J in Food Co UK LLP (t/a Muffin Break) v Henry Boot Developments Ltd [2010] EWHC 358 (Ch) [177]. 143 Rose (n 137).

16 Illegality: Pleading, Proof, and Presumptions WILLIAM SWADLING

I. Introduction Until the decision of the UK Supreme Court in Patel v Mirza,1 an exception to the defence of illegality to claims in private law arose where a claimant could by its pleadings make out the elements of its cause of action without having to rely on its own illegal conduct. The leading case was Bowmakers Ltd v Barnet Instruments Ltd,2 a common law claim in tort for the conversion of goods. However, this non-reliance exception was rejected in Patel because it was seen to operate in an arbitrary fashion and Bowmakers was accordingly overruled.3 In its place was substituted a ‘range of factors’ approach, with the court being asked to weigh against one another three different things: the underlying purpose of the prohibition which had been transgressed and whether that purpose would be enhanced by denial of the claim; any other relevant public policy on which the denial of the claim might have an impact; and whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts.4 This chapter will argue that the arbitrariness of the Bowmakers rule came not from anything in the case itself, but from the rule laid down in Gascoigne v Gascoigne5 and Chettiar v Chettiar6 – a rule which was both interpreted by the majority of the House of Lords in Tinsley v Milligan as concerned with the admission of evidence in equity, and which was assimilated to the common law Bowmakers non-reliance rule.7 The argument which will be made is that while the Supreme Court was right in Patel to overrule Tinsley, the Bowmakers rule is not subject to the same objection of arbitrariness and should not have been tarred with the same brush. It should be stressed that this chapter does not purport to argue in favour of the Bowmakers non-reliance rule;8 its purpose is merely to remove one unjustified criticism.

1 Patel v Mirza [2016] UKSC 42, [2017] AC 467. 2 Bowmakers Ltd v Barnet Instruments Ltd [1945] KB 65 (CA). 3 Patel (n 1) [110] (Lord Toulson). 4 ibid [101], [120] (Lord Toulson). Lord Toulson gave the leading judgment, with which Lady Hale, Lord Kerr, Lord Wilson and Lord Hodge agreed. Lord Neuberger gave a concurring judgment. Lords Mance, Clarke and Sumption disagreed with Lord Toulson’s approach, but not as to the result of the appeal. 5 Gascoigne v Gascoigne [1918] 1 KB 223 (KB). 6 Chettiar v Chettiar [1962] AC 294 (PC). 7 Tinsley v Milligan [1994] 1 AC 340 (HL). 8 The decision itself is trenchantly criticised by CJ Hamson, ‘Illegal Contracts and Limited Interests’ (1949) 10 CLJ 249.

280 William Swadling One question is whether there is any point making this argument post-Patel. The short answer is, yes. First, the ‘range of factors’ approach adopted in Patel may well turn out to be unworkable, in that it forces courts to weigh up incommensurables.9 If that happens, then the non-reliance exception may need to be revisited. Secondly, it is only England and Wales which has adopted the range of factors approach; indeed, it has been expressly rejected in Singapore post-Patel.10 Where the question is still open, therefore, it will be of benefit to know that the Bowmakers non-reliance rule is not as bad as has been painted. The chapter will proceed as follows. It will first set out the non-reliance rule in the common law case of Bowmakers, and show that it is a rule of substantive law and not, as often said, procedure. It will then discuss the equitable rule seen in Gascoigne and Chettiar, asking likewise whether it is a rule of procedure or substance. Like the Bowmakers rule, it is commonly perceived as procedural, though, as such, it operates in an arbitrary fashion. However, it too will be seen to be substantive. Next, it will be shown how a majority of the House of Lords in the leading authority on illegality prior to Patel, Tinsley v Milligan, made two errors: first, in seeing both rules as procedural; and secondly, in seeing the latter as the equitable instantiation of the former, thereby infecting the Bowmakers rule with the arbitrariness of Gascoigne and Chettiar. The conclusion reached will be that the Bowmakers rule has come in for unfair criticism and should not have been so quickly dismissed in Patel.

II. The Bowmakers Rule The Bowmakers non-reliance rule provides that illegality is no defence to claims in private law where the claimant’s pleaded case is complete without the need to rely on any illegal conduct. In other words, it is a rule that collateral illegality is no defence in private law. The rule in fact comes from the much-cited eighteenth-century decision of Holman v Johnson,11 where a contract was made in France between two French residents for the sale and purchase of tea, which the purchaser, to the knowledge of the vendor, intended to smuggle into England. The tea was delivered by the vendor to the purchaser but never paid for. When sued for the purchase price, the purchaser defended the suit on the ground that since the illegal purpose of the contract was known to the vendor, he should be denied the assistance of the court. The argument was rejected on the ground that the making of the contract involved no criminal conduct. Lord Mansfield said: This is an action brought merely for goods sold and delivered at Dunkirk. Where then, or in what respect is the plaintiff guilty of any crime? Is there any law of England transgressed by a person making a complete sale of a parcel of goods at Dunkirk, and giving credit for them? The contract is complete, and nothing is left to be done. The seller, indeed, knows what the buyer is going to do with the goods, but has no concern in the transaction itself. It is not a bargain to be paid in case the vendee should succeed in landing the goods; but the interest of the vendor is totally at an end, and his contract complete by the delivery of the goods at Dunkirk … Therefore, I am clearly of opinion, that the vendors of these goods are not guilty of any offence, nor have they transgressed against the provisions of any Act of Parliament.12

9 The

minority in Patel saw it as amounting to little more than an exercise of judicial discretion. Trading Ltd v Chua Siok Lui (trading as VIE Import & Export) [2018] SGCA 5, [2018] 1 SLR 363. 11 Holman v Johnson (1775) 1 Cowp 341, 98 ER 1120. 12 ibid 344; 1121. 10 Ochroid

Illegality: Pleading, Proof, and Presumptions 281 Why, then, is this the leading case on illegality? It is because of a dictum of Lord Mansfield that the result would have been different had the contract itself been illegal. No court, he said: will lend its aid to a man who founds his cause of action upon an immoral or an illegal act. If, from the plaintiff ’s own stating or otherwise, the cause of action appears to arise ex turpi causa, or the transgression of a positive law of this country, there the court says he has no right to be assisted.13

The prohibition, therefore, was on courts assisting those whose cause of action was founded on illegal or immoral acts: ex turpi causa non oritur actio (from a disgraceful cause, no action arises). Holman v Johnson, however, was not such a case; nor was Bowmakers Ltd v Barnet Instruments Ltd.14 In Bowmakers, the defendants wished to purchase a title to machine tools from a man called Smith. To finance the deal, it was arranged for Smith to sell his title to the claimants, who would then hire the goods to the defendants, the defendants having an option to purchase the title at the end of the hire period. Both the contract of sale between Smith and the claimants and the hire-purchase contract between the claimants and the defendants were unlawful due to wartime regulations, though none of the parties knew this. The tools were delivered by Smith to the defendants, who later sold their (inferior) title to some of the goods to a third party. The defendants also fell into arrears in making the instalment payments, but refused to deliver up the rest of the goods to the claimants when this was demanded. When sued in conversion, the defendants conceded that title to the goods had passed from Smith to the claimants,15 but defended the suit on the ground of illegality, relying on the ex turpi causa dictum of Lord Mansfield in Holman v Johnson.16 The Court of Appeal held that the defence was inapplicable. The claimants were not founding their claim on any illegal or immoral act, but on their admitted title to the goods. Du Parcq LJ, giving the judgment of the court,17 said: The question, then, is whether in the circumstances the plaintiffs are without a remedy. So far as their claim in conversion is concerned, they are not relying on the hiring agreements at all. On the contrary, they are willing to admit for this purpose that they cannot rely on them. They simply say that the machines were their property, and this, we think, cannot be denied … Why then should not the plaintiffs have what is their own? No question of the defendants’ rights arises. They do not, and cannot, pretend to have had any legal right to possession of the goods at the date of the conversion. Their counsel has to rely, not on any alleged right of theirs, but on the requirements of public policy … Prima facie, a man is entitled to his own property, and it is not a general principle of our law (as was suggested) that when one man’s goods have got into another’s possession in consequence of some unlawful dealings between them, the true owner can never be allowed to recover those goods by an action. The necessity of such a principle to the interests and advancement of public policy is certainly not obvious.18

13 ibid 343; 1121. 14 Bowmakers (n 2). 15 Query whether this concession was properly made. It seems that the goods never passed through the claimants’ hands but were delivered directly from Smith to the defendants. The only way title could have passed, therefore, was through s 17 of the Sale of Goods Act 1893, where intent alone is sufficient. However, we then have to question the effect of illegality on the contract. If it rendered the contract void, which does not seem clear, then it was not a contract of sale at all (cf, albeit in a different context, Shogun Finance Ltd v Hudson [2003] UKHL 62, [2004] 1 AC 919), so the Act would have no application. If, however, the contract was not void but merely unenforceable, then it was a contract of sale and title would have passed. Another argument might be that even if the contract was void, there was delivery from the claimants because Smith was acting as their agent. 16 See text to n 13 above. 17 Which also comprised Scott LJ and Uthwatt J. 18 Bowmakers (n 2) 69–70.

282 William Swadling The position would have been different, said du Parcq LJ, if possession of the goods themselves had been illegal, but that was not the case here. Bowmakers was followed and applied by the Privy Council in Singh v Ali,19 where a title to a lorry was bought by the defendant from the British Military Disposals Board. The defendant had agreed to sell the title on to the claimant, but before that was done, and pursuant to an agreement with him, the defendant procured the registration of the lorry in his own name,20 acquiring a haulage permit for the lorry on that basis by falsely representing to the authorities that he would be using it in his own business. The intent, however, was for the claimant to use the lorry in his haulage business. The reason for the subterfuge was that only the defendant was in a position to obtain a permit. The ruse succeeded and the haulage permit was granted to the defendant. Possession of the lorry, however, remained throughout with the claimant, who used it for his own haulage business, which involved an illegality on both sides. After a falling out between them, the defendant seized the lorry and the claimant brought an action for detinue and trespass. The trial judge, Smith J, found the facts alleged by the claimant to be proved by evidence, but held that the parties were ‘fellow conspirators engaged in practising a deceit on the public administration of the country’ and dismissed the claim on the basis of the ex turpi causa rule. The Court of Appeal of the Federation of Malaya applied Bowmakers and allowed the claimant’s appeal, and the Privy Council held that it was right to do so. Title had passed to the claimant, which gave him all he needed to maintain his claim. The fact that both parties were engaged in a conspiracy against the state was irrelevant. Lord Denning, giving the advice of the Board, said: the transferee, having obtained the property, can assert his title to it against all the world, not because he has any merit of his own, but because there is no one who can assert a better title to it. The court does not confiscate the property because of the illegality – it has no power to do so – so it says, in the words of Lord Eldon: ‘Let the estate lie where it falls’; see Muckleston v Brown. This principle was applied by the Court of Appeal recently in Bowmakers Ltd v Barnet Instruments Ltd. The parties to the fraud are, of course, liable to be punished for the part they played in the illegal transaction, but nevertheless the property passes to the transferee.21

Lord Denning’s reference to Muckleston v Brown22 is notable as that case concerned the application of the same rule in equity. As will be seen, Lord Eldon LC there held, differently to the common law, that collateral illegality was a bar in equity.

III. What Sort of Rule is the Bowmakers Rule? What sort of rule is that contained in Bowmakers? Although usually described as ‘procedural’,23 this is a misnomer; it is in fact a rule of substantive law. Substantive law is that part of the law concerned with the rights people have in the world. Examples are the right to bodily integrity, the right to possession of land or goods, and the right to the performance of a contract. Procedural law, by contrast, is concerned with the machinery by which those rights are enforced in court.

19 Singh v Ali [1960] AC 167 (PC). 20 The register was one not of title, but of use. 21 Singh (n 19) 176–77. 22 Muckleston v Brown (1801) 6 Ves Jun 52, 31 ER 934. 23 See, eg Tinsley (HL) (n 7) 374 (Lord Browne-Wilkinson); Nelson v Nelson [1995] HCA 25, (1995) 184 CLR 538, 609 (McHugh J); Patel (n 1) [111] (Lord Toulson).

Illegality: Pleading, Proof, and Presumptions 283 The distinction was well put by Lush LJ, who said in Poyser v Minors24 that procedure was the sum total of rules governing ‘the mode of proceeding by which a legal right is enforced’, while substantive law was that which ‘gives or defines the right, and which by means of the proceeding the Court is to administer’.25 More recently, in the Hong Kong case of Li Tze Cho (No 3) v Ching Hua Co (HK) Ltd, Reece J said that the ‘procedure of the Court is merely the machinery by which the case is brought to a conclusion, the mode of enforcing the legal right’.26 On this basis, the point at issue in each of Holman v Johnson, Bowmakers and Singh v Ali was substantive, for the focus was on the content of the claimant’s cause of action – a phrase said by Brett J in Cooke v Gill to denote ‘every fact which is material to be proved to entitle the plaintiff to succeed, – every fact which the defendant would have a right to traverse’.27 In this context, the question is whether the claimant can make out a good cause of action, in other words, plead sufficient facts, without the need to rely on facts which reveal an illegality. It is therefore a debate about rights, not the machinery for the enforcement of rights. Although it concerns what facts the claimant needs to plead, it has nothing to do with procedure, with what happens at trial. Whether the claimant in Bowmakers, for example, could succeed depended on the constituent elements of the tort of conversion, issues addressed in substantive law textbooks such as Winfield and Jolowicz,28 not procedural ones such as the White Book.29 The Bowmakers non-reliance rule raises no question as to the proof of facts – undoubtedly a question of procedure. The question raised concerns which facts need to be proved in order to make out a cause of action independent of illegality, not how those facts are proved. Holman v Johnson, Bowmakers and Singh v Ali are all decisions at common law. Does the Bowmakers non-reliance rule also apply in equity? This was the issue at the heart of the decision of the House of Lords in Tinsley v Milligan. It turns on how we interpret the rule coming out of another line of cases, that in Gascoigne v Gascoigne and Chettiar v Chettiar, which the majority in Tinsley saw, it will be argued wrongly, as the equitable embodiment of Bowmakers.

IV. The Gascoigne/Chettiar Rule In Gascoigne v Gascoigne,30 the claimant husband alleged that the grant of a lease of land to his wife by a third party was paid for by him, and that his wife had agreed to hold it for him on trust, the reason behind the arrangement being to hide the husband’s interest in the lease from his creditors. When they later parted company, the husband’s demand for a conveyance of the lease to him was refused. He thereupon sued for an order that his wife be compelled to convey the lease to him, relief to which he would normally be entitled under the rule in Saunders v Vautier.31 At trial, the wife relied upon the presumption of advancement, a supposed presumption of gift, but the judge accepted the husband’s oral evidence of the trust arrangement in rebuttal and

24 Poyser v Minors (1881) 7 QBD 329 (CA). 25 ibid 333. 26 Li Tze Cho (No 3) v Ching Hua Co (HK) Ltd (1954) [1961] HKLR 201, 206. See generally W Swadling, ‘Substance and Procedure in Equity’ (2016) 10 Journal of Equity 1. 27 Cooke v Gill (1873) LR 8 CP 107 (CP) 116. 28 J Goudkamp and E Peel, Winfield and Jolowicz on Tort, 19th edn (London, Sweet & Maxwell, 2014). 29 G Vos et al (eds), The White Book Service 2019: Civil Procedure (London, Sweet & Maxwell, 2019). 30 Gascoigne (n 5). 31 Saunders v Vautier (1841) 4 Beav 115, 49 ER 282.

284 William Swadling granted the order sought. The Divisional Court allowed the wife’s appeal. Relying on three cases, Cottington v Fletcher,32 Muckleston v Brown,33 and Davies v Otty (No 2),34 Lawrence and Lush JJ held that the judge was wrong to allow ‘the plaintiff to rebut the presumption which the law raises by setting up his own illegality and fraud, and to obtain relief in equity because he has succeeded in proving it’.35 The same result obtained in the Privy Council decision of Chettiar v Chettiar,36 where a father alleged that he bought a title to 40 acres of land on which there was a rubber plantation. There were government limits on production for those whose holdings exceeded 100 acres, and as the father already held a title to 99 acres, he allegedly transferred his title to the 40-acre parcel to his son on trust for himself, the conveyance falsely stating that the son paid $7,000 in return. Fifteen years later, the father requested a reconveyance of the title so that he could sell it to a third party. On the son’s refusal to do so, the father sought a declaration from the court that the son held the title for him on trust and an order that the son execute a valid and registrable transfer of the title to him. At trial, the son neither appeared nor gave evidence. Smith J, who had been the trial judge in Singh v Ali, found the father’s allegations proved by evidence. Applying the decision of the Court of Appeal in Singh v Ali,37 in which, it will be recalled, that court rejected Smith J’s use of the ex turpi causa rule to bar the conversion claim, Smith J found for the father. Though the father had practised a deceit on the public administration of the country in order to obtain a benefit for himself, Singh v Ali showed, said the judge, that the father’s turpitude was no reason to deny him relief.38 Although Smith J’s decision was upheld by the Court of Appeal, it was once again overturned by the Privy Council. Although Gascoigne was not cited to the court, Lord Denning, giving the opinion of the Board, applied similar reasoning: [T]he plaintiff had of necessity to disclose his own illegality to the court and for this reason: He had not only to get over the fact that the transfer stated that the son paid $7,000 for the land. He had also to get over the presumption of advancement, for, whenever a father transfers property to his son, there is a presumption that he intended it as a gift to his son; and if he wishes to rebut that presumption and to say that his son took as trustee for him, he must prove the trust clearly and distinctly, by evidence properly admissible for the purpose.39

As to the trial judge’s reliance on Singh v Ali, the case was distinguished on the ground that the plaintiff there: founded his claim on his right of property in the lorry and his possession of it. He did not have to found his cause of action on an immoral or illegal act. He was held entitled to recover. But in the present case the father has of necessity to put forward, and indeed, assert, his own fraudulent purpose, which he has fully achieved. He is met therefore by the principle stated long ago by Lord Mansfield ‘No court will lend its aid to a man who founds his cause of action upon an immoral or an illegal act,’ see Holman v Johnson. Their Lordships are of opinion that the courts should not lend their aid to the father to obtain a retransfer from the son.40

32 Cottington

v Fletcher (1740) 2 Atk 155, 26 ER 498. (n 22). 34 Davies v Otty (No 2) (1865) 35 Beav 208, 55 ER 875. 35 Gascoigne (n 5) 226. 36 Chettiar (n 6). 37 The advice of the Privy Council had not at this stage been delivered. 38 Cited in Chettiar (n 6) 300–01 (Lord Denning). 39 ibid 301–02. 40 ibid 303. 33 Muckleston

Illegality: Pleading, Proof, and Presumptions 285

V. What Sort of Rule is the Gascoigne/Chettiar Rule? What sort of rule is being laid down in Gascoigne and Chettiar? At first sight, it is procedural in that it concerns the admissibility of evidence in the rebuttal of presumptions. However, if seen as procedural, the rule produces arbitrary results – with the result of litigation turning on the extraneous issue of the location of the burden of proof at trial. In light of this, it will be argued that, properly understood, the Gascoigne/Chettiar rule can only be substantive: that is, differently to the common law, collateral illegality is a defence in equity. That conclusion is also supported by a review of the authorities relied on in Gascoigne. To understand the argument, we need first to appreciate how presumptions work. Presumptions are part of the law of procedure, an aspect of the law concerned with the proof of facts at trial. There are four methods of proof of facts: evidence; admission; judicial notice; and presumption. Where presumptions apply, their application depends on proof at trial of a certain primary fact or set of facts, either by evidence or (rarely) admission. Once that primary fact or set of facts is proved, the burden of proof of the alleged secondary fact, the subject matter of the presumption, shifts from the person alleging the truth of the alleged fact to the person denying it. The denier must adduce evidence which satisfies the tribunal of fact of the untruth of the allegation. It is, in other words, a burden of disproof, of rebuttal. There are two presumptions in play in the case law on trusts created for illegal purposes: the presumption which triggers the resulting trust and the presumption of advancement. As to the former, the primary facts needed to be proved by evidence or admission so as to raise the presumption, at least in England and Wales,41 are an inter vivos transfer of rights from A to B, where A is not B’s husband or father or person standing in loco parentis to B, or the provision of the purchase money by A to C for a transfer by C to B, again, where A is not B’s husband, father or person standing in loco parentis. As to the presumption of advancement (or gift),42 it arises where, in the above cases, A is either a husband or father of B or a person standing in loco parentis. However, as will be seen below,43 it is doubtful whether the latter presumption exists. A narrow view of the Gascoigne/Chettiar rule would be that it is a rule peculiar to the rebuttal of presumptions, such that, though evidence of illegality is normally admissible in equity, there is a bar on its admission in the context of the rebuttal of presumptions. However, such a view would lead to arbitrary results, with everything now turning on the location of the burden of proof rather than the illegality itself. The result, in other words, would turn on how the facts of the case were to be proved, not on the facts themselves. This can be demonstrated by reversing the position of the parties in Gascoigne. Suppose that it is now the wife suing her husband for a conveyance of the lease. She will allege that she paid the third party to grant the lease to her husband and that her husband agreed to hold it for her on trust. The husband, unlike his wife, will not be given at trial the benefit of the presumption of advancement. Thus, applying this narrow procedural view of the rule, the wife will succeed, for she will not be seeking to rebut any presumption by evidence of illegality. Her iniquity, however, is precisely the same as the husband’s in Gascoigne itself. The Gascoigne/Chettiar rule does not fare any better if we view it as a wider procedural rule, which more generally precludes evidence of illegality in equity. Take again the case of Gascoigne with the parties reversed. Since the wife provided the purchase price for the lease, she will be given the benefit of the other presumption, that which triggers a resulting trust in her favour. Although

41 There

are different rules in Australia and Canada. v Antoni [2007] UKPC 10, [2007] WTLR 1335; Equality Act 2010, s 199. the discussion at text to nn 51–56.

42 Antoni 43 See

286 William Swadling there is debate as to the nature of the fact proved by presumption, the better view is that of Lord Nottingham LC in Cook v Fountain,44 that the person providing the purchase money directed the transferee to hold the rights in question for them on trust. As a consequence, the wife will have no need to lead evidence to make good the second part of her allegation of a declaration of trust and so will not fall foul of a wider procedural version of the rule. Once again, though the iniquity is the same, because it is a wife to husband and not a husband to wife case, a wide procedural version of the rule also generates the opposite result to Gascoigne itself. The same thinking applies to Chettiar, though with a possible difference of result depending on whether a wide or narrow procedural version of the rule is adopted. Suppose it is now the son who alleges a transfer of the title to the rubber plantation to his father on trust to enable the son to hide his interest from the state. The father later refuses to retransfer the title to the son, who seeks an order from the court to compel a reconveyance. For the sake of argument, assume that the litigation takes place in England. There are two possible scenarios that need to be considered. In the first, the son will obtain at trial the benefit of the presumption which triggers the resulting trust. Assuming once again that the fact proved by this presumption is that the conveyance of the rights from son to father was accompanied by a declaration of trust by the son for himself, the son will not have to do what Lord Denning says he cannot and lead evidence of his illegal purpose. He will not, in other words, fall foul of the narrow procedural version of the rule because he is not trying to rebut a presumption. Of course, the father might raise the illegal purpose of the trust in his defence, but unless he is to commit perjury, his evidence will only confirm the allegation of the declaration of trust by his son. The crucial point, however, is that the son will succeed even though his illegality is identical to that of the father in Chettiar itself. Suppose, however, a second scenario, which will happen if the presumption which triggers the resulting trust does not arise because of the operation of section 60(3) of the Law of Property Act 1925,45 which has sometimes been said to have abolished the presumption in voluntary transfer (though not purchase money) cases.46 Now there is no presumption in play at all. It would then be for the son to lead evidence of his agreement with his father that the latter was to hold the title for him on trust. There are two issues here. The first is section 53(1)(b) of the Law of Property Act 1925, assuming it was pleaded,47 which limits the evidence admissible to make good allegations of declarations of trust of titles to land to writing signed by the person able to declare such trusts.48 There might be written evidence, in which case the problem disappears. But even

44 Cook v Fountain (1676) 3 Swans 585, 591; 36 ER 984, 987; W Swadling, ‘Explaining Resulting Trusts’ (2008) 124 LQR 72, 79–80. 45 LPA 1925, s 60(3) provides: ‘In a voluntary conveyance a resulting trust for the grantor shall not be implied merely by reason that the property is not expressed to be conveyed for the use or benefit of the grantee.’ 46 Hodgson v Marks [1971] Ch 892 (CA); Lohia v Lohia [2001] EWCA Civ 1691; Ali v Khan [2002] EWCA 974, [2009] WTLR 187. The better view, as expressed by Chief Master Marsh, sitting as a Deputy High Court judge in National Crime Agency v Dong [2017] EWHC 3116 (Ch), [2018] BPIR 477, is that s 60(3) is only a word-saving provision and has no effect on the incidence of the presumption. However, for the sake of argument, let us assume that it did change the law of procedure. 47 Neither defendant in Gascoigne or Chettiar appears to have taken this point. 48 It has been assumed so far that we are concerned with express trusts, ie ones arising because of a declaration of trust. However, there is an odd comment of Lord Goff in Tinsley (HL) (n 7) 355, that in the case ‘where A puts his property in the name of B in order to conceal his (A’s) interest in it for a fraudulent purpose … it is most unlikely that A will have constituted B an express trustee of the property’. This seems, however, to be based on a misreading of s 53(1)(b) of the LPA 1925, with Lord Goff only saying it was unlikely that the declaration of trust would have been put into writing. The notion that these cases involve express trusts is accepted by many cases, including in the Court of Appeal in Tinsley v Milligan [1992] Ch 310 (CA) 337 (Lloyd LJ), Tribe v Tribe [1996] Ch 107 (CA) and Collier v Collier [2002] EWCA Civ 1095, [2002] BPIR 1057. Though some cases describe the trust as resulting, there is no logical opposition between express and resulting trusts.

Illegality: Pleading, Proof, and Presumptions 287 if there was not, the son’s oral testimony would normally be admissible under the rule in Rochefoucauld v Boustead,49 that a statute designed to prevent fraud cannot be relied upon to commit a fraud.50 The second issue is that, though the son will not obtain the benefit of proof by presumption of a declaration of trust, he will not be leading evidence to rebut any presumption either, so that the narrow procedural version of the rule in Gascoigne and Chettiar would present no bar. However, an application of the wider procedural view would mean that his claim would fail. We have so far assumed that there is such a thing as a presumption of advancement. However, suppose there is not. If there is no presumption to overcome, then a narrow version of the procedural rule, one concerned with the rebuttal of presumptions, would have no purchase. An argument can be made that this is in fact the case, which proceeds as follows. It is undoubtedly true that the presumption which triggers the resulting trust, the presumption of declaration, does not arise where an inter vivos transfer is made from husband to wife, father to child, or person standing in loco parentis to the transferee. In these situations, there is said to operate a counter-presumption, a presumption of ‘advancement’ or gift. However, the High Court of Australia has said there is no such thing. In Martin v Martin, Dixon CJ, McTiernan, Fullager, and Windeyer JJ said: ‘It is called a presumption of advancement but it is rather the absence of any reason for assuming that a trust arose.’51 Likewise, in Calverley v Green, Murphy J described the presumption of advancement as having ‘always been a misuse of the term presumption’ and therefore ‘unnecessary’.52 It is submitted that these views are correct,53 as is demonstrated by the following example. Suppose that in litigation A alleges that he transferred shares to B to hold for him on trust. B denies this. If A proves by evidence both the fact of the transfer and that he and B are siblings, he will obtain the benefit of the presumption which triggers a resulting trust. Because of that, he will have no need to lead evidence of the alleged declaration, which will be proved by presumption. Suppose further that B leads no evidence in rebuttal.54 The result will be that the court must find A’s allegations proved.55 Suppose now an identical version of the allegation and proof by evidence of facts, save that A is not B’s sibling, but his father. In such a case, A will not be given at trial the benefit of the presumption which triggers the resulting trust. As a consequence, if A leads no evidence of the alleged declaration of trust in his favour, his claim will be dismissed. However, this is not because B has the benefit of any presumption operating in his favour, but simply because A bore the burden of proof on all matters relevant to his claim and that burden was not discharged.56 The High Court of Australia was therefore correct when it said that the presumption of advancement is nothing more than a misleading shorthand used to describe situations where the presumption which triggers the resulting trust is absent. Now apply such thinking to Gascoigne and Chettiar. Recall that in Gascoigne, the husband proved by evidence that he paid a third party to grant a lease to his wife, and that his wife agreed to hold it for him on trust. On the assumption that the presumption of advancement is no presumption at all, he would have no need to lead evidence in ‘rebuttal’ since there was no fact 49 Rochefoucauld v Boustead [1897] 1 Ch 196 (CA). 50 We see this doctrine successfully invoked in Davies v Otty (n 34), one of the authorities relied on in Gascoigne (n 30). 51 Martin v Martin [1959] HCA 62, (1959) 110 CLR 297, 303. 52 Calverley v Green [1984] HCA 81, (1984) 155 CLR 242, 265. 53 See further W Swadling, ‘Legislating in Vain’ in A Burrows, D Johnston and R Zimmermann (eds), Judge and Jurist: Essays in Memory of Lord Rodger of Earlsferry (Oxford, Oxford University Press, 2013). 54 As happened in The Venture [1908] P 218 (CA), a case of joint purchase by two brothers of a title to a yacht but that title being vested in one brother alone. 55 ibid. 56 Thinking that there genuinely is a presumption of advancement has led to error. Examples are the decision of the Court of Appeal in Warren v Gurney [1944] 2 All ER 472 (CA) and the UK Parliament’s enactment of s 199 of the Equality Act 2010, wherein the ‘presumption’ of advancement is supposedly abolished.

288 William Swadling proved by presumption to rebut. The same thinking applies to Chettiar. A father alleges a transfer to his son on trust for himself. He leads evidence to prove both transfer and declaration, which the trial judge finds convincing. As there is no such thing as a presumption of advancement, a rule about the type of evidence admissible in rebuttal cannot be relevant. On the basis of a narrow procedural version of the rule, therefore, both cases were wrongly decided if we accept that there is no such thing as a presumption of advancement. Finally, it is not even clear that the husband in Gascoigne and father in Chettiar needed to adduce evidence of their illegality in order to rebut the presumption of advancement, assuming there is such a thing. Recall that in Chettiar, Lord Denning said that the father would have to disclose his own illegality in order to rebut the presumption of advancement.57 This, however, goes too far. In the High Court of Australia decision of Nelson v Nelson, Dawson J said that it was incorrect to say: that a party seeking to rebut the presumption of advancement in the case of a transaction for an illegal purpose is forced to rely upon his own illegality. What must be established in order to rebut the presumption is that no gift was intended. There may be an illegal purpose for the transfer of the property and that may bear upon the question of intention, but it is the absence of any intention to make a gift upon which reliance must be placed to rebut the presumption of advancement. Intention is something different from a reason or motive. The illegal purpose may thus be evidentiary, but it is not the foundation of a claim to rebut the presumption of advancement.58

Thus, all that is needed to rebut the presumption is proof by evidence of facts which tend to prove the untruth of the fact proved by presumption. In Chettiar, proof by evidence by the father that the transfer was on trust for himself would have been enough to rebut the presumption as it is clearly inconsistent with the notion that the father was making a gift to his son. There would be no need to go further, as Lord Denning asserts, and explain why the trust was being created. The same thinking applies to Gascoigne. Cases such as Gascoigne and Chettiar therefore operate in an arbitrary fashion if seen as based on a rule of procedure, whether narrow or wide. That no issue of procedure was involved is further demonstrated by the cases relied on in Gascoigne, none of which are analysable in such terms.

VI. The Gascoigne Authorities Three cases were relied on by the Divisional Court in Gascoigne v Gascoigne: Cottington v Fletcher, Muckleston v Brown, and Davies v Otty (No 2). The most important is Cottington v Fletcher.

A. Cottington v Fletcher In Cottington v Fletcher,59 the claimant granted a 99-year lease of an advowson, the right to present to a vacant benefice, to the defendant, to be held on trust for himself. The reason he did so was that, as a Catholic, there was a possibility of forfeiture of the advowson by statute should a

57 Chettiar 58 Nelson

(n 6) 303. The relevant passage is set out above, in the text to n 39. (n 23) 580. (n 32).

59 Cottington

Illegality: Pleading, Proof, and Presumptions 289 non-conformist make a presentation. He later conformed to the Church of England. Later still, a vacancy arose and the defendant, on the claimant’s instructions, presented a man called Loggin. The claimant then asked the defendant for a reconveyance of the lease of the advowson, which was refused, whereupon he applied to the court for an order that the defendant reconvey, pleading that the transfer was on trust for himself, set up to avoid the penalties of the statute. The defendant admitted both the transfer on trust and the presentment after the claimant conformed to the Church of England, but defended the suit on the ground that the trust was unenforceable for want of writing under section 7 of the Statute of Frauds 1677. Given that section 7 was a rule of evidence rather than enforceability, and that the defendant had admitted the declaration of trust, the plea was obviously bad; Lord Hardwicke LC rightly dismissed the objection and allowed the claim. However, towards the end of his judgment, he said this: If the defendant had demurred to this part of the bill, it might have been of a different consideration. For as this assignment was done in fraud of the law, and merely in order to evade the statute … I doubt at the hearing whether the plaintiff could be relieved, such fraudulent conveyances being made absolute against the grantor.60

This passage is difficult and is open to a number of interpretations. In Nelson v Nelson, Deane and Gummow JJ in the High Court of Australia read it as follows: The Lord Chancellor went on to say that the result might have been different if, rather than having made these admissions, the … defendant had demurred. For it would then simply have appeared from the bill itself that the plaintiff had assigned the advowson in trust for himself in order to avoid the operation of the legislation. The case thus turned upon the operation of statute upon express and resulting trusts.61

Although it is true that Lord Hardwicke does seem to be saying that, on its face, the claimant was pleading his own illegality and that if the defendant had argued there was no case to answer he would have succeeded, it is hard to agree that the case ‘thus turned upon the operation of the statute upon express and resulting trusts’, at least as concerns issues of proof. While it is true that, immediately before the passage cited above, Lord Hardwicke made reference to resulting trusts, it would appear to have been a reference to those arising by operation of law. He said: If the admission and confession by the answer amounts to an admission and confession of a trust for the defendant Loggin as to the first avoidance, the consequence of this is a resulting trust for the plaintiff after the presentment to Loggin is performed … And this is the case upon the statute of Frauds and Perjuries, where the admission of an express trust to one person, is likewise the admission of a resulting trust for another (Loyd v Spillet …).62

At the risk of appearing anachronistic, the resulting trust of which Lord Hardwicke is speaking in this passage would seem to be one arising ‘automatically’ on the failure of a trust, rather than because of the operation of a fact proved by presumption. Lord Hardwicke seems to see the purpose of the trust as being exhausted once the presentment of Loggin was made, with the consequence that there was then an automatic resulting trust for the claimant. This is borne out by the citation of his own decision in Lloyd v Spillet, where he sees there being only two types of resulting trust: those arising where A pays C for the purchase of a right in the name of B and

60 ibid

156; 498. (n 23) 560. (n 32) 156; 498.

61 Nelson

62 Cottington

290 William Swadling those where the trust fails to exhaust the beneficial interest. Lord Hardwicke seems to have been of the view that the presumption triggering a resulting trust arising on a voluntary transfer of real property (of which an advowson was a species) from A to B inter vivos no longer arose after the enactment of the Statute of Frauds 1677. This leads to the conclusion that, since Cottington v Fletcher was not a purchase money case, he can only have been speaking of a failure to exhaust resulting trust.63 As we know, failure to exhaust resulting trusts have nothing to do with matters of proof64 and so cannot have any relevance to a case such as Gascoigne if that case is seen as laying down a procedural rule. This lends support to the notion that the Gascoigne rule can only be substantive.

B. Muckleston v Brown A mistaken turn seems to have been made in the second case relied on in Gascoigne, Muckleston v Brown,65 and it is this which has caused much of the confusion. The facts of the case are not relevant, only Lord Eldon’s interpretation of Cottington v Fletcher, of which he said (obiter): Lord Hardwicke means to say, that if the Defendant admits the trust, though against the policy of the law, he would relieve: but if he does not admit the trust, but demurs, he would do, what does not apply in the least to this case; the Plaintiff stating, he had been guilty of a fraud upon the law, to evade, to disappoint, the provision of the Legislature, to which he is bound to submit, and coming to equity to be relieved against his own act, and the defence being dishonest, between the two species of dishonesty the Court would not act; but would say, ‘Let the estate lie, where it falls.’ That is not this case.66

Under Lord Eldon’s interpretation, everything now turns on the admission of the alleged declaration of trust. Admission is, of course, a method of proof; like the operation of presumptions, it is a form of proof without evidence. But such reasoning would make no sense. Whether a declaration of trust is proved by admission, presumption, or evidence, it is the same trust which in the end is enforced. The better explanation of Lord Hardwicke’s dictum in Cottington v Fletcher, therefore, is the point made above: that he was concerned only with the fact that the defendant did not demur to the bill on the ground of illegality.

C. Davies v Otty (No 2) The final case relied on in Gascoigne is Davies v Otty (No 2),67 where the claimant alleged that he conveyed a title to land to the defendant to hold for him on trust because he feared prosecution for bigamy. When the threat of prosecution passed, he sought a reconveyance, which was refused. At the trial of the action, the defendant, like the defendant in Cottington v Fletcher, relied on section 7 of the Statute of Frauds 1677, there being no written evidence bearing his signature of the agreement to hold on trust. The court, applying the rule later enshrined

63 Campbell points out that the case is more fully reported on this point sub nom Lloyd v Spillit (1740) Barn C 384, 27 ER 689, where Lord Hardwicke reasoned that because the conveyance was one to the grantees to their own use, there could be no presumed resulting trust, so supporting the point made in the text: J Campbell, ‘The Consequences of Rebutting a Presumption of Advancement’ (2018) 46 Australian Bar Review 229, fn 10. 64 Vandervell v IRC [1967] 2 AC 291 (HL). 65 Muckleston (n 22). 66 ibid 68–69; 942. 67 Davies v Otty (n 34).

Illegality: Pleading, Proof, and Presumptions 291 in Rochefoucauld v Boustead,68 held that it would be fraudulent for the defendant to rely on the statute, and admitted the claimant’s oral testimony. As to the illegality, the court held that there was none because the claimant was entitled to think that his wife was dead when he ‘married’ the second woman. However, despite what was said in Gascoigne about the court approving what Lord Eldon said in Muckleston, there was no mention whatsoever of either that case or Cottington v Fletcher. There is thus no case in logic and little in authority to support either narrow or wider procedural versions of the rule. The only possibility, therefore, is that the Gascoigne/Chettiar rule is substantive: in contrast to the position at common law, the rule in equity is that collateral illegality is a defence. It is, in other words, the rule expressed in the maxim, ‘he who comes to equity must do so with clean hands’. Although Lord Goff recognised this in Tinsley in dissent, it was denied on spurious grounds by the majority. It is to that decision which we now turn.

VII. The False Assimilation of the Two Rules in Tinsley v Milligan We saw that the common law rule in Bowmakers is concerned with whether a claimant need rely on their illegality to make out a good cause of action; if not, their illegality presents no bar. We also saw how the Divisional Court in Gascoigne and the Privy Council in Chettiar laid down a rule which on its face appeared to be procedural but can only have been substantive – a rule whereby, once a court becomes apprised of illegality, equitable relief will be denied. In Tinsley,69 the Gascoigne/Chettiar rule was falsely seen by the majority as procedural and, worse, wrongly assimilated to the Bowmakers rule. Ms Tinsley and Ms Milligan conceived a plan to defraud the state of housing benefits. They both contributed to the purchase of a title to land for themselves to occupy but, in order to make it appear that Ms Milligan had no interest in the title, which would have disqualified her from entitlement to benefits, they instructed the vendor to transfer the title to Ms Tinsley alone, though it was agreed between them that Ms Tinsley would hold the title on trust for them both in equal shares. On that basis, Ms Milligan successfully and fraudulently claimed benefits for a number of years. Ms Milligan later repented of her fraud and informed the Department of Social Security of her activities. Later still, Ms Tinsley and Ms Milligan fell out, and Ms Tinsley moved out of the house. She sought an order for possession against Ms Milligan, who counterclaimed for a declaration that Ms Tinsley held the title for them both on trust in equal shares and an order for sale. Ms Tinsley defended the counterclaim on the ground of illegality. The trial judge dismissed Ms Tinsley’s claim for possession and allowed Ms Milligan’s counterclaim. He thought the matter one for his discretion and held that the balance was ‘heavily in favour of allowing the defendant, despite the illegality of some of her proceedings, to succeed in her claim to a moiety of the equity in the house’.70 A majority on the Court of Appeal, Ralph Gibson LJ dissenting, agreed with the result.71 Before that court, the claimant had relied on Gascoigne and Chettiar as authority for the proposition that equity would not lend its aid to

68 Rochefoucauld

(n 49). (HL) (n 7). 70 Cited in Ralph Gibson LJ’s judgment in the Court of Appeal: Tinsley (CA) (n 48) 328. 71 Tinsley (CA) (n 48). 69 Tinsley

292 William Swadling those involved in illegality, that he who comes to equity must do so with clean hands – thus seeing the cases as laying down the substantive rule outlined above. This view was seemingly accepted by all members of the court. However, and somewhat confusingly, Nicholls LJ also saw the clean hands doctrine as the equitable counterpart of the ex turpi causa rule at common law.72 But if that was really so, the case should have presented no difficulty, for Ms Milligan’s claim was simply to enforce a trust of real property, which in itself involves no illegality. This seems to have been realised by Lloyd LJ, who said that the illegality here was collateral and incidental, and that the ex turpi causa rule was not therefore engaged.73 However, Nicholls LJ thought the unclean hands rule prima facie barred Ms Milligan’s claim, though he avoided its application by saying that the rule was not absolute: the test was whether allowing relief would shock the public conscience, which, on the facts, it did not. Gascoigne and Chettiar were distinguished as falling on the wrong side of the line.74 Though the House of Lords was unanimous in rejecting Nicholls LJ’s public conscience gloss, they were divided on whether to allow the counterclaim. Lord Goff,75 in dissent, said that the court should give Ms Milligan no assistance, holding that the Bowmakers rule had no application in equity and relying on Lord Eldon’s dictum in Muckleston v Brown. The rule in equity, as exemplified by Gascoigne, was that once illegality was brought to the attention of the court, whether through pleadings, evidence in chief, or even evidence adduced in cross-examination, the clean hands principle applied and relief would be denied.76 Although Gascoigne concerned the rebuttal of presumptions, other authorities, Lord Goff said, raised no procedural issues at all.77 By contrast, Lord Browne-Wilkinson, with whom Lords Jauncey and Lowry agreed, said it would be anomalous should the Bowmakers rule not apply to both common law and equitable claims. His Lordship managed to find a way of making it apply by holding that the rule laid down in Gascoigne and Chettiar was its equitable embodiment. He clearly saw Gascoigne and Chettiar as concerned with matters of procedure, for he said that the presumption of resulting trust was ‘crucial’ in considering the authorities and that it was on that presumption that the answer to the vital question whether the clamant had to rely on the underlying illegality hinged. Where the presumption of resulting trust applied, said Lord Browne-Wilkinson, the claimant had no need to rely on his illegality. Unless it was a case in which the presumption of advancement applied, the claimant could establish his interest under the trust ‘without relying in any way on the underlying illegal transaction’. As to the case itself, he said: Miss Milligan as defendant simply pleaded the common intention that the property should belong to both of them and that she contributed to the purchase price: she claimed that in consequence the property belonged to them equally. To the same effect was her evidence in chief. Therefore Miss Milligan was not forced to rely on the illegality to prove her equitable interest. Only in the reply and the course of Miss Milligan’s cross-examination did such illegality emerge: it was Miss Tinsley who had to rely on that illegality.78 72 ibid 324. 73 ibid 338. 74 ibid 323–24. 75 With whom Lord Keith agreed. 76 Tinsley (HL) (n 7) 357–58. 77 His Lordship instanced Curtis v Perry (1802) 6 Ves Jun 739, 31 ER 1285; ex parte Yallop (1808) 15 Ves Jun 60, 33 ER 677; Roberts v Roberts (1818) Dan 143, 159 ER 862; Groves v Groves (1829) 3 Y & J 163, 148 ER 1136; Childers v Childers (1857) 3 K & J 310, 69 ER 1126; Re Great Berlin Steamboat Co (1884) 26 Ch D 616 (CA); Crichton v Crichton (1895) 13 R 770; McEvoy v Belfast Banking Co Ltd [1934] NI 67; Re Emery’s Investments Trusts, Emery v Emery [1959] Ch 410 (Ch); Preston v Preston [1960] NZLR 385; Tinker v Tinker [1970] P 136 (CA); Cantor v Cox (1975) 239 EG 121 (Ch). 78 Tinsley (HL) (n 7) 371–72.

Illegality: Pleading, Proof, and Presumptions 293 Though the present case did not concern the operation of the presumption of advancement, if it had, said Lord Browne-Wilkinson, that would have led to a different result: in such a case, the claimant ‘has himself to lead evidence sufficient to rebut the presumption of gift and in so doing will normally have to plead, and give evidence of, the underlying illegal purpose’.79 Lord Eldon’s bar on the enforcement of the equitable claim in Muckleston v Brown was therefore, in Lord Browne-Wilkinson’s view, not as absolute as expressed. It was subject to a non-reliance exception: [T]he explanation for this departure from Lord Eldon’s absolute rule is that the fusion of law and equity has led the courts to adopt a single rule (applicable both at law and in equity) as to the circumstances in which the court will enforce property interests acquired in pursuance of an illegal transaction, viz, the Bowmakers rule … A party to an illegality can recover by virtue of a legal or equitable property interest if, but only if, he can establish his title without relying on his own illegality. In cases where the presumption of advancement applies, the plaintiff is faced with the presumption of gift and therefore cannot claim under a resulting trust unless and until he has rebutted that presumption of gift: for those purposes the plaintiff does have to rely on the underlying illegality and therefore fails.80

Lord Browne-Wilkinson said this position was ‘well illustrated’ by two decisions of the Privy Council, Singh v Ali and Chettiar. In his view, the Privy Council was: applying exactly the same principle in both cases although in one case the plaintiff ’s claim rested on a legal title and in the other on an equitable title. The claim based on the equitable title did not fail simply because the plaintiff was a party to the illegal transaction; it only failed because the plaintiff was bound to disclose and rely upon his own illegal purpose in order to rebut the presumption of advancement. The Privy Council was plainly treating the principle applicable both at law and in equity as being that a man can recover property provided that he is not forced to rely on his own illegality.81

While it is easy to sympathise with Lord Browne-Wilkinson’s attempt to align the common law and equitable positions and to correct the anomaly he identifies of different rules operating in the two jurisdictions, it is unfortunate that he does so by praying in aid the decisions of Gascoigne and Chettiar – for now the Bowmakers rule is tarred with the arbitrariness objection seen above, of which Gascoigne and Chettiar are obvious examples. Indeed, the arbitrary operation of Lord Browne-Wilkinson’s rule was soon noticed. Thus, in the High Court of Australia’s decision the following year in Nelson v Nelson,82 Dawson J said that the difference in results the application of the presumptions generated was ‘entirely fortuitous being dependent upon the relationship between the parties and is wholly unjustifiable upon any policy ground’,83 while McHugh J described the results as ‘essentially random’.84 Likewise, Toohey J said it amounted to a ‘triumph of procedure over substance’85 and that allowing ‘the result in such a situation to be determined by the procedural aspects of a claim for relief is at odds with the broad considerations necessarily involved in questions of public policy’.86 Tinsley was accordingly rejected in Australia.

79 ibid

372–73. 375. 81 ibid 376. 82 Nelson (n 23). 83 ibid 580. 84 ibid 609. 85 ibid 592–93. 86 ibid 597. 80 ibid

294 William Swadling Similarly, in the English case of Tribe v Tribe,87 where a father transferred shares to his son to be held on trust for himself, the purpose of the trust being to deceive creditors, Millett LJ said that the rule emerging from Tinsley: does not conform to any discernible moral principle. It is procedural in nature and depends on the adventitious location of the burden of proof in any given case. Had the plaintiff transferred the shares to a stranger or distant relative whom he trusted, albeit for the same dishonest purpose, it cannot be doubted that he would have succeeded in his claim.88

Millett LJ was, however, able to find in favour of the father by the invocation of a further exception, that of the locus poenitentiae, which he held applied despite the fact that there was no repentance by the father, but simply because, on the facts, no creditors had been deceived.89 Similar criticisms of Lord Browne-Wilkinson’s speech are present in Patel v Mirza itself, where both majority and minority describe the Tinsley rule as ‘arbitrary’.90 This was because, said Lord Sumption, it made the illegality principle depend on ‘adventitious procedural matters, such as the rules of pleading, the incidence of the burden of proof and the various equitable presumptions’. The equitable presumptions, he said, ‘operate wholly procedurally, and have nothing to do with the principle which the court is applying in illegality cases’.91

VIII. The Gascoigne/Chettiar and Bowmakers Rules Distinguished It is important to understand, if it is not already clear, that despite Lord Browne-Wilkinson’s assimilation of the two rules in Tinsley, the procedural version of the Gascoigne/Chettiar rule he adopts is not the same as the rule laid down in Bowmakers. Take the case of Gascoigne, being sure to distinguish issues of pleading (substance) and proof (procedure). For the husband to have made out a good cause of action to collapse the trust under the rule in Saunders v Vautier, he would have had to plead four things: the agreement between himself and his wife that the lease would be held by her for him on trust; the grant of the lease to the wife by the third party; a demand for conveyance by the husband; and a refusal by the wife to do so.92 He would have had no need to explain why the trust was being set up. Applying the Bowmakers rule, the husband’s illegality should not have barred his claim. Yet he lost, the reason being that his collateral illegality caused his claim to be dismissed. Likewise, in Chettiar, all the father needed to plead to make out a good cause of action was a transfer of title to the rubber plantation to his son, an accompanying declaration of trust for

87 Tribe (n 48). 88 ibid 134 (Millett LJ); see also ibid 118 (Nourse LJ). 89 ibid 135. 90 eg in Patel (n 1), Lord Toulson cited with approval the view of the English Law Commission (expressed in Illegal Transactions: The Effect of Illegality on Contracts and Trusts (Law Com CP No 154, 1999); The Illegality Defence (Law Com No 320, 2010)) that the rule applied in Tinsley was arbitrary in that the question whether the illegality affected the recognition or enforcement of the trust depended not on the merits of the parties, nor the policies underlying the illegality defence, but on a procedural issue: ibid [110]. Moreover, the effect of applying the reliance principle in cases involving the presumption of advancement gave that presumption an overriding importance which it was never intended to have: ibid [24]. 91 Patel (n 1) [237]. 92 There is no need for the husband to plead that he paid for the lease unless he wants to trigger the presumption, which is in any case unavailable to him.

Illegality: Pleading, Proof, and Presumptions 295 himself, a demand for reconveyance, and a refusal by the son to do so. Once again, there was no need for the father to mention in his pleadings why he was setting up a trust for himself, and thus no need to plead the illegality. Applying Bowmakers, he too should have won. The reason he lost, as in Gascoigne, had nothing to do with whether his pleadings disclosed a good cause of action without reliance on the illegality; it was because of the collateral illegality. Lord Browne-Wilkinson was therefore wrong to equate the Bowmakers rule with that in Gascoigne and Chettiar.

IX. Conclusion It is vital in the law of illegality to keep separate rules of substance and procedure. Both the rule in Bowmakers and that in Gascoigne and Chettiar are substantive, not procedural. When seen in that light, they are clearly not the same, but opposites. The real issue in Tinsley, obscured in Lord Browne-Wilkinson’s speech though not in that of Lord Goff, was whether the equitable rule should be changed to align it with that at common law. Lord Goff thought it a matter for consideration by the Law Commission, but Lord Browne-Wilkinson tried to effect the change himself. However, by confusing issues of substance and procedure – and, more particularly, issues of pleading, proof and presumption – he did it such a way as to make the law more arbitrary than before. Everything now turned on the location of the burden of proof, whether the claim was at law or in equity. In Lord Sumption’s dissenting judgment in Patel, he said of the cases concerned with trusts set up for illegal purposes that the application of the illegality principle should depend on what facts the court must be satisfied about in order to find an intention giving rise to an equitable interest and not how those facts were established.93 This has been questioned by Burrows, who asked: ‘On what logical basis can one simply ignore the standard law on presumptions?’94 This criticism is misconceived. There is no lack of logic in what Lord Sumption says; his Lordship is not ignoring the standard law on presumptions, merely drawing a perfectly logical and muchneeded distinction between pleading and proof.

93 Patel (n 1) [238]. 94 A Burrows, ‘A New Dawn for the Law of Illegality’ in S Green and A Bogg (eds), Illegality after Patel v Mirza (Oxford, Hart Publishing, 2018) 31–32.

296

17 Negotiating Damages ANDREW BURROWS*

I. Introduction In this chapter, I want to focus on a contract case which has important ramifications for restitution for wrongs. This was the Supreme Court decision in Morris-Garner v One Step (Support) Ltd,1 which dealt with what have traditionally been most commonly called ‘Wrotham Park damages’, but which the Supreme Court has said are best called ‘negotiating damages’. Whatever the label – and I will from here on call them ‘negotiating damages’ – these are damages for a civil wrong, whether a tort, breach of contract or an equitable wrong such as breach of confidence, that are assessed according to the price which, prior to the wrong, the claimant could reasonably have charged the defendant for releasing the defendant from the duty that has been broken. To give an example, take the case of Wrotham Park Estate Ltd v Parkside Homes Ltd2 itself. Houses had been built in breach of a restrictive covenant. Brightman J refused to grant a mandatory injunction ordering the houses to be pulled down given the social waste that that would have involved. But he did make an award of substantial damages. If compensatory damages were assessed in the normal way, by considering what the claimant’s position would have been had there been no breach of covenant – ie had the houses not been built – the damages would have been nominal only: the diminution in value of the claimant’s land caused by the breach of covenant was zero. Nevertheless, substantial damages were awarded by Brightman J by applying the approach of assessing damages according to the price which the claimant could reasonably have demanded for releasing the defendant from the restrictive covenant had the defendant approached the claimant prior to building the houses. In Wrotham Park, the damages were actually given in substitution for the mandatory injunction under what was Lord Cairns’ Act 1858 and is now section 50 of the Senior Courts Act 1981, so that they were what has been termed ‘equitable damages’, rather than standard common law damages. But while, as I shall explain later, there may be a particular reason for awarding

* This chapter draws heavily on the relevant chapter in my book, Remedies for Torts, Breach of Contract, and Equitable Wrongs, 4th edn (Oxford, Oxford University Press, 2019), and I would like to thank Oxford University Press for its permission to use much of that chapter here. An early version of this chapter was presented at the Centre for Commercial Law, University of Edinburgh, on 16 October 2018, and I would like to thank Parker Hood, who organised that seminar, the chair, Lord Glennie, and the participants for making me feel so welcome and for their helpful comments. 1 Morris-Garner v One Step (Support) Ltd [2018] UKSC 20, [2019] AC 649. 2 Wrotham Park Estate Ltd v Parkside Homes Ltd [1974] 1 WLR 798 (Ch).

298 Andrew Burrows negotiating damages where the damages are equitable damages of this sort, no one is suggesting that negotiating damages cannot be awarded at common law or can only be awarded in lieu of an injunction. Indeed, there have been many cases subsequent to Wrotham Park in which damages (whether equitable or common law) have been assessed in the same way for a civil wrong.3 Moreover, there were earlier cases in which damages were assessed according to what was considered a reasonable fee for the wrongful use of property (whether real or personal property and including intellectual property). These cases were effectively taking the same approach to the assessment of damages as was subsequently taken in Wrotham Park.4 Although often referred to as ‘user damages’ or as damages awarded according to the ‘user principle’, they can now all be labelled ‘negotiating damages’. Having clarified the type of damages we are looking at, the principal puzzles revolve around two main linked uncertainties. First, what is the purpose of negotiating damages? This has spawned a large academic literature.5 Are they compensatory (ie covering the claimant’s loss)? Or is it more realistic to regard negotiating damages as restitutionary (ie reversing gains made by the defendant) rather than compensating loss? Or do they support Robert Stevens’s thesis – his so-called ‘substitutive damages’ thesis – that the basic measure of damages is always the value of the right infringed, and compensatory and restitutionary damages are merely dealing with consequential losses and gains if any?6 Secondly, when can such damages be awarded (ie what is their availability)? 3 eg cases on tortious trespass to land (Bracewell v Appleby [1975] Ch 408 (Ch); Jaggard v Sawyer [1995] 1 WLR 269 (CA); Eaton Mansions (Westminster) Ltd v Stinger Compania de Inversion SA [2013] EWCA Civ 1308, [2014] HLR 4); the tort of nuisance by infringement of rights to light (Carr-Saunders v Dick McNeil Associates Ltd [1986] 2 All ER 888 (Ch); Tamares (Vincent Square) Ltd v Fairpoint Properties (Vincent Square) Ltd (No 2) [2007] EWHC 212 (Ch), [2007] 1 WLR 2167); infringement of intellectual property rights (General Tire & Rubber Co Ltd v Firestone Tyre & Rubber Co Ltd [1975] 1 WLR 819 (HL) (patent); 32Red plc v WHG (International) Ltd [2013] EWHC 815 (Ch) (trade mark)); breach of restrictive covenant affecting land (Amec Developments Ltd v Jury’s Hotel Management (UK) Ltd (2001) 82 P & CR 22 (Ch)); breach of contractual obligation restricting use of master tapes (Experience Hendrix LLC v PPX Enterprises Inc [2003] EWCA Civ 323, [2003] 1 All ER (Comm) 830); landlord’s breach of covenant of quiet enjoyment (Lunn Poly Ltd v Liverpool & Lancashire Properties Ltd [2006] EWCA Civ 430, [2006] 2 EGLR 29); breach of obligations in joint venture (Pell Frishmann Engineering Ltd v Bow Valley Iran Ltd [2009] UKPC 45, [2011] 1 WLR 2370); breach of obligation to allow driving time (Giedo van der Garde BV v Force India Formula One Team Ltd (formerly Spyker F1 Team Ltd (England)) [2010] EWHC 2373 (QB) [499]–[559], esp [505]–[507]; and breach of contractual and/or equitable obligations of confidence (Force India Formula One Team Ltd v 1 Malaysia Racing Team Sdn Bhd [2013] EWCA Civ 780, [2013] RPC 36; MVF 3 APS (formerly Vestergaard Frandsen A/S) v Bestnet Europe Ltd [2016] EWCA Civ 541, [2017] FSR 5; Marathon Asset Management LLP v Seddon [2017] EWHC 300 (Comm), [2017] ICR 791). 4 Whitwham v Westminster Brymbo, Coal & Coke Co [1896] 2 Ch 538 (CA) and Penarth Dock Engineering Co Ltd v Pounds [1963] 1 Lloyd’s Rep 359 (QB) (both trespass to land by, respectively, tipping spoil on to the claimant’s land and refusing to remove a floating pontoon from the claimant’s dock); Meters Ltd v Metropolitan Gas Meters Ltd (1911) 28 RPC 157 (CA) 164–65 (patent infringement); Watson, Laidlaw & Co Ltd v Pott, Cassels & Williamson (1914) 31 RPC 104 (HL) (patent infringement); Strand Electric Engineering Co Ltd v Brisford Entertainments Ltd [1952] 2 QB 246 (CA) (tort of detinue by retaining and using the claimant’s theatre equipment); Seager v Copydex Ltd (No 2) [1969] 1 WLR 809 (CA) (breach of confidence). 5 eg R Sharpe and S Waddams, ‘Damages for the Lost Opportunity to Bargain’ (1982) 2 OJLS 290; M McInnes, ‘Gain, Loss and the User Principle’ [2006] Restitution Law Review 76; A Burrows, ‘Are “Damages on the Wrotham Park Basis” Compensatory, Restitutionary, or Neither?’ in D Saidov and R Cunnington (eds), Contract Damages (Oxford, Hart Publishing, 2008); C Rotherham, ‘The Conceptual Structure of Restitution for Wrongs’ [2007] CLJ 172; C Rotherham, ‘Wrotham Park Damages and Account of Profits: Compensation or Restitution?’ [2008] Lloyd’s Maritime and Commercial Law Quarterly 25; C Rotherham, ‘Gain-Based Relief in Tort after Attorney General v Blake’ (2010) 126 LQR 102; K Barker, ‘“Damages Without Loss”: Can Hohfeld Help?’ (2014) 34 OJLS 631; K Low, ‘The User Principle: Roshomon Effect or Much Ado about Nothing?’ (2016) 28 Singapore Academy of Law Journal 984; J Edelman, Gain-Based Damages (Oxford, Hart Publishing, 2002); R Stevens, Torts and Rights (Oxford, Oxford University Press, 2007) ch 4; K Barnett, Accounting for Profit for Breach of Contract (Oxford, Hart Publishing, 2012). 6 See Stevens (n 5) ch 4.

Negotiating Damages 299 In Morris-Garner v One Step (Support) Ltd, the Supreme Court sought to answer both questions in the context of a claim for breach of contract.

II. Morris-Garner v One Step (Support) Ltd The claimant, One Step (Support) Ltd, was in the business of providing support for young people leaving care. Karen Morris-Garner, the first defendant, was a 50 per cent shareholder and director of that company and the second defendant, Andrea Morris-Garner, was its manager. When both left the company, and the first defendant sold her shares in it, each of them entered into three-year covenants with the company promising not to compete with it or solicit the company’s clients and to keep certain information about the company confidential. The defendants had in fact already set up a competing company, Positive Living Ltd, of which they were the sole shareholders. After the defendants left One Step, Positive Living started to trade in competition with One Step. Through that trading, the defendants were acting in breach of their non-competition and non-solicitation covenants, and the first defendant was acting in breach of the confidentiality clause. Positive Living was very successful and was sold by the defendants, some three years after starting trading, for £12.8 million. One Step, meanwhile, suffered a significant downturn in its business. In the claimant’s action for breach of contract, it was estimated by the claimant’s expert that the loss to the claimant’s business (ie the loss of profits) caused by the defendants’ breach of the non-competition covenant – and, incidentally, the other covenants – was between £3.4 million and £4.6 million. But the claimant argued that it was entitled to ‘Wrotham Park damages’, with the hypothetical release fee from the covenants being estimated by the claimant’s expert at between £5.6 million and £6.3 million. The claimant was therefore seeking ‘negotiating damages’ in a situation where compensatory damages assessed in the ordinary way would yield a very substantial sum of damages. And they were doing that because negotiating damages would give them an even higher sum. Why was that? There are several possible reasons. For example, applying a conventional approach, the loss of profits might have been difficult to prove, especially given the need to take into account loss of goodwill. But, in any event, negotiating damages may yield more because at the time just before breach – when it is assumed that the hypothetical bargain would have been struck – it might reasonably have been assumed that the claimant’s losses from the competition, or the defendant’s gains from being able to compete, would be far higher than they in fact turned out to be, so that the reasonable release fee would be higher than loss as conventionally assessed. While at first instance and in the Court of Appeal the claimant succeeded in its argument that it was entitled to elect for negotiating damages, rather than normal compensatory damages, that argument was unanimously rejected by the Supreme Court. In remitting the assessment of the damages back to the trial judge, the Supreme Court held that negotiating damages were not here available as an alternative measure of damages – albeit that a hypothetical release fee might be used as evidence in calculating ordinary compensatory damages for the claimant’s loss of business. Lord Reed, with whom Lady Hale, Lord Wilson and Lord Carnwath agreed, gave the leading judgment, on which I will focus. Lord Sumption gave a separate judgment concurring with the result and adopting broadly similar reasoning to Lord Reed. The only other separate judgment was given by Lord Carnwath, who explained, inter alia, why he preferred the judgment of Lord Reed to that of Lord Sumption.

300 Andrew Burrows In relation to our two puzzles, Lord Reed held as follows. The purpose of negotiating damages is to compensate the claimant for a loss. They are not restitutionary and they are not a non-compensatory measure reflecting the value of the right infringed. As regards availability, Lord Reed identified two categories of case in which negotiating damages can be awarded. First, where the damages are being given in substitution for an injunction or specific performance under section 50 of the Senior Courts Act 1981; and secondly, where the wrong in question constitutes the infringement of a property right or a closely analogous right. On the facts of Morris-Garner, neither category was in play. Hence negotiating damages could not be awarded. I now want to look in more depth at each of those two issues.

III. The Purpose of Negotiating Damages In Morris-Garner, Lord Reed expressly or impliedly rejected both the restitutionary analysis, that these damages are really concerned to strip profit, and Robert Stevens’s analysis that these damages are not compensatory but are concerned with the value of the right infringed. Speaking of negotiating damages for wrongful use of property, Lord Reed said that, although the loss is ‘not loss of a conventional kind’, ‘a compensatory analysis need not be regarded as strained or artificial’.7 This was because ‘the person who makes wrongful use of [another’s] property prevents the owner from exercising his right to obtain the economic value of the use in question, and should therefore compensate him for the consequent loss’.8 Extending this more broadly to cases of breach of contract, such as the breach of a restrictive covenant over land or breach of a contractual right of confidence, Lord Reed said: ‘The claimant has in substance been deprived of a valuable asset, and his loss can therefore be measured by determining the economic value of the asset in question.’9 In viewing the damages as compensatory, we can therefore see that Lord Reed effectively says that the relevant loss is the loss of the right to obtain value or the loss of the value of the right/asset of which the claimant has been deprived. This most directly equates to the ‘loss of opportunity to bargain’ analysis put forward by Sharpe and Waddams in a famous article in the 1982 volume of the Oxford Journal of Legal Studies.10 Prior to Morris-Garner, many commentators argued that it was often fictional to regard the claimant as having suffered a loss of bargaining opportunity (or a loss of the value of a right) because the claimant would never have agreed to bargain away its rights or at least would not have done so at the reasonable fee fixed.11 In other words, the view being taken was that while compensation for a loss of bargaining opportunity was sometimes a realistic analysis of negotiating damages, often it was not; in truth, what the courts were often doing in awarding negotiating damages was stripping away some of the gains made by the defendant rather than compensating the claimant. In support of that approach, one could point to explicit reasoning in many negotiating damages cases where the courts accepted that they were effecting restitution of gain rather than compensating loss.

7 Morris-Garner (n 1) [30], [66]. 8 ibid [30]. 9 ibid [92]. See, similarly, ibid [95] point (10): ‘Negotiating damages can be awarded for breach of contract where the loss suffered by the claimant is appropriately measured by reference to the economic value of the right which has been breached, considered as an asset.’ 10 Sharpe and Waddams (n 5). 11 See, eg the articles by Burrows and Rotherham, and the books by Edelman and Barnett, cited in n 5 above.

Negotiating Damages 301 For example, in Strand Electric Engineering Co Ltd v Brisford Entertainments Ltd,12 which concerned a claim in the tort of detinue for the retention and use of theatre equipment, and in Penarth Dock Engineering Co Ltd v Pounds,13 which concerned trespass to land by a floating pontoon, Lord Denning said that the damages were concerned not with what the claimants had lost, but with the benefit the defendants had gained from their wrongful use of the claimant’s goods or land. In Ministry of Defence v Ashman,14 in which a tenant had wrongfully ignored a notice to quit Royal Air Force accommodation because she and her children had nowhere else to go, a majority of the Court of Appeal (Kennedy and Hoffmann LJJ) accepted that the claimant landlord was entitled to restitutionary damages for the trespass. The Penarth Dock case was relied on, and it was held that the damages should be assessed according to what it would have cost the tenant to rent alternative local authority accommodation had any been available. Hoffmann LJ said: A person entitled to possession of land can make a claim against a person who has been in occupation without his consent on two alternative bases. The first is for the loss which he has suffered in consequence of the defendant’s trespass. This is the normal measure of damages in the law of tort. The second is the value of the benefit which the occupier has received. This is a claim for restitution. The two bases of claim are mutually exclusive and the plaintiff must elect before judgment which of them he wishes to pursue. These principles are not only fair but, as Kennedy LJ demonstrated, also well established by authority. It is true that in earlier cases it has not been expressly stated that a claim for mesne profit for trespass can be a claim for restitution. Nowadays I do not see why we should not call a spade a spade. In this case the Ministry of Defence elected for the restitutionary remedy.15

In Attorney-General v Blake,16 the House of Lords laid down that an account of profits can be awarded, in exceptional circumstances, to strip a contract-breaker of profits. An account of profits is indisputably a restitutionary remedy rather than a compensatory remedy; but the important point here is that the reasoning of Lord Nicholls appeared to support a restitutionary analysis of tort ‘user damages’ and of the damages awarded in Wrotham Park as a bridge to the account of profits awarded in that case. Following on from Blake, it was accepted that an award of ‘negotiating damages’ could be made on a restitutionary basis in Experience Hendrix LLC v PPX Enterprises Inc.17 There had been a contractual settlement in 1973 of a dispute between the rock star Jimi Hendrix and the defendant record company. By the terms of that settlement, it was agreed that certain master tapes could be used for recording purposes by the defendant, but that the rest should be delivered up to Jimi Hendrix. In breach of that contract, the defendant used master tapes that should have been delivered up. The claimant, representing the estate of Jimi Hendrix, sought damages (or an account of profits) for that breach of contract. The claimant was held entitled to damages based not on compensating the claimant’s loss, but on what was a reasonable sum taking into account the gains

12 Strand Electric Engineering Co Ltd v Brisford Entertainments Ltd [1952] 2 QB 246 (CA). 13 Penarth Dock Engineering Co Ltd v Pounds [1963] 1 Lloyd’s Rep 359 (QB). 14 Ministry of Defence v Ashman (1993) 66 P & CR 195 (CA). See also Ministry of Defence v Thompson [1993] 2 EGLR 107 (CA); Gondal v Dillon Newsagents Ltd [1998] NPC 127 (CA); Ramzan v Brookwide Ltd [2011] EWCA Civ 985, [2012] 1 All ER 903 [67]. See generally E Cooke, ‘Trespass, Mesne Profits and Restitution’ (1994) 110 LQR 420. cf Inverugie Investments Ltd v Hackett [1995] 1 WLR 713 (PC). 15 Ashman (n 14) 200–01. 16 A-G v Blake [2001] 1 AC 268 (HL). 17 Experience Hendrix LLC v PPX Enterprises Inc [2003] EWCA Civ 323, [2003] 1 All ER (Comm) 830. For a helpful case note, see M Graham, ‘Restitutionary Damages: The Anvil Struck’ (2004) 120 LQR 26.

302 Andrew Burrows made by the defendant from its use of the forbidden master tapes. Peter Gibson LJ’s judgment, taking a restitutionary analysis, was particularly clear. He said: In my judgment, because (1) there has been a deliberate breach by PPX of its contractual obligations for its own reward, (2) the claimant would have difficulty in establishing financial loss therefrom, and (3) the claimant has a legitimate interest in preventing PPX’s profit-making activity carried out in breach of PPX’s contractual obligations, the present case is a suitable one (as envisaged by Lord Nicholls) in which damages for breach of contract may be measured by the benefits gained by the wrongdoer from the beach. To avoid injustice I would require PPX to make a reasonable payment in respect of the benefit it has gained.18

But, as I interpret Lord Reed’s judgment in Morris-Garner, he has now said that all of that is incorrect. Certainly, in relation to breach of contract, he explicitly rejected the idea that, rather than being compensatory, negotiating damages are restitutionary, designed to strip the defendant of some or all of the wrongdoer’s gains. In his Lordship’s view, it is incorrect to think that there is a sliding scale by which an account of profits and negotiating damages are both gain-based, with negotiating damages stripping away a proportion, rather than the whole, of the defendant’s profits, depending on, for example, whether the breach of contract was deliberate or whether the claimant had a ‘legitimate interest’ in preventing the profit-making activity.19 This is a denial at the highest level that negotiating damages are restitutionary at least in the sense – which is the most realistic alternative to any compensatory analysis – that they are concerned to strip away a proportion of the profit obtained by the wrongdoer.20 It follows from this that it was irrelevant in Morris-Garner that the breach of the no-competition clause was cynical and was a breach by which the defendant made substantial profits, which they would now keep. According to the Supreme Court, we are concerned with the claimant’s loss, not the defendant’s cynically acquired gains. I should say, to balance up the references to authority, that Lord Reed is not the first influential English judge to favour a compensatory analysis and to reject a restitutionary analysis. For example, in Jaggard v Sawyer,21 where the facts concerned a continuing trespass to land and breach of covenant by using the claimant’s road as a right of way, Sir Thomas Bingham MR and Millett LJ were presented with obiter dicta of Steyn LJ, who had said the following in Surrey County Council v Bredero Homes Ltd: ‘The object of the award in the Wrotham Park case was not to compensate the plaintiff for financial injury, but to deprive the defendants of an unjustly acquired gain.’22 18 ibid [58]. See also Marathon Asset Management LLP v Seddon [2017] EWHC 300 (Comm), [2017] ICR 791, in which Leggatt J, in a tightly reasoned judgment in which he surveyed most of the relevant authorities, concluded that a compensatory or a restitutionary analysis might be realistic, depending on the facts. The case concerned breach of a contractual (or equitable) duty of confidence by electronically copying some files. The claimants were denied substantial damages and were instead awarded nominal damages. The claimant had not suffered a loss (because, for example, it would not have bargained away the right in question) and the defendant had not made a gain (because it had not used the electronic files and had certainly made no profits, nor would it make any profits, from using them). Importantly, Leggatt J regarded many of the past cases on ‘negotiating damages’ as having awarded restitutionary damages that strip a part of the defendant’s profits. But on the facts of the instant case, ‘negotiating damages’ should not be awarded as they would neither compensate for loss nor strip gain. 19 Morris-Garner (n 1) [90]. 20 There are a few passages that might be interpreted as indicating that Lord Reed has not entirely shut the door on a restitutionary analysis. So his emphasis in Morris-Garner (n 1) [30], [92] on ‘taking something for nothing’ might be thought to indicate that it is important that the defendant has gained rather than that the claimant has lost. Again, at [79], Lord Reed accepted that in the wrongful use cases one might say that negotiating damages have been measured by the benefit gained by the wrongdoer provided the benefit was taken to be the objective value of the wrongful use: but, he said, ‘The courts did not … adopt a benefits-based approach, but conceived of the awards as compensating for loss’. 21 Jaggard v Sawyer [1995] 1 WLR 269 (CA). See also, eg Severn Trent Water Ltd v Barnes [2004] EWCA Civ 570, [2004] 2 EGLR 95. 22 Surrey CC v Bredero Homes Ltd [1993] 1 WLR 1361 (CA) 1369.

Negotiating Damages 303 Sir Thomas Bingham MR and Millett LJ explicitly rejected that analysis. Wrotham Park damages were compensatory not restitutionary, and the relevance of the reference to the profits made by the defendant was only that it helped to fix the reasonable fee that the claimant would reasonably have charged for the release of the defendant from its obligation. According to Millett LJ: there is no reason why compensatory damages for future trespasses and continuing breaches of covenant should not reflect the value of the rights which she has lost, or why such damages should not be measured by the amount which she could reasonably have expected to receive for their release.23

One might also say, in favour of a compensatory analysis, that if one is unwilling to recognise that there has been a loss where the claimant would not have been willing to release the rights, one appears to reach a paradoxical result: the more the claimant values the right infringed – such that the claimant would never have been willing to release the right at any price, let alone a reasonable price – the less one would say the claimant has lost (and the lower the compensatory damages would become). It might also perhaps be argued that it is irrelevant whether or not in reality the claimant would have bargained away its rights. The fact that it could have done so, and that it was reasonable to have done so, is sufficient. After all, if we take the analogy of the permanent deprivation of the claimant’s property – as where the claimant’s goods have been stolen and disposed of and the claimant brings an action in the tort of conversion – we would assess damages according to the market value of the goods irrespective of whether it is realistic to think that the claimant would ever have sold the goods or was intending to replace them. An alternative possible way of meeting the argument that a loss of bargaining opportunity is often fictional is to think of the counter-factual that we have in mind when considering the claimant’s loss. The relevant counter-factual that we apply in a conventional assessment of compensatory damages is the position the claimant would have been in if the defendant had not carried out the relevant wrongful conduct. But plainly that counter-factual is unrealistic in the sense that the defendant has precisely carried out that wrongful conduct. So, in relation to negotiating damages, one might say that all one is doing is positing a different counter-factual, based on the claimant’s opportunity to bargain away its legal rights, and that it is equally irrelevant whether or not that counter-factual is realistic. Drawing these threads together, the apparent rejection of a restitutionary analysis is disappointing: as a matter of authority and in principle, a restitutionary analysis can be seen as providing a useful alternative measure of damages in many situations. Nevertheless, there are certainly plausible arguments to justify the Supreme Court’s compensatory analysis based in effect on a loss of bargaining opportunity.

IV. The Availability of Negotiating Damages In Morris-Garner, Lord Reed can be interpreted as having identified two wide-ranging categories of case in which negotiating damages have been awarded and are appropriate. The first is where ‘equitable damages’ are being given in substitution for an injunction or specific performance, as in Wrotham Park itself. Lord Reed argued that the refusal to grant specific relief produces the same effect as if the claimant had given up its right so that measuring

23 Jaggard

(n 21) 291.

304 Andrew Burrows damages by what the claimant would have reasonably charged for giving up that right is particularly apt: The rationale is that, since the withholding of specific relief has the same practical effect as requiring the claimant to permit the infringement of his rights, his loss can be measured by reference to the economic value of such permission.24

However, all compensatory damages, whether common law or equitable, are in one sense a monetary substitute for the rights infringed. It is therefore not immediately obvious why one should be carving out negotiating damages as being especially apt where they are awarded as a substitute for specific relief. The approach which links negotiating damages to damages awarded in substitution for specific relief may also mean, oddly, that the availability of negotiating damages may turn on the timing of the commencement of proceedings. Imagine that in Morris-Garner, One Step had commenced proceedings seeking an injunction to restrain the breach of contract prior to the expiry of the three-year covenants. Even if the case came to court after the expiry of the three-year period, so that an injunction would not be granted because it would come too late to have any effect, it would still appear that negotiating damages could have been awarded in substitution for the injunction because, at the commencement of the proceedings, there was jurisdiction to grant an injunction. Indeed, although Lord Reed indicated some doubts about this,25 Lord Walker said in Pell Frischmann Engineering Co Ltd v Bow Valley Iran Ltd26 that the court would have jurisdiction to award damages in substitution for an injunction, and hence to award negotiating damages, even if an injunction had not been sought. My inclination, therefore, is to think that it is misconceived to regard negotiating damages as more appropriate just because the damages are equitable rather than common law. However, although not articulated in this way by Lord Reed, there is a powerful argument that where equitable damages are being given as compensatory damages for a future or anticipated wrong, rather than for a wrong that has already been committed, then negotiating damages, in contrast to compensatory damages assessed in the normal way, provide a very helpful technique for avoiding the need to speculate as to the future. Trying to put the claimant into as a good a position as it would have been in had a future or anticipated wrong not occurred is doubly speculative: both the wrong and the losses lie in the future. In contrast, it may be thought much easier to assess a reasonable release fee so that there will be no future or anticipated wrong. On the facts of Morris-Garner, this first category, whether or not narrowed in the way I have just explained, was not in play. The damages were common law, not equitable, and the wrong was in the past, not the future. Plainly, therefore, the facts fell outside Lord Reed’s first category. The second category of case identified by Lord Reed comprises what, for shorthand (although his Lordship did not use this terminology), can be labelled ‘interference with proprietary rights and closely analogous rights’. This covers proprietary and closely analogous rights, whether created by operation of law or by contract. Lord Reed’s judgment indicated that, prior to their more widespread use in breach of contract cases after Blake,27 the main application of negotiating damages was in three types of case: (i) wrongful use of another’s real or personal property (the ‘user’ cases in tort); (ii) wrongful interference with intellectual property rights, whether by a tort

24 Morris-Garner

(n 1) [95], point (4). (n 1) [45]. 26 Pell Frischmann Engineering Co Ltd v Bow Valley Iran Ltd [2009] UKPC 45, [2011] 1 WLR 2370. 27 A-G v Blake (n 16). 25 Morris-Garner

Negotiating Damages 305 or by the equitable wrong of breach of confidence; and (iii) breach of a restrictive covenant over land, as in Wrotham Park itself. Lord Reed appeared to regard those three types of case as being relatively stable and defensible, but that that stability had been undermined by the reasoning in Blake: ‘in Blake the wider availability of such awards was signalled, but the seeds of uncertainty were sown’.28 Accordingly, while not challenging the decision in Blake as to the award of an account of profits, Lord Reed appeared to wish, in respect of common law negotiating damages, to turn the clock largely back to the position prior to Blake. Perhaps because it is not easy in principle to see why one should be looking just at proprietary rights – indeed, the definition of what is a proprietary right is far from straightforward, as shown by the debates over the proper characterisation of a breach of confidence – Lord Reed went on to include closely analogous rights. He envisaged that such damages were appropriate provided ‘the breach of contract results in the loss of a valuable asset created or protected by the right which was infringed’;29 or, expressed in a slightly different way, provided the contractual right is of such a kind that its breach can result in an identifiable loss equivalent to the economic value of the right, considered as an asset, even in the absence of any pecuniary losses which are measurable in the ordinary way.30

Lord Reed used that reasoning to justify the award of negotiating damages in, for example, Experience Hendrix,31 where, in breach of contract, the defendant had licensed the use of certain master tapes of Jimi Hendrix which were in its possession. Lord Reed also indicated that the award of negotiating damages in Pell Frischmann32 might be justified on this basis because the relevant contractual right in a joint venture agreement – the right to participate in a business opportunity – was ‘treated … as a commercially valuable asset, of which the claimant had been effectively deprived’.33 However, the limits of this idea become problematic. If Pell Frischmann was correctly decided, why could one not say on the facts of Morris-Garner that the purpose of the non-competition covenant was to protect the claimant against the loss of business profit, including goodwill, and hence to protect against ‘the loss of a valuable asset … protected by the right which was infringed’? Indeed, if one just focused on the goodwill, surely one could have said that the no-competition clause was designed to protect that valuable asset. Again, why could one not say that the noncompetition covenant was a ‘contractual right of such a kind that its breach could result in identifiable loss equivalent to the economic value of the right’? Yet Lord Reed held that this kind of right fell on the wrong side of the line for an award of negotiating damages. The upshot is that, in relation to this second category, it is very hard to see why, applying the explanatory logic of Lord Reed’s reasoning, negotiating damages should not always be available where the right in question protects an economic rather than a personal right and why, indeed, negotiating damages were not available on the facts of Morris-Garner itself. The Supreme Court has sought to draw a distinction between some economic rights (those that are or are close to proprietary rights) and others. Ultimately, that line-drawing appears to be based more on pragmatism than principle, and to rest on a desire not to allow negotiating damages to be available at the claimant’s election for most instances of breach of contract. But it is hard to understand where the precise line is being drawn.

28 Morris-Garner

(n 1) [48]. [92]. 30 ibid [93]. 31 Experience Hendrix (n 17). 32 Pell Frischman (n 26). 33 Morris-Garner (n 1) [83]. 29 ibid

306 Andrew Burrows I wonder whether it might help to consider whether damages assessed in the conventional way would be inadequate in the sense that the claimant has a non-financial interest in upholding the right infringed that would not be reflected in a conventional award of damages. For example, in Wrotham Park,34 ordinary damages based simply on the diminution of value of the claimant’s land were inadequate because they did not reflect the claimant’s environmental interest in maintaining the land as it was without the buildings. It should be noted that this ties in with the first of Lord Sumption’s three categories of case35 where negotiating damages may be awarded, namely where the claimant has an ‘interest extending beyond financial reparation’. Applying this to Morris-Garner, there would be no difficulty in explaining why ‘negotiating damages’ were unavailable. Damages assessed in the normal way would be adequate because the claimant’s interest in relation to the no-competition clause was purely financial. A final wrinkle with Lord Reed’s judgment is that he accepted – and Lord Sumption made a lot of this point in his judgment – that a reasonable release fee could be used in Morris-Garner as evidence in assessing the claimant’s loss of profit. In other words, a reasonable release fee could be used, and it would appear can always be used by a judge if thought helpful, not as an alternative measure to the conventional measure of compensation, but as an evidential technique or tool, as Lord Sumption expressed it, in applying the conventional measure. This distinction between evidential tool and measure is not straightforward, but in practice what it means is that a claimant should not knowingly be put in a better position, by using the reasonable release fee evidence, than the conventional measure would dictate. In short, one cannot recover more than the loss of profit, known to have been suffered, by claiming a higher reasonable release fee. One sees here a parallel with the relationship between the expectation loss measure and a claimant’s reliance losses.36 In Morris-Garner, the claimant had precisely sought negotiating damages as an alternative measure, and not as an evidential tool in assessing loss of profit, with a view to recovering a higher sum as negotiating damages than the loss of profit recoverable by applying a conventional measure. It was that which was impermissible. Finally, it is of interest that in Turf Club Auto Emporium Pte Ltd v Yeo Boong Hua37 the h ighest Singaporean court, the Singapore Court of Appeal, has also been grappling with the purpose and justification of ‘Wrotham Park damages’ in the context of the breach of a settlement agreement of a commercial dispute. While agreeing with the Supreme Court that the purpose of those damages is compensatory not restitutionary, the Singaporean Court of Appeal was strongly critical of Lord Reed’s approach to the availability of such damages and regarded the ‘property or analogous rights’ category as too uncertain. The Singaporean Court of Appeal’s preference was, in part,38 to focus on there being a ‘remedial lacuna’, which is similar to my above suggestion that the inadequacy of normal compensatory damages might be a helpful guide to the availability of these damages.

34 Wrotham Park (n 2). 35 Lord Sumption’s other two categories were ‘damages in lieu of an injunction’ and where the ‘notional release fee is [evidence of the claimant’s] pecuniary loss’: Morris-Garner (n 1) [110]–[123]. 36 Omak Maritime Ltd v Mamola Challenger Shipping Co (The Mamola Challenger) [2010] EWHC 2026 (Comm), [2011] 1 Lloyd’s Rep 47. See A Burrows, Remedies for Torts, Breach of Contract, and Equitable Wrongs, 4th edn (Oxford, Oxford University Press, 2019) ch 6. 37 Turf Club Auto Emporium Pte Ltd v Yeo Boong Hua [2018] SGCA 44. 38 Two other criteria for the availability of ‘Wrotham Park damages’ were laid down in addition to there being a ‘remedial lacuna’: (i) that, in general, there has been the breach of a negative, not a positive, covenant (but it is hard to see the force of this); and (ii) that the hypothetical bargain would not be illegal.

Negotiating Damages 307

V. Conclusion No one would suggest that these issues are easy. Some aspects of the relevant law have been clarified by the Supreme Court, including what we should call these damages and that they have a compensatory purpose. As regards their availability, I have suggested that there is a particular reason for their award – ie avoiding a doubly speculative normal assessment of loss – where they are substituting for an injunction to prevent a future or anticipated wrong. We also know that they are certainly available for proprietary wrongs and for breach of confidence, but, beyond that, the position remains unclear. I have suggested that it might help to consider whether damages assessed in the conventional way would be inadequate in the sense that the claimant has a nonfinancial interest in upholding the right infringed that would not be reflected in the usual award of compensatory damages; but there is admittedly no support for this proposal in the reasoning of Lord Reed. While, in some respects, Morris-Garner represents a step forward, it remains to be seen whether it is a case of one step forward, two steps back.

308

18 The Myth of Common Law Tracing SARAH WORTHINGTON

I. Introduction The focus of this chapter is on common law tracing. This is an intriguing area for lawyers, especially those with a commercial bent. In this small part of the legal landscape there is widespread support for discarding the common law rules of tracing and adopting wholesale the more interventionist and flexible equitable rules of tracing. The ‘equitable’ tracing rules would simply become ‘the’ tracing rules. The merits and likely success of this mooted development are the concern of this chapter.1 The whole subject of tracing can provide endless fascination. Tracing is the key element in a strategy designed by the common law (including, and primarily coming from, equity) not only to protect proprietary interests, but also to prolong and sometimes even enlarge the proprietary status of an individual’s original rights. As Lord Millett explained extrajudicially, One of the rules of our law of property, common to both equity and the common law, is that the owner of a thing can claim ownership of its traceable proceeds. The rule is not a rule of natural law. It is not universal. We do not have to have such a rule. We choose to have it. Most civilian systems do not.2

This is a powerful proposition. Taken at face value, and focusing on the common law aspects, it maintains that it is an immutable aspect of our property law that the legal owner of an asset (A1) can claim legal ownership of its traceable proceeds (A2 or A3, etc), and can do so whether or not the substitutions or exchanges delivering those traceable proceeds are effected by the original owner of the asset or by someone else, whether authorised or not.3

1 Indeed, some would say the development has taken place already: see Foskett v McKeown [2001] 1 AC 102 (HL) 113 (Lord Steyn), 128–29 (Lord Millett), both obiter, both asserting that only one set of tracing rules should be used both at common law and in equity. See also Trustee of the Property of FC Jones & Sons v Jones [1997] Ch 159 (CA) 169; FHR European Ventures LLP v Cedar Capital Partners LLC [2014] UKSC 45, [2015] AC 250 [44]. In the same vein, see Bracken Partners Ltd v Gutteridge [2003] EWHC 1064 (Ch), [2003] 2 BCLC 84 [31] (Peter Leaver QC); Barros Mattos Junior v MacDaniels Ltd [2005] EWHC 1323 (Ch), [2005] ILPr 45 [135] (Lawrence Collins QC). Affirming the continuing distinction between common law and equity, see Shalson v Russo [2003] EWHC 1637 (Ch), [2005] Ch 281 [103]–[104] (Rimer J); London Allied Holdings Ltd v Lee [2007] EWHC 2061 (Ch) [256]–[257] (Etherton J). 2 P Millett, ‘Proprietary Restitution’ in S Degeling and J Edelman (eds), Equity in Commercial Law (Sydney, NSW, Law Book Co, 2005) 314. For Millett LJ’s and Lord Millett’s earlier judicial reinforcement, see FC Jones (n 1); Foskett (n 1). 3 FC Jones (n 1), throughout the judgment, but see especially 170. However, this seems to be the only case to go as far as holding that the original owner acquires legal title to the substitute asset. The analogy with the rights of an equitable owner, who can typically claim equitable ownership of the traceable proceeds, is obvious: see Foskett (n 1). The language of ‘equitable ownership’ is common, although it can encourage analogies with ‘common law ownership’ that are not

310 Sarah Worthington That makes the advantages of tracing very clear. When it is possible to trace in equity, a claimant who can trace her original property interest into its substitute assets obtains continuing insolvency protection against the defendant’s general creditors, she can claim the windfall profits inherent in any lucrative investments,4 she can follow her original property or its traceable proceeds into the hands of third parties and she may be able to assert claims against them. Moreover, in identifying those traceable proceeds, the claimant can rely on presumptions to enable her to trace through mixed substitutions, thus preserving value even when the evidence cannot unequivocally support the view taken.5 In this chapter, an argument is presented for doubting the application of this rule to legal owners. Indeed, elsewhere, I have questioned the rule’s unqualified application to those with equitable interests.6 To make that case, the focus is on the simplest factual scenarios, and the principles that would necessarily be needed to support common law tracing in such cases. If the arguments for these simplest cases cannot be made out, then there is little reason to suppose that the more complicated cases will prove amenable to such analysis. The problems in this area are compounded by lack of secure authorities on common law tracing. It follows that much of the argument is made from principle rather than precedent, and moreover from principles that are aligned with the modern approach to equitable tracing.

II. Leading Cases The two leading English cases on tracing are Trustee of the Property of FC Jones v Jones7 and Foskett v McKeown,8 the first concerning tracing at common law, the second tracing in equity. Lord Millett gave the leading judgments in both cases, as Millett LJ in the Court of Appeal in the first case and as Lord Millett in the House of Lords in the second. The power of tracing can be seen in dramatic form in Foskett v McKeown.9 The claimants were beneficiaries of a commercial trust set up to speculate in property. In breach of trust, the trustee used at least £40,000 of trust funds to pay a number of premiums on a life insurance policy he had taken out on his own life. Later he committed suicide, and the payout under the policy was £1 million. The beneficiaries successfully claimed a pro rata share of these proceeds. In short, they were held to be entitled to the secondary profits derived from the successful but unauthorised

necessarily apt. It is preferable to think of beneficiaries as having interests, not ‘ownership’, and their interests give them an entitlement to the economic benefit, or the economic value, that can be derived from the assets held on trust. See the discussions in S Worthington, ‘Revolutions in Personal Property: Redrawing the Common Law’s Conceptual Map’ in S Worthington, A Robertson and G Virgo (eds), Revolution and Evolution in Private Law (Oxford, Hart Publishing, 2018); S Worthington, ‘The Commercial Triple Helix: Contract, Property and Unjust Enrichment’ in P Devonshire and R Havelock (eds), The Impact of Equity and Restitution in Commerce (London, Bloomsbury Publishing, 2019). 4 And if the investments were not successful, she could instead claim against the trustee for equitable compensation for the loss caused to the fund by the trustee’s breaches: see AIB Group (UK) Plc v Mark Redler & Co Solicitors [2014] UKSC 58, [2015] AC 1503. 5 eg presumptions favouring the victim over the wrongdoer. 6 See especially S Worthington, ‘Justifying Claims to Secondary Profits’ in EJH Schrage (ed), Unjust Enrichment and the Law of Contract (London, Kluwer Law International, 2001) 451; S Worthington, Equity, 2nd edn (Oxford, Oxford University Press, 2006) ch 4. 7 FC Jones (n 1). 8 Foskett (n 1). 9 ibid.

The Myth of Common Law Tracing 311 investment of funds which they owned in equity. The court held it to be immaterial that the trustee had settled the benefit of the life insurance policy on his children. That settlement meant that the parties in dispute were the innocent beneficiaries of the commercial trust and the innocent beneficiaries of the family trust. However, the latter beneficiaries, being mere donees, could be in no better position than the donor, the defaulting trustee. The court adopted a property justification for its conclusion, a justification in line with the assertion recorded at the start of this chapter. As Lord Millett put it, This is a textbook example of tracing through mixed substitutions … The transmission of a claimant’s property rights from one asset to its traceable proceeds is part of our law of property, not of the law of unjust enrichment. There is no ‘unjust factor’ to justify restitution (unless ‘want of title’ be one, which makes the point). The claimant succeeds if at all by virtue of his own title, not to reverse unjust enrichment. Property rights are determined by fixed rules and settled principles. They are not discretionary. They do not depend upon ideas of what is ‘fair, just and reasonable’. Such concepts, which in reality mask decisions of legal policy, have no place in the law of property.10

For present purposes, the point to note here is the potential parallels with common law tracing. What needs to be considered is whether those parallels are justified.

III. The Basics of Tracing A great deal has been written on tracing.11 What follows is an outline of the specific issues that might inform the common law focus adopted in this chapter. Given that (equity’s) tracing and claiming techniques were first developed to deal with defaulting trustees, it is perhaps predictable that the trust model came to define their essential prerequisites. Seventy years ago, the orthodox view was that a claimant could ‘trace’ only if she could show that she was both the equitable owner of the original asset in the defendant’s hands and the defendant owed her fiduciary obligations in dealing with that asset.12 Furthermore, little if any distinction was ever drawn between tracing and claiming; indeed, it was common simply to refer to the claimant’s ‘tracing claim’. Now the distinction is clearly recognised, and slippage in terminology is seen as indicative of sloppy analysis.13 ‘Tracing’ is a distinct technique for identifying new assets that can legitimately be regarded as the exchange products or value substitutes for the claimant’s old asset. Especially with modern eyes, the particular logic and intuitive legal sense of this focus on value substitutes is readily apparent in the context of equitable tracing. As we shall see, it is not so apt in the context of common law tracing. Consider a simple scenario where a car (A1) is held on trust for X but then exchanged without authority by her trustee, T, for a horse (or for £2,000, it matters little – A2), and the horse is 10 ibid 126–27 (Lord Millett). Lord Millett is publicly credited with persuading Lord Browne-Wilkinson to change his mind (see 108), thus turning the 3:2 outcome of the case around in favour of the commercial beneficiaries. 11 Beyond the standard texts on aspects of common law tracing see in particular: M Scott, ‘The Right to “Trace” at Common Law’ (1966) 22 University of Western Australia Law Review 463; R Goode, ‘The Right to Trace and Its Impact in Commercial Transactions Parts I and II’ (1976) 92 LQR 360, 525; F Babafemi, ‘Tracing Assets: A Case for the Fusion of Common Law and Equity in English Law’ (1971) 34 MLR 12; RA Pearce, ‘A Tracing Paper’ [1976] Conveyancer and Property Lawyer 277; S Kurshid and P Matthews, ‘Tracing Confusion’ (1979) 95 LQR 78; P Birks, ‘The Necessity of a Unitary Law of Tracing’ in R Cranston (ed), Making Commercial Law: Essays in Honour of Roy Goode (Oxford, C larendon Press, 1997); NH Andrews and J Beatson, ‘Common Law Tracing: Springboard or Swansong’ (1997) 113 LQR 21; LD Smith, The Law of Tracing (Oxford, Clarendon Press, 1997). 12 Re Diplock [1948] Ch 465 (CA); affirmed sub nom Ministry of Health v Simpson [1951] AC 251 (HL). 13 The clear distinction emerged in Smith (n 11). This monograph predated FC Jones (n 1).

312 Sarah Worthington then exchanged for a parcel of shares now worth £10,000 (A3). An assertion that T holds the car on trust for X, or that X is the equitable owner of the car, is simply an assertion that X (not T) is entitled to all the economic benefits that can be derived from the car: in short, X is entitled to its economic value. T must hold the car for that purpose, even though T is the legal owner and usually has the right to possession unless the trust deed specifically gives it to X. Given that T is obliged to hold the car for X’s economic benefit,14 there is an intelligent logic in X’s search for an alternative asset to stand in its place when T disposes of the car without authority. The equally understandable step is to demand as that asset whatever property T acquired in exchange for the car. On the simple facts here, the answer is easy: the clean substitute for the car is the horse (or the £2,000), and the clean substitute for the horse (or the £2,000) is the shares.15 The facts are rarely so accommodating, although even when there has been a mixing of assets from different sources, the underlying objective of protecting X’s interest in the economic benefits that can be derived from an asset remains as described here. It might be noted even at this early stage that there is no easy equivalent to this justificatory argument in the context of common law tracing, where the outcome is assertion of legal title to the substitute asset. Returning to equitable tracing, the measure of that protection is now seen as delivered by way of a separate, distinctive second step, which is ‘claiming’. Modern orthodoxy holds that this second step is separate from the first step, the tracing step, since the tracing step neither indicates nor delivers any particular outcome. Notice, though, that this sharp distinction between tracing and claiming sits rather oddly with the bald assertion set out at the start of this chapter that ‘the owner of a thing can claim ownership of its traceable proceeds’.16 That assertion is of an automatic right to claim whatever is traced: the impression given is that this transactional transmission of rights from one asset to another lies ‘in the very nature of things’.17 That is clearly not the case,18 even though, empirically, ‘ownership of the traceable proceeds’ is indeed the typical outcome in equitable tracing cases. A beneficiary tracing into asset exchanges in the hands of her defaulting trustee is routinely assumed to be entitled to claim equitable ownership of the last asset at the end of the tracing chain, and little effort is ever expended in explaining why that might be so.19 This encourages an all too easy assumption that tracing plus claiming simply allows the claimant’s rights to be transposed to any traceable property. Indeed, in Foskett v McKeown, that conclusion seemed to be positively endorsed: Tracing … is merely the process by which a claimant [does various things, including] justifies his claim that the proceeds can properly be regarded as representing his property … [Tracing] enables the claimant to substitute the traceable proceeds for the original asset as the subject matter of his claim. But it does not affect or establish his claim … He will normally be able to maintain the same claim to the substituted asset as he could have maintained to the original asset.20

14 This simple description of equitable ownership as the distinctive right to the economic benefit of an asset seems to have great utility. It also allows for useful comparisons between equitable ownership and other derivative property interests in assets, such as rights to possession and security interests. See n 3 above. 15 With the critical assumption that the relevant assets are identifiable as delivering specific contributions in the chain of substitutions, and that – in the simplest of cases – these are provable as ‘clean’ (unmixed) substitutions. 16 See n 2 above. 17 C Hare, ‘Tracing Value and the Value of Tracing’ (2013) 24 Journal of Banking and Finance Law and Practice 249, 266. 18 T Cutts, ‘Tracing, Value and Transactions’ (2016) 79 MLR 381. 19 The leading cases roundly confirm this robust rule of a distinction between tracing and claiming, but their analyses routinely gloss over the second step and its separate justifications: see Foskett (n 1); FC Jones (n 1). 20 Foskett (n 1) 127–28 (Lord Millett), emphasis added. See also Rimer J in Shalson (n 1) [102], who explained tracing as ‘the process by which a claimant seeks to show that an interest he had in an asset has become represented by an interest in a different asset’.

The Myth of Common Law Tracing 313 This accurately describes the outcome with our defaulting trustee: if T disposed of the trust assets for the horse (or the £2,000), then X’s original equitable ownership of those assets (the car) would be maintained, after tracing, as equitable ownership of the horse (or the £2,000), and still later as equitable ownership of the further substituted shares. For decades, little attention was given to the question of why the claimant was entitled to these windfall benefits inherent in the claims to traceable substitutes.21 The cases typically concerned fiduciaries, and there was scant discussion of the rationale for tracing into the substitute assets in their hands and claiming an interest in them. The leading case on equitable tracing, Foskett v McKeown,22 was such a case, and there the House of Lords made it plain that tracing is simply an incident of English property law.23 This is considered below in the context of common law tracing.24

IV. Early Common Law Tracing Approaches The common law approach is asserted to be similar, at least so far as clean substitutions are concerned. This conclusion seems to be based on a literal reading of the words of Lord Ellenborough in Taylor v Plumer,25 notwithstanding that this case has now been exposed as a case relating to equitable title and, more particularly, to claims to equitable title as against a defaulting fiduciary.26 With this leading case on common law tracing in tatters as a precedent, it follows that the historical truth may well be that there never was common law tracing, but it was asserted in several cases and so now exists,27 as Millett LJ explained must be the case in FC Jones.28 What Lord Ellenborough said in Taylor v Plumer was: The … property of a principal entrusted by him to his factor for any special purpose belongs to the principal, notwithstanding any change which that property may have undergone in point of form, so long as such property is capable of being identified, and distinguished from all other property … It makes no difference in reason or in law into what other form different from the original, the change may have been made … for the product of or substitute for the original thing still follows the nature of the thing itself, as long as it can be ascertained to be such, and the right only ceases when the means of ascertainment fail, which is the case when the subject is turned into money, and mixed and confounded in a general mass of the same description …29

Taking this as an accurate description of common law tracing rather than equitable tracing, the assertion is that the product or substitute ‘belongs to the principal’, just as the original property

21 This was always the framework for the claim to traceable proceeds in equity, since, if the substitutes were not more valuable, then the claimant would ignore the tracing possibilities and simply claim equitable compensation for losses to the managed fund caused by her fiduciary’s unauthorised dealings: AIB Group v Mark Redler (n 4). 22 Foskett (n 1). 23 See the criticisms in A Burrows, ‘Proprietary Restitution: Unmasking Unjust Enrichment’ (2001) 117 LQR 412 (criticised on the basis that this is an assertion, not a reason; what is needed is an explanation of why this is an incident of property). See also P Birks, ‘Overview: Tracing, Claiming and Defences’ in P Birks (ed), The Frontiers of Liability, vol 2 (Oxford, Oxford University Press, 1994). 24 See section V below. 25 Taylor v Plumer (1815) 3 M & S 562, 105 ER 721. 26 Kurshid and Matthews (n 11); L Smith, ‘Tracing in Taylor v Plumer: Equity in the Court of King’s Bench’ [1995] Lloyd’s Maritime and Commercial Law Quarterly 240; Smith (n 11) 168. 27 Birks (n 11), 246–47. 28 FC Jones (n 1) 169. 29 Taylor (n 25) 575; 726.

314 Sarah Worthington did, provided only that the means of ascertainment does not fail. This accords with the assertion set out at the start of this chapter. Moreover, this ‘transmission’ of rights from original asset to substitute does not seem to require any positive decision by the claimant, whether by ratification or by election of remedies. This appears to be confirmed by FC Jones,30 despite the apparent inconsistency with the pointed practical problems identified by Lord Goff in Lipkin Gorman.31 This power of common law tracing is, however, subject to severe practical constraints when the subject matter is mixed and confounded in a general mass, so the picture often drawn of the different capabilities of common law and equitable tracing is of ‘that poor mutt, the common lawyer, being able to grasp the identity of specific coins, but retiring mouth agape in baffled amazement once they are mixed with other coins’.32 It is these practical limitations that are behind the call for the common law to leave behind its own rules and adopt equity’s more sophisticated tracing rules. But notice too that Lord Ellenborough’s words are rather vague, and, until FC Jones, the older cases asserting this possibility of common law tracing never concluded with an outright assertion that the claimant had legal title to the asset at the end of the tracing/following exercise.33 There may have been descriptions of the location of ‘property’, as in Taylor v Plumer itself, but the end result was typically a claim for money had and received, a claim which is now seen as distinct from a proprietary claim, and, indeed, dependent not on the claimant retaining legal title, but on the defendant acquiring legal title to the asset in question, and thus acquiring an enrichment, albeit one regarded as an unjust enrichment given the context. These early cases all emerged in an era where unjust enrichment was not yet a principle adopted by the courts34 and the jurisdiction was only slowly emerging. Put more generally, even where the language may appear significant, these older cases may not be describing what we are now looking for. The language used was very loose, yet this was done with no hint of analytical error or sloppiness. As noted earlier in the equitable context, ‘following’ and ‘claiming’ were used without discrimination (eg ‘follow’ the ‘money’ cases), and commentators such as Scott saw no harm in using ‘tracing’ to describe both ‘tracing hand to hand’ and ‘tracing into substitutions’.35

V. The Modern Common Law Authority: FC Jones That description of the position at common law might be said to have ended with Trustee of the Property of FC Jones v Jones.36 The facts in FC Jones are analogous to the illustration set out earlier, but concern cash and investments in potato futures, not cars, horses and shares. Secondly, and potentially importantly, the person effecting the substitutions was not a thief, but an innocent third party. The defendant, Mrs Jones, received approximately £11,000 from her husband. He paid this sum using cheques drawn on his partnership. Mrs Jones invested these funds successfully on 30 FC Jones (n 1). See also section V below. 31 Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL) 573, Lord Goff held that the claimants could trace or follow their ‘property into its product’ since this ‘involves a decision by the owner of the original property to assert his title to the product in place of his original property’ (emphasis added). In particular, the interest could not have arisen automatically, otherwise the defendants would have been liable in conversion. 32 Scott (n 11) 470. 33 eg Taylor (n 25); Banque Belge pour L’Etranger v Hambrouck [1921] 1 KB 321 (CA). Contrast FC Jones (n 1). 34 That occurred in Lipkin Gorman (n 31). 35 Scott (n 11) 463–64. 36 FC Jones (n 1).

The Myth of Common Law Tracing 315 the potato futures market, so the proceeds of the investment amounted to more than £50,000. Shortly afterwards, however, the partnership went into bankruptcy. Because of the doctrine of relation back, the trustee in bankruptcy was held to have legal title to the funds that had been paid out by the husband to his wife, so he could confer no title on her,37 and she was therefore utilising assets in which she ‘never acquired any title’38 in order to make her investments. Having adopted this view of where title lay, the Court of Appeal in FC Jones did not then ‘follow’ the partnership assets, now legally belonging to the trustee in bankruptcy, into the hands of any parties receiving them in order to recover them for the benefit of the estate. This is what earlier cases had done. In particular, this is what had been done by the Court of Appeal in Re Gunsbourg,39 the case on which Millett LJ placed great reliance in determining the consequences of relation back. In that case, furniture that had been transferred by the bankrupt was followed into third party hands, much as might be done with assets stolen by a thief and then sold on to innocent third parties who nevertheless would obtain no title, given the constraints of nemo dat. Those assets, or their value, could then be recovered for the benefit of the estate. Mrs Jones was quite prepared for this claim, and accepted that in the circumstances she would need to repay the original £11,000 received from her husband, or its value, plus interest. But the Court of Appeal in FC Jones went further – much further – holding that the trustee in bankruptcy’s legal title to the chose in action representing the money in the partnership bank account entitled the trustee to claim legal title to the cheques transferred to Mrs Jones by her husband, then to the choses in action which represented Mrs Jones’s various dealings with those cheques and, finally, to the chose in action representing the £50,000 proceeds.40 No authority was cited in support of this ability of the trustee in bankruptcy to trace into the proceeds and claim legal title to them, rather than merely follow the original assets affected by relation back. Indeed, the difference between tracing and following was not stressed. This blurring of the two different processes is all too common with ‘money’, where it can be tempting to suggest that the £11,000 asset identified as the partnership’s chose in action (by relation back) is then ‘located … first, in the defendant’s [the wife’s] account with the commodity brokers and, later, in the defendant’s account at Raphaels’.41 This might suggest ‘the money’ is like ‘the red car’, initially located in the partnership’s hands, then in the commodity brokers’, then in Raphaels’s, and then finally brought into court for a determination of what ought to be done with it. But this would be wrong. The ‘money’ is not being followed into the hands of different people, with claims then advanced against those people; rather, it is being traced into its substitutions in the defendant’s hands: partnership assets are transferred to Mrs Jones (and can be followed into her hands) and then those assets – cheques which Mrs Jones is entitled to draw against partnership funds – are traced into their successive substitutes in Mrs Jones’s hands, with the £50,000 fund finally paid into court being the traceable proceeds of Mrs Jones’s final chose in action, being her claim against Raphaels.42 37 ibid 164. The bankrupt husband’s title in this context is rather strange, according to Millett LJ (ibid 165), citing from his own earlier judgment in Re Dennis (A Bankrupt) [1996] Ch 80 (CA) 104: the husband did not have ‘a defeasible title, but either an indefeasible title if the act of bankruptcy was not followed by adjudication or no title at all if it was. Outside the law of bankruptcy no similar ambulatory title was known to the law.’ This, in turn, had been derived from the Court of Appeal decision in Re Gunsbourg [1920] 2 KB 426 (CA) esp 446. 38 FC Jones (n 1) 163. 39 Re Gunsbourg (n 37). 40 FC Jones (n 1) 170. 41 ibid 170. 42 See also Banque Belge (n 33), where the Court of Appeal was also dealing with a sum which had been paid into court. Bankes LJ, at 328, apparently perceived this situation as involving a purely proprietary claim at common law and he was anxious to discover that the second recipient, Mlle Spanoghe, whose money had been paid into court by her bank, had not

316 Sarah Worthington But why does such common law tracing entitle the trustee in bankruptcy to claim the legal title (or even property in) the £50,000 proceeds? The ability to trace in equity and claim the proceeds in a fiduciary’s hands was noted by the court, but the court held that Mrs Jones was definitely not a trustee or fiduciary.43 Instead, entitlement to the proceeds was justified by the principle against unjust enrichment: ‘If [Mrs Jones] were to retain the profit made by the use of the trustee’s money, then, in the language of the modern law of restitution, she would be unjustly enriched at the expense of the trustee.’44 This has an intuitively compelling morality to it – no one should be allowed to profit from a wrong45 – but in an action in unjust enrichment for receipt of the £11,000 partnership cash, even if run as a claim in money had and received where the defendant acquires title to the funds received, but that receipt is seen as an unjust enrichment, it would be irrelevant what the defendant had done with the money after she received it.46 See too Lord Millett’s later disavowal of unjust enrichment as a basis for any claim arising at the end of a process of tracing in equity.47 Similarly in tort, a claim in conversion, resting on the trustee’s retention of legal title to the asset, provides a money remedy assessed on the basis of the value of the initial asset converted and the value of its lost use to the claimant, whether or not the defendant still retains that asset, and regardless of what secondary benefits have been generated from it by the defendant. If the defendant no longer has the asset, the claim in conversion does not become a claim relating to the value of the traceable proceeds. This would upset long lines of authority dealing with stolen goods and the claims that can be asserted by their true owners against all subsequent recipients. If this justification lacks the necessary force, the alternative is to turn to later paragraphs in the judgment, where it might be inferred that a simple ‘property’ argument will deliver the same ends: the trustee’s legal title to the original asset entitles him to assert legal title to any traceable proceeds derived from its use.48 But this too surely will not do. Not only does the assertion lack compelling explanation, it also contradicts reality. All of Mrs Jones’s arrangements with her bank, her brokers and other third parties were in her name: she was the legal owner of the respective choses in action. She may have been liable to transfer those benefits to the claimant (a personal claim);49 she may have been liable to novate the accounts (a proprietary response); or she may have been held to be a constructive trustee of the benefits for the claimant (another proprietary response, and the usual response in equity to assertions that the claimant is entitled to ‘the property’ in issue). But it cannot be the case that the trustee in bankruptcy simply becomes the legal owner, without more, when relation back kicks in. Whether these claims arose automatically or only after the trustee’s election, this would put the banks in an untenable position.50 Similarly mixed the plaintiff ’s money in her own account. Atkin LJ also acutely noticed this point, at 332–33, but then acquiesced in the trial judge’s blurring of the action as one ‘for money had and received’: Andrews and Beatson (n 11) 25. 43 FC Jones (n 1) 167. 44 ibid 168. 45 But, unless the wrong is one for which the remedy is disgorgement of profits (as in the fiduciary cases), this stripping of profits is achieved by statute, for the benefit of the state: see the Proceeds of Crime Act 2002, which provides for the confiscation or civil recovery of the proceeds from crime. 46 FC Jones (n 1) 168. See also A Burrows, ‘In Defence of Unjust Enrichment’ [2019] CLJ 52 and the discussion below at section VI.B, indicating that these forms of secondary ‘consequential benefits’ are self-evidently not made ‘at the expense of ’ the claimant, so cannot be recovered by the claimant from the defendant in unjust enrichment. Admittedly, this was not the view taken by unjust enrichment scholars at the time Millett LJ was deciding this case, although Millett LJ himself appears to adopt the now current modern view, at least in relation to personal claims in unjust enrichment: see FC Jones (n 1) 168. 47 Foskett (n 1). 48 FC Jones (n 1) 169–70. 49 Pragmatically, this was the outcome in the case. Arguably only £11,000 plus interest should have been transferred, but that detail is beyond the scope of this chapter although see below at section VI.8. 50 See, eg D Fox, Property Rights in Money (Oxford, Oxford University Press, 2008) 194; Smith (n 11) 328–30.

The Myth of Common Law Tracing 317 with other assets, such as land or shares: it would throw into doubt any ability to rely on the land register, or the share register, as indicating the location of legal title. This is simply not how legal title works. If the holder of legal title (here, the wife) is not entitled to retain the benefit of the assets she is holding, then the remedy is a personal one for restitution for unjust enrichment at common law, or a proprietary one by way of constructive trust in equity.51 On the other hand, if the claim is by the legal owner of assets now seen to be in the possession of third parties,52 then the claim is in conversion, with the consequences already noted. This case appears to be the sole English common law authority on tracing into proceeds and asserting legal title to them, and its analysis appears difficult to defend. This issue is picked up in the discussion below.53

VI. Reasons against Adopting the Same Tracing Rules at Common Law and in Equity As outlined above, early notions of the equitable ‘tracing claim’ – or the sum of the claimant’s rights against the defendant – were unquestionably dependent on whether the claimant had an initial equitable proprietary interest and on whether the defendant owed the claimant fiduciary obligations. No one thought to ask whether other claimants might also have protected rights; no effort was made to uncover the core principles underlying tracing and claiming. This possibility of generality is important. Can the equitable tracing rules be used even when there is no hint of either equitable property or a fiduciary relationship? Can they be used by anyone, including a legal owner, and should they be able to be used in this way? If tracing is decoupled from claiming, then at first sight it appears that this question can, logically, only be answered in the affirmative.54 However, that view now seems mistaken, for reasons outlined below. The principal argument in favour of having only one set of tracing rules is that tracing is merely an evidential necessity prior to claiming. This idea is manifest in Lord Steyn’s words in Foskett v McKeown:55 ‘In truth tracing is a process of identifying assets: it belongs to the realm of evidence. It tells us nothing about legal or equitable rights to the assets traced.’ This is the point that lies at the heart of calls to ‘unify’ the rules of tracing at common law and in equity. In Birks’s words, An exercise of identification either can or cannot be conducted. It would be absurd to suppose that it could be conducted vigorously and resourcefully only on Mondays and Thursdays, and it is prima facie no less absurd to assert that it can be so conducted only by a plaintiff who has managed to attract the attention of equity. ‘Equity’ is not a reason. The prima facie absurdity can only be displaced by finding a convincing reason why some plaintiffs, but not others, should be allowed to overcome evidential difficulties.56 51 Fox (n 50) 194; L Smith, ‘Simplifying Claims to Traceable Proceeds’ 125 LQR 338, 347; J Edelman, ‘Two Fundamental Questions for the Law of Trusts’ (2013) 129 LQR 66, 71–72. 52 An assertion which is only possible in relation to goods. Fraudsters gaining control of the claimant’s intangibles do so by fraudulently getting themselves on the register, not by obtaining control of assets that remain in the legal ownership of their original owner: see Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2013] Ch 156. Indeed, this is also the process adopted in respect of land. 53 See section VI. 54 And so I accepted in Worthington, Equity (n 6) 101. Similarly Smith (n 11) 11, 120–30. See also the cases cited at n 1. 55 Foskett (n 1) 113. 56 Birks (n 11) 243; P Birks, ‘On Taking Seriously the Difference between Tracing and Claiming’ (1997) 11 Trust Law International 2; Birks, ‘Overview: Tracing, Claiming and Defences’ (n 23) 291; C Mitchell, P Mitchell and S Watterson (eds),

318 Sarah Worthington Admittedly we are not yet at that point.57 But, more importantly, there seems to be no practical advantage or doctrinal justification supporting endeavours to reach that point.

A. Problem 1: The Practical Impossibility of Asserting Legal Title to Traceable Proceeds The recent judicial and academic calls for the common law to adopt the equitable tracing rules are calls rooted in the modern conception of equitable and common law tracing, where tracing is merely a means of identification of an appropriate substitute asset. In these circumstances, it is suggested that the identification rules for tracing at common law should be the same as the sophisticated rules operating in equity. But that assertion, while logical as a self-standing proposition, has to be seen in its proper context. As a preliminary point, consider first the process of following. It is trite to observe that X’s ability to ‘follow’ her red sports car from hand to hand over a given period of time says nothing about whether she still has any proprietary interest in the car. In order to establish that, she will have to examine each transfer, starting with the first in the chain of transfers from hand to hand, to see if there is any one of them where X loses her ability to assert that she retains her interest in her car. As lawyers, we would never start down the evidential trail of following the red sports car if was clear at the outset that X would have no proprietary interest in the car at the end of the trail, perhaps because, at the very first step, there was a perfectly proper sale or gift of the car by X to some third party. Similarly, we would not start down the route of ‘tracing at common law’, even using the sophisticated equitable tracing rules, if it were clear that at the end of the tracing trail there was never any prospect of establishing a proprietary interest in the traceable substitutes. This is where a pause for thought is warranted. Tracing in equity is done because, at the start of the exercise, X has an equitable interest in certain assets and wants to establish that she has an equitable interest in some traceable substitute asset that will enable her to assert – or claim – a proprietary interest in the final traceable asset, with all the benefits and protections that affords. It is the same with common law tracing. And since it is common law tracing, the end point will be the assertion of a common law proprietary interest, not an equitable one, and will therefore be either ownership or co-ownership, recognised by the assertion of legal title.58 However, as noted earlier, and barring FC Jones, this is never the outcome,59 and it would seem impossible for it to be the outcome. The reason is evident from our earlier example. If a thief steals X’s car, sells it and buys shares with the proceeds, those shares will be registered in the thief ’s name, and will belong legally to the thief, whatever we might say about X’s entitlement to the economic benefit of those shares (ie whether the shares should be regarded as held on trust or not). This illustration may seem easy because of the rules of registration. But it is the same with bank accounts held in the name of the thief: the thief is the legal owner of the chose in action,

Goff & Jones: The Law of Unjust Enrichment, 9th edn (London, Sweet & Maxwell, 2016) para [7-18]. See also Andrews and Beatson (n 11) 26, suggesting that the best option is to elide both systems and thus create a unified set of rules and a common phraseology (as suggested by Denning J in Nelson v Larholt [1948] 1 KB 339 (KB) 342–43). 57 See n 1. 58 Notice that common law rights to possession – rights which are derivative rights, as are equitable interests under a trust – are not traced. That tells us something too. 59 See, eg Taylor (n 25); Banque Belge (n 33).

The Myth of Common Law Tracing 319 whether or not the benefit of that chose is held on trust for others or is owned absolutely by the thief, and whether or not the thief is subject to any personal claims as a result of the theft. Put bluntly, as a matter of strict property law, it is simply not true that legal title to A1 automatically entitles the owner to legal title to A2, its substitute. Title is not so easily acquired by one owner or lost by another; mere exchange does not suffice.60 There are too many contract and property cases where exchange, even with the intention to deliver title, is ineffective. If this property analysis were taken as literal truth, then we would have none of the complex rules on the assignment of property as we currently know them.61 The only case of common law tracing seeming to contradict these assertions is FC Jones,62 considered earlier. Thus, the first argument against adoption of the same tracing rules at common law and in equity is simply that this could not deliver the proprietary ends so commonly assumed. Further arguments from doctrine and from practice are equally against the advance, and are considered next.

B. Problem 2: The Doctrinal Impossibility of Justifying at Common Law either Persistent Insolvency Protection of the Original Claim or Disgorgement of the Proceeds It is clear from what has been said earlier that equitable owners can typically trace their original asset into its substitute assets and can claim equitable ownership of those substitutes, thus capturing the windfall investment gains made from the investments. The approach in Foskett63 that delivers this outcome might equally seem to suggest that if a thief steals X’s car (where X was the absolute legal owner) and exchanges it for cash and then buys shares, then X’s original legal ownership of the car will emerge as legal ownership of the traceable proceeds (assuming these clean substitutions are traceable at common law), and the thief will simply be treated as though he had ‘stolen’ the claimant’s shares and will have to pay damages accordingly. The claimant’s rights as legal owner of the car become transmuted to rights as legal owner of the substituted shares. Indeed, this transmission of legal ownership into substitutes was the outcome on different facts in FC Jones.64 But surely this is not how the rules work.65 ‘Tracing’ identifies substitute assets, certainly. But if ‘claiming’ truly is a separate exercise, independent of the tracing step, it should not simply transpose the claimant’s original rights into the same form of rights in the substitute assets. That would provide no justification at the claiming stage, and would, in one sweep, completely eliminate any potential impact that the nature of the claim might have on the outcome. 60 See Re Diplock (CA) (n 12) 519–20: ‘Sovereigns in A’s pocket either belonged in law to A or they belonged in law to B. The idea that they could belong in law to A and that they should nevertheless be treated as belonging to B was entirely foreign to the common law. This is the reason why the common law doctrine finds its typical exemplification in cases of principal and agent.’ In those circumstances, the common law ‘proceeded on the basis that the unauthorized act of purchasing was one capable of ratification by the owner of the money’. By contrast, the ‘metaphysical approach of equity’ allowed it to perceive a mixed asset – such as the bank account – ‘as an amalgam constituted by the mixture of two or more funds each of which could be regarded as having, for certain purposes, a continued separate existence’. 61 See the relevant sections on assignment in M Bridge et al, The Law of Personal Property, 2nd edn (London, Sweet & Maxwell, 2017); M Smith and N Leslie, The Law of Assignment, 3rd edn (Oxford, Oxford University Press, 2018). 62 Above n 1. 63 See section II above. 64 As was the case, on different facts, in FC Jones (n 1). 65 See the careful distinction between tracing and claiming in Boscawen v Bajwa [1996] 1 WLR 328 (CA) 334 (Millett LJ).

320 Sarah Worthington Notwithstanding this, ownership of A1 is said to carry with it the same rights (ie ownership) to traceable substitutes. This idea is clearly evident in the case extracts cited earlier. However, this common – in fact, seemingly universal – assertion elides tracing and claiming in a fashion which the courts have explicitly decried.66 If tracing really is neutral as to rights, then the same rights (ie rights reflecting ownership of A1) must persist in the traceable substitute, not merely the same type of rights (ie ownership, first of A1 then of A2). The same type of right is unquestionably a new right, a right apparently generated automatically by the tracing exercise.67 If we are to adhere to modern judicial and academic assertions about tracing, then it is the initial claim to A1 that must remain the same; tracing simply allows this initial claim to persist, with its proprietary (preferential) status intact because there is substitute property which will provide the necessary protection of that status. More explicitly, ownership of A1 may entitle X to specific recovery of A168 or recovery of its value. If the defendant no longer has A1, then specific recovery cannot be demanded. Instead, X can demand payment of its value and, if there are traceable proceeds, can insist that the claim is secured (by lien) against those proceeds.69 In this way, tracing is neutral as to rights: the initial right and its preferred proprietary status are preserved; and no new rights are claimed. The commonly advocated claim to ownership of the proceeds is a new right, one which must be justified by something other than mere tracing. Notice too, going back to the equitable tracing cases, that in these types of cases, X’s initial claim, as it might have been asserted when matters first went wrong, is not ‘I own that car (in equity)’. That was the state of the facts before there was a call for X to make any claim at all. Now that things have gone wrong, X’s claim is that ‘I used to own that car (in equity), and now it has gone; its disposition is a wrong, and I therefore make some claim (to be specified), and in aid of that claim I have my eye on the traceable substitutes in the defendant’s hands’. But what is the aid that X is entitled to in respect of the traceable substitutes? Put in the familiar language, what claim is she entitled to make in respect of the traceable substitutes? It is the same in the common law example: X’s claim at the outset is not that ‘I own that car (at law) – so I can have it back’. The common law does not have a vindicatio. Instead, her claim is that I own that car and you have converted it; I therefore have a personal claim in conversion.70 Whether that claim can be protected in some way by identifying it with the substitute assets is another matter. As a matter of principle and logic there seem to be two possibilities. First, X may be seeking continued insolvency priority for her initial claim that responded to the unauthorised disposal by the defendant of ‘her asset’.71 That claim is of the form: ‘I used to own that car (in equity), and now it has gone; I therefore claim compensation for its loss, and in aid of that claim I have my eye on the traceable substitutes as an asset against which my compensation claim can be secured in order to preserve my insolvency priority.’ This claim is secured by an equitable lien against the substitute asset.72 But permitting that to happen requires some justification that insolvency 66 See, eg ibid. 67 See PJ Millett, ‘Restitution and Constructive Trusts’ (1998) 114 LQR 399, 409: ‘[T]he effect of a successful tracing exercise is to confer on the parties the same rights mutatis mutandis in respect of the substituted asset as they or their predecessors in title to the original asset.’ 68 If the right is based on X’s legal title, then this possibility is severely limited: see the Torts (Interference with Goods) Act 1977, s 3. 69 For further details, see S Worthington, The Law of Personal Property: Text and Materials (Oxford, Hart Publishing, 2000) ch 6. 70 With the court having a discretion to order return of the specific asset: Torts (Interference with Goods) Act 1977, s 3. (And surely not discretion in relation to traceable substitutes.) 71 Here I use ‘her asset’ deliberately, even though it is preferable to avoid terminology which fails to indicate exactly what is meant. Here the meaning is plain, but this chapter seeks to expose how easy it is to slide from loose words to assumed outcomes. 72 Foskett (n 1). See also J McGhee (ed), Snell’s Equity, 33rd edn (London, Sweet & Maxwell, 2015) para 30-055.

The Myth of Common Law Tracing 321 priority should be maintained, and secured against the traceable substitute, notwithstanding that the original property (the initial trust asset) in which the claimant had a proprietary interest, and which gave her insolvency protection to that extent against her trustee, is now no longer in her trustee’s hands. The most obvious justification is that at the start of the story the beneficiary was entitled to the economic benefit of certain assets in the trustee’s hands, despite their legal ownership by the trustee, and was entitled to that benefit in priority to the trustee’s general creditors should the trustee become insolvent. On the facts here, before any breach by the trustee, the trustee would have held legal title to the car, and the beneficiary would have been entitled to advance a proprietary claim to the car (with preserved insolvency priority), to the detriment of the trustee’s general creditors. If English law is to have any version of tracing at all, then there are good arguments to suggest that the strongly protective elements of this fiduciary relationship between trustee and beneficiary should be preserved for so long as there are traceable assets against which X’s original claim – ie a claim to the first call on the ‘economic benefit’ inherent in the car – might still be made against its substitutes. But this is a claim that only affords proprietary protection to the extent of the original claim – ie to the extent of the value of the economic benefit of having the car held on trust for the beneficiary. This is secured not by equitable ownership, but by means of a lien against the substitute assets (the shares).73 It is not the same where X is the legal owner of the car. Then her claim is of the form: ‘I used to own that car (at law), and now it has gone; I therefore claim damages in conversion for its loss, and in aid of that claim I have my eye on the traceable substitutes as an asset against which my damages claim can be secured in order to preserve my insolvency priority.’ The original claim in conversion against the initial recipient of the car does come with some measure of insolvency protection. In particular, an insolvent defendant – or, more specifically, his trustee in bankruptcy – cannot obtain possession of X’s car, dispose of it for the benefit of the general creditors and simply pay X a pro rata sum in damages for conversion along with the defendant’s other general creditors. This is because the defendant’s right to elect whether to pay damages in conversion or return the car74 delivers an election to pay damages in full for the conversion, at which stage the defendant is treated as the owner in law, and it is only then that the trustee in bankruptcy can treat the car as an asset belonging to the insolvent defendant and able to be sold so that its proceeds can be distributed to the unsecured creditors. This means X has insolvency protection while the defendant still has her car. But once the car has been disposed of, matters change. At common law,75 X will have no asset giving her insolvency protection against this defendant, now he no longer has her car, unless she obtains legal ownership of the substitute assets and this ownership affords her the protection desired. It seems impossible to argue that this should happen, however, for the practical reasons noted in the previous sub-section, and for the doctrinal reasons addressed next. Moreover, there is no common law equivalent of an equitable lien, and indeed no justification advanced as to why X’s damages claim should be protected against the insolvency of the defendant regardless of whether he has her car. Instead, X’s insolvency protection is delivered by her original asset – her car – and persists in her claims in conversion against subsequent defendants

73 See the discussion in Worthington, Equity (n 6) 99–100, 104–06. 74 Torts (Interference with Goods) Act 1977, s 3. 75 It is a different matter whether X is afforded any protection in equity: see the possibility of claimants asserting equitable interests in substitute assets acquired by thieves, discussed in Westdeutsche Landesbank Girozentrale v Islington LBC [1996] UKHL 12, [1996] AC 669 (HL); Black and Black v S Freedman and Company [1910] HCA 58, (1910) 12 CLR 105 (HCA). These claims can be justified as disgorgement claims responding to the knowing and fraudulent wrongdoing of the defendants in stealing the claimants’ assets; these equitable remedies are not of general application to all cases of conversion. And clearly these remedies do not rely on any concept of tracing at common law.

322 Sarah Worthington while they still have the car; it is not preserved more generally at common law by granting X legal title to the substitute assets. Any suggestion that common law tracing (and claiming) could or should help here seem flawed. The second alternative is that, rather than simply seeking insolvency protection for her initial claim, X may instead advance distinct and different primary claims in respect of the windfall gains represented by the substitute assets. In pursuit of these claims, the technique of tracing is used in equity to identify assets, or gains, that were, as a matter of physical evidence (or, with mixed substitutions, a matter of legal presumption), derived from the initial trust assets in the trustee’s hands. Returning to our trust beneficiary, there are at least four reasons for allowing her to claim equitable ownership of the shares, having established that these are the traceable substitutes of her car. They represent increasingly liberal approaches to primary claims to windfall secondary profits. First, the result may follow because the trustee has a duty to invest the trust funds on behalf of the beneficiary and a discretion to select the investments. In these circumstances, the law may allow the beneficiary to ‘ratify’ or ‘adopt’ any unauthorised investment and insist that the trustee has invested the funds for the beneficiary and not for himself. This approach is automatically proprietary, with all the advantages for the beneficiary that flow from that,76 but it depends on quite specific facts.77 Secondly, the result may follow because the defendant trustee is a fiduciary. All fiduciaries must behave loyally; they are not allowed to make secret unauthorised profits out of their position, and in particular they are not allowed to appropriate the trust assets for their own personal purposes, using them to generate secondary profits, or investment profits, for their own benefit. Defaulting fiduciaries are obliged to disgorge these disloyal profits to the beneficiary, and this obligation to disgorge is proprietary: the beneficiary is regarded as the equitable owner of any profits (eg the shares) that the trustee must transfer.78 These two rationalisations are unquestionably part of the existing law. If this is as far as the law goes, then claims to windfall secondary profits are conditional on proving fiduciary wrongdoing – and, indeed, fiduciary wrongdoing of a very particular type. Moreover, the tracing rules have a particular purpose, so we can expect the content of those rules (and any adopted presumptions) to be directed at delivering those ends. This is indeed the dominant feature of the tracing and claiming cases typically presented in textbooks on ‘tracing’. Thirdly, as in Foskett, the law could simply insist that claims to substitute assets delivering windfall secondary profits follow automatically from the law of property, without the need for any further justification. The logic is simply this: if the claimant owns the original asset, whether at law or in equity, then she automatically owns all the profits derived from its use. The reason itself compels the conclusion that the right is proprietary. This analysis would justify giving X the equitable ownership of the shares. All this has the advantage of simplicity, but its ramifications are profound and, as I have argued before, largely undesirable.79 A fourth and final justification for permitting claims to windfall secondary profits is based on unjust enrichment, although, as explained below, this view is now perhaps relevant only as a matter of historical record. According to this analysis, profits made from the unauthorised use

76 Foskett (n 1) could be analysed like this, but was not by the House of Lords. 77 The same approach might legitimately be applied where agents have management of their principal’s assets and make unauthorised substitutions, even when the principal, not the agent, has legal title to the original asset, but the agent will acquire title to the traceable substitute. 78 A-G for Hong Kong v Reid [1994] 1 AC 324 (PC); FHR European Ventures (n 1). 79 The detail is in Worthington, ‘Justifying Claims to Secondary Profits’ (n 6).

The Myth of Common Law Tracing 323 of another’s property were seen to be, as of their very essence, unjust enrichments that should be returned to the original owner of the property.80 Little beyond assertion supported this view that it is necessarily unjust to keep the profits derived from the use of another’s property. Instinct might suggest it is, until it is pointed out that the rule will bite even when the defendant is completely innocent and ignorant of the fact that the property is not his, and even when, had he known, he would simply have returned the claimant’s property and used his own for the transaction. If these concerns are brushed aside, then the rationalisation seems merely to reiterate the property rationalisation favoured by the judiciary, and does not directly concern itself with the problems that the law of unjust enrichment is designed to remedy.81 As it is, unjust enrichment scholars now also disavow this approach, doing so on the basis that these forms of secondary ‘consequential benefits’ are self-evidently not made ‘at the expense of ’ the claimant, and so cannot be recovered by the claimant from the defendant in unjust enrichment.82 Of these four bases for allowing the claimant a proprietary claim to the traceable proceeds, the law favours the third justification; this is the justification mandated by the House of Lords. This ‘property’ analysis suggests that a claimant will always be entitled to assert a proprietary interest in the traceable proceeds even if that delivers an unexpected windfall gain. It is exemplified in the quotation cited at the start of this chapter. The ‘wrongs’ analysis is much more restrictive; it suggests that disgorgement is only apt if the defendant has breached his management investment agreement or his fiduciary obligations of loyalty. All of this is clearly significant in considering the common law tracing analogies. The House of Lords’ preferred ‘property’ analysis has already been criticised,83 but it fails further in the common law context because it seems impossible to concede that a claimant can simply assert legal ownership of the traceable proceeds in this way.84 That leaves the unjust enrichment analysis, already conceded as inapplicable in this context, and the wrongs analysis, applicable only when the defendant is a fiduciary. This final context is apt to describe agents and company directors dealing with assets belonging legally to their principals, for example, but in these contexts the remedies are inevitably awarded in equity by way of constructive trust, and not by awarding the principal a legal title to any traceable proceeds of the wrongdoing. Finally, one would have thought that the proprietary or personal nature of the claiming response would also need further careful justification. In the context of primary claims to disloyal 80 See especially Birks, ‘Overview: Tracing, Claiming and Defences’ (n 23) ch 1; Burrows (n 23), but criticised in C Rotherham, Proprietary Remedies in Context (Oxford, Hart Publishing, 2002) 000. See also P Birks, ‘Mixing and Tracing: Property and Restitution’ (1992) 45 Current Legal Problems 69. This approach was roundly dismissed in Foskett (n 1); although note the references to unjust enrichment in FC Jones (n 1) in explaining why the defendant could not keep the profit element of her successful investment in potato futures. 81 See FC Jones (n 1) 168 to that effect, even though the court’s conclusion was assertion of a common law property interest and therefore entitlement to money that had been paid into court, not a remedy in unjust enrichment by way of money had and received. 82 See especially Burrows (n 46), citing Investment Trust Companies (in liq) v Revenue and Customs Commissioners [2017] UKSC 29, [2018] AC 275 and Prudential Assurance Co Ltd v Revenue and Customs Commissioners [2018] UKSC 39, [2018] 3 WLR 652. See also A Burrows, ‘“At the Expense of the Claimant”: A Fresh Look’ [2017] Restitution Law Review 167. Prudential Assurance Co arguably goes too far in overruling Sempra Metals Ltd v Inland Revenue Commissioners [2007] UKHL 34, [2008] 1 AC 561 and denying recovery of the ‘use value’ of funds, or other assets to which an orthodox claim in unjust enrichment might have been advanced. After all, the unjust enrichment claims seek to restore to the claimant the economic value lost by the claimant (‘at the expense of ’) and unjustly gained by the defendant (the ‘enrichment’): see Worthington, ‘The Commercial Triple Helix’ (n 3). By contrast, these secondary profits could perhaps be recovered on the basis of some alternative claim, such as a claim in wrongdoing, but unjust enrichment scholars have long held that they cannot be recovered simply on the basis of the proprietary claim as asserted by the House of Lords in Foskett (n 1): see Burrows (n 23), 417–19, 424–28. 83 See n 79. 84 See the discussion above at sections V and VI.A (Problem 1).

324 Sarah Worthington profits, the vigour of such arguments is very recent.85 Yet with tracing interposed, it seems possible to move from a claimant with ‘assets held on trust/at law’ to one with ‘traceable proceeds held on trust/at law’ with far less fuss and bother.86 That is surprising, to say the least.87

C. Problem 3: The Practical Impossibility of Common Law Tracing through Mixtures One of the appeals of equitable tracing is its ability to trace through mixtures. Common law tracing lacks that ability, according to the authorities.88 It is occasionally suggested that this is not true, and, more insistently, that it should not be true, and that the common law should adopt the equitable rules. Consider first the assertion that the common law does already trace through mixtures. The authorities typically cited are mixing/identification cases such as Spence v Union Marine Insurance Co Ltd,89 FS Sandeman & Sons v Tyzack & Branfoot Steamship Co Ltd90 and Indian Oil Corp Ltd v Greenstone Shipping Co SA (Panama) (The Ypatianna).91 However, these are not tracing cases; they are cases where the legal ownership of the original asset has to be established (eg the bales of cotton, minus their identification markings). Given that there is a bulk, and that the owners of the components of that bulk must come from the restricted pool of those who put their bales of cotton on the ship, it follows that if none of them own the cotton, then the cotton will be bona vacantia. The common law resists this counter-intuitive conclusion and adopts a pragmatic approach, holding that all the unmarked bales of cotton are owned in common by all the contributors to the unmarked bulk. What remains unanswered, however, is whether the equitable rules for tracing through mixtures could have any application in a common law context. In this context, we should be wary of assuming that the equitable tracing rules merely state an empirical conclusion about the location of assets, and whether those assets physically – or in value terms – stand for the earlier assets for which they are the traceable substitute. In clean substitutions, that may well be the case, but the sophistication of equity’s tracing rules lies precisely in their ability to overcome evidential difficulties. And here it is plain that what is in issue goes beyond mere scientific value-neutral substitute-asset identification. A simple illustration suffices. All that is needed to make tracing a more complicated exercise is to raise the prospect of mixed substitutions. This is where, somewhere along the chain of exchanges, the original asset or its traceable proceeds are mixed with other assets before the next exchange takes place. The likelihood of this is self-evident, and almost all the equitable tracing rules are directed at resolving problems in these contexts. To tweak the earlier example, suppose once again that T holds a car on trust for X and without authority sells it for £2,000; he deposits

85 See the extensive debates reported in FHR European Ventures (n 1), which settled the law in favour of a proprietary response. 86 Foskett (n 1); FC Jones (n 1). 87 Recall that if the claimant were to acquire legal title to the traceable proceeds at the end of a common law tracing exercise, then the common law’s personal remedy will come with insolvency protection: see above, text at nn 74 and 75. 88 See Taylor (n 25). 89 Spence v Union Marine Insurance Co Ltd (1867–68) LR 2 CP 427 (CP) 436–39 (Bovill CJ). 90 FS Sandeman & Sons v Tyzack & Branfoot Steamship Co Ltd [1913] AC 680 (HL) 694–99 (Lord Moulton). 91 Indian Oil Corp Ltd v Greenstone Shipping Co SA (Panama) (The Ypatianna) [1988] 1 QB 345 (Com Ct) 360, 368–71 (Staughton J).

The Myth of Common Law Tracing 325 this sum in his bank account, which is already in credit to the tune of £2,000; he then uses £2,000 to purchase shares and a further £2,000 to fund a long-planned skiing trip. The shares are now worth £10,000, but the sole residue of the skiing trip is a number of happy memories. Will the tracing rules allow X to say that ‘her’ £2,000 (the traceable substitute of ‘her’ car), or at least some part of it, is now represented in the value of the shares, or will they compel her to accept that ‘her’ £2,000, or some part of it, was spent on a skiing holiday, so that there is now no substitute asset? The answer is that the applicable tracing rules differ depending on who has contributed to the mixed fund. Here, the funds have come partly from X and partly from the defaulting trustee. But in other scenarios, the £2,000 initially in T’s account may itself have come from a different trust, again without authority. The authorities are familiar and uncontroversial. Their essential points can be summed up quite simply. Mixtures make absolute proof of the provenance of a substitution impossible. In ‘guessing’ or assigning the provenance of different substitutions, equity simply treats innocent parties equally, but allows the claimant to make self-interested assumptions against wrongdoers.92 It follows that in this example X will be able to cherry-pick the better substitutions from amongst all the alternatives that could possibly have arisen on the facts. As a result, she can insist that it was her £2,000 that spent on the shares and T’s contribution that was spent on the ski trip. But if the £2,000 contributions to the mixture had both come from trust beneficiaries, then none of the presumptions that are possible against wrongdoers would apply, and the tracing rules would allocate half of each beneficiary’s contribution to the share acquisition and half to the skiing trip; that is, as equally innocent parties, they would each share both the gains and the losses pro rata according to their inputs to the fund being expended.93 Thus, even with equitable tracing, it is clear that Birks’s qualification, cited earlier, really bites. In these cases, the courts have long seen that there is indeed ‘a convincing reason why some plaintiffs, but not others, should be allowed to overcome evidential difficulties’.94 It is telling – perhaps telling in the extreme – that even in equity, and even though tracing is acknowledged to be an evidential process, the tracing rules nevertheless incorporate strong normative elements, as described above.95 Indeed, extrajudicially, Lord Millett has noted that tracing against innocent recipients, even in equity, and even where there are ‘clean substitutions’, can be difficult to justify.96 If those normative elements are to be transferred to common law tracing, would the translation simply differentiate between common law wrongdoers and innocent parties in an analogous way? And who is to be classified as a wrongdoer and who as an innocent party? Common instinct suggests a thief would be classified as a wrongdoer, but what of an innocent donee from the thief or, worse still, an innocent purchaser from the thief? All these people will

92 Tracing against the wrongdoer fiduciary/trustee: Re Hallett’s Estate (1880) 13 Ch D 696 (CA); Re Oatway [1903] 2 Ch 356 (Ch). Contrast Turner v Jacob [2006] EWHC 1317 (Ch), [2008] WTLR 307 with Shalson (n 1) [144], with the latter approach preferred here as being more consistent with general principle and with Foskett (n 1) 132 (Lord Millett). Tracing involving competing innocent parties: Barlow Clowes International Ltd v Vaughan [1992] 4 All ER 22 (CA); Shalson (n 1). End of the tracing trail/lowest intermediate balance, etc: James Roscoe (Bolton) Ltd v Winder [1915] 1 Ch 62 (Ch); Bishopsgate Investment Management Ltd v Homan [1995] Ch 211 (CA); but see also Federal Republic of Brazil v Durant International Corp [2015] UKPC 35, [2016] AC 297. 93 ibid. 94 Birks (n 11) 243. 95 For important detail and comment, see P Sales, ‘The Future of Tracing: Practical and Conceptual Issues’ (2017) 25 Restitution Law Review 183, 185; Cutts (n 18). 96 Millett (n 2), 316–17.

326 Sarah Worthington be liable in conversion to the true owner, so the nature of the common law claim is not suggestive of differential wrongdoing. If these parties mix the proceeds and then substitute, what is the purpose of tracing? As the law now stands, the distinction between wrongdoers and innocent defendants is material only to the presumptions applied during the tracing process, but not to the final claim to the entire secondary assets pool, even if the pool constitutes a windfall profit. That approach is doubted (see the previous section), and that doubt is further supported by extrajudicial comment.97 If these nuances in the equitable domain are translated to the common law domain, and are then also combined with the findings from the immediately previous section, it then becomes plain why there are no cases where the common law traces through mixtures. Indeed, even with clean substitutions, it has already been pointed out that the common law cases, other than FC Jones, do not conclude that the claimant obtains legal title to the traceable substitute.98 This lends further support to the findings of the previous sections, where the conclusion was that there is little point tracing at common law since it is not possible – either practically or doctrinally – to justify a finding that the claimant has legal title to the substituted assets. If the courts routinely resist finding that claimants obtain legal title to the traceable substitutes, what do they do instead? For the most part, and perhaps predictably, it seems they focus on investigating the existence of common law unjust enrichment claims. In this context, causality in delivering the enrichment is crucial, not tracing and technical attribution or substitution. By contrast, and although it is beyond the scope of this chapter, it is suggested that if the claimant’s proprietary interest in the final traceable assets is important to the claim being advanced, then the better solution is that equity should be left to do all the work99 – and, indeed, the cases suggest that it does already do all the work.100

VII. Conclusion The very purpose of tracing is to enable assertion of a proprietary interest in substitute assets. With equitable tracing, the assertion is of an equitable proprietary interest; with common law tracing, the assertion is necessarily of a common law proprietary interest, that interest being title (no other common law proprietary interest being possible in the context of tracing). It is illogical to assert that common law tracing is possible and yet also hold that it is impossible to claim legal title to the traceable substitutes. If the second assertion is true, then common law tracing is a myth. And the second assertion does appear to be true.101 Given this fundamental stumbling block to common law tracing, it follows that it then becomes immaterial to decide whether the common law rules should be preserving insolvency priority in substitute assets or affording disgorgement of windfall profits, and doing the latter for any of the four reasons typically articulated in the context of equitable tracing.102 It might

97 ibid

316–17. n 33 above. 99 See Sales (n 95), 191–92. 100 See Smith (n 51). 101 See above, especially sections V and VI.A (Problem 1). 102 See above, section VI.B (Problem 2). 98 See

The Myth of Common Law Tracing 327 be noted in the context of common law claims, however, that even if the initial issue of asserting legal title was not an impediment, it would in any event seem impossible to justify either of the two core goals of the common law’s (and equity’s) protective tracing regime, being preservation of insolvency priority and disgorgement of secondary gains. This is because the original common law claim does not itself have such protection built in by virtue of the duties imposed on the relevant defendants in these cases.103 Imposing a fiduciary-style regime in the common law context, with its remedies of specific redelivery and profits disgorgement, would certainly remove the incentive to steal property and reap the rewards from it. However, such a regime is not warranted merely because it induces better behaviour and provides better remedies. Much of private law would be subsumed within fiduciary law if this were adequate justification. Fiduciary law is not directed at protecting property, but at protecting its loyal management. With this as its goal, the only way the claimant’s legitimate interests can be adequately protected is by deterrence regimes that redirect disloyal profits to the intended beneficiary. The law of conversion is quite different. It is directed at protecting property, and the remedies reflect that. If added deterrence is warranted in cases of theft, then it should be provided through rules on wrongdoing, including criminal law,104 not by the technical process of tracing. If the issues in this broad area are to be generalised, it might be noted that property’s primary protection comes from rules that enable disputing parties to resolve title conflicts and priority disputes. Absolute justice is unlikely; the law must simply decide who has the stronger claim. When an asset is stolen and resold, for example, the original owner and the new purchaser cannot both own the asset: the law must choose between two innocent victims of the thief ’s dishonesty. The various legal rules expose the tension between protecting property rights and protecting the security of transactions in general trade and commerce. The current matrix of common law and equitable rules reflects a consistent and coherent response to these issues. Property’s secondary protection is more controversial. Equity’s tracing and claiming strategies allow claimants to assert new proprietary rights in substitute assets. These rights preserve the insolvency priority of the initial proprietary claim or secure entitlements to windfall secondary profits. Neither of these options follows inevitably from the initial assertion of a proprietary right. Any insolvency priority inherent in the ownership of an asset need not be preserved in substitutes: to do so weights security of property over security of transactions, and dramatically widens the class of third parties affected by the initial property interest. Proprietary claims to windfall secondary profits operate even more aggressively. There may be much to be said for the view that these expansive equitable tracing and claiming rules are appropriate in the trust situation, where they were conceived, but not outside that. In a trust context, these rules effect a necessary division between the trustee’s fiduciary and personal patrimony, allocating both primary and substituted assets to one or the other. Outside these limited circumstances, it is arguable that the reach of property should not be unnecessarily enlarged. Certainly, one of the most critical and controversial modern debates concerns the wisdom of preserving insolvency priority in substitutes and requiring the handing over of

103 Note that although insolvency priority is preserved at common law in a conversion claim while the defendant has the converted assets in his possession, it is not preserved thereafter, and there would seem to be no justification for that protection since following the asset into the next recipient’s hands will deliver that same protection in another form. 104 Statutory rules already require certain criminals to disgorge the proceeds of their crimes. This mirrors fiduciary deterrent strategies, although the profits of these crimes are paid not to the victims, but to the state, as befits the interest being protected. See n 45 above.

328 Sarah Worthington indfall secondary profits merely because a claimant starts out on the road with a property interw est in a completely different asset, and notwithstanding the absence of any wrongdoing on the part of the defendant. To that end, it is perhaps reassuring that this chapter, if it persuades, suggests that common law tracing into substitutes is a non-starter for various reasons, the primary one being that r evision of the allocation of legal title on such a fortuitous basis would completely disrupt the security of domestic and commercial arrangements.

19 Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies STEPHEN WATTERSON

I. Introduction The emergence of cryptocurrencies such as Bitcoin has grabbed news headlines worldwide. Much of this popular attention has focused on their astonishing price inflation and fluctuations, and on the riches that might appear to be had from speculative cryptocurrency investments. Other attention has focused on the many public policy implications of their emergence – eg concerns about the facilitation of black-market activity/asset laundering, investor protection or the macro-economic implications of these privately created currencies becoming widely adopted as a medium of exchange. To date, remarkably little close attention has been devoted to private law analysis of cryptocurrencies, and of transactions involving them. Within the English legal literature, an important advance has recently been made with the publication of the major collaborative work, Cryptocurrencies in Public and Private Law.1 Nevertheless, its private law coverage remains incomplete in one important respect: it contains no account of the challenges presented by these cryptocurrencies for the law of restitution/unjust enrichment. On examination, the challenges are considerable. Indeed, it quickly becomes apparent that any sure-footed account of the applicability of this body of law – whether it is English law or some other law that is governing – necessitates a considerable amount of preliminary contextual and conceptual foundation laying. That job will be undertaken in this chapter, which prepares the ground for a detailed analysis of possible restitutionary liabilities arising from cryptocurrency transactions that the author will undertake in a separate paper. The discussion also has a much broader significance, however, as analysis of the issues that are considered here is needed before one can make sense of any private law claim involving cryptocurrencies.

1 D Fox and S Green (eds), Cryptocurrencies in Public and Private Law (Oxford, Oxford University Press, 2019). cf also the briefer treatment in S Gleeson, The Legal Concept of Money (Oxford, Oxford University Press, 2018) ch 9.

330 Stephen Watterson

II. A Brief Introduction to Cryptocurrencies Any account of the legal treatment of cryptocurrencies requires at least an outline understanding of their character. This is not straightforward to achieve – recent years have seen the emergence of a diversity of cryptocurrency systems, whose technical underpinnings and operations differ. What follows is a rudimentary account of salient features of the Bitcoin system2 – the most well known, and the first to achieve widespread acceptance. In broad terms, Bitcoin, if a ‘currency’, is a species of ‘digital currency’ – a purely digital representation of value. It is unlike fiat currencies, in that it is not denominated in the unit of account of any state’s monetary system – eg the UK’s pounds sterling, US dollars. Instead, the value is denominated in the Bitcoin system’s own units (bitcoins), or fractional units.3 These digital units are neither created by a state nor under state authority – they have a private genesis, in a privately authored software protocol. As a purely digital representation of value, they are unlike money in its tangible form – coins or notes. But they are also unlike so-called ‘bank money’ – which consists in law of a chose in action, denominated in fiat money, against an intermediating, account-holding bank.4 A key challenge for creators of such currency systems – if they are to attract widespread public use and acceptance, and thereby value – is how to achieve system integrity. This requires a trustworthy historic record which allows for the reliable identification of present entitlements, as well as adequate security for those entitlements, such that a disposition of the currency is recognised only where it has been effected by its lawful holder. The Bitcoin system seeks to achieve this via a combination of several features. These consist above all of a particular form of decentralised/distributed digital ledger, together with public– private key cryptographic methods. This explains why Bitcoin, and other currencies operating in a similar way, have come to attract the further designation: ‘cryptocurrencies’. Within the Bitcoin system, there is no central authority or other intermediating entity that maintains a record of entitlements – the Bitcoin system is a deliberately decentralised system, which operates via a peer-to-peer computer network. Copies of the digital ledger are shared, stored and maintained across all of the ‘nodes’ of this network. This digital ledger – the ‘transaction record’ – discloses the full transactional history of all bitcoin units that have ever entered circulation. It is also devised so that it is, for practical purposes, ‘append only’: it is only practically possible to add new transactions to the record, and it is not practically possible to alter the historic record. The method of achieving this is complex. In simplified terms, the process of ‘writing into’ the transaction record comprises a process of bundling up valid transaction data into ‘blocks’ with an identity which is unique, and which is uniquely tied to the preceding blocks already in the digital ledger (the so-called ‘blockchain’),5 so that any attempt to modify any part of the historic transaction record will be immediately

2 For a full technical introduction to the Bitcoin system and its underlying technology, see, eg A Narayanan, J Bonneau, E Felten, A Miller and S Goldfeeder, Bitcoin and Cryptocurrency Technologies – A Comprehensive Introduction (Princeton, Princeton University Press, 2016). For a non-technical exploration of the genesis and future of Bitcoin and other cryptocurrencies, see P Vigna and MJ Casey, Cryptocurrency – The Future of Money (London, Vintage, 2016). 3 The smallest fractional unit permitted is 0.00000001 bitcoin (a ‘satoshi’, named after ‘Satoshi Nakamoto’, a pseudonym for the (so far unidentified) creator(s) of the Bitcoin system). 4 See section IV.A(i) below, where it is explained that the digital unit/fractional unit does not have the character of a claim-right/chose in action. 5 A useful exploration of ‘blockchain’, its wider applications and the implications of these for legal regulation is P de Filippi and A Wright, Blockchain and the Law – The Rule of Code (Cambridge, MA, Harvard University Press, 2018).

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 331 verifiable.6 The process of transaction validation and bundling requires the expenditure of computing power, which members of the system are incentivised to deploy on a competitive basis. The success of these so-called ‘miners’ is rewarded by newly issued bitcoin.7 Within the context of this ledger system, the concept of a ‘transaction’ needs further attention. The transaction record associates bitcoin units/fractional units – representing the unspent outputs of prior transactions – with a particular digital ‘address’. The real-world address holder/user is not visible within the transaction record, and so to that extent, the record maintains a degree of anonymity. However, it is only a degree – a ‘pseudonymity’. It is not impossible, albeit with greater or lesser difficulty, to link an address holder/user and associated transactions to a realworld actor. The nature of such digital public addresses also requires explanation: it lies at the heart of the system’s strategy for securing bitcoin entitlements. In short, the Bitcoin system uses public–private key cryptographic methods with a view to ensuring that transactions are validated only when they are authorised by the sending address holder. The digital ‘address’ with which bitcoin units/fractional units are associated is a version of a public key, which has a paired private key. A ‘transaction’ involving a transfer of some or all of those units/fractional units from one address to another involves the generation of a transaction ‘message’, which identifies another ‘destination’ address for the transfer. Crucially, however, it is necessary to ‘sign’ the transaction ‘message’, by using the paired private key. The resulting transaction message will be broadcast across the peer-to-peer network, and the system is meant to work in such a way that it will only treat as valid – and write to the transaction record, as part of a new block in the blockchain – transactions that use unspent transaction outputs associated with the relevant sending address, which have been duly authenticated by the application of the associated digital signature. In light of this, it should be apparent that the security of the private key is paramount. If the private key is lost, the relevant address holder will cease to be able to dispose of any unspent transaction outputs associated with the paired address. Equally, if a private key’s secrecy is compromised, an unauthorised third party may be able, by its use, to effect transactions involving the outputs associated with the paired address. System users may have any number of addresses with which bitcoin units/fractional units – as unspent outputs of prior transactions – may be associated. In practice, system users may use a variety of ‘wallet’ solutions to assist with the management of these addresses. These include local software solutions or online wallet services which may, in particular, enable the calculation of balances associated with particular addresses and provide a vehicle for the storage of the associated private keys.

III. Contextual Challenges When answering the concrete legal questions that may be presented by cryptocurrency transactions, assumptions must inevitably be made regarding the applicable law. This chapter’s major concern is to lay the foundations for understanding private law claims insofar as they are governed 6 A transaction written into the blockchain in this way is said to be ‘confirmed’. This will not guarantee that a different transaction history ends up prevailing – but, in practice, within a relatively short period, if successive ‘blocks’ of transaction data are built upon the transaction block containing the transaction in question, this becomes vanishingly unlikely. Within the Bitcoin system, a rule of thumb is that after six confirmations – six successive blocks, built upon the block containing the relevant transaction, it can be assumed that it will hold good. 7 For full exploration, see Narayanan et al (n 2) ch 5.

332 Stephen Watterson by English law. However, whether or not that is a correct starting point in any particular case, cryptocurrencies present several key challenges that need to be faced by any legal system.

A. Immutability and (Ir)reversibility A key component of technical descriptions of blockchain – and of cryptocurrency systems, like Bitcoin, that use this technology – is the ‘immutability’ of the blockchain transaction record. This denotes the practical difficulty – even the practical impossibility – of rewriting/altering the historic blockchain transaction record reflected in the distributed ledger.8 What is the significance of this for the availability of restitutionary relief in respect of cryptocurrency transactions? On closer investigation, the immutability of the historic transaction record does not – and should not – entail ‘irreversibility’ by restitutionary mechanisms. Immutability is a design feature that supports the integrity of the blockchain ledger system. The historic blockchain transaction record is made more or less immune from ex post facto alteration – an improper ‘rewriting’ of history – and as a corollary of this, the risk of ‘double spending’ is substantially reduced. However, none of this requires that a legal system must accept the consequences of that blockchain transaction history as justified, and as incapable of ‘correction’ for the future, via appropriate means. In practice, the technical features of blockchain-based cryptocurrency systems are no barrier to commonplace transactional flaws that are the familiar basis for restitutionary remedies based on underlying ‘off-blockchain’ events, such as an intended ‘transfer’ of cryptocurrency vitiated by mistake or duress, or made on a condition which is not satisfied, an unintended transfer that occurs through another’s unauthorised intervention, or a transfer procured via another’s wrongdoing. There is no compelling reason9 why a legal system should disregard such flaws which infect transactions recorded in the blockchain ledger, and which normally justify restitutionary relief.10 It is also evident that the immutability of the historic transaction record is no barrier to the availability of familiar restitutionary remedies that would serve to reverse an unjust/wrongful transaction in money or, less often, in specie. If the ‘recipient’ address holder is identifiable, then they are in principle susceptible to a monetary restitutionary remedy, consisting of an order that they pay the value in money of the benefit unjustly/wrongfully received. Further, if the relevant cryptocurrency unit is a res that can be located, then it might be the subject matter of a specific/proprietary restitutionary mechanism which might result, if necessary, in a ‘retransfer’ of the specific unit to the restitution claimant.11

8 It is also frequently noted that within a distributed peer-to-peer blockchain ledger system, such as that used for Bitcoin, there is no intermediary that might be able to reverse an ‘on-blockchain’ transaction. This feature alone would afford a practical finality to such transactions which is not present in traditional bank-intermediated payment systems. A different analysis is required for ‘off-blockchain’ transactions of the nature discussed in section III.G below. 9 This might only be acceptable in the unlikely event that there was a system rule, agreed to by users of the cryptocurrency system, that the transaction record should be conclusive of lawful entitlements and that ordinary legal remedies afforded to reverse defective transfers are ousted. 10 A similar conclusion is reached, in the context of the proprietary implications of cryptocurrency transactions, in D Fox, ‘Cryptocurrencies in the Common Law of Property’ in Fox and Green (n 1) esp paras 6.05–6.07. 11 This must be read subject to the technical caveat that, as explained below, the true analysis may be that the digital res is consumed in the ‘transaction’ and replaced by a new digital res, which represents the transaction’s output: see section IV.A(ii) below.

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 333

B. The Currency of ‘Monetary Value’ for Monetary Restitutionary Remedies The labelling and operation of cryptocurrency as ‘currency’ – as a potential medium of exchange, and as a measure of the value in exchange of goods or services – raises a pressing question when it comes to the form of restitutionary remedies. In English law, the standard restitutionary remedy for a cause of action in unjust enrichment is a personal, monetary restitutionary remedy – whatever the nature of the benefit received by the defendant, restitution is generally effected via a liability to pay its value in money. This quickly prompts a further question: what is the currency in which this monetary liability falls to be measured? Historically, in cases coming before the English courts, this was pounds sterling. More recently, and consistently with wider developments in judicial practice,12 the English courts have been willing to make monetary restitutionary awards denominated in a foreign currency in appropriate cases13 – most obviously, but not only, where the defendant’s immediate enrichment consists of a receipt of foreign currency from the claimant.14 Might the courts ever go one step further, and allow a particular cryptocurrency to be the measure of a benefit’s monetary value, and potentially the currency of any resulting monetary restitutionary award?

C. Value Volatility and Monetary Restitutionary Remedies If any monetary restitutionary remedy must be measured in sterling or in an appropriate foreign currency, then even an apparently simple case involving a defendant’s receipt of an amount of cryptocurrency could present some difficult valuation-related questions if the cryptocurrency exhibits significant value/price volatility – a familiar characteristic of the cryptocurrencies, like Bitcoin, that have emerged to date. A monetary value – in pounds sterling or in an appropriate foreign currency – would need to be attached to the defendant’s enrichment arising from the cryptocurrency’s receipt. During periods of significant value/price volatility, the cryptocurrency’s value/price could fluctuate substantially, up or down, in the wake of the defendant’s initial receipt. One unavoidable consequence is that the measure of the defendant’s liability – and of the claimant’s recovery – could vary substantially depending upon the exact time when the defendant’s restitutionary liability is understood to arise and the enrichment’s value is determined. These are issues that require resolution in a principled manner, in light of the general nature, basis and essential components of the relevant type of restitutionary liability. Although not yet conclusively settled, the prevailing view in the English literature, reinforced by recent judicial decisions, is that, in the context of personal liability in unjust enrichment, the monetary value of the enrichment is determined at the time of the defendant’s initial receipt – meaning, in the present context, the time of the defendant’s enriching receipt of the cryptocurrency.15

12 The landmark decision was Miliangos v George Frank (Textiles) Ltd [1976] AC 443 (HL). For a comprehensive survey of the background to this decision and later developments, see M Howard, J Knott and J Kimbell, Foreign Currency: Claims, Judgments and Damages (London, LLP, 2016). 13 The leading discussion remains BP Exploration Co (Libya) Ltd v Hunt (No 2) [1979] 1 WLR 783 (QB) 838–45 (Robert Goff J). 14 ibid 840. 15 See esp Prudential Assurance Co Ltd v Revenue and Customs Commissioners [2018] UKSC 39, [2018] 3 WLR 652.

334 Stephen Watterson

D. Reification and Specific Restitutionary Mechanisms The availability of a monetary restitutionary remedy does not depend upon the precise characterisation of the ‘benefit’ involved in a receipt of cryptocurrency – such relief is capable of accommodating any sort of benefit which is susceptible to valuation in money. In contrast, more analytical work is certainly required if the law is to afford a specific restitutionary mechanism in the context of cryptocurrency transactions – by which a claimant might be able to recover identifiable units of cryptocurrency. It is not necessary at this point to explore the basis and conditions on which such relief is potentially available – a highly contested question. It is sufficient to observe that for such relief to be feasible, it will be necessary to be able to ‘reify’ a unit of cryptocurrency – to understand it as a thing/res which can be the subject of rights that are susceptible to analysis in terms of familiar legal concepts.16 Within the English legal system, this would probably mean conceiving of an identifiable cryptocurrency unit as a digital res to which one (or more) person(s) might hold legal title, and to which others might have lesser claims arising at law, or more likely in equity, in the shape of a trust or lien.

E. (En)title(ment)s, Unjust Enrichment, Wrongdoing and Title Vindication If a unit of cryptocurrency can indeed be understood as a digital res, and as the potential subject of rights, then further testing questions will arise regarding the nature and operation of those entitlements, if we are to understand the cause(s) of action that might provide a basis for restitutionary relief – in the law of unjust enrichment, in the law of wrongs, grounded in the ‘vindication’ of titles or otherwise. For example, understanding whether the law of unjust enrichment is engaged by a transaction involving cryptocurrency is likely to require closer analysis of the character in law of a particular ‘holding’ of cryptocurrency – whether direct or via indirect or intermediated means – and how this holding might be affected by the transaction.17 In what sense, if any, did this transaction involve an enrichment of the apparent recipient at the prior holder’s expense? A satisfactory answer is also likely, in turn, to require closer analysis of the relationship between entitlements to cryptocurrency and the blockchain transaction record, and of the impact of transactional ‘flaws’ on the transmission of entitlements.18 According to one contested vision of the law, if an apparent transaction has no effect on the prior holder’s entitlement – which persists through the transfer – then the appropriate cause of action is a claim to vindicate that title, in money or in specie, and not a cause of action in unjust enrichment.19 Similarly, closer analysis of the nature in law of a holding of cryptocurrency is necessary for a proper understanding of what wrongs might be committed by a third party’s unauthorised ‘interference’ with the holding. In particular, if it is properly regarded as a ‘pure’ intangible asset, then it will be controversial whether the common law’s tort of conversion – which has historically protected rights to possess

16 See section IV below. For a searching analysis of how common law property law analysis might apply to cryptocurrencies, see Fox (n 10). 17 See further section III.F below. 18 See further section IV.B(ii) below. 19 As assumed in Armstrong DLW GmbH v Winnington Networks Ltd [2012] EWHC 10 (Ch), [2013] Ch 156. The modes by which a legal or equitable entitlement to a specific cryptocurrency holding might be susceptible to ‘vindication’ in this way are considered in Fox (n 10) para 6.101–6.106.

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 335 tangible moveable assets – is capable, via reinterpretation or extension, of providing a basis for relief.20

F. Direct and Intermediated Holdings Legal analysis of the cryptocurrency landscape is made even more complex because there is more than one way in which cryptocurrency can be received, held and dealt with. Users might hold and deal with cryptocurrency directly – themselves holding the address with which specific units of cryptocurrency are associated and the private key that enables dealings with that holding. In practice, however, many users of cryptocurrency systems enlist the services of entities on terms which may require a different understanding of the legal relationships involved. Whilst generalisation is fraught with risks, it seems likely that there will be a material distinction between (i) cryptocurrency exchanges and (ii) wallet service providers. As regards exchanges, it seems very probable that the resulting legal relationship will look more akin to that between a bank and its account-holding customer. Thus, although the user may believe that he ‘owns’ identifiable units of cryptocurrency, it is more likely that the user merely has contractual rights against the service provider to, inter alia, receive an amount of cryptocurrency, equal to the user’s ‘balance’ for the time being, and to have such an amount transferred to third parties on the user’s instructions. This difference will have implications for the bases on which any restitutionary claim might be rationalised, including the important question of who – user or service provider – might bring a claim in unjust enrichment against a recipient of cryptocurrency. The legal relationships involved in the use of an online wallet service are rather more difficult to determine.21 The most appealing analysis is that the user would still be understood as owning the specific cryptocurrency holding which is identified with the addresses held within the wallet – as such, the associated holding would not be at the disposal of the wallet service provider, or available to its creditors. However, depending on the services associated with the wallet, the contractual relationship between user and online wallet service provider would certainly bear some similarities to the service relationships that can sometimes exist between a customer and a bank. It is, above all, likely to entail a custodian relationship, involving a contract for the provision of safe-keeping services in relation to the wallet holder’s private keys. It may also be overlain by an agency relationship if the wallet service provider offers currency exchange services or services in executing payment transactions, on the user’s behalf, via the Bitcoin network, involving the user’s cryptocurrency holding.

G. On- and Off-Blockchain Transactions The phenomenon of intermediated methods of holding cryptocurrencies also requires attention to be given to a wider feature of the cryptocurrency landscape – that transactions involving cryptocurrency may take place both ‘on’ and ‘off ’ the blockchain. For present purposes, an 20 cf OBG Ltd v Allan [2007] UKHL 21, [2008] 1 AC 1; Armstrong v Winnington Networks (n 19); Fox (n 10) para 6.102 (rejecting the application of the possessory torts, in light of OBG v Allan); S Green and J Randall, The Tort of Conversion (Oxford, Hart Publishing, 2009) (advocating an extension of possessory torts to intangible assets). 21 See the very helpful analysis in C Hare, ‘Cryptocurrency and Banking Law: Are There Lessons to Learn?’ in Fox and Green (n 1) paras 9.04–9.17.

336 Stephen Watterson ‘on-blockchain’ transaction is a transaction recorded on the blockchain, which involves a movement of cryptocurrency between addresses – typified by a transaction initiated by a user who directly holds and deals in the cryptocurrency. In practice, however, many transactions may occur ‘off-blockchain’ – involving an adjustment to derivative entitlements only, and not any change in the location of, and title to, any specific units of cryptocurrency. To take one example, a user with an ‘account’ with a cryptocurrency exchange, who gives instructions for the transfer of some or all of its cryptocurrency balance to an account held by another user with the same exchange, is engaging in a transaction that can occur entirely off-blockchain.22 The transfer can be effective via an internal adjustment at the exchange, consisting of a cancelling/discharge of the service-provider’s liability to the initiating user and a corresponding assumption of liability to the recipient user – much like a routine in-house funds transfer between accounts held with the same bank, which does not engage any inter-bank payments system.

H. Pseudonymity Finally, the pseudonymity23 of cryptocurrency systems, like Bitcoin, also presents obvious practical challenges for any system of restitutionary relief. This is not a problem of the substantive law; neither is it a problem that uniquely affects the law of unjust enrichment/restitution. Nevertheless, unless it is feasible in an individual case satisfactorily to identify the individual(s) with whom ‘sending’ and ‘recipient’ addresses are associated, it will be impossible for a given claimant to establish their entitlement to relief from a given defendant.

IV. Conceptual Foundations Moving now from context to legal concepts, further analysis is required if any account of the availability of restitutionary relief for cryptocurrency transactions is to proceed on a secure footing. Above all, it is necessary to investigate the character in law of cryptocurrencies, and the nature and operation of the entitlements that may be held in respect of them.

A. Characterising Cryptocurrencies Insofar as cryptocurrency is susceptible to being ‘held’, ‘transferred’ and ‘received’, what precisely is its proper characterisation in law? The answer will be important for all possible bases of restitutionary liability. What is the nature of the defendant’s ‘enrichment’ ‘at the claimant’s expense’ if a claim is brought alleging unjust enrichment via a receipt of cryptocurrency? Can we conceive of a unit of cryptocurrency existing as an identifiable res, to which a person might hold a ‘title’ which is susceptible to vindication by appropriate legal means? And, reflecting the quality in law of a cryptocurrency holding, what wrong (if any) might be committed by an unauthorised interference with – or appropriation of – cryptocurrency?

22 A point brought out in R Anderson, I Shumaliov, M Ahmed and A Rietmann, ‘Bitcoin Redux’ (University of Cambridge Computer Laboratory, 28 May 2018), www.cl.cam.ac.uk/~rja14/Papers/bitcoin-redux.pdf. 23 See section II above.

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 337 To the extent that cryptocurrency like bitcoin is directly received and held, this receipt and holding probably coincides with a confirmed24 blockchain transaction record of the transfer of a certain amount of cryptocurrency from the address with which it was previously associated to a new address identified with the recipient. From the point of view of the blockchain transaction record, there might then appear to be a transfer to a different digital location. But what exactly is the nature in law of what has been received in this way?

(i) An Intangible ‘Digital’ Asset? Cryptocurrency user-communities commonly talk of cryptocurrency as something which is susceptible to being ‘owned’, ‘bought’ and ‘sold’. Although this presents some challenges for property lawyers, the English common law does seem sufficiently flexible to allow a unit of cryptocurrency (or ‘cryptocoin’) to be reified – to be treated as an asset, of monetary value, which is susceptible to analysis in terms of familiar property law concepts.25 On any view, a unit of cryptocurrency is a purely intangible asset with a digital existence/form. What is described as a ‘unit’ or ‘coin’ of cryptocurrency is an ideational entity only, represented in the real world by a data string.26 This presents classificatory challenges. A unit of cryptocurrency – or its real-world manifestation as a data string – does not have the character of a chose in action, narrowly understood: a direct holding27 of cryptocurrency does not involve a legal claim-right against any particular person which is susceptible to realisation by legal proceedings. It is also not a ‘chose in possession’ in the conventional sense, which encompasses only tangible assets capable of being physically possessed. It has recently been suggested that the new reality of the digital world may justify an extension of these traditional legal categories to accommodate ‘virtual’ or ‘digital’ ‘choses in possession’ as a new form of property,28 but such straining29 is probably unnecessary. The better view is that the category of intangible property is not limited to choses in action, narrowly understood,30 and that a unit of cryptocurrency displays features that make its reification both possible and appropriate. It is not difficult to conceive of a unit of cryptocurrency as a digital res which is associated for the time being, via the relevant digital address, with a particular person, and which is susceptible to exclusive control and disposition by that person (assuming appropriate care in maintaining the custody and confidentiality of the associated private key), within the parameters established by the relevant cryptocurrency system and by general property law.

(ii) Susceptibility to Transfer? A related question is whether a transaction which involves an apparent transfer of a cryptocurrency unit between addresses is properly analysed as a transfer of the same digital asset, albeit 24 For the meaning of ‘confirmation’, see n 6 above. 25 For a comprehensive exploration of this, see Fox (n 10). 26 ibid. 27 See further section III.F above on the differences between direct holdings and indirect/intermediated/account-based holdings. 28 See Financial Markets Law Committee, Issues of Legal Uncertainty Arising in the Context of Virtual Currencies (July 2016) 6–9. 29 This characterisation requires, above all, a glossing of the traditional understanding of ‘possession’. 30 See esp Armstrong v Winnington Networks (n 19) (carbon emissions allowances). See further Fox (n 10) paras 6.28–6.44. cf L Chambers and C Buckingham, ‘Intangible Property and Proprietary Restitution in the High Court: Armstrong v Winnington’ [2013] Lloyd’s Maritime and Commercial Law Quarterly 296, who appear to argue that the term ‘chose in action’ is an umbrella term for all forms of intangible property.

338 Stephen Watterson with a new digitally identified location, associated with a different person.31 If it is, then such a transfer might seem analogous to a transfer of tangible money – coins or notes. Although this conception may hold sway in the popular imagination of user-communities, a closer technical analysis of the nature of the digital asset suggests that the legal reality may be otherwise. As David Fox has recently explained,32 the true position may be that the output of a cryptocurrency transaction is a new digital asset – a data string – which is different from the input, which is consumed in the transaction. Legal consequences may follow, if that is indeed the true position. In particular, a transfer of a unit of cryptocurrency will then bear a closer resemblance to an inter-account transfer of bank money – which does not involve any transfer of an asset, but only matched adjustments of the rights of the transferring customer (which are diminished) and of the recipient customer (which are correspondingly enlarged). As a further corollary, the process of ‘tracking’ a cryptocurrency unit or fractional unit from address to address might then need be understood as a process of tracing from one digital asset into a different digital asset, identifiable as its substitute via the recorded transactional links – rather than a process akin to ‘following’ precisely the same tangible asset from hand to hand. Nevertheless, even if this is the case, this difference may not matter substantially from the point of view of title analysis, if equivalent rules regarding derivative title transfers are applied by analogy to determine the nature and quality of the title which is acquired by the recipient to the new digital asset – as Fox has argued is the case for transfers of bank money.33 On his analysis, this would mean, for example, that if recipient B received a unit of cryptocurrency via a transaction which was the result of fraud practised on sender A, then B would hold the transaction output associated with his address – a new digital asset – subject to A’s equity to rescind; and C, a later recipient via a transaction from B, might in turn acquire title to the transaction output – a new digital asset – subject to a similar defect (if not a bona fide purchaser).34

(iii) ‘Money’ or a ‘Non-money’ Commodity? It follows that a unit of cryptocurrency might well be viewed as a form of intangible digital asset, susceptible to transfer, so as to afford the recipient ‘title’ to that asset, in place of the prior ‘holder’. Yet, for legal purposes, some further classifications may assume importance. In particular, should a cryptocurrency unit be regarded as ‘money’, or is it better regarded as a different form of nonmoney asset/commodity? Sociologists, anthropologists, economists and lawyers will disagree about how money is to be identified, and the premises on which cryptocurrencies might warrant that label. There may not be much disagreement that, to be money, an asset should serve as a (generally accepted) medium of exchange – and related to that, a unit of account and store of value. However, views may well differ as to whether money must be issued by or under the authority of a state, or at least be denominated by reference to the currency of that state.35 31 cf an alternative account, which would suggest that the ‘transferee’ in all cases obtains a new digital asset, which is a product of, or substitute for, the old: see the discussion immediately below, in the text to n 32. 32 See Fox (n 10) paras 6.12–6.19. 33 A fundamental theme of D Fox, Property Rights in Money (Oxford, Oxford University Press, 2008) (dealing with ‘intangible’ ‘bank money’); see further Fox (n 10) paras 6.53–6.56 (extending the same thinking to cryptocurrencies). cf, however, the different perspective on the process of ‘tracing’ money through bank transfers recently offered in T Cutts, ‘Dummy Asset Tracing’ (2019) 135 LQR 140. 34 For exploration of the legal meaning of ‘money’ and theories surrounding it, see, eg Fox (n 33) ch 1; C Proctor, Mann on the Legal Aspect of Money, 7th edn (Oxford, Oxford University Press, 2012) ch 1; Gleeson (n 1); C Proctor, ‘Cryptocurrencies in International and Public Law Conceptions of Money’ in Fox and Green (n 1) esp paras 3.01–3.15. 35 cf the ‘State’ and ‘Societary’ theories of money discussed in Proctor, Mann on the Legal Aspect of Money (n 34) ch 1.

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 339 Such general definitions present several familiar problems for the classification of cryptocurrencies as money. First, at present, these ‘currencies’ function as a medium of exchange only within a limited user-community at best. Indeed, in practice, cryptocurrency may be acquired more frequently by users as a speculative commodity investment – and only in a minority of cases used as a medium of exchange in the purchase of goods or services. Secondly, leading cryptocurrencies like Bitcoin – a ‘self-anchored’36 currency which is not denominated in a fiat currency – have a history of dramatic value/price volatility. If this persists, it is likely to inhibit the w idespread adoption of such currencies as a medium of exchange, and their acceptability as a reliable measure – or store – of value. Thirdly, to date, such cryptocurrencies have not been issued by or under the authority of any state; they are not denominated in the unit of account of any state; and they are not legal tender in any state in which they presently circulate. These circumstances might not remain unchanged forever. Forms of cryptocurrency could conceivably emerge whose genesis, denomination and acceptability/usage enable them to more readily satisfy such general definitions. Consider, for example, the possible emergence of a state-issued/authorised digital currency, denominated in the fiat currency and redeemable at par, which is backed by legal tender rules.37 It would then become particularly pressing to understand why the designation ‘money’ might matter in practice. Many lawyers will probably observe that no single general definition is likely to be useful or appropriate in law.38 Rather, it may be better to ask: what is the specific legal purpose for which any definition is sought? When it comes to the availability of restitutionary remedies – the primary concern of this chapter – there are three distinct reasons why the appropriate characterisation of a unit of cryptocurrency, as money or as something else, could assume particular salience. First, within the modern law of unjust enrichment, it may be material when establishing a defendant’s enrichment to know whether the benefit received is money or is something else which is merely susceptible to valuation in money. It is commonly suggested that a defendant’s enrichment is incontrovertible when money is received, whereas other non-money benefits can present greater difficulty when it comes to determining their objective value in money and because of their susceptibility to so-called subjective devaluation by a defendant. Secondly, within the modern law of unjust enrichment, the standard restitutionary remedy is a monetary award, which reflects the value in money of the enrichment received by the defendant. In the English courts, this generally means the value in pounds sterling; but in an appropriate case, a valuation in foreign currency may be required, or permitted, instead. Might a particular cryptocurrency, if properly regarded as a currency, ever serve the same purpose – affording, on appropriate facts, the measure of monetary value, and of the resulting monetary restitutionary award? Thirdly, for more than two centuries, the English common law has been exceptionally supportive of transfers of money and related instruments when it comes to its rules governing the transfer/acquisition of title – their free commercial circulation is facilitated by generous

36 See B Geva, ‘Disintermediating Electronic Payments: Digital Cash and Virtual Currencies’ (2016) 31 Journal of International Banking Law & Regulation 661. 37 As mooted by the Bank of England, in its One Bank Research Agenda (February 2015) 6. See too the extended discussion of ‘central bank digital currency’ in Gleeson (n 1) ch 8. 38 eg S Green, ‘It’s Virtually Money’ in Fox and Green (n 1), arguing for the adoption of a peculiarly private law conception of ‘money’, which would allow for the recognition of cryptocurrencies as ‘money’, eg for the purposes of regarding an exchange of goods for cryptocurrency as a contract of sale, based only on whether the cryptocurrency is being used as a medium of exchange.

340 Stephen Watterson nemo dat exceptions/title rules, which readily afford a ‘good’ new legal title to bona fide recipients for value.39 May good faith recipients of units of cryptocurrency attract equivalent treatment?40 Whether such benevolent rules apply can have a vital bearing on whether the recipient can obtain a good legal title, despite prior transactional flaws, and, linked to this, on the recipient’s susceptibility to incurring some form of restitutionary liability to a prior ‘owner’ – on the ground of unjust enrichment, wrongdoing or otherwise.

B. Analysing Entitlements to Cryptocurrency as the Subject of Rights If a unit of cryptocurrency is properly treated as an asset, then it must be possible to explain how entitlements to cryptocurrency work. Several pressing questions arise. What is the nature and basis of such title(s)? How might such title(s) be acquired by others, by transfer or o therwise? What circumstances might prevent/impair such acquisition? And what is the relationship between the location and quality of title and the blockchain transaction record? These are not strictly questions for the law of unjust enrichment/restitution – they are (mostly at least) questions of property law. Nevertheless, for reasons that will become apparent, the answers given to these questions might have an important bearing on the nature and basis of restitutionary relief.

(i) Modes of Acquiring Entitlements When it comes to understanding title(s) to cryptocurrencies, there is no obvious reason why – subject to appropriate modifications for the peculiarities of the cryptocurrency asset and the operation of the specific cryptocurrency system – cryptocurrencies should not be susceptible to analysis in terms of familiar property law frameworks. For example, one can readily contemplate: (i) that a legal title might be acquired originally or derivatively; (ii) that within a system like the Bitcoin system, the first legal title to a particular unit of cryptocurrency can be tracked to its genesis as a reward for ‘mining’;41 (iii) that thereafter, legal title might primarily be acquired derivatively, via a legally sufficient dispositive act by a present holder; (iv) that a legal title might also be acquired non-derivatively due to other events;42 and (v) that a familiar range of derivative entitlements might be created in respect of any legal title – eg in the form of a trust or charge – by familiar means. Although this may sound familiar, there may be special features of a unit of cryptocurrency, as a pure intangible asset, that require closer attention. In particular, it is probably necessary to conceive of a unit of cryptocurrency as a res with a present digital location which coincides with the address with which it is presently associated in the blockchain transaction record. It is difficult to conceive of a unit of cryptocurrency as having any existence except as part of the data string that, for the time being, associates the relevant unit with that particular digital location. This immediately suggests another puzzle. As a result of a flawed transaction which is nevertheless recorded in the blockchain, a unit of cryptocurrency becomes disassociated with one 39 eg Miller v Race (1758) 1 Burr 452, 97 ER 398; Clarke v Shee & Johnson (1774) 1 Cowp 197, 98 ER 1041. For analysis, see esp D Fox, ‘Bona Fide Purchase and the Currency of Money’ [1996] CLJ 547; Fox (n 33) chs 2 and 8. 40 For extended consideration, see Fox (n 10) paras 6.57–6.66. 41 See section II above. 42 eg from an original act of taking control by appropriation of cryptocurrency to an ‘address’ identified with and controlled by another.

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 341 person’s address and is newly associated with another’s address. What has happened here in fact and in law? For the time being, the latter has certainly obtained control and at least de facto dispositional capability in relation to the digital res, which the former no longer enjoys – the latter has the capability, now denied to the former, to effect transactions using the private key associated with the new address.43 But what is the significance of these events for the relevant parties’ legal entitlements? This requires closer investigation of the significance of the blockchain transaction record for the allocation of entitlements to cryptocurrency.

(ii) The Significance of the Blockchain Record for Questions of (En)title(ment) On investigation, there are two main ways of understanding how (legal) titles might work within the blockchain system.44 Much may turn on which is preferred. According to the first (and less conventional) title analysis, the blockchain transaction record is conclusive as to the present location of legal title to the digital res. As such, legal title is exclusively identified with the person who can be identified as the ‘holder’ of the address with which the relevant unit of cryptocurrency is presently associated in the blockchain transaction record. This would be so even if there was an off-blockchain flaw in the immediate transaction, pursuant to which the transaction record was altered, or in an earlier transaction in the transaction chain. Whatever the underlying (off-blockchain) facts, if a transaction recorded in the blockchain has resulted in the disassociation of a unit of cryptocurrency from one address and its association with a new address, then legal title will effectively pass to the new address holder – it would vest in the new holder and, as a necessary corollary, would be lost by any prior holder. Following this first title analysis, the title regime applicable to cryptocurrencies would be ‘mono-titular’ at the level of legal title. The blockchain transaction record, which presently associates the relevant unit of cryptocurrency with a particular address, would be the sole basis of legal title for the time being, regardless of the underlying transactional history and the integrity of any underlying (off-blockchain) transaction(s). In practice, the implications of this first title analysis, if adopted, need not be as dramatic as might at first appear. Whilst legal title would invariably pass as a result of the association of the relevant cryptocurrency unit with a new address in the blockchain transaction record, this would not mean that the transaction pursuant to which that shift in title occurred was proper, justified or lawful, and that its effects would be left uncorrected. Indeed, there is no reason to think that the law would universally prioritise the security of receipt of recipients of cryptocurrency, whatever the underlying transactional facts, by leaving them free from liability in all circumstances. On the contrary, it is substantially more likely, and consistent with the general law, that the law would respond to a material flaw in the underlying (off-blockchain) transaction by appropriate corrective means. If the claimant had ‘lost’ title and the defendant had ‘gained’ title through a flawed transaction, then restitutionary relief could be readily justified via the principles of the law of unjust enrichment. This would yield at least a monetary restitutionary remedy. It might also, in an appropriate case, yield specific restitutionary relief – for example, pursuant to the imposition of a trust over the recipient’s title to the cryptocurrency asset in favour of the claimant. The availability of such restitutionary mechanisms would be an important counterbalance to the rigidity of the cryptocurrency legal title regime – on this first analysis – which could readily 43 If Fox’s analysis is correct, then technically, the digital asset – ie the data string, representing the unspent transaction output formerly associated with the former holder’s address – is consumed by the transaction, and the transaction output now associated with the recipient holder’s address is a new digital asset: see section IV.A(ii) above. 44 cf the analysis in Fox (n 10) esp paras 6.49–6.52.

342 Stephen Watterson result in a misallocation of legal title. However, even if softened in this sort of way, the regime would still strongly favour the security of subsequent recipients of cryptocurrency at the expense of prior holders. The first recipient of cryptocurrency pursuant to a flawed transaction would run a significant risk of incurring a restitutionary liability, but subsequent recipients would be far less exposed. This is for a combination of three reasons. First, each subsequent recipient, like the first, would acquire (the) legal title. Secondly, it might prove difficult to reach such remoter recipients via a claim in unjust enrichment, given the newly emphasised requirement for a ‘direct’ enrichment, or equivalent thereof:45 in the absence of a coordinated sequence of transactions,46 this probably requires demonstration of a persisting title-link.47 Thirdly, on appropriate facts, equity might afford the prior holder a new equitable title capable of persisting through later transactions, but this would in practice be readily defeated by a plea of bona fide purchase. Under this first title analysis, any persisting title of a prior holder would be equitable only; and since legal title would vest in successive subsequent recipients without more, they could rely on equity’s general bona fide purchase defence, whatever the proper characterisation of cryptocurrency.48 It is not obvious that the free commercial circulation of cryptocurrencies is so important that later parties should be afforded protection so readily at the expense of (innocent) prior holders. In light of the preceding discussion, a second title analysis might well be preferred. According to this approach, the blockchain record is not definitive of the location of legal title. The present association of a cryptocurrency unit with a particular address in the blockchain transaction record might raise an evidential presumption that the recipient is the owner, having the best legal title.49 It might also, more controversially, be an independent source of a legal title. However, if that is conceivable, such a title would not necessarily be paramount. Consistently with general property law, title acquisition would primarily be a derivative process. As such, the quality of the title of the present holder would depend above all upon the integrity of the underlying (off-blockchain) transaction history. It should be easier to accept the lesser claim involved in this second title analysis – ie that the blockchain record is not conclusive of the location of the legal title, but might be an independent source of a legal title. The law has long recognised that possessors of tangible assets – goods and land – have a status which is independently worthy of protection vis-à-vis strangers, and that their possession is the source of an independent, original possessory title. Might a person’s status as a holder of a unit of cryptocurrency, which arises from its association with their address, bring similar legal consequences? A unit of cryptocurrency is not a tangible asset and, as such, is not susceptible to exclusive physical control in the same way as goods or land. Nevertheless, the association of a unit of cryptocurrency with a particular address is likely to give the address holder control, akin to that enjoyed by a possessor of goods or land, as well as dispositional capabilities. Might this status as holder yield an analogous original control-based title which is not dependent on the integrity of the transaction chain through which any prior holder’s title might have been transmitted?50 If the law did recognise such original, control-based titles to units of cryptocurrency, then it would be operating a multi-titular title regime at the level of legal title. It would allow for the 45 See esp Investment Trust Companies (in liq) v Revenue and Customs Commissioners [2017] UKSC 29, [2018] AC 275. 46 cf Relfo Ltd (in liq) v Varsani [2012] EWHC 2168 (Ch); [2014] EWCA Civ 360, [2015] 1 BCLC 14. 47 cf Lipkin Gorman (a firm) v Karpnale Ltd [1991] 2 AC 548 (HL). 48 As such, there would be no need to create or extend any special common law nemo dat exception, akin to that conventionally applied to money/money-like instruments, in order to protect subsequent good faith recipients for value. 49 See further Fox (n 10) paras 6.50–6.52. 50 cf Fox (ibid), whose analysis suggests that this is a false assumption, since, on a narrow traditional understanding, it is not possible to ‘possess’ an intangible asset, and it is impossible to conceive of a ‘possessory’ title to such an asset. A more accommodating approach might follow from the different premises of Green and Randall (n 20).

Contextual and Conceptual Foundations of Private Law Claims Involving Cryptocurrencies 343 possibility of coexisting (competing) legal titles to the same unit of cryptocurrency, which would then need to be ranked relative to one another on familiar assumptions. How might this work? Within a cryptocurrency system like Bitcoin, the first title to any new ‘coin’ would vest in the successful ‘miner’, as a reward for the miner’s computational effort.51 Thereafter, the primary source of title would be a derivative source – title would derive from a legally effective transfer by the prior holder. This would necessarily bring traps, familiar to property lawyers, which are inherent in systems of derivative title acquisition and are liable to compromise the security of title of successive recipients. Ex hypothesi, under this second title analysis, the paramount (legal) title would not be definitively associated with the present holder of a unit of cryptocurrency, as identified by the blockchain transaction record. That holder might conceivably have an original control-based title, arising from the association of the digital res with their address for the time being, which the law would recognise and protect vis-à-vis strangers. However, whether they were also invested with the paramount (legal) title or only something more fragile would depend on the underlying (off-blockchain) transactional history, viewed through the lens of applicable transactional/ property law. As such, it might depend not only upon the integrity of the immediate transaction, pursuant to which title might immediately have been acquired, but also on preceding transactions in the transaction chain. Consider the possibility of a flaw in an earlier cryptocurrency transaction that, on general principles, would be likely to prevent the transmission of a prior holder’s (paramount) legal title – most likely, because the relevant transaction, pursuant to which the cryptocurrency was disassociated from the prior holder’s address and associated with another’s, occurred wholly without the prior holder’s knowledge and consent, as a consequence of a third party’s misappropriation and misuse of their private key.52 For the time being, the prior holder would retain their better (paramount) legal title and, at least in theory, this (paramount) legal title would remain susceptible to enforcement by appropriate legal means against successive holders of the cryptocurrency. The associated vulnerability in the present holder’s title would not be apparent from the blockchain transaction record, and could only be identified by difficult investigations into the underlying (off-blockchain) facts. This hidden risk for subsequent holders would only be mitigated by one or more familiar devices: (i) the availability of some form of title-clearing rule – in which case, the present holder or any relevant prior holder benefiting from the rule might acquire an unchallengeable paramount title; (ii) rules on limitation or laches – which might prevent the enforcement of the prior title (and associated claims) after the effluxion of a period of time; or (iii) the practical obstacles facing a prior holder in reliably tracking ‘their’ cryptocurrency through later transactions, and in proving the circumstances that might justify their retention of a persisting title in the face of those later transactions.53

V. Conclusion Cryptocurrencies are an emerging and diverse phenomenon. As yet, it is uncertain whether any will attract widespread acceptance as a medium of exchange, or whether they will remain

51 See n 7 above. 52 cf the facts of Armstrong v Winnington Networks (n 19). 53 For a wide-ranging, penetrating analysis of the legal rules that might govern the ‘tracing’ of cryptocurrency movements, see Fox (n 10) paras 6.67–6.100.

344 Stephen Watterson rimarily a speculative investment vehicle. Whatever the path of their future development, p however, it is inevitable that, before long, cases will find their way to court in which private law is looked to for a remedy by those who claim that ‘their’ cryptocurrency holdings have ended up, without justification, in other people’s hands.54 This chapter’s overriding concern has been to lay the contextual and conceptual foundations for subsequent detailed analysis of the availability of unjust enrichment claims under English law. Nevertheless, its discussion has obvious wider relevance for understanding the nature and viability of other forms of private law claim involving cryptocurrencies, whether based on wrongdoing, on the vindication of title or otherwise. It is impossible in the space of this short conclusion to anticipate all questions that will need to be faced. It is sufficient to observe that the exercise of working through the significance of unjust enrichment law for cryptocurrency transactions is likely to prove doubly illuminating. First, this exercise throws into sharp relief some key differences in cryptocurrency ecosystems which are not always sufficiently appreciated and emphasised. Above all, a satisfactory account must give centre stage to the distinction between on-blockchain transactions, involving direct holdings of cryptocurrency, and off-blockchain transactions, involving indirect holdings mediated via an account held with a service provider. This distinction is vital to understanding how in practice a defective transaction involving cryptocurrency can take place – with or without the claimant’s involvement – and the reasons why any resulting transaction might be defective and susceptible to reversal via a restitutionary remedy. However, the same divide will bear on other key issues relevant to liability in unjust enrichment – in particular, whether and in what sense the defendant has been enriched at the claimant’s expense. The difference between direct and indirect holdings is essential for understanding whether and how the claimant’s legal position may be adversely affected via any transaction – direct and indirect holdings are diminished via fundamentally different mechanisms. It also bears on what the defendant receives via the transaction – an accretion to a direct holding is conceptually different from an accretion to an indirect account-based holding. Secondly, the same exercise also compels closer interrogation of some under-tested orthodoxies of unjust enrichment law. The most testing questions probably concern the nature and quantum of any restitutionary liability. It is not controversial that the standard restitutionary remedy is a monetary award, but the disputed designation of cryptocurrencies as currencies forces us to confront the under-examined question of what can and should be selected by the courts as the currency of monetary restitution. If the answer is that the award cannot be made in cryptocurrency, but must instead be made in pounds sterling or another state currency, then the value volatility of cryptocurrencies makes it peculiarly important to be able to provide clear answers to several other basic questions raised within the enrichment inquiry. When is cryptocurrency received by a defendant? At what time precisely does the monetary value of this enriching receipt fall to be determined? And how is that receipt to be valued in money? This last question requires closer examination of the nature of money as an enrichment, and of the basis and limits of the long-standing assumption that a receipt of money is incontrovertibly beneficial.

54 See, analogously, the recent case of Armstrong v Winnington Networks (n 19), which considered the remedies available against a recipient of intangible EU carbon emissions allowances, which had been misappropriated from the claimant’s account with the German registry and transferred, by an imposter, to the defendant’s account with the UK registry.

20 Silence and Solidarity? The Duties of Individual Directors Minded to Speak Out about their Board’s Decision-making and Governance PETER WATTS

I. Introduction There is very little Commonwealth case law on board collegiality as a legal concept, or indeed on specific points from which one might try to extrapolate some general principles about duties of collegiality, if any. This chapter is focused on such restrictions as the law might place on a director’s ability unilaterally to reveal to a shareholder or shareholders information as to the conduct of board meetings (including the director’s own contribution) and other problems the director perceives with the company’s governance. We have two accessible decisions, but little else. Reference is often made to Lord Woolf MR’s remark in Re Westmid Packing Services Ltd that ‘the collegiate and collective responsibility of the board of directors of a company is of fundamental importance to corporate governance under English company law’.1 But that dictum was just the preface to Lord Woolf ’s main point, which was that directors’ duties are imposed on directors individually. Commonwealth companies’ statutes have also tended to say little or nothing on the subject of collegiality. What statutory provisions do often make clear is that, subject to delegation by the board, decisions made by directors on behalf of the company have to be made collectively. Thus, a majority of directors cannot simply dispense with holding meetings, thereby closing out what might be dissenting voices.2 Such provisions, however, do not usually address directors’ personal conduct, nor say anything about the confidentiality of board proceedings. This lack of legal material is perhaps as it should be. So, the paucity of jurisprudence may reflect the fact that the players themselves usually deal with the problems that arise. If a director

1 Re Westmid Packing Services Ltd (No 2) [1998] 2 BCLC 646 (CA) 653. 2 See, eg United Kingdom Companies (Model Articles) Regs 2008 (SI 2008/3229), sch 1: Model Articles for Private Companies Limited by Shares, arts 7, 8.

346 Peter Watts has failed to maintain comity with his or her colleagues, then shareholders might act to remove the director. Other factors are likely to be at play as well. It will, for instance, often be difficult for a company to prove any loss from directors noising the fact of board division. Conversely, the matter may be relatively open and shut because there may exist a binding code of conduct or other contractual provision that clearly covers the ground. Often the disaffected or dissenting director will simply choose to resign. On the other hand, there are reasons for thinking that there could be a greater body of case law than there is on the scope and content of what we might label a director’s ‘duty of collegiality’. Particularly where a non-collegial director not only holds office as such but is also an employee, unilateral action by the company to terminate the employment contract is likely to evoke litigation by the director for wrongful termination. Another factor is the development by modern stock exchanges of what are usually termed ‘continuous disclosure duties’ on listed companies and their boards. These duties may license, and perhaps even require, individual directors, if the board as a whole fails to do so, to communicate to shareholders the fact of board disharmony, particular or general.3 This development may yet generate litigation. This chapter focuses on the simplest case of a director who is not an employee, and on a company that neither has a binding code of conduct nor is listed on a stock exchange. N evertheless, the more complicated variations will not be ignored altogether for two reasons: first, one of the two main cases to be discussed involved an employee-director and a company that was listed on a stock exchange; and secondly, the evolution of continuous disclosure requirements on stock exchanges may come to colour our assessment of the common law position.

II. Some High-Level Observations At this point, a few observations, or tentative propositions, may be useful in approaching the questions we are addressing. First, we are concerned with parties who are in voluntary relationships with one another, and so we might be looking for default rules that conform to what we think are the most generally useful minimum terms the key players would settle for themselves. Insofar as the basis for legal intervention is express or implied undertaking, we are concerned with a process of construction. In the present context, the conclusions the law reaches may be relatively weak presumptions, since even the features of the particular company may be enough to cause a court to depart from the ‘normal’ position. Secondly, the fact that we are trying to identify constructed rules does not mean that we should disown there being any moral dimension to the task. So, it is not necessary to be a human rights advocate to recognise that we are concerned with constraints on a person’s freedom of speech, including their talking about their own actions and experiences. At the same time, it is relatively easy at common law to ‘contract out’ of the right to reveal information. The phrase ‘contract out’ is put in quotation marks to reflect the fact that duties of confidentiality can arise without a formal contract. But still, such duties normally need to be assumed. Ordinarily, one cannot impose a duty of confidence without the recipient of the information’s express or inferred assent. Against this moral concern, a range of pragmatic factors are likely to be relevant.

3 For an example of this from the USA, see Re Hewlett-Packard (SEC Release 55801, 23 May 2007), www.sec.gov/ litigation/admin/2007/34-55801.pdf.

Silence and Solidarity? 347 These include the fact that, without a common understanding of confidentiality, members of a board are likely to be inhibited in what they say and do. Connected to the moral dimension, in setting default rules it is relevant to consider which party is likely to have the whip hand when it comes to modifying the base position. It seems reasonable to assume that that party will be the company. It is, of course, not inconceivable that a particular director might decline appointment unless given an express right to speak out about board conduct and governance. But, without testing the question empirically, it seems more probable that the company will be the party that takes the initiative on rules as to board confidentiality. One might anticipate too that companies will have more experience of negotiating rules of conduct than a director, although it is accepted that not much weight can be put on that assumption. Certainly, codes of conduct for directors that include rules of confidentiality and collegiality are very common. These background observations connect to the fact that, historically at least, the legal relationships between the board and shareholders, and within those two groups of persons, are grounded in the intentions of the incorporators, expressed through a company’s constitution, and, after incorporation, of those parties who control the constitution.4 The controlling parties are usually the shareholders. This does not entail that directors might not attempt to create their own f reedoms through a contract with the company, so long as the constitution does not forbid such contracting. At this point, we should advert to the famous dicta of Greer LJ in John Shaw & Sons (Salford) Ltd v Shaw, which feature strongly in one of the cases we will be scrutinising: If powers of management are vested in the directors, they and they alone can exercise these powers … [Shareholders] cannot themselves usurp the powers which by the articles are vested in the directors any more than the directors can usurp the powers vested by the articles in the general body of shareholders.5

Two things must be noted about these dicta. First, it is often overlooked that the dicta start with the conditional ‘if ’. The delegation of management power to directors is merely a default rule. So, Greer LJ is quite specific that incorporators can, if they wish, give themselves management power by providing for that in the constitution. Other authorities support the proposition that shareholders can do that for a single transaction or for a class of transaction.6 Secondly, it is by no means clear, as a matter of construction, that a delegation to the board of ‘powers of management’ entails that it is the board that gets to govern the relationship between the board and its individual members, on the one hand, and the shareholders, on the other.

III. The Two Cases to be Studied The two cases that we will be looking at provide a striking contrast to one another. They are Stobart Group Ltd v Tinkler,7 a decision in the England and Wales Commercial Court, and

4 For more detail, see P Watts, N Campbell and C Hare, Company Law in New Zealand, 2nd edn (Wellington, LexisNexis, 2016) 230–36. 5 John Shaw & Sons (Salford) Ltd v Shaw [1935] 2 KB 113 (CA). 6 See Automatic Self-Cleansing Filter Syndicate Co Ltd v Cuninghame [1906] 2 Ch 34 (Ch and CA) 38, 41; Bamford v Bamford [1970] Ch 212 (CA) 220. See further P Watts, ‘The Power of a Special Majority of Shareholders, or of All Shareholders Acting Informally, to Override Directors’ [2015] Company and Securities Law Bulletin 89. 7 Stobart Group Ltd v Tinkler [2019] EWHC 258 (Comm).

348 Peter Watts National Roads and Motorists’ Association Ltd v Geeson,8 a decision of the New South Wales Court of Appeal. Stobart not only finds that there are legal duties of collegiality, but also treats a breach as involving breach of fiduciary duty. In NRMA, on the other hand, the court declined to prevent a director reporting to members the way in which a board meeting had been run by the chair, including the content of resolutions that she had put forward but the chair had not allowed to be voted on, and reporting the criticisms she had made of him at the meeting. I proceed by giving an account of Stobart, followed by a critique, confined at that point to the issue of breach of fiduciary duty. An account of NRMA is then given, followed by an assessment of how the approach in that case might assist in addressing the proper scope, if any, of legal duties of collegiality.

IV. Stobart Group Ltd v Tinkler Stobart Group Ltd v Tinkler9 was a complex matter. The judgment of Judge Russen QC is 186 pages in length, and there were 10 formal questions to be answered. These questions arose out of various claims and counterclaims, only a few of which concern the subject matter of this chapter. Important though Stobart is, it contains two features that make it less than a paradigm case. The first is the complication, already alluded to, that the relevant director was also a senior employee subject to employment duties. The second is that the company’s constitution contained a provision that enabled the board to remove a colleague from it. That provision played a key role in the outcome of the case.

A. The Facts of Stobart Stobart Group Ltd (Stobart, or the company) is a Guernsey company listed on the London Stock Exchange. Its business involves owning and supporting various types of infrastructure assets, including London Southend Airport and Carlisle Lake District Airport. Mr Tinkler was one of the founders of the company and at all material times had (and at the time of writing still has) a substantial shareholding in it (7–8 per cent). For 10 years, between 2007 and 2017, he had been the chief executive officer of the company. At the end of that period he remained an executive director, giving half of his time to the company and half to a related business. On 14 June 2018, the company purported to terminate his employment and his directorship. He was then re-elected to the board by the shareholders at the company’s annual general meeting on 6 July 2018, but purportedly removed again the next day by his colleagues, using a provision in the company’s articles of association, Article 89. These events were the main focus of the litigation. The company commenced the proceedings, seeking damages against the director for various alleged breaches of duty while in office,10 including duties of confidentiality, and adding a request for a declaration of its entitlement to dismiss him. Some of the allegations of breach of duty (including for incurring unjustified expenses as employee) were dropped by the company at the conclusion of the

8 National Roads and Motorists’ Association Ltd v Geeson [2001] NSWCA 343, (2001) 40 ACSR 1 (hereafter NRMA (NSWCA)). 9 Stobart (n 7). 10 ibid [720].

Silence and Solidarity? 349 evidence in the trial. The company also sued Mr Tinkler and a number of other defendants on the basis that they had conspired to injure the company by unlawful means. Mr Tinkler sought a counter-declaration that he remained a director, and an injunction against the other directors from taking steps to remove him using Article 89. Mr Tinkler’s counterclaims included a challenge to the board’s transfer of treasury stock to the trustees of the company’s employee-benefit trust. The effect of that had been to re-enliven the voting rights in the shares, thereby diluting the voting power of Mr Tinkler and his supporters. We do not need to concern ourselves further with the detail of this aspect of the case, save to note that Mr Tinkler’s challenge to this transfer, on the basis of the so-called ‘proper purpose’ doctrine, succeeded. The factual narrative of the company’s complaints about Mr Tinkler is presented seamlessly in the judgment, but, in order to focus only on the issue of alleged breaches of boardroom collegiality, not all the picture will be presented here. There are some dangers in doing this, since the court rightly was concerned with the holistic question of whether the company was justified in terminating Mr Tinkler’s employment contract. However, this chapter does not have the space for a complete traversal of the facts. What matters is that while much of the judge’s focus was on the termination of Mr Tinkler’s employment status, in relation to the second time the board purported to remove Mr Tinkler from office, the judge concluded that even if Mr Tinkler had been only a director, his removal would have been justified.11 A central part of the story is that Mr Tinkler and the chairman of the board, Mr Ferguson, fell out in a major way. This was in part due to a personality clash, but also resulted from differing views as to the merits of particular company decisions and more broadly the strategic direction of the company. No criticism can be made of the neutrality with which the judge set out the detailed narrative of what happened. Ultimately, however, the judge came down strongly on the side of Mr Ferguson and against Mr Tinkler in terms of the rights and wrongs of the clash. He, of course, had the benefit of observing them both as witnesses and sitting through all the evidence. Nonetheless, given that one of the principal breaches of duty found against Mr Tinkler was his lack of collegiality in approaching several substantial shareholders to voice his criticisms of Mr Ferguson, the CEO (Mr Brady) and the board as a whole, and to seek the support of those shareholders for the removal of Mr Ferguson, it is not clear from a reader’s perspective that Mr Ferguson’s conduct in trying to have (and succeeding in having) Mr Tinkler removed from the board was much more admirable. We will see that it was accepted by the judge that both parties thought they had the company’s best interests at heart. Mr Tinkler was found by the judge to have been in breach of duty by voicing his unhappiness with the chairman and the board’s decision-making to only some of the company’s shareholders: ‘briefing against the board’. But Mr Ferguson also spoke to several of the same shareholders and others. Perhaps he was justified in doing so, on the basis that he needed to check on Mr Tinkler’s wrongs. But he also, from early on in the burgeoning clash, commenced secret preparations to have Mr Tinkler removed from the board, including consulting with some of his board colleagues and the company secretary, and drafting resolutions for removal. If directors are supposed to act solely as part of a board, as Judge Russen suggested, neither side did that in Stobart. The fact is that clandestine moves are only to be expected in a clash such as occurred in Stobart. Neither side will be open with the other when there is a power struggle. It is not obvious, at least in relation to a director who is not an employee, that the law should not just allow such clashes to play out, finding neither party in breach so long as there has not been inappropriate disclosure of commercially sensitive information about the company’s business.

11 ibid

[46] for submission by the company; ibid [935]–[946] for the holding on the proper purpose challenge.

350 Peter Watts The same might be said as to what occurred when matters started to come to a head in mid-May 2018. On the very same day as Mr Tinkler informed Mr Ferguson that he thought he had sufficient support to requisition a shareholder resolution to remove Mr Ferguson, the company’s CEO informed Mr Tinkler that a Stock Exchange announcement would be made of Mr Tinkler’s removal from the board if he proceeded with the requisition. Mr Ferguson expressly stated that it was untenable for Mr Tinkler as a fellow director to requisition a resolution for his removal. On this the judge seemed, again, to agree.12 But was he right? Vigorous conduct by both sides also occurred once the clash broke out into the open. We have already noted the board’s transfer of treasury stock to the trustees of the company’s employee-benefit trust, which the judge did indeed conclude was wrongful. The board also issued a four-page announcement to the Stock Exchange (the 29 May 2018 Regulatory News Service (RNS) announcement) highly critical of Mr Tinkler, which later led to separate, unsuccessful, defamation proceedings.13 It followed this up with a circular to the company’s employees informing them that the board disharmony, now made public by both sides, should not prevent ‘business as usual’, and that they should relay the same message to the company’s customers and suppliers where appropriate. Mr Tinkler responded by sending to major shareholders and then to most employees a letter that was highly critical of the company’s management. The judge was to find all of these communications to have been inappropriate in content in one or more respect,14 although he concluded that only Mr Tinkler was in breach of his duty to act in the best interests of the company. The final stage of the drama has already been mentioned. The board did purport to remove Mr Tinkler as director, but he managed to get himself re-elected by the shareholders. That victory was short-lived, because the board then purported to remove him again using Article 89. The present proceedings followed directly thereon. There can be little doubt that board disharmony leading to a battle for control of the board can be disruptive to a company’s business. The company’s claim for damages in Stobart was confined to the costs of increased management time, public relations costs and collateral costs incurred in dealing with Mr Tinkler’s actions.15 It might conceivably also have complained about damage to the value of the company’s business. The trial was, however, confined to liability issues, and so the details of compensation were not addressed. The judge did note that the market capitalisation of the company fluctuated considerably, from a high of about £1 billion to a low of about £800 million, in the period of disharmony,16 but he made no findings as to whether any blame for that could be laid at the feet of Mr Tinkler. The question again, however, is whether this sort of event in the life of a company should trigger any response from the law.

B. Reasoning in Stobart as to the Law Judge Russen’s conclusions that a legal response did indeed flow from the way the disharmony played out need now to be described. The judge set the legal scene as he saw it early in the judgment. He opened by stating that ‘this litigation raises some elementary principles of c orporate

12 ibid

[415], [756] and [950]. Tinkler v Ferguson [2019] EWCA Civ 819. 14 See Stobart (n 7) [788] as to the 29 May RNS announcement and [756] as to Mr Tinkler’s letters. 15 Stobart (n 7) [720]. 16 ibid [11]. 13 See

Silence and Solidarity? 351 governance’.17 Insofar as the company was the initiator of the litigation, he then said the following: [T]he court will have well in mind the usual hesitancy in interfering in the internal affairs of companies. Ordinarily, it will not attempt to second-guess the actions of a company’s present directorship, acting either unanimously or by an effective majority. Whether or not the company’s affairs are being conducted in a way that might cause the majority of shareholders to undertake a review of the board’s composition at the next opportunity to do so (including at an EGM requisitioned by a qualifying shareholding) is another matter.18

This led the judge to take a point of principle, which he was to return to later in the judgment, as to the relationship between the directors and the shareholders of a company: The facts that have emerged at trial only serve to highlight what should [be] the clear demarcation between directorial responsibility and investor interest. As the Company’s executive, the Board is collectively responsible for making decisions upon business strategy which it believes to be for the good of the company’s enterprise … On the other side of the line are the Company’s shareholders and employees who have every interest in the outcome of such decisions but no (or no direct) responsibility for them. These same facts highlight the potential harm for the company when one member of the collective not only forms an independent view upon the matter, as is his right, but then seeks to involve shareholders and employees in matters of corporate strategy that are not within their direct or immediate remit.19

The judge then referred to a submission of the company that Mr Tinkler’s actions had involved him ‘breaching the constitutional divide between management and shareholders (and employees) and going behind the backs of the remainder of the Board so as to undermine it’.20 This was a submission that the judge essentially went on to uphold. Much later in his judgment, the judge turned more specifically to directors’ duties. Dealing with a Guernsey company, he was concerned with the common law rather than the Companies Act 2006 (UK). He turned first to the duty to act in good faith and in the company’s best interests. In respect of this, Judge Russen stated: In keeping with what I have said above about the company being the true beneficiary of the director’s fiduciary duties, the interests of the company are usually associated with those of its members as a whole without discriminating between the interests of any majority and minority factions that may exist.21

No quibble can be taken with this last dictum, but, taken together, these dicta can be said to present a director-centred, or establishment-cementing, view of company law. It is true that traditionally courts have been hesitant to interfere in a company’s internal affairs and it is also true that they decline to second-guess the decisions of the board. But this was a case where the board was asking the courts to interfere in the company’s internal affairs – a very different situation. Naturally enough, the company, as claimant, had promoted a director-centred conception of company law. It had relied upon Greer LJ’s dicta in John Shaw & Sons,22 set out earlier in this chapter. The company was applying the dicta, not to a typical management decision, but to the

17 ibid

[4]. [42]. 19 ibid [45]. 20 ibid [46]. 21 ibid [394]. 22 See text to n 4 above. 18 ibid

352 Peter Watts board’s decision to remove one of its members from office, a person who had been elected by shareholders only the day before. John Shaw & Sons was indeed concerned with the power of a board to cause its company to sue some of its directors, even where those being sued held a majority of the shares in the company. But, unlike Stobart’s internally focused decision, John Shaw & Sons involved the recovery of a debt that the board alleged the directors owed the company. Collecting debts is undoubtedly a management matter. In fact, the board’s action failed, and Greer LJ was the only member of the Court of Appeal who addressed the question of a separation of powers. Judge Russen accepted the company’s line of argument.23 Although he included the passage in Greer LJ’s dicta that envisaged that shareholders can take powers away from the board,24 one is left to infer that, to the extent that the judge noticed it at all, that passage was regarded as an exception to a rule. There was no recognition that, historically speaking at least, the true starting point is the reverse. That is, it is the incorporators who determine the company’s rules of governance, and the so-called constitutional divide is little more than a presumption as to their taste. It is for this reason that there is no inconsistency between John Shaw & Sons and the older case, Isle of Wight Railway Co v Tahourdin,25 where nothing in the constitution prevented shareholders from intervening in what were otherwise management matters and the court upheld their intervention. In Stobart, the company otherwise put most of its emphasis on the director’s duty to act in the best interests of the company. In response, Mr Tinkler relied on the duty of individual directors to exercise independent judgement as justifying his conduct. Both sides accepted that there was such a duty, and the judge noted that it appears in the statutory duties to be found in the Companies Act 2006 (UK). Judge Russen traversed the considerable number of cases that each side relied upon under this head. From them, he purported to draw some quite firm propositions that went against the position of Mr Tinkler, whilst rightly accepting that none of the cases bore much relevance to the fact pattern before him. It is necessary to set out his key conclusions at some length: In my judgment, the authorities cited to me by each side support the unsurprising proposition that the duty to exercise independent judgment is one that operates upon each director in the context of him operating as a member of the board of directors. This obligation comes with the office of director and does not carry with it some kind of entitlement or licence for an individual director to go off and do his own thing, independently of the board, in relation to matters that fall within the sphere of management of the company’s business. It is only as a member of the board that the director has been entrusted [with] information about management matters in the first place. Therefore, any discussion by him of those matters with shareholders should either be in the presence of the rest of the board or with the prior approval of the board. If the latter, I would think it generally unlikely that the board would delegate to him the task of speaking on such matters only to some shareholders, but not all, or do so without agreeing the terms of his message to them. As it is difficult to see a case for discriminating between shareholders, the former scenario will usually involve the board presenting its views in a general meeting or in a circular to all shareholders. The individual director should therefore (as appropriate) raise, debate, reflect upon and then decide upon his own position on such matters at the level of the board, either as part of the majority or as a dissenting voice. Only by doing so will he facilitate his fellow directors’ compliance with their own duties to exercise an independent judgment and to act in the best interests of the company. The duty upon each

23 Stobart

(n 7) [46] and [395]. [395]. of Wight Railway Co v Tahourdin (1883) 25 ChD 320 (CA).

24 ibid 25 Isle

Silence and Solidarity? 353 director to exercise an independent judgment exists in order to support the board’s management of the company’s business in an efficient and competent manner. By invoking it to justify what might be, or border upon, freelance activity on his part, a director is likely to hinder rather than contribute to the board’s management of the business.26

Judge Russen went on to say that directors who find themselves with irreconcilable differences with their colleagues can resign, but: For so long as a person occupies the office of director, I do not see how the duty to exercise an independent mind can carry with it any entitlement to speak or act as if he were not a member of the board without responsibilities to that collective decision-making body.27

Endorsing the company’s view of the law, he further stated: Therefore, to the extent that the Company draws a justified distinction between independence of mind and independence of action on the part of an individual director it is not, in my view, seeking to exert a military style conception of corporate authority as Mr Tinkler suggests. On the contrary, it is correctly pointing out that the existence of the duty does not justify renegade action being taken by an individual director in a manner inconsistent with proper management of the Company’s business by the Board.28

Then, finally in this section of his judgment, Judge Russen elaborated on his view that if individual directors were to approach shareholders with their concerns, it must be to all the shareholders in a formal way, and then only after the director has first tried his or her best to address differences through the board. Again, it is necessary to lay out the full passages: The [textbook] statement provides no support for the idea that a director may take any grievance he may have about management matters straight to the shareholders, and still less only to some of them through private conversations … There is, in my judgment … no place for an individual director seeking to ‘pick off ’ particular shareholders, in advance of any EGM, by making private approaches to them, individually, and then airing his own views upon board management matters. The risk of a resulting imbalance of information amongst shareholders, especially if communicated orally at private meetings with only some of them, becomes obvious if that course is adopted. That risk carries with it a real danger that the director will fall foul of his duty to act in the best interests of the company (i.e. for the benefit of its members as a whole) and, at least in terms of the expression of what is in his mind, his duty to the company to exercise an independent mind. That would be so even if the director’s views have been properly aired by him at board level, without them making a sufficient impression on his colleagues. Obviously, something has gone very seriously wrong if the director seeks to short-circuit the board by taking his ‘issues’ over management direct to just some of the shareholders. A director who does that will have forgotten that the company (again, the members as a whole) has delegated management matters to the board of which he forms part.29

Some 300 paragraphs later, Judge Russen then applied the law as he saw it to the facts. Given that the judge had already covered the territory in a generic way, he had little difficulty in concluding that Mr Tinkler had acted in breach of both fiduciary duty and contractual duty. Mr Tinkler had broken his duty to act in the best interests of the company by: • speaking to major shareholders in order to criticise board management and the strategy the board was adopting, and to agitate for the removal of the board chair (Mr Ferguson);30

26 Stobart

(n 7) [413]–[414] (original emphasis). [415]. 28 ibid [417]. 29 ibid [421], [425]. 30 ibid [734]–[735]. 27 ibid

354 Peter Watts • not first raising his concerns with his colleagues on the board to permit them to address those concerns, and if the shareholders approached him first, he should have curtailed discussion rather than proffered his dissenting opinions;31 • confining his discussion to only some of the shareholders: ‘in my judgment, the selective communications which Mr Tinkler had with chosen shareholders prior to [9 June 2018] were all about himself and his personal aim of getting rid of Mr Ferguson’;32 • giving confidential information about a business matter to a person whom Mr Tinkler was intending to propose as a new board chair;33 and • writing a one-sided and misleading letter to all shareholders as part of a personal campaign to get rid of Mr Ferguson,34 then forwarding it to employees in response to a relatively neutral letter the board had sent urging ‘business as usual’.35

C. Critique of the Finding of Breach of Fiduciary Duty in Stobart There is no doubt that Judge Russen states his view of the law very eloquently. But if the account appears persuasive, it is not because it is supported by precedent. And it is respectfully suggested it is too board-centred. This does not entail that it will be argued that Mr Tinkler was not in breach of his employment contract – even senior managers can be taken to be subject to the duty of obedience that is a hallmark of the employment relationship;36 rather, it is in relation to the company law issues that the objections that follow are addressed. As to those issues, the criticisms of Judge Russen’s reasoning fall into two categories. First, it will be argued that Mr Tinkler’s actions were in principle not properly susceptible to challenge under the basic directors’ duty to act in good faith in the company’s interests. Secondly, on the merits of directorial freedom, it is argued that the judgment takes too limited a view of what an individual director is entitled to do if unhappy with board governance and company direction. In other words, assuming there could be some basis for court intervention other than breach of fiduciary duty, what should the law’s default position be? These points of disagreement are obviously connected. So, if one concludes that, at least in relation to some aspects of his conclusions, the judge was incorrect as a matter of principle or policy in finding that the director had acted wrongfully, then those things should certainly not be stigmatised as involving breach of fiduciary duty. But, as signalled at the beginning of this chapter, for the purposes of exposition, the issue of fiduciary duty will be dealt with first; consideration of the proper scope of a duty of collegiality will be deferred until we have looked at NRMA. Before turning to general principle, one should note that in the UK there is now a statutory version of the duty to act in the best interests of the company, found in section 172 of the Companies Act 2006. This provision is not declaratory of the common law and was consciously reformative. The UK has to live with it, but it betrays little understanding of the duty of loyalty as applicable to fiduciaries in general in equity, the company law duty being simply a variant.

31 ibid [736]. 32 ibid [737]–[738]. 33 ibid [747], [749]. 34 ibid [757]. 35 ibid [759]. 36 See British Telecommunications plc v Ticehurst [1992] ICR 383 (CA); University of Nottingham v Fishel [2000] ICR 1462 (QB).

Silence and Solidarity? 355 Happily, because Stobart involved a Guernsey company, only the common law was relevant, and this chapter focuses on the common law still predominant in most of the Commonwealth. One should note, nonetheless, that among the things that section 172 demands that a director have regard to is ‘(e) the desirability of the company maintaining a reputation for high standards of business conduct’. At common law, and in contrast to the section 172 formulation, the duty of a director to act in good faith and as the director perceives is in the best interests of the company is essentially a negative duty.37 The subjective entitlement of the director to judge what is in the company’s interests and the concomitant unavailability of judicial review of a director’s actions38 generally foreclose a curial analysis of what it means to act in a company’s best interests in a positive sense. This entails, or at least tends to entail, that only actions where a director makes no pretence to be acting in the company’s interests are caught by the duty. Mismotivation is the core concept. Even the pursuit of self-interest is not by itself enough to establish a breach, since the pursuit of the company’s interests and one’s own are not incompatible things. So much was made clear by Lord Selborne LC in Hirsche v Sims: If the true effect of the whole evidence is, that the Defendants truly and reasonably believed at the time that what they did was for the interest of the Company, they are not chargeable with dolus malus or breach of trust, merely because, in promoting the interest of the Company they were also promoting their own.39

Where there is mere self-interest, equity’s rules about conflicts of interest may apply, but the consequences of those rules are far less severe, both for the director and third parties, than is the case for a breach of the duty to act in the best interests of the company. In all of this, company law is following the general principles applicable to fiduciary obligations. Thus, much the same point was made by Millett LJ in relation to solicitors in his classic exposition of fiduciary duties in Bristol & West Building Society v Mothew: ‘Breach of fiduciary obligation, therefore, connotes disloyalty or infidelity. Mere incompetence is not enough. A servant who loyally does his incompetent best for his master is not unfaithful and is not guilty of a breach of fiduciary duty.’40 In Stobart, Judge Russen relied on a well-known dictum of Arden LJ in Item Software (UK) Ltd v Fassihi41 for an expansive approach to the duty to act in the company’s best interests. It is useful to set it out here: The duty is expressed in these very general terms, but that is one of its strengths: it focuses on principle, not on the particular words which judges or the legislature have used in any particular case or context. It is dynamic and capable of application in cases where it has not previously been applied but the principle or rationale of the rule applies. It reflects the flexible quality of the doctrines of equity.

The significant finding that Arden LJ made in Fassihi was that there was a positive duty on the relevant director to reveal to the company (effectively his board colleagues) that he had established a competing business and was attempting to divert to it a business opportunity that his company was interested in. But for present purposes, nothing in what Arden LJ said supports the use of the duty where the director, however misguidedly, believes what he or she is doing is in the

37 See

too R Nolan, ‘Controlling Fiduciary Power’ [2009] CLJ 293, 296. Smith Ltd v Ampol Petroleum Ltd [1974] AC 821 (PC) 832. 39 Hirsche v Sims [1894] AC 654 (PC) 660. See also Mills v Mills [1938] HCA 4, (1938) 60 CLR 150, 163. 40 Bristol & West Building Society v Mothew (t/a Stapley & Co) [1998] Ch 1 (CA) 18. 41 Item Software (UK) Ltd v Fassihi [2004] EWCA Civ 1244, [2005] ICR 450 [41]. 38 Howard

356 Peter Watts interests of the company. In Fassihi, the director was undoubtedly not pursuing the company’s best interests and would not have pretended to be doing so. It should also be borne in mind that what the best-interests duty requires is loyalty to the company, not to a director’s colleagues. If there is a legal duty of collegiality, then it should be isolated and described, and not slipped in using the best-interests duty. It is true that in Fassihi, Arden LJ was concerned that there not be a proliferation of duties,42 but she must have meant an unnecessary proliferation. Otherwise, a host of dangers arise if we treat honest but misguided conduct as disloyalty. One can return again to Millett LJ’s analysis in Mothew:43 The expression ‘fiduciary duty’ is properly confined to those duties which are peculiar to fiduciaries and the breach of which attracts legal consequences differing from those consequent upon the breach of other duties. Unless the expression is so limited it is lacking in practical utility. In this sense it is obvious that not every breach of duty by a fiduciary is a breach of fiduciary duty.

No typically fiduciary consequences were in play on the Stobart fact pattern. One can pause to note that it is reasonably clear that directors do not owe fiduciary duties directly to one another. While there is little case law, the issue arose in the New Zealand case, Mathias v Pearce.44 This is a case that has resonance for the Stobart fact pattern. Here, a meatprocessing company, Co X, was formed as a result of a coming together of two sets of people: farmers (some 170 of them) and marketing companies. The farmers held roughly 60 per cent of the shares and the marketers the rest. Under the company’s constitution, the farmers were entitled to appoint three directors and the marketers two. One of the marketing companies, two of whose directors were the two marketing directors of Co X, acquired some shares from farmers. The farmer-directors, alarmed at this development, formed a new company, Co Y, of which they were the only directors. Co Y secretly made offers to farmers to buy their shares conditional on a certain percentage of acceptances. They hoped by this to concentrate control of Co X in their hands. The marketing directors commenced this current litigation, arguing that the farmer-directors were acting in breach of fiduciary duty not only to Co X, but also to them as co-directors. Fisher J held that no breaches of duty had taken place, and, more specifically, that there were no fiduciary duties owed between the directors. He stated:45 In some cases such a relationship is presumed from the inherent nature of the relationship itself, eg the relationship between a solicitor and his or her client, and between a director and his or her company. It has never been suggested that one director owes to a co-director any presumptive fiduciary duties of that nature. Nor has it been suggested that there was any transaction, dealing, discussion or other specific incident between the plaintiffs and the defendants in this case akin to that found in Coleman v Myers where the specifics of the occasion created the necessary relationship of reliance. Since there was no fiduciary duty owed to the plaintiffs there could be no breach.

The facts and result of Mathias suggest that directors do not always have to work as a team, with full candour. It needs to be noted that in Stobart, Judge Russen did not overlook the subjective element to the duty to act in the best interests of the company.46 Later in his judgment, the judge indeed found as a fact that Mr Tinkler intended his actions to be in the best interests of the

42 ibid

[41]. (n 40) 16. 44 Mathias v Pearce (1992) 6 NZCLC 68,102 (NZ HC). 45 ibid 68,111. 46 Stobart (n 7) [396]. 43 Mothew

Silence and Solidarity? 357 company: ‘As I have just said above in relation to intention, I believe Mr Tinkler thought he was acting in the best interests of the Company, at least when viewed in the longer-term.’47 But in the face of these conclusions of law and fact, Judge Russen still found Mr Tinkler’s active uncollegiality to have involved a breach of the loyalty duty. He concluded: Mr Tinkler’s discussions with shareholders were therefore pieces of covert action on his part which he cannot be heard to say somehow involved him acting in the best interests of the Company or exercising and expressing an independent mind. On the contrary, when considered in the light of the other matters addressed below, they can be seen to have had a de-stabilising effect upon the Company’s management.48

One is left to conclude that the judge considered that subjective belief as to the pursuit of the company’s interests did not preclude a finding of disloyalty against a director. One also infers that the judge may have determined that no reasonable director in the position of Mr Tinkler could have believed that his actions were in the interests of the company, and applied a test analogous to the concept of ‘objective dishonesty’ used in the action for dishonest assistance in a breach of fiduciary duty.49 Even if, however, such an approach were appropriate, it is very difficult to see that no reasonable director could have regarded all the types of thing that the judge held Mr Tinkler had done wrong as involving action necessarily contrary to a company’s interests. In this regard, it is important to observe that there was little that was either irrational or immoral in what Mr Tinkler did. It is submitted that the judge’s approach to the alleged wrongs, by deploying the loyalty duty, involved a serious miscategorisation. We should also note the basis on which Judge Russen dismissed the company’s conspiracy action against the parties that Mr Tinkler approached to voice his complaints and to seek their assistance in having the chairperson replaced. The judge did so on the basis that, at least as pleaded, the conspiracy action required the claimant to show that Mr Tinkler was acting purely for selfish purposes.50 But the judge does not appear to have treated as significant the fact that at least one of the alleged conspirators was a representative of a substantial shareholder in the company. One can accept in principle that a shareholder could be liable for conspiring to cause a director to breach fiduciary duties, or for dishonestly assisting such a breach. But it is alarming to contemplate that a shareholder who has been personally approached by a disaffected director to support changing the chairperson of the board could become liable to the company in conspiracy or for dishonest assistance if he or she acts on the director’s approach. Certainly, in the absence of active inducement to breach any duty of collegiality that may exist, it is suggested that the duty should sound only against the director and not against third parties, let alone against shareholders. This is the position, for instance, with the duty of care and skill.51 This goes to Millett LJ’s point, above, about the special nature of fiduciary duties. Fiduciary duties demand the respect of third parties, but they are narrow in compass. As for precedents for his approach, it has already been noted that Judge Russen was not able to cite much authority. Counsel for the company in Stobart cited a number of cases,52 but counsel for Mr Tinkler had little difficulty in showing that none was centrally about duties of collegiality.53 There is in fact precious little case law of any description addressing such duties.

47 ibid

[715]. See also [712]. [740]. Barlow Clowes International Ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 All ER 333 [15]. 50 Stobart (n 7) [712]. 51 Motorworld Ltd v Turners Auctions Ltd [2010] NZHC 113. 52 Stobart (n 7) [405]–[406]. 53 ibid [407]–[408]. 48 ibid 49 See

358 Peter Watts The most pertinent case cited by counsel for the company was Super-Max Offshore Holdings Ltd v Malhotra.54 Popplewell J, sitting in the Commercial Court, concluded that the defendant, who was the claimant company’s executive chairman, had acted inappropriately in conducting a sustained campaign against the CEO and senior executives, and then purporting to suspend them and substitute himself as the CEO. The company, in turn, purported to terminate the defendant’s employment contract using provisions in a shareholders’ agreement that permitted a designated sub-committee of the board to do so. The judge stated:55 ‘In short, Mr Malhotra allowed his sense of grievance as a shareholder to override any respect for the proper corporate governance of the business or the appropriate treatment of senior management.’ The judge’s reasoning was principally directed at determining whether the defendant was in breach of his employment contract, but at the very end of the liability part of his judgment he stated:56 I should record that his conduct also constituted breaches of his fiduciary duties as a director of SMOH. That is not relied on as a separate ground justifying termination, but it may be relevant to the form of relief granted.

The defendant director’s conduct in Super-Max was considerably more egregious than Mr Tinkler’s in Stobart. He purported to make major decisions, including dismissing and appointing the company’s CEO, without involving the board, and in breach of the provisions of the shareholders’ agreement. Whether persistent actions taken without authority by a director on behalf of a company amount to breaches of the duty of loyalty need not be determined here, but the case can provide only marginal support for Judge Russen’s judgment in Stobart. Moreover, it appears that, notwithstanding his removal as an executive director, the defendant in Super-Max remained an ordinary director even after judgment had been given. The shareholders’ agreement did not permit more than termination of the defendant’s status as employee. The defendant retained control of the majority of shares, so he could maintain his position on the board. If the duty of loyalty to the company is not the right place to integrate a duty of collegiality, how might a duty of collegiality be classified? One candidate is the action for breach of confidence. Certainly, much information imparted in the boardroom will be subject to a legal duty of confidence. It is also at least arguable that some actions that have taken place within the boardroom will attract duties of confidence.57 How individual directors voted on particular issues might be one candidate, although it is less clear that directors ought not to be able to reveal how they themselves voted. That subject will be touched upon again, below. But it cannot straightforwardly be said – at least, not on current case law – that non-specific information, such as that there is disharmony on the board, is protected by the action for breach of confidence. Nor is it clear that that action is intended to prevent people expressing their own opinions about the wisdom, or lack of it, of decisions that are already known to the parties with whom they are communicating. Then, there is the consideration whether it is in shareholders’ interests that there is a general rule that they are to be kept in the dark about board disharmony unless the board as a whole resolves to reveal it. This chapter is premised on the traditional view that a solvent company’s interests are to be identified with those of its shareholders as a whole.58

54 Super-Max Offshore Holdings Ltd v Malhotra [2017] EWHC 3246 (Comm). 55 ibid [79]. 56 ibid [118]. 57 Participation in private events has in other contexts been protected by equity: see, most famously, Duchess of Argyll v Duke of Argyll [1967] Ch 302 (Ch). 58 For case law, see Watts et al (n 4) 375–82.

Silence and Solidarity? 359 There is a decision of the New South Wales Supreme Court where the action for breach of confidence was invoked in relation to the disclosure of board deliberations by a disaffected director. It is Canterbury-Hurlstone Park RSL Club Ltd v Roberts.59 It is of limited utility because there is little doubt that the minutes of a board meeting that the defendant director had circulated to non-members contained confidential financial information. The defendant had also conceded that board deliberations were governed by the confidence duty, ‘except to the extent they may contain trivial matters or matters of common or public knowledge’. The body also had a code of conduct that expressly made board deliberations confidential. Nonetheless, Palmer J considered that the defendant’s concession was rightly made. He did not discuss NRMA, the tenor of which is rather different, as we shall see. A stronger basis for a general duty of collegiality would be an implied undertaking. Orthodoxy, however, has it that the relationship between non-executive directors and their company is not contractual.60 It is for this reason that equity, rather than the common law of contract, developed the jurisprudence that set the director’s duty of care.61 As Millett LJ was at pains to point out in Mothew, this equitable duty of care is not a fiduciary duty: It is similarly inappropriate to apply the expression [fiduciary] to the obligation of a trustee or other fiduciary to use proper skill and care in the discharge of his duties. If it is confined to cases where the fiduciary nature of the duty has special legal consequences, then the fact that the source of the duty is to be found in equity rather than the common law does not make it a fiduciary duty.62

It is possible that one might even treat a duty of collegiality as part of the duty of care and skill, since conscious action can be careless. Such an analysis would signal that an allegation that a director’s non-collegiality had broken a duty of care needs to be assessed on its particular facts. Where, however, a director openly criticises colleagues and their decisions, then the duty might be better seen as a stand-alone undertaking. Where the board itself has not complied with its duties to listen seriously to the director’s views, an approach based on an undertaking might also allow the director to argue that the board’s own breach has released the director from the undertaking.

V. National Roads and Motorists’ Association Ltd v Geeson The corporation at the centre of the NRMA case63 was the National Roads and Motorists’ Association Ltd, a company limited by guarantee. Its principal function was to render roadside and other assistance to its 1.8 million members. Its structure and very large membership made it unusual. Although a private body, it was described by the first instance judge in the case as ‘part of the general organisation of society in New South Wales’.64 Its board became disharmonious and, in particular, divided between supporters and opposers of its president. That person was

59 Canterbury-Hurlstone Park RSL Club Ltd v Roberts [2008] NSWSC 845. 60 See, eg Base Metal Trading Ltd v Shamurin [2004] EWCA Civ 1316, [2005] 1 WLR 1157 [76]. 61 See Base Metal Trading (n 60). Admittedly, the interrelationship between common law and equitable duties of care is somewhat problematic in this area. For the possibility of common law liability, see Lagunas Nitrate Co v Lagunas Syndicate [1899] 2 Ch 392 (CA) 435. See too JD Heydon, ‘Are the Duties of Company Directors to Exercise Skill and Care Fiduciary?’ in S Degeling and J Edelman (eds), Equity in Commercial Law (Pyrmont NSW, LawBook Co, 2005). 62 Mothew (n 40) 16–17. 63 NRMA (NSWCA) (n 8). 64 ibid [7].

360 Peter Watts Mr Nicholas Whitlam, the son of former Australian Prime Minister, Gough Whitlam. At the relevant time, Nicholas Whitlam was the subject of separate proceedings by the Australian Securities and Investments Commission (ASIC) relating to alleged contraventions of the Corporations Act 2001 (Cth). Although one cannot tell from the judgment in the case, Mr Whitlam was later exonerated as a result of an appeal to the New South Wales Court of Appeal in those proceedings.65 Three of the defendants were members of the board; the fourth was the publisher of the Sydney Morning Herald newspaper. Two of them had given an interview on television in relation to the affairs of the company, although little of what they said in that interview is revealed in the judgment. A week after the television programme went to air, the board held a meeting (‘the 17 September meeting’) which prompted the present proceedings. At the meeting, the company’s CEO read a statement criticising the appearance of the directors on television, which he said had damaged public perception of the company and unsettled its employees. Once the meeting had proceeded to general business, the first defendant moved a motion that the president step aside during the ASIC proceedings. The president took the view that he could only be removed, not suspended. The second defendant indicated that she would make the way the board voted on the resolution a matter of public record. This led the president to state that if he did not receive an undertaking from every director not to disclose what took place in the boardroom, he would seek an immediate injunction. The three defendant directors declined to give the undertaking, whereupon the president terminated the meeting. The company immediately proceeded to seek an injunction against the defendants in fairly broad terms, including an order against their making public the events that took place at the 17 September meeting. The company relied on its directors’ code of conduct, which contained confidentiality provisions, and on the directors’ duties found in the Corporations Act, including the fiduciary duty of loyalty. The company failed to obtain the injunction at first instance,66 and lost its appeal. The only substantive judgment on appeal was given by Justice Ipp, Acting Judge of Appeal. Early in his judgment, Ipp AJA set the tone for what was to follow. He stated: ‘Not everything said at a board meeting is necessarily confidential and an obligation of confidence does not arise in respect of every item discussed at a meeting of a company’s board of directors.’67 Having dealt with issues of fact, which led to a conclusion that only the second defendant was likely to want to ‘spill the beans’, the judge turned to address the right of the defendants to tell the members of the company about the motion to have the president step aside during the ASIC investigation and about the events which led to the closure of the meeting by the president. Ipp AJA did not strongly differentiate between the company’s reliance on its code of conduct and its invocation of general principles of directors’ duties. Nor did he say anything about whether the company’s use of the loyalty duty was appropriate. The absence of extended discussion is explained by the hearing being an appeal from injunction proceedings, with the judgment delivered less than a fortnight later. Ipp AJA regarded the following factors as justifying the dissenting director’s entitlement to speak to the company’s members about the events at the 17 September meeting:68 the information did not involve trade secrets or details of the company’s business; reporting board dissension would not necessarily be wrong (the existence of some dissension was

65 Whitlam v Australian Securities and Investments Commission [2003] NSWCA 183, (2003) 57 NSWLR 559. 66 National Roads and Motorists’ Association Ltd v Geeson [2001] NSWSC 832, (2001) 39 ACSR 401. 67 NRMA (NSWCA) (n 8) [16]. The cases cited do not, however, provide strong authority for the propositions which they are said to support. 68 ibid [39].

Silence and Solidarity? 361 already public knowledge, but the judge did not seem to regard that as crucial); to the extent that the information would cause embarrassment to individual board members, that was irrelevant to the company’s interests; and an election for half the board membership was to take place the following month, in relation to which election the actions of the president and the extent to which those actions were supported by particular board members might be of legitimate interest to those voting. Ipp AJA then stated: Underlying the view expressed in the preceding paragraph is the proposition that it would be in the interests of the applicant as a whole, particularly in the light of the pending election, for members to be fully informed as to these events so that they would be able to consider, properly, what action should be taken in that connection. Achieving such a situation might readily be thought to be beneficial to the long-term health of the company, notwithstanding any exacerbation of the damage of the kind to which [the CEO] referred in the course of the meeting of 17 September.69

Although Ipp AJA did not in fact base his reasoning on the NRMA’s code of conduct, we should not leave the case without remarking on the code. The code was notable for striking a balance between the board’s interests and that of a dissenting director in communications with shareholders. This reinforces the point that the courts should not set default duties of collegiality too high if, in practice, companies themselves take a more relaxed approach. Under the code, a dissenting director had to discuss any intention to go to the shareholders with the chair first, but did not have to get the chair’s consent. The code even contemplated a director speaking more broadly than just to shareholders. The detail taken from Ipp AJA’s judgment is as follows: Clause 14 provided that directors must not disclose the content of discussion at board meetings ‘outside appropriate and responsible circles within the company with a legitimate interest in the subject of the disclosure, unless that disclosure has been authorised by the company …’ Clause 15 provided that, subject to prior discussion with the president and notification to the board, ‘in the exceptional circumstances where it is in the interests of the company as a whole for disclosure of particular discussions [of board meetings] to be made public’, a director may publicly disclose the contents of those discussions ‘if the director honestly and reasonably believes that it is in the best interests of the company as a whole to do so’. Clause 16 provided that ‘where a decision is not unanimous, a dissenting director may disclose the fact that he/she dissented’.70

VI. Returning to the Substantive Issues and Seeking Answers It will be obvious that both the facts and reasoning in NRMA are of more limited compass than those in Stobart. Nonetheless, the case is very useful for bringing out the idea, expressed tentatively here, that shareholders may have a legitimate interest in knowing about board dissension, and that withholding information about board dissension from shareholders may not be in their long-term interests, even if they take a short-term hit upon receiving the information. It is also possible that, once the existence of dissension is known, fuller disclosure may in fact be in a company’s short-term interests. Next, it would be hard to deny that judges are not best placed to assess where a company’s best interests lie. Companies that want to back extra-legal solutions (such as removing a director)

69 ibid 70 ibid

[41]. [35].

362 Peter Watts with legal ones could always adopt express rules. Such a solution would force courts back into the picture, but if a court made a mess of applying a particular collegiality provision (whether in a contract or constitution), at least no general principle would be thereby established. It may be too that stock exchanges would regulate such provisions on the basis that they are impediments to continuous disclosure, and it is generally in shareholders’ interests to be informed about board division. It is noteworthy that in NRMA the court does not appear to have found it objectionable that the second defendant indicated that she would have revealed how the board had voted on the issue of the president’s suspension. It is not clear whether she intended to reveal how individual directors had voted, or just the numbers in favour or against. One assumes that she would have revealed how she herself had voted, although that could be inferred anyway. It can certainly be accepted that even where the board has made available to shareholders the outcome of a board vote on a particular issue, sensitivities can remain as to how individual board members voted. For example, where the board has voted on an appointment to a senior management position, considerable embarrassment could be caused to individual directors were the candidate or candidates to learn how each director had voted. Nonetheless, it would not follow that there should be a legal rule that directors may not reveal how they voted on an issue where the outcome itself is not secret. The appropriateness of a director’s revealing how they voted could be left for shareholders to judge. Directly revealing how others voted would often, perhaps usually, be inappropriate.71 Even if NRMA is notable for the latitude it afforded a director to reveal to shareholders the precise division of voting on a controversial issue, the case did not involve the problem Judge Russen identified in Stobart, of a director consulting about board problems with only some of the company’s shareholders. In NRMA, the second defendant was not planning to be selective as to which shareholders she would communicate with; she indicated that she would make a public statement about the relevant parts of board business. But Judge Russen’s strong deprecation of directors communicating their unhappiness about their colleagues to only some of the shareholders should be addressed. One might start by conceding that Judge Russen’s all-shareholders-or-none rule is one that might plausibly be adopted by a company. However, one would not put it any higher than that. Such a rule is unlikely to meet the various tests common law jurisdictions apply for the implication of particular terms. It is doubtful, too, whether such a rule should be a term implied in law. In the context of board dysfunction, it is only natural that directors might want to speak to shareholders they know or whom they consider can do something about their concerns. This is particularly the case where, like Mr Tinkler, the director is a substantial shareholder him or herself. It probably ought not to change things that the communicating director is in fact more the problem than the solution. In the context of a battle for control of the board, it seems quite unrealistic to think that, even before any election process for board positions is under way, incumbent directors cannot speak to individual shareholders to seek support for their own or another’s appointment or for the removal of a sitting director. A cone of silence otherwise imposed on directors about perceived difficulties in the running of the company, or about perceived mistakes in a company decision when the decision itself is known to shareholders, is not an obvious default position. Or at least, to repeat the theme of this chapter, it is not self-evident that the law should intrude in the absence of express rules adopted by the relevant organisation.

71 See

Harkness v Commonwealth Bank of Australia Ltd (1993) 32 NSWLR 543, 553.

Silence and Solidarity? 363 One should acknowledge that some statutory insider-trading regimes may inhibit, and even preclude, directors from voicing to individual shareholders their unhappiness with colleagues.72 Regimes that go that far would usually apply only to listed companies, at least where the speaking director was not encouraging the recipient of the information to trade in the company’s shares. What is reasonably clear is that the drafter of the regime will not have intended that it be the vehicle by which incumbent directors make it difficult for others to build support for their removal from office. Although it is only peripheral to the subject matter of this chapter, some comment should also be made about the board’s use in Stobart of the power it had been given to remove colleagues, using Article 89 of the company’s constitution. Sometimes company constitutions give boards power to co-opt members, and where that is the case one might expect the board to have a power of removal. But it is also not uncommon, at least in unlisted companies, for boards to be given a power such as Article 89. In the well-known Privy Council case, Lee v Chou Wen Hsien, it was held that such a power is a fiduciary power and therefore subject to the best-interests duty.73 Nothing was said there, and it does not seem to have been argued, as to whether the power is not also subject to the more rigorous ‘proper purposes duty’.74 That case also did not involve the removal of a director who had been appointed by general meeting; the relevant director was one of the founding directors. There is certainly something arresting about a board’s being given and using a blanket power to remove elected colleagues, let alone a director who was elected by the shareholders only the day before. It is true that the directors who did the removing had also been re-elected by the shareholders, but the shareholders can scarcely have anticipated that those parties would so quickly eject one of the successful candidates. In Stobart, Judge Russen took the view that Article 89 was, like any other power, to be exercised as the board thought fit in the company’s interests:75 ‘[the directors’] duties were owed to the Company, not to the 51.44 per cent of shareholders who had voted in favour of Mr Tinkler’s election’. The parties had, in fact, agreed that the power was also subject to the proper purposes duty.76 Judge Russen nonetheless appeared to disagree with the proposition that that duty required account to be taken by the directors of the fact that the power to decide the composition of the board of directors is ordinarily a shareholder function.77 The judge suggested that the board could legitimately take the view that some shareholders voting for Mr Tinkler had not taken a ‘considered or balanced’ view.78 It is legitimate to ask whether it can really have been the intention of the drafter of Article 89 that a majority of the board could act on nothing more than a bona fide view that the company’s interests would be better served without one of the directors whom the shareholders had just voted onto the board? Admittedly, in Lee v Chou Wen Hsien, Lord Brightman, a very experienced Chancery judge, took a sanguine view of the removal power. Nonetheless, his failure to advert to the proper purposes doctrine was, it is suggested, a serious weakness in the decision. It seems likely that the decision will need careful scrutiny in the light of the subsequent case law on the proper purposes doctrine, not least Éclairs Group Ltd v JKX Oil & Gas plc.79 Moreover, the focus 72 See, eg Financial Markets Conduct Act 2013 (NZ), s 242. The standard insider-trading provision, applicable to all companies, is much narrower: see Companies Act 1993 (NZ), s 149. 73 Lee v Chou Wen Hsien [1984] 1 WLR 1202 (PC). 74 See Companies Act 2006 (UK), s 171(b); Companies Act 1993 (NZ), s 133. 75 Stobart (n 7) [888]. 76 ibid [937]. 77 ibid [941]. 78 ibid [889]. See also [941]. 79 Éclairs Group Ltd v JKX Oil & Gas plc [2015] UKSC 71, [2015] Bus LR 1395.

364 Peter Watts of the judgment was on the relevant director’s argument that the decision to remove him was a nullity, an untenable proposition. It must be arguable that the Article 89 power was intended for use in exceptional circumstances, such as where the director had committed a serious civil or criminal wrong. Of course, Judge Russen did consider that Mr Tinkler had committed serious wrongs in relation to his employment contract. But he did not confine the power to such circumstances. To put it bluntly, the judge’s approach disempowered voters, and was condescending to them, insofar as the shareholders were treated as not having exercised sound judgement. In relation to listed companies, it is noteworthy that the Listing Rules of New Zealand’s main stock exchange provide that no listed company can put restrictions (other than shareholding qualifications) on the personnel who can be nominated for election to the board.80 There do not appear to be similar restrictions in London. One could also say that the pro-establishment approach to Article 89 taken in Stobart is calculated to kill dynamism in corporate affairs.

VII. Conclusions Some brief conclusions should now be drawn. The paucity of case law on the existence and scope of legal duties of collegiality makes it difficult to be emphatic on anything. Even if there were more case law to go by, the many countervailing considerations involved in dealing with disagreement amongst directors, or any other types of co-agent for that matter, make it unlikely that firm rules could safely be laid down. There will be some role for the law of breach of confidence, but it should not be assumed that it has a primary role in the maintenance of collegiality. Otherwise, where no issues of employment law are involved, the courts should be cautious before involving themselves in matters which may best be sorted out by the parties themselves. It is wrong in principle to use fiduciary law to impose duties of collegiality, as happened in Stobart. An honest, if misguided, lack of collegiality does not make a director disloyal. A less preemptive starting point is also likely to be more efficacious. One of the solutions that may bring a court back into the picture is for the company to adopt a binding code of conduct. Construing formal documents is a task which courts are well used to. The court may find that the code already recognises that a balance should be maintained between the interests of the group and the individual director. In that regard, the code in NRMA might provide something of a model. A well-functioning board will be collegial, certainly. But if disharmony arises, it should not be assumed that that fact should be kept from shareholders or that only the chair (or the majority’s spokesperson) can speak to shareholders, whether individually or collectively. There has not been space in this chapter to look at particular continuous disclosure rules frequently now used by stock exchanges. Such rules do not normally give express rights to dissenting directors to go direct to shareholders if they feel the shareholders should know something, but neither do they put collegiality above the shareholders’ rights to know material facts about the company’s operations. The same can be said for the prima facie rights some companies’ statutes give to shareholders to pose questions of the board.81

80 See

81 See,

NZSX/NZDX Listing Rules (6 August 2010), r 3.3.5. eg Companies Act 1993 (NZ), s 178.

Silence and Solidarity? 365 Stobart is wrong, it is suggested, in adopting a blanket prohibition on individual directors speaking to individual shareholders about their concerns with their colleagues, and seeking the support of those shareholders for change in the composition of the board. While most company statutes provide for regular board elections, they also enable stated percentages of shareholders at any time to requisition shareholders’ meetings for the appointment or removal of directors. It is true that many of the standard duties of directors are designed to make directors answerable to the shareholders as a whole, but it simply does not follow that directors ought not to communicate with individual shareholders (particularly when the directors are themselves shareholders). Important too as concerns about insider trading can be, it would be an unfortunate by-product if such concerns worked only to save the skins of incumbent directors. Again, it is suggested, the common law at least should take a hands-off approach.

366