Criminal Sentencing as Practical Wisdom 9781509902613, 9781509902644, 9781509902637

Citation preview

CRIMINAL SENTENCING AS PRACTICAL WISDOM How do judges sentence? In particular, how important is judicial discretion in sentencing? Sentencing guidelines are often said to promote consistency, but is consistency in sentenc­ ing achievable or even desirable? Whilst the passing of a sentence is arguably the most pub­ lic stage of the criminal justice process, there have been few attempts to examine judicial ­perceptions of, and attitudes towards, the sentencing process. Through interviews with Scottish judges and by presenting a comprehensive review and analysis of recent scholarship on sentencing­—including a comparative study of UK, Irish and Commonwealth sentencing jurisprudence—this book explores these issues to present a systematic theory of sentencing. Through an integration of the concept of equity as par­ ticularised justice, the Aristotelian concept of phronesis (or ‘practical wisdom’), the concept of value pluralism, and the focus of appellate courts throughout the Commonwealth on sentencing by way of ‘instinctive synthesis’, it is argued that judicial sentencing methodo­ logy is best viewed in terms of a phronetic synthesis of the relevant facts and circumstances of the particular case. The author concludes that sentencing is best conceptualised as a form of case-orientated, concrete and intuitive decision making; one that seeks individualisation through judicial recognition of the profoundly contextualised nature of the process.

ii 

Criminal Sentencing as Practical Wisdom

Graeme Brown

OXFORD AND PORTLAND, OREGON 2017

Hart Publishing An imprint of Bloomsbury Publishing Plc Hart Publishing Ltd Kemp House Chawley Park Cumnor Hill Oxford OX2 9PH UK

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2017 © Graeme Brown 2017 Graeme Brown has asserted his right under the Copyright, Designs and Patents Act 1988 to be identified as Author 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/, 1998–2017. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-50990-261-3 ePDF: 978-1-50990-263-7 ePub: 978-1-50990-262-0 Library of Congress Cataloging-in-Publication Data Names: Brown, Graeme, (Lawyer), author. Title: Criminal sentencing as practical wisdom / Graeme Brown. Description: Portland, Oregon : Hart Publishing, 2017.  |  Includes bibliographical references and index. Identifiers: LCCN 2016059422 (print)  |  LCCN 2016059680 (ebook)  |  ISBN 9781509902613 (hardback : alk. paper)  |  ISBN 9781509902620 (Epub) Subjects: LCSH: Sentences (Criminal procedure)—Great Britain.  |  Criminal justice, Administration of—Great Britain. Classification: LCC KD8406 .B76 2017 (print)  |  LCC KD8406 (ebook)  |  DDC 345.41/0772—dc23 LC record available at https://lccn.loc.gov/2016059422 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.

For Clair

vi 

PREFACE AND ACKNOWLEDGEMENTS

I would like to thank a number of people for their help and support during the time it has taken to research and write this monograph. My sincere thanks go to the 17 sheriffs and 8 judges who participated in the study and who gave generously of their time during the interviews. The frankness with which respondents discussed their views on sentencing was striking; it is a credit to the Scottish judiciary that they were prepared to discuss the issues so openly. I am extremely grateful to the two former Lords Justice General, the Rt Hon Lord Hamilton and the Rt Hon Lord Gill (both now retired), who granted me permission to undertake the research; to the present Lord Justice General, the Rt Hon Lord Carloway, and the present Lord Justice Clerk, the Rt Hon Lady Dorrian, for their continued support; and to the six Sheriffs Principal (Sheriffs Principal Edward Bowen QC, Alastair Dunlop QC, Bruce Kerr QC, Brian Lockhart, James Taylor and Sir Stephen Young Bt QC, now retired) who granted me permission to approach sheriffs within their sheriffdoms with invitations to participate in the research. Without the kind co-operation of Lord Gill, Lord Hamilton and the Sheriffs Principal, this study would not have been possible. In his capacity as Lord Justice Clerk, and latterly as Lord Justice General, Lord Gill was extremely supportive of my research. His Lordship took great interest in the research and gave me much encouragement. I would like to thank him for his input and for all his sup­ port over the years. This monograph is based on my PhD undertaken at the University of ­Edinburgh. I was privileged to have Professor Richard Sparks (University of ­Edinburgh) and Professor Jac­ queline Tombs (Glasgow Caledonian University) as my supervisors. Professor Tombs’ expe­ rience and expertise in sentencing research was of enormous benefit. I learned a great deal from both her published work and from her guidance. Her advice was invaluable during the planning stages of the empirical research. The input of Professor Tombs and Professor Sparks is gratefully acknowledged and much appreciated. I am also grateful to my two PhD examiners, Professor Andrew Ashworth CBE, QC (University of Oxford) and ­Professor Gerry Maher QC (University of Edinburgh), for their insightful comments and contin­ ued support. Thank you also to Sheriff Frank Crowe, Sheriff of Lothian and Borders at ­Edinburgh. Despite all the assistance I have received from such eminent judges and jurists, any errors and omissions that remain in the book are solely my responsibility. My doctoral studies were generously funded by a studentship from the Scottish Centre for Crime and Justice Research (SCCJR). I am extremely grateful to the SCCJR and to its directors, past and present, for their support. I would also like to express my thanks to Bill Asquith and Francesca Sancarlo at Hart Publishing for their help and encouragement in preparing this book. At the Supreme Courts in Edinburgh, my thanks go to Joseph Moyes, Deputy Principal Clerk of Justiciary, for arranging periods of research leave and to my friend and colleague Ashley Brown for her forbearance during these periods of leave. I have benefited from the

viii 

Preface and Acknowledgements

expertise of Jennifer Findlay and Kayleigh McGarry at the Judges’ Library; their help in obtaining Commonwealth material was particularly appreciated. A project of this size places considerable demands upon one’s family. Both my parents and my wife, Clair Woods-Brown, have been tremendous sources of support and encour­ agement. My heartfelt thanks go to Clair for her love, selflessness and understanding. This book is dedicated to her. I have endeavoured to state the law in Scotland, England and Wales, and the various com­ mon law jurisdictions whose sentencing jurisprudence is considered in this monograph, as at 1 October 2016. It has, however, been possible to take account of certain developments in case law since that date. Dr Graeme Brown Supreme Courts, Edinburgh, January 2017

CONTENTS

Preface and Acknowledgements���������������������������������������������������������������������������������������������� vii

1. Introduction�����������������������������������������������������������������������������������������������������������������������1 Previous Studies of Judicial Decision Making in Sentencing������������������������������������������1 ‘Protecting the Mysteries of their Craft’—Sentencing Research in the English Courts���������������������������������������������������������������������������������2 Previous Research on Scottish Judges and Sentencing������������������������������������������������6 ‘The Human Face of Judging’—Sentencing Research in other Commonwealth Jurisdictions���������������������������������������������������������������������8 Conclusion�����������������������������������������������������������������������������������������������������������������������10 2. Context and Methodology����������������������������������������������������������������������������������������������13 The Importance of Interviewing Sentencers������������������������������������������������������������������13 Methodology��������������������������������������������������������������������������������������������������������������������13 Respondents—Judges and Sheriffs�����������������������������������������������������������������������������13 Gaining Access�������������������������������������������������������������������������������������������������������������14 The Judicial Interviews������������������������������������������������������������������������������������������������17 Thematic Analysis: The Use of Grounded Theory���������������������������������������������������������17 Coding the Data�����������������������������������������������������������������������������������������������������������19 Data Saturation������������������������������������������������������������������������������������������������������������20 Transcribing the Interviews���������������������������������������������������������������������������������������������20 Some Cautions on the Use of the Interview Data����������������������������������������������������������20 The Use of Case Law��������������������������������������������������������������������������������������������������������21 Conclusion�����������������������������������������������������������������������������������������������������������������������23 3. The Instinctive Synthesis and Wise Blending of Penal Aims: A Comparative Study of Sentencing Methodology��������������������������������������������������������25 Introduction���������������������������������������������������������������������������������������������������������������������25 Sentencing Methodology in Canada�������������������������������������������������������������������������������25 The Traditional Approach to the Sentencing Task�����������������������������������������������������25 ‘A Very Human Process’—R v Hamilton and the Modern Methodological View��������������������������������������������������������������������������������28 Sentencing Methodology in Australia�����������������������������������������������������������������������������29 The Instinctive Synthesis versus the Two-Stage Approach to Sentencing—the Decisions in Williscroft and Young�����������������������������������������30 The Instinctive Synthesis Affirmed—the Majority Judgment in Markarian v The Queen���������������������������������������������������������������������������������������33

x 

Contents The Dangers of the Staged Approach and the Primacy of Judicial Discretion—Justice McHugh’s Judgment in Markarian�������������������������������������������������������������������������������������������35 The Need for Individualisation�������������������������������������������������������������������������������36 The Likelihood of More Punitive Sentences����������������������������������������������������������37 The Illusory Nature of the Staged Approach���������������������������������������������������������39 Considerations of Proportionality��������������������������������������������������������������������������40 The Superiority of the Instinctive Synthesis����������������������������������������������������������40 The Importance of Judicial Instinct�����������������������������������������������������������������������41 A Need for Greater Transparency? Justice Kirby’s Dissent in Markarian�����������������43 Developments Post-Markarian—the High Court’s Continued Emphasis on the Instinctive Synthesis�������������������������������������������������45 The Instinctive Synthesis in the Scottish Courts������������������������������������������������������������46 The General Approach������������������������������������������������������������������������������������������������46 A Cautionary Lesson—the Decision in Petch and Foye���������������������������������������������47 Conclusion�����������������������������������������������������������������������������������������������������������������������49

4. Equity and the Rule of Law in Sentencing����������������������������������������������������������������������51 Introduction���������������������������������������������������������������������������������������������������������������������51 The Instinctive Synthesis and the Rule of Law���������������������������������������������������������������51 The Real Meaning of ‘Instinctive Synthesis’���������������������������������������������������������������51 The Rule of Law�����������������������������������������������������������������������������������������������������������54 The Rule of Law and Judicial Discretion��������������������������������������������������������������������55 In Defence of the Instinctive Synthesis—Equity and the Rule of Law��������������������������57 Equity and a Virtue-Centred Theory of Judging�������������������������������������������������������58 Reconciling Equity in Sentencing with the Rule of Law��������������������������������������������60 Judicial Objectivity, the ‘Postmodern’ Sentencer and the Concept of Habitus��������������������������������������������������������������������������������62 Equity in the Sentencing Practice of the Scottish Courts������������������������������������������64 Statutory Minimum Sentences—Judicial Views on Drugs Offences��������������������64 Statutory Minimum Sentences—Shrieval Views on the Scottish Parliament’s Aborted Proposals for Minimum Sentences for Knife Possession���������������������������������������������������������������������������������������������66 The Iniquity of Mandatory Minimum Sentences��������������������������������������������������69 Culpable Homicide��������������������������������������������������������������������������������������������������72 Serious Assault���������������������������������������������������������������������������������������������������������73 Sexual Offences��������������������������������������������������������������������������������������������������������76 Leniency and Mercy in Sentencing�����������������������������������������������������������������������������79 Exercising Leniency in Sentencing��������������������������������������������������������������������������79 Mercy in Sentencing������������������������������������������������������������������������������������������������80 A Place for Mercy in Sentencing?����������������������������������������������������������������������������82 The Application of Mercy���������������������������������������������������������������������������������������83 Judicial Views on Leniency and Mercy—the Findings of the Present Study�����������������������������������������������������������������������������84 Equity and the Imposition of Condign Punishment�������������������������������������������������88 The Problem of the ‘Non-Virtuous’ Sentencer�����������������������������������������������������������94

Contents

 xi

Individualised Notions of Justice��������������������������������������������������������������������������������96 Pursuit of the Chimera—Questions of Consistency and Uniformity����������������������97 Conclusion���������������������������������������������������������������������������������������������������������������������102 5. The Practical Wisdom of Sentencing: A Rationalisation of Intuitive Decision Making and Judicial Discretion�������������������������������������������������107 Introduction�������������������������������������������������������������������������������������������������������������������107 Berlin’s Value Pluralism�������������������������������������������������������������������������������������������������107 Value Pluralism in Sentencing���������������������������������������������������������������������������������������113 The Pluralist Virtues in Sentencing��������������������������������������������������������������������������120 Pluralism and the Need for Autonomy in Decision Making�����������������������������������122 Pluralism and Practical Wisdom�������������������������������������������������������������������������������123 The Ideals of Episteme, Techne and Phronesis���������������������������������������������������������������124 From Novice to Expert—the Dreyfus Model of the Human Learning Process��������������������������������������������������������������������������������127 The Novice Actor�������������������������������������������������������������������������������������������������������127 The Advanced Beginner��������������������������������������������������������������������������������������������128 The Competent Performer����������������������������������������������������������������������������������������128 The Proficient Performer������������������������������������������������������������������������������������������129 The Expert Actor��������������������������������������������������������������������������������������������������������129 Intuitive Decision Making versus Sentencing Information Systems: The Sentencer as an Expert Actor������������������������������������������������������131 Judicial Wisdom—Phronesis and Sentencing���������������������������������������������������������������133 A Reconceptualisation of Sentencing as ‘Art’: The ‘Phronetic Synthesis’����������������136 The ‘Art’ of Sentencing������������������������������������������������������������������������������������������136 A Balancing Exercise����������������������������������������������������������������������������������������������138 The Phronetic Synthesis����������������������������������������������������������������������������������������140 6. Structuring the Sentencer’s Discretion�������������������������������������������������������������������������141 Introduction�������������������������������������������������������������������������������������������������������������������141 The Pursuit of Consistency in Sentencing—Alternatives to the Phronetic Synthesis������������������������������������������������������������������������������������������142 The English Approach to Structuring Discretion through Sentencing Guidelines�����������������������������������������������������������������������������142 The First ‘Phase’—Appellate Guidance����������������������������������������������������������������142 The Second ‘Phase’—the Sentencing Advisory Panel������������������������������������������143 The Third ‘Phase’—the Sentencing Guidelines Council�������������������������������������143 The Fourth ‘Phase’—Presumptively Binding Guidelines and the Sentencing Council������������������������������������������������������������������������������146 The Continued Importance of Individualised Sentencing��������������������������������������151 A Blending of the Phronetic Synthesis with Sentencing Guidance—Principled Discretion in Scottish Sentencing�����������������������������������������������������������������������������154 Appellate Review��������������������������������������������������������������������������������������������������������155 Guideline Judgments�������������������������������������������������������������������������������������������������155 Judicial Receptiveness to Guideline Judgments�������������������������������������������������������157 The Shortcomings of Guideline Judgments�������������������������������������������������������������158 The Scottish Sentencing Council������������������������������������������������������������������������������160

xii 

Contents The Merits of Judicial Self-Regulation���������������������������������������������������������������������162 Principled Discretion�������������������������������������������������������������������������������������������������163 Ripping up ‘the Telephone Book’: Respondents’ Desire for Guidance—not Guidelines—in Sentencing���������������������������������������������������171 Hedgehogs and Foxes on the Bench������������������������������������������������������������������������������176 The English Hedgehog����������������������������������������������������������������������������������������������176 Foxes on the Bench—Value Pluralist Sentencers in other Common Law Jurisdictions��������������������������������������������������������������������179 Judicial Conceptions of the Sentencing Process�����������������������������������������������������������180 The Findings of Previous Studies�����������������������������������������������������������������������������180 England and Wales������������������������������������������������������������������������������������������������180 Scotland������������������������������������������������������������������������������������������������������������������181 Australia�����������������������������������������������������������������������������������������������������������������182 Ireland��������������������������������������������������������������������������������������������������������������������183 Phronetic Synthesisers and Intuitive Over-Riders: The Findings of the Present Study�����������������������������������������������������������������������������������������������183 Structured Sentencing�������������������������������������������������������������������������������������������184 The Instinctive Approach��������������������������������������������������������������������������������������184 Intuitive Over-Riding: The Mixed Approach�������������������������������������������������������187 Conclusion���������������������������������������������������������������������������������������������������������������������191

7. A ‘Seedy Little Bargain with Criminals’? Judicial Discretion and the Guilty Plea Discount����������������������������������������������������������������������������������������195 Introduction�������������������������������������������������������������������������������������������������������������������195 Sentence Discounting for Guilty Pleas: Law and Practice in Scotland������������������������������������������������������������������������������������196 The Traditional Approach to Guilty Plea Discounting��������������������������������������������196 The Statutory Provisions�������������������������������������������������������������������������������������������196 Guidelines on Guilty Plea Discounting—the Decisions in Du Plooy and Spence����������������������������������������������������������������������������������������������196 The Discretionary Nature of Guilty Plea Discounting—the Decisions in Gemmell and Murray�����������������������������������������������������������������������������������������198 Discounting as a Discretionary Process—Gemmell v HMA��������������������������������198 The Discretionary Nature of Discounting Affirmed—Murray v HMA��������������201 Discounts and the Nature of Sentencing Guidelines�������������������������������������������202 Discretion in Discounting—the Findings of the Present Study������������������������������203 Can the Guilty Plea Discount be Justified?�������������������������������������������������������������������206 The Rationale for the Guilty Plea Discount�������������������������������������������������������������206 The Practical Risks of Guilty Plea Discounting�������������������������������������������������������210 The Incentive on the Innocent to Plead Guilty����������������������������������������������������211 Public Confidence in the Criminal Justice System and the Credibility of Sentences������������������������������������������������������������������������������������212 The Penalising of Accused who Exercise their Right to Trial������������������������������215 The Development of ‘a Guilty Plea Culture’ amongst Defence Counsel���������������������������������������������������������������������������������216

Contents

 xiii

‘Double Dipping’—the Interaction of the Discount with Charge/Fact Bargaining����������������������������������������������������������������������������216 The ‘Inherent’ or ‘Unseen’ Sentence Discount�����������������������������������������������������219 Discounting and Individualised Justice����������������������������������������������������������������221 Conclusion���������������������������������������������������������������������������������������������������������������������225 8. The Phronimos and the Metronomic Clockwork Man������������������������������������������������227 How Judges Sentence�����������������������������������������������������������������������������������������������������228 The Phronetic Synthesis���������������������������������������������������������������������������������������������228 The Importance of (Principled) Judicial Discretion������������������������������������������������230 Phronetic Synthesis and Principled Discretion in Practice—Sentencers’ Views����������������������������������������������������������������������������������231 Towards a New System of Guilty Plea Discounting—Suggestions for Reform���������������������������������������������������������������������233 The Phronimos on the Bench�����������������������������������������������������������������������������������������240

Bibliography���������������������������������������������������������������������������������������������������������������������������241 Index��������������������������������������������������������������������������������������������������������������������������������������269

xiv 

1 Introduction The passing of sentence by a judge on a convicted offender comprises the most public stage of the criminal justice process. The sentence imposed is the product of the sentencer’s rea­ soned judgment. Whilst these decisions are often criticised by many—by the public and by academic commentators alike—the actual decision making process in sentencing is poorly understood. This book attempts to address our lack of understanding about how judges sentence. It presents a study of the judicial methodology of sentencing, along with judicial percep­ tions and attitudes towards the sentencing process. The book examines and critically analy­ ses approaches to sentencing through an examination of interviews with 25 serving Scottish judges. It elicits respondents’ views on the nature of the sentencing process; how the judges come to their decisions in sentencing offenders; the importance of judicial discretion; the aims and purposes of sentencing; and the operation of the practice of allowing an offender a discount in sentence in return for tendering a plea of guilty. Following O’Malley (2011: 5) it will be argued that, in order to comply with the demands of justice, sentencing must remain discretionary and the selection of sentence in specific cases must remain exclusively a judicial task. By reference to Isaiah Berlin’s theory of value pluralism and the Aristotelian concept of phronesis (or ‘practical wisdom’), it will be con­ tended that the experience both of the law and of society which judicial officers take to the Bench is such that sentencers should be afforded a significant amount of discretion in discharging the sentencing task. Sentencing, it will be argued, is best conceptualised in terms of what the High Court of Australia has termed an ‘instinctive synthesis’. It is, as the Canadian courts have observed, a very human process: a delicate art based on competence and expertise involving a ‘wise blending’ of penal aims. In recent years, however, sentencers have become obliged to consider a purpose—the so-called ‘utilitarian value’ of an offender’s guilty plea—that is often not seen as a legitimate purpose of sentencing.

Previous Studies of Judicial Decision Making in Sentencing In 1995, Ashworth made the following comments on the need for sentencing research: [T]he social importance of sentencing is a powerful argument in favour of careful research. More ought to be known about the motivation of judges and magistrates. Such knowledge would assist in the formation of sentencing policy, and might also help to extend a form of accountability into this sphere of public decision-making … [T]he public policy arguments for research must be ­confronted. They are surely overwhelming (Ashworth 1995a: 263–64)).

2 

Introduction

It has been suggested, however, that almost as much has been written on the dearth of ­information on judicial views on sentencing as on the findings of such research (Bartels, 2009: 45; 2008: 89). Michael Kirby, former Justice of the High Court of Australia, notes that there is little research on judicial reasoning and decision making that goes beyond that contained in the reasons of the courts in formal published judgments (Kirby, 1999). His Honour continues: ‘Judges are reticent because bound by convention to leave their inner thoughts to the words written in their published opinions. Yet it is impossible to give all of one’s reasons in the manageable space available for published reports’ (ibid). A continuing feature of sentencing as a social practice is the paucity of empirical research on the decision making processes of sentencers (Ashworth, 2002: 232). We do not generally know what judges think about sentencing and how they go about it since the judiciary are infrequent contributors to scholarly debates on sentencing. This clouds the transparency of the process of sentencing and is an undesirable outcome in sentencing systems where significant levels of discretion are retained by the judiciary (Mackenzie, 2003: 288; 2001: 1). Thus, what judges think about sentencing and how they approach the task are largely ‘missing links in sentencing research’ (Mackenzie, 2006a: 1, 2005: 2, 2001: 2; see also Roach Anleu and Mack, 2010: 565). Although the information would clearly be of significant ben­ efit to sentencing research, comparatively little research has been undertaken into judicial methodology in sentencing compared to other aspects of the process (Mackenzie, 2005: 5, 2001: 30). Ashworth (1995a: 263) observes that research into why judges do what they do has long been advocated as a prerequisite of the successful development of sentencing policy. Whilst questions about the extent to which judicial officers exercise discretion and the extent to which their decision making is rule governed can only be answered empirically (Hutton, 2013a: 90), sentencers in many countries seem to resist academic research. As demonstrated by the following review of previous sentencing research, there are very few studies which involve interviews with sentencers (Ashworth, 2002: 232).

‘Protecting the Mysteries of their Craft’—Sentencing Research in the English Courts In England and Wales, there is a history of the judiciary refusing to co-operate in academic research, particularly with research on sentencing. Writing in 1987, ­Pannick considered that: ‘[Judges] discourage sociological studies of their activities, such as sentencing. Like members of the Magic Circle who face expulsion if they explain how the trick is done, judges are eager to protect the mysteries of their craft’ (Pannick, 1987: 10). Judges have traditionally been disinclined to engage in research which might shed light on their working practices. The unwillingness on the part of judges to be interviewed has hindered the development of academic research as interview data is a key tool in under­ standing the workings and culture of powerful groups (Malleson, 1999: 197). The prevail­ ing attitude amongst English judges was generally that academic research had little to offer the judiciary, that it was potentially damaging and that they were not required to co-operate (ibid 198). The best known example of judicial non-co-operation with sentencing research occurred in December 1981. During the previous year, in response to an initiative from the Centre

Previous Studies of Judicial Decision Making

 3

for Criminological Research at the University of Oxford, the Home Office agreed to fund a three-year research programme into sentencing policy and practice in the Crown Court. The study was led by Andrew Ashworth and began in October 1980. Having noted the ‘dearth of reliable information about sentencing practices in the Crown Court’ (Ashworth et al, 1984: 1), Ashworth and his colleagues were concerned about ‘the general level of sen­ tencing and the pressure on the prison system’ (ibid 2). The study aimed to convey a more accurate picture of judicial decision making to the public and to those who might influence penal policy making. It appeared that such a study was opportune given that, in delivering the judgment of the Court of Appeal in R v Bibi in July 1980,1 the Lord Chief Justice, Lord Lane, had acknowledged public concern about consistency in sentencing and had called for sentencers to adopt ‘uniformity of approach’ to the sentencing of crimes of moderate seriousness (Ashworth, 2003: 320). In January 1981, Lord Lane granted permission for a pilot study, the aim of which was to determine the most appropriate research methods and to explore the issues which would be the focus of the main inquiry. His Lordship stated that he would consider the results of the pilot study when deciding whether to permit the main project to go ahead. The pilot study comprised, inter alia, interviews with 25 judges; however, the respondents were not selected randomly but were selected either by the Lord Chief Justice or by the local Presiding Judges (Ashworth et al, ibid 9–10; Ashworth, ibid 320–21). A short report on the pilot stage, together with proposals for the main stage of the pro­ ject, was prepared with advice from the Home Office Research Unit and was subsequently sent to the Lord Chief Justice in October 1981. In December 1981, his Lordship refused permission for the research to proceed any further (Ashworth et al, ibid 5). Ashworth and his colleagues were informed by the Lord Chief Justice that his Lordship did not consider the research to be ‘worth the expenditure of judicial time and public money’. His Lordship took the view that many of the points raised in the pilot study were ‘well known among judges’ and that further research on these issues would therefore be of ‘no assistance’ to the judiciary (ibid 64). Ashworth et al further reported that the Lord Chief Justice considered that ­neither the press, the public, nor politicians would take any notice of the research findings and so, in his Lordship’s view, systematic research of the kind proposed could not be justified (ibid). Lord Gifford QC, a practising barrister and member of the House of Lords, later decried this as ‘self-satisfied complacency … ­characteristic of the legal world’ (Gifford, 1986: 32). A further objection raised by the Lord Chief Justice was that: [R]esearch into the attitudes, beliefs and reasoning of judges was not the way to obtain an accurate picture: sentencing was an art and not a science, and the further judges were pressed to articulate their reasons the less realistic the exercise would become … In his view the available textbooks gave a fairly clear account of the factors which judges take into account in sentencing, and he could not think of any aspects of judicial sentencing upon which research might prove helpful (Ashworth et al, ibid; see also Fox, 1987: 219).

As Ashworth later noted, this amounted to a total rejection of the idea of the c­ riminological study of sentencing, dismissing the possibility that sentencing might be influenced by

1 

(1980) 2 Cr App R (S) 177.

4 

Introduction

s­entencers’ differing views on the aims or the effectiveness of sentences (Ashworth, ibid 322). When the time came to publish the findings of the pilot study, substantial pressures were exerted on Ashworth and his colleagues by the Home Office in an attempt to avoid further alienation of the judiciary. When the report of the pilot study was finally published in 1984, it was denounced by the senior judiciary; the Lord Chancellor even gave a speech in which he questioned the integrity of the researchers (ibid 322–23). Lord Lane’s decision to refuse permission for the study to continue and the reaction of senior judges to the report of the pilot study had a profound effect on empirical research of the judiciary. The pilot study was seen as ‘a shipwreck in the channel’ around which future applicants would require to navigate (Ashworth, ibid 322). Criminologists decided that they were not welcome in the Crown Courts. Any research that was conducted was based only on court records, or observation of court proceedings from the public benches without any interviews or interaction with judges (ibid 323). Following the blocking of the Oxford pilot study, the judiciary established a procedure of its own to deal with applications from researchers for access to the Crown Court. Very few researchers were granted permission to interview judges (ibid). In 1992, for example, Hood published the results of a statistical study into whether race was a factor that influenced sentencing in the Crown Court (Hood, 1992). Hood studied the outcome of a large number of cases involving ethnic minority defendants who had been convicted and sentenced at five Crown Courts serving a metropolitan police force area in the West Midlands in 1989 (ibid 29). Hood received permission from the Lord Chancellor’s Department to obtain information on these cases (ibid 30). The study involved only an examination of Crown Court case files (ibid 35–36). On the issue of conducting interviews with the judges concerned, Hood said this: [S]ome insight into whether judges perceived any special problems relating to the ­sentencing of ethnic minorities would have been valuable in interpreting the findings of the statistical analysis. Unfortunately they were, I was told, instructed ‘by the powers that be’ in the judiciary not to co-operate with this part of the planned study (ibid 37).

For a time at least, to even attempt to undertake research on the professional judiciary was seen as a futile exercise: It was supposed that no one could break into the courts if the staff of the University of Oxford ­Centre for Criminological Research, working with Home Office money and support, could not do so … A criminological taboo remained for a time and the few attempts made to breach it were not very successful (Rock, 1993: 5).

Thus, as Baldwin (2008: 385) notes, members of the senior judiciary have generally been hostile to empirical research projects, tending to view requests from researchers to partici­ pate almost as an impertinence. Although he cites a number of studies in which sentenc­ ing decisions have been examined, Baldwin a­ cknowledges that they have either proceeded in an anonymous, statistical manner and not on a judge by judge basis or, alternatively, have been conducted in experimental settings (ibid). Those researchers who have tried to go beyond the statistics and talk directly to judges have, Baldwin reports, generally been frustrated (ibid 386). The professional judiciary’s supposed ‘distaste for research’ has meant that criminologists and other researchers in England and Wales have only been able to conduct studies with lay magistrates (ibid). Whilst the English Magistrates’ Association and

Previous Studies of Judicial Decision Making

 5

local benches of magistrates have participated in criminological research and have given ­serious consideration to some of the results, the professional judiciary—sentencers who deal with the most serious cases—have traditionally not allowed researchers to get close to their working practices and assumptions in this way (Ashworth, 2003: 319). It appears, however, that the English judiciary is becoming more amenable to academic research. In a study funded by the Esmée Fairbairn Foundation, Hough et al (2003) explored the process by which sentencing decisions are made by judges and magistrates, particularly in relation to cases that are on the borderline between custody and community sentences. The authors conducted focus groups with 80 magistrates and one to one, semi-structured interviews with 48 Crown Court judges, recorders and district judges, along with five mem­ bers of the senior judiciary (ibid x; 4–5). The study emphasised the importance of mitigat­ ing factors in sentencing. This led Jacobson and Hough (2007) to undertake a further study to examine the role of personal mitigation in sentencing in the Crown Court. This study involved observation of sentencing in open court and one to one interviews with sentencers in five Crown Court centres across England. Interviews were conducted with 27 full time judges and 13 recorders (ibid viii; 5–6; see also J­ acobson and Hough, 2011). Also in 2007, Drewry, Blom-Cooper and Blake published their research into the recent history and present day operation of the Civil Division of the Court of Appeal (Drewry et al, 2007). Although the authors conducted semi-structured interviews with 10 of the then 37 Lords Justices of Appeal (ibid 8), the study focused exclusively on the workings of the Court’s Civil Division (ibid 5–6). The most recent studies are those by Dhami (2013a), Fitz-Gibbon (2013), and Darby­ shire (2011). Dhami’s study sought to ascertain the views of Crown Court judges on the operation of English sentencing guidelines. A self-administered ­survey was issued during a judicial training event in 2010. A total of 89 Crown Court judges participated in the study (Dhami, 2013a: 298).2 Fitz-Gibbon’s study sought to examine whether English legal practitioners and the English judiciary considered the mandatory life sentence for murder to be in the best interests of justice (Fitz-Gibbon, 2013: 507). Fitz-Gibbon interviewed six members of the English judiciary, along with 20 practising barristers and three policy rep­ resentatives, to consider arguments for and against the mandatory life sentence for murder (ibid). This study was part of a larger research project examining the comparative effects of homicide law reform in the English, New South Wales and Victorian criminal justice sys­ tems in which in-depth interviews were conducted with 81 judges and practitioners across the three jurisdictions (ibid; see also Fitz-Gibbon, 2016). In the course of research sponsored by the Nuffield Foundation, Darbyshire spent seven years interviewing and shadowing 40 judges from across the full range of courts in ­England and Wales and interviewing a further 37 judges (Darbyshire, 2011: 5–6). Darbyshire reports that her ethnographic study of the judiciary was conducted with the support and a­ ssistance of Sir Igor Judge, the former Lord Chief Justice of England and Wales (ibid). The study aimed to describe by observational research a sample of judges in their working lives, examining, inter alia, the judges’ career backgrounds and aspirations; their day to day work and work­ load; the job of judging; their attitudes towards recent and proposed changes in p ­ rocedure;

2 

The structure and operation of the English sentencing guideline system is considered in Chapter 6.

6 

Introduction

their attitude towards proposed changes to the trial structure; and their ­relationship with the jury (ibid 2). Darbyshire’s report of the way in which her fieldwork was conducted stands in stark contrast to earlier studies of the English judiciary: The amount of information gleaned over the years was overwhelming. I was granted u ­ nlimited access to the research judges and many others and everything that impinged on their work. There was absolute transparency. There was almost no activity from which I was debarred and no ­documentation denied (ibid 10).

Darbyshire considers such unprecedented access to the professional judiciary ‘a prize beyond jewels’ (ibid). Crucially, however, Darbyshire explicitly states that hers was not a study in either judicial decision making or sentencing (ibid 15). She states that there are ‘ample sentencing studies in the UK [sic], which interview magistrates, the primary sen­ tencers’ and cites the same studies listed by Baldwin (2008). In so doing, Darbyshire appears to disregard the fact that Scotland has its own legal system3 and fails to note that a small number of important studies on the sentencing practice of the professional judiciary have been conducted in this jurisdiction.

Previous Research on Scottish Judges and Sentencing Insofar as sentencing research on the judiciary is concerned, it is probably fair to say that more has been achieved in Scotland than in England and Wales (Ashworth, 2003: 326). In 1988, for example, a group of academics at the then Department of Jurisprudence, ­University of Glasgow, published the results of a study examining the use made by the Scot­ tish criminal courts of their power to award payments of compensation from convicted offenders (Maher et al, 1988). The research, funded by the Scottish Office, was seen as mak­ ing a contribution to general issues of the use of sentencing powers and, in particular, to the question of how sentencers adapt to a new disposal (ibid 15). The research was carried out over a three year period between October 1981 and December 1984 (ibid 16). Interviews with various criminal justice professionals were con­ ducted during the second half of the project (ibid). Prior to embarking on the main study, the researchers undertook a pilot study involving semi-­structured interviews with, inter alios, 11 sheriffs across six sheriff courts (ibid 17). Following the s­ ubmission of an interim report based on the pilot study in November 1982, the researchers commenced the main study which involved interviewing 48 sheriffs in 23 towns and c­ ities across Scotland (ibid 18). The range of topics in the questionnaire included g­ eneral attitudes towards compen­ sation via the ­criminal courts and the perceived effects of the system on offenders and victims; the types of offence for which compensation orders were, and were not, made; the relationship between compensation orders and other methods of disposal; and the factors which inclined sentencers to award compensation (ibid 18).

3  As Garland (1999a: xiv) notes, the Scottish criminal justice system differs from the standard r ­ eference points for scholarly discussion which tend still to be the US and the UK and, all too often in such scholarly discourse, ‘the UK’ really means England and Wales. See also Croall et al (2010: 5–8) and McAra (2008: 481).

Previous Studies of Judicial Decision Making

 7

In a later review of the methodology used in the research project, the lead researcher commented: A point which is worthy of public record is the extensive co-operation given to the researchers by those we interviewed and also the considerable goodwill and ­interest shown towards the project. To say the least, our experience augurs well for future co-operation between academics and legal practitioners in the field of criminal justice research (Maher, 1988: 34).

The Scottish Office continued to fund sentencing research. The study by Carnie (1990) sought to record and assess the perceptions of Scottish sentencers towards community ser­ vice as a disposal (ibid 1). Carnie interviewed two High Court judges, 21 sheriffs and 14 lay justices of the peace between June and August 1989. The interviews addressed sentencers’ perceptions of community service in relation to other sentencing options, their views as to the purpose and philosophy of the community service order, and their attitudes towards the operation of community service schemes (ibid 1–2). In 1993 the senior judiciary approached academics at Strathclyde University to assist in developing a Sentencing Information System, or ‘SIS’ (Ashworth, 2003; Hutton, 2013b and 2003: 326; Hutton and Tata, 2000: 316–22; Tata, 2013: 245 and 2010a: 210). Tata reports that implementation of the SIS was phased-in during the 1990s and handed over to the Scottish Courts Service in 2003. It was decided that clerks of court would assume responsibility for maintaining the system and by keeping it up to date by recording infor­ mation on sentenced cases. The quality assurance processes recommended by the univer­ sity team were not, however, followed by the Scottish Courts Service; the entry of data after 2003 was not reliable and the SIS ‘appears to have been left to quietly wither away’ (Tata, 2013: 247 and 2010a: 210; see also Ashworth, 2015: 453). In any event, as Ashworth (2003: 326) notes, even the development of the SIS fell short of the kind of fundamental qualitative research required in order to better understand judicial decision making. Tata et al (2008) undertook a four year qualitative study on the preparation and use of pre-sentence (or social enquiry) reports. The study was funded by the Economic and Social Research Council and examined pre-sentence social enquiry reporting from the perspec­ tive of both social workers and sheriffs (ibid 839). The authors conducted 17 interviews with sheriffs in two sheriff courts after sentencing diets had been observed and conducted five focus group discussions with sheriffs throughout Scotland (McNeill et al, 2009: 425). A total of 26 sheriffs participated (ibid). The aim of the authors was to explore in-depth the communication process between the producers of reports (criminal justice social workers) and their principal consumers (sentencers) (ibid 424; Tata et al, ibid 839; Tata, 2010b: 243). An earlier study by Tata and Hutton (1998) focused on the issues of consistency and dispar­ ity in sentencing. The study, funded by the Scottish Office, was initiated by 10 sheriffs who sat in three separate courts within the same s­ heriffdom. These sheriffs decided that they would find information about their own sentencing practices useful and approached the authors for assis­ tance (ibid 342).The custodial sentencing patterns of the 10 sheriffs were examined by collecting retrospective information on all custodial sentences passed by them over a two year period. The objectives of the study were, firstly, to provide the respondents with information about their own sentencing practices and, secondly, to devise a means of examining the nature and extent of variation in custodial sentencing practices by comparing the sentencing practices of each sheriff (ibid). As these two studies focused on, respectively, pre-sentence

8 

Introduction

reports and sentencing disparity, neither was designed to specifically consider the nature of the decision making process by examining issues such as the methodology of sentenc­ ing, the aims of sentencing, judicial discretion, reception of appeal court judgments in the sheriff court and guilty plea discounting in detail. In any event, and as Hutton himself later noted, the study was conducted on a small scale and cannot be generalised to the country as a whole (Hutton, 2003: 322). One of the most insightful studies of judicial perceptions of the sentencing process and of judicial methodology in sentencing is A Unique Punishment—Sentencing and the Prison Population in Scotland, published by Tombs in 2004. The study was designed to be compa­ rable with the earlier study by Hough et al (2003). It aimed to understand how sentencing practices affect the prison population in Scotland; it sought to identify what might dis­ courage sentencers from making use of custody and what might encourage them to make use of alternative community sentences. In addition to an analysis of government statistics on sentencing and the prison population over the preceding 10 years and an observation of decision making in the sheriff courts, Tombs undertook one to one semi-structured inter­ views with five judges, 34 sheriffs and a stipendiary magistrate. The interviews took place at 16 sheriff courts across Scotland’s six sheriffdoms (Tombs, 2004: 15). In order to gain insight into the factors that influence sentencers in deciding whether or not to impose a custodial sentence, Tombs’ interviews with sheriffs included questions about how they had made specific sentencing decisions in borderline cases involving adult offenders where they had recently passed (i) a prison sentence after giving serious consid­ eration to specific non-custodial options, and (ii) a non-custodial sentence, having seri­ ously considered a short prison sentence (ibid 16). Further work was published based on the research in which Tombs examined the ways in which policy rhetoric has become a reality for sentencers and how this framed sentencers’ practices (Tombs, 2008); in which Tombs and Jagger explored the ways in which sentencers made their decisions to imprison and called for a better understanding by policy makers and legislators of sentencers’ logic (Tombs and Jagger, 2006); and in which Millie, Tombs and Hough examined how sentenc­ ers in Scotland, England and Wales differed in their sentencing decision making. This last study concluded that, despite differences between the two legal systems and criminal jus­ tice structures, the decision making process in the two jurisdictions was in fact remarkably similar (Millie et al, 2007).

‘The Human Face of Judging’—Sentencing Research in other Commonwealth Jurisdictions Important studies on judicial methodology in sentencing have been conducted in Common­ wealth jurisdictions, particularly Canada and Australia. One of the earliest such studies was conducted by Hogarth. Hogarth’s study focused on the sentencing behaviour of 71 magistrates in the Canadian province of Ontario and was published in 1971 as Sentencing as a Human ­Process. Hogarth explained that his research was ‘directed towards formulating generalizations about observed regularities in behaviour’ and, to this end, three principal questions were asked: ‘[a] What does sentencing mean to each individual judge? [b] What common patterns of meaning exist among the judges as a group? [c] What does sentencing mean to us as ­students of the process?’ (Hogarth, 1971: x).

Previous Studies of Judicial Decision Making

 9

Mackenzie (2001: 32) notes that Hogarth’s research remains of interest and importance because of its depth and relative rarity. The three questions posed by Hogarth are substan­ tially the same as those posed both by Mackenzie (ibid) and by the present study, over 40 years later. Three further Canadian studies are also of relevance. Firstly, research into the C ­ anadian judicial system was conducted by McCormick and Greene in the late 1970s and early 1980s; the results were published in 1990 (McCormick and Greene, 1990). Interviews were con­ ducted with 91 judges at all levels of court in Alberta and Ontario; basic biographical data was also collected for a sample of 277 judges across Canada (ibid vi). The authors analysed the appointment of judges, judges’ backgrounds, their attitudes to sentencing and their role in decision making through interviews with the judiciary in Alberta and Ontario. S­ econdly, the Canadian Sentencing Commission conducted a written survey of judges’ opinions on various sentencing issues. The Commission noted in its 1987 report that judges from across Canada responded to the invitation and that many judges included thoughtful com­ ments in their responses (Canadian Sentencing ­Commission, 1987: 281). Finally, Doob and Beaulieu conducted research on the sentencing of young offenders in the early 1990s (Doob and Beaulieu, 1992). A set of four hypothetical cases was sent to approximately 60 youth court judges across ­Canada. Respondents were asked to indicate what sentence they would impose and to answer a series of questions relating to the case (ibid 40). ­Completed ­questionnaires were returned by 43 judges (ibid 39). In Australia, Seifman interviewed 22 county court judges and 16 Supreme Court judges for his study on plea bargaining in the state of Victoria (Seifman, 1982). More recently, ­Bartels conducted interviews with 16 Tasmanian judicial officers for her 2008 doctoral the­ sis at the University of Tasmania entitled Sword or Feather? The Use and Utility of Suspended Sentences in Tasmania (Bartels, 2008; see also Bartels, 2009). Bartels interviewed all six of the judges of the Supreme Court of Tasmania and 10 of the 12 magistrates, canvassing their views on the operation of suspended sentences (Bartels, 2009: 46–47, 2008: 92–93). The study of sentencing behaviour in a jurisdiction outwith the UK which has most rel­ evance to the present research, however, is that conducted by Mackenzie for her PhD thesis at the University of New South Wales entitled A Question of Balance: A Study of Judicial Methodology, Perceptions and Attitudes in Sentencing (Mackenzie, 2001)—later published in How Judges Sentence (Mackenzie, 2005). Mackenzie examined judicial perceptions and attitudes to the sentencing process by conducting interviews with 21 judges of the District Court of Queensland and 10 judges from the Supreme Court of Queensland (both the Trial Division and the Court of Appeal) at Brisbane in November and December 1998 (Mackenzie, 2001: 9; 112–13). Respondents were questioned about how they saw the sentencing function and their role in the process; judicial discretion; the aims and purposes of sentencing, and public opinion and the media. Mackenzie analysed the interview transcripts in detail by comparing and contrasting respondents’ statements on various issues; she was then able to draw conclu­ sions on how the judges in the study perceived the sentencing process and how they went about making sentencing decisions from the analysis of the results (ibid 9–10). This study provides an extremely valuable insight into the minds of sentencers. What Mackenzie says of her interview data is this: The answers are frank, and sometimes provocative and unexpected. Above all, they demonstrate that judges are thoughtful, generally reflective as to their own practice, and at times have strong

10 

Introduction

opinions on contentious issues. In short, what is revealed is the human face of judging, together with a unique insight into how judges go about making sentencing decisions (ibid 37; see also Mackenzie, 2005: 12).

The same can be said of the data collected from the Scottish judges and sheriffs for the present study. Mackenzie reports that, despite geographical distance, cultural differences and the passage of time, the conclusions of her study and those of the English Crown Court study conducted 17 years earlier by Ashworth et al (1984) were remarkably similar. The results of the present study with the Scottish judiciary are, in turn, remarkably similar in places with those of Mackenzie.

Conclusion There has been very little research exploring the issues of judicial sentencing methodology and judicial perceptions of, and attitudes towards, the sentencing process. Such studies as do exist and which have attempted to gain information through the use of judicial inter­ views have tended to have been undertaken by researchers in England and Wales, or in Australian jurisdictions. In Scotland, the most recent research in this vein was published over 10 years ago and so, despite their capacity to contribute to the knowledge and debate in this area, the voices of Scottish sentencers are rarely heard. The present study critically analyses the responses of the 25 judges and ­sheriffs, together with case law (domestic and Commonwealth), findings from previous studies, and the results of a comprehensive review of the literature on ­sentencing—both the substantive law of sentencing and the philosophical ­justifications for a wide sentencing discretion. Con­ clusions are then drawn on the nature of the sentencing task and judicial perceptions and attitudes towards ­sentencing. In so doing, a unique insight is provided into how sentencers in ­Scotland make their decisions. Chapter 2 introduces the qualitative empirical study which grounds the research. The chapter begins with a discussion of the rationale for conducting interviews with the profes­ sional judiciary. The qualitative, empirical research is then considered. Questions of access to the judiciary are examined. A discussion of the conduct of the judicial interviews is pro­ vided, along with certain issues relating to the thematic analysis of the interview data. Chapter 3 examines judicial sentencing methodology from a comparative perspective. The Canadian courts’ emphasis on individualised sentencing and the requirement of pro­ portionality are considered. There follows an examination and critique of Australian sen­ tencing methodology, specifically the tension between the discretion-orientated ‘instinctive synthesis’ approach and the ‘staged’ or ‘two-tiered’ approach which aims to make sentencing more transparent and consistent. The use of the preferred approach—that of the instinctive synthesis—in the Scottish courts is then considered. In Chapter 4 the justification for advancing a view of sentencing based on the i­ nstinctive synthesis model is further considered by reference to its compatibility with the rule of law. The principal argument in favour of the structured approach to sentencing (that the instinctive synthesis is contrary to the rule of law) is addressed by reference to Lawrence Solum’s notion of equity in the application of the rule of law and by recourse to his ‘virtuecentered’ theory of judging (Solum, 2003, 1994). The findings from previous studies on

Conclusion

 11

judicial conceptions of the nature of sentencing are then examined and contrasted with the findings of the present study. In Chapter 5, the operation of a wide judicial discretion in sentencing is explored through an application of Isaiah Berlin’s concept of value pluralism and of the Aristotelian con­ cept of phronesis, or ‘practical wisdom’, as recently re-interpreted by Bent Flyvbjerg (2001). The links between intuition, expert knowledge, information processing and decision mak­ ing are examined in the context of sentencing law and practice. Chapter 6 considers ways in which the discretion afforded to sentencers can be struc­ tured. Through a comparison of the current system of English sentencing guidelines and a consideration of the system of appellate sentencing guidance in Scotland, it is demon­ strated that the judges and sheriffs interviewed for the study are value pluralist decision makers. In sentencing offenders, they routinely employ the concept of phronesis (although they never articulate the sentencing task in such terms). It will be argued that wide judicial discretion in sentencing can be justified because the sentencing task is essentially one of phronetic synthesis with a premium placed on the practical experience, practical wisdom and reasoning of the sentencing judge. In Chapter 7, the guilty plea discount principle is used as a lens through which to examine the core themes discussed throughout the book. The chapter reviews the controversy sur­ rounding the guilty plea discount. It analyses the law and practice in various common law jurisdictions and considers it in relation to the contrast between guideline sentencing and dis­ cretionary sentencing. In particular, the sentence discount jurisprudence is contrasted with the importance of practical wisdom in sentencing, the importance of context and of narra­ tive. By reference to the judicial interviews, it is demonstrated that the system of guilty plea discounting is a matter of concern to sentencers, particularly in the High Court. It is ­demonstrated that defence agents and counsel have come to expect standardised discounts; that the sentencers interviewed for the study regarded the appellate guidance on discount­ ing as too prescriptive; and that the guilty plea discount can in certain cases—especially in tandem with plea/charge bargaining—result in what are perceived by the respondents to be disproportionately low sentences. Chapter 8 outlines the major themes which emerge from the study. It offers certain sug­ gestions for reform of the practice of guilty plea discounting. It also draws conclusions about the nature of judicial sentencing and the sentencing methodology employed by judges in Scotland and in other Commonwealth jurisdictions.

12 

2 Context and Methodology The Importance of Interviewing Sentencers The value of conducting interviews with legal professionals as a means of gaining insight into the operation of the law is well recognised throughout legal scholarship (Fitz-­Gibbon, 2016: 49). Such interview data provides a direct insight into the experiences of those charged with the daily operation of the criminal justice system. Interviews provide firsthand accounts of how the law operates, and offer a unique opportunity for researchers ‘to go beyond the pages of trial transcripts, sentencing judgments and legislation to gain valuable insight into the operation of the criminal courtroom from those who operate it’ (Fitz-Gibbon, 2014: 248–49). With regard to sentencing in particular, the opinions of judges are important and neces­ sary in order to gain insight into the sentencing process (Mackenzie, 2001: 6). Conducting interviews with sentencers illuminates their experiences, motivations and attitudes, whilst also providing a ‘human face’ to judging and the sentencing process (Fitz-Gibbon, ibid 249; Mackenzie, ibid 37). The way in which sentencers perceive the sentencing process, and thus go about their role, is critical both to understanding the process itself and to informing other processes such as law reform (Fitz-Gibbon, ibid 248; Mackenzie, ibid 126; Partington, 2005: 139).

Methodology Respondents—Judges and Sheriffs This book examines and critically analyses the approach to sentencing through an exami­ nation of interviews with 25 serving members of the professional judiciary in Scotland: 17 sheriffs and eight judges. Judges—or, more properly, Senators of the College of Justice (Lords Commissioners of Justiciary when exercising their criminal jurisdiction)—are the senior judicial officers in Scotland. At the time of writing there were 31 judges: the Lord Jus­ tice General and the Lord Justice Clerk (the two most senior judges) and 29 Senators (Judi­ cial Institute for Scotland, 2016: 3). Appointments tend to be made from amongst the ranks of senior advocates practising at the Scottish Bar. Judges sit in the High Court of Justiciary, the ­highest criminal court in Scotland. The High Court is the trial court for major crime, where a single judge sits with a jury of 15 people; it also has appellate jurisdiction in respect

14 

Context and Methodology

of solemn criminal appeals against both conviction and sentence.1 The High Court has exclusive jurisdiction to try murder and rape. It can also try any common law or ­statutory offence except where its jurisdiction has by statute been expressly excluded. In practice the High Court hears cases where, due to the gravity of the crime or the previ­ ous convictions of the accused, the prosecutor is of the opinion that a ­sentence in excess of five years’ imprisonment (the maximum period of ­imprisonment that may be imposed in the sheriff court) would be merited ­(Dickson and Vannet, 2008: 57; 321–22). In common law cases, the High Court’s powers of sentencing are unlimited. It may impose imprison­ ment for any term up to and including life. The High Court’s power of imprisonment may be limited by statute when it is dealing with a statutory offence (Dickson and Vannet, ibid). Most prosecutions take place in the sheriff courts; Scotland’s intermediate c­ riminal courts (see Scottish Government, 2016: 48). The country is divided into six sheriffdoms. Each sher­ iffdom is headed by a Sheriff Principal who is ­responsible for, amongst other things, the efficient disposal of court business in the sheriffdom. The sheriffdoms are in turn divided into 39 sheriff court districts, each with its own sheriff court building. The number of sher­ iffs sitting in each court depends on the volume of business (Dickson and Vannet, ibid 58). At the time of writing, there were 131 permanent sheriffs sitting across Scotland (Judicial Institute for Scotland, ibid). Advocates or solicitors of 10 years’ standing are eligible for appointment to the shrieval Bench. In practice, however, few are appointed with less than a couple of decades of experience in legal practice (White and Willock, 2006: 99). The sheriff court hears cases by way of both summary procedure (where the sheriff sits alone) and solemn procedure (where the sheriff sits with a jury of 15). The sheriff court has a wide criminal jurisdiction. A sheriff sitting summarily may impose a sentence not exceed­ ing 12 months for a common law offence.2 The maximum period of imprisonment on indictment in the sheriff court is five years3 although a case may be remitted by the sheriff to the High Court for sentence.4

Gaining Access In light of what has been discussed in Chapter 1, securing access to the judiciary in order to undertake sentencing research would appear a daunting task. At least one previous attempt to secure such access did not meet with success: Malleson’s doctoral research on the English Court of Appeal (conducted between 1993 and 1996) had to be based on interviews with only retired members of the judiciary after Lord Taylor, Lord Lane’s successor as Lord Chief Justice, refused access to sitting judges (Malleson, 1999: 231, fn 13). No reasons were given for this refusal (Ashworth, 2003: 326). Meanwhile in Scotland, at least one request made by senior academics in the early 2000s to interview judges and sheriffs on their sentencing practices at judicial training events was refused.

1  In September 2015 a Sheriff Appeal Court (‘SAC’) was established. The SAC hears appeals against conviction and sentence arising from summary criminal proceedings. The Bench consists of two or three appeal sheriffs sit­ ting together, depending on the nature of the appeal. 2  Criminal Procedure (Scotland) Act 1995, s 5(2). 3  Criminal Procedure (Scotland) Act 1995, s 3(3). 4  Criminal Procedure (Scotland) Act 1995, s 195.

Methodology

 15

Requests for permission to undertake the present study and for permission to approach judges and sheriffs with invitations to participate were made to the then Lord Justice Gen­ eral (the Rt Hon Lord Hamilton) in October 2009. Permission was also sought at the same time from the then Lord Justice Clerk (the Rt Hon Lord Gill). Permission was immediately granted, Lord Hamilton stating that the ­decision on whether to participate in the study would be one for the individual judge concerned. Permission to conduct the study having been granted, all six Sheriffs Principal were then written to in similar terms later that month. It was explained that permission to conduct interviews had been granted by the Lord Justice ­General. Permission was sought from each Sheriff Principal to approach sheriffs within his sheriffdom with invitations to participate in the research. P ­ ermission to conduct interviews was granted by all six Sheriffs Principal in O ­ ctober and early November 2009. It was originally intended to interview 15 sentencers (three judges and 12 sheriffs). The judicial response to the invitations to participate was, however, far more positive than anticipated. A total of 28 sentencers agreed to participate in the research, comprising nine judges (just over a quarter of the total number of judges at the time) and 19 sheriffs. In the event, 25 sentencers were interviewed (eight judges and 17 sheriffs). Thirteen of the 41 judicial officers (four judges and nine sheriffs) invited to participate in the study did not do so. Ashworth suggests that judicial reluctance to welcome researchers may be due to three substantive reasons: (i) a fear of unfairly adverse criticism, (which is related to the conviction that it is difficult for those outside the judiciary to grasp the nature of the judi­ cial task and the range of issues that have to be taken into account); (ii) apprehension about the use of research findings by the Government; and (iii) beliefs about judicial independ­ ence (Ashworth, 2003: 327). It is possible that some of the 13 judicial officers who declined to participate erroneously considered the research to have been undertaken at the behest of the senior judiciary and/or that it comprised some form of ‘assessment’ of their sentencing practices or ‘test’ of their knowledge of sentencing principles. The end result, however, was that almost 70 per cent of those judicial officers approached agreed to participate in the study. This compares favourably with the study by ­Mackenzie where it is reported that 31 of the 52 judges invited to take part in the study did so, an acceptance rate of almost two thirds ­(Mackenzie, 2001: 112). The participation rate in the present study well exceeds that of a recent study of the Irish judiciary, where only 28 per cent of judges accepted the researcher’s invitation to take part in her study focus­ ing on consistency in ­sentencing ­(Maguire, 2010: 32). It also exceeds the response rate of Dhami’s study on judicial views of English sentencing guidelines, in which 51 per cent of the Crown Court judges invited to participate completed the self-administered survey (Dhami, 2013a: 298). The interviews for the present study took place between late November 2009 and early November 2010. Although it was never intended to compare findings between courts, sher­ iffs were interviewed in five separate sheriff courts in four of the six sheriffdoms across Scotland. Four of the sheriff courts were located in urban areas; the fifth sheriff court was located in a smaller town in a p ­ redominantly rural area. A wide range of judicial experience was reflected in the interviews. The most recently appointed judicial officer had been on the Bench for a little over a year; the most experienced had sat for well in excess of 15 years. Of the eight judges interviewed, six had had mixed practices at the Bar undertaking both criminal and civil work; all six had also served as Advocates Depute, prosecuting in the High Court. The remaining two judges had ­practised

16 

Context and Methodology

exclusively at the civil Bar. Of the 17 sheriffs, four had spent either their whole careers or a significant part of their careers prior to their appointment to the shrieval Bench as prosecu­ tors. Two s­ heriffs had practised at the Bar prior to appointment (one in civil practice, the other in criminal defence). A further three sheriffs began their careers as solicitors before passing advocate, two of whom had a mixed practice and one of whom practised at the civil Bar. The majority of sheriffs had practised as solicitors with either a civil or a mixed civil and criminal practice before their appointment to the Bench. It was never intended that the study should include any form of gender analysis but, for the sake of completeness, four of the 25 judicial officers ­interviewed were female.5 One of the Sheriffs Principal requested a list of sheriffs in one particular sheriff court whom it was intended to approach. The list was duly supplied. The ­Sheriff Principal then wrote back suggesting (but not directing) that a small number of sheriffs should not be approached, principally due either to their imminent ­retirement or relative inexperience in criminal matters. He suggested the names of several other sheriffs who might be willing to participate. As with the correspondence with the Lord Justice General, this Sheriff Principal stressed, however, that the decision of whether to participate had to lie with the individual sheriff. The sheriffs suggested by this Sheriff Principal comprised, in the main, experienced judicial officers who had generally sat in the sheriff court for a number of years. Thus, in respect of one of the five sheriff courts visited during the fieldwork, respondents were to an extent ‘controlled’ by a member of the higher judiciary. A similar experience is reported by Ashworth et al (1984: 9). Ashworth and his col­ leagues report that their intention was to randomly select some 25 judges to approach with a request for an interview. It was made clear to the authors, h ­ owever, that they could not proceed in this way on the South Eastern circuit; they were informed that it would not be appropriate to interview any deputy circuit judges and they were informed that it would be ‘in their interests’ to interview the senior judge at each court they visited. Ashworth et al were thus presented with a list of experienced circuit judges and recorders, of whom they were permitted to request an interview. The authors report that they felt able to accept this as part of their pilot study; however, they state that such ‘pre-selection’ would not have been acceptable in the fuller enquiry which they intended to conduct (ibid). Although in the present study a small number of respondents at one sheriff court were suggested by one of the Sheriffs Principal, it was decided to accept this ‘pre-selection’ for a variety of reasons. Firstly, the scope of the study was limited to 25 members of the profes­ sional judiciary—the number of respondents being the same as the pilot study by Ashworth et al (ibid) in which judicial pre-selection was accepted. Secondly, unlike the study by Ash­ worth et al, the number of pre-selected respondents comprised only a fraction of the total number of sentencers interviewed. Thirdly, it was never the intention to randomly select respondents; respondents tended to be approached on the basis of their being known to me or their having previously discussed the research with me. Fourthly, there was in the event little difference in terms of judicial experience between those sheriffs I had intended to approach for the grant of an interview and those sheriffs suggested by the Sheriff Principal. Fifthly, I did not want the Sheriff Principal concerned to withdraw permission to interview sheriffs at this particular sheriff court or at any other sheriff courts within the particular 5  In order to preserve anonymity, all references to quotes from the four female judicial officers use the mascu­ line personal pronoun.

Thematic Analysis: The Use of Grounded Theory

 17

sheriffdom. Finally, such ‘pre-selection’ as did take place was done by a senior member of the judiciary and not by the executive. ‘Pre-selection’ of respondents would have been completely unacceptable had the exercise been undertaken by civil servants. As Darbyshire notes, for a civil servant to grant or withhold consent to conduct interviews with the judici­ ary would constitute a breach of judicial independence (Darbyshire, 2011: 4).

The Judicial Interviews In-depth, face to face, semi-structured interviews were conducted with the 25 r­ espondents. Semi-structured interviews allow for the inclusion of both direct and open-ended ­questions. This in turn permits comparative analysis of responses whilst still providing sufficient flexibility to seek elaboration and clarification from the respondent (Flynn and Fitz-­ Gibbon, 2011: 908–09; Foddy, 1993: 127–28; Seale and Filmer, 2004: 130). All of the shrieval interviews took place in private in the individual sheriff ’s chambers. All of the interviews with judges took place in private in the ­individual judge’s chambers in Parliament House, Edinburgh. Permission was sought from all respondents to record the interviews. Only two respondents refused, both of whom were judges. In these two cases, detailed notes were taken which were w ­ ritten up immediately after conclusion of the inter­ views. The length of the interviews ranged from 35 minutes to two hours and 15 minutes. Interviews with judges tended to be shorter than those with sheriffs: the average duration of an interview with a judge was just over 50 minutes; with sheriffs the average duration was just over an hour and 15 minutes. In order to preserve anonymity, respondents are identi­ fied only as ‘Judge’ or ‘Sheriff ’ along with a number; for example, ‘Judge 1’, ‘Sheriff 2’, etc. The number allocated to each judicial officer is random and is unconnected with either his or her seniority or the order in which respondents were interviewed. In interviews with judicial officers, gaining the trust and co-operation of one’s respond­ ents is crucial (Mackenzie, 2001: 121). I agree with Mackenzie (ibid) that my being known to a number of respondents as a legal practitioner greatly assisted in securing access to, and subsequently securing the co-operation of, the professional judiciary for the purposes of the research. The overall impression gained of the interviewing process was of engage­ ment and commitment by all respondents. All the judges and sheriffs who participated appeared pleased to have been given the opportunity to discuss the issues. Many respond­ ents expressed a genuine interest in the research both prior to the interview and after its conclusion.

Thematic Analysis: The Use of Grounded Theory An iterative, or recursive, approach to data collection and analysis was used in conducting the judicial interviews, in which the main elements of Glaser and Strauss’s grounded theory were employed. The approach developed by Glaser and Strauss (1967) has been defined as: ‘… theory that was derived from data, systematically gathered and analyzed through the research process. In this method, data collection, analysis, and eventual theory stand in close relationship to one another’ (Strauss and Corbin, 1998: 12).

18 

Context and Methodology

Grounded theory was devised to contest the domination of social science in the 1960s by a small number of ‘grand theories’ (such as that of Karl Marx); it also sought to ­challenge social science’s concomitant reliance on deductive and quantitative methodologies ori­ entated to the verification of existing theory, inductive and qualitative approaches being largely dismissed as unsystematic and unreliable at this time. Of particular importance, grounded theory encouraged researchers to generate their own new theories from empiri­ cal data (Hodkinson, 2008: 84). Grounded theory operates more or less in a reverse direction to traditional research. It starts with data collection and, as data emerge, makes constant comparisons between data and theory (Wakefield, 2011: 81). As Hodkinson explains: [T]he purpose of ongoing data collection in grounded theory is not simply to provide additional cases relating to elements of the theory that have already been demonstrated, but rather to develop the theory further by identifying and comparing cases which illustrate new variations (Hodkinson, ibid 86).

There was thus a repetitive interplay between the collection and analysis of data. Analysis started after transcription of the first two interviews (Hodkinson, ibid 85). The implications of that analysis then shaped the succeeding steps in the data collection process (Bryman, 2012: 566). Data collection and analysis proceeded in tandem and repeatedly referred back to each other (ibid 387; 571). It is also important, however, to recognise the limitations of grounded theory in the context of the present study; in particular, the extent to which theory neutral interview­ ing is possible. For example, one must ask of any study whether the researcher’s personal identity has any significance for the way in which interviewees respond to him or to her (Fitz-Gibbon, 2014: 255). In highlighting the importance of such reflexive inquiry, Cooper makes the point that an adequate conceptualisation of the social world has to include the activity of researching it; the researcher is not simply observing from a position of detach­ ment (Cooper, 2008: 18). It is questionable whether grounded theory does indeed offer us the chance to access untainted truths about the social world.Grounded theorists maintain that through exhaustively following the correct techniques, researchers can minimise the influence of existing theories or assump­ tions, nullify subjective biases and draw conclusions based only on the social realities illus­ trated in their data; that they can, in short, ‘transcend their situatedness’ (Hodkinson, ibid 94). As H ­ odkinson notes, however, researchers will always hold values, assumptions and biases which will always shape their research (ibid; see also Fitz-Gibbon, ibid). Stanley and Wise put it thus: [A]ll knowledge, necessarily, results from the conditions of its production, is contextually located, and irrevocably bears the marks of its origins in the minds and intellectual practices of those lay and professional theorists and researchers who give voice to it (emphasis per original), (Stanley and Wise, 1990: 39).

Strauss and Corbin themselves acknowledge that researchers ‘draw on their own experiences when analysing materials’ (Strauss and Corbin, 1998: 5). As a legal practitioner with previous experience in the areas of criminal law and sentencing who had worked with the Scottish judici­ ary for some eight years as a professional support lawyer, I already knew a considerable amount about sentencers’ logic and the pressures under which they operate. I thus tried as far as possi­ ble to remain self-critical throughout the process of conducting the judicial interviews and to

Thematic Analysis: The Use of Grounded Theory

 19

be sensitive to existing conceptualisations in order to focus the investigations (Bryman, 2012: 574).

Coding the Data Coding refers to the ongoing process of assigning conceptual labels to different segments of data in order to identify themes, patterns, processes and relationships (Hodkinson, 2008: 87). Once interviews were transcribed, the transcripts were repeatedly read in order to ensure immersion in the data: the closer the researcher is to the data, the more likely he or she is to develop a ‘feel’ for it and what it means (King and Wincup, 2008: 36). During the review of the transcripts, labels were assigned to component parts that seemed to be of potential significance (Bryman, 2012: 568). The coding frame thus emerged from the open responses of respondents (Semmens, 2011: 67). These codes assisted in the labelling, sepa­ ration, compilation and organisation of the data (Charmaz, 1983: 186). Data were constantly compared as the interviews were completed and transcribed. Inter­ view data were treated as potential indicators of concepts. Throughout the interview pro­ cess and subsequent transcription exercise, the themes, issues, principles and problems discussed by the judges and sheriffs were comparatively examined to determine the con­ cepts with which they best fitted (Bryman, ibid; Strauss, 1987: 25). Such building of theory ‘from the ground up’ (Charmaz, 1994: 37) ensured that eventual conclusions emerged from, and were traceable back to, the data (Hodkinson, ibid). The coding practice advocated by Charmaz (2006) was followed in analysing the data. This involved the use of two main forms or phases of coding: initial coding and selective, or focused, coding (see also Lofland and Lofland, 1995; cf Strauss and Corbin, 1990). The initial coding stage involved a line by line analysis of the transcripts and the assignation of a code to either every line of text or, latterly, to every paragraph of text in order to provide initial impressions of the data. The subsequent process of focused coding involved empha­ sising the most common codes. The focus turned to those codes that were seen as most revealing about the data and so some initial codes were dropped at this stage (see Bryman, ibid 569). The initial coding exercise produced in excess of a dozen codes. Whilst interesting in themselves and providing a valuable insight into judicial concerns about the wider crimi­ nal justice system, many of these codes were not directly relevant to the aims of the study. Such codes included: inadequate training by the Crown Office and Procurator Fiscal Ser­ vice of procurators fiscal in the sheriff courts; concerns about Crown prosecution policies in respect of certain offences (particularly domestic abuse and certain sexual offences); and the attitude of the police to certain offences. Other codes were of more relevance to the study but time constraints precluded further investigation; for example, the need for defini­ tive guidance from the Appeal Court on the imposition of orders for lifelong restriction,6 the use of victim statements under section 14 of the Criminal Justice (Scotland) Act 2003, and the role of public opinion and the media in sentencing.

6  Such guidance was in fact duly issued by the Appeal Court after the empirical research had c ­ oncluded in the decision of Petch and Foye v HMA 2011 JC 210, discussed in Chapter 3.

20 

Context and Methodology

At the stage of focused coding many of the initial codes were dropped. This left four codes that made the most analytic sense with which to categorise the data inclusively and completely (Charmaz, ibid 57–58), namely the concepts of: (i) sentencing methodology (comprising the categories of the conceptualisation of sentencing as either an instinctive synthesis, a structured process, or a mixture of the two approaches); (ii) sentencing aims (comprising the categories of the ­judicial approach to punishment, denunciation, public protection and rehabilitation); (iii) aggravation and mitigation in sentencing (comprising the categories of remorse, family issues, previous good character and mercy); and (iv) sen­ tence ­discounting (comprising the categories of supposed justifications for the practice and judicial attitudes towards it).

Data Saturation This iterative approach to the collection and analysis of data proceeded until the four con­ cepts were ‘saturated’ with data. Interviews continued until: (i) no new or relevant data emerged; (ii) each category became well developed in terms of its properties and dimen­ sions demonstrating variation; and (iii) the relationships amongst the categories became well established and validated (Bryman, 2012: 421; Strauss and Corbin, 1998: 212). As Glaser and Strauss explain, if an instance merely replicates a feature already identified then ‘it only adds bulk to the coded data and nothing to the theory’ (Glaser and Strauss, 1967: 111). Thus, this stage of the research ceased when new data gathered from the inter­ views no longer suggested new insights into the ­emergent theory or no longer suggested new dimensions of theoretical categories (Bryman, ibid).

Transcribing the Interviews All 25 interviews were transcribed personally. This was done to ensure full i­mmersion in the data. Transcribing interviews also brings interviewers closer to the data, encourages them to begin identifying key themes, and ensures that they are alert to similarities and differences between respondents’ accounts (Barnes, 2012; Fielding and Thomas, 2008: 258). As the study comprised a relatively small sample of 25 respondents, all of the interviews were transcribed verbatim. This ensured that all possible analytic uses were allowed for; no data was lost which may later have become significant (Fielding and Thomas, ibid 257).

Some Cautions on the Use of the Interview Data Analysis of the interview data must be subject to certain cautions on the interpretation and use of the material identified by Mackenzie in the course of her own research on the Queensland judiciary. Due to the relatively unstructured nature of the interviews, some of the comments made by respondents were offered spontaneously and not in response to a particular question. It should not be assumed therefore that because few comments were

The Use of Case Law

 21

made on a particular issue, that other respondents might have held an opposite view, or even no view. Given the nature of the study, no meaningful quantitative analysis can be ­carried out with qualitative data of this sort (see Mackenzie, 2005: 10, 2001: 124). Mackenzie also makes the important point that one must remember that the interview responses are not written responses but spoken comments. Thus, the responses often have an informality which one would not see in legal writing. As Mackenzie notes of her own data derived from judicial interviews: The comments can seem quite casual and ‘chatty’ at times. It must be borne in mind that these are not court judgments tempered by the constraints of formal legal writing, but personal inter­ views where the judges were free to speak their own opinions in an anonymous way, and did so ­(Mackenzie, 2005: 11; see also Mackenzie, 2001, ibid).

Many judges and sheriffs made very similar comments on various issues. As such, any assumptions as to the identity of the judge or sheriff who made a particular comment would in all likelihood be inaccurate.

The Use of Case Law This study also comprises a detailed analysis of sentencing case law. As Roberts et al (2000: 4) observed in their research on the attitudes of Canadian sentencers to the then new sanc­ tion of conditional sentences, one of the ways to understand aspects of judicial sentenc­ ing is through an analysis of case law (see also Roberts and La Prairie, 2000: Chapter 2). More recently, Lord Hope of Craighead, former Deputy President of the United Kingdom Supreme Court, has noted that the way the law works in practice is best discovered by studying cases (Hope of Craighead, 2010: xv). The domestic case law to which reference is made throughout the study is primarily that of the High Court of Justiciary sitting as an appellate court. Decisions of this court are nor­ mally binding on High Court judges presiding at trials, and on sheriffs (Walker, 2001: 446). When acting in its appellate capacity, the High Court is often called the Court of Criminal Appeal or the Justiciary Appeal Court, although these nomenclatures are not used in Parts VIII or X of the Criminal Procedure (Scotland) Act 1995 which deal, inter alia, with appeals against sentence in solemn and summary proceedings respectively (Stewart, 1997: 264). For the sake of brevity, the term ‘Appeal Court’ is used to refer to the High Court of Justiciary exercising its appellate jurisdiction in both solemn and summary sentence appeals. The sentencing jurisprudence of other common law jurisdictions is also considered. The sentencing jurisprudence of the English courts and the Supreme Courts of Common­ wealth jurisdictions (particularly Australia and Canada) is arguably more sophisticated and more refined than in Scotland. The decisions of these courts are typically thorough in their analysis of both doctrinal and theoretical arguments; they frequently take academic and empirical research into consideration, as well as previous case law (Stark, 2012: 92;7 see also Warner, 2007a). In particular, the High Court of Australia has since the 1970s 7  Stark makes this observation in the context of appeals against conviction, but it applies equally to sentence appeals.

22 

Context and Methodology

become active in the development and refinement of sentencing principles (Colvin, 2003: 102; Edney, 2005a: 3–4, 20–29; see also Bagaric and Edney, 2016: 147–52 and Odgers, 2012: 18). It must be appreciated, however, that studies based on analysis of case law are subject to certain limitations. Roberts et al identify three weaknesses with what they term the ‘case law approach’ (Roberts et al ibid). Firstly, only a very small ­percentage of sentences imposed will be captured by the reporting services. S­ econdly, as only noteworthy cases are reported, they are not representative of the majority of sentences imposed. Thirdly, the authors s­ uggest that: [T]he underlying judicial reasoning [in reported decisions] has, to a large extent, to be inferred, as the judgement is not usually comprehensive enough to explain all the reasons giving rise to the sanction. Trial judges rarely have the time to write judgements that explain all the relevant factors considered at the time of sentencing (ibid).

To this, we may add another weakness. As Ashworth observes, the courts in many jurisdic­ tions receive only a selection of cases, and receive them only in the form in which they are constructed by others in the criminal justice system (Ashworth, 1995a: 264). The practices of the prosecution and of defence agents and defence counsel are, as Ashworth notes, influ­ ential in most systems (ibid fn 50). In the Scottish context, one can take issue with the view of Roberts et al (ibid) that sen­ tencing judgments are not comprehensive. In recent years appellate sentencing judgments have increasingly examined important points of principle by reference both to Scottish authority and to English and Commonwealth jurisprudence (see Lord Gill, 2016: 17–18). The authors’ point regarding the reporting of trial judges’ sentencing judgments, however, is well made. Although the majority of sentences are passed in the sheriff court, a sheriff ’s reasons for imposing the sentence that he or she did are likely to come to light only if the sentence is appealed. In such cases, excerpts from the sheriff ’s report may be included in the opinion of the Appeal Court. Few sentence appeals from the sheriff courts are, how­ ever, reported in the main law reports. Similarly, it is only rarely that first instance sen­ tencing decisions from the sheriff courts are reported. In recent years, the only such cases to appear have both concerned the operation of the notification requirements under the Sexual Offences Act 2003.8 The position is the same in the High Court where very few first instance s­ entencing deci­ sions are reported, the only recent example being HMA v M (sub nom HMA v McArthur).9 This is not to suggest that reported decisions of the Appeal Court are not in daily use by sentencers or that they are not a valuable source of information on the sentencing practice and culture of the courts;10 rather it is an acknowledgment that judgments, or ‘opinions’ as they are termed by the Scottish courts (Duncan, 1992: 79; O’Rourke, 2014: 90), are written on only a very small percentage of all cases sentenced in Scotland. Although it is vital to give documentary data such as the sentencing case law of domes­ tic and Commonwealth courts due weight and appropriate analytic attention, reported

8 See

Wylie v M 2009 SLT (Sh Ct) 18 and Cottam v AB 2010 SCCR 669. 2010 SCL 1309. 10  On the considerable advances made in recent years with regard to the reporting of criminal cases in Scotland, see Christie, 2000: viii; Farmer, 2010: 97; Lord Rodger of Earlsferry, 2010: xviii; and Lord Reed, 2014: 165. 9 

Conclusion

 23

s­ entencing decisions are not ‘surrogates for other kinds of data’ (Atkinson and Coffey, 2011: 79–80). Thus a researcher who wishes to use case law as a means of understanding judicial decision making will have to buttress an analysis of domestic and Commonwealth sentenc­ ing jurisprudence with other sources of data. As Bryman states: ‘[I]f we want to treat docu­ ments as telling us something about an underlying reality, we are likely to need to employ other sources of data regarding that reality and the contexts within which documents are produced’ (Bryman, 2012: 555). It is for this reason that the present study was not limited to a doctrinal analysis of sen­ tencing jurisprudence and academic literature. Whilst doctrinal research on the canons of sentencing law was an important aspect of the study, it was buttressed with the qualitative empirical research undertaken with judges and sheriffs. The result is the most extensive survey of Scottish sentencers undertaken in recent years.

Conclusion As judges in this comparatively small jurisdiction retain a wide discretion in sentencing, Scotland provides an excellent base for a study of judicial views on the sentencing process. The present study illustrates the importance of listening to sentencers and to what they say about the sentencing process. The data from the present study are discussed in Chapters 4 to 7, and conclusions are drawn about the nature of sentencing and its practical imple­ mentation by judges and sheriffs. Before the data are examined, however, the nature and exercise of judicial discretion in sentencing is considered by reference to practice in Canada, Australia and Scotland.

24 

3 The Instinctive Synthesis and Wise Blending of Penal Aims: A Comparative Study of Sentencing Methodology Introduction One of the most difficult and fundamental problems in sentencing is how the judge ought to approach the sentencing task in terms of a methodology or s­ ystem of decision m ­ aking (Corns, 1990: 145). An attempt is made to address this issue through a comparative e­ xamination of judicial sentencing methodology in Canada and Australia—two j­urisdictions in which the courts, like those in S­ cotland, have routinely emphasised the importance of a wide judicial discretion in sentencing and where the sentencing process is geared towards the attainment of individualised justice. Following a review of the sentencing approach of the Canadian courts and their focus on judicial discretion and proportionality, the sentencing method­ ology of the Australian courts is considered in depth. This is achieved by examining the tension between the two models that have been developed through appellate sentencing jurisprudence: the discretion-orientated ‘instinctive synthesis’ approach, and the ‘staged’ or ‘two-tiered’ approach which has consistency as its p ­ rimary aim. The chapter concludes by examining the Scottish courts’ use of the instinctive s­ ynthesis approach.

Sentencing Methodology in Canada The Traditional Approach to the Sentencing Task In common with the position in other common law jurisdictions, discretion has been the hallmark of sentencing in Canada. Judicial discretion is at the heart of understanding sen­ tencing in this jurisdiction where for many years the task of articulating principles was left entirely to the judiciary (Doob and Webster, 2016: 402; Manson, 2001: 56; 61; Manson et al, 2016: 38). Discretion has been regarded as a necessary precondition for introducing suf­ ficient flexibility into the sentencing process, thereby ensuring that the sentence is precisely tailored to meet the needs of the individual offender (Healy, 2013: 294–95; Krasnostein, 2015a: 16–17; Krasnostein and Freiberg, 2014: 4; Pomerance, 2013: 307; Young, 1988: 97). It has long been recognised that whilst marked disparity in sentences for similar crimes should be avoided, it is primarily to the individual offender himself and his s­ urrounding

26 

A Comparative Study of Sentencing Methodology

circumstances that the sentencer must look in determining the appropriate disposal1 (Ruby et al, 2008: 751). The traditional approach to the sentencing task was articulated by Mackay J in the ­decision of the Ontario Court of Appeal in R v Willaert.2 The appellant was convicted of rape and sentenced to life imprisonment. In his subsequent ­application for leave to appeal against sentence, Mackay J said this: I am respectfully of opinion that the true function of criminal law in regard to punishment is in a wise blending of the deterrent and reformative, with retribution not entirely disregarded, and with a constant appreciation that the matter concerns not merely the Court and the offender, but also the public and society as a going concern. Punishment is, therefore, an art—a very difficult art— essentially practical, and directly related to the existing needs of society … It is therefore impossible to lay down hard and fast and permanent rules (at 176, emphasis added).3

This dicta envisages sentencing as an individualised and discretionary process. Instead of ‘hard and fast rules’, the sentencer relies on his or her experience—experience gained both whilst at the Bar and since appointment to the Bench—to select the appropriate purpose for the sentence (Manson, ibid 62). This view, confirmed in a number of Supreme Court ­decisions,4 has resulted in a non-interventionist and deferential approach to appellate review (Manson, 1997: 278; Quigley, 1999: 5). As the Supreme Court observed in R v M (C A), appellate courts must respect the skills and experience of trial judges gained on the ‘front lines of criminal litigation’.5 The Willaert principle continues to represent the law in the sense that it provides a fair explanation of the approach of the Canadian courts: a proper sentence must be ‘fit’ for both the offence and the offender. This result is achieved through a wise blending of penal aims (Manson et al, ibid 41). The decision in Willaert has been cited for many decades as the leading example of the role of judicial discretion in sentencing and is still cited by the Canadian courts today.6 This discretion confers upon the sentencer the ability to choose the specific sentencing disposal in any given case, as well as the ability to determine the ‘fit’ or appropriate aims of sentencing in the particular case. Thus, the Willaert approach provides what Manson et al describe as an open and flex­ ible template for sentencing judges, a template which leaves their discretion dramatically unconstrained (Manson et al, ibid 41). Uniformity in sentencing, it has been said, would be too rigid to meet the needs of individual justice (Ruby, 1986: 447; Ruby et al, 2008: 751). This is illustrated by the major­ ity judgments of the Supreme Court of Canada in R v Ipeelee and R v Lacasse,7 and by the unanimous judgment of the Court in R v M (C A).8 In delivering the majority ­judgment in Ipeelee, LeBel J noted that: ‘The determination of a fit sentence is … a highly i­ ndividualized

1 

R v Switlishoff (1950) CCC 132 at 136. (1953) 105 CCC 172. 3  See also R v Natanson (1927) 49 CCC 89 at 90. 4  R v Shropshire [1995] 4 SCR 227; R v M (C A) [1996] 1 SCR 500; R v McDonnell [1997] 1 SCR 948; R v Nasogaluak [2010] 1 SCR 206; R v Nur [2015] 1 SCR 773; R v Lacasse [2015] 3 SCR 1089 and R v Lloyd [2016] SCC 13. 5  R v M (C A), ibid at [91]; see also Bloos and Renke, 1997: 801 and R v Lacasse, ibid at [11]. 6  See, for example, R v MacDonald 2011 BCPC 453 at [33] and R v Harvey 2016 BCPC 247 at [17]. 7  R v Ipeelee [2012] 1 SCR 433 and Lacasse (n 4). 8  R v M (C A) (n 4). 2 

Sentencing Methodology in Canada

 27

process. Sentencing judges must have sufficient manoeuvrability to tailor sentences to the circumstances of the particular offence and the particular offender’.9 Writing for the majority in Lacasse, Wagner J stated that: The determination of a just and appropriate sentence is a highly individualized exercise that goes beyond a purely mathematical calculation. It involves a variety of factors that are difficult to define with precision … [E]verything depends on the gravity of the offence, the offender’s degree of responsibility and the specific circumstances of each case.10

In the earlier decision of R v M (C A), meanwhile, Lamer CJ, writing for the Court, also considered sentencing to be an individualised process and noted that a degree of disparity is to be expected: It has been repeatedly stressed that there is no such thing as a uniform sentence for a particular crime … Sentencing is an inherently individualized process, and the search for a single appropriate sentence for a similar offender and a similar crime will frequently be a fruitless exercise of aca­ demic abstraction. As well, sentences for a particular offence should be expected to vary to some degree across various communities and regions in this country, as the ‘just and appropriate’ mix of accepted sentencing goals will depend on the needs and current conditions of and in the particular community where the crime occurred.11

The approach of the Canadian courts has thus made individualisation the key perspective for judicial decision making (Manson et al, ibid 39; Healy, ibid: 294; Berger, 2015: 6–7). In an article on the work of the Canadian Sentencing Commission, former Commissioner Anthony Doob appears, initially, to criticise the notion of individualised sentencing as a type of ‘judicial fetish’; however, he very candidly proceeds to recount a discussion that he and a colleague, Jean-Paul Brodeur (the Commission’s research director), entered into with an unnamed but ‘very important’ member of the Canadian judiciary that gives the lie to such criticism (Doob, 2011: 285). Doob recounts how, in the early days of the Canadian Sentencing Commission, he and Brodeur were asked by the judge what evidence existed for unwarranted disparity in sentencing. Doob and Brodeur referred to sentencing exercises conducted with the judiciary. They suggested that unwarranted disparity might be evident if, when faced with identical cases, one sentencer imposed a relatively straightforward and short probation order whilst another suggested a moderately long prison sentence. The judge, having ‘looked at [Doob and Brodeur] patiently’, offered an alternative conclusion when he suggested that perhaps both judges were right. Doob admits that neither he nor his colleague could fault the judge’s reasoning: if both sentences could be logically justified and individually tailored to the person being sentenced then, states Doob, perhaps both were right (ibid; see also Manson et al, ibid 69–70). This pragmatic, individualised approach to the sentencing task involves the sentencer borrowing from the philosophical justifications of punishment in general. The set of justi­ fications in turn becomes the set of potential purposes of the sentence. The sentencer must therefore consider the individual circumstances of both offence and offender to determine which objectives should be pre-eminent or subordinate in the particular case (Manson, ibid 61). 9 

10  11 

Ipeelee (n 7) at [38]. Lacasse (n 4) at [58]. R v M (C A) (n 4) at [92].

28 

A Comparative Study of Sentencing Methodology

The traditional Canadian sentencing methodology, guided by the choice of objective or objectives which suit the circumstances of the individual case, is described by M ­ anson as an ‘amalgam approach’ since all the objectives of punishment are translated into the potential purposes of the individual sentence. The description of the methodology as an amalgam captures both the combination of factors and the basic fluidity of the meth­ odology (Manson, ibid 62). For Manson, the methodology is inherently amorphous. In an earlier work, Cole and ­Manson observe that whilst this blending of penal objectives permeates the sentencing process and provides both systemic and individual justifica­ tions, it generates its own inevitable tensions due to the inherent contradictory nature of some of the objectives. At the same time, however, the blending of objectives in dis­ charging the sentencing task hides the uncomfortable observation that precision and consistency may be unobtainable goals (Cole and Manson, 1990: 16).

‘A Very Human Process’—R v Hamilton and the Modern Methodological View The decision of the Ontario Court of Appeal in R v Hamilton,12 a case involving the ­importation of cocaine by two impoverished Jamaican drugs mules, is notable for the discussion by Doherty J of the modern methodological view of the sentencing task. His ­Honour observed that the process has a narrow focus: it aims at imposing a sen­ tence that reflects the circumstances of the specific offence and the attributes of the specific offender.13 His Honour began his examination of the analytical framework of sentencing with an important observation: Sentencing is a very human process. Most attempts to describe the proper judicial approach to sentencing are as close to the actual process as a paint-by-numbers landscape is to the real thing … [T]he fixing of a fit sentence is the product of the combined effects of the circumstances of the specific offence with the unique attributes of the specific offender.14

Doherty J noted that the proportionality requirement as set out in the Canadian Criminal Code—section 718.1 of which requires a sentence to be proportionate both to the grav­ ity of the offence and to the degree of responsibility of the offender—had long been a touchstone of Canadian sentencing law which accepted the ‘just deserts’ rationale for stateimposed punishment. A sentence must first and foremost fit the specific crime and the specific offender, whatever other ends the sentencer may hope to achieve.15 Taking both of the constituent parts of the relevant section of the Code in turn, the ‘gravity of the offence’, his Honour noted, referred to the seriousness of the offence in a generic sense as reflected by the potential penalty imposed by Parliament and any specific features of the com­ mission of the crime which may tend to increase or decrease the harm or risk to the community

12 

(2004) 186 CCC (3d) 129. Ibid at [2]. 14  Ibid at [87]. 15  Ibid at [89]. On the importance of proportionality in Canadian sentencing, see Roberts and Cole, 1999: 10; Berger, 2015: 17–18; O’Malley, 2016: 67; and Doob and Webster: 2016: 374 and 406. See also the decisions of the Supreme Court of Canada in Lacasse (n 4) at [12] and [127]–[128]; and Ipeelee (n 7) at [36]–[37]. 13 

Sentencing Methodology in Australia

 29

occasioned by the offence. The ‘degree of responsibility of the offender’, meanwhile, referred to the offender’s culpability as reflected in the essential substantive elements of the offence— especially the fault component—and any specific aspects of the offender’s conduct or back­ ground that tended to increase or decrease his or her personal responsibility for the crime.16 Doherty J also stated that, in addition to complying with the principles of sentencing, sentences must promote one or more of the objectives identified in section 718 of the Crim­ inal Code, viz the denunciation of unlawful conduct; specific and general deterrence; inca­ pacitation (where necessary); rehabilitation; reparation to victims or to the community; the promotion of a sense of responsibility in offenders, and acknowledgement of the harm done to victims and to the community. How is this to be achieved? Doherty J said no more than that the relevance and relative importance of each of the objectives identified in sec­ tion 718 of the Code will vary according to the nature of the crime and the circumstances of the offender.17 This part of the decision in Hamilton is criticised by Manson et al who appear to ­suggest that the judgment lacks specification when they comment that ‘[s]urely, our judges are ­entitled to more guidance than what is now available’ (Manson et al, 2008: 115). ­Nonetheless, the Canadian courts continue to view the sentencing process as a delicate art based on com­ petence and expertise. The sentencer is thus required to undertake a task which, as we shall see, the Australian courts have characterised as involving an ‘instinctive synthesis’ of the facts and circumstances both of the offence and of the offender.

Sentencing Methodology in Australia In Australia, the so-called ‘art’ of sentencing has metamorphosed into what the courts have described as an ‘instinctive synthesis’ (Mackenzie, 2005: 17; 2001: 139) under which the fac­ tors bearing on a sentencing decision are aggregated and assessed in a single, global process of reasoning (Abbs, 2008: 136; Bagaric, 2015: 79–80; Bagaric and Alexander, 2015: 534; Bagaric and Edney, 2016: 19–20; Freiberg, 2016: 427; Hewton, 2010: 80; Krasnostein, 2015a: 36; Krasnostein and Freiberg, 2014: 18; Sentencing Advisory Council (Victoria), 2016: 14). A sentencer may have recourse to a number of guides, but the most important is his or her own intuition regarding the factors pertaining both to the instant offence and the offender (Krasnostein and Freiberg, ibid and 2013: 268). As the confluence of these factors is, as Kaye AJA noted in Russell v The Queen,18 ‘unique’ to each case, determining their relative weight and translating that weight into a sentence must be similarly unique (Krasnostein, ibid; Krasnostein and Freiberg, 2013: 268–69). Sentencing under the instinctive synthesis method is not entirely unconstrained. Any positive rules of law must be observed, but essentially the sentencer will balance and weigh all the circumstances of the particular case and make a judgment as to what is the appropri­ ate sentence to be imposed (Freiberg, 2014: 228; Hammond, 2007: 214). The basis on which

16  17  18 

Hamilton (n 12) at [90]–[91]. Hamilton (n 12) at [101]–[102]. [2011] VSCA 147 at [57].

30 

A Comparative Study of Sentencing Methodology

the sentencer reaches the end result need not be neatly and ‘correctly’ laid out: what matters is the final ‘result’ (Hammond, ibid). The alternate approach is that of the ‘two-tiered’, ‘sequential’, ‘structured’ or ‘staged’ method. These expressions are taken to describe the same concept and are used interchange­ ably in the Australian jurisprudence (Traynor and Potas, 2002). Under this approach, the decision making process is compartmentalised and particular factors isolated for the pur­ pose of calculating their specific impact on the ultimate disposal (Abbs, ibid; Bagaric, ibid 80; Bagaric and Alexander, ibid; Bagaric and Edney, ibid 20; Hewton, ibid 83; Krasnostein, ibid 37–38; McClellan, 2012: 14; Sentencing Advisory Council, ibid 14–15). The first ‘stage’ or ‘tier’ is to identify the facts that relate to ‘objective offence seriousness’ or the ‘objective circumstances’ of the case in order to determine what a proportionate sentence might be. The ­second ‘stage’ is to take into account any aggravating or mitigating circumstances and to increase or decrease the sentence accordingly in order to impose an appropriate sentence in the particular case (Freiberg, ibid 229). The debate over the instinctive synthesis versus the two-tiered approach raises broader issues, however, including how to accommodate different (and often competing) sentenc­ ing philosophies. In particular, it raises the issue of the importance of judicial discretion in the sentencing process (Traynor and Potas, ibid; O’Malley, 2000: 124).

The Instinctive Synthesis versus the Two-Stage Approach to Sentencing—the Decisions in Williscroft and Young In the State of Victoria, the task of the sentencer has always been seen as an ‘instinctive’ or ‘intuitive’ synthesis of all the aims of sentencing: retribution, rehabilitation, deterrence and incapacitation (Freiberg, 2014: 228–229; Freiberg, 2008: 155; Freiberg and Ross, 1999: 202; Krasnostein, 2015b: 43). The Victorian courts have traditionally treated sentencing as an essentially pragmatic exercise (Freiberg, 2014: 234). The Court of Appeal has emphati­ cally and repeatedly denied that the process can be dissected into its component parts and has disapproved attempts either to reduce sentencing to separate stages, or to quantifi­ able elements (Fox and Freiberg, 1999: 195; see also Krasnostein, ibid 47). The apotheosis of the approach is found in the Court of Criminal Appeal’s decision in R v Williscroft19 (Fox and Freiberg, ibid 196; Freiberg and Ross, ibid 203; McClellan, 2012: 5–6). In what was to become perhaps the most famous—or, depending upon one’s view of the ­desirability of judicial discretion, infamous—judicial statements in Australian sentencing jurisprudence, Adam and Crockett JJ described the nature of the sentencing decision thus: Now, ultimately every sentence imposed represents the sentencing judge’s instinctive synthesis of all the various aspects involved in the punitive process. Moreover, in our view, it is profitless … to attempt to allot to the various considerations their proper part in the assessment of the particular punishments presently under examination.20

This, as McHugh J of the High Court of Australia was later to observe, amounted to a candid recognition of the fact that in the end sentencing depends on the judge’s assessment of what 19 

20 

[1975] VR 292. Ibid at 300.

Sentencing Methodology in Australia

 31

is the correct sentence.21 The Court in Williscroft was not suggesting that the s­ entencing process was one of guesswork; rather, it was referring to the complex process involved in the exercise of the sentencer’s discretion whereby each mitigating and ­aggravating circum­ stance must be considered and accorded such weight as it deserves in combination with all the other relevant circumstances (see Krasnostein, 2015a: 39). It is an approach that draws upon conventional legal reasoning skills from other areas of the law where all the relevant facts are marshalled and the appropriate rules and principles applied (Hutton, 2013a: 88; Hutton, 2002: 553; Hutton and Tata, 2000: 310). In Williscroft, Adam and Crockett JJ considered that the sentencing judge had not taken a sufficiently serious view of the offences (armed robbery) and that he had placed undue emphasis on the rehabilitation of the offenders in reaching his decision. Their Honours recognised that this conclusion rested upon what they described as ‘essentially a subjec­ tive judgment largely intuitively reached by an appellate judge as to what punishment is appropriate’.22 This aspect of the judgment is critical as it provides the central basis for how a sentence is rendered through the instinctive synthesis method; it also expressly endorses the view that the sentencer’s subjective judgment is an acceptable basis for determining sentence (Bagaric and Edney, 2003: 123; Hutton, 2002: 553). Adam and Crockett JJ referred to the earlier decision of R v Geddes23 where, having attempted to construct a rational principle for determining whether a sentence was inad­ equate, the Supreme Court of New South Wales was able to find a solution only in the use of an epigrammatic device. In Geddes, Jordan CJ observed that it was easier to see when a wrong principle had been applied than to lay down rules for solving particular cases, and concluded that ‘… the only golden rule is that there is no golden rule’.24 Adam and Crock­ ett JJ stressed that a judgment as to what is appropriate by way of sentence must depend upon knowledge of sentences for the same or similar offences, this being derived from ‘personal experience’.25 Thus, the judgment in Williscroft suggests that sentencing cannot be undertaken in a systematic manner; sentencing is not a process involving the application of authoritative and ascertainable norms to a particular factual situation, rather it is the result of a subjective assessment by the judicial officer (Bagaric and Edney, ibid). The desirability of the instinctive synthesis approach was later affirmed by the Victorian Court of Criminal Appeal in R v Young.26 Young raised important questions concerning the manner in which the task of sentencing offenders should be approached. In a lengthy report to the Appeal Court, the County Court judge explained how he had approached the sentencing task in two stages. Firstly, he determined the sentence which was proportionate to the gravity of each of the charges. The judge then determined the actual sentence to be imposed—what he described as ‘the appropriate sentence’—by having regard to factors peculiar to the particular offender (such as the individual’s previous convictions), to policy matters (such as the tendering of a guilty plea), to the purposes sought to be achieved by the sentence, and to the principle of totality. In seeking leave to appeal, counsel for the ­appellants criticised the methodology employed by the sentencing judge. 21 

See his Honour’s judgment in Markarian v The Queen (see n 33 below) at [66]. Williscroft (n 19) at 200. 23  (1936) 36 SR (NSW) 554. 24  Ibid at 555. 25  Ibid at 301. 26  [1990] VR 951. 22 

32 

A Comparative Study of Sentencing Methodology

The question for the Appeal Court was whether it was open to a sentencing judge to proceed in the ‘staged’ manner adopted by the judge at first instance. The Court stressed the discretionary and individualised nature of the sentencing task. As the circumstances of particular offences and particular offenders are ‘infinitely various’, the Court noted that the task of the sentencing judge had never been regarded as capable of being confined within rigid formulae; to hold otherwise would, it was said, result in injustice.27 The Court then affirmed the instinctive synthesis approach to sentencing articulated in Williscroft. The Court noted that the decision in Williscroft had never been questioned and that it was frequently cited and followed in the Appeal Court. Indeed, the Court noted that what it termed ‘the Williscroft approach’ had been adopted on numerous occasions by the High Court of Australia.28 Having reviewed the authorities, the Appeal Court concluded that there was nothing to suggest that a sentencing judge should approach the task of arriving at an appropriate sentence in two stages or by other formalised steps.29 The two-stage approach stemmed from a notion that the principle that a sentence must not be disproportionate to the offence charged of itself imposes an order in which a sentencer is to consider the factors relevant to his or her task. This approach dictated that the sentencer must first fix a sentence which is proportionate to the crime, based purely on the objective circumstances of the offence. The term ‘objective’ is itself difficult to define but, in essence, refers to the overt, clearly demon­ strable characteristics of the offence such as the age of the victim or the level of violence used (Corns, 1990: 147; Nicholson, 2010: 6; Traynor and Potas, 2002; Ryan v The Queen).30 The sentencer who follows the two-stage approach must then adjust the proportionate sen­ tence by taking into account matters personal to the offender and other subjective factors such as prior record and motive (Corns, ibid; Fox, 2006: 16; Mildren, 2006: 2). The Appeal Court in Young disapproved this approach in trenchant terms, stating that there was ‘no justification whatsoever’ for it and that its adoption would lead to injustice. What is a proportionate sentence is, the Court noted, a matter of discretion; in most cases there must be a range of sentences open to a sentencing judge which are proportionate to the offence. There cannot be said to be a sentence which is the proportionate sentence, as the sentencing judge had purported to fix. Thus the Court considered that to attempt to fix a proportionate sentence before deciding upon the final sentence to be imposed would ‘only multiply the possibilities of error’.31 The Court concluded: [W]e can find no warrant in authority or justification or advantage from a practical point of view in the adoption of an artificial process for arriving at an appropriate sentence or any process which unnecessarily limits further the discretion of a sentencing judge. We think that the adoption of such a process is calculated to lead to error and injustice. Until Parliament or the High Court indi­ cates to the contrary we are clearly of the opinion that artificial processes or methods should not be adopted in Victoria (ibid.; cf Fox and Freiberg, 1999: 196; Fox, 2006: 16, and Freiberg, 2006: 9).

The Court in Young made it clear that since the staged or two-tiered approach would vitiate the sentencer’s discretion, the Williscroft instinctive synthesis was the only legitimate approach 27 

Ibid at 954–55. Ibid at 955. 29  Ibid at 958–60. 30  [2001] HCA 21 at [142]–[150]. 31  Young (n 26) at 960. 28 

Sentencing Methodology in Australia

 33

to the sentencing task. The unequivocal rejection of the staged approach cemented the highly discretionary instinctive synthesis approach and represented a formidable barrier to the development of alternative sentencing methodologies (Corns, ibid 150). The significance of the decision in Young lies in the fact that this methodological debate goes to the heart of attempts to guide sentencing discretion in common law countries by methods short of US-style sentencing ‘grids’ (Freiberg, 1995: 62). Following the decision in Young, the term ‘instinctive synthesis’ entered into common usage as a description of the sentencing process, particularly in Victoria but also in other State and Territory Supreme Courts and the High Court (Bagaric and Edney, ibid 122). It became the standard sentencing approach in most ­Australian jurisdictions ­(Bagaric, 2001: 14). A possible explanation for the continuing influence of the instinctive syn­ thesis is that the majority judgment in ­Williscroft did no more than simply articulate the way in which the judiciary themselves understand the process of sentencing (Edney and Bagaric, 2007: 17). In any event, ‘instinctive synthesis’ has since come to represent the sentencing process itself: the Court’s descriptive statement in Williscroft has, over time, metamorphosed into a normative principle, anchored to the notion that it is the instinctive synthesis alone that is the correct approach to the exercise of the sentencing discretion (Abbs, 2008: 138; Mackenzie, 2001: 139–140; Wong v The Queen).32 The High Court of Australia had traditionally refused to be drawn into the ­methodological debate and to rule on the correct approach to sentencing (Fox, ibid; Fox and Freiberg, ibid 199; Freiberg, 1995: 63; Morgan, 1994: 54). On later occasions when the issue arose, the High Court failed to provide a definitive answer (Leader-Elliot, 2002: 8). The Court eventu­ ally addressed the issue in Markarian v The Queen,33 a decision which represents the high water mark of judicial examination of sentencing methodology and in which the High Court both affirmed and reconceptualised the notion of instinctive synthesis (­Hewton, 2010: 84; Mackenzie and Stobbs, 2010: 28).

The Instinctive Synthesis Affirmed—the Majority Judgment in Markarian v The Queen The appellant pleaded guilty in the District Court to supplying heroin. Following a success­ ful Crown appeal against the sentence imposed, in which his sentence was increased by the New South Wales Court of Criminal Appeal from two and a half years’ imprisonment to eight through the Court’s application of a staged—and indeed, a mathematical—approach involving the making of explicit deductions from the statutory maximum sentence in respect of factors such as the nature of the offence and his guilty plea, the appellant sought special leave to appeal to the High Court of Australia. Leave was granted, the key question in the appeal being whether the correct approach to the sentencing task was the staged, sequential or two-tiered approach or, alternatively, the instinctive synthesis. Although Gleeson CJ, Gummow, Hayne and Callinan JJ (delivering the lead majority judgment) declined to set down a ‘universal rule’ enshrining instinctive synthesis as the

32  33 

[2001] HCA 64 at [74]. [2005] HCA 25.

34 

A Comparative Study of Sentencing Methodology

only permissible sentencing methodology,34 they emphasised that sentencing involves a discretionary judgment.35 The majority noted that the process of ‘instinctive synthesis’ may be wrongly understood as denying the requirement that a sentencer give reasons for the sentence passed; identifying ‘instinctive synthesis’ and ‘transparency’ as antonyms in the debate mis-described the area for debate.36 They described the nature of the instinctive synthesis as follows: ‘In general, a sentencing court will, after weighing all of the relevant factors, reach a conclusion that a particular penalty is the one that should be imposed’.37 Their Honours then set out and adopted a lengthy extract from the earlier decision of the High Court in Wong v The Queen.38 Wong held that the ‘two stage’ approach to sentenc­ ing (in the sense of a mathematical approach to sentencing in which there are to be incre­ ments to, or decrements from, a predetermined range of sentences) should not be adopted as it was both apt to give rise to error and was an approach that departed from principle. The ‘two stage’ approach failed to take account of the fact that there are many conflicting and contradictory elements which bear upon sentencing an offender and thus attributing a particular weight to some factors, while leaving the significance of all other factors substan­ tially unaltered, may be quite wrong.39 It was stressed in Wong that a sentencer must take account of all the relevant factors to arrive at a single result which takes due account of them all. So long as a sentencer must, or may, take account of all of the circumstances of both the offence and the offender, the Court in Wong considered that use of a staged approach distorted the already difficult balancing exercise required of the sentencer.40 The Court thus objected to the placement of an exag­ gerated emphasis on some sentencing factors by attaching numerical values to them in a way tending to skew the final sentence (Colvin, 2003: 102–03; Hewton, 2010: 89). The lead majority in Markarian then confirmed the desirability of the instinctive syn­ thesis approach to sentencing stating that, following Wong, sentencing courts may not ‘add and subtract item by item from some apparently subliminally derived figure, passages of time in order to fix the time which an offender must serve in prison’.41 Their Honours appeared to qualify this statement, however, by suggesting that ‘indulgence in arithmetical ­deduction’ should not be ‘absolutely forbidden’ in simple cases where the circumstances of the crime have to be weighed against one or a small number of other important matters. The instant appeal, however, was not such a case because of the number and complexity of the ­considerations which had to be weighed by the sentencing judge.42 As Edney and Bagaric (2007: 24) note, this section of the judgment is intriguing. ­Anderson (2006: 212) suggests that the lead majority were searching for a ‘middle ground’; a flexible approach that may be sequential or singularly instinctive depending on the complexity of the individual case (see also McClellan, 2012: 16–17). Edney and Bagaric

34 

Ibid at [36]. Ibid at [27]. 36  Ibid at [36]. 37  Ibid at [37]. 38  Wong (n 32). 39  Wong (n 32) at [75], approved in Markarian (n 33) at [37]. 40  Wong (n 32) at [76]. 41  Markarian (n 33) at [39]. 42  Markarian (n 33) at [39]. 35 

Sentencing Methodology in Australia

 35

regard as s­ ignificant the references in the judgment to the importance of ‘transparency’ and ­‘accessible reasoning’ in the sentencing process for these attributes are, they maintain, more likely to be evident in the staged approach to sentencing than under the instinctive synthe­ sis model (ibid). Although the High Court appeared to suggest that a staged approach to sentencing may be appropriate in the context of ‘simple’ cases, whereas more complex cases will require the instinctive synthesis to be invoked, it did not identify what is a ‘simple’ case and what is a ‘complex’ case (ibid 25; Bagaric and Edney, 2016: 23; Edney, 2005b: 53). Thus the joint reasons of the majority in Markarian embrace instinctive synthesis as the pre­ ferred approach to sentencing without endorsing it as the only legitimate approach (Warner, 2005: 360). The majority judgment does not object to structured reasoning in sentencing per se. The judgment does, however, oppose structured reasoning that is badly conceived and applied, such as where the sentencer quantifies certain factors at the expense of others in a way that leads to a misplaced emphasis on the particular circumstances of the offence and the offender (Hewton, ibid). Nevertheless, the decision comprises ‘a powerful r­ estatement of the wide discretion’ enjoyed by the sentencer,43 conclusively demonstrating that ‘the entire selection, fashioning and eventual imposition of any sentence … remain[s] at the discretion and responsibility of the sentencing judge’.44 The High Court, however, provided no practical criteria as to when a sentencer may legitimately resort to the ‘indulgence’ of an arithmetical process in carrying out the sentenc­ ing task. Abbs (2008: 142) suggests that little prospective methodological guidance can be derived from the joint reasons of the lead majority in Markarian because the exceptions to the approved instinctive synthesis approach are so poorly defined. Whilst the categorical rejection of the staged approach to sentencing set out in Wong may have been diluted in Markarian, resulting in a ‘watered-down’ version of instinctive synthesis as the preferred sentencing methodology (Warner, ibid 355), it is not clear to what degree or to what end (Abbs, ibid). Abbs also notes that the joint reasons in Markarian have been cited in State courts throughout Australia in support of the proposition that a staged approach to sentencing constitutes legal error on the sentencer’s part; however, they also allow some space for a staged approach to take place (ibid).

The Dangers of the Staged Approach and the Primacy of Judicial Discretion—Justice McHugh’s Judgment in Markarian McHugh J delivered a separate judgment in Markarian in which, by reference to cogni­ tive psychology and the jurisprudential history of sentencing under ­Australian criminal law, he mounted a significant defence of the instinctive synthesis approach to sentencing (Edney, 2005b: 54). His Honour began by defining the central concepts of the two-tier and instinctive synthesis approaches: two-tiered sentencing involves (i) the judge determin­ ing a sentence by reference to the ‘objective circumstances’ of the case, then (ii) increas­ ing or reducing the hypothetical sentence incrementally or decrementally by reference to

43  44 

Moore v The Queen [2005] NSWCCA 407 at [39]. Western Australia v Reynolds [2006] WASC 31 at [67].

36 

A Comparative Study of Sentencing Methodology

other ­factors, usually—but not always—personal to the accused. The instinctive synthesis, ­meanwhile, is the method of sentencing by which the judge identifies all the factors that are relevant to the sentence, evaluates their significance and then makes a value judgment as to what is the appropriate sentence given all the factors of the case. In this approach, the sentence is determined only at the end of the process.45 Those who support the instinctive synthesis, his Honour noted, contend that two-tier sentencing mistakes an illusion of exactitude for the reality of sentencing because there is no method of sequential arithmetical reasoning that produces the correct sentence for any case. As a sentence can only be the product of human judgment, based on all the facts of the case, the judge’s experience, the data derived from comparable sentences and the guidelines and principles authoritatively laid down in statutes and authoritative judgments, the instinctive synthesiser asserts that sentencing is not an exercise in linear reasoning: the result of each step in the sentencing process is not the logical foundation for the next step in the process. The circumstances of criminal cases, so the argument in favour of the instinctive synthesis goes, are so various that they cannot be the subject of mathematical equations. Sociological variables do not easily lend themselves to what McHugh J termed ‘mathematization’.46

The Need for Individualisation Justice McHugh gave four core reasons for favouring the instinctive synthesis approach. Firstly, the two-tiered approach does not allow for sufficient individualisation of the sen­ tence. In setting the notional or ‘benchmark’ sentence, the judge will invariably give undue, or even decisive, weight to only some of the relevant factors in the case.47 McHugh J alluded to this problem in his dissenting judgment in the earlier decision of AB v The Queen48 when he noted that, in applying the two-tiered approach, abstraction replaces the ­offender’s concrete circumstances in determining the appropriate sentence and that discretionary judgments require the weighing of elements, not the formulation of adjustable rules or benchmarks.49 McHugh J’s position is compelling. As Tata observes, the operational a­bstraction of ‘offence’ and ‘offender’ from the whole case is impracticable (Tata, 2002: 412). Divorcing the objective and subjective circumstances of the offence and the offender can be a difficult task and does not necessarily represent an accurate p ­ icture of what has actually occurred: context is all important in sentencing (Brown, 2010a: 8; Traynor and Potas, 2002). This aspect of the two-tiered approach raises some of the same problems identified with USstyle sentencing guidelines. A staged approach is apt to replace the individual, contextual­ ised offender with what Tonry identifies as ‘two dimensional crime-and-criminal history amalgams’ (1996: 20). The two-stage approach may operate in such a way as to steer the sentencer’s considerations towards ‘generic offenders’ (ibid). Offenders who are sentenced

45 

Markarian (n 33) at [51]. Markarian (n 33) at [52]. See also Hudson v The Queen [2010] VSCA 332 at [32]; Freiberg, 2014: 229; Freiberg and Krasnostein, 2011: 74; and Krasnostein and Freiberg, 2013: 269. 47  Markarian (n 33) at [53]. 48  [1999] HCA 46. 49  Ibid at [15]–[16]. 46 

Sentencing Methodology in Australia

 37

by a judge employing the two-stage method may well be transformed from persons to ‘kinds of persons, abstract entities … their concrete existence systematically ignored and thus ­nullified’ (Stith and Cabranes, 1997: 1263, emphasis per original).

The Likelihood of More Punitive Sentences Secondly, McHugh J considered that the two-tiered approach is likely to lead to more puni­ tive sentences. By concentrating on the objective circumstances of a crime, the sentencer is giving effect—and ultimately greater weight—to the ­retributive or deterrent theory of sentencing. The sentencer who starts with a notional sentence will, consciously or uncon­ sciously, downplay the importance of mitigation, reformation and rehabilitation in the sen­ tencing process.50 This argument is also persuasive. It is the principal reason for the move away from start­ ing point sentences in Canadian sentencing jurisprudence. Starting point sentences were pioneered by the Alberta Court of Appeal in the 1970s in an attempt to provide sentenc­ ers in the lower courts with a place to start the ­consideration of what sentence to impose (Manson, 2001: 67); the approach was then extended to a number of offences over the following 10 years (Manson, ibid 69; Manson et al, 2016: 76). It is difficult in retrospect to know whether the goal was to introduce greater uniformity in sentencing, to provide a more clearly explicable methodology, or to encourage a harsher sentencing regime (Manson et al, ibid 77). The Canadian starting point approach comprises three stages: firstly, the appellate court categorises offences into ‘archetypical’ cases described with as much precision as possible; secondly, the court attaches a starting sentence to the category; and thirdly, the sentencing judge applies the relevant aggravating and mitigating factors to vary the sentence up or down from the starting point (Manson, ibid 68). The starting point methodology has been criticised by academics and practitioners as leading to rigidity and inflexibility in the sentencing process (Healy, 2005: 269; Ruby, 1986: 451). As with McHugh J’s criticism of the Australian two-tiered approach, the starting point methodology was seen as resting exclusively on concerns relating to the objective gravity of the offence without making allowance for sufficient consideration of the offender (Healy, ibid 271). Starting points were also criticised as producing the natural effect of bunching sentences around a median rather than spreading them across a range to suit individualised circum­ stances (Manson, 1997: 282). The result was that the set of sentences for a particular offence were moved up the penalty scale and the overall rate of imprisonment was thus increased (ibid). Allan Manson, for example, undertook a comparison of the sentences imposed for cases of sexual assault across Canada (ibid 283). Manson found that the proportion of such offences resulting in custodial sentences was higher in the provinces that had adopted the starting point methodology (Alberta, Manitoba and Saskatchewan) than in the rest of the country and was significantly—three times—higher than in Ontario and British Columbia, two provinces which had eschewed the starting point approach in favour of

50  Markarian (n 33) at [54]; cf Edney (2005b: 55) and Edney and Bagaric (2007: 25) who criticise the supposed absence of empirical evidence for this view (compare the study by Manson (1997) discussed later).

38 

A Comparative Study of Sentencing Methodology

more flexible ranges of sentence. This led Manson to conclude that although the ­starting point m ­ ethodology might reduce disparity in sentencing, it tended to increase rates of ­incarceration (ibid). The starting point methodology rests upon a notion of ‘intrinsic blameworthiness’ (Lacelle, 1996). Under the banner of promoting consistency in sentencing, it begins with the court creating a category or label to describe a particular offence. The aim is then to impose uniform sentences on all offenders whose conduct can be placed within this ­category (ibid). This involves the judicial creation of a sentencing algorithm which imposes a supposedly rational structure upon what is essentially a discretionary art (Kennedy, 1994: 166). The primacy of the individual is replaced by a practice that emphasises offence characteris­ tics as the prime determinant of sentence (Young, 1988: 97). Starting point sentencing thus does not guide judicial discretion; rather, the reverse seems true (Kennedy, ibid 169). In an argument that accords precisely with McHugh J’s warnings against the Australian two-tiered methodology, Bloos and Plaxton (2003) consider starting points to be a ­methodology that subjects the sentencing process to ‘an unnatural compartmentalization of factors’ that either over complicates the process or forces the sentencer to give certain factors more or less weight than is appropriate. The starting point approach serves to justify what Cole and Manson (1990: 17) term a ‘prearticulated penal response’ by the courts: prison sentences become the norm as starting points harden into de facto minimum sentences and imprisonment is reinforced as the only fitting punishment for certain offences.51 As was stated in the report of the Canadian Sentencing Commission: ‘The logic [of the starting point methodology] appears to be that there can be too little punishment—but there cannot be too much’ (Canadian Sentencing Commission, 1987: 85). In R v McDonnell,52 however, a majority of the Supreme Court of Canada rejected the Alberta Court of Appeal’s starting point approach. Writing for the majority, Sopinka J considered that the Alberta Court of Appeal had taken the view that the starting point approach was a matter of legal principle53 and had been wrong to do so. The majority held that it could never be an error in principle in itself ‘to fail to place a particular offence within a judicially created category … for the purposes of sentencing’;54 nor would a sentencer err in principle by failing to refer to a particular starting point classification.55 The recognition of starting points—or ­‘presumptive sentences’ as they were also termed56—as legal principles would diminish the deference owed to sentencing judges. The decision of the majority in McDonnell, thus dealt a significant blow to the use of the starting point approach by stressing the importance of individualised s­entencing. The majority adopted what Manson describes as a diluted form of starting point, one that provides guidance but which does not entrench a template that would trigger ­appellate

51  Quigley, 1999: 3; see also R v Beaudry 2000 ABCA 243 at [72]; and R v Kain 2004 ABCA 127 at [32]–[33] (Berger J). 52  McDonnell (n 4). 53  McDonnell (n 4) at [30]. 54  McDonnell (n 4) at [32]. 55  McDonnell (n 4) at [24], [32] and [42]–[43]. 56  McDonnell (n 4) at [18].

Sentencing Methodology in Australia

 39

intervention if it was not applied (Manson, 2001: 73).57 Following the Supreme Court’s decision in McDonnell, Berger JA of the Alberta Court of Appeal (a staunch critic of starting point sentencing) provided a damning critique of starting points in a series of judgments between 1998 and 200858 in which he referred to the importance of sen­ tencers being able to select from the full range of sentencing options in the ‘judicial arsenal’.59 The decision of the Supreme Court in McDonnell demonstrates how the concept of starting points in sentencing undermines and severely limits sentencing discretion. To extend Justice Berger’s metaphor, the use of starting points (as applied by the Alberta Court of Appeal) empties the powder kegs of the judicial sentencing arsenal. In so doing, they replace the powerful howitzer of judicial discretion with an ineffectual pop-gun of de facto minimum sentences.60 As McHugh J recognised in Markarian, this can only result in insuf­ ficient attention being paid to offenders’ personal circumstances and, consequently, in the imposition of more punitive sentences.

The Illusory Nature of the Staged Approach Justice McHugh’s third reason for preferring the methodology of instinctive synthesis strikes at the very heart of the two-tiered approach to sentencing: the first of the two stages— concerning the objective circumstances of the offence—is itself an exercise in intuition: One fact that critics of the instinctive synthesis approach do not face up to—assuming they are aware of it—is that the first tier of the two-tier approach—unless it is the maximum sentence—is itself derived by an instinctive synthesis of the ‘objective circumstances’ of the case.61

Whether the two-tiered sentencer derives his or her starting point from the objective ­circumstances or from a sentence proportionate to the offence, the correctness of the sen­ tence will always depend on the correctness of the value judgment involved in assessing the first-tier sentence.62 Even if such a sentencer can correctly assess the first-tier sentence, he or she must still correctly assess the quantum of the increment or decrement for each factor in the process—a process which, as the High Court noted in Wong v The Queen,63 is ‘apt to give rise to error’.64 His Honour continued: The belief that two-tier sentencing is the preferable method is principally based on the idea that it promotes transparency of sentencing. Certainly, it shows a series of numbers. But they are more likely than not to be erroneous numbers. Each time the judge adds or subtracts another number the chance of ultimate error increases exponentially.65

57  See, however, the decision of the Alberta Court of Appeal in R v Arcand 2010 ABCA 363 where the majority of the Court sought to harden the use of starting points in a way which has been criticised as not fully respectful of the Supreme Court’s repeated admonitions that deference must be the standard of review for sentence appeals (see Quigley, 2011: 5 and Rudin, 2011: 1008; see also the judgment of Berger JA in R v Lee 2012 ABCA 17). 58 Namely R v Waldner 1998 ABCA 423 at [37]–[42]; R v Beaudry (n 51) at [66]–[72]; R v Kain (n 51) at [17]–[40], and R v White 2008 ABCA 328 at [37]. See also his Honour’s judgment in R v Lee, ibid. 59  Waldner, ibid at [29] and [39], and Beaudry (n 51) at [80]. 60  On the problems raised by statutory minimum sentences, see Chapter 4. 61  Markarian (n 33) at [55]. 62  Markarian (n 33) at [56]. 63  Wong (n 32). 64  Markarian (n 33) at [56], citing Wong (n 32) at [74]. 65  Markarian (n 33) at [71].

40 

A Comparative Study of Sentencing Methodology

As the major premise of the two-tiered sentence is a value judgment based on the judge’s instinct or intuition,66 McHugh J’s analysis revealed that the two-tiered approach’s appear­ ance of ‘objectivity and unfolding reason’ was simply illusory;67 it is an approach, his Honour stated, which is arbitrary and which has an almost exponential capacity for error.68 Even Edney and Bagaric—ardent opponents of the instinctive synthesis approach—accept that McHugh J’s criticism of this aspect of two-tiered or structured sentencing has ‘consid­ erable force’ (Edney and Bagaric, 2007: 26; see also Bagaric, 2015: 100 and Edney, 2005b: 55).

Considerations of Proportionality Finally, McHugh J also criticised the two-tiered approach as difficult, if not impossible, to reconcile with the principle of proportionality, a fundamental principle of sentencing law. The proportionality principle requires the sentencer to make a judgment concern­ ing the relationship of the penalty to the facts of the case. This, his Honour noted, is a value judgment based on the sentencer’s experience and instinct derived after taking into account all the facts and circumstances of the particular case. At the end of the process, the two-tier sentencer must ask whether the result of the additions and subtractions from the objectively determined sentence is proportionate to the accused’s offence. If the sentencer concludes that it is not (and in Justice McHugh’s view it would almost be a miracle if it was), then judicial tinkering with the quantum of each component in the sentence, in an attempt to achieve a result which is compatible with the concept of proportionality, renders the twotiered approach meaningless, if not a charade.69

The Superiority of the Instinctive Synthesis Whilst McHugh J acknowledged some of the criticisms of the instinctive synthesis approach and was frank in admitting that the approach is far from perfect, his Honour observed that—short of introducing US-style sentencing grids—it is the best that the judiciary can do.70 His Honour’s final view, however, was that the instinctive synthesis is the superior approach. Nothing in the judgments that have used it suggest that the two-tiered approach produces sentences that are more acceptable or ‘better’ than the sentences produced by the instinctive synthesis method.71 McHugh J also considered that, insofar as its proponents claim that two-tier sentencing promotes predictability, they ‘mistake the illusion for the reality’.72 His Honour was also scathing in his criticism of those who contend that sentencing is made more ‘scientific’ by the two-tiered system: [I]f two-tier sentencing is science, its results … suggest it is junk science … There is no Aladdin’s Cave of accurate sentencing methodology, the door to which can be opened by chanting the magic words, ‘two-tier sentencing’. There is only human judgment based on all the factors of the case, the

66 

Markarian (n 33) at [58]. Markarian (n 33) at [56]. Markarian (n 33) at [64]; see also Mildren, 2006: 5. 69  Markarian (n 33) at [69]. 70  Markarian (n 33) at [72]. 71  Markarian (n 33) at [70]. 72  Markarian (n 33) at [71]. 67  68 

Sentencing Methodology in Australia

 41

judge’s experience, the data derived from comparable sentences and the guidelines and principles authoritatively laid down in statutes and authoritative judgments.73

McHugh J considered that critics of the instinctive synthesis method placed too much emphasis on the ‘instinct’ and too little on the ‘synthesis’.74 By reference to the decision in Veen v The Queen (No 2),75 his Honour noted that in the process of synthesising all relevant factors, greater and lesser weight will be allocated to some factors depend­ ing on their relevance to both the offence and the offender. Ultimately, community and legal values are translated into a number of years, months and days—a process which must involve an instinctive judgment.76 In commenting on the decision in Markarian, the Queensland Court of Appeal was later to remark in R v McDougall and Collas that the exercise of sentencing discretion is ‘an integrated process directed to the determina­ tion of a just sentence’;77 Spigelman CJ of the New South Wales Court of Criminal Appeal explained in Mulato v The Queen that adoption of the two-tiered approach was unten­ able because it ‘impermissibly confines the sentencing discretion’;78 and Weinberg JA of the Court of Appeal of Victoria observed in Trajkovski v The Queen that the instinctive synthesis ‘is in no way incompatible with transparency and perfectly conducive to analytical rigour’.79 Edney and Bagaric, however, maintain that two-tier sentencing is preferable as it exposes precisely why a particular approach to the circumstances of the offence and the determina­ tion of the final sentence have been taken by the sentencer (Bagaric, 2016: 4; Edney, 2005b: 55; Edney and Bagaric, 2007: 26). The approach favoured by Edney and Bagaric may well result in the reasons for a particular sentence being more transparent; however, the authors simply assume that such a sentence will be ‘correct’ in the sense of being proportionate. Justice McHugh’s critique aptly demonstrates that although the sentencer’s reasoning may be more transparent, it is most likely that the outcome will not, in fact, be a just one. In ascribing mathematical values to sentencing purposes such as, for example, retribution and deterrence, separated from the myriad other purposes that are to be synthesised in the sen­ tencing outcome as the lower court did in Markarian, the final sentence is unlikely to be either proportionate or just.

The Importance of Judicial Instinct McHugh J further emphasised that the sentencer’s instinct is the aspect that is most trou­ bling to those who value predictability and transparency in sentencing; they see it as entirely subjective, personal, arbitrary and unconfined.80 The sentencer does not, however, select the sentence from thin air: the judicial air is, McHugh J noted, thick with trends, statistics, and appellate and statutory guidance.81 His Honour concluded his defence of

73 

Markarian (n 33) at [71]. Markarian (n 33) at [73]. 75  (1988) 164 CLR 465. 76  Markarian (n 33) at [73]. 77  R v McDougall and Collas [2006] QCA 365 at [17]. 78  Mulato v The Queen [2006] NSWCCA 282 at [13]. 79  Trajkovski v The Queen [2011] VSCA 170 at [83]. 80  Markarian (n 33) at [76]. 81  Markarian (n 33) at [76]. 74 

42 

A Comparative Study of Sentencing Methodology

the i­nstinctive synthesis approach by offering five observations on the nature of judicial instinct in sentencing. Firstly, McHugh J noted that a sentencer almost never imposes a sentence for an offence that has been committed for the first time.82 Through exposure to cases the sentencer ­develops, through experience and expertise, a sense of the relative gravity of offences and the relative circumstances of offenders that dictate the weighing of different factors in the sentencing process.83 His Honour continued: No one suggests that the judicial robe carries in its seams the wisdom of Solomon, but judicial experience in sentencing is a skill to be respected by the community and other judges. Repeated exercise in synthesising sentencing factors can only hone the instinct required to translate such ­factors into just numerical outcomes.84

Secondly, sentencers are sensitive to legislative trends. Changes in the maximum penalty for an offence may, for example, indicate a shift in the values to be applied when sentencing for that offence.85 McHugh J also referred in this regard to the use of appellate court guideline judgments to inform the sentencing task.86 Warner suggests that this is surprising as she considers guideline judgments to be ‘a highly developed application of two-tiered sentencing’ (2005: 358; see also Abadee, 2006: 15 and Lovegrove, 2002: 194). Justice McHugh’s acceptance of guideline judgments is, however, in no way inconsistent with his defence of the instinctive synthesis approach to sentencing. The use of guideline judgments is fully considered in Chapter 6. For present purposes, however, it is sufficient to note that the Appeal Court in Scotland has emphasised that such guidelines provide a structure for, but do not remove, judicial discretion; that guidelines should neither be applied rigidly nor lead to a mechanistic approach; and that they only assist sentencers in exercising their discretion.87 It is also recognised in Australia that the aim of such guidelines ‘is not to usurp the role of judges as sentencers, nor to deny the subjective elements in their decision-making, but to assist them in their difficult and thankless task’ (Fox, 2006: 16). Justice McHugh himself stated that—pace Warner, ibid and Lovegrove, ibid—the existence of a particular sentencing guideline does not mean that the sentencer must start with a specific number; knowledge of the guideline simply ‘guides the judicial instinct’.88 Thus, far from being inconsistent with the instinctive synthesis approach, guideline judg­ ments have the capacity to complement this method of judicial decision making by making it clearer and more intelligible to the parties and to the wider public. Guideline judgments serve to strike a balance between unbounded discretion and tight constraint; they allow for an instinctive synthesis that is more transparent, comprehensible and consistent (Crilly, 2005: 54). Justice McHugh’s third observation was that a sentencer will always know that the ­sentence imposed is subject to review on appeal and will thus pay close attention to 82 

Markarian (n 33) at [77]. Markarian (n 33) at [77]. 84  Markarian (n 33) at [78]. 85  Markarian (n 33) at [80]. 86  Markarian (n 33) at [80]. 87  HMA v Graham 2011 JC 1; Mitchell v HMA 2012 JC 13; Jakovlev v HMA 2012 JC 120; Rippon v HMA 2012 SCCR 699; and Neill v HMA [2014] HCJAC 67. 88  Markarian (n 33) at [80]. 83 

Sentencing Methodology in Australia

 43

s­ entencing guidance provided by appellate courts.89 Fourthly, sentencers will, for the sake of criminal justice generally, attempt to impose sentences that accord with legitimate com­ munity expectations.90 Finally, the principles of proportionality and consistency commonly operate as final checks upon a sentence proposed by a judge: the ultimate control on the judicial sentencing discretion is the requirement that the sentence be proportionate to the gravity of the offence committed.91 Justice McHugh concluded with the following remarks: The acceptance of the role of instinctive synthesis in the judicial sentencing process is not opposed to the concern for predictability and consistency in sentencing that underpins the rule of law and public confidence in the administration of criminal justice … [J]udicial instinct does not operate in a vacuum of random selection. On the contrary, instinctive synthesis involves the exercise of a discretion controlled by judicial practice, appellate review, legislative indicators and public opin­ ion. Statute, legal principle and community values all confine the scope in which instinct may oper­ ate. The judicial wisdom involved in the instinctive synthesis approach is therefore likely to lead to better outcomes than the pseudo-science of two-tier sentencing.92

These sentiments are echoed by Sir Gerard Brennan, former Chief Justice of the High Court of Australia. His Honour states that in undertaking the instinctive synthesis, there is no material element that a sentencer can leave out of the calculation; sentencers must be aware of the broad level of sentences current at the particular time in their particular commu­ nity and they must test any tentative view about an appropriate sentence to see whether it is proportionate to the circumstances of the particular offence (Brennan, 2006: 4). Thus, whilst ‘[t]he breadth and untrammelled nature of the sentencing discretion may seem like a charter of judicial freedom … in truth it evokes the most anxious consideration’ (ibid). The decision of the High Court in Markarian was not, however, unanimous in approving the instinctive synthesis approach. Kirby J dissented on the core issue of whether a sen­ tencer should undertake an instinctive synthesis of all the relevant factors in the discharge of the sentencing task.

A Need for Greater Transparency? Justice Kirby’s Dissent in Markarian In a dissenting judgment critical of the instinctive synthesis, Kirby J rejected the complaint that the lower court had erred in adopting a staged approach. His Honour asserted that, far from the two-tier or staged approach being inconsistent with the principle of proportion­ ality, the principle actually required an approach in which the sentencer first determines the outer limits of the sentence by reference to the objective factors and then applies any mitigating factors to arrive at the correct sentence.93 Kirby J also considered the instinc­ tive synthesis to be inconsistent with what he termed ‘statutory transparency’, or the trend to spell out in legislation specific considerations to be taken into account in sentencing;

89 

Markarian (n 33) at [81]. Markarian (n 33) at [82]. 91  Markarian (n 33) at [83], citing Veen (No 2) (n 75) at 472. 92  Markarian (n 33) at [84]. 93  Markarian (n 33) at [111]–[112]. 90 

44 

A Comparative Study of Sentencing Methodology

this, in his view, was a trend which suggested the need for adjustment up or down from a hypothesised norm.94 Justice Kirby’s primary reason for preferring the staged approach to sentencing, however, was that of transparency. He argued that adherence to the instinctive synthesis sends the wrong signals for the law of sentencing and runs contrary to the tendency in other areas of the law to expose the use of public power by public officials to scrutiny: ‘Talk of “instinctive synthesis” is like the breath of a bygone legal age. It resonates with a claim, effectively, to unexplainable and unreviewable power’.95 Kirby J considered that appellate courts should encourage the revelation at least of the important adjustments made by the sentencer.96 The courts should not, his Honour stated, encourage the thought that upon appointment to the Bench, a mystical ‘instinct’ or ­‘intuition’ descends upon the judicial officer that ensures he or she will get the sentence right ‘instinctively’. He considered that such an approach discourages explanation of the ‘logical and rational process’ that led to the sentence, so far as it can reasonably be given.97 To suggest that it is safer, wiser or even essential to keep the process of judicial reasoning ‘secret’ is ‘good neither for the parties, nor for the community, nor for the discharge of the functions of sentencing, nor for appellate review’.98 As Mildren J notes, however, it is already common practice for sentencing judges to indi­ cate, in general terms, what weight they have attached to the more important aspects of their task (Mildren, 2006: 6). This may amount to little more than a comment to the effect that, ‘I give little weight to X’ and, ‘I have given considerable weight to Y’ but, nevertheless, this is surely enough (ibid)—particularly, it is suggested, in the context of busy lower and intermediate courts such as the sheriff courts in Scotland where the presiding sheriff may be required to deal with several dozen cases in one sitting. Kirby J concluded: Australian judges must now express their obeisance to an ‘instinctive synthesis’ as the explana­ tion of their sentencing outcomes … The lofty and absolute prescriptions of Williscroft and Young remain in place like the two vast and trunkless legs of stone of Ozymandias. But, with all respect, they are now beginning to look just as lifeless. One day, I expect that travellers to the antique land of this part of the law of sentencing will walk this way without knowing that the two proscriptions once were there.99

Even Kirby J, however, conceded the importance of judicial discretion in the ­sentencing task. His Honour agreed both with the majority and with certain dicta in AB v The Queen and Wong v The Queen,100 that sentencing is not a mechanical, numerical, arithmetical or rigid activity in which one starts from the maximum fixed by the legislature and then works down in mathematical steps.101 The sentencing process is not so scientific and, because sentencers must take so many—often competing—factors into account, Kirby J considered

94 

Markarian (n 33) at [125]–[127]. Markarian (n 33) at [129]. 96  Markarian (n 33) at [130]. 97  Markarian (n 33) at [130]. 98  Markarian (n 33) at [131]; see also [135]. 99  Markarian (n 33) at [139]. 100  AB v The Queen (n 48); and Wong v The Queen (n 32). 101  Markarian (n 33) at [133]; see also his Honour’s comments to the same effect in Ryan v The Queen (n 30) at [89] where he also noted that the process involves ‘intuition and judgment’. 95 

Sentencing Methodology in Australia

 45

the evaluation to be necessarily imprecise.102 There is sometimes a legitimate role for differ­ ences of judicial view, favouring either the extension of leniency or the advocating of a more punitive view of all the relevant considerations in the particular case.103 Kirby J accepted that there are limits to the explanation of reasons for a given sentence.104 Short of noting that judgment is inevitably invoked in the passing of all sentences except those fixed by law,105 his Honour did not, however, expand upon this concession. Thus, strong as his criticism of the instinctive synthesis was, Kirby J accepted the need for individualisation in sentencing and the importance of discretion (Bennett and Broe, 2007: 81; Hewton, 2010: 91–92). Where Kirby J differed from the majority was in holding that such individualised sentences should be imposed following a process of transparent reasoning rather than through an application of instinctive synthesis.

Developments Post-Markarian—the High Court’s Continued Emphasis on the Instinctive Synthesis In Hili v The Queen and Barbaro v The Queen,106 the High Court made a number of obser­ vations regarding the nature of the sentencing task which re-affirmed instinctive synthesis as the correct approach. In Hili the majority stated that whilst consistency in sentencing is important, the consistency that is sought is consistency in the application of the relevant legal principles, not some numerical or mathematical equivalence; the sentencer must have regard not just to what has been done in other cases but why it was done.107 The majority also approved the observations of Simpson J in DPP (Cth) v De La Rosa108 that past sen­ tences ‘are no more than historical statements of what has happened in the past. They can, and should, provide guidance to sentencing judges, and to appellate courts, and stand as a yardstick against which to examine a proposed sentence’.109 In Barbaro, meanwhile, the Court stated that: Fixing the bounds of a range within which a sentence should fall or within which a sentence that has been imposed should have fallen wrongly suggests that sentencing is a mathematical exercise. Sentencing an offender is not, and cannot be undertaken as, some exercise in addition or subtrac­ tion. A sentencing judge must reach a single sentence for each offence and must do so by balancing many different and conflicting features. The sentence cannot, and should not, be broken down into some set of component parts.110

As Warner notes, Barbaro reflects the High Court’s disapproval of a mathematical approach to sentencing and its dislike of numerical guidelines first expressed in Wong v The Queen,111 (Warner, 2014: 366, 369). The High Court’s decisions in Hili and B ­ arbaro 102 

Markarian (n 33) at [133]. Markarian (n 33) at [133]. 104  Markarian (n 33) at [130]. 105  Markarian (n 33) at [130]. 106  Hili v The Queen [2010] HCA 45 and Barbaro v The Queen [2014] HCA 2. 107  Hili, ibid at [18]; see also Green v The Queen [2011] HCA 49 at [29]. 108  [2010] NSWCCA 194. 109  Ibid at [304] (emphasis added), cited in Hili (n 106) at [54]. 110  Barbaro (n 106) at [34], references omitted. 111  Wong (n 32). 103 

46 

A Comparative Study of Sentencing Methodology

are therefore consistent with its earlier decisions in Wong and Markarian in endorsing the instinctive synthesis approach to sentencing (Bagaric, 2015: 80–81, 2011: 14). Hili and Barbaro continue what Warner describes as the High Court’s ‘conservative approach’ to sen­ tencing methodology, one which favours individualism over consistency (Warner, ibid 371). The decisions strongly assert the inherent value of a broadly unfettered judicial ­sentencing discretion and the notion that such discretion should be as wide as possible within the parameters of the maximum penalty, the limiting principle of proportionality, and any statutory constraints (Bagaric and Alexander, 2015: 535; Freiberg, 2016: 428; K ­ rasnostein, 2015b: 42). The decisions in Hili and Barbaro reflect the High Court’s continued view that the fairness of sentencing outcomes ‘is in a direct relationship with the amount of discre­ tion accorded to individualise sentences’ (Krasnostein, ibid 47; see also Krasnostein, 2015a: 184). Sentencing in the Australian courts therefore continues to be regarded as an attempt above all else to attain individualised justice (Bagaric and Edney, 2016: 27).

The Instinctive Synthesis in the Scottish Courts The General Approach The question of the appropriate sentencing methodology has not been the subject of detailed discussion in Scottish case law. Such authority as does exist, however, indicates a general preference (at least amongst appellate judges) for the instinctive synthesis over the staged or two-tiered approach. Reference to the instinctive synthesis methodology has featured in a number of cases concerning the allowance of a discount in sentence following the tendering of a guilty plea by an accused in terms of section 196 of the Criminal Procedure (Scotland) Act 1995. The issue of guilty plea discounts is considered in detail in Chapter 7; however, for present pur­ poses it is sufficient to note that in Du Plooy v HMA112 (in which sentencing guidelines for guilty plea discounts were first set out), the Court referred with approval to the judgment of Spigelman CJ of the New South Wales Court of Appeal in R v Thomson113 where his ­Honour emphasised that the instinctive synthesis was the correct approach to sentencing.114 The following year, in his dissenting opinion in RB v HMA,115 Lord Philip declared himself to be ‘in agreement with this analysis of the sentencing process’.116 The dicta of ­Spigelman CJ was again cited with approval by the Appeal Court in Paterson v HMA.117 Paterson involved the rape of a physically handicapped 11-year-old child with severe learn­ ing difficulties by her own learning impaired cousin. The Court acknowledged that it was a complex case and considered it to be a prime example of the sentencing judge having

112 

Du Plooy v HMA 2005 1 JC 1. [2000] NSWCCA 309. R v Thomson ibid at [55] and [57], cited in Du Plooy (n 112) at [12]. 115  2004 SCCR 443. A dissent which was later approved by the Lord Justice Clerk (Gill) in Gemmell v HMA 2012 JC 223 at [80]. 116  RB v HMA, ibid at [27]. 117  High Court of Justiciary on appeal, 24 February 2005, unreported. 113  114 

The Instinctive Synthesis in the Scottish Courts

 47

to employ an instinctive synthesis of a wide range of incommensurable and conflicting ­objectives in arriving at the final sentence.118 Most recently—and most importantly—instinctive synthesis was stated to be the correct approach to sentencing by Lord Justice Clerk Gill (as he then was) in Gemmell v HMA.119 By reference to the decision of the High Court of Australia in Markarian, Lord Gill stated that the assessment of a sentence is not a matter of precise arithmetical calculation; rather, it involves the making of an overall judgment from a consideration of numerous factors based on judicial experience.120 These comments were later endorsed by Lord Gill’s succes­ sor as Lord Justice Clerk, Lord Carloway, in delivering the opinion of the Court in McGill v HMA121 and in the course of delivering the leading opinion in Ferguson v HMA.122 In his opinion in Ferguson, Lord Carloway also noted—again by reference to Lord Gill’s opinion in Gemmell—that although the sentencing decision often involves ‘a complex matrix of factual and legal material’, it is a decision which ‘will be instantaneous, if not quite instinc­ tive … once the material is ingathered and understood’.123 Lord Carloway also referred to the importance of the sentencer’s ‘knowledge and experience’ in making the decision as to sentence.124 Thus, Scotland’s two most senior judges have stressed the centrality of judicial experience and discretion in sentencing (Brown, 2014a: 5–6).125

A Cautionary Lesson—the Decision in Petch and Foye There are only two situations in which judges in Scotland are expressly directed to f­ollow a staged approach to sentencing: the allowance of a discount in sentence for a guilty plea under section 196 of the Criminal Procedure (Scotland) Act 1995, and the imposition of Orders for Lifelong Restriction (OLRs) or punishment parts in discretionary life sentences under, respectively, section 210F of the 1995 Act and section 2(2) of the Prisoners and Criminal Proceedings (Scotland) Act 1993. These latter provisions—the complexity of which is criticised by Thomson (2011: 110; 2012: 23)—require the court to fix a period of time which it considers appropriate to satisfy the requirements of retribution and deter­ rence (ignoring any period which may be necessary for the protection of the public), which the prisoner must serve before the question of parole can be considered. As noted above, the issue of guilty plea discounting is fully considered in Chapter 7. Insofar as discretionary life sentences and OLRs are concerned, a Bench of seven judges gave guidance on the issue in Petch and Foye v HMA.126 What the majority of the Appeal Court in Petch and Foye decided was this: in fixing the punishment part the sentencer should (i) identify the determinate sentence which n ­ otionally

118 

Ibid at [4]. Gemmell (n 115). 120  Gemmell (n 115) at [59]. 121  2014 SCCR 46 at [13]. 122  2014 SCCR 244 at [103]. 123  Ibid at [104]. 124  Ibid. 125  Lord Carloway was himself appointed Lord Justice General in January 2016, with Lady Dorrian being appointed Lord Justice Clerk in April 2016. 126  2011 JC 210. 119 

48 

A Comparative Study of Sentencing Methodology

might have been imposed if a life sentence had not been imposed, (ii) strip out of that notional sentence the whole of any element for public protection including any protective element which might be included in any determinate sentence, and (iii) select half of the resultant period, subject to a possible power to select a longer period (less than two thirds) as an aspect of the sentencer’s residual discretion under section 2(2) of the 1993 Act as a whole.127 In reality, however, a fourth step is required as the judge must also consider the sentence discount provisions of section 196 of the 1995 Act.128 A fifth step may even be required, namely that of deciding whether (as the Court itself did in Mr Petch’s subsequent hearing) to fix a higher initial determinate sentence than would be justified in a ‘real’ determinate sentence fixed for the offence, taking into account the normal sentence for such offences (see Gordon 2011a: 368). Thus, even a cursory glance at the rubric of Petch and Foye reveals that this staged, formulaic approach is completely at odds with the oft-cited judicial view that sentencing is not a mathematical exercise. A variety of disagreements on this and related points amongst the other judges led the Lord Justice General (Hamilton) and certain of his colleagues to suggest that a clear legislative solution was called for (Stoddart, 2011: 43). All of the judges in Petch and Foye agreed that the statutory regime lacked clarity and could give rise to results that were objectively impossible to justify. As Thomson observes, there is an inherent artificiality in ‘stripping down’ a sentence into its punitive and protective elements, then applying a notional proportion to one element (Thomson, 2011: 109); there was also a degree of public disquiet that the ‘stripping down’ exercise gave rise to excessively short punishment parts (Thomson, 2013: 265). As Lord Justice Clerk Carloway (as he then was) noted in a recent address, the difficulty which judges faced in fixing punishment parts using this methodology was that none had hitherto thought in terms of compartmentalising parts of a custodial sentence into discrete periods (Carloway, 2013: 6). In response, the Scottish Parliament passed the Criminal Cases (Punishment and Review) (Scotland) Act 2012 which sets out a new series of detailed rules for the calculation of such punishment parts. The rules comprise a series of stages through which the sentencer must work and are expressed in a way, which Sheriff Alastair Brown argues, tend ‘to make the eyes glaze over’ (AN Brown, 2012: 5). The three (in reality four, or perhaps even five) stage approach set out by the Appeal Court in Petch and Foye for the calculation of punishment parts in discretionary life ­sentences arose from the complexity of the 1993 Act and can be seen as an example of the trend referred to by Kirby J in Markarian of spelling out in legislation certain specific con­ siderations to be taken into account in sentencing. Nevertheless, the Petch and Foye staged approach amounts to the antithesis of the instinctive synthesis method of sentencing. The approach of the majority in Petch and Foye was criticised by Lord Emslie who, in his dis­ senting opinion, considered it regrettable that the Court saw fit to impose such a complex regime on sentencers; his Lordship considered that the regime undermined the importance of judicial discretion in sentencing.129 Lord Emslie’s concerns regarding the complexity of the regime were shared by Scotland’s leading authority on criminal law, Sir Gerald Gordon

127  128  129 

Ibid at [43]–[46]. Thomson, 2011: 106; Ross v HMA 2013 SCL 1054 at [28] and [33]. Petch and Foye (n 126) at [110].

Conclusion

 49

QC, who noted in the opening words of his commentary to the decision in Petch and Foye: ‘Do not be troubled, dear reader, if you have difficulty in understanding this case—I think that your difficulty is shared by many people, including, certainly, myself ’ (Gordon, 2011b: 239). The staged approach prescribed by Petch and Foye was such that in McCluskey v HMA,130 a decision which applied the post-Petch and Foye law, the Appeal Court sought to explain the punishment part imposed by the sentencing judge by reference to a mathemati­ cal equation: ‘[E]mploying the formula dictated by Petch and Foye v HM Advocate (supra), [the judge] selected a punishment part of 3 years and 4 months (12−2 = 10 ÷ 2 = 5−1/3)’.131 Although the staged approach applies only to the determination of the punishment part of discretionary life sentences and OLRs, and thus concerns only sentencers in Scotland’s High Court (and even then will arise for consideration in only the most serious sexual and ­violent crimes),132 the new regime has nonetheless introduced a rigid and unyielding ­template for the imposition of such sentences.

Conclusion In examining judicial sentencing methodology from a comparative perspective, it has been demonstrated that sentencing in Canada and Australia is seen by the courts as an inher­ ently individualised process—as a human process.133 It is a process characterised by high levels of judicial discretion and involves the judge relying on his or her professional experi­ ence (gained whilst in legal practice and since appointment to the Bench) in order to craft an appropriate sentence. Such judicial discretion introduces flexibility into the sentencing process; it ensures that the sentence is tailored to meet the needs of the individual offender whilst also reflecting the seriousness of the particular offence. The final sentence is thus determined through a process which the Canadian courts have described as comprising a ‘wise blending of penal aims’.134 The overarching sentencing methodology in Australia—recently considered and ­confirmed by the High Court of Australia—is that of the ‘instinctive synthesis’.135 ­Sentencing by way of instinctive synthesis involves the judge using his or her experience to aggregate and assess all the relevant factors in a single, integrated process of reasoning. The High Court of Australia has rejected the alternative, staged approach to sentencing in which the judge sets an appropriate sentence commensurate with the severity of the offence, before making allowances up and down in light of relevant aggravating and mitigating factors. Thus, the Australian courts do not consider arithmetical exactitude to be either possible or desirable in sentencing because of the inherently multifaceted nature of the task.

130 

2012 SCCR 719. Ibid at [4]; see also Reid v HMA 2013 SCCR 70 at [19]. Ferguson v HMA (n 122) at [26]; see also Chalmers v HMA 2014 JC 220 and, for commentary, Brown, 2014b. 133  Hamilton (n 12). 134  Willaert (n 2). 135  Markarian (n 33). 131 

132 See

50 

A Comparative Study of Sentencing Methodology

The Appeal Court in Scotland has approved the instinctive synthesis method of sentenc­ ing in a series of decisions, most recently in opinions delivered by the country’s two most senior judges.136 Recent case law and legislation governing the calculation of punishment parts in discretionary life sentences has served to illustrate the problems and perceived injustices that may arise following the imposition of a formalised and staged sentencing regime. As such legislation pertains to only a fraction of cases sentenced by the Scottish courts, however, sentencing in Scotland routinely accords with practice in Canada and Aus­ tralia. As in those jurisdictions, sentencing in Scotland is an individualised process which has at its core judicial discretion and experience. These are the key factors to understanding the nature of the sentencing task. Whilst the importance of judicial discretion is such that the Australian, Canadian and Scottish courts have favoured the instinctive synthesis approach to sentencing, this method­ ology has been criticised as insufficiently transparent by Kirby J in his dissenting judgment in Markarian v The Queen. The next c­ hapter considers a further criticism of the instinctive synthesis made by academic ­commentators—that such an approach violates the rule of law.

136 

Gemmell (n 115); McGill (n 121) and Ferguson (n 122).

4 Equity and the Rule of Law in Sentencing Introduction This chapter seeks to justify the instinctive synthesis model of sentencing by c­ onsidering its compatibility with the rule of law. The claim that sentencing by way of instinctive synthesis is contrary to the rule of law is addressed by reference to Lawrence Solum’s notion of equity (Solum, 2008, 2003 and 1994) and his ­concept of a ‘virtue-centred’ theory of judging. It is argued that the concept of equity in the sense of doing particularised justice—justice as fairness, distinguished from ­justice as lawfulness—has particular relevance to the sentencing task. By considering the findings of previous sentencing studies, domestic and Common­ wealth ­sentencing jurisprudence and the findings of the present study, it is demonstrated that the instinctive synthesis approach allows judges to achieve individualised ­justice by basing their sentencing decisions on the particular offence and on the individual, contex­ tualised offender. The importance of individualised sentencing is further illustrated by examining how equity is practised by judges and sheriffs in Scotland. Judicial responses to restrictions on sentencing discretion in the form of statutory minimum sentences are considered, along with the roles of leniency and mercy in the sentencing process. The chapter concludes by exploring respondents’ attitudes towards the issue of consistency in sentencing and whether it is in fact attainable in a system which prioritises judicial discretion.

The Instinctive Synthesis and the Rule of Law The Real Meaning of ‘Instinctive Synthesis’ The notion of sentencing as an ‘instinctive synthesis’ has been criticised by ­Ashworth as an ‘inscrutable idea’, sanctioning a ‘free for all’ approach to the purposes of punishment in which ‘no weight at all is given to rule-of-law values’ (Ashworth, 2015: 81, 2010a: 77). Only one respondent in the present study (Sheriff 12) agreed with Ashworth’s criticism, stating that the notion of sentencing as an ‘art’ was ‘just an excuse for prolonging a kind of unregulated area’. As we shall see, however, in arguing for a renewed academic focus on individualised justice in sentencing, I suggest that the danger lies not in subscribing to the instinctive

52 

Equity and the Rule of Law in Sentencing

synthesis approach per se, but rather in the wholesale adoption of this idea as an analytical term. The Canadian Sentencing Commission noted that, as an art rather than a science, sen­ tencing is a process which cannot be reduced to the mechanical application of a few rigid formulae but cautioned nevertheless that, if stretched too far, the analogy with art could generate significant misunderstandings (Canadian Sentencing Commission, 1987: 281). In a similar vein we must be clear what we mean by the use of terms such as ‘instinctive synthesis’ or ‘intuitive decision making’ in interpreting sentencers’ views. As Basten JA noted in Sanchet v DPP (Cth),1 sentencing involves both a synthesis of inter-related considerations and a value judgment, which should be articulated so far as possible.2 That attempt, his Honour stated, should not be avoided by describing the outcome as ‘instinctive’ as sentencing is inev­ itably instructed by a knowledge of comparable cases and relevant sentencing principles.3 As a result based on knowledge and experience—two factors which, it will be argued, are key in achieving individualised justice—sentencing is thus the antithesis of ‘instinctive’; nonetheless, the term is used to describe the effect of weighing a number of disparate ele­ ments to reach a considered conclusion.4 This is by its nature an idiosyncratic and difficult task (Connolly, 2006: 2). From an interpretive perspective,5 sentencers are self-conscious and reflective social actors, making decisions and choices within certain boundaries (Hutton, 2006a: 157–158; 161). Sentencers are actively engaged in interpreting material—including the Crown narrative, the plea in mitigation, any schedule of previous convictions and any criminal justice social work report—to try and ascertain what sort of person is before them (Jacobson and Hough, 2007: 48–49; Wheeler et al, 1988: 14). Sentencers draw on their past experience, on the facts of the particular case and on the normative lens through which they judge offenders to actively construct a particular accused to be the kind of person for whom a particular sanction is justified (Wheeler et al, ibid). The decisions which sentencers are called on to make are fact-specific and i­ndividual offences and offenders are infinitely complicated (Padfield, 2011: 99; see also Bingham, 2000a: 310 and 2000b: 46). The High Court of Australia noted in Veen v The Queen (No 2) that: [S]entencing is not a purely logical exercise, and the troublesome nature of the ­sentencing discretion arises in large measure from unavoidable difficulty in giving weight to each of the purposes of punishment. The purposes of criminal punishment are various: ­protection of society, deterrence of the offender and of others who might be tempted to offend, retribution and reform. The purposes overlap and none of them can be considered in isolation from the others when determining what is an appropriate sentence in a ­particular case. They are guideposts to the appropriate sentence but sometimes they point in different directions.6

1 

[2006] NSWCCA 291. Ibid at [17]. 3  Ibid at [17]. 4  Ibid at [17]; see also Hantzis v The Queen [2006] NSWCCA 387 at [14]–[16]. 5  The interpretive perspective views social action largely as consisting of the meaning-orientated behaviour of social actors, the sources of this meaning being the different social groups within which the actors are socialised, and holds that interpretation of the world inevitably takes place from within this group-informed perspective (Tamanaha, 1997: 3); in short, that the social world is an active p ­ roduction of social action (Hutton, 2006a: 161). 6  Veen v The Queen (No 2) (1988) 164 CLR 465 at 476. 2 

The Instinctive Synthesis and the Rule of Law

 53

In the earlier decision of R v Rushby, in what was to become a frequently cited dictum, Street CJ observed that the determination in any given case of the appropriate sentence involves ‘an adjudicative balancing of a number of differing and not entirely consistent elements’.7 In R v Engert, Gleeson CJ, having discussed the decision in Veen, stated that: A moment’s consideration will show that the interplay of the considerations relevant to sentenc­ ing may be complex and on occasion even intricate … It is therefore erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.8

More recently, in Elias v The Queen, the High Court explained: As this Court has explained on more than one occasion, the factors bearing on the determina­ tion of sentence will frequently pull in different directions. It is the duty of the judge to balance often incommensurable factors and to arrive at a sentence that is just in all of the circumstances. The administration of the criminal law involves individualised justice, the attainment of which is acknowledged to involve the exercise of a wide sentencing discretion.9

It is the weighing of all relevant factors to arrive at a single result which takes due account of them all which is encapsulated in the term ‘instinctive synthesis’. As the High Court of Australia said of the term in Wong v The Queen, the expression is used, not as might be sup­ posed, to cloak the task of the sentencer in mystery, but to make it clear that he is required to reach a single sentence which balances many different and conflicting features.10 Use of the term ‘instinctive synthesis’ thus does not connote anything occult or meta­ physical (see Berlin, 1997: 46–47). It does not mean a judicial magic eye able to penetrate into something that we of ordinary mind (we who are seated not on the raised Bench, resplendent in horsehair wig, gown and falls, but in the well of the court, the public benches or—heaven forbid—the dock) cannot apprehend. The instinctive synthesis is perfectly ordinary and quasi-aesthetic in the way that it works (Berlin, ibid). As we have seen, the instinctive synthesis is essentially a human judgment based on all the factors of the case, the sentencer’s experience, data derived from comparable sentences, and guidelines and princi­ ples authoritatively laid down in statute and case law. It is a technique allowing experienced sentencers to make sense of a complicated area which embraces matters of philosophy, public policy, sentencing law and moral issues, along with simple common sense (Davies and Tyrer, 2003: 264–65). The instinctive synthesis was born of a legal system whose courts remain firmly of the view that the retention of a wide judicial discretion is necessary and that individual justice is possibly more important than some more abstract notion of systemic fairness (Freiberg, 2001: 35; Krasnostein, 2015a: 18; Warner, 2014: 371). This accords with the approach in jurisdictions such as Scotland and Canada whose sentencing jurisprudence encourages judges to see every offender as an individual for whom the appropriate and just sentence cannot simply be pulled from a mould (Brown, 2013a: 8; Hutton, 2006a: 156; 7 

R v Rushby [1977] 1 NSWLR 594 at 597. R v Engert (1995) 84 A Crim R 67 at 68 (emphasis added). 9  Elias v The Queen [2013] HCA 31 at [27] (references omitted). 10  Wong v The Queen [2001] HCA 64 at [75]. See also Russell v The Queen [2011] VSCA 147 at [58]. 8 

54 

Equity and the Rule of Law in Sentencing

Hutton, 1999: 173; Manson, 2001: 81; Nicholson, 1992: 177). The instinctive synthesis is the method most compatible with the principle of individualised justice, ensuring that the judge imposes a sentence that is just and appropriate in all the circumstances of the particu­ lar case (Australian Law Reform Commission, 2006: 155; Laws, 2004: 64; Spigelman, 2008: 450–51). As Mahoney JA noted in delivering the leading j­udgment in Kable v DPP (NSW): What may and should be done [in sentencing] is not to be determined by the articulation of gen­ eralities. The law must be effective to do what needs to be done in the individual case. If justice is not individual, it is nothing.11

The sentencer’s broad discretion is required because, as the High Court of Australia noted in Weininger v The Queen, sentencing is not a syllogistic process but is ultimately ‘a synthesis of competing features which attempts to translate the complexity of the human condition and human behaviour to … units of punishment usually expressed in time or money’.12 Whilst the term ‘instinctive synthesis’ masks the sentencer’s knowledge and experience (and to this extent can actually be seen as something of a misnomer), this terminology arguably encapsulates the essence of what is widely acknowledged to be a most complex, daunting, and thankless task (see Ashworth et al, 1984: 55; Hogarth, 1971: 303–04; Judge, 2014: 678 and 2015: 270; and Mackenzie, 2005: 29–31).

The Rule of Law Edney and Bagaric consider that the major problem with the instinctive synthesis is that it allows ‘too much scope for the personal sentiments of sentencers to play a role in sentenc­ ing’ (2007: 28 et seq; see also Bagaric, 2015: 77–78 and 104–08). They consider that the possibility exists of a disposal in any given case being influenced by the ‘I’ or the ‘me’ of the decision maker. This, they state, would be unacceptable in all other areas of the law; it would give rise to complaints of lack of predictability and certainty. They claim that it would result in the inability of citizens to know with reasonable certainty their obligations under the law and consider that there is nothing in the subject matter involved in criminal proceedings that distinguishes them from other branches of the law in terms of the need for clarity and transparency (Edney and Bagaric, ibid 29). Most importantly, the authors claim that the instinctive synthesis approach to sentencing violates the rule of law (ibid; see also Bagaric, ibid 104; Bagaric and Pathinayake, 2013: 410; Hutton, 2013b: 115; Manson et al, 2016: 41; and Roberts et al, 2011: 528). The rule of law is the formal name given to the state of affairs in which a legal system is legally in ‘good shape’ (Finnis, 1980: 270).13 Although there are several conceptions of the rule of law (Solum, 2008: 143), it was defined by Lord Bingham of Cornhill as the propo­ sition that all persons and authorities should be bound by and entitled to the benefit of laws publicly made, taking effect (generally) in the future and publicly administered by the 11 

Kable v DPP (NSW) (1995) 118 FLR 444 at 464. Weininger v The Queen [2003] HCA 14 at [24]. 13  It is standard within legal theory to separate conceptions of the rule of law into formal and substantive branches (Tamanaha, 2004: 91). Formal theories focus on the proper sources and form of legality, whilst sub­ stantive theories also include requirements about the content of the law, usually that it must accord with moral principle or justice (ibid 92). 12 

The Instinctive Synthesis and the Rule of Law

 55

courts (Bingham, 2010: 8). For present purposes, it can be read in the sense that sentenc­ ers shall be both ruled by the law and subject to it (Posner, 2008: 70). The personal will or arbitrary decisions of government officials should not serve as the basis for imposing penal sanctions (the notion of ‘a rule of laws and not of men’). The inspiration underlying this ideal is that to live under the rule of law is not to be subject to the unpredictable vagar­ ies of other individuals; it is to be shielded from familiar human weaknesses such as bias, prejudice, error, ignorance or whim (Tamanaha, 2004: 122). The legal system should meet the requirement of regularity: decisions should reflect the precept that similar cases should be treated similarly (the corollary, often overlooked insofar as sentencing is concerned, being that different cases should be treated differently—see Green, 1996: 123, and Ash­ worth, 1987a: 24) and judges ought to justify the distinctions they make between persons by reference to the relevant legal rules and principles (Solum, 1994: 122). The basic intuition from which the doctrine of the rule of law derives is that the law must be capable of guiding the behaviour of its subjects (Raz, 2009: 213–14). There are a number of principles that can be derived from the basic idea of the rule of law, including the requirements that all laws should be prospective, open and clear; that laws should be relatively stable; that the making of particular legal orders should be guided by open, stable, clear and general rules; that the independence of the judiciary must be guaranteed; and that the principles of natural justice must be observed (Bagaric and Pathinayake, ibid; Edney and Bagaric, ibid 29; Fuller, 1969: 39 et seq; Raz, ibid 214–19; Waldron, 1990: 47). As Wal­ dron notes, although these principles or requirements may seem straightforward enough when applied to particular laws, they may become very demanding when applied to the web-like apparatus of an entire system of legal regulation (Waldron, ibid 48).

The Rule of Law and Judicial Discretion For Edney and Bagaric, the indeterminacy that is inherent in the instinctive ­synthesis approach to sentencing cuts across these principles, or virtues, derived from the rule of law (2007: 30). The authors consider that the problem is compounded by the failure of judges and legisla­ tures to clearly define uniform aims and goals of sentencing. They acknowledge the argu­ ment that this can be seen as no more than an illustration of the typical tension between the desirability for ­certainty and flexibility in the law, with the requirement for a greater degree of flexibility in sentencing law being necessitated by the greater number of ­considerations which are potentially relevant to the sentencing decision. The authors c­ ontend, however, that the wide discretion enjoyed by sentencers is more fundamental than a ‘mere’ conflict between certainty and flexibility (ibid).14 Edney and Bagaric decry the primacy of judicial discretion in sentencing. They observe that sentencing is the one area of law where judges have attempted to steer clear not only of inflexible rules, but also of broad principles (ibid 31). In a damning critique, Edney and Bagaric state that: There is no doubt that due to the limited nature of human foresight, future situations cannot all be anticipated in advance and, accordingly, in any area of the law a tenable argument can be made for

14 

See also Bagaric and Pathinayake, 2013: 410–11 where the same argument is repeated almost verbatim.

56 

Equity and the Rule of Law in Sentencing

maintaining some degree of discretion. But sentencing practice is so nebulous and unconstrained that even the outcome of stock in trade cases is unpredictable. As the situation presently stands, sentencing law is so indeterminate that judges are free to switch from one rationale to another as they choose, according to the case or type of case before them, and this amounts to a liberty to determine and to switch policy at a whim (ibid).15

The authors cite Ashworth’s work in support of this critique. Although they refer only to the second edition of Ashworth’s Sentencing and Criminal Justice (1995b: 60), Ashworth has consistently maintained that the rule of law should be brought as far into sentencing decisions as possible (Ashworth, 2010a: 76, 2005: 72–73, 2000: 63). More recently, Ash­ worth has contended that although it is one thing to agree that judges should be left with discretion, allowing them to adjust the sentence to fit the particular combination of facts in an individual case, it is quite another to suggest that they should be free to choose what rationale of sentencing to adopt in particular cases or particular types of case (Ashworth, 2015: 80). Freedom to select from amongst the various rationales of sentencing is a freedom to determine policy, not a freedom to respond to the unique circumstances of each case; it is tantamount to a licence to judges to pursue their own penal philosophies (ibid 80–81; see also Duff, 2005: 1167–68; and Krasnostein, 2015a: 116–17). Thus for Edney and Bagaric, the imprecise state of sentencing law results in a situation in which the rule of law virtues of consistency and fairness are often trumped by the idiosyn­ cratic intuitions of sentencers. Bagaric, in particular, goes so far as to denounce sentencing by way of instinctive synthesis as ‘pro-hunch and anti-law’ (Bagaric, 2015: 110; see also Bagaric and Edney, 2016: 34). Edney and Bagaric maintain that the instinctive synthesis approach to sentencing means that it is not possible to subject the decision to impose a par­ ticular sentence to logical evaluation as we do not know why and how the sentencer arrived at this decision (ibid).16 In denouncing the instinctive synthesis, Edney and Bagaric echo Judge Marvin Frankel’s complaint of ‘lawlessness in sentencing’ made over 40 years ago (Frankel, 1972, reproduced in von Hirsch et al, 2009: 237–42). Frankel’s concern was that decisions on sentence were made in the seeming absence of the protections of the rule of law: no standards for sen­ tencing existed and, as there was no requirement to give reasons, the accused was afforded no protection from inconsistent or unprincipled decisions (Adelman and Deitrich, 2008: 242–45; Ashworth, 1998a: 212; Roberts, 2009: 229; von Hirsch et al, ibid 237). As Frankel saw it, sentencing made a fetish of vagueness; it was a wasteland in the law characterised by the evil of unbounded discretion vested in officials ill equipped to undertake the task (Adel­ man and Deitrich, ibid; von Hirsch et al, ibid 240–42). More recently, Hutton has criticised the wide scope of judicial discretion in sentencing (Hutton, 2006a, 2003). Hutton considers that discretion is the exercise of choice about which rule is relevant where there is no institutional means of forcing the sentencer to make one choice rather than another. For Hutton, discretion in sentencing is not about deciding which rule is relevant as there are no rules; he sees sentencing decisions as naked expressions of value preference or freely exercised choice which are not determined by law 15  The argument is repeated verbatim in Bagaric and Pathinayake, ibid 411. The argument is also repeated almost word for word in Bagaric, 2015: 111–12. 16  The argument is repeated verbatim in Bagaric and Pathinayake, ibid 411. The argument is also repeated almost word for word in Bagaric, ibid: 112.

In Defence of the Instinctive Synthesis

 57

and are ‘hardly recognisable as legal decisions at all’ (2006a: 169; see also 2016a: 146–47). Hutton considers sentencing to be ‘an ad hoc decision which is announced by the judge and justified on the basis of the authoritative office of the judge rather than by reference to any set of rational rules or principles’ (2003: 323). Hutton goes so far as to conclude that ‘[i]ndividualised sentencing has no rational justification’ (ibid 324).

In Defence of the Instinctive Synthesis—Equity and the Rule of Law Is the sentencing landscape really a wasteland as bleak and as lawless as Frankel, Hutton, Edney and Bagaric would have us believe? In particular, can Edney and Bagaric’s claim that the instinctive synthesis violates the rule of law be sustained? Raz observes that when a political ideal captures the imagination of large numbers of peo­ ple, it is not uncommon for its name to become a slogan used by supporters of ideals which bear little or no relation to the one it originally designated (2009: 210). Raz stresses that the rule of law is just one of the virtues which a legal system may possess and by which it is to be judged. It is not to be confused with, inter alia, justice, human rights of any kind, or respect for persons or for human dignity (ibid 211). Whilst the doctrine of the rule of law rests on the basic idea that the law should be capable of providing effective guidance, Raz cautions that the principles deriving from the rule of law do not stand on their own; rather, they must be constantly interpreted in the light of the basic idea (ibid 218). In a similar vein, Fuller states that the stringency with which the principles should be applied, as well as their priority of ranking inter se, will be affected by the branch of the law in question, as well as by the kinds of legal rules that are under consideration (1969: 93). It is sometimes best to take advantage of, and to incorporate into the law, what Fuller describes as ‘common sense standards of judgment’ that have grown up in ordinary life lived outside the legislative halls (ibid 64). Conceptions of the rule of law adhered to by critics such as Edney and ­Bagaric are, as Tamanaha notes, grounded upon fear and distrust of others, reflecting a choice to prefer rule by law to unrestrained rule by another—even by a wise person—out of concern for the potential abuse that inheres in the power to rule (Tamanaha, 2004: 122; in this regard see especially Edney and Bagaric, 2016: 33 and Hutton, 2003: 323). Although powerful, the idea of ‘a rule of laws and not of men’ has been dogged by the problem that laws are neither selfinterpreting nor ­self-applying. The operation of law cannot be sequestered from human ­participation ­(Tamanaha, ibid 123; see also Sedley, 2011: 168). The ­inevitability of such participation, writes Tamanaha, provides the opportunity for the ­reintroduction of the very weaknesses sought to be avoided by resorting to law in the first place (Tamanaha, ibid). The indeterminacy of law and language suggest that this o ­ pening can never be shut completely, for, at the moment of application, legal rules cannot be applied without what Tamanaha describes as ‘the injection of human reason, insight and judgment’ (ibid 126). Edney and Bagaric’s position is arguably an example of what Raz describes as the ‘pro­ miscuous’ use made in recent years of the expression ‘the rule of law’ (Raz, ibid 211). In raising their objections to the instinctive synthesis, Edney and Bagaric have fallen foul of the tendency against which Raz has cautioned. The result is that Edney and Bagaric have

58 

Equity and the Rule of Law in Sentencing

framed their objections around a totem v­ ersion of the rule of law; a totem far removed from its original designation and meaning. It is recognised that convicted persons are entitled to the protection of the rule of law (O’Malley, 2016: 47, 2011: 10, 2003: 135 and Krasnostein, 2015a: 3–4); however, the dis­ charge of the sentencing task must necessarily be marked by a wide discretion due to its complexity. As judicial discretion is the defining aspect of the sentencing task, it exposes the shortcomings of legalism as a description of what the sentencer is called upon to do. As Judge Richard Posner notes, the task of the sentencer has almost nothing to do with legal analysis. The determinants of how severely an offender should be punished depend, rather, on the sentencer’s attitudes towards such large, contested and broadly ideological issues as personal versus social responsibility for the offending behaviour, the morality of retri­ bution, the feasibility of rehabilitation and the deterrent effects of the sentence imposed (Posner, 2008: 71). The manner in which a sentencer decides to dispose of a case will be a result of legal doctrine (of what offence has the accused been convicted?); institutional constraints (such as the five year maxima in the sheriff court); policy (such as the discount for a guilty plea); strategic considerations (is there scope for probation; has the offender complied with community service or community payback orders in the past?), and the equities of the case. These factors are themselves all mixed together and are all mediated by judicial experience (ibid 84–85). Thus, in the sphere of sentencing, conventional, ‘rule of law’ conceptions of how the judicial system operates are completely inadequate (ibid 90). Sentencing requires a form of detailed and flexible decision making that responds pragmatically to circumstance, not slavishly to principle (Waldron, 1990: 51).

Equity and a Virtue-Centred Theory of Judging In Western legal systems there is an equity tradition, traceable to Aristotle at least, which authorises—or indeed requires—judges to depart from legal rules in order to do justice in particular cases (Shapiro, 1994: 5). Solum argues that properly understood, the practice of equity—in the sense of doing particularised justice, when the just result is not required by, or is contrary to, the result required by the set of applicable legal rules;17 justice as fairness, distinguished from justice as lawfulness—is broadly consistent with the ideal of the rule of law (Solum, 2008: 145, 2003: 205, 1994: 120 and 123). By reference to Aristotle’s Nicomachean Ethics (Aristotle, 1976), Solum notes that the core of Aristotle’s view of epieikeia (‘equity’ or ‘fair-mindedness’ that allows flexible situational judgment) is that equity is just, but not what is legally just: it is a rectification of legal justice. This is necessary because all law is universal and there are some things about which it is not ­possible to pronounce rightly in general terms (Solum, 2008: 146, 2003: 205, 1994: 123–24). As Meyer explains, justice is inherently particular and contextual, not merely the con­ sistent application of rules (Meyer, 2007: 86). Any general legal rule may, for example, be over-inclusive or under-inclusive with respect to the goal that the rule is meant to achieve

17  Solum cites the dicta of Douglas J in Hecht v Bowles 321 US 321, 329 (1944) that in doing equity, the judge will ‘mold each decree to the necessities of the particular case’ (Solum, 1994: 123; see also Solum, 2008: 145).

In Defence of the Instinctive Synthesis

 59

(Solum, 2003: 205, 1994: 124). As Tamanaha notes, legal rules are set forth in general terms in advance and cannot anticipate or account for every eventuality that might arise (Tamanaha, 2006: 229). It is impossible to anticipate adequately the countless ways in which factors interweave and bear upon one another in human reality (Nussbaum, 1993a: 119). Thus, sentencing is inevitably a discretionary activity since no set of rules or guidelines can possi­ bly anticipate all the unique circumstances of a particular case in the future (Bottoms, 1998: 69; Nussbaum, 2001: 301 and 1990: 71). For Solum, equity corrects the law’s generality by filling gaps in the law, by adjusting con­ flicts and tensions amongst legal provisions, and—crucially for the sentencer—by making exceptions in cases in which the rule leads to unjust results (Solum, ibid; see also Nuss­ baum, 1993a: 92–93 and 1990: 69). In the context of sentencing, equity is predicated on the unique features of the particular case; it tailors the law to the requirements of the case and the demands of the particular situation (Solum, 2008: 147, 2003: 205–06, 1994: 124–25 and 136; see also Brien, 1998: 90; Nussbaum, 1993a: 85; and Sitze, 2007: 196). Solum advocates what he terms a ‘virtue-centred’18 practice of equity and t­heory of judging. For Solum, departing from the rules in order to achieve particularised justice is consistent with the rule of law when legal decision makers possess certain judicial virtues— characteristics of mind and will that produce excellence in judging. Two of the key virtues of a good judge, Solum notes, are judicial intelligence and judicial wisdom. Judicial intelligence is a form of sophia, or theoretical wisdom, and comprises an ability to understand and theorise about the law. The good judge must be able to engage in sophisticated legal rea­ soning and, of particular importance in sentencing, must have an ability to grasp the facts of cases that may involve other disciplines such as sociology, criminology and economics (Farrelly and Solum, 2008: 14; Solum, 2003: 191–92, 1994: 130–31). Judicial wisdom, meanwhile, is the particularly adjudicatory form of the intellectual vir­ tue of phronesis, or practical wisdom (Solum, 2008: 156, 2003: 192, 1994: 120–21). The con­ cept of phronesis as it applies to sentencing is examined in detail in Chapter 5 by reference to the work of Bent Flyvbjerg. For present purposes, however, it is sufficient to note that insofar as sentencing by way of instinctive synthesis is concerned: The person of practical wisdom knows which particular ends are worth pursuing and knows which means are best suited to achieve those ends. Judicial wisdom is simply the virtue of practical wis­ dom as applied to the choices that must be made by judges. The practically wise judge has developed excellence in discerning what goals to pursue in the particular case and excellence in choosing the means to accomplish those goals’ (Solum, 1994: 131, emphasis added; see also Solum, 2008: 156 and 2003: 192).

The key to effective sentencing is knowing how to construe a particular case, how to describe and classify what is before one (Sherman, 1989: 29, cited in Solum, 2008: 157). As a phronimos, or a person of practical wisdom, the sentencer has—or ought to have—the experi­ ence to discern what is morally relevant about a particular situation (Nussbaum, 2001: 306; Richardson, 1997: 184; Solum, 2008: 157, 1994: 136). The sentencer is thus able to discern those situations in which a departure from the letter of the law is consistent with its spirit

18 Originally ‘virtue-centered’ (Solum, 1994), later anglicised to ‘virtue-centred’ (Solum, 2003) and most recently appearing again as ‘virtue-centered’ (Solum, 2008). The anglicised spelling is employed throughout the present work.

60 

Equity and the Rule of Law in Sentencing

and hence with the rule of law (Solum, 1994: 121; 145). Law cannot but speak through people and sentencers must possess the judgment, wisdom and character encapsulated in the concept of phronesis. Were this not so, then sentencing, as a branch of the criminal law, would be dull-minded, vicious and oblivious to its consequences (Tamanaha, 2004: 125). Solum’s virtue-centred theory of judging creates a conceptual link between the correctness of a legal decision (for present purposes, the sentence imposed) and the decision that would be made by a virtuous judge (Solum, 2003: 199, 1994: 129). Three definitions specify the nature of the link: (i) a virtuous decision (a legal decision is right if, and only if, a virtuous judge would make that decision under the circumstances); (ii) a virtuous judge (one who occupies the legal role of adjudicator and who possesses and acts in accordance with the judicial virtues); and (iii) judicial virtue (a dispositional quality of mind or will that promotes excellence in judging, including judicial integrity and fidelity to the law but also, as we have seen, judicial intelli­ gence—sophia or theoretical wisdom—and judicial wisdom—or phronesis—which includes an ability to assess the consequences of one’s decisions and take them into proper account) (Solum, 1994: 129–30). As equity is the tailoring of the law to the demands of the particular sit­ uation, a process which requires both moral vision or legal perception and the ability to com­ prehend the morally and legally relevant features of the particular case, it can (or should) be done only by a phronimos: only a virtuous judge can do equity (Solum, 2008: 157, 2003: 206). Thus Solum’s model acknowledges the relevance of both legal rules and the consequences that follow from the decision, but neither is critical. Good decisions are made in accordance with the judicial virtues (the key virtues in the sentencing context being those of sophia and phronesis) to produce an individualised decision, tailored or crafted to the particular constellation of facts and circumstances of the individual case. This is well illustrated by Nussbaum in her own description of equity: The world of epieikeia or equity … is a world of imperfect human efforts and of complex obstacles to doing well, a world in which humans sometimes deliberately do wrong, but sometimes also get tripped up by ignorance, passion, poverty, bad education, or circumstantial constraints of various sorts. It is a world in which bad things are sometimes simply bad, sometimes extremely bad, but sometimes—and more often, when one goes into them—somewhat less bad, given the obstacles the person faced on the way to acting properly. Epieikeia is a gentle art of particular perception, a temper of mind that refuses to demand retribution without understanding the whole story […]’ (Nussbaum, 1993a: 91–92).

The sentencer as a phronimos must closely examine the offence and the offender in order to fully comprehend the particular circumstances of the individual case (Nussbaum, ibid 97–98). The sentencer’s assessment will be searchingly particular, devoted to a deep and internal understanding of each concrete case (ibid 99). As Nussbaum explains, on this view of sentencing, the inner world of the offender is treated as deep and complex; it instructs the sentencer to investigate that depth (ibid 111; see also 2001: 300). Equity and Solum’s virtue-centred approach to j­ udging act as a kind of jurisprudential diving bell which allows the sentencer to investigate these depths.

Reconciling Equity in Sentencing with the Rule of Law Solum contends that the virtue-centred account of judging serves as the basis for two main arguments that the practice of equity can be reconciled with the rule of law. Firstly, when

In Defence of the Instinctive Synthesis

 61

equity is done by a virtuous judge, it is not an arbitrary departure from the rules: the pur­ poses and values immanent in the law itself still hold sway, despite the judge’s departure from the letter of the law (Solum, 1994: 136; see also Posner, 2008: 13). As Shklar observes, justice is the constant disposition to act fairly and lawfully (Shklar, 1987: 3, cited in Solum, ibid; see also Solum, 2008: 158, and 2003: 194–98). A judge who adopts Solum’s virtue-cen­ tred theory will consider the particular facts and circumstances of the case and will carefully consider the values expressed in the rule of law. If the judge considers that doing equity is appropriate in the circumstances, the virtue-centred theory tells him that he should exercise his practical wisdom and depart from the rules (Solum, 1994: 137). Thus equity done by a judge who is a phronimos does not substitute the rule of an individual for the rule of law (Solum, 2008: 158). The second argument for the reconciliation of the practice of equity and the rule of law is that the notion of moral and legal vision that grounds the practice of equity is also neces­ sary to understanding the application of legal rules (Solum, 1994: 139). As Solum explains: Rule application can only take place after a judge comprehends the facts of the case. Moral and legal vision is required in order to reveal that a case is governed by a rule. For this reason, both the application of legal rules and the practice of equity require the virtue of judicial wisdom, or phronesis (ibid; see also Solum, 2008: 160).

Any ethical reflection will first involve a perceptual grasp of what is morally ­significant and what is not; it is only with such a grasp that the practice of deliberation can get off the ground (Tadros, 2006: 180). Thus there can never be a simple set of rules which decide the issues facing the sentencer, for even the application of rules will require some prior moral comprehension of the complexities of any given situation (ibid). As Posner states, wisdom in applying rules requires a sense of when the rules run out and (what is not necessarily a different question) when it would be a serious mistake to apply a particular rule ‘as written’ (Posner, 1993: 143). The practice of equity may thus actually reinforce, rather than undermine, the values of predictability and regularity that support the ideal of the rule of law (Solum, 2008: 161, 1994: 139–40; see also Brien, 1998: 90 and Posner, 2008: 252). Nussbaum stresses that we do not have to choose between the doing of equity and the rule of law as understandings of what justice demands (Nussbaum, 1993a: 96). The point of the rule of law is to bring us as close as possible to what equity would discern in a variety of cases. As we have seen, however, legal rules can be too general and insufficiently sensitive to the particular case. When a strict application of such rules leads to manifest error it is, as Nussbaum states, justice itself, and not a departure from justice, to use equity’s flexible standard (ibid; see also Nussbaum, 1990: 99–100). Nussbaum illustrates the idea of ethical flexibility by recounting Aristotle’s m ­ etaphor of the ‘Lesbian Rule’. The decision maker who persists in choosing according to a rigidly predetermined standard is like an architect who tries to use a straight ruler on the intri­ cate curves of a fluted column. Instead, the good architect will, like the ancient builders of Lesbos, measure with the Lesbian Rule—a device still used today comprising a flexible strip of metal that is not fixed and which bends to the shape of the stone (Nussbaum, 2001: 301, 1993a: 93, 1990: 69–70; see also Crowder, 2002: 191–92). As Nussbaum explains, good deliberation, like the Lesbian Rule, ‘accommodates itself to the shape that it finds, respon­ sively and with respect for complexity’ (Nussbaum, 1990: 70; see also 2001: 301).

62 

Equity and the Rule of Law in Sentencing

Judicial Objectivity, the ‘Postmodern’ Sentencer and the Concept of Habitus Postmodernism drove doubts about the possibility of objectivity into society and legal culture (Tamanaha, 2006: 239). Postmodernism suggests that: ‘The human subject is an embodied agent, acting and judging in a context that can never be wholly objectified, with orientations and motivations that can never be fully grasped or controlled’ (Tarnas, 1991 cited in Tamanaha, ibid; see also Tamanaha, 1997: 3–10). Under this view, although judges may have a sincere desire to decide cases objectively, they subconsciously see the law through an ideologically coloured lens (Tamanaha, 2006: 239). Tamanaha accepts this as a basic proposition, but resists its sceptical implications. Judges do indeed approach the law from the standpoint of their personal views and, more immediately, see the law both from within the lens of the legal tradition in which they have practised and from within the conventions of legal practice and judging in which they con­ tinue to participate. ­Crucially for Tamanaha, however, the totality of the legal tradition— which includes the legal language, the corpus of legal rules, concepts, principles and ideas, legal p ­ rocesses and practices, hierarchical legal institutions, and the craft of lawyering—has the effect of stabilising legal meaning and providing restraints on the influence of the sub­ jective views (ibid; see also Tamanaha, 1997: 223–24). These ‘regularities in behaviour’ are due to the fact that individuals within a community who are similarly socialised often have shared ideas, beliefs and r­easoning patterns and operate under similar constraining conditions (Tamanaha, 1997: 68; 199). Sentencers share the mind and ways of doing that make up both the community of judges and the practice of sentencing (ibid 223). This is illustrated by the comments of one respondent in the present study on the importance of adhering to guidance from the Appeal Court: I have to be obedient to what the Appeal Court have told me. I am a great believer in the principle of stare decisis—if there’s a case in point then you follow it … I think that is something that many people talking about sentencing overlook—obedience to the authority of higher courts. I am obe­ dient (Sheriff 9).

Consonant with Tamanaha’s notion of ‘regularities in behaviour’ amongst similarly social­ ised individuals within a community, the Appeal Court has held that sheriffs can, in appropriate cases, take the frequency of particular crimes in their district into account in sentencing.19 Thus, another respondent referred to the operation of unofficial ‘going-rates’ for particular offences in individual locales: When I was first appointed and sat at [name of town] it was not long after there had been a lot of stabbings there. There was a lot of knife crime—particularly in and around this nightclub—and there was actually an unofficial policy that if someone was found in possession of a knife, they went to jail for three months. Everyone in the street knew that … And everyone knew where they were so knife crime dropped enormously at that point in [name of town] … So I was in Rome and did what the Romans did! I was told they get three months for possession and that’s it! So everyone expected that. I even heard offenders talking about it—‘Ken [“you know” in colloquial Scots], you get three months for that’ (Sheriff 15).

19  Blair v Hawthorn 1945 JC 17 at 20; Campbell v Johnston 1981 SCCR 179 at 182; Murray v McGlennan 1991 SCCR 18 at 19; McCleary v Walkingshaw 1996 SCCR 13 at 14–15; Donnelly v HMA 2009 SCCR 512 at [12]; see also Nicholson, 1992: 185.

In Defence of the Instinctive Synthesis

 63

Tamanaha notes that law is a socially produced and shared activity in which p ­ articipants are not free to do as they choose and that judges who stretch legal rules beyond recogni­ tion risk disapproval or rebuke from appellate courts (2006: 239–40). This is demon­ strated by the decision in Kane v HMA20 where the ­sentencing sheriff was criticised by the Appeal Court for erroneously stating that the Court had repeatedly made it clear that the only possible sentence for robbery at ­knife-point was a lengthy custodial one. Such factors are the social and ­institutional mechanisms of perpetuating and enforcing conformity in the interpretation of legal rules. Thus Tamanaha’s account incorporates the postmodern insight about the influence of background views on how people see the world, adding the reminder that the legal tradition itself is such a body of background views which becomes an integrated aspect of the judge’s own perspective (ibid 240). Hutton (2006a) draws on Bourdieu’s concept of habitus to explore the ­routinised day to day practices of sentencers.21 Habitus is ‘the durably installed generative principle of regu­ lated improvisations which produce practices’ (Bourdieu, 1977: 78, cited in Hutton, ibid 162). It has been described as a ‘feel for the game’ (Webb et al, 2002: 38, cited in Hutton, ibid) which, as Hutton notes, sits easily with s­ entencers’ own perception of sentencing as an intuitive process (ibid; see also Hutton, 2016a: 154). Judges’ legal training and experience in practice at the Bar teaches them how to think, argue and make decisions in a judicial way. This, combined with the fact that they work within the same criminal justice institutions, patterns their sentencing behaviour. These social factors enable judges to make sentencing decisions whilst also setting limits on those decisions (Hutton, 2006a: 162). As Hutton explains: Judges have a limited repertoire from which to draw in making their sentencing ­decisions, but this repertoire can be drawn upon in varied and sometimes novel ways. There is scope for different judges to make different interpretations or evaluations of the same facts. This means that although sentencing is patterned, the pattern is far from uniform, and inconsistencies are found alongside the consistencies (ibid).

Thus, sentencing decisions emerge from the habitus. Whilst for most experienced sentenc­ ers, the habitus is an ‘unthinking common sense approach that becomes second nature’, for Bourdieu it is a socially constructed set of values, motives, m ­ ethods of analysis, judgemen­ tal criteria, assumptions, classifications, and ­categories which are distinctively the condi­ tions of existence of judges (­ Hutton, ibid 163).22 That habitus enables judges to draw from their sentencing ­repertoire in potentially novel ways is demonstrated by Scottish sentencers’ crafting of ­appropriate disposals through the application of equity.

20 

2003 SCCR 749 at [10]. by means of purely doctrinal analysis, without conducting any qualitative, empirical research with sentencers. 22  See also Giddens’ concept of ‘practical consciousness’ (Giddens, 1984, cited in Bottoms, 1998: 69–70). In fac­ ing a novel situation, sentencers draw on their experience of similar situations in the past, plus their training and the formal rules and procedure of sentencing law. The action that the sentencer takes then itself becomes some­ thing that may be drawn on (implicitly or explicitly) in future situations. Thus, the sentencer develops a practical consciousness of how to ‘go on’ in a given situation, although (s)he cannot articulate this kind of knowledge with any precision. These notions are what Bottoms describes as the ‘deep structures’ of consciousness in the occupa­ tion in question (Bottoms, ibid 70). 21  Albeit

64 

Equity and the Rule of Law in Sentencing

Equity in the Sentencing Practice of the Scottish Courts The practice of equity as dispensing individualised justice is nowhere more evident than in judicial sentencing practice in Scotland. In Scotland, there has traditionally been no concept of a ‘tariff ’ for specific offences with the courts preferring a pragmatic and flex­ ible approach to questions of sentence (Kelly, 1993: 1; Millie et al, 2007: 250; Nicholson, 1992: 177–78; Tata, 2010a: 197; Tombs, 2004: 35). Although this has changed to an extent in recent years with the Appeal Court now being more willing to issue both sentencing guideline judgments23 and to consider questions of principle in sentencing (Brown, 2012a: 4), Scottish sentencing practice still accords with that of Canada and of Australian jurisdic­ tions in that it remains an individualised process in which discretion continues to be the hallmark. Tombs explains: In producing coherent sentencing narratives, sentencers consider the ‘facts’ of the cases presented to them, the information provided about the circumstances surrounding the crime and offender in question, and create interpretive frameworks within which those facts and circumstances con­ sidered relevant are evaluated … [I]n order to sentence individuals (as opposed to legal categories of offenders), [judges and sheriffs] continue to draw on traditional sentencing values (for example, showing mercy) and their creative abilities in constructing meaningful stories about the crimes committed by contextualised individuals (Tombs, 2008: 84–85; see also Aas, 2005: 78–80 and 86–87).

This individualised approach, together with Solum’s criticism of general legal rules as being over-inclusive with respect to the goal that the rule is meant to achieve, is well illustrated by considering how statutory minimum sentences o ­ perate in practice.

Statutory Minimum Sentences—Judicial Views on Drugs Offences In Scotland, statutory minimum sentences are prescribed for murder (section 205 of the Criminal Procedure (Scotland) Act 1995), certain firearms offences (section 51A of the Fire­ arms Act 1968)24 and certain drugs offences. Under section 205B of the 1995 Act, where an offender aged 18 or over is convicted in the High Court of a third Class A drug trafficking offence, then he must be sentenced to a minimum of seven years’ imprisonment (section 205B(2)) unless there are specific circumstances relating to any of the offences or the offender which would make that sentence unjust (section 205B(3)).25 A ‘drug trafficking offence’ is defined in section 205B(5). As Shiels et al observe, it has a much wider meaning than merely dealing in or being concerned in the supply of controlled drugs and includes, for example, permitting premises to be used for drug related activity (Shiels et al, 2016: 522). Further, ‘conviction’ is defined by section 205C of the 1995 Act in such a way that the automatic sen­ tence of seven years for a third drug trafficking offence will apply even if one, or both, of the earlier convictions was disposed of by way of a probation order (ibid 523). Thus, the broad

23 

Discussed in Chapter 6. For commentary, see Thomas, 2012a: 399, 2011a: 170, 2007: 4, 2006: 656 and 2005a: 882. See also Cochrane v HMA 2011 SCCR 63 (cf HMA v McGovern 2007 JC 145 and Lowe v HMA 2008 SCCR 760). 25  For the approach of the Court of Appeal (Criminal Division) in England and Wales to the corresponding English provision, see the decisions in R v McDonagh [2006] 1 Cr App R (S) 111; R v Stenhouse [2000] 2 Cr App R (S) 386; R v Turner [2006] 1 Cr App R (S) 95, and the discussion by Ashworth, 2012: 83–85. 24 

In Defence of the Instinctive Synthesis

 65

intention of section 205B is to limit the sentencing discretion available to judges when deal­ ing with recidivist drug offenders. One of the judges interviewed (a relatively recent appointment to the High Court Bench) acknowledged that the operation of section 205B and the imposition of minimum sen­ tences could be ‘tricky’; however, he considered that in such instances he ‘just [had] to go with what Parliament has decreed is the appropriate sentence’ (Judge 6). This view can be contrasted with the position taken by two more experienced judges who had between them sat in the High Court for a period well in excess of 20 years and who had encountered many cases struck at by section 205B. The first judge emphasised the importance of ‘tailoring the sentence to the specifics of the case’. He stressed that the current offence and the two previ­ ous convictions ‘may be chalk and cheese’ thereby necessitating frequent recourse to the statutory ‘escape route’ of section 205B(3): The two previous convictions might have been 10, 15, 20 years ago and they might equally have been tiny amounts [of drugs] or have been possession with intent to supply—not a big distribu­ tion amount. In between, the person may have no convictions whatsoever. They may be seriously valuable members of the community in other respects … Yet someone else who is subject, appar­ ently, to the same minimum sentence may have major convictions under section 4(3)(b) [of the Misuse of Drugs Act 1971—being concerned in supply], or import offences which are usually bigger amounts … The differences between these two cases are absolutely enormous, but the statute says that the minimum sentence has got to be applied, subject to the escape route. And the escape route is one that we have to take all the time because the minimum sentence simply does not fit all—and it can’t (Judge 3; see also Spohn, 2009: 253).

The second judge also stressed the importance of considering ‘not only the seriousness of the offence but the personal circumstances of the accused and the circumstances surround­ ing the commission of the offence’. He began by noting that the kinds of cases now indicted in the High Court have changed dramatically over the past 30 years. Until about the 1970s most accused who appeared in the High Court were, as he put it, ‘criminals—bank robbers, people who made a career out of housebreaking, gang leaders’. This changed in the late 1970s and early 1980s: Drugs have transformed the situation. Drugs have had two effects: firstly, the drug trade has brought youths and women into the criminal justice system; it’s not just the traditional criminals, if I can call them that, any more. Secondly, drugs have brought people into the system at a very high level. In the past, there was a definite trend of the criminal classes working their way through the system. They’d start with appearances in the Police Courts, proceed to the Sheriff Courts and even­ tually end up in the High Court. That’s all changed with drugs. Many first offenders now enter the system at High Court level when it comes to being concerned in the supply of drugs. The Crown takes a strong line (Judge 5).

This judge criticised the minimum sentence provisions for drug trafficking as being liable to lead to injustice: I don’t think dealing in Class A drugs and prison should be inextricably linked. You want to know a great deal more about the person than in the past. You have to know about their personal cir­ cumstances and their background. With drugs offences, they may be a villain, a major importer of drugs. But many offenders do just succumb to pressure. For example, you sometimes get young or even quite elderly offenders who want to make easy money. These people are targeted by the Mr Bigs to take packages from down south back up here. I’ve also heard it said that the Mr Bigs

66 

Equity and the Rule of Law in Sentencing

sometimes offer up a sacrifice to keep the police happy. So you get this old guy, or it’s often a naïve youngster, who drives up from England, he’s been instructed to pull into a service station for the drop and all of a sudden he’s surrounded by armed police! You have to look at the offender’s motivation (Judge 5).

The approach of the more experienced judges to sentencing in drug supply cases where the mandatory minimum sentence applies accords with the position in England and Wales. Ashworth, for example, reviewed certain key decisions of the Court of Appeal and con­ cluded that the overall sense is that applying a ­minimum sentence would be unjust if a sin­ gle strong mitigating factor, which tends to ­negative a purpose of the minimum sentence, is present (Ashworth, 2012: 84).26

Statutory Minimum Sentences—Shrieval Views on the Scottish Parliament’s Aborted Proposals for Minimum Sentences for Knife Possession It is not only judges who expressed disquiet over the operation of minimum sentences. As with Mackenzie’s study (2005: 49, 2001: 231–33), there was no support for mandatory sen­ tencing amongst the sheriffs in the present study. At the time that the interviews were being conducted, the Criminal Justice and Licensing (Scotland) Bill was progressing through the Scottish Parliament. In a paradigm example of Bottoms’ populist punitiveness (1995: 39– 41), Labour MSP Richard Baker lodged an amendment at the stage 2 consideration of the Bill seeking to introduce a minimum sentence of six months’ imprisonment for any adult convicted of carrying a knife in a public place unless exceptional circumstances existed (McCallum, 2011: 4). An alternative amendment lodged by Conservative MSP Bill Aitken sought to establish a minimum sentence of two years’ imprisonment (ibid). As O’Malley observes of legislators’ tendency to introduce statutory minimum sentences: In the comfort of the parliamentary chamber and emboldened by the high-minded rhetoric of their colleagues, legislators can easily reach the conclusion that everyone who commits a given offence should receive sure and severe punishment. Yet, even a modicum of exposure to day-to-day proceedings in the criminal courts would quickly displace this presumption (O’Malley, 2008a: 12).

At its meeting on 2 March 2010, the Scottish Parliament’s Justice Committee agreed to call for written evidence on, inter alia, the amendments concerning sentencing for knife crime. Informed opinion in the form of submissions received from academics and from crimi­ nal justice stakeholders was unanimous in rejecting the introduction of such minimum sentences. Hutton and McNeill noted that there was little credible evidence to support the view that changes in criminal sanctions affect crime rates; they suggested that knife carry­ ing is better understood as part of a problematic culture amongst certain young males which the imposition of minimum sentences would do nothing to change (Hutton and McNeill, 2010: 1–2). Tata observed that mandatory sentencing is, by definition, insensi­ tive to individual differences between cases, in terms of both the relative seriousness of the

26  See also the decision of the Supreme Court of Canada in R v Lloyd [2016] SCC 13 in which a majority of the Court struck down a statutory minimum sentence of 12 months’ imprisonment for certain repeat drug offenders (section 5(3)(a)(i)(D) of the Controlled Drugs and Substances Act 1996). See especially [28] to [36] of the judg­ ment of the majority and the hypothetical examples discussed therein.

In Defence of the Instinctive Synthesis

 67

offence and the individual brought to court; it ‘turns judges into machines’ with sentenc­ ing becoming rigid, leading to perverse results (Tata, 2010c: 2; see also Aas, 2005: 31 and Spohn, 2009: 253). The Scottish Police Federation cautioned against ‘imprisoning young adults for what could be a moment of madness and foolishness’ resulting in ‘a fool entering prison only to leave six months or two years later as an educated criminal with a grudge against society’ (Steele, 2010: 2). The Association of Chief Police Officers in ­Scotland, mean­ while, stated that in its view ‘discretion in relation to sentencing should remain with sheriffs who would consider the full circumstances of each offence and determine an appropriate sentence’ (Scott, 2010: 2). Despite the Labour amendment being agreed at stage 2 by a majority of Justice Commit­ tee members, the provision was subsequently removed from the Bill by further amendment at stage 3, the removal being agreed by a small majority of the whole Parliament, (McCa­ llum, ibid). Although a mandatory minimum sentence for possession of a knife was, in the event, not introduced into Scots criminal law, the issue gave rise to some illuminating shrieval comments on the efficacy of mandatory sentences. All of the eight sheriffs who discussed sentencing knife crime stressed the importance of taking all the facts and circumstances of the offence and of the offender into account. Five sheriffs specifically criticised the proposal for a six month minimum sentence for possession of a knife. As one respondent put it: People commit offences for many, many different reasons. One needs a flexibility in disposal which reflects the variety—the constellation if one likes—of things which give rise to the commission of the offence. If you seek to create a specific disposal for a specific offence, that carries with it the implication that the factors which give rise to the offence and the disposal are uniform. And they are not uniform, nor will they ever be (Sheriff 10).

Another respondent discussed a recent case in which the offender had been prosecuted under section 49 of the Criminal Law (Consolidation) (Scotland) Act 1995 for possession of an article with a blade or point. The article in question was a craft knife of a design which, as the respondent noted, was readily available in the local branch of the discount retailer Poundland and which comprised an 11mm long triangular blade mounted on a length of folding plastic. In discussing the proposed minimum sentence and the sentencing of the particular offender, the respondent said this: You really can’t have a routine sentence. That case was correctly classified as a section 49 but it’s totally different from a kitchen knife, or a Bowie knife, or a diver’s knife … To simply say six months for that is just not fair. It’s not justice … I think there is sufficient opportunity for us to do justice within the existing [framework]. The real difficulty will come if you have mandatory sentences of six months for possession of a blade. That’s nuts. That’s just got to be nuts. What will happen is that there will be trials galore … I think there will be sheriffs who will be—ahem—‘unconvinced by the evidence’. That’s the reality of the situation … It’s not possible really to legislate for the huge variety of circumstances that might occur, and it’s just not worth trying (Sheriff 6).

The respondent cited another recent example where the accused—who had spent the evening visiting various pubs in the city centre—was found in possession of a knife which he used in the course of his work. The knife was recovered from a bundle containing the accused’s work clothes and equipment: So he’s staggering off, gets involved in an altercation and gets picked up. The police find his work knife in his jacket. Should he get six months for that? This was one of the few cases where, actually,

68 

Equity and the Rule of Law in Sentencing

there was justification on the assumption that you believe his story—he was collecting his bundle of work gear to go home and it wasn’t even in the pub he had this stuff, it was left somewhere else, in someone’s flat and he had gone to collect it. But he was prosecuted. Come on! It was completely unjust (Sheriff 6).

Another respondent (Sheriff 1) illustrated the injustice that can arise from a s­ystem of statutory minimum sentences for knife possession by discussing a recent case in which a 16 year old youth (a first offender) had committed a breach of the peace by threatening his social worker with a knife. The respondent explained that the youth was a member of a notorious local ‘criminal family’ and that he had succumbed to the knife carrying cul­ ture prevalent amongst his peer group. The respondent had been provided with what he described as the best social enquiry report he had ever read; a testimonial from the local Army Careers Office was also submitted along with a letter from the victim of the offence requesting that the respondent ‘give him a chance’. The youth was said to have come to realise the seriousness of what he had done and had shown what was described as ­genuine remorse. In order to allow the youth to pursue his intention to join the Army, the respondent imposed a probation order with a community service component of 150 hours. The youth completed the community service in a month—which the respond­ ent considered ‘an extraordinary achievement’—and was accepted into the Army. As the respondent explained: If you had looked at it in the kind of stark way that some politicians want you to do, you’d say ‘Well, he had a knife, so that’s two years. He brandished it at a social worker. That’s a very serious aggravating factor, so therefore we’ll start at two and a half years’. But that doesn’t allow the sen­ tencer to factor in all the things about the offender that even the v­ ictim was saying—‘Please give him a chance’. The concern was, if you send him to prison, he’s doomed. It’s a university of crime. He had a name to maintain that would involve him acting the hard man and not playing ball. He’d come out as the complete criminal … So that, for me, was an example of where anything between community service and three years would have been reasonable … And that’s the kind of thing that emphasises the danger of taking one component and saying, well, this is the decisive ­factor in this case and nothing else matters (Sheriff 1).

This respondent considered that if mandatory sentences are in force, ‘it doesn’t take long for manifestly unfair decisions to come up’. The respondent stressed the importance of sen­ tencers having a range of disposals from which to choose; this, he stated, allows the sentence to be individually tailored to meet the needs of the individual offender whilst still reflecting the seriousness of the offence. Such an approach was, the respondent stated, preferable to ‘the kind of American ­sentencing grid which just loses people’ (Sheriff 1). One respondent considered that, given the prevalence of knife crime in his jurisdiction, deterrence was an important factor in sentencing. However, he noted that such cases can be difficult to sentence: If we jailed everyone then the Director of Prisons would be turning on us. We can’t jail everyone for a first offence, but equally when it gets to their second offence then they’re appearing on indict­ ment. You’re a pretty stupid character if you’ve got one previous conviction and you do it again. The trouble about it is that that’s where you get the difficult ones coming up. You get the very dif­ ficult ones with the young women who, for some reason—whether to impress the boys, or whether they’re coerced into it—carry knives for their boyfriends on the basis that it’s more difficult to search a woman going into a club than it is to search a guy. These can be very anxious sentencing exercises (Sheriff 9).

In Defence of the Instinctive Synthesis

 69

As another respondent (Sheriff 12) noted, quite apart from the proposed s­tatutory mini­ mum sentence for possession of a knife, certain prosecution policies can themselves result in injustice. In particular, he criticised the Crown’s policy of prosecuting all second offences of knife possession on indictment as too rigid. The respondent discussed a recent case in which, following his ejection from hostel accommodation, the offender (a homeless man who had at least one previous conviction for carrying a knife) had shouted and swore in the street. The police were called. They proceeded to search the offender’s duffel bag (which contained all the offender’s possessions) and found a Swiss army knife. The respondent continued: The next thing was he was on indictment and I think he had been remanded in custody for a while. At the end of the day I thought the right thing to do was admonish him and I never heard any more about it (Sheriff 12).

The shrieval views on this matter illustrate the dangers of penal populists allowing ‘the electoral advantage of a policy to take precedence over its penal effectiveness’ (Roberts et al, 2003: 5). As Freiberg notes, when votes are seen as more important than the promotion of justice, the exercise can become ‘malignant’ and ‘dangerous’ (Freiberg, 2003: 223).

The Iniquity of Mandatory Minimum Sentences Regimes of statutory minimum sentences are thus based on the nomenclature of crimes rather than on the seriousness of individual acts (Aas, 2005: 20). As demonstrated by the findings of the present study, such regimes have proved unpopular with Scottish sentencers since they have great difficulty in accounting for the fact that offences with the same name may be very different with regard to their moral seriousness (see Aas, ibid). Mandatory sentences constrain j­udicial s­entencing discretion too tightly, producing the injustice of treating differently ­situated offenders similarly (Freiberg, 2016: 428; Krasnostein, 2015a: 227 and 2015b: 52–53; O’Malley, 2008a: 11). As Sir Gerald Gordon puts it: ‘mandatory sentences are Parliament’s way of stopping judges being soft or, if you prefer it, humane’ (Gordon, 2000). These concerns are shared by sentencers in other C ­ ommonwealth jurisdictions. Having compared the more significant mandatory minimum sentencing schemes in South Africa and Australia, Terblanche and Mackenzie conclude that such schemes are rarely a good idea (Terblanche and Mackenzie, 2008: 413; see also Terblanche, 2009: 278; cf Bagaric, 2000 and 2002, and the reply by Morgan, 2002a). There is little to commend the adoption of such schemes (Mackenzie, 2002: 85; Terblanche, 2001: 19; see also Mackenzie and Stobbs, 2010: 138), the main criticism being that they ignore the sentencing principle that offenders must be examined individually and that prescribed penalties should be miti­ gated in some cases depending on the outcome of the examination (Findlay et al, 2014: 276; see also O’Malley, 2006: 55 and Young, 2014: 156). Terblanche and Mackenzie note that mandatory minimum sentencing schemes contribute to prison overcrowding (ibid 410), thus any jurisdiction considering implementing such a scheme must ask whether its criminal justice system has the requisite prison capacity (ibid 413). It also appears that such schemes lead to inflationary drift in sentencing; the provisions of the South African legislation, for example, have been interpreted by the courts as indicating a general approach by the legislature that increased prison terms are appropriate, with the result that sentences for offences not included in the legislation have also increased (ibid

70 

Equity and the Rule of Law in Sentencing

410–11; Australian Law Reform Commission, 2006: 540). Terblanche and Mackenzie thus consider that mandatory minimum sentencing schemes act ‘like a double-barrelled gun’ and institutionalise bad sentencing practices (ibid 411). Another criticism of mandatory sentencing is a lack of proportionality, resulting in less consistency and fairness as the sentencer is not able to take all the facts and circumstances of a particular case into account (ibid 412; Australian Law Reform Commission, ibid; Fielding, 2011: 112–13; O’Malley, 2008a: 12; Roberts and Baker, 2008: 571; Tonry, 2002: 101; and Young, ibid). In discussing the operation of the Northern Territory scheme, for example, Morgan cites, inter alia, the cases of a young Aboriginal man sentenced to 12 months’ imprisonment for ­stealing $23 worth of biscuits and cordial; a 17 year old yo-yo thief sentenced to 14 days’ imprisonment, and the particularly poignant example of a home­ less man sentenced to 12 months’ imprisonment for taking a $15 towel from a washing line to use as a blanket (Morgan, 2000: 165; see also Santow, 2000: 299 and Warner, 2007b: 344). In R v Nur,27 the Supreme Court of Canada considered the constitutional validity of certain legislation that imposed mandatory minimum sentences for the offence of pos­ sessing a prohibited or restricted firearm when the weapon is loaded or is kept with readily accessible ammunition.28 A six-to-three majority of the Court struck down the minimum sentences as unconstitutional. Writing for the majority, McLachlin CJ set out the objections to mandatory minimum sentences: Mandatory minimum sentences, by their very nature, have the potential to depart from the prin­ ciple of proportionality in sentencing. They emphasize denunciation, general deterrence and ret­ ribution at the expense of what is a fit sentence for the gravity of the offence, the blameworthiness of the offender, and the harm caused by the crime. They function as a blunt instrument that may deprive courts of the ability to tailor proportionate sentences at the lower end of a sentencing range. They may, in extreme cases, impose unjust sentences, because they shift the focus from the offender during the sentencing process in a way that violates the principle of proportionality. They modify the general process of sentencing which relies on the review of all relevant factors in order to reach a proportionate result. They affect the outcome of the sentence by changing the normal judicial process of sentencing.29

Mandatory sentences thus ‘cast their net’ over a wide range of potential conduct30 and ‘iron out’ the differences between offences of the same type (Morgan, ibid 176). For example, had the proposed amendments to the Criminal Justice and Licensing (Scotland) Bill been accepted then the offender referred to by Sheriff 6, who inadvertently left his work knife in his bundle of work clothes, may have received the same sentence as a young gang member who goes out ‘tooled up’, in common parlance, to a pre-arranged gang fight. Such a result cannot be just (Tata, 2010c: 3). As one of the respondents in Mackenzie’s study put it: Not having judicial discretion in sentencing leads to harsh results. It doesn’t allow any real scope for any proper assessment of the quality of the conduct concerned. You can never predict, look into the future, and assess the types of circumstances which might arise (Mackenzie, 2005: 50).

27 

[2015] 1 SCR 773. Namely section 95(2)(a) of the Canadian Criminal Code, RSC 1985 which imposed sentences of three years’ imprisonment for a first offence and five years for a second or subsequent offence when prosecuted on indictment (Nur, ibid at [2] and [148]). 29  Nur (n 27) at [44]. 30  Nur (n 27) at [82]. 28 

In Defence of the Instinctive Synthesis

 71

Fixed penalty regimes often lead to different cases being sentenced in the same way (Rob­ erts, 2009: 229). ‘Rule of law’ virtues are thereby undermined as a result of the fact that public decision making by judges assumes less importance than largely unaccountable pretrial decision making (Morgan, 2002a: 143; 151; Tata, ibid: 3), this also being identified as a problem by certain respondents in the present study. The Australian mandatory sentencing regimes, meanwhile, have been criticised as leading to grossly unjust outcomes (Cuneen, 2002: 322) whilst the South African experience of mandatory sentencing also revealed harsh and unintended consequences (Terblanche and Mackenzie, ibid 415; Terblanche, 2001: 2–3). Terblanche notes, for example, that sentencers in South Africa have become so used to imposing the extremely long sentences that are contained in the legislation that many are losing the required sense of proportionality (Terblanche, 2009: 278). This recently led South Africa’s Supreme Court to state that ‘courts are not vehicles for i­ njustice’31 and to reiterate that ‘a court is not compelled to perpetrate injustice by imposing a sentence that is disproportionate to the particular offence’.32 Doing justice means having the capacity to reflect differences between offenders and not forcing different cases into one straitjacket (Morgan, 2002b: 300). Mandatory sentencing schemes have also failed to show any crime r­eduction effect (Ashworth, 2009: 252–53; Ashworth, 1998a: 215–16; Morgan, 2002b: 299, 2000: 174, 1999: 271–73; Roberts, 2009: 232; Tata, ibid 3; Terblanche and Mackenzie, ibid 411). The manda­ tory minimum sentence legislation in South Africa, for example, did not reduce crime rates (Sloth-Nielsen and Ehlers, 2005, cited in Terblanche and Mackenzie, ibid 408; Terblanche, 2001: 18–19). Writing in 1992, Tonry considered the evidence to be ‘clear and weighty’ that enactment of mandatory penalty laws had either no deterrent effect or a modest deter­ rent effect that soon wasted away (1992: 42; see also Tonry, 2007: 13, cited in ­Terblanche and Mackenzie, ibid 404).33 It appears that certainty of arrest, conviction and ­punishment serves as a possible deterrent rather than severity of punishment only (Ashworth, 2015: 84–85; Edney and Bagaric, 2007: 61–62; Morgan, 2002a: 148–49; Terblanche and Macken­ zie, ibid 407) and so an increase in sentence severity is unlikely to generate a proportion­ ate increase in deterrence (Manson, 2001: 43–46). That mandatory penalties have little or no impact on more serious offenders is due to the fact that serious offenders receive long sentences as a matter of the normal sentencing process, quite apart from the mandatory laws (Morgan, 2002b: 300 and Morgan, 2000: 173; 176). One of the most compelling reasons against introducing a system of mandatory sen­ tencing was articulated 40 years ago by Zimring (1977, reproduced in Duff and Garland, 1994: 164–73). Zimring noted that when the legislature determines sentencing ranges, it is operating at a level of abstraction far removed from individual case dispositions (ibid 171). In arguing for a system of individualised justice in which sentencers are able to take account of the particular circumstances of the individual case, Zimring observed: ‘[W]e lack the capacity to define into formal law the nuances of situation, intent, and social harm that condition the ­seriousness of particular criminal acts’ (ibid 169). More recently, the Australian Law Reform Commission examined the issue of man­ datory sentencing in some detail (2006: 538–41) and concluded that the prescribing of 31 

S v Vilakazi 2009 (1) SACR 552 at [18]. Ibid at [20]. 33  See also Nur (n 27) at [113]–[114]. 32 

72 

Equity and the Rule of Law in Sentencing

mandatory terms of imprisonment for federal offences was ‘generally incompatible with sound practice and principle in this area’ as it had the potential to offend against the prin­ ciples of proportionality, parsimony and individualised justice (ibid 541). The Commission considered that the judiciary should retain its traditional sentencing discretion to enable justice to be done in individual cases (ibid). It also stressed that the legislature should not prejudge the appropriate minimum penalty for an offence in legislation without regard to the facts of individual cases (ibid). Given that ‘the maintenance of individualised justice and broad judicial discretion are essential attributes of [the] criminal justice system, out­ weighing the potential deterrent effect that mandatory minimum sentencing might have’ (ibid 542), the Commission’s recommendation was that no mandatory minimum terms of imprisonment be prescribed for any federal offence (ibid). As Morgan notes, mandatory sentences are anathema to a progressive criminal justice system (2000: 168; see also 1999: 277). Mandatory sentences fail on almost every crimino­ logical criterion upon which they are measured (Freiberg, 2001: 42; Stobbs et al, 2014: 158). They do not assist in the appropriate structuring of judicial discretion, but restrict it in a way that severely limits the ability of the courts to do justice (Mackenzie, 2002: 84; Stobbs et al, ibid 157), thus demonstrating that it is far better to leave sentencing to the courts than for the legislature to pass this kind of legislation (Terblanche, 2001: 19). As the former Chief Justice of the Supreme Court of New South Wales explains: The degree of rigidity involved in [mandatory minimum sentences] inevitably leads to injustices arising in individual cases. Sooner or later those systems collapse under the weight of those injus­ tices. The principle of consistency has been served at the expense of other sentencing principles, especially the principle of individualised justice (Spigelman, 2008: 454).

A regime of statutory minimum sentences seeks to transform sentencing into a formal and standardised process. Such regimes introduce tariff based sentencing at the expense of individualised justice. The Scottish courts have always been opposed to such approaches. In Edward & Sons v McKinnon, for example, the Lord Justice Clerk (Cooper) stated that: ‘… for the purposes of estimating the gravamen of the offence and the proper penalty to impose, it is necessary to look beneath the technical form to the substance of the [offence]’.34 It is clear that this view is still one to which modern judges and sheriffs adhere. Thus the Scottish courts have always sought to do right in particular cases, as opposed to rigidly following universal rules. Although a study that specifically examined the issue of manda­ tory sentences would yield more representative views on the subject, the comments of both judges and sheriffs in the present study strongly suggest that the Scottish judiciary would be intensely opposed to any proposals to widen the use of mandatory minimum sentencing as being contrary both to the practice of equity and to the attainment of individualised justice.

Culpable Homicide Offences attracting statutory minimum sentences are certainly not the only occasions in which a sentencing judge is required to consider the practice of equity; it must also be con­ sidered in the sentencing of common law offences. Sentencing for the offence of culpable homicide, for example, has been acknowledged as notoriously difficult since the crime may 34 

Edward & Sons v McKinnon, 1943 JC 156 at 168 (emphasis added).

In Defence of the Instinctive Synthesis

 73

be committed in a wide variety of circumstances.35 In Walsh v HMA,36 however,—a case in which the deceased and the appellant had agreed to settle certain differences by way of a ‘square go’37—the Appeal Court appeared to suggest that custodial sentences are inevitable in cases of culpable homicide (‘No one can dispute that when a death is involved, custody is invariably an inevitable consequence’38). In contrast to the dicta in Walsh, one of the judges (who had discussed minimum sentences for drug trafficking offences) spoke of sentencing a case of culpable homicide where, unusually, the circumstances of the offence and of the offender were such as to allow him to depart from the usual practice of imposing a custodial sentence: I had a case of culpable homicide where a father killed his son in an argument. The son went for him, provoked him, and the chap picked up the nearest thing that came to hand—a kitchen knife— and stabbed him through the heart. I looked at the circumstances and the fact that he’d lost his son; he had to live with this. He’d led a blameless life up until this point. I admonished him. Now, that takes balls doesn’t it? (Judge 5).

Balls were indeed required by his Lordship in this case given that the Appeal Court held in HMA v Jamieson39 that serious injury, let alone death, inflicted using a weapon will usually require the imposition of a custodial sentence. Nevertheless, in conceptualising his sentenc­ ing decision in such terms the judge was implicitly referring to the moral courage required to do equity in disposing of this case. A further example is the decision of Lady Smith in HMA v Edge40—a tragic case in which the 80 year old accused pleaded guilty to the culpable homicide of his 85 year old wife. The accused and the deceased had been married for more than 50 years; the deceased had developed senile dementia and the accused had had difficulties in coping with her illness. He placed a pillow over her face and smothered her without any resistance before immedi­ ately telephoning the police. The circumstances of the case are such that one can question whether it really was in the public interest to prosecute the accused at all. In all the circum­ stances, however, the sentencing judge took the equitable decision to admonish the accused, a decision which the Crown did not appeal.41

Serious Assault Similarly, in the sheriff courts one sheriff made reference to a case of assault to severe injury and permanent disfigurement—an offence most often disposed of by way of custody— where the circumstances were such as to allow him to admonish the offender. The respondent (Sheriff 1) explained that the accused—a first offender—had been provoked into an argu­ ment with her teenaged daughter in the course of which she bit her daughter on the face. Whilst the respondent acknowledged that this was ‘an appallingly savage way of reacting’, 35 

HMA v McMillan [2005] HCJAC 119 at [12]. High Court of Justiciary on appeal, 10 November 2005, unreported. 37  For readers furth of Scotland who may be unfamiliar with the term, see Smart v HMA 1975 JC 30 at 34 (Lord Justice Clerk Wheatley). It is essentially an unarmed ‘duel’ by members of the criminal classes. 38  Walsh (n 36) at [3] (Lord Johnston). 39  1996 SCCR 836. 40  2005 GWD 20-360. 41  See also Murphy v HMA 2002 SCCR 1029 and Black v HMA 1998 GWD 34-1745 for ­examples of custodial sentences for culpable homicide being overturned on appeal and probation orders substituted. 36 

74 

Equity and the Rule of Law in Sentencing

he explained that the offence was put in ‘a completely different context’ when one consid­ ered the social enquiry report and the stress which the accused was facing at the time. The respondent continued: With the other things that were going on, I felt that in a sense the shame for her of being in the dock, the fact that the social work department were then involved in relation to whether she was a suitable mother or not, were enough to mean that the criminal justice system didn’t need to express displeasure or punishment. So I admonished her, or deferred sentence with a view to admonition … The whole other effects of being in the criminal justice system meant that you could look at a sentence that would, on paper, be a surprisingly lenient one, for the reasons I explained (Sheriff 1).

Another sheriff gave similar reasons for his decision to admonish three young offenders following their pleas of guilty to charges of assault: I had a solemn case some years ago where there were three relatively young guys. They were from [a local private, fee paying secondary school]. They had got out of hand one night and given someone a doing. They were all of impeccable background, no previous convictions. I was told by each of the agents—and I believed them—that this had been such an experience for them that they would never, ever be back in the courts. So in relation to the indictment matter I deferred sentence on them for a period of a year. At the end of it I admonished all three of them (Sheriff 8).

A further example was given by another sheriff who described a recent case where the accused had remonstrated with a youth who had made racist remarks about a passenger on a bus. The accused then head-butted the complainer and pleaded guilty to assault. The respondent (Sheriff 2) continued the case for three months for the accused to be of good behaviour: Technically he’d committed an offence … but I thought the circumstances were such that the foulmouthed youth deserved all he got! I mean, he wasn’t injured, you know, like a broken nose or anything like that. So after three months I admonished the guy. Now, I think that’s an example of leniency in the interests of justice (Sheriff 2).

An example of the Appeal Court’s practice of equity in an even more serious case of assault can be found in the decision in Calem v HMA.42 Here, the 22 year old appellant pleaded guilty to a charge of assault to severe injury and permanent impairment. Following a drunken altercation initiated by the complainer, the appellant was punched in the face. The appellant retaliated by repeatedly striking the complainer on the head and body with a dis­ carded piece of metal tubing with which he had armed himself in an effort to scare off the complainer. The consequences for the complainer were severe: he sustained two fractures to his jaw which required the insertion of metal plates and continued to suffer numbness in his jaw which was likely to be permanent. The social enquiry report disclosed that the appellant was a first offender and came from a good family; he had recently started university and was utterly ashamed of his conduct. The sheriff concluded, however, that only a custodial disposal was appropriate and sentenced the appellant to 12 months’ imprisonment. The sentence was subsequently quashed on appeal. The Appeal Court held that whilst the assault was of such gravity that the sheriff had been right to give consideration to a custodial sentence, he had given inadequate weight to the circumstances in which the inci­ dent arose. It was held that in all the circumstances, including in particular the appellant’s 42 

High Court of Justiciary on appeal, 9 June 2005, unreported.

In Defence of the Instinctive Synthesis

 75

previous good character, the gravity of the crime could be properly marked by imposing the maximum number of hours of community service and a community service order of 300 hours was imposed.43 Perhaps the paradigm example of the practice of equity in sentencing an offence that occasioned serious injury is the decision of Lord Turnbull in HMA v M (sub nom HMA v McArthur).44 The facts of the case were that the 39 year old accused, who was in employ­ ment and who had no previous involvement with the courts, pleaded guilty in the High Court to a charge of culpable and reckless conduct towards his three month old son. The accused’s wife of seven years suffered from a depressive illness which increased the levels of domestic strain. Having been left to attend to both his infant son and his three year old daughter one evening, the accused inserted a baby wipe into his son’s bottom when chang­ ing his nappy. The accused then tended to his daughter and forgot about the baby wipe. On ­checking the next morning, and being unable to locate the wipe, the accused inserted his finger into the baby’s bottom to locate it, without success. Some three days later, the baby was admitted as an emergency to hospital, where he was found to have sustained a perfora­ tion to his bowel, causing widespread and life threatening infection. The accused’s son had to undergo emergency surgery, at which point he was close to death. A police enquiry was subsequently initiated by medical staff and both children were taken into care by social services. In his written opinion on his decision on sentence, Lord Turnbull stated that this was without question the most difficult sentencing exercise which he had ever undertaken.45 The accused presented as a distraught and flawed parent whose understanding and regret for what he had done was obvious. Having regard to the offence committed, and balancing the factors of the accused’s work ethic and support of his family, Lord Turnbull considered that the case fell to be dealt with as an act of wholly misguided parental intervention occur­ ring at a time of considerable stress, rather than as an act of malice directed at the child.46 With all that had happened, Lord Turnbull considered that the concept of punishment seemed to be ‘somewhat out of place’.47 The accused was, his Lordship noted, a law abid­ ing man who was unlikely ever to offend again, and substantial punishment had already befallen him with the knowledge of what he had done and the loss of both his ­children to foster care. In the circumstances, Lord Turnbull considered that a custodial sentence would not be appropriate despite the fact that a focus on the harm done to the child might seem to some to lead inevitably to that outcome.48 His Lordship considered that, in the correct circumstances, compassion can properly feature in judicial assessment.49 Thus in doing justice to the particular facts of the case—in doing equity—Lord Turnbull admonished the accused.50

43  Ibid at [9]–[11]; see also the decisions in HMA v Q 2013 SCL 440; HMA v Gilmour 2004 SCCR 117; Macintosh v HMA High Court of Justiciary on appeal, 11 December 2013, unreported—see the case note by Brown, 2014c—and Muirhead v HMA 2014 SCL 702. 44  2010 SCL 1309. 45  Ibid at [23]. 46  Ibid at [22]. 47  Ibid at [23]. 48  Ibid. 49  Ibid at [25]. 50  Ibid.

76 

Equity and the Rule of Law in Sentencing

Sexual Offences The judicial practice of equity can also be seen in the decision of HMA v Currie.51 The respondent (a 17 year old first offender) was convicted in the High Court of the statutory offence of having unlawful sexual intercourse with a 13 year old girl and of the then com­ mon law offence of attempted rape52 of another 13 year old girl. Neither of the charges libelled any acts of violence towards the complainers or of their having been caused any physical injury. The complainer in the first charge gave evidence that she had willingly engaged in intercourse with the ­respondent; she had no complaint to make against him, she had not wanted to report the i­ncident and appeared to have lied about her age, claim­ ing to have been older than she actually was. The evidence in respect of the second charge was unclear. The assault appeared to have consisted of the respondent lying down on the complainer when she was asleep. When she awoke and told the respondent to desist, the incident came to an end. In her report to the Appeal Court, the sentencing judge considered that, in view of the respondent’s age, his status as a first offender and the circumstances of the offences, any custodial sentence would have been at what she described as the low end of the scale for attempted rape. She noted that the respondent was a young, immature teenager who had experienced substantial difficulties in his upbringing; he was now at a crucial stage of devel­ opment and had demonstrated a willingness to address his underlying behavioural prob­ lems. Her Ladyship continued: I was concerned that a custodial sentence would not achieve the necessary change in attitude, maturity and behaviour of the respondent. It would have been an easy solution to give a custodial sentence. But in my opinion custody will ‘harden’ the respondent and the longer term risks are likely to increase as a result.53

The respondent was sentenced to three years’ probation with an additional condition that he perform 200 hours of unpaid work in the community. The Crown appealed against the sentence as unduly lenient on the ground, inter alia, that the judge had failed to have suf­ ficient regard to the seriousness of the attempted rape. In presenting its appeal, the Crown submitted that the respondent had been convicted of the serious offence of attempted rape and on that basis alone the sentence selected was unduly lenient.54 Thus it could arguably be said that the Crown was seeking the introduction of a minimum sentence or, more prop­ erly, the introduction of a sentencing policy to the effect that a conviction for the offence of attempted rape should always result in the imposition of a custodial sentence. This conten­ tion was rejected by the Appeal Court. In delivering the opinion of the Court, Lord Osborne said this: The crime of attempted rape may range from an incident involving, as here, no violence, on the one hand, to one which might involve extremes of violence and consequent serious injury, on the other. The consequences of the crime may therefore vary enormously according to circumstances. We do not think that it can be maintained, as a general proposition, that a conviction for attempted 51 

2009 SCCR 48. The Scottish law of sexual offences was later re-written and put on a statutory basis by the Sexual Offences (Scotland) Act 2009. For commentary to the 2009 Act see AN Brown, 2015. 53  Currie (n 51) at 49. 54  Currie (n 51) at [8]. 52 

In Defence of the Instinctive Synthesis

 77

rape must always be visited by the imposition of a custodial sentence, regardless of the particular circumstances of the case.55

This dicta accords with the observation of the Appeal Court in the earlier ­decision of Baillie v HMA (a case involving a series of indecent assaults by a physician upon female patients) that a sentencing court must set such sexual offences at an ­appropriate point in ‘the spectrum of serious offences’.56 In particular, it was said that where sexual offences do not include penile penetration or physical violence, the range of the penalty must reflect that consideration.57 It was held that the sentence imposed on the respondent Currie was within the range of disposals which the sentencing judge was entitled to adopt in the circumstances. The Crown’s appeal was refused.58 The important point here is that, in rejecting the sugges­ tion by the Crown that a custodial sentence should automatically follow a conviction for attempted rape, the Appeal Court both confirmed the importance of individualised sen­ tencing and approved the sentencing judge’s decision to do equity in the particular cir­ cumstances of the case. Ferguson and McDiarmid criticise the decision as failing to reflect the gravity of the harm caused (Ferguson and McDiarmid, 2014: 329). The decision could, however, be seen as an example of particularised justice. The judge in Currie imposed a sentence that, whilst unusual for such serious offences, was considered by her to be the most appropriate method of addressing the circumstances of the specific offences and the attributes of the specific offender. To refrain from imposing a custodial sentence for what the Appeal Court recognised was a very serious offence, required the practice of equity.59 One respondent in the present study (an extremely experienced sheriff) held particularly strong views on the prosecution of young men for sexual offences such as contraventions of section 5(3) of the Criminal Law (Consolidation) (Scotland) Act 1995 (intercourse with a girl aged between 13 and 15) committed where the complainer consented—or in circum­ stances indicative of consent—and in which the complainer is of the same, or of similar, age to the accused:60 What Holyrood [the Scottish Parliament] really ought to be getting itself on to is ­charging girls art and part in these cases [i.e. charging them in concert]. It is absolutely disgraceful the extent to which young men are being criminalised in a very nasty way (and I mean nasty in terms of conse­ quences) for behaviour which is not so far out of the ordinary. A 16 year old boy and a 15 year old, or a 14 year old, girl. And it comes out as a matter of playground gossip and then a teacher hears about it or a parent hears about it and all hell’s let loose. And it’s the boy who’s prosecuted and the girl isn’t. Some of our agents have got wise to it. They go on to Facebook now and download

55  Currie (n 51) at [13]; see also Hunter v HMA High Court of Justiciary on appeal, 22 October 2008, unre­ ported (another case of attempted rape) where Lord Osborne stated: ‘We can only agree that the offence of which the appellant was convicted was a serious offence, but it is not an offence which we consider, in all circumstances, must inevitably be visited with a custodial imposition’ (at [4]). 56  Baillie v HMA 2007 JC 161 at [20]. 57  Ibid. 58  Currie (n 51) at [13]. 59  See also Paton v HMA 2002 SCCR 57 and Dalgarno v PF, Elgin High Court of Justiciary on appeal, 22 May 2009, unreported (sentence of three months’ imprisonment following conviction for the s­tatutory offence of engaging in indecent behaviour towards a girl aged between 12 and 16 quashed in circumstances where the 24 year old appellant engaged in consensual sexual activity with a 15 year old girl, and an admonition substituted). 60  An offence at the time of the interview; see now section 37 of the Sexual Offences (Scotland) Act 2009.

78 

Equity and the Rule of Law in Sentencing

Facebook photographs of these 15 year olds, who would quite easily pass for 24. That is an area where … you’ve got to be very sure of the background circumstances but I have very little sympathy for these prosecutions because of the totally disproportionate effect on the boy for the rest of his life … All because some wee lassie didn’t like being groped by the bicycle sheds. Well, what was she doing behind the bicycle sheds in the first place? They are either very Victorian, or they didn’t have very exciting childhoods, our politicians. And I’m afraid that applies to the Lord Advocate as well because that’s where prosecution policy comes from. I think it’s disgraceful, I really do … I have for a long time been very, very concerned about this propensity for prosecuting a boy for what is really nothing more than adolescent behaviour (Sheriff 5).

Following conviction in such circumstances, the practice of equity would also result in the imposition of a sentence that may appear ex facie to be extremely lenient. In Thomas and Waddell v HMA,61 for example, the Appeal Court quashed the sentences of 12 and nine months’ detention imposed on the 16 and 15 year old appellants following their pleas of guilty to a statutory charge of having u ­ nlawful sexual intercourse with a 14 year old girl and substituted community service orders of 200 and 180 hours respectively. In so doing, the Court referred to the relative ages of the parties, the appellants’ stable family backgrounds, the ­favourable social enquiry reports, and their status as first offenders.62 The same approach appears to be taken in England and Wales by the Court of Appeal (Criminal Division), particularly in cases where the complainant ostensibly consented and/or where the offender reasonably believed her to be 16 or over (see Rook and Ward, 2016: 273–81, 2014: 109–13 and 2010: 168–72). Although a Definitive Guideline on sexual offences had been issued by the Sentencing Guidelines Council (SGC, 2007),63 the Court in R v Larcombe noted that the sentencing of sexual offences demanded flexibility.64 Refer­ ence was made in Larcombe to the Definitive Guideline on sexual offences, which stressed that the starting points and ranges within the Definitive Guideline were not rigid and that movement between ranges would depend on the circumstances of individual cases, par­ ticularly aggravating and mitigating features.65 In R v Corran (a decision which pre-dated the Definitive Guideline), the Court gave what it described as ‘preliminary, non-­prescriptive guidance’ on sentencing what were then new sexual offences against children under the Sexual Offences Act 2003, in particular for rape of a child under 13.66 The principles set out by the Court67 were then developed in the Definitive Guideline itself (Selfe, 2011: 6). The 2007 Definitive Guideline provided: [T]here will be cases involving victims under 13 years of age where there was, in fact, consent where, in law, it cannot be given. In such circumstances, presence of consent may be material in relation to sentence, particularly in relation to a young offender where there is close proximity in age between the victim and offender or where the mental capacity or maturity of the offender is impaired. Where there was reasonable belief on the part of a young offender that the victim was 16, this can

61 

1997 SCCR 77. Ibid at 82. 63  English sentencing guidelines are discussed in Chapter 6. On 12 December 2013 the Sentencing Council published its new Sexual Offences Definitive Guideline (Sentencing Council, 2013) which replaced the existing sexual offences guideline on 1 April 2014. 64  R v Larcombe [2008] EWCA Crim 2310 at [11]. 65  SGC, 2007: 5; see Larcombe, ibid at [11]. 66  R v Corran [2005] 2 Cr App R (S) 73 at [2]. 67  Ibid at [6]–[10]. 62 

In Defence of the Instinctive Synthesis

 79

be taken into consideration as a mitigating factor (SGC, 2007: 21, emphasis in original; see now Sentencing Council, 2013: 28).

In such cases the court is required to undertake a significant weighing of interests (Selfe, ibid). The Definitive Guideline did not amplify the point about consent and it was left to the sentencer to decide how significant the presence of actual consent should be in any given case (Thomas, 2008: 159). This decision could require the practice of equity and, as in Scotland, resulted in the imposition of sentences which would appear to an observer unfamiliar with the particular facts to be lenient, perhaps unduly so. This is well illustrated by several recent decisions of the Court of Appeal involving convictions for the rape of a child under 13. In R v Charles, the Court quashed custodial sentences of two years imposed on six young men of previous good character and imposed suspended sentences of 12 months with 12 months’ supervision. The facts were said to be ‘wholly different from those envisaged in the guidelines’.68 The two complainants had been willing participants (one of whom consented to intercourse with five of the six appellants) and the appellants had believed that they were both 16 years old. In allowing the appeals, the Court discussed this particular sentencing problem in terms redolent of the practice of equity: [T]his Court, as the judge was, is faced with what appears to be a tension between the desire of Par­ liament to see that young girls are protected by providing an absolute ban on sexual activity with a girl under 13, whatever the belief of the perpetrator, and at the same time doing justice to these young men, who believed and were accepted as believing that they were committing no criminal offence at all.69

In R v B,70 the appellant and the complainant had corresponded via Facebook when he was 18 and she was 12. The complainant claimed to be 16; she consented to, and indeed initi­ ated, sexual intercourse with the appellant. The suspended sentence of six months’ impris­ onment with a 12 month supervision requirement was quashed and a conditional discharge substituted. In R v Cleverley71 and R v Mooney,72 meanwhile, short custodial sentences were imposed for the offence.

Leniency and Mercy in Sentencing Exercising Leniency in Sentencing It is accepted by the courts that justice can, given the right circumstances, be best served by lenience in sentencing. Lord Woolf, for example, states that both justice and the public interest can in some cases be best served in this way (Woolf, 2007). An experienced sen­ tencer should be in a position to identify the cases which fall into this category and sentence accordingly (ibid). The Canadian judge Justice Renee Pomerance explains: Many of us have seen cases in which an offender ‘on paper’ is deserving of a jail term, but ‘in per­ son’ is deserving of another chance. Testimony and character references may attest to the offender’s 68 

R v Charles [2012] 1 Cr App R (S) 74 at [26]. Ibid at [33]. 70  [2011] EWCA Crim 165. 71  [2011] 1 Cr App R (S) 84. 72  [2010] 2 Cr App R (S) 97. 69 

80 

Equity and the Rule of Law in Sentencing

rehabilitative progress. Perhaps the offender has finally overcome a ­substance abuse problem that led in the past to his other criminal activities. ­Whatever the circumstances, cases will arise in which it will be appropriate for the court to ­recognize hope and offer leniency so that it might flourish (Pomerance, 2013: 314).

In delivering the judgment of the Court in Attorney-General’s Reference (No 8 of 2007), Lord Phillips CJ noted that: ‘[l]eniency where the facts justify it is to be commended, not condemned’.73 In Attorney-General’s Reference (Nos 32, 33 and 34 of 2007), Hughes LJ described ‘conscious leniency’ as being a legitimate function of a judge in a proper case.74 In the leading Scottish decision of HMA v Bell, in which the Appeal Court set out the defi­ nition of ‘undue leniency’ in the context of Crown appeals against sentence, it was noted that there may be cases where, in the particular circumstances, a lenient sentence is entirely appropriate.75 The studies by Hough et al (2003) and Tombs (2004) emphasise that sentencers argue that they only use custody as a last resort (Hough et al, ibid 35; Tombs, ibid 45). The Austral­ ian study by Lovegrove, meanwhile, demonstrates that when fully informed of the circum­ stances of a particular case and of a particular, contextualised offender, the public are well able to appreciate the breadth, complexity and importance of personal mitigation (Love­ grove, 2011: 199, 201).

Mercy in Sentencing The role of mercy in the sentencing process is more problematic. Fox, for example, poses the following question: if the law of sentencing allows for all mitigating factors to be taken into account, why should a sentencer, having considered all such matters, deliberately order less than what is called for by the criminal law (Fox, 1999: 2)? One possible answer is that in order to humanise the world and to make it possible for people to thrive, the law needs the company of virtues such as mercy (Hussain and Sarat, 2007: 1). In Attorney-General’s Reference (No 4 of 1989), for example, the Lord Chief Justice (Lane) made the following observation: [I]t must always be remembered that sentencing is an art rather than a science; that the trial judge is particularly well placed to assess the weight to be given to various competing considerations; and that leniency is not in itself a vice. That mercy should season justice is a proposition as soundly based in law as it is in literature.76

This dicta has been referred to many times by the Court of Appeal in subsequent appeals against sentence, often in the context of Crown appeals against sentence on the grounds of undue leniency.77 In Attorney-General’s Reference (No 16 of 2005), the statement was said to be ‘as valid today as when it was first enunciated 16 years ago’.78

73 

Attorney-General’s Reference (No 8 of 2007) [2008] 1 Cr App R (S) 1 at [16]. Attorney-General’s Reference (Nos 32, 33 and 34 of 2007) [2008] 1 Cr App R (S) 35 at [21]. 75  HMA v Bell 1995 SCCR 244 at 250; see also HMA v Gordon 1996 SCCR 274 at 277; HMA v MacPherson 1996 SCCR 802 at 806–07; and HMA v Smith 2014 SCCR 39 at [17]–[23]. 76  Attorney-General’s Reference (No 4 of 1989) (1989) 11 Cr App R (S) 517 at 521; see also Judge, 2015: 268. 77  See, for example, Attorney-General’s Reference (No 73 of 2006) [2006] EWCA Crim 2508 at [20] and AttorneyGeneral’s Reference (No 153 of 2004 and Nos 14 and 2 of 2005) [2005] EWCA Crim 954 at [2]. 78  Attorney-General’s Reference (No 16 of 2005) [2006] 1 Cr App R (S) 28 at [18]. 74 

In Defence of the Instinctive Synthesis

 81

Similar sentiments have been expressed by the Appeal Court in Scotland. For example, in HMA v McKay (in which the respondent had pleaded guilty to being in possession of drugs in prison with intent to supply, another prisoner having threatened him with vio­ lence unless he moved the drugs from one part of the prison to another), the Court stated that it was in complete agreement with the sentencing judge’s view that the case was one in which ‘justice should be tempered with mercy and humanity’.79 In delivering the opinion of the Court, the Lord Justice General (Hope) said this: We wish to make it clear to the Lord Advocate, and to impress upon anyone else who may seek to press us in that direction, that this court will continue, so long as the power to do so rests with us, to assert the right of the judges to exercise leniency in the matter of sentencing wherever this is appro­ priate. We shall also continue to assert their right to take account of exceptional circumstances in deciding not to impose immediate custodial sentences in cases where a custodial sentence would otherwise be inevitable.80

The Australian courts have also explicitly recognised the role of mercy in the exercise of sentencing discretion (Edney and Bagaric, 2007: 311–312; Freiberg, 2014: 460–62). In delivering his judgment in Cobiac v Liddy, Windeyer J noted the importance of mercy and explained how the concept operates: The whole history of criminal justice has shewn that severity of punishment begets the need of a capacity for mercy … This is not because mercy, in Portia’s sense, should season justice. It is that a capacity in special circumstances to avoid the rigidity of inexorable law is of the very essence of justice.81

On this view, mercy is not so much the antithesis of justice as a component of it (O’Malley, 2006: 107). Whether or not mercy arises for consideration in a particular case will depend on all the circumstances both of the offence and of the offender (Edney and Bagaric, ibid 312; Fox, ibid 18–19; R v Miceli82). The factors, or combination of factors, that will lead a court to consider the application of the sentencing principle of mercy are potentially innu­ merable, although generally such factors would have to be significant enough to counter the strong claims for punishment that ordinarily make a sentence of imprisonment appropriate (Edney and Bagaric, ibid; see also Pomerance, 2013: 314–15). For the Court of Appeal in England and Wales at least, any reduction in sentence will be justified as an exceptional act of mercy when it accepts that the punishment will be more onerous because of the offender’s personal circumstances (Piper, 2007: 142). There is, how­ ever, no clearly articulated penal justification for allowing such mitigation to have effect (ibid; see also Bagaric and Alexander, 2013: 159). Walker considers that the appellate courts tend to resort to what they describe as ‘mercy’ when they have a vague compassion for the offender but cannot articulate a precise justification for reducing the severity of the sentence (Walker, 1999: 220). In a recent review of cases in which the Court of Appeal has imposed sentences at levels below those suggested by applicable guidelines on the grounds of mercy, Ashworth concludes that recourse to mercy is made either where there is an unusually

79 

HMA v McKay 1996 JC 110 at 113. Ibid at 115. 81  Cobiac v Liddy [1969] HCA 26 at [3]; see also R v Osenkowski (1982) 30 SASR 212 at 212–13. 82  [1997] VSC 22. 80 

82 

Equity and the Rule of Law in Sentencing

powerful accumulation of mitigating factors83 or where there is an exceptionally strong showing of one or two mitigating factors84 (Ashworth, 2012: 88). In this way, mercy can be said to operate as a residual ‘safety valve’ in sentencing (Fox, ibid 4).

A Place for Mercy in Sentencing? For some theorists, such as Harrison (1992) and Murphy (1988), mercy simply has no place in the criminal justice system. Mercy, on this view, is either part of justice or is opposed to it (Murphy, ibid 166–68; Steiker, 2007: 22). What the courts have articulated as ‘mercy’ is actually the same as doing justice and so there is no independent need for the concept of mercy (Steiker, ibid). As Murphy puts it: Judges in criminal cases are obligated to do justice … There thus simply is no room for mercy as an autonomous virtue with which their justice should be tempered. Let them keep their sentimentality to themselves for use in their private lives with their families and pets (Murphy, ibid 174).

Steiker explains that under this sceptical view, justice embraces a piece (perhaps a very large piece) of what is commonly called ‘mercy’. Examples sometimes offered of appropriate exercises of mercy (including the imposition of reduced sentences) are really the pursuit of moral justice within the retributivist tradition, which demands that offenders be punished according to their moral desert (Steiker, ibid). The sceptical view has been challenged by theorists such as Brett (1992), Brien (1998), Muller (1993), and Steiker, ibid. Steiker, for example, asks whether it really can be the case that justice (especially justice in punishment) never needs tempering by mercy? A resonant literary tradition joins common intuition in questioning this assertion (Steiker, ibid 24). For Muller, Murphy’s retributivist account reaches the wrong conclusion about mercy as it works from an unrealistic and incorrect impression of the sentencing task in real cases (Muller, ibid 290). Muller was an assistant US attorney before entering academia and is thus eminently familiar with the handling of such cases. He elaborates: Is Murphy describing something real? … Murphy has chosen to build his theory of mercy on an ‘either/or’ sentencing scenario: either the judge imposes a harsher sentence (the ‘just’ one), or he imposes a more lenient sentence. This is simply not an accurate model of the overwhelming major­ ity of sentencing decisions that sentencers are called upon to make … [V]irtually no sentencing decisions are ‘either/or’ decisions at all. Rather, in most cases, the sentencer’s task is to choose a sen­ tence from within a continuum of authorized punishments (ibid 302; see also Brett, 1992: 93–94).

Muller points out that it is unrealistic to build an account of mercy on a model of sentenc­ ing that imagines judges simply to be selecting one of two possible sentences (ibid 303). It is not the sentencer’s task to make a stark choice between a ‘just’ sentence and a milder one; the task involves the making of a subtle selection that seems most just from amongst a group of just possibilities (ibid; Fox, 1999: 11). Given the increasing rates of imprison­ ment (see Tombs, 2005: 21–27, 2004: 17–41), Steiker argues that the ideal of mercy can, in appropriate cases, serve as a necessary counterbalance (Steiker, ibid 31). By exercising

83 See R v Hussain [2010] 2 Cr App R (S) 60; Attorney-General’s Reference (No 95 of 2009) [2010] 2 Cr App R (S) 83 and Mooney (n 72). 84 See R v Schumann [2007] 2 Cr App R (S) 73; Attorney-General’s Reference (No 11 of 2007) [2008] 1 Cr App R (S) 6; R v Clarke [2010] 1 Cr App R (S) 26; and R v Foster [2010] 1 Cr App R (S) 36.

In Defence of the Instinctive Synthesis

 83

mercy, sentencers can give voice to humanitarian considerations which may still be lacking within the criminal justice system (Fox, ibid 23; Tasioulas, 2006: 312, cited in Maslen and Roberts, 2013: 125; cf Maslen and Roberts, ibid 135).

The Application of Mercy Walker suggests five criteria for the application of mercy in sentencing: (i) the main rea­ son for its application must be compassion for someone (thus distinguishing mercy from mitigation); (ii) it must not be whimsical or random, but consistent; (iii) it must not be improper in other ways; (iv) it should not be leniency of a kind required by justice; and (v) it should not be motivated solely by expediency (which would exclude, for example, leniency in return for pleading guilty) (Walker, 1999: 226). Hampton, meanwhile, defines mercy in the following terms: ‘[M]ercy is the suspension or mitigation of a punishment that would otherwise be deserved as retribution, and which is granted out of pity and compas­ sion for the wrongdoer’ (Hampton, 1988: 158). As Bottoms notes, such a definition does not, in and of itself, establish any principle of being free to treat like cases in an unlike manner (Bottoms, 1998: 68). For Bottoms, this framework allows for what he describes as a rational exercise of mercy (ibid). Ashworth considers it best to treat mercy as referring to exceptional situations, so that courts exercise mercy in sentencing only where there is a particularly u ­ nusual and com­ pelling factor which is not adequately provided for by normal principles of mitigation (Ashworth, 2015: 201–02, 2011: 35–36 and 2010a: 190–91). Ashworth notes how the Court of Appeal has in the past used the ­concept of mercy in an attempt to avoid setting a prec­ edent (Ashworth, 2011: 36). In R v Bernard, for example, the Court stated that where a prisoner is suffering from a terminal illness or other serious medical condition the sen­ tence may be reduced ‘as an act of mercy in the exceptional circumstances of a particular case rather than by virtue of any general principle’.85 This dicta was clearly intended to negate imposing a duty of mercy upon sentencers (von Hirsch and Ashworth, 2005: 169). Von Hirsch and Ashworth consider this aspect of the decision in Bernard to be ‘plainly objectionable’ as it leaves mitigation of sentence as a discretionary matter, rather than seek­ ing to resolve the issue at the level of principle (ibid; see also Ashworth, 2012: 89– 90). More recently, Ashworth states: ‘[T]he importance of being able to mitigate for “mercy” should not be obscured by suggesting that it can operate outside the realms of principle. The court’s stated justification for sentence is particularly important in this type of case’ (Ashworth, 2011: 36; see also Ashworth, 2015: 195, 201–02 and 2010a: 187 and 191; Easton and Piper, 2016: 230–31; Piper, 2007: 154–55; and Warner, 2011: 138). Part of the difficulty with decisions such as Bernard is the courts’ use of the term ‘mercy’, which tends to connote a highly exceptional, one-off exercise of a special dispensing power (von Hirsch and Ashworth, ibid). As Fox notes, however, the continuing opportunities for mercy in sentencing carries an expectation that those who are appointed to judge others are selected not only for their legal knowledge, but also for their possession of traits which allow them to express benevolence and concern for others in acts of mercy in appropriate cases (Fox, 1999: 27). A merciful sentencer is one who has both the capacity and the willingness

85 

R v Bernard [1997] 1 Cr App R (S) 135 at 139; see also R v Hall [2013] 2 Cr App R (S) 68.

84 

Equity and the Rule of Law in Sentencing

to consciously yet impartially consider the impact of the available disposals from the per­ spective of the offender (ibid). In R v Kane, the Supreme Court of Victoria explained: [M]ercy must be exercised upon considerations which are supported by the evidence and which make an appeal not only to sympathy but also to well-balanced judgment. If a court permits sym­ pathy to preclude it attaching due weight to the other recognized elements of punishment, it has failed to discharge its duty.86

Thus, while questions of principle remain, the sentencer as a phronimos—a person of prac­ tical wisdom—is arguably best placed to determine when mercy should be extended to an offender through the practice of equity.

Judicial Views on Leniency and Mercy—the Findings of the Present Study Consistent with the importance of achieving individualised justice in sentencing, every one of the 25 respondents in the present study agreed that justice and the public interest can, in appropriate cases, be best served by lenience in s­ entencing. It was agreed by all respondents that lenience would be appropriate in cases where, for example, an individual had acted out of character; where the risk assessment undertaken as part of the social enquiry report indicated that the individual presented a low risk of re-conviction; or where the offender exhibited genuine remorse: In certain circumstances, having heard a case, you maybe think it’s deserving of custody, but on the other hand, you have somebody who has acted entirely out of character, who has an unblemished record and who is clearly genuinely remorseful for what has happened. You get a feeling that this person is unlikely to ever be back in court again. You begin to wonder whether the family impact might have a greater sway; the fact they’ve got a conviction for this might considerably affect their future prospects. It may not be necessary to impose custody (Sheriff 14). You should never be afraid to be lenient where leniency is appropriate. It involves a good degree of courage on the part of the judge. It’s much easier to be severe. You have to look at all the circum­ stances and ask what is the correct sentence (Judge 4).

Another judge illustrated the concept by reference to the crime of culpable homicide: Culpable homicide, just as an example, is an offence which is so varied that the level of culpability can be close to the level of murder but it can also be at a very, very low level. Whilst you do have to reflect the tragic fact that someone has died, we should not be sentencing based on the conse­ quences. We should be sentencing based on the culpability of the offender and the offence. In those kinds of situations I think leniency can really serve the interests of justice (Judge 2).

Whilst recognising that cases of causing death by dangerous driving, contrary to section 1 of the Road Traffic Act 1988, are no longer prosecuted in the sheriff court,87 one sheriff discussed the appropriateness of leniency by reference to conviction for such offences: I think what it means is that as a human being, as well as a sentencer, you recognise there may be cases where perhaps leniency is justified because of, for example, the desperate circumstances of the offender. You do get these terrible cases where as a result of someone’s dangerous driving a close relative or loved one is killed. I’ve never, for the life of me, really understood why it ought to be

86  87 

R v Kane [1974] VR 759 at 766. It has been the Crown’s practice in recent years to indict all such cases in the High Court.

In Defence of the Instinctive Synthesis

 85

regarded as almost inevitable that somebody should go to prison for a minimum of three or four years in cases like that. I’ve often thought that the punishment of the loss of your relative is enough. I just can’t get my head round this idea that just as a matter of pure public policy, a significant sen­ tence of imprisonment should be imposed (Sheriff 17).88

Another sheriff stressed the importance of leniency in the sentencing of summary offences: My whole approach to sentencing is leniency at summary level. I think I have a reputation of being quite a heavy sentencer at solemn level because I think serious crime needs to be treated seriously and that’s what the public needs protection from. But leniency is the watchword for my summary sentencing process (Sheriff 3).

Thirteen sheriffs made specific reference to lenient disposals often being appropriate in cases where the offender suffers from mental health problems. Such cases were seen as ‘cries for help’ (Sheriff 7) by ‘poor souls’ (Sheriffs 5, 9 and 16) or by ‘poor, inadequate souls’ (Sheriff 8): Many of the people who come before the courts are here because of misfortune, mental illness, a combination of them, and I don’t see quite a lot of them as needing to be punished, as distinct from turned from their criminal ways (Sheriff 4). People are still charged with breach of the peace by attempting to commit suicide by jumping off a bridge … Well, I mean, what’s the point in imposing a penalty for that? You might get a social enquiry report just to put them in touch with social services. But there’s really no point in imposing a penalty for that sort of thing (Sheriff 5).

Five sheriffs considered that many such offenders—‘the legions of the inept, the handi­ capped, the resentful, and the poor’ (Ruby, 1999: viii)—should not be dealt with in the criminal justice system at all. One respondent (Sheriff 1) considered that the criminal justice system was ‘filling a vacuum left by social services’. This sheriff noted that much offending behaviour is a symptom of mental health problems or alcohol or drug dependency: You have to be realistic about what the criminal justice system can achieve. As soon as it’s pointed out to you that there’s this vacuum that the criminal justice system has to fill, you begin to identify the cases where that applies … They’re not people who society thinks should be punished. There’s sometimes just an inability to just fit in. They then fall foul of the criminal law in circumstances where the notion of punishment really just doesn’t fit (Sheriff 1; on this point see also Lacey, 2002: 30 and Bingham, 2000a: 308).

Another respondent (Sheriff 9) considered that the authorities are often too quick to pros­ ecute those with mental health difficulties, especially for public order offences: The answer is not to prosecute them. Yet we do. Society’s put these folks in the community. Many of them are poor souls. You don’t punish the weak. You don’t punish the sad, the mad. It’s a terrible way of speaking these days but that was the way I was brought up. You don’t punish the sad or the mad, only the bad. As a principle, it’s not a bad one. The word ‘mad’ may be inappropriate but the idea of only punishing the bad still remains (Sheriff 9). 88  The Appeal Court has approved judicial recourse both to the English Court of Appeal’s sentencing guidelines in cases involving death by dangerous driving (R v Cooksley [2004] 1 Cr App R (S) 1 and R v Richardson [2007] 2 Cr App R (S) 36) and the Definitive Guideline issued by the Sentencing Guidelines Council for causing death by driving (SGC, 2008). See, respectively: HMA v MacPherson 2004 SCCR 579; HMA v Roulston 2006 JC 1; Wright v HMA 2007 JC 119; and HMA v Noche 2012 SCL 329; Grant v HMA 2013 SCCR 201; Fleming v HMA 2013 SCL 386; HMA v McCourt 2014 JC 94; Geddes v HMA 2015 SCCR 230; and HMA v McKeever 2016 SCL 564.

86 

Equity and the Rule of Law in Sentencing

The respondent Sheriff 13 noted that many offenders have previously been subjected to physical or sexual abuse; they will often either be homeless or live in dreadful conditions, and will often be dependent on drugs or alcohol. The respondent considered that, given many offenders’ dire personal circumstances, sentences will often be imposed which appear lenient or merciful to those who are unfamiliar with the full circumstances: We deal with a lot of people here who are very broken. It’s not for the judiciary to comment on policy but the reality is that we deal with so many more damaged people than we ever used to. We used to deal with these people in a different way. I’m baffled as to why we think we’re a humane society. We tend to prosecute the disadvantaged rather than deal with them in other ways. Accord­ ingly, I think the court is probably in a position where more often than in the past it has to consider whether a sentence that might be seen as being lenient or merciful is to be imposed (Sheriff 13).

The respondents Sheriff 4 and Sheriff 5 discussed two examples of offenders who had learn­ ing difficulties; they both questioned the appropriateness of prosecuting such individuals and noted the difficulties they faced at the sentencing stage: Antisocial behaviour orders. What a waste of time those are! He, [name of offender], is a simple soul in the old fashioned, classical, simple sense. Or, he is ‘differently learning abled’. He can’t really manage by himself, so every now and again because he’s in his own tenancy, he just loses the plot and gets drunk. Is he really a criminal? Hmmm. So he’s got two charges and we’re spending a lot of money at the moment having a psychological assessment done on him, the psychiatrists having decided there was nothing they could do because he has a learning disability. When he’s sober he’s a nice soul (Sheriff 5). The unhappiest examples you find are people who are just above the learning disabled level who cannot get any of the routine support as those who are seen as learning disabled get, but who need it … I can think of someone—he was just above the level at which he would get fairly constant social work support—who had picked up a conviction that led to his being placed on the sexual offenders register. He kept coming back charged with breaching the notification requirements. He did things like getting on the wrong bus … so he couldn’t find a police station to sign on and notify, so he ended up breaching the notification requirements. I had a pattern of admonishing him because I saw him as struggling to cope … I just felt he could not cope. So, that’s the sort of offender where I’m not really looking to punish, and clearly punishment did not a jot of good anyway, save to cost society a great deal of money (Sheriff 4).

One judge, however, whilst agreeing that lenience can be appropriate in certain cases, cau­ tioned that leniency ‘is usually a synonym for softness’ (Judge 3). The respondent was criti­ cal of certain judges and sheriffs for being too lenient in their sentencing; he considered that they habitually imposed lenient sentences to avoid appeals being taken against their deci­ sions. The respondent noted that, despite their appalling records of previous convictions, some offenders would never receive a custodial sentence until they appeared in his court: These people come up in front of me and I say ‘Well, you’re going to jail’. What do they say? [The respondent at this point affects a high-pitched, whining tone] ‘Ohhh, but you can’t send me to jail, I’ve never had a prison sentence before; I’m doing a drugs testing thing’ and you say ‘Well, too bad. You can’t come along and say “I only did it because I was high on drugs” and expect that your choice to be on drugs should entitle you to some special favour’. It’s the ordinary people who need our protection. And in a lot of cases we’re not giving it. That’s my feeling (Judge 3).

Twenty two respondents considered mercy to be qualitatively different from leniency, whilst three considered that the concepts ‘blended into each other’ (Judge 6), that they were ‘part

In Defence of the Instinctive Synthesis

 87

of the same spectrum’ (Sheriff 1) and that they were ‘difficult to separate’ (Sheriff 16). Con­ sistent with Walker’s observation of how mercy is applied by the English Court of Appeal (1999: 220), many respondents acknowledged that they had difficulty in articulating a pre­ cise justification for reducing the severity of a sentence on grounds of mercy. There was general agreement, however, that mercy would be a relevant consideration where ‘there are compelling circumstances over and above normal mitigating circumstances’ (Sheriff 17); that the decision whether to apply it is ‘an entirely subjective matter’ (Sheriff 3); and that mercy ‘goes beyond mitigation’ (Sheriff 15). As the last named respondent elaborated: Mitigation gets you to your headline sentence, taking everything into account. Mercy would then come in when, having considered everything, you think, well, that is the sentence that is appropri­ ate for that person in all the circumstances of this case. But there would have to be some other supervening reason, over and above that, for mercy. But there must be room for it. It would be an impoverished system if there wasn’t (Sheriff 15).

As one very experienced judge put it: Mercy is an important part of the judicial armoury. All sentencers should be as lenient as the particular offence allows. The imposing of exemplary sentences belongs to a much cruder age of penology. The judge or the sheriff should impose the lowest sentence consistent with the circum­ stances of the offence and of the offender (Judge 5).

These sentiments were echoed by two experienced sheriffs: ‘I think it’s entirely legitimate in a human justice system to include considerations of compassion, humanity and mercy’ (Sheriff 11) and ‘Mercy has a role to play. It must have a role to play in any civilised society’ (Sheriff 3; see also Carloway, 2013: 15 and Hooper, 2012: 23). One sheriff considered that in summary cases (where the sheriff is both the trier of fact and the sentencer) mercy can, in certain circumstances, dictate that an accused be found not guilty where a conviction would otherwise be warranted. In discussing what Brien (1998: 97) describes as ‘mercy verdicts’, the respondent referred to the decision in Barile v Griffiths, a case where a schoolteacher was convicted on summary complaint of assaulting two pupils in his class. Although Mr Barile’s appeal against conviction was refused, the Appeal Court criticised the Crown for prosecuting the appellant, given that he had acted under ‘extreme provocation’ in the face of ‘disgraceful’ behaviour by the pupils.89 The respondent’s view of the case was this: That would be an example where one would perhaps have just not convicted him by just the simple cop out of not believing the school boys because the school teacher was obviously at the end of his tether with these bastards! … I think one’s way out of that situation might just be to disbelieve the boys. That would be the best solution (Sheriff 2).

The appellant Barile was admonished following his conviction, although press reports indi­ cated that he was subsequently dismissed from his employment (AN Brown, 2010b: 3). Mr Barile’s subsequent appeal against sentence was granted and an absolute discharge was imposed.90 This, I suggest, is an excellent example of the practice of equity in sentencing. As another respondent said of the case: He lost so much that the absolute discharge that was ultimately imposed seemed the right decision, despite the fact that on paper a teacher assaulting a pupil is terrible. But when you look at the context 89  90 

Barile v Griffiths 2010 SCCR 97 at [15]. Barile v PF, Dundee High Court of Justiciary on appeal, 11 March 2010, unreported.

88 

Equity and the Rule of Law in Sentencing

of it, it’s a situation where mercy would call for a disposal which wouldn’t necessarily fit with what the sentencing books say (Sheriff 1).

The views of 21 respondents in the present study accorded with the position taken by theo­ rists such as Brett, Muller, Brien, and Steiker: mercy is an essential component of the law. Only four respondents (two judges and two sheriffs) did not consider that it was their place to show mercy to an offender. As one of the judges explained: I don’t really think it’s for me to show mercy. I think it’s for the individual families [of the victim(s)] to indicate. The fact that they have indicated a merciful response may or may not have its effect in leniency, but it’s not for the court to do it (Judge 2).

This respondent viewed mercy as necessarily individual and the province only of discrete victims of crime (see Steiker, 2007: 20). One of the two sheriffs explained his misgivings about mercy in sentencing at the sheriff court level: Mercy, I think, is for the High Court. I don’t think that we, at our level, have to operate that way … I think the High Court judges in the cases they have and the sentences they have to impose have more scope for being merciful. I’m not sure that I would ever pretend to be so grand as to be merciful in the sheriff court … But there is a place for it, however you define it … I’m not sure the sheriffs will have a need to exercise mercy, but leniency is more common (Sheriff 14).

Equity and the Imposition of Condign Punishment Although discussion has thus far focused on the practice of equity as mitigating sentences, it must be appreciated that equity as the doing of particularised justice in the case of a contextualised individual can equally result in the imposition of more severe sentences (see DPP v Drought).91 As Nussbaum explains by reference to Plato’s Laws (Plato, 1988): [T]he equitable [judgment] will sometimes be more lenient than the generality of the law, but sometimes harsher. For, as that not-very-merciful philosopher Plato puts it in the Laws, sometimes the offender turns out to be unusually good for an offender of that sort, but sometimes, too, unusu­ ally bad (Nussbaum, 1993a: 87, citing Plato, Laws, 86-7d.).

There was a degree of consensus amongst respondents on factors that serve to aggravate offences. In addition to the statutory aggravations (such as offending whilst on bail, racial aggravation, or aggravation by religious prejudice) respondents referred to the following factors: previous convictions (particularly analogous previous convictions—10 respond­ ents); breach of a position of trust (six respondents); planning or premeditation in com­ mission of the offence (six respondents); targeting vulnerable complainers, particularly the young and the elderly (five respondents); unprovoked and/or gratuitous violence (five respondents); the repeated nature of an offence, either in the sense of the length of time over which the offence was committed (ie the period libelled in the charge) in, for example, drugs offences or sexual offences or, alternatively, in the sense of repetition of the criminal conduct on a single occasion (four respondents); voluntary intoxication (three respond­ ents); and lack of remorse (three respondents).

91 

[2007] IEHC 310 at [12].

In Defence of the Instinctive Synthesis

 89

Although regarded by respondents as something of a truism, the aggravating effect of previous convictions would appear to be viewed by the 10 respondents who cited it in terms of ‘the recidivist sentencing premium’. This theory holds that as an offender accumu­ lates more convictions, the sentence imposed at ­subsequent sentencing hearings becomes progressively more severe (Roberts, 2008: 3).92 It may be, however, that the type of offence or the facts of a particular case will result in a custodial sentence even for an offender of previous good character. One respondent (Sheriff 15) spoke of such offences as being ‘just beyond the pale of human behaviour’, offences that ‘have to be marked as such’ and that ‘cry out for custody, where anyone would think that they must merit a custodial sentence’. The respondent elaborated: Attacks on complete strangers. Unprovoked attacks. Assault to severe injury, maybe with a weapon. Real violence. One that sticks in my mind is biting someone’s ear off, then [the offender] getting on the bus with the ear in his hand and showing it to all the appalled passengers. The complainer was permanently disfigured. We never found out what happened to the ear. Just really depraved violence. He’d been to prison before, but even if it had been a first offender, I’d still have sent that person to prison for that kind of offence’ (Sheriff 15).

An example given by another respondent featured what he described as a ‘white c­ ollar’ case. This respondent accepted that young people ‘have always had ­something of a propensity to get into fights under the influence of alcohol’ but was of the opinion that the level of violence and the gratuitous nature of the assault in this particular case warranted a custo­ dial sentence: It was an otherwise respectable university student in company with friends on [name of street in city centre]. He was approached by some other, probably quite unpleasant, drunken youth. I can’t quite remember how the brawl first developed but in any event the complainer (who had clearly been the instigator of the interaction) ended up on his knees in front of the accused. A witness described seeing the accused kick the complainer’s head ‘like taking a drop goal at rugby’. I sentenced him to prison. I had no hesitation (Sheriff 5).

The respondent was critical of the Appeal Court’s decision to allow the offender’s subse­ quent appeal: I can’t remember what community based disposal their Lordships saw fit to impose, but I thought that was disgraceful. The Appeal Court said ‘It’ll ruin his life’. Well—duh! I mean, he could have killed him! But as I say, he was an otherwise respectable, if drunken, young man. I’ll impose a cus­ todial sentence in situations where it smacks of ‘I’m in control now. I’ve got you on the ground’ and the red mist descends and it becomes vicious (Sheriff 5).

These views were shared by another sheriff (Sheriff 17) who sat in the same court. This sheriff expressed concern at the increasing levels of drunken violence and disorder in the jurisdiction. The respondent stated that in such cases where serious injuries are inflicted,

92  Cf the ‘cumulative sentencing’ model (where previous convictions should always be used at sentencing unless they are too old or are considered to have little predictive utility—see Roberts, 2008: 8–9); the ‘flat-rate sentencing’ model (which holds that previous convictions should never count on the grounds that the criminal law should punish people only for their current actions—see Roberts, ibid 9); and the theory of ‘progressive loss of mitiga­ tion’ (where first offenders receive a discounted or mitigated sentence, but then lose this discount when they have lost their first offender status—see Freer, 2013; O’Malley, 2008b: 8; Roberts, 2010; Roberts, ibid 9–10; Roberts and Baker, 2008: 556–57; Ryberg, 2010; and von Hirsch, 2010).

90 

Equity and the Rule of Law in Sentencing

even first offenders will often be jailed. This view seems to have been shared by the Appeal Court in more recent years. It has been held that, unless there are exceptional mitigating circumstances, drunken street violence involving kicking and stamping on a victim lying prone on the ground will result in a custodial sentence.93 One respondent explained that he considered certain offences as ‘self-aggravating’: Theft by housebreaking, personal assaults on strangers in the streets—these offences themselves aggravate the sentence, if you like. Matters of that kind. Also, people who are exposing others to harm generally—drug dealing. I will normally, without a second thought, lock up a 4(3)(b) [being concerned in the supply of drugs] even on summary complaint. That’s one of my exceptions to the rule of keeping people out of prison. If it’s commercial drug dealing, then that’s a self-aggravating offence. Whatever the alternatives are I am really needing to be persuaded not to lock someone up (Sheriff 3).

Although not articulated by the respondent in these terms, this is arguably a further exam­ ple of equity in sentencing: the nature of the offence dictating that a severe sentence is required. The issue is considered by Ashworth in the context of progressive loss of mitiga­ tion. Ashworth notes that at common law, the courts have traditionally applied a form of ‘flat-rate’ sentencing to grave crimes (2015: 214).94 Ashworth cites the decision in R v Turner where, in setting out sentencing levels for armed robbery and for grave crimes in general, Lawton LJ stated that ‘the fact that a man has not much of a criminal record, if any at all, is not a powerful factor to be taken into consideration when the court is dealing with cases of this gravity’.95 The justification for this restriction, Ashworth notes, must be along the lines that little concession to human weakness should be made where there is egregious wrongdoing. Whilst there should be some concession to human frailty, allowing less serious offences to be seen as unfortunate lapses, there is, in the absence of compelling mitigation, less room for tolerance of those who succumb to the temptation to commit grave crimes. Ashworth proposes that this may suggest a kind of sliding scale, with the general ‘concession to human frailty’ approach to first offenders gradually giving way to a harder line towards egregious wrongs (ibid; see also Ashworth, 2013a: 621–22). This was illustrated by a respondent in the present study who made the following obser­ vations concerning sexual offences: Now, I obviously don’t do rape cases but we do assault with intent to ravish and to me that’s always jail. I’m sore on sex crimes. I think it shows an enormous lack of respect for women and I think that’s got to be punished very severely. I’m sore on that. There are some crimes that are just wrong … I’ve always thought that there is a line that people must not cross. I’ve never abused a child, it wouldn’t occur to me to do so. But if I found myself doing it, I would know before I did it that I was about to cross a line (Sheriff 9).

The respondent provided the following example: I sentenced a gentleman recently who had sexually abused his daughter over a period of three years. He had been abusing this girl regularly and it reached the stage of digital penetration. The big plea

93  Clark v HMA [2013] HCJAC 32 at [5]–[6]; Wishart v HMA [2013] HCJAC 116 at [15]– [17]; and Robertson v PF, Aberdeen High Court of Justiciary on appeal, 18 February 2010, unreported—see the case note by Brown, 2010a. 94  Not to be confused with the ‘flat-rate’ model discussed by Roberts, 2008: 9. 95  R v Turner (1975) 61 Cr App R 67 at 91.

In Defence of the Instinctive Synthesis

 91

in mitigation was ‘He’s very devout, he’s very ashamed’ and so on. He was in good employment. But you can’t have that. The child was distraught. The accused’s wife wasn’t having him back; she was unforgiving for what he’d done to their daughter. In my view you’ve got to protect children from this … I don’t see that there’s any alternative to a significant sentence there. That’s what you’ve just got to do (Sheriff 9).

These comments accord with certain decisions of the Appeal Court which have recognised that equity requires the imposition of custodial sentences for particular classes of offence. Insofar as sexual offences are concerned, it was held in HMA v Brough96 that custodial sen­ tences will be ‘well-nigh inevitable’ for teachers who commit sexual offences against their pupils, whilst in both HMA v Stopper97 and HMA v Fallan98 it was held that unprovoked sexual attacks committed against strangers in the street late at night are of such gravity that custodial sentences should normally be imposed. The Appeal Court has stated that custodial sentences should normally be imposed for offences involving trafficking in Class A drugs.99 Even a conviction for such an offence may, however, result in a non-custodial disposal if the mitigating factors are sufficiently compelling.100 So-called ‘social’ supply of such drugs should also normally attract a custodial sentence.101 Custodial sentences should also be imposed for those convicted of cannabis farming.102 The Appeal Court has made it clear that an offender who races his vehicle on public roads and who causes death or serious injury to innocent members of the public can expect a custo­ dial sentence, however young he may be and however good his character.103 With regard to crimes of violence, a conviction for armed robbery will result in a custo­ dial sentence in almost every case.104 Sentencers have been directed to ‘keep prominently in mind’ the need for deterrent sentences in cases involving the unlawful use of prohibited firearms, particularly against serving police officers.105 The Appeal Court has taken what Shiels describes as a ‘strict approach’ in sentencing knife crime (Shiels, 2013: v). The fact that custodial sentences will often be required for offences involving the use of knives is evident from the Appeal Court’s dicta in HMA v Boyle: [A]t the present time knife crime is a scourge in the Scottish community and … the court should be acting, and be seen to be acting, in a way which discourages the carrying of sharp weapons, the use of which may lead to needless deaths.106

96 

1996 SCCR 377 at 381. 2002 SCCR 668 at [9]. 1996 SCCR 80 at 85. 99  HMA v McPhee 1994 SCCR 830 at 836–37; HMA v Lee 1996 SCCR 205 at 212–13; and more recently Slaven v Spiers 2005 SCCR 308 at [7]; Irvine v PF, Dornoch [2005] HCJAC 53 at [6]; and B ­ unton v HMA High Court of Justiciary on appeal, 1 November 2005, unreported (cf judicial views on the operation of the statutory minimum sentence of seven years for a third Class A drug trafficking offence discussed above). 100  Mitchell (Kyle) v HMA High Court of Justiciary on appeal, 29 August 2012, unreported (see the case note by Brown, 2013c) and Smart v HMA [2016] HCJAC 73. 101  Mitchell (Keith) v PF, Aberdeen High Court of Justiciary on appeal, 27 April 2010, unreported—see the case note by Brown, 2010b. 102  Lin v HMA 2008 JC 142 at [12]–[13]. 103  HMA v Stalker 2003 SCCR 734 at [12]. 104  Reilly v HMA 1983 SCCR 311 at 313; HMA v Smith (n 75) at [17]. 105  HMA v Murray 2004 SCCR 585 at [20]. 106  HMA v Boyle, 2010 JC 66 at [16]. 97  98 

92 

Equity and the Rule of Law in Sentencing

Similarly, in McNeilage v Griffiths,107 the Court noted that it was the most common experi­ ence of those working in the criminal courts that the carrying of knives frequently led to the commission of the most appalling offences, up to and including murder. The carrying of knives was thus a practice which had to be severely discouraged.108 It was held in HMA v Jamieson109 that where serious injury has been caused in an unpro­ voked attack by the use of a weapon, it will be only in the most u ­ nusual case that the court will be able properly to refrain from imposing a custodial ­sentence on the offender. The respondent Sheriff 9 provided a particularly poignant example of the application of this approach. The case (a plea of guilty to a charge of assault to severe injury and per­ manent disfigurement) was one in which, despite the offender’s previous good character, her favourable employment record and her success in overcoming certain difficulties in her own life, the level of violence inflicted meant that, in practising equity, the respond­ ent was left with no option but to impose a significant custodial sentence. The respondent explained that the 18 year old accused (‘a poor wee soul’ and ‘a lovely, nice, pleasant wee girl’) had endured a terrible upbringing in which she had been raped by her carer. She had, however, performed well at school and was working in an office at the time of the offence. The respondent continued: But it’s the way with the young, sometimes. She was in her house and she was taking drugs and drink. Another girl came to the house and for some reason she took offence. She had a sword hang­ ing on the wall—a samurai sword. She took it off the wall and, well … This girl was just maimed. The scars! There wasn’t a plastic surgeon in the world who could fix what she did to this girl. She had to go into custody. I remember the case very well. I got the plea in mitigation and I said I’ll have to think about this, in fact I’ll have to think about this overnight actually. So I said ‘9.30 tomorrow’. And she said, ‘Oh, I can’t come, I’ve got work tomorrow’. Oh no! She had no previous involvement with the criminal justice system—only as a victim. But she had completely destroyed the face of another girl and I couldn’t see a way round a custodial sentence. I think it was about three years she had to get. That was because of the nature of the offence. It was just awful. She was very lucky to be indicted in the sheriff court. I mean, permanent disfigurement doesn’t come into it; this was a maiming! There was no way the victim here would ever present to anyone as anything other than severely disfigured. Nothing could be done. Now, that had to be marked and you can’t mark that by a non-custodial sentence. ­Sometimes the job is tough. […] I broke my heart over that wee girl. I bent over backwards to try and find some way of keeping her out [of prison] but I concluded ultimately that, had I not sent her to prison, the Crown would have appealed. It would have been hanging over her head for even longer. The Crown would have appealed because they would have to have appealed. E ­ veryone would have wrung their hands in the Appeal Court but they could not have avoided sending her to prison … Then they’d have had to hammer her. So she’d have waited longer to get this thing all past and that was the comfort I took from it ultimately.110 You do your job and it’s not always a pleasant job, and sometimes it’s a very unpleasant job, but that’s what you do (Sheriff 9). 107 

2004 GWD 7-142. also McGartland v Miller 2004 GWD 8-185; Rotchford v Carnegie 2005 GWD 19-345; Pelosi v Normand 1993 SCCR 45 at 46; and Shiels, 2013: v and 64–65. 109  Jamieson (n 39) at 840. 110  On this point, see also Robertson, 2007: 5—‘A “weakly merciful” sentence at first instance may make the judge feel good but is cruel to the defendant as the correct sentence will be imposed some months later in the Court of Appeal’. 108 See

In Defence of the Instinctive Synthesis

 93

Whilst such cases may feature a number of compelling mitigating factors, they must be set against both the serious nature of the offence and considerations of retribution and deter­ rence.111 The doing of particularised justice in such cases through the imposition of what can be lengthy custodial sentences despite a host of mitigating factors is itself an example of judicial recourse to equity. There are also circumstances, albeit rare, in which a severe sentence may be imposed because, paradoxically, the mitigating circumstances are so compelling. This was demon­ strated by Sir Igor Judge (as he then was) in an address delivered at Lincoln’s Inn. His Lord­ ship recounts an incident he experienced as a newly called barrister in the early 1960s where his client threw a brick through a shop window and waited for the police to arrive. The client had been a prisoner of war and, upon release, found that he was incapable of living in the community. He would regularly commit petty crimes in an attempt to be sent back to prison. The judge imposed the maximum sentence of five years’ imprisonment—described by Sir Igor Judge as an horrendous but merciful sentence as the judge (who was fully aware of the individual’s background) understood that he was sending the offender back to a place where he was able to live (Judge, 2007: 1). This theme was also taken up by the last-mentioned respondent (Sheriff 9) in a discus­ sion of his approach to sentencing drug addicted female offenders who are actively seeking a prison sentence to detoxify. The respondent explained that on several occasions he had sent such offenders to prison (in circumstances where he would not otherwise have con­ sidered custody) simply because the offender had herself requested a prison sentence. In such cases, the offender would typically explain that she owed drug debts, that her dealer was pressuring her to prostitute herself in order to pay off the debt, and that she wanted to ‘get clean’. It was well known amongst the criminal classes, the respondent explained, that the best place to detoxify was in prison—‘if you want to get off, get inside’. The respondent considered it ‘an act of compassion’ to send such offenders to prison: So, yes I have contributed to the prison population in that way and I am bound to tell you that I will make absolutely no apology for so doing. It’s an act of compassion. These people are before the court for sentence. If you fine them they will not pay the fine. I have always thought, for example, that fining a prostitute … is immoral. How does she pay off the fine? Well, we know how she gets the money to pay off the fine! It’s a completely wrong-headed approach in my view. Many of these ladies who are in that situation, if they’re not prostitutes then they’re drug addicts certainly and they’re skirting the fringes and are reaching the stage of desperation where they do prostitute them­ selves and that must be an awful thing for a woman to have to do (Sheriff 9).

Although this approach does not strictly accord with the dicta of the Appeal Court in Duffy v HMA112 to the effect that the criminal justice system cannot impose restrictions on a person for his or her own good or for that of others if these go beyond what is retribu­ tively justifiable, it is yet another example of a sheriff doing equity in sentencing individual, contextualised offenders.

111  HMA v Trainer High Court of Justiciary on appeal, 5 October 2010, unreported at [6]; HMA v Jamieson High Court of Justiciary on appeal, 22 June 2012, unreported at [8]. 112  2005 SCCR 697 at [7].

94 

Equity and the Rule of Law in Sentencing

The Problem of the ‘Non-Virtuous’ Sentencer That a strict adherence to the letter of the law can produce arbitrary and unjust results is well illustrated by judicial views on the operation of minimum sentences (see above) and by the case law relating to a variety of offences considered above. This, as Solum notes, takes us back to Aristotle’s discussion of the basis for equity: no general rule can be framed with sufficient precision to yield the expected outcome in all particular cases (Solum, 2008: 161, 1994: 139; see also Crowder, 2002: 58). Solum acknowledges, however, that the arguments he offers for the reconciliation of equity and the rule of law do not eliminate all of the tensions between them. A virtue-cen­ tred theory of judging will not, he states, ‘work magic’ and there will, for example, be cases in which doing equity produces ‘surprise or unfairness’ (Solum, 2008: 155, 161, 1994: 139). Furthermore, not all judges are virtuous and allowing the practice of equity will permit some judges to act arbitrarily and unjustly (Solum, 2008: 161, 1994: 140). Solum explains: ‘Lacking judicial wisdom, a judge who attempts to do equity will not perceive the relevant features of the particular case and so in departing from the rule is more likely to do harm than good’ (2008: 158). This point deserves particular consideration in the context of sentencing. What we value is the virtue of judicial wisdom itself, rather than a naïve assumption that everyone appointed to the Bench somehow becomes endowed with it (O’Malley, 2011: 32). Not all sentencing judges have the requisite experience (Krasnostein, 2015a: 116). As McClellan J (as he then was) observed in the course of his judgment in R v Whyte: The structure of the modern legal profession, which demands specialisation by practitioners in particular areas of the law, will have the effect that a person who is, without doubt, appropriate for judicial appointment, may not have any, or significant experience, in the sentencing process.113

The law reports are, for example, replete with cases in which, following appeals by the Crown, sentences have been overturned by the Appeal Court as unduly lenient, following the test in HMA v Bell114 that the sentence selected fell outside the range of sentences which the first instance judge could reasonably have considered appropriate. Some of the most notable cases from recent years include HMA v Clark,115 HMA v Trainer,116 HMA v Campbell117 and HMA v SSK.118 Clark featured a Crown appeal against the sentences of 40 months’ detention imposed on the respondents following their conviction for attempted murder; the complainer sustained catastrophic injuries and would require 24 hour care for the rest of his life. In allowing the appeal, the Court substituted sentences of five years’ detention. In Trainer the Crown appealed the sheriff ’s disposal of 18 months’ probation imposed on the respondent follow­ ing his plea of guilty to a charge of assault to severe injury. The Court noted that the assault

113  R v Whyte [2002] NSWCCA 343 at [264]; see also R v Bangard [2005] VSCA 313 at [33] and McClellan, 2012: 26. 114  Bell (n 75) at 250. 115  2010 JC 90. 116  Trainer (n 111) 117  High Court of Justiciary on appeal, 25 September 2012, unreported. 118  2016 SCCR 74.

In Defence of the Instinctive Synthesis

 95

had been ‘a vicious attack on a helpless victim lying on the ground, involving repeated punches and kicks to his head and body’;119 the complainer suffered multiple fractures to his face and was then abandoned whilst still unconscious late at night in freezing condi­ tions. The Court had regard to the serious nature of the offence, the grave injuries inflicted, and the respondent’s record of analogous offences, noting that in imposing probation the sheriff had focused on rehabilitation to the apparent exclusion of other essential sentencing objectives such as retribution and deterrence.120 The appeal was allowed and a sentence of four years and three months (discounted from a starting point of five years—the maximum sentence available in the sheriff court) was imposed.121 The respondent in Campbell, meanwhile, was sentenced to 10 months’ imprisonment following his plea of guilty to a charge of assault to severe injury and permanent disfigure­ ment by repeatedly stabbing a complete stranger in the street. The Appeal Court was ‘well persuaded’ that the sentence was unduly lenient, noting that such an offence would justify at least three years’ imprisonment (see the case note by Brown, 2012c). Finally, in HMA v SSK122 the trial judge imposed an extended sentence of seven years, comprising a custodial term of five years and an extension period of two years, on the respondent following his conviction of a series of sexual offences committed against his former partners and their children. In particular, the respondent was convicted of lewd practices and indecent assault committed against a nine year old boy (including oral pen­ etration and attempted sodomy); lewd practices against a 12 year old girl (including digital penetration); the repeated indecent assault and the repeated rape of one former partner; and the rape of a second partner on a single occasion. The Crown successfully appealed the sentence imposed by the trial judge. The Appeal Court imposed an extended sentence of 12 years, comprising a custodial term of eight years and an extension period of four years.123 The decision in SSK is also notable for the comments that the sentencing judge set out in his report to the Appeal Court in which his Lordship described the offences against the adult victims as ‘essentially non-violent relationship rapes’. The sentencing judge also clas­ sified the level of criminality and the level of harm suffered by the second adult victim as ‘minor’, involving only ‘a transient sense of violation’; his Lordship referred to the first adult victim as ‘condoning’ or ­‘acquiescing in’ the rapes; and stated that he might have considered ­admonishing the respondent for the rapes committed against the first adult victim had the convictions involving her stood alone.124 These comments were disapproved and criticised by the Appeal Court.125 In an observation that has particular resonance for the practice of sentencing, Solum states that a truly excellent judge needs the wisdom that can only be developed through experience (2008: 163). The cases cited above are notable for the fact that arguably none of the judges was a phronimos in the sphere of sentencing. The sentencing judge in Clark had been recently appointed to the Bench following a career as an advocate specialising in commercial law, local government law and medical negligence. The sentencing sheriff in 119 

Trainer (n 111) at [2]. Trainer (n 111) at [6]. Trainer (n 111) at [7]. 122  SSK (n 118). 123  SSK (n 118) at [27]–[28]. 124  SSK (n 118) at [5] and [7]. 125  SSK (n 118) at [22] and [23]. 120  121 

96 

Equity and the Rule of Law in Sentencing

Campbell had similarly specialised in civil litigation before his appointment to the Bench. The sentencing sheriff in Trainer was a recent appointment to the shrieval Bench, having previously specialised in family law and mental health law. The trial judge in SSK had also had a largely civil practice prior to his appointment. Whilst all were extremely learned and were experts in their respective fields, it would seem arguable that none had the experience in criminal law which Solum stresses is required before one can achieve excellence in this aspect of judging. Lest I be charged with the common law offence of slandering—or ‘mur­muring’—judges (Alison, 1832: 573–74; Christie, 2000: 750–51; Hume, 1844: 341; 405–06), it should be made clear that the sentencing judges in the cases considered above were ‘unvirtuous’ in the nar­ row term used by Solum: they imposed unduly lenient sentences because, by virtue of their areas of practice prior to their appointment to the Bench, they were not yet, at this stage in their judicial careers, able to exercise practical wisdom—phronesis—in the disposal of such difficult cases.126 It is thus contended that in the sentencing context, Solum’s concerns regarding the ‘unvirtuous judge’ are addressed by the right of both a convicted person to appeal against the sentence imposed and of the Crown to appeal against sentence on the grounds of undue leniency.

Individualised Notions of Justice Together, these cases, respondents’ views on the operation of statutory minimum sentences, and judicial recourse to equity in sentencing, all accord with the findings of Millie et al (2007). Having drawn together the findings from two related studies of sentencing in England and Wales and Scotland (Hough et al, 2003 and Tombs, 2004) the authors found that, con­ trary to many of the assumptions underlying contemporary penal policies and despite the legal and cultural distinctiveness of the two jurisdictions, legal categories were rarely signifi­ cant in determining whether a custodial sentence was imposed (Millie et al, ibid 254; see also Ashworth, 1987: 32–33). It was found that the most important factor for the sentencer was when ‘the time had come’ when they had ‘no alternative’ but to impose a custodial sentence. This ‘time’ was less to do with legal categories of offence than to do with the inter­ action between the offender’s criminal history and the seriousness with which sentencers viewed the offence in question (Millie et al, ibid). As Tombs and Jagger explain: [T]he individualized notions of justice embedded in forms of legal argument mean that some offences have an ambiguous status within discourse. The same offence may attract a custodial or community sentence. Moreover, offenders themselves (as individuals) do not fit neatly into the limited number of categories discursively available: they are only ever approximations of these categories … [W]hen making their decisions, sentencers consider not only the severity and nature of the offence, but also the characteristics of the offender (2006: 806).

Thus, a ‘space for judgment’ is opened up by the ambiguous status of some offences, together with the fact that offenders are merely approximations to categories of offences (ibid 807). This is the paradigm example of what Judge Richard Posner terms judicial decision making

126  The placing of such a high premium on judicial practical reasoning means, of course, that great care must be taken in the selection of judges (see O’Malley, 2011: 32–33 and Tamanaha, 2004: 124–25).

In Defence of the Instinctive Synthesis

 97

in ‘the open area’—where a judge is required to make a decision in a non-routine case, in an area where conventional legal materials (such as statutes and precedent) offer no—or very little—assistance (Posner, 2016: 22, 180–81, 2013: 106, 2008: 13, 15). There are no ‘clear, unambiguous criteria for a correct solution’ (Lloyd-Bostock, 1988: 61). This indicates a desire on the part of judges when sentencing such cases to do equity by giving reasons for their decisions based on the facts of the particular case and the circumstances of the partic­ ular, contextualised offender—the offender as ‘a unique nonreplaceable being, a being not like anyone else in the world’ (Nussbaum, 1990: 72)—and not on rules framed in general terms (Solum, 1994: 126). Comments from respondents in the present study on the importance of focusing on the individual offender, his background and history, included the following: When I was being trained as a prosecutor, one of the first things that my boss said to me was: ‘It’s prosecution in the public interest’. It’s the same with sentencing. It’s s­ entencing in the public interest. Never forget that the accused is a member of the public. So you c­ annot possibly, in my view, either prosecute or sentence without considering the i­ ndividual who committed the crime (Sheriff 5). Every individual’s got their own story, their own background. So that’s always taken into account … It has to be about the individual. What’s appropriate for that individual? What sentence on the individual would serve society best? (Sheriff 15). I don’t think the kind of pepper shot approach is helpful. You have to be focused on the individual, focused on their recent personal history (Sheriff 7).

Above all, the judicial comments on the importance of equity, together with the decisions in Edge, McArthur and Currie,127 along with the English decisions on cases under section 5 of the Sexual Offences Act 2003, demonstrate that judges and sheriffs endeavour to sentence according to values that privilege their own ­professional knowledge and experience as judges, rather than the policy goals of politicians (Tombs, 2008: 97). As Solum notes, the practice of equity is ad hoc, done on ­particular facts and requires the judge to focus on the uniqueness of ­particular fact patterns (Solum, ibid). Judicial recourse to equity ensures that offenders are treated as whole people and not as the ‘two-dimensional crime-and-criminal-history amalgams’ described by Tonry (1996: 20).

Pursuit of the Chimera—Questions of Consistency and Uniformity Perfect consistency is hardly ever achievable within any discretionary system (Lothian, 2006: 18; O’Malley, 2011: 13). To accommodate the practice of equity in sentencing, con­ formity to the rule of law in this area must be a matter of degree rather than the absolute requirement as apparently suggested by critics such as Edney and Bagaric (2007) (see Raz, 2009: 222). I suggest that the application of the rule of law to sentencing as proposed by Edney and Bagaric results in an approach that prioritises uniformity and consistency of decisions at the expense of individualisation (Aas, 2005: 104). Raz, for example, acknowl­ edges that complete conformity to the rule of law is impossible; some vagueness is inescap­ able and maximal possible conformity is on the whole undesirable since some controlled 127 

Edge (n 40); McArthur (n 44); and Currie (n 51).

98 

Equity and the Rule of Law in Sentencing

discretion is better than none (Raz, ibid). The rule of law always has to be balanced against competing claims of other values (ibid 228)—values which, it is suggested, include the need for individualisation in sentencing and flexibility in undertaking the sentencing task, both achieved through the sentencer doing equity. Raz concludes: Conflict between the rule of law and other values is just what is to be expected. Conformity to the rule of law is a matter of degree, and though, other things being equal, the greater the conform­ ity the better—other things are rarely equal. A lesser degree of conformity is often to be preferred precisely because it helps realization of other goals (ibid 228).

Raz considers that conformity to the rule of law is not itself an ultimate goal; he cautions against disqualifying the legal pursuit of major social goals in the name of the rule of law (ibid 229; see also Posner, 2008: 89). In sentencing, such social goals include the traditional aims of sentencing—protection of the public, denunciation, deterrence, and rehabilita­ tion.128 Raz also warns that the law may become barren and empty if too many social goals are sacrificed on the altar of the rule of law (ibid). Ashworth considers that the unprincipled nature of sentencing practice has resulted in a ‘cafeteria system’ of sentencing, which permits the sentencer to pick and choose whatever sentencing rationale seems appropriate at any given time (Ashworth, 2000: 63; 1995: 252; 1989: 350; see also Bagaric and Alexander, 2013: 139; Bagaric and Pathinayake, 2013: 399; Edney and Bagaric, 2007: 208; Hutton, 1995: 555; Krasnostein, 2015a: 222–23; Mackenzie, 2001: 292–94; and Renaud, 2012: 321). The cafeteria metaphor is clearly pejorative,129 but given the nature of the sentencing task, the ability to choose the rationale appropriate to the given case is arguably a distinct advantage. Walker and Padfield, for example, note that: … unless a sentencer is of the academic type which in every case has a single aim in view—just punishment, the reduction of crime or ritual satisfaction—he must be an eclectic. An eclectic selects his aim in the light of the circumstances of each case in which he has to sentence. His selec­ tion may be governed either by his emotional reaction to his information, or—preferably—by rules or principles which make his selection rational. One sensible rule … is that aims which have some chance of being achieved should be preferred to aims which do not’ (Walker and Padfield, 1996: 122; see also Padfield, 2013: 33–34 and 2003: 35–36).

As the Lord Justice General, Scotland’s most senior judge, noted in a recent address, the function of the first instance court in sentencing is ‘generally a practical one of applying the law at its raw end; not to invent or contemplate theoretical ideals’ (Carloway, 2013: 7). Sentencing practice may be akin to a cafeteria, but at least it is a cafeteria that is open and that has a wide selection from which to choose. A cafeteria which—in Raz’s terms—is bar­ ren and empty is no use to anyone, yet this is exactly what would result from a system that restricted judicial discretion and curtailed the judicial focus on individualisation of sen­ tences through regimes of mandatory sentences and rigid, presumptively binding, numeri­ cal sentencing guidelines.

128  For a full discussion of the way in which judges weigh and balance the often competing aims of sentencing, see Chapter 5. 129  Ashworth goes so far as to describe this situation as ‘a prescription for sentencing anarchy’ (1989: 350); see also Ashworth 1987: 31 (‘[t]he position is one of choice with little guidance, a veritable invitation to disparity’).

In Defence of the Instinctive Synthesis

 99

The inevitable subjectivity of the sentencing decision (Cowdery, 2006: 16; Findlay et al, 2009: 261; Hough et al, 2003: 42) means that it is practically impossible to eliminate dispar­ ity without resorting to such mandatory minimum sentences or rigid guidelines. Consistent sentencing, however, is not necessarily just ­sentencing (Cooper, 2008: 279). It is difficult to compare two cases even if they appear, ex facie, to be similar because of the many and var­ ied factors relating both to offender and offence that arise (Quigley, 1999: 11). As Manson observes: ­‘Intrinsic diversity among offenders and the myriad ways in which offences can be ­committed will necessarily combine to produce disparate sentencing results. Disparity itself is not a problem; only unjustified disparity demands scrutiny’ (Manson, 2004: 338; see also Chasse, 2010: 97). Quigley concludes that concern about disparity in sentencing has had altogether too much influence. He makes the important point that whilst in other legal contexts we are able to see equality concerns in a rich and contextual way, for some reason we are unable to do so in the sentencing context, with some commentators regarding standardised penalties as fair, just and therefore necessary (Quigley, ibid; see also Green, 1996: 119). Sentencers are, however, individual human beings each with their own notions about what constitutes serious criminal conduct, what the aims of sentencing should be, and what is appropriate as a sanction. This simply cannot be regulated (Quigley, ibid 12). As one respondent in the present study put it: ‘There’s no panacea, there’s no obvious answer to any of it and we’re all just doing our best within the confines of a system staffed by humans’ (Judge 3). Thus the appellate courts should recognise that true parity in sentencing is next to impos­ sible to achieve (Quigley, ibid 22; see also Hough et al, 2003: 42). The ideal of uniformity in sentencing presupposes a uniformity of crime and criminals which simply does not exist (Karlton, 1991: 186). As Morgan and Murray observe, whilst it is easy to talk in general terms about ‘consist­ ency’ in sentencing, it is much harder to explain what consistency might mean (Morgan and Murray, 1999: 95). It might mean consistency of ‘purpose’ or ‘philosophy’ (agreement on the basic aims of sentencing); consistency in ‘approach’ (taking account of the same factors and giving similar weight to those factors); consistency in ‘outcome’ or ‘result’ (imposing the same type and quantum of sentence); or a variant or combination of these meanings (ibid). Consistency is thus ‘a fluid and slippery concept’ (Padfield, 2013: 32). The following observations of Spigelman CJ, delivered in the course of handing down the first guideline judgment by the New South Wales Court of Criminal Appeal130 are particularly apposite: The existence of multiple objectives in sentencing—rehabilitation, denunciation and deterrence— permits individual judges to reflect quite different penal philosophies. This is not a bad thing in a field in which ‘the only golden rule is that there is no golden rule’ (Geddes, (1936) 365 SR (NSW) 554, 555 per Jordan CJ). Indeed, judges reflect the wide range of differing views on such matters that exists in the community.

Thus Spigelman CJ appears to encourage diversity rather than consistency of ­purpose on the part of sentencers (Morgan and Murray, ibid 96; see also O’Malley, 2011: 240; Padfield, ibid 33–34; and Spigelman, 2008: 450). The same position was taken by respondents in the present study. Only three respondents (one judge and two sheriffs) considered consistency

130 

R v Jurisic (1998) 45 NSWLR 209 at 221.

100 

Equity and the Rule of Law in Sentencing

in sentencing to be either desirable or achievable. A further three respondents recognised that, at a very broad level, like cases should be treated alike; as one judge put it: ‘generally speaking, people who commit similar murders or kill by dangerous driving, broadly we kind of understand the range within which the sentence should fall’ (Judge 6). Even those respondents who held this view, however, continued to emphasise the importance of indi­ vidualised sentencing: Although consistency of sentence is desirable as a generality, there are situations in which a degree of inconsistency has value. Housebreaking in rural communities in the Highlands is regarded much more seriously than housebreaking in urban areas and will attract a different type of penalty (Judge 7). Clearly at the broadest level of generalisation, like cases should be treated alike. But when one thinks of the many different variegations of human behaviour—and though one incident appears the same as another incidence—then one puts into the equation the distinctive features of the indi­ vidual accused, the individual victim and so on, then it’s difficult to know how or why you should aim to have consistency at a micro level … [At that level] it’s neither desirable nor achievable … There must be tolerance for the factors unique or singular to that case (Judge 6). I’m not sure that consistency equates necessarily with justice … I think we should attempt to achieve a degree of consistency—I’ve got no problem with that, provided it’s balanced against the just result in the individual case (Sheriff 16).

Indeed, one of the three respondents who referred to the importance of consistency acknowledged the impossibility of achieving it: Consistency in sentencing has to be achieved as far as possible but it’s got to be recognised that you’re never going to get it entirely (Sheriff 9).

The view of six judges and 16 sheriffs in the present study was that consistency simply can­ not be achieved: The only way you’ll get consistency is by using a grid structure like they have in the States. What does consistency mean? Does it mean that everyone is to get the same sentence? Certainly not! Does it mean that the sentences we can pass should be defined in such detail that we can only sentence within narrow bands? No, it doesn’t mean that either. The circumstances of cases are infinitely vari­ able. Consistency is a chimera (Judge 5). I have a problem with the concept of consistency in sentencing. Consistent with what? If we go back to the question of what do you take into account in sentencing, it doesn’t m ­ atter where you start, every single crime is different, at least to some extent. Because each criminal is different, each victim is different, the consequences for each—the accused and the victim—are different, the demands of any particular locale are different … It’s a buzzword, a politician’s buzzword that doesn’t actually mean anything. And it’s not something that should be aimed for (Sheriff 5). I certainly don’t think consistency is achievable and I think I would have to go the length of say­ ing it’s neither necessary nor desirable. I don’t think it’s achievable because no cases are the same. The facts of the offence are generally different and certainly the personal circumstances and all the surrounding factors involved in the case are completely different in every case. So you cannot have consistency. People shouldn’t be going on about it and it would be wrong in principle to have it because that would mean you would be ignoring factors that you really ought to be taking into account (Sheriff 2).

In Defence of the Instinctive Synthesis

 101

Absolute consistency is entirely inconsistent with a just outcome and that is the difficulty with sentencing (Sheriff 13).

Similar sentiments were voiced by the Sheriffs’ Association in its submission to the Scottish Parliament’s Justice Committee during the passage of the Criminal Justice and Licensing (Scotland) Bill 2009, noting that since ‘a sentence is a decision in an individual case, then consistency is a pipe dream, an unachievable goal unless one takes no account of individual circumstances’ (Sheriffs’ Association, 2009: paragraph 2.4, cited in Tata, 2013: 254). We have already seen how, in stressing the need for individualisation in sentencing, the Supreme Court of Canada has stressed that there is no such thing as a uniform sentence for a par­ ticular crime.131 In delivering the leading judgment in Rees v R, Garling J noted that: ‘There is no such thing as perfect consistency in sentencing. A search for perfect consistency is to look for the unattainable and will frequently be an exercise of academic abstraction’.132 Judge Richard Posner also accepts that consistency in sentencing is difficult, if not impos­ sible, to achieve (Posner, 2013: 314–15). One of the most forceful articulations of the need for individualised sentencing which consequently results in sentences that may appear inconsistent to those unfamiliar with the particular facts and circumstances, came from Sheriff 16, a particularly experienced sheriff who had sat for over 10 years: It all emphasises the importance of context, both of the offender and of the offence. I’ve had this discussion with [named academic] and [said academic] is very much in favour of taking sentenc­ ing away from judges. This is the really funny thing. Taking sentencing away from the judges! Well who’s going to do it? It’s part of a judge’s job. You might not like it but what are you going to do? Put it into a computer to feed in the information and ask the computer and it does it? I’d be wholly against that. You are going to get people saying, ‘Well, he got six months and he got nine months, or community service’ or whatever. If you just take the bare facts, the bare details, you are going to get apparent inconsistencies but it doesn’t necessarily mean the sentence was wrong in each individual case (Sheriff 16).

These findings are consistent with previous studies of the Scottish judiciary’s views on sen­ tencing (see Millie et al, 2007: 250–51). Moreover, several commentators have emphasised that the exercise of sentencing discretion must remain in the hands of the judiciary; com­ puters, for example, cannot be programmed to register the ‘feel’ of a case or the impact that an offender makes upon the sentencer (Cooke, 1987: 58; Murray, 2002: 198). Thus an important argument in favour of judicial sentencing discretion is that judges are uniquely positioned to impose appropriate sentences (van Zyl Smit, 2000: 198) as their experience allows them to do equity. One respondent explained: I mean honestly—if you’ve been in a trial and you’ve watched the accused and thought about it and been steeped in the law, it’s not very helpful for someone to tell you what to do who either hasn’t got that general experience or the particular experience about how you should sentence that individual in front of you (Judge 6133).

131  R v M (CA) [1996] 1 SCR 500 at [92]. Discussed in Chapter 3 under the heading ‘Sentencing Methodology in Canada’. 132  Rees v R [2012] NSWCCA 47 at [50]. 133  See also the observations of the Supreme Court of Canada in R v Lacasse [2015] 3 SCR 1089 at [11].

102 

Equity and the Rule of Law in Sentencing

As a former Chief Justice of the High Court of Australia observes, whilst the judge is limited by the armoury of penalties available to do justice in the particular case, nobody is better equipped with information and submissions to weigh up the competing purposes of sen­ tencing (Brennan, 2006: 7; for similar judicial views, see Spigelman, 2008: 452 and 2006, and Bell, 1999: 71).

Conclusion A sentencing judge’s decision making is instructed by a knowledge of comparable cases, of relevant sentencing principles, and of experience of the criminal law and so can be seen as the very antithesis of ‘instinctive’ decision making. Nevertheless, the term ‘instinctive synthesis’ remains the term used by the Australian courts, and increasingly by the Scot­ tish courts, to describe the effect of balancing a variety of often conflicting mitigating and aggravating factors to reach a considered conclusion as to the appropriate sentence in factspecific, individual cases. The sentencing jurisprudence of appellate courts throughout the Commonwealth focuses on individualised justice by encouraging sentencers to see each offender as an individual for whom the appropriate and just sentence cannot simply be pulled from a mould. The instinctive synthesis is the method most compatible with both the principle of individualised justice and the requirement for a broad judicial discretion in sentencing. In examining academic criticism of the instinctive synthesis, it has been shown that commentators such as Edney and Bagaric (2007) and Hutton (2006a) consider that this sentencing methodology violates the rule of law. Critics have raised concerns regarding the predictability, certainty, clarity and transparency of sentencing decisions made by way of instinctive synthesis. It has been claimed that the idiosyncratic intuitions of sentencers often trump the rule of law virtues of consistency and fairness. On this view, sentencing decisions are simply expressions of value preference made by the individual sentencer; they are neither determined by law nor recognisable as legal decisions (Edney and Bagaric, ibid; Hutton, 2003). Such criticism of the instinctive synthesis sentencing methodology is predicated upon a flawed conception of the rule of law. The operation of the rule of law cannot be seques­ tered from human participation (Tamanaha, 2004). In sentencing, as in all other areas of law, legal rules and principles cannot be applied without human reason, insight and judg­ ment. Whilst convicted persons are entitled to the protection of the rule of law, sentenc­ ing must be marked by a wide judicial discretion due to its complexity. This discretion is not, however, completely unconstrained: the sentence imposed by the judge will be a result of legal doctrine, institutional constraints, policy, strategic considerations, and the equi­ ties of the individual case. These factors are both mixed together and mediated by judicial experience in a form of detailed and flexible decision making that allows the sentencer to respond appropriately to the seriousness of the crime and the individual circumstances of the offender (Posner, 2008; Waldron, 1990). In his work on the concept of equity in judicial decision making, Solum (2008, 2003, 1994) regards equity as the doing of particularised justice—justice as fairness, distinguished from justice as lawfulness. Although all law is universal there are some things about which it

Conclusion

 103

is not possible to pronounce rightly in general terms (Meyer, 2007; Solum, ibid; Tamanaha, 2006). Departing from legal rules in order to achieve particularised justice is consistent with the rule of law when legal decision makers possess the virtues of judicial intelligence—a form of sophia (or theoretical wisdom)—and judicial wisdom, the adjudicatory form of the intellectual virtue of phronesis, or practical wisdom (Solum, 2003 and 1994). In applying this ‘virtue-centred’ theory of judging to the judicial task of sentencing, it has been argued that as a phronimos, or a person of practical wisdom, the sentencer ought to have the experience to discern what is morally relevant about a particular case and thus be able to discern those situations in which a departure from the letter of the law is consist­ ent with its spirit and hence with the rule of law (Solum, 1994). This is not an arbitrary departure from the legal rules governing sentencing. The purposes and values inherent in the law itself still hold sway and, perhaps more importantly, the moral and legal vision that grounds the practice of equity is also necessary to understanding the application of such legal rules. On this view, the practice of equity in sentencing may actually reinforce, rather than undermine, the values of predictability and regularity that support the ideal of the rule of law. Judges are also constrained by habitus, or certain ‘regularities in behaviour’ which pattern their sentencing. Sentencers will have their decisions overturned on appeal if they seek to stretch legal rules too far, in the sense of imposing sentences which are unduly leni­ ent or disproportionately severe (Tamanaha, 1997). This is a particularly important factor in a small jurisdiction such as Scotland. The practice of equity as dispensing individualised justice has been illustrated by refer­ ence to the sentencing jurisprudence and the sentencing practice of the Scottish courts. Domestic and Australian academic studies (Mackenzie, 2005; Tombs, 2008), case law, and judicial views on the operation of proposed and current legislation that seeks to curtail judicial discretion through the imposition of mandatory sentences, all demonstrate that sentencing in Scotland accords with that of the Commonwealth jurisdictions examined in Chapter 3 in that it remains an individualised process characterised by high degrees of judicial discretion. By reference to the operation of certain statutory minimum sentencing schemes in South Africa and Australia, it has been demonstrated that such regimes are deeply problematic for a variety of reasons: they lack proportionality and take insufficient account of the circum­ stances of individual offences leading to grossly unjust outcomes; they contribute to prison overcrowding; they lead to inflationary drift in sentencing; they undermine ‘rule of law’ virtues as largely unaccountable pre-trial decision making by prosecutors comes to assume more importance than public decision making by judges; and they fail to show any crime reduction effect. In short, as the Australian Law Reform Commission recently concluded, the prescribing of mandatory terms of imprisonment by the legislature offends against the principles of proportionality, parsimony and individualised justice (Australian Law Reform Commission, 2006). Statutory minimum sentencing schemes are anathema to Scottish sentencers. Judges and sheriffs continue to practice equity through the imposition of indi­ vidually crafted sentences, as opposed to rigidly following universal rules. This demon­ strates the way in which, as Meyer notes, justice is inherently particular and contextual, not merely the consistent application of rules (Meyer, 2007). As a further illustration of the importance of individualised justice, the place of leniency and mercy in sentencing and the imposition of condign punishment as equity have been considered. Through an examination of domestic case law, Commonwealth case law and

104 

Equity and the Rule of Law in Sentencing

previous academic studies along with the qualitative empirical data from the present study, it has been shown that both leniency and mercy will feature in the exercise of sentencing discretion in appropriate cases. Every respondent in the present study agreed that justice and the public interest can, in appropriate cases, be best served by lenience in sentencing— especially with less serious cases prosecuted on summary complaint and also in cases where the offender suffers from mental health problems. Although 22 respondents saw mercy as being qualitatively different from leniency, they still considered it a legitimate—indeed, an essential—consideration, but again only in appropriate cases. Many respondents, however, also acknowledged a difficulty in articulating a precise justification for reducing the severity of a sentence on this ground. Similarly, we have seen the ways in which the practice of equity—as the doing of particu­ larised justice in the case of a contextualised, individual offender—can result in the imposi­ tion of severe sentences. The present study found there to be a general consensus amongst respondents on factors that serve to aggravate offences. They included previous convictions (seen in terms of ‘the recidivist sentencing premium’ described by Roberts, 2008); breach of a position of trust; planning or premeditation; the targeting of vulnerable complainers; unprovoked and gratuitous violence; repeated offending; voluntary intoxication; and lack of remorse. It has also been shown, by reference both to appellate sentencing jurisprudence and the findings of the present study, that the type of offence or the facts of a particular case can result in custody even for offenders of previous good character. This, it is submitted, is a further example of equity in sentencing: the nature of the offence dictating that a severe sentence is warranted despite strong mitigating factors. As the present research also demon­ strates, equity can on occasion even extend to the imposition of custodial sentences beyond what would normally be considered retributively justifiable in order to allow drug addicted offenders, for example, to undertake rehabilitative and detoxification programmes. A discretionary approach to sentencing is preferable to a regime of mandatory sentences which are based on the nomenclature of crimes rather than on the seriousness of individual acts (Aas, 2005). The findings of this aspect of the present study accord with the earlier con­ clusions of Tombs and Jagger (2006) and Millie et al (2007) that, amongst sentencers, legal categories are rarely significant in determining whether to impose a custodial sentence. The present study is also consistent with Tombs’ (2008) findings that judges and sheriffs in Scot­ land sentence offenders according to values that privilege their own professional knowledge and experience as judges, rather than the policy goals of politicians. Individualised justice is achieved by the judge using ‘equity’s flexible standard’ (Nussbaum, 2001, 1993a) to tailor the sentence to meet the demands of the particular facts and circum­ stances of the individual case. Sentencing ought to be seen as inherently contextual and particularised, with a concomitant need for a wide judicial discretion. Although perfect consistency in sentencing will never be achieved through this approach—by its very defi­ nition, the sentencing methodology of instinctive synthesis will produce outcomes upon which reasonable minds will differ—consistent sentencing is not necessarily just sentencing (Cooper, 2008). Indeed, 22 of the 25 respondents in the present study considered that con­ sistency in sentencing simply cannot be achieved. Judicial discretion and recourse to equity ensures that offenders are treated as individuals—as whole people—rather than as Tonry’s (1996) ‘two-dimensional crime-andcriminal-history amalgams’. The approach to sentencing advocated by critics such as Edney, Bagaric and Hutton, however, puts a premium on the transparency and standardisation

Conclusion

 105

of penal decisions. Such critics prioritise uniformity and consistency in sentencing at the expense of individualisation. In their pharisaic veneration for certainty and predictability in sentencing, commentators such as Edney, Bagaric and Hutton ignore the motivations and constraints operating on sentencers and the judicial mentality that results (see Posner, 2008: 377). It is to the practical wisdom of sentencing that the discussion now turns.

106 

5 The Practical Wisdom of Sentencing: A Rationalisation of Intuitive Decision Making and Judicial Discretion Introduction The social character of sentencing is now examined in more depth by reference to the con­ cept of value pluralism, the central idea in the work of Isaiah Berlin (2013a–d, 2002a–b), and by the Aristotelian concept of phronesis—or ‘practical wisdom’—as recently re-­interpreted by Bent Flyvbjerg (2001). Berlin’s value pluralism is contrasted with the concept of moral monism. Pluralism is then considered in the context of sentencing as a means by which, in imposing punishment, potentially competing and incommensurable societal values and values particular to the individual offender are weighed and balanced. The concept of phronesis is then introduced. The discussion proceeds by applying the Dreyfus model of learning (Dreyfus and Dreyfus, 1986) to the practice of sentencing in an attempt to understand the process, from the perspective of sentencers, as a form of rational social action. It is argued that it is phronesis as practical wisdom that allows judicial recourse to equity in sentencing: individualised justice is best achieved through a phronetic synthesis of all the relevant facts and circumstances of the particular case.

Berlin’s Value Pluralism Deciding how much to punish is an agonizing process in which conflicting aspirations compete. The best one can do is decide which aspirations, on balance, appear to be most important—and build one’s theory on them (von Hirsch, 1986: 59).

As the High Court of Australia noted in Norbis v Norbis,1 judicial discretion in sentenc­ ing involves the making of value judgments in respect of which there is room for reason­ able differences of opinion, no particular opinion being uniquely right.2 This is due to the dilemmas and conflicts inherent in the task itself: the multiplicity of possible goals (utilitarian, moral, or administrative) mean that even with hindsight it is not possible to

1  2 

[1986] HCA 17 at [4]. See also R v MacNeil-Brown [2008] VSCA 190 at [141].

108 

The Practical Wisdom of Sentencing

state definitively that a sentence was the ‘right’ one (Pennington and Lloyd-Bostock, 1987: 2; see also Krasnostein, 2015a: 70). Proper exercise of the sentencing discretion, and the making of such value judgments, nevertheless involves balancing and accommodating ‘all the manifold competing circum­ stances and influences’3 with the result that ‘[v]iews on sentencing outcomes will almost always vary, depending significantly, although not exclusively, upon the perspective of the commentator’.4 Ashley JA of the Supreme Court of Victoria elaborates: Judges and lawyers speak about a ‘sentencing discretion’. What it means is this. In deciding what sentence is to be imposed in any particular case, a judge or magistrate has to bring to account many circumstances of the offence, the offender and the victim, which are likely to point in different directions—some towards a heavier sentence, some towards a more lenient sentence … [I]t is for the judge, balancing out all the circumstances, to select what seems to be the appropriate sentence. It has repeatedly been said by courts of appeal, and rightly, that no one sentence is correct. There is a range of sentences, all of which are open. Any number of judges might conscientiously e­ xamine the same circumstances and yet consider that somewhat different sentences were ­appropriate. ­Sentencing is not simply an arithmetical process.5

Thus sentencing involves the judge weighing or balancing both societal values, such as jus­ tice, liberty and community, and values particular to the individual offender: denunciation, retribution, specific deterrence, rehabilitation and perhaps also mercy. How are such com­ peting claims and values to be reconciled? In his famous essay on Tolstoy’s historical thought, The Hedgehog and the Fox, Isaiah Berlin quotes the Greek poet Archilochus: ‘The fox knows many things, but the hedgehog knows one big thing’ (Berlin, 2013a: 436). As Crowder explains, these words suggest a pro­ found distinction between two kinds of thinker (or, for present purposes, two kinds of sen­ tencer): on the one hand those who, like the hedgehog, ‘relate everything to a single central vision, one system, less or more coherent or articulate, in terms of which they understand [and] think’; and on the other, those who, like the fox, ‘pursue many ends, often unrelated and even contradictory’, who think: on many levels, seizing upon the essence of a vast variety of experiences and objects for what they are in themselves, without, consciously or unconsciously, seeking to fit them into, or exclude them from, any one unchanging, all-embracing, sometimes self-contradictory and incomplete, at times fanatical, inner vision (Crowder, 2004: 1, citing Berlin, ibid 436–37).

The contrast that Berlin draws between ‘hedgehogs’ and ‘foxes’ is a metaphor for a c­ rucial distinction at the heart of his political thought, namely the distinction between moral ­monism and value pluralism (Crowder, 2003: 333).6 Monism has been the mainstream 3 

R v Burns [2007] NSWCCA 228 at [36]. Ibid. 5  DPP v McMaster [2008] VSCA 102 at [63]–[64], references omitted. 6  As Crowder (2004: 126–27) explains, although Berlin never offers any systematic analysis of the concept of value pluralism, preferring to allow it to emerge from discussions of other ideas or from studies of particular thinkers like Vico and Herder, the pluralist theme nevertheless persists throughout Berlin’s writing. Berlin himself confirms pluralism as one of his central concerns when he comes to reflect on his life’s work in his last essays and interviews (see Crowder, ibid 127 and references cited therein). Although the notion of pluralism is intimated by several earlier Western thinkers, Berlin is the first writer to formulate the idea explicitly as the centrepiece of a sustained political philosophy (Crowder, ibid 190; see also Crowder, 2002: 17, fn 2). 4 

Berlin’s Value Pluralism

 109

belief of Western philosophy (Crowder, 2007: 208, 2004: 190). It involves ‘faith in a s­ ingle criterion’ (Berlin, 2013b: 241), a ‘single ­embracing vision’ (Berlin, 2013a: 466), and the belief in a single overriding end to which everything else must be subordinated (Crowder, 2007: 208, 2004: 26, 2002: 4). On the monist view, there is a single right way of answering any moral or political question. The monist considers that all moral values must somehow fit together in a single, coherent system—a ‘single systematic account of reality’ (Crowder, 2004: 125). All true moral judgments dovetail neatly within this single, exceptionless and rationally comprehensible system (Crowder, 1994: 294). For the monist thinker, all values can be commensurated by subordination to a summum bonum, or super-value, or some other scheme of ranking independent of context (Crowder, 2002: 69, 1994: 295). It is this system that yields the correct answer to any moral problem (Crowder, 2004: 4); for if there is one overriding value or set of values or formula, then there will be one in-principle solu­ tion to all conflicts of values and interests: ‘Do whatever will realise or maximise the supervalue’ (Crowder, 2002: 92). It is the task of the monist to uncover the path that leads to this single correct answer (Berlin, 2000: 5, 2002a: 47)—the answer that is seen as uniquely superior to all alternatives (Crowder, 2004: 73). Berlin also emphasises the dominant role played in modernity by the distinctively ­science-based monism of the Enlightenment in which scientific investigation of the human world (as of the natural world) will yield a harmonious set of laws which can be formulated by experts (2002a: 115). On this view, all human goods will be realised within a single moral and political system, authorised, administered and enforced by these same experts (ibid; see also 128). For the pluralist, however, basic human goods are seen as multiple, conflicting and incommensurable (Berlin, 2013a: 466; Crowder, 2004: 1–2, 2002: 2 and 46–56, 1994: 293). This notion of the incommensurability of values lies at the heart of pluralism (Crowder, 2004: 126). Values are incommensurable in the sense that they are so different from one another that each has its own character and force, untranslatable into the terms of any other (Crowder, ibid 4; 2002: 52). For the value pluralist, no basic value is inherently more important, more authoritative, or weightier than any other; each value is seen by the plural­ ist decision maker as ‘a separate jewel in the crown, valuable in its own right’ (Nussbaum, 1990: 63). No one value embraces or summarises all other values (Crowder, 2002: 2–3), thus any faith in a single static pattern or system of values is ‘blown to pieces’ (Hausheer, 2013: xlii). The incommensurability of values means that there is no common measure by which they can be weighed inter se; rather each good is its own measure (Crowder, 2007: 207, 2004: 74, 2003: 336; Duff, 2005: 1176). Each good is ‘equally ultimate’ as it possesses its own intrinsic value, its own weight and its own independent moral force (Berlin, 2013b: 239; Crowder, 2004: 138; Hausheer, ibid xl; Nussbaum, ibid 68–69). As Berlin explains: [H]uman goals are many, not all of them commensurable, and in perpetual rivalry with one another. To assume that all values can be graded on one scale, so that it is a mere matter of inspec­ tion to determine the highest, seems to me to falsify our knowledge that men are free agents, to represent moral decision as an operation which a slide-rule could, in principle, perform (Berlin, ibid 241).

Pluralists consider value conflicts to be both inescapable and on occasion hard—although not always impossible—to resolve rationally (Crowder, 2002: 55). The best that can be said

110 

The Practical Wisdom of Sentencing

is that some values will have priority in some cases, whilst other values will have priority at other times (Crowder, ibid 7). Plural and incommensurable values may, in particular cases, come into conflict with one another. They may be incompatible or mutually exclusive, such that one particular value can only be realised at the cost of sacrificing another (Crowder, ibid 3, 54; Williams, 1978: xvi). For the value pluralist, the difficulty of many of the moral choices we must make stems from the deeply fragmented nature of the ends we pursue (Crowder, ibid 5). This results in hard choices for the decision maker—perhaps choices that are agonising and tragic7—that cannot be resolved by any simple, abstract rule (Crowder, 2004: 74, 138, 190), or by a single formula that can be used in all cases (Crowder, 2007: 208–09; Hausheer, ibid xxxviii). Berlin elaborates: The world that we encounter in ordinary experience is one in which we are faced with choices between ends equally ultimate, and claims equally absolute, the realisation of some of which must inevitably involve the sacrifice of others. Indeed, it is because this is their situation that men place such immense value upon the freedom to choose; for if they had assurance that in some perfect state, realisable by men on earth, no ends pursued by them would ever be in conflict, the necessity and agony of choice would disappear, and with it the central importance of the freedom to choose (Berlin, ibid 239).

It is precisely because values conflict that we have to choose among them, and therefore that we value the freedom with which to choose (Crowder, 2002: 81). Crucially, however, the pluralist considers that there can be no incontestable scheme for harmonising such difficult decisions (Williams ibid). If goods are incommensurable or unrankable then when they conflict—which they will do frequently—there is no single way of deciding between them that will be uniquely correct in all cases (Crowder, 2002: 259–60; Kronman, 1987: 847–48; Williams, ibid). Many different ways of ranking incommensurable goods in individual, con­ crete cases, or in particular types or category of case, may be equally reasonable (Crowder, 2004: 75, 130–31, 139). We cannot devise a single formula or apply any simple rule that either reduces rival goods to a common denominator, or that arranges them in a single hierarchy that applies in all cases (Berlin, ibid; Crowder, ibid 4; 1994: 294). Nussbaum, for example, stresses that excellent choice cannot be captured in universal rules because it is a matter of fitting one’s choice to the complex requirements of a concrete situation, taking all of its contextual fea­ tures into account. Complexity and contextual variety call for what she describes as ‘respon­ siveness and yielding flexibility, a rightness of tone and a sureness of touch that could not be adequately captured in any general description’ (Nussbaum, 2001: 303–04; see also 1990: 71–72). The good judge, Nussbaum explains, will not decide by subsuming a case under antecedently fixed rules, and neither will he or she accept that there is ‘a general proce­ dure or algorithm’ for computing what to do in every case: ‘The appropriate response is not arrived at mechanically; there is no general procedural description that can be given ­concerning how to find it’ (Nussbaum, ibid 73).

7  See, for example, Darbyshire, 2011: 207 and the decisions in Neil v HMA 2003 GWD 33-951; HMA v Hegarty 2002 SCCR 1022; and HMA v KH 2014 JC 195.

Berlin’s Value Pluralism

 111

The dichotomy between the pluralist and monist views of decision making is highlighted by Berlin and Williams: It is true that there are some conflicts, particularly involving issues of public policy, that are better decided by simple and publishable rules than by individuals’ judgements of importance. Equally, there are other matters that are better left to judgements of importance. Moreover, there is, inescapably, no mechanical procedure for deciding which are which. None of this in itself sheds much light on the powers and limits of reason except to remind us that practical decision could not in principle be made completely algorithmic, and that a conception of practical reason which aims at an algorithmic ideal must be mistaken … [I]t is not a requirement of reason that there should be one value which in all cases prevails over the other (Berlin and Williams, 1994: 307, emphasis added).

This does not mean that choices amongst incommensurables are necessarily non-rational. Pluralism, Crowder explains, is not to be confused with either subjectivism or relativism. Rather, it implies a particular set of objective norms that ought to guide our choices uni­ versally (Crowder, 2002: 261–62; see also 2003: 338). Nor does it mean that no such choice can be more justified than any other; however, it does mean that pluralist choices tend to be ‘complex and often painful’ (Crowder, 2004: 4–5). The pluralist seeks to accommodate the choices and disagreements inherent in decision making, rather than trying to transcend them (Crowder, 2003: 337). In particular, and as Crowder (2004: 139–40) explains, although incommensurability may place problems in the way of reasoned decision making, it does not rule out reasoned choice within a particular context. Choosing amongst ­incommensurables is always hard as one’s choice cannot be guided adequately by any simple monist standard and, furthermore, the choice always involves regrettable loss (Crowder, ibid 131–32; Kronman, ibid 851). Nev­ ertheless, decisive reasons to choose in one direction rather than another will be generated by the context of the particular case (see also Crowder, 2002: 73 and Nussbaum, 1990: 63). As Berlin puts it, ‘the concrete situation is almost everything’ (Berlin, 2013c: 15; see also Nussbaum, 1993b: 258). We are, as Nussbaum observes, ‘lost without the fuller context’ (Nussbaum, 1990: 88). For Berlin, successful decision making does not involve adhering to a set of abstract rules derived from a putative ‘political science’; rather ‘what matters is to understand a particular situation in its full uniqueness’ (Berlin, 1997: 44–45). Good choices always involve close attention to the particulars of the case (Crowder, 2004: 140). To identify a value conflict as a conflict of incommensurables is not necessarily to abandon hope of a reasoned resolution; such a resolution can appeal to the context of the particular case (ibid 188). Value pluralism thus implies the inescapability of choice in all spheres of human experi­ ence including, for our purposes, sentencing decisions taken by judges. Pluralism entails the impossibility of moral and political perfection, and the inevitability of disagreement and conflict. On the pluralist approach, we must accept fundamental disagreement about the good and seek to contain and manage such disagreement (Crowder, ibid 5). In a passage that has particular resonance for the sentencing judge, Crowder explains: Utopian thinking, made possible most fundamentally by moral monism, is an invitation to justify any means by reference to an end that is, by definition, ultimately and absolutely desirable. The value pluralist view denies that there can be any such ultimate end, insists rather that there are

112 

The Practical Wisdom of Sentencing

many different ends to be balanced, and counsels care and moderation in seeking that balance (ibid, emphasis added).8

Williams, for example, makes the point that if there are many and competing genuine values, then the greater the extent to which a society tends to be single-valued, the more genuine values it neglects or suppresses (Williams, ibid xvii). For Williams—as for Berlin— more, to this extent, must mean better (ibid; see also Crowder, 2007: 220). Thus Berlin’s pluralist ‘fox’ is concerned with ‘the concrete details of human life, its precise contours and textures … [its] irreducible multiplicity and complexity’ (Crowder, 2004: 31). The pluralist is committed to a form of decision making that accepts conflicts and hard choices as per­ manent features of human experience that must simply be coped with (Crowder, 2002: 78). In this way, Berlin makes the case for pluralism, complexity and nuance in decision ­making (Marr, 2013: xvi). For Berlin, moral monism is false: the monist is blind to reality (Berlin, 2002a: 47, 1997: 53) as no single formula can resolve all ethical conflicts (Crowder, 2007: 208). Whilst human goods are deeply plural rather than unitary, inherent in the mon­ ist view are ‘the errors of omission and s­ implification, of succumbing to the illusion that one value can override all others and restructure everything’ (Williams, ibid xvi). ­Monism fails to do justice to the depth and persistence of conflict in the moral experience of human beings—experience­that teaches us, firstly, that we are frequently faced with choices amongst competing goods, and secondly that no clear answers to such hard choices can be derived from simple monist rules (Crowder, 2004: 4; 130; see also Nussbaum, 2001: 301). What is more, Berlin regards the monist outlook as positively dangerous: to suppose that moral and political perfection is possible, even in principle, is to invite the thought that its realisation justifies the employment of any efficient means (Berlin, 2013c: 12–13; Crowder, ibid 4).9 As Crowder explains, once a single value (for example, consistency in sentencing) or a relatively narrow range of values is held up as overriding or commensurating all other concerns (for example, equity and individualised justice in sentencing), the way is open to the marginalisation, downgrading and suppression of much that is in fact important to people (­Crowder, 2002: 92, 97). In order to realise the goal of the perfect society, no sac­ rifice, whether of liberty or of other values, will be too great for the monist: as the monist goal is overriding, it justifies by definition its enforcement at any cost (Crowder, 2004: 74; Nussbaum, 1990: 101). In this way, monism can be seen as ‘the ultimate killer of freedom’, as ‘a terrible error’ and as ‘a universal source of cruelty’ (Marr, ibid xii). Whilst monism makes ‘absurd claims, promising the impossible’ (Berlin, 2013a: 495), it also leads to ‘absurdities in theory and barbarous consequences in practice’ (Berlin, 2002a: 48). Monism encourages an expectation of the possibility of a total convergence of values and interests that is unrealistic and that has consequences that are ultimately inhumane (Crowder, 2002: 97). For Berlin: [T]he results of [monists’] labours, even when they are not nugatory, tend to take unpredicted, often catastrophic, directions—and all this because they will not, being vain and headstrong, admit

8 

On the notion of sentencing as a ‘balancing exercise’, see below. of Berlin’s work, written during the Cold War, can be understood as a response against twentieth century totalitarianism, particularly communism (see Crowder, 2004: Ch 3 and Hausheer, 2013: xli–xliv). In this regard, Berlin sees moral monism as the ultimate premise of the authoritarian outlook (see Crowder, ibid 73). 9  Much

Value Pluralism in Sentencing

 113

that too many factors in too many situations are always unknown, and not discoverable by the methods of natural science (Berlin, 2013a: 495).10

Berlin thus considers that ‘liberty may be incompatible with, and better than, too much effi­ ciency’ and that ‘pluralism and untidiness are, to those who value freedom, better than the rigorous imposition of all-embracing systems’ (Berlin, 1998: 257). Berlin’s thesis has pro­ found implications for the methodology of, and the achievement of justice in, sentencing.

Value Pluralism in Sentencing It is a truism that a judge must have regard to a variety of aims and purposes in imposing sentence. The sentencing decision is a complex one, containing many distinct features and dimensions of the offence, the offender and the victim (see Crowder, 2002: 83). In jurisdic­ tions that accord judges high levels of discretion in sentencing, in which sentencing is seen as ‘a delicate art based on competence and expertise’11 involving ‘a wise blending of penal aims’,12 values such as justice, ­liberty and community are not monistic in structure but plu­ ral. They are irreducibly multiple and incommensurable. The sentencing decision involves an inescapably difficult set of choices amongst conflicting aims, values and goods (Crowder, ibid 89; von Hirsch, 1986: x). Although, as Mackenzie and Stobbs observe of sentencing practice in Australia, the pur­ poses of sentencing are not infrequently overlooked in everyday practice, with the courts tending to take a more pragmatic and practical approach, the aims and purposes of criminal punishment provide an important theoretical and policy basis for sentencing (­Mackenzie and Stobbs, 2010: 42; see also Nicholson, 1992: 177; O’Malley, 2016: 44; and Walker and Padfield, 1996: 122 in relation to practice in Scotland, the Republic of Ireland and England and Wales respectively). In England and Wales the purposes of sentencing are set out in statute. Section 142(1) of the Criminal Justice Act 2003 provides that any court dealing with an adult offender in respect of his offence must have regard to the following ‘purposes of sentencing’: (a) the punishment of offenders; (b) the reduction of crime (including its reduction by deterrence); (c) the reform and rehabilitation of offenders; (d) the protection of the public;13 and (e) the making of reparation by offenders to persons affected by their offence. A separate statutory framework of aims exists for sentencing those aged under 18. Under section 142A(2) of the 2003 Act, a court sentencing a young offender is required to have regard to: (a) the principal aim of the youth justice system, which is to prevent offend­ ing and reoffending; (b) the welfare of the offender; and (c) the ‘purposes of sentencing’,

10  A comparison of analytical, scientific knowledge (episteme) and technical knowledge (techne) with practical wisdom (phronesis), as applied to the task of sentencing, is given later in the chapter. 11  Gemmell v HMA 2012 JC 223 at [59]; Ferguson v HMA 2014 SCCR 244 at [103]. 12  R v Willaert (1953) 105 CCC 172 at 176. See also Chapter 3. 13  The sentencing legislation in most Australian jurisdictions contains statements of the purposes of sentencing and typically includes matters such as deterrence, community protection, denunciation and rehabilitation (for a full discussion, see Mackenzie and Stobbs, 2010: 42, fn 14 and the statutory provisions cited therein; Bagaric and Edney, 2016: Ch 7; Edney and Bagaric, 2007: 5 and Ch 3; Krasnostein, 2015a: 215–17; and Mackenzie, 2005: Ch 4).

114 

The Practical Wisdom of Sentencing

which mirror those for adult offenders14 (section 142A(3)). O’Malley (2016: 44–45, 2011: 238) notes that sentencing statutes in other common law countries frequently include a similar menu of penal purposes. These aims are familiar to all sentencers and have always influenced sentencing decisions (Tarling, 2006: 39). As Lord Igor Judge observes, all these purposes have to be balanced in each sentencing decision; however, the purposes of sentencing in section 142 are not described as having any particular priority over each other (Judge, 2015: 263). Whilst section 142 provides a list of sentencing purposes, it does not provide a hierarchy between them (Dingwall, 2008: 402; Koffman, 2006: 296; Tarling, ibid; Tiarks, 2016: 37, 58; and von Hirsch and Roberts, 2004: 642). What is more, the purposes are not always consistent: reform, rehabilitation and pun­ ishment do not always go together and are not always likely to produce the same conclusion (Judge, ibid; see also Padfield, 2013: 33, and Tiarks, ibid 65, 74, and 117). In short, it is ‘as plain as a pikestaff ’ that these purposes may conflict in their application to any given case (Ashworth and Player, 2005: 825). In criticising section 142 as ‘­fatuous’, Ashworth explains that ‘it is simply a list that provides no means of resolving the inevitable conflict between purposes, and establishes no order of priority’ (Ashworth, 2004: 528). For Lord Judge, three essential factors to weigh in the balance in any sentencing decision are: (i) the crime itself and how it is perceived in contemporary society; (ii) the victim and the impact of the crime on that victim; and (iii) the accused: the circumstances in which he or she came to commit the crime, and his or her attitude to the offence (Judge, ibid). As Judge also notes, not all of these three factors are objective, and indeed two of them include at least an element of the subjective (ibid; see also Duff, 2005: 1178; Millie et al, 2007: 256–57; and O’Malley, ibid 239–41). There is, in fact, an enormous amount of information which has to be taken into account by a sentencer in deciding what sentence to impose: first, the facts of the case itself; second, the general aims of punishment and the current attitudes of the legislature, judiciary and the public towards the appropriate balance between the retributive, deterrent and rehabili­ tative goals of the criminal law; third, information about the offence itself, including the harm done, the prevalence of the particular offence and the offender’s degree of participa­ tion; fourth, information about the offender’s personal characteristics, including age, sex, character, mental health, family and employment circumstances, and general social history; fifth, issues relating to how the offender has responded to the charges, including displays of remorse, whether he has assisted the authorities, and whether there has been any confisca­ tion of property; sixth, the effect of the sentence on the offender; seventh, factors that do not bear directly on the offence or offender but which relate to the investigatory or court processes themselves, such as delay or the availability or otherwise of treatment services for the offender; and eighth, the question of whether the court ought to show mercy to an offender when all other legally relevant factors have been taken into account (see Freiberg, 2014: 219–20). In her study of matters that appear to influence the courts at sentencing, ­Shapland, for example, identifies some 229 factors of relevance (1981: 55), whilst in a study of ­Magistrates’ Courts in the State of Victoria, Douglas identifies 292 such factors (Douglas, 1980: 55, cited in Freiberg, ibid 220). Thus, as the ­Victorian Court of Criminal Appeal observed in

14 

With the exception of the purpose of the reduction of crime.

Value Pluralism in Sentencing

 115

R v Williscroft: ‘The purposes of p ­ unishment are manifold and each element will assume a different significance not only in different crimes but in the individual commission of each crime’.15 Dingwall and Tarling both note that given the number of relevant s­ entencing considerations and the lack of a specified hierarchy, it seems reasonable to assume that, consciously or subconsciously, sentencers will have to set their own priorities and develop their own interpretations in sentencing particular cases (Dingwall, ibid; Tarling ibid; see also O’Malley, 2016: 44; and Tiarks, ibid 71, 119). For Ashworth, section 142 of the Criminal Justice Act 2003 is thus ‘self-evidently­a ­recipe for inconsistency of approach’ (Ashworth, ibid 529). Padfield also acknowledges that section 142 may well have led to greater inconsistency as it allows the sentencer to choose from the list of sentencing purposes: ‘“in this case, reform and rehabilitation is more impor­ tant”, “in this case, public protection gets priority”’ (Padfield, ibid). For Padfield, however, the difficulty lies not in the fact that section 142 embodies, in Ashworth’s terms, ‘the worst of “pick-and-mix” sentencing’ (Ashworth, 2015: 82), but rather that the statutory scheme makes it difficult to assess consistency and the efficacy of measures aimed at aiding such consistency, such as sentencing guidelines16 (Padfield, ibid 33–34). In particular, Padfield (an academic lawyer, barrister and part-time recorder) considers it ‘perfectly acceptable that different aims should take priority in different cases’ (ibid; see also O’Malley, 2016: 44–45 and 2008a: 9–10). Similarly, the Sentencing Guidelines Council’s guideline ‘Overarching Principles: Seriousness’ (SGC, 2004) expressly directs sentencers to decide for themselves between the five potentially competing aims of sentencing in section 142 of the 2003 Act: The Act does not indicate that any one purpose should be more important than any other and in practice they may all be relevant to a greater or lesser degree in any individual case—the sentencer has the task of determining the manner in which they apply (SGC, ibid 3, emphasis added).

In what can be seen as a classic statement of Berlinian value pluralism, the Supreme Court of Canada noted in R v Nasogaluak that: The determination of a ‘fit’ sentence is, subject to some specific statutory rules, an individualized process that requires the judge to weigh the objectives of sentencing in a manner that best reflects the circumstances of the case. No one sentencing objective trumps the others and it falls to the sentencing judge to determine which objective or objectives merit the greatest weight, given the particulars of the case … The judge’s discretion to decide on the particular blend of sentencing goals and the relevant aggravating or mitigating factors ensures that each case is decided on its facts […].17

The task of determining which purpose or purposes to apply in any given case is one for the judge and the judge alone because, through his or her professional experience and legal training in the values set out in section 142, the judge knows what to look for when faced

15 

R v Williscroft [1975] VR 292 at 299. The use of sentencing guidelines in various jurisdictions is discussed in Chapter 6. 17  R v Nasogaluak [2010] 1 SCR 206 at [43], emphasis added and citations omitted; see also R v Lacasse­ [2015] 3 SCR 1089 at [1] and R v Rushby [1977] 1 NSWLR 594 at 597. See also R v Hamilton (2004) 186 CCC (3d) 129 and the judgment of McHugh J in Markarian v The Queen [2005] HCA 25 at [72], both considered in Chapter 3, and the decisions in R v Rushby, ibid at 597; Veen v The Queen (No 2) (1988) 164 CLR 465 at 476; and Elias v The Queen [2013] HCA 31 at [27], all considered in Chapter 4. 16 

116 

The Practical Wisdom of Sentencing

with a new case. As a result, the judge is adept at ‘sorting out’ the new situation and is able to single out the features of ethical relevance (Nussbaum, 2001: 315). Thus it is up to the sentencing judge to properly weigh the various sentencing principles and objectives whose relative importance will necessarily vary with the nature of the crime and the circumstances in which it was committed.18 There is no list of statutorily defined sentencing purposes in Scotland. Although in HMA v AB19 the Lord Justice General observed that the purposes of sentencing set out in section 142 of the 2003 Act ‘are not directly applicable in Scotland’, sentencers in Scotland are nevertheless guided by exactly the same philosophical principles of punishment as those of other common law jurisdictions (Hutton, 1999: 166). For example, over 35 years ago the purposes that guided Scottish sentencing were stated to be desert or retribution, public protection, deterrence, denunciation, rehabilitation, restitution, reduction of crime, and parsimony or economy of resources (Nicholson, 1981: 201–06; see also Goring, 2008: 29; Lothian, 2004: 44; and Nicholson, 1992: 178–82). Thus it may be assumed that all the tra­ ditional purposes of punishment apply by virtue of the common law; indeed each of them has occasionally been recognised in Appeal Court jurisprudence. Thus in sentencing an offender, the judge will have regard to what Ashworth (ibid 83) describes as six contemporary rationales of sentencing, namely deterrence, rehabilitation, incapacitation, desert, reparation or restoration, and social theories20 which emphasise the social context of offending.21 In the course of his or her career on the Bench, and as a sen­ tencer in particular, the judge is faced with what Berlin terms ‘the variety of life—the differ­ ences, the contrasts, the collisions of persons and things and situations, each apprehended in its absolute uniqueness’ (Berlin, 2013a: 466). It is trite that in applying these rationales the judge is also required to have regard to the relevant aggravating and mitigating circum­ stances of the particular case22—the ‘dark mass of factors’ that constitutes human existence (Gardiner, 1997: xv). Considered as a whole, these purposes, considerations and factors are, or have the poten­ tial to be, in Berlinian terms, plural and incommensurable values. If the aims and purposes of sentencing are indeed plural and incommensurable (and the sentencing case law from the UK and other common law jurisdictions gives every indication that this is indeed the case) then how can the sentencer choose rationally between the relevant purposes when, as is so often the case, they conflict? Berlin explains: The way out must … lie in some logically untidy, flexible and even ambiguous compromise. Every situation calls for its own specific policy, since ‘out of the crooked timber of humanity’, as Kant once remarked, ‘no straight thing was ever made’. What the age calls for is not … more scientific

18 See

Lacasse, ibid at [54]. 2016 SCCR 47 at [13]. as the ‘mixed theory’ advocated by Hudson (1995) which seeks to acknowledge ‘both the social pur­ poses of punishment and the realities of the population who find themselves caught up in the penalising circuit’ (ibid 75); the work of Lacey (1998) which focuses on the state’s duty to foster a sense of community amongst its citizens with punishment being justified as reinforcing the values that it has been decided will be protected through the criminal law; and Braithwaite and Pettit’s (1990) ‘republican theory’ of criminal justice. 21  For a full discussion of each rationale, see Ashworth, 2015: 83–100; Freiberg, 2014: 250–68; O’Malley, 2016: 32–44; von Hirsch et al, 2009: Chs 1 to 4; and Walker and Padfield, 1996: Ch 9. 22  For full discussions of aggravating and mitigating factors in sentencing, see Ashworth, 2015: Ch 5; Bagaric and Edney, 2016: Chs 8 and 9; Edney and Bagaric, 2007: Chs 6 and 7; Manson, 2001: Ch 7; O’Malley, 2016: Chs 6 and 7; Roberts, 2011b; Walker, 1999; and Walker and Padfield, 1996: Ch 4. 19 

20  Such

Value Pluralism in Sentencing

 117

organisation. Rather it is the opposite—less Messianic ardour, more enlightened scepticism, more toleration of idiosyncrasies, more frequent ad hoc measures to achieve aims … What is required is a less mechanical, less fanatical application of general principles, however rational or righteous, a more cautious and less arrogantly self-confident application of accepted, scientifically tested, ­general solutions to unexamined individual cases (Berlin, 2002b: 92, citing ‘Elender Behelf ’ in the Critique of Practical Reason: Kant’s gesammelte Schriften, vol 8, p 23, line 22).

Berlin thus considers that the right policy for dealing with individual, concrete cases cannot be arrived at in a mechanical or deductive fashion (Berlin, 2002a: 47). For Berlin, ‘there are no hard-and-fast rules to guide us; conditions are often unclear, and principles incapable of being fully analysed or articulated’ (ibid; see also Nussbaum, 1990: 97). Berlin suggests that we must be guided by ‘the general pattern of life in which we believe’; in making such hard choices we simply ‘do the best we can’ (ibid; see also Berlin, 2013c: 16; Crowder, 1994: 303 and Hausheer, 2013: xlv). We just have to ‘plump in some sort of way’ (Berlin, 1975, cited in Crowder, 2004: 139). On the value pluralist account, what the sentencer should do is purely a matter for judgment in the particular case (Crowder, 2002: 7). This is a world of moral conflict, of ‘disagreement and dilemma’ (Crowder, 2007: 208, 2004: 130). As in other forms of decision making, sentencing is an exercise that requires trade-offs, compromises and moderation (Crowder, 2004: 93, 1994: 300). Rules, values and principles must yield to each other in varying degrees in specific situations (Berlin, 2013c: 15). In short, in sentencing ‘one cannot have everything, in principle as well as in practice’ (Berlin, ibid 14). Berlin’s pluralist outlook therefore dictates that fundamental value conflicts in sentenc­ ing between, for example, retribution and rehabilitation, or denunciation and the applica­ tion of mercy should be seen for what they are, with all the difficulty which that entails. Different values or ends will have strong claims in different circumstances (Crowder, 2004: 115): a value which has more weight in one set of circumstances may have less in another (Berlin and Williams, 1994: 307; see also Judge, 2015: 265–66). To paraphrase Crowder (ibid 74), discretion is discretion, consistency is consistency, and justice is justice: no amount of one wholly compensates for the loss of another. For the pluralist sentencer, such con­ flicts should not be ignored or disguised by the rhetoric of consistency or uniformity in sentencing. On a monist view of sentencing, however, it could be conceded that values such as denunciation, retribution, rehabilitation and mercy are deeply incompatible, and that con­ flicts amongst them are real and unavoidable, yet still argue that all such conflicts have a single, correct resolution. The monist could argue that even if values cannot be commen­ surated they can still be ranked, so that the superior good always overrides the subordinate. It might be claimed, for example, that aggravating features of a particular offence should always outrank, or trump, any personal mitigation. The monist might also claim that there is a single, rationally defensible procedural rule that can be used to resolve value conflicts without recourse to the ranking of values that is inherent in the pluralist approach (see gen­ erally Crowder, ibid 127–28). By recourse to the single procedural rule or overarching prin­ ciple, the monist can accommodate conflicting values into ‘an all-things-considered order’ (Crowder, 1994: 294). It is my claim that this monist approach accords with contemporary sentencing practice in England and Wales through the use of the formal, prescriptive, and presumptively binding sentencing guidelines issued by the Sentencing Council.23 23 

See Chapter 6.

118 

The Practical Wisdom of Sentencing

These accounts are rejected by pluralist sentencers—or, in Berlinian terms, judicial ‘­foxes’—such as Scottish, Canadian and Australian judges. The rational pluralist judge ­cannot rely on ready-made monist procedures to resolve the deep moral conflicts inherent in sentencing (Crowder, 2004: 167). As Duff, for example, explains: Whether we are dealing with goods such as justice, truth, and economic well-being; or with evils such as cruelty, dishonesty, and hatred; or with wrongs such as murder, rape, theft, and perjury; or with punitive burdens such as imprisonment, fines, community service orders, and different types of probation, we cannot represent them all as involving different quantities or degrees of some ­single basic value, nor can we formulate a mechanism that would allow us to compare or rank them all on a single scale of value (Duff, 2005: 1176).

The rational pluralist judge must go behind the perspectives of such monist procedures to weigh the values they embody for himself (Crowder, ibid). In order to ‘wisely blend’ the various penal aims in any given case, sentencers must have sufficient discretion to, within agreed limits, go their own way in deciding matters ‘for which there is no abso­ lute b ­ lueprint’ (Crowder, 2002: 90; see also 227–28 and Berlin, 2013b: 196). In this regard ­Scottish, C ­ anadian and Australian sentencers—Berlinian foxes all—must be ‘free, autono­ mous, unpredictably creative, self-interpreting­’ decision makers (see Hausheer, ibid xliv). The wide range of sentencing aims implies a concomitantly wide range of legitimate permutations of those aims, that is, of reasonable rankings. There is consequently room for reasonable sentencers to disagree on reasonable grounds (Berlin and Williams, 1994: 307; Crowder, 2004: 167; Duff, ibid 1177). For example, appellate courts frequently hear conflict­ ing submissions in Crown appeals against sentence from the prosecution and from counsel for the respondent which may all seem reasonable up to a point. The submissions for the Crown and for the respondent may all appeal to one or more important aspects of our individual or collective goods, or to certain of our deep seated moral beliefs. It is the court’s duty to accommodate these conflicting submissions by doing justice in the individual case, rather than by transcending them by recourse to a monist super-value such as consistency (see Crowder, 2002: 99). In HMA v V,24 for example, the respondent (a first offender) was convicted after trial of murdering his wife in the family home, following his discovery that she was having a homo­ sexual affair with a close family friend. The respondent had stabbed the deceased four times with a kitchen knife. The trial judge imposed the mandatory sentence of life imprisonment. In terms of the relevant legislation,25 the judge was required to set the ‘punishment part’ of the sentence, being the period considered appropriate to satisfy the requirements of retribu­ tion and deterrence (taking into account the seriousness of the offence) which the respond­ ent must serve before being considered eligible for parole. Sentencing guidelines issued five years earlier by the Appeal Court in HMA v Boyle26 stated that other than in exceptional circumstances, punishment parts of at least 16 years (or significantly longer, depending on the circumstances) should be imposed for murders involving the use of knives. In V’s case, however, the trial judge set the punishment part at 12 years as he considered the murder to

24 

2015 SCCR 50. Section 2(2) of the Prisoners and Criminal Proceedings (Scotland) Act 1993. 26  2010 JC 66 at [16]. 25 

Value Pluralism in Sentencing

 119

have been ‘a unique episode which occurred in very unusual circumstances’.27 In p ­ articular, the judge had regard to the respondent’s previous status as an upstanding, law abiding member of the community (he had been a career civil servant with no previous involve­ ment in the criminal justice system); to the fact that the incident had occurred at a time of immense stress in the respondent’s life; that news of the deceased’s affair had come com­ pletely out of the blue and in distressing circumstances (the couple’s son had inadvertently accessed intimate e-mail correspondence between his mother and her lover on his new iPad); and that the respondent had initially obtained the knife used to murder the deceased with a view to killing himself. The Crown appealed on the grounds of undue leniency submitting, inter alia, that the trial judge had erred in determining that the case was ‘exceptional’ such as to justify departure from the general guidance given by the Court in Boyle.28 In refusing the Crown’s appeal, the Appeal Court considered that the sentencing judge had not sought to reduce the wickedness of the respondent’s acts but had appropriate regard to the whole circumstances of the murder and to the respondent’s personal circumstances, in particular to the strong mitigatory ­circumstances.29 Looking at the case in the round, the Court was satisfied— ‘with some hesitation’—that although the sentence was lenient, the circumstances were such that it was open to the trial judge to select a punishment part at a lower level than might be expected in a case of murder involving the use of a knife.30 The considerations bearing upon sentence in HMA v V are typical of the conflicting ­values with which the courts must grapple, albeit in the context of a particularly serious and tragic case. The Crown’s submissions that the punishment part of 12 years failed to reflect the seriousness of the offence were, up to a point, perfectly reasonable given the use of a lethal weapon to inflict multiple wounds on the deceased, the commission of the offence in a domestic context, and the respondent’s failure to seek medical attention for the victim. Nevertheless, the compelling mitigatory factors—including the respondent’s previous good character, the stress he was under at the time of the offence and his own state of mind— were such that an admittedly lenient sentence well below that indicated by the guideline judgment in Boyle, was held to be within the range open to the sentencing judge.31 Thus on a value pluralist view, the conflicts between sentencing purposes such as desert or retribution, deterrence, denunciation and rehabilitation are not all ultimately r­ esolvable in the way that monists suppose (Crowder, 2002: 6). The best explanation of such conflicts is that they result from the diversity of goods, each of which makes its own claims on the sentence—claims that cannot be translated into any other terms (ibid). Typically, as Galston observes in a different context, these considerations cannot be reduced to a single, common measure and nor will there be any obvious way to give one moral claim priority over the others (Galston, 1999: 880). When the various sentencing purposes are recognised as incom­ mensurable, it follows that each has, in a rough sense, an equal claim on the sentencer’s attention. Only in particular circumstances does it seem possible to rank plural values

27 

HMA v V (n 24) at [13]. Boyle (n 26). 29  HMA v V (n 24) at [24]–[25]. 30  HMA v V (n 24) at [25]. 31  See also the unsuccessful Crown appeals against sentence in HMA v Q 2013 SCL 440; HMA v Smith 2014 SCCR 39; and HMA v Graydon 2016 SCL 418. See also HMA v Leith 2016 SCL 331. 28 

120 

The Practical Wisdom of Sentencing

such as the purposes of sentencing. In these cases the sentencer’s reasons for so ranking must, crucially, attend to the particulars of the case in hand, rather than rely on an abstract ­super-value or formula (Crowder, ibid 96).

The Pluralist Virtues in Sentencing Crowder argues that to cope well with choices amongst incommensurables is to exercise a specific set of skills or virtues, virtues which provide further guidance to pluralist prac­ tical reasoning (Crowder, 2004: 141, 2002: 185). As pluralism imposes hard choices, the sentencer must develop certain dispositions or traits of character. A person confronted by a hard choice amongst incommensurables will be assisted in choosing well if he or she informs their reflection with certain attitudes of mind, or ‘virtues’ (2004: 166). On Crowder’s account, the four pluralist virtues comprise generosity or open minded­ ness, realism, attentiveness, and flexibility. Whilst Crowder defines g­ enerosity as an ability to acknowledge that many goods and ways of life are objectively valuable, involving respect for the full range of human goods and lives, including those we cannot accommodate within our own decisions (2004: 166, 2002: 189–90), it is arguably the latter three virtues that are of particular importance in the sentencing context. The second pluralist value—realism—involves ‘a feeling for the real costs of moral and political decisions, conditioned in particular by the implications of incommensurability’ (Crowder, 2004: 166). The quality of mind required for this aspect of the pluralist outlook is, Crowder explains, honesty or even courage, for those sentencers who genuinely adopt a pluralist point of view cannot fool themselves that conflicts between the purposes of sentencing can be resolved without cost (see Crowder, 2002: 190). Thus judges must be realists in the sense that they are required to face the depth and permanence, as well as the pervasiveness, of value conflict in sentencing (ibid). The third virtue—attentiveness—involves the decision maker being attentive to the rele­ vant details of the choice situation, including the claims and circumstances of those affected by the choice. This virtue, Crowder explains, is required due to the pluralist rejection of neat abstract rules and its insistence on paying close attention to the particulars of the con­ crete situation (2004: 166; see also Berlin, 2013c: 10 and 14). Crowder elaborates: It is only by specifying the precise facts and values that constitute the context for choice that the pluralist chooser can specify what is most important to him or her in the situation. Only then can the chooser arrive at reasons to subject the contending considerations to some kind of ranking (2002: 191).

For Crowder, the idea of particularist attentiveness can be seen to have three different, yet closely connected, aspects. Applied to the sentencing context, we see that, firstly, attention must be paid to the distinctive character of the different goods involved in a choice situation. Secondly, the sentencer must attend to the distinctive particularity of the situation, which is constituted partly by the values at stake, but also by the relevant facts of the case. Thirdly, the sentencer must attend to the individual persons involved in the situation (for example, the offender, the victim, and their families). The sentencer must attend to their claims and to their needs (ibid). Finally, the fourth value—flexibility—is required since, as pluralism requires that the decision maker attends to the particularities of concrete situations rather than insisting

Value Pluralism in Sentencing

 121

always on the application of general rules, he or she must be able to respond to those ­situations (ibid 198). In the absence of decisive monist rules, pluralists need to be flexible in ‘tailoring their judgement closely to the situation to which they attend’ (Crowder, 2004: 166). In deciding upon the appropriate sentence in any given case, the judge must not insist on trying to resolve the situation by rigid application of a general rule; rather, he or she must be prepared to reach a balance between general rule and particular judgment tailored to the circumstances (see Crowder, 2002: 191). This does not mean that rules are irrelevant and that the pluralist sentencer’s decisions are arbitrary or ad hoc. General rules such as appellate sentencing guidance or the form of sentencing guidelines handed down by the appellate courts in Scotland, the Republic of Ireland and several Australian jurisdictions32 have useful roles to play so long as they ‘keep their place’ (Nussbaum, 1990: 73). In particular, such guidelines comprise useful summa­ ries of decision making experience (­Crowder, ibid 192). Such sentencing guidance is itself the product of reflection on concrete situations (ibid 197) or, as Nussbaum puts it, ‘a sum­ mary of wise decisions’ (Nussbaum, ibid 69; see also Nussbaum, ibid 75, 84–85, 97 and 104, and 1993b: 258) which assists the judge in identifying the salient features of the particular case (Nussbaum, 2001: 299 and 304–05; 1990: 73). Yet, as Crowder explains: [R]ules will not by themselves determine particular decisions. Rather, such decisions will require a process of ‘interplay’ or ‘conversation’ between the general rules and concrete particulars in which each may be modified by the other … The pluralist chooser must be prepared to balance back­ ground commitments, including those summarized in the form of general principles … against considerations brought forward by attention to the concrete situation (ibid 192, emphasis per original).

The pluralist sentencer must recognise, like the trial judge in HMA v V,33 that general rules contained within appellate sentencing guidelines are open to modification in the right set of circumstances (Nussbaum, 1993b: 258). Whilst such guidelines are useful for a variety of purposes, they remain valid only to the extent to which they correctly describe good con­ crete judgments, and they are to be assessed, ultimately, against these (Nussbaum, 1990: 68). On Nussbaum’s account, there is still room for the judge, in applying any set of sentencing guidelines, to recognise as ethically salient any new or surprising features of the particular case before him—features that have not been anticipated in the guidelines, or even features (perhaps those relating to the personal circumstances of the offender, the victim, or both) that could not in principle be captured in any rule (Nussbaum, ibid 68–69). Thus pluralist sentencers must cultivate the ability to perceive and correctly describe the particular case before them finely and truly; their perceptual grasp must include even those features of the case that are not covered under existing sentencing guidelines ­(Nussbaum, 1993b: 258). By so doing, the pluralist sentencer captures the ‘fine detail of the concrete ­situation’ by confronting it ‘as a complex whole’ (Nussbaum, 2001: 300–01). In this way, pluralist sentencers respond to ‘nuance and fine shading, adapting [their] judgment to the matter at hand in a way that principles set up in advance have a hard time doing’ ­(Nussbaum, ibid 301).

32  33 

Discussed in Chapter 6. HMA v V (n 24).

122 

The Practical Wisdom of Sentencing

The problem that value pluralism presents for reasoned argument in general is why one particular value should be ranked ahead of another or, in a broader sense, how one can choose rationally amongst options given value incommensurability (Crowder, 2002: 10, 56–57). For the value pluralist, the kind of ranking that becomes a problem because of incommensurability is what Crowder describes as abstract ranking, or ranking irrespective of context: If values are incommensurable in the relevant sense, then it seems on the face of things that we shall have no reason to rank them the same way in every case or even in most cases … Under value incommensurability it looks as though we shall have no good reason to construct hierarchies of values in the abstract—hierarchies abstracted, that is, from particular cases and contexts … [R]eason seems to permit multiple legitimate rankings of plural values depending on the circum­ stances (Crowder, ibid 53).

In sentencing, as in other forms of decision making, specification of context reveals the values that are most important, hence the values that guide choice. Conversely, to get clearer about the values guiding the sentencer’s choice in a particular case involves the sentencer specifying the choice situation or context (Crowder, ibid 57–58). This may be done, for example, by the sentencer’s remarks when passing sentence, by the issuing of a sentencing statement or, in jurisdictions such as Scotland and Victoria, by the report on the case pre­ pared by the sentencer for the appeal court in the event of the disposal being appealed (see respectively McCluskey and McBride, 2000: 55–56 and Freiberg, 2014: 169–71). Attention to the particular context of a decision involves the specification of both the relevant facts, which delimit possibilities, and the relevant values, which provide a standard for deciding what matters most in any given set of circumstances; that is, for deciding how to rank the values that concern the sentencer in the particular case (Crowder, ibid 62). Thus for the pluralist sentencer, guidance can be sought in the features of the particular context (ibid). It is only through this kind of particularism that specific rational choices are possible under pluralism (ibid 63).

Pluralism and the Need for Autonomy in Decision Making As Crowder explains, value incommensurability implies a particularist approach to ethics. In sentencing, the pluralist judge is required to decide value-related questions concerning the type of disposal (a custodial or non-custodial sentence) and the quantum of punishment (how many years’ imprisonment; how many hours of unpaid work as part of a community payback order; what level of fine) not by attending to the guidance of abstract rules—or to a general rule which is regarded as authoritative across many different situations—but rather by attending to the particular circumstances of the case (see Crowder, 2007: 210 and 2002: 53). In sentencing, attention to the context of the individual case enables the judge to specify the elements of that context, its constitutive facts and values, and so to specify what mat­ ters most in that particular case: which values should come first and in what combination (Crowder, 2002: 188). In this way, in any given case the sentencer can weigh the values at stake against one another and rank them without recourse to impossible or superficial commensuration: Pluralists … are obliged … to think for themselves in a strong sense. Ideally, they should be ­prepared to deal with each choice on its own terms, weighing all relevant competing considerations … They

Value Pluralism in Sentencing

 123

should be able to stand back from received rules and customs, recognise the value rankings these embody, and critically assess their application in the circumstances. This may involve appeal to background values such as personal and collective conceptions of the good, but these too should be subject to revision. Pluralists ought, that is, to be capable of autonomy when they are faced with such fundamental conflicts (Crowder, 2007: 225).

In order to choose rationally amongst such plural values, the sentencer is required to be able to choose autonomously, or critically (ibid). Adoption of the pluralist outlook commits the sentencer to meeting the necessary conditions for making hard choices. Prominent, if not pre-eminent, amongst those conditions is personal autonomy: it is not choice alone that imposes this requirement, but the peculiarly demanding nature of choice under value pluralism (Crowder, 2002: 210; Hausheer, 2013: xlii). In other words, the case for autonomy is grounded not in the mere necessity of choosing amongst incommensurables, but rather in the goal of choosing well (Crowder, 2007: 225). It is for this reason that the Scottish courts—and indeed the appellate courts in many other common law jurisdictions—have eschewed systems of formal, prescriptive and presumptively binding sentencing guidelines, with the present system of such guidelines in England and Wales being unique in the com­ mon law world (Krasnostein, 2015a: 165; O’Malley, 2013a: 219–20; Roberts and Ashworth, 2016: 309–10; Wiles, 2015: xi).34 The importance of personal autonomy for the pluralist decision maker is reflected in the very wide sentencing discretion traditionally enjoyed by Scottish judges (Hutton, 1999: 173; Moody, 1995: 219–20); by the absence of any concept of a ‘tariff ’ or ‘going rate’ for specific offences in Scotland (Hutton and Tata, 2010: 272); and by the Scottish Appeal Court’s ­historic reluctance to issue any form of sentencing guidelines (AN Brown, 2010a: 162; ­Hutton, 2008: 144; Nicholson, 1992: 177; Sentencing Commission for Scotland, 2006: 15; Tata, 2013: 238). Thus for Scottish judges, a strong form of autonomy, or sentencing discretion, has been necessary in order to cope rationally with the exigency of choices which cannot be decided by the straightforward application of simple rules such as prescriptive, numerical sentencing guidelines (see Crowder, ibid 211).35 For C ­ rowder, such autonomy, or, in the sentencing context, judicial discretion, can also be seen to enhance the pluralist virtues. As autonomy answers to pluralist flexibility, required by the need to respond to the ­particulars of the given case, it follows that autonomy also reinforces the pluralist virtue of attentiveness to the details of the situation (the particular offence) and of the persons involved in it (including the offender and his dependents, the victims and their families— see Crowder, ibid).

Pluralism and Practical Wisdom As Crowder notes, the mere fact that values are plural tells us nothing about which of the vast range of values known to us from human experience are the values we ought to choose in any given situation: pluralism ‘tells us that we must choose but not what to choose’ (Crowder, 1994: 303). In the sentencing context, as we have seen, although various ­common

34  35 

See the discussion in Chapter 6. See the discussion in Chapter 6.

124 

The Practical Wisdom of Sentencing

law jurisdictions have statutory statements of sentencing purposes which are required to be b ­ alanced in the individual case, little or nothing is said about how the rationales of ­deterrence, rehabilitation, incapacitation, desert, reparation or restoration, and various social theories are to be reconciled (see Ashworth, 2015: 83; and Judge, 2015: 263). Yet, value pluralism does not exclude reasoned value judgment (Crowder, 2002: 57). Under pluralism, practical reason is needed to organise choices among incommensurable goods. In the absence of such practical reasoning, sentencers’ decisions would indeed be arbitrary, incoherent and perhaps even self-defeating (Crowder, 2004: 165). In order for the sentencer to cope adequately with choices under pluralism—to impose a just sentence—he or she must cultivate the kinds of skill that particularistic practical reasoning requires (Crowder, 2002: 63, 73, 196). In short, as the purposes of sentencing are plural and incommensurable, then when they come into conflict in particular cases the sentencer can choose rationally amongst them only by engaging in the particularistic kind of practical reasoning described by the Aristotelian tradition (Crowder, ibid 187; see also Annan, 2013: xxx; Crowder, 2007: 226; and Nussbaum, 2001: 294). Thus, although Berlin does not specifically make the connection, the Berlinian account of the superiority of value pluralism over moral monism accords with the Aristotelian concept of phronesis.

The Ideals of Episteme, Techne and Phronesis In presenting his thesis that context and judgment are irreducibly central to the under­ standing of all human action, Bent Flyvbjerg (2001) offers a contemporary interpretation of the Aristotelian concept of ‘phronesis’. We no longer have a word for this intellectual virtue, one which Aristotle and other founders of the Western tradition regarded as a nec­ essary condition of successful social organisation, and the most important prerequisite to such organisation (Flyvbjerg, 2006: 371). The accepted translation, however, is ‘prudence’, ‘practical wisdom’, ‘practical common sense’ (Flyvbjerg, 2001: 2, 56; Schram, 2012: 16) or ‘practical judgment’ (­Steinberger, 1993). Phronesis must be distinguished from the Aristotelian concepts of ‘episteme’ and ‘techne’. Episteme—found in the modern words ‘epistemology’ and ‘­epistemic’ (Flyvbjerg, ibid 3)— can be regarded as analytical, scientific knowledge. It is context-independent knowledge that concerns universals. Epistemic knowledge is achieved with the aid of analytical ration­ ality. As a concept, it corresponds to the modern scientific ideal as expressed in natural science (Flyvbjerg, ibid 55–56; 2006: 371). Flyvbjerg notes that this ideal has come close to being the only legitimate view of what constitutes genuine science. For Flyvbjerg, intel­ lectual activities such as social science are not, and probably can never be, scientific in the sense of episteme. The dominance of epistemic knowledge, however, has resulted in the social sciences being compelled to strive for, and to try and legitimate themselves in terms of, this ideal (Flyvbjerg, 2006: 371, 2001: 56; see also Nussbaum, 2001: 302 and 1990: 55). Techne—found in the modern words ‘technology’ and ‘technical’ (Flyvbjerg, ibid 3)—meanwhile, is a form of technical knowledge or know-how. It can be translated into English as ‘art’, in the sense of ‘craft’. As an activity, it is concrete, variable and contextdependent: the objective of techne is the application of technical knowledge and skills according to what Flyvbjerg terms a ‘pragmatic instrumental rationality’ (ibid 56).

The Ideals of Episteme, Techne and Phronesis

 125

Phronesis, however, goes beyond these concepts. Korthagen and Kessels contrast the knowledges of episteme and phronesis thus: [E]pisteme aims primarily at helping us to know more about many situations, while the emphasis of phronesis is mostly on perceiving more in a particular situation and finding a helpful course of action on the basis of strengthened awareness (1999: 7, emphasis per original).

If episteme denotes theoretical ‘know-why’ and techne denotes technical ‘­ know-how’, ­phronesis emphasises practical knowledge and practical ethics (­Flyvbjerg, ibid 56; Schram, ibid 16). It is, as Berlin explains, a capacity for seeing ‘what fits with what, [and] of what can­ not exist with what’ (Berlin, 2013a: 485). Phronesis is best understood as embodied judge­ ment linking knowledge, virtue, and reason: it is not simply a form of knowledge but an amalgam of knowledge, virtue and reasons that enables people to decide what they should do (Coulter and Wiens, 2002: 16). ­Phronesis is thus the most important of the intellectual virtues as it is needed for the management of human affairs (Flyvbjerg et al, 2012a: 1). Coulter and Wiens explain that Aristotle contrasted two conceptions of practice that are often conflated in English: ‘poiesis’ (practice as craft) and ‘praxis’ (practice as moral-political action). They note that while practice as poiesis is means-ends activity in which knowledge and skill (techne) are used to accomplish ends decided by the exercise of theoretical wisdom (or ‘sophia’), practice as praxis aims at a different kind of end: a good and worthwhile life (‘eudaimonia’) where the means are integral to the end. A different form of wisdom is needed in deciding what counts as a good life, in acquiring the requisite knowledge and virtue, and in matching that knowledge and virtue to particular situations correctly: that form of wisdom is phronesis (Coulter and Wiens, ibid). Phronesis is thus based on practical value-rationality. A reflexive analysis and discussion of values and interests is at the core of phronesis (Flyvbjerg, ibid 3). Flyvbjerg notes that even if both phronesis and techne involve skill and judgement, one type of intellectual virtue can­ not be reduced to the other: phronesis, states Flyvbjerg, is about value judgement, not about producing things (2006: 372). He elaborates: The person possessing practical wisdom (phronimos) has knowledge of how to behave in each particular circumstance that can never be equated with or reduced to knowledge of general truths. Phronesis is a sense of the ethically practical rather than a kind of science … [F]or Aristotle [rational persons] are moved by a sense of the proper order among the ends we pursue. This sense cannot be articulated in terms of theoretical axioms, but rather, is grasped by phronesis (2001: 57; see also Flyvbjerg, 2006: 372).

As Berlin explains, practical wisdom is not scientific knowledge, but ‘a special sensitive­ ness to the contours of the circumstances in which we happen to be placed’ (2013a: 489). It requires ‘an awareness of the interplay of the imponderable and the ponderable, of the ‘shape’ of things in general or of a specific situation, or of a particular character’ (ibid 489–90). The phronimos decides what ought to be done in a given situation, how exactly the competing considerations should be balanced, through accumulation of experience in dealing with comparable situations (Crowder, 2004: 141). Experience enables the phronimos to develop skill in practical reasoning (ibid; Richardson, 1997: 184). The phronimos apprehends the best course in any particular given case through his or her experience of ethical questions and responses in the past (Crowder, 2002: 58). Flyvbjerg considers phronesis to be the intellectual activity most relevant to praxis since it focuses on what is variable. Phronesis both focuses on that which cannot be encapsulated

126 

The Practical Wisdom of Sentencing

by universal rules and, crucially for our discussion of sentencing, on specific cases. Phronesis requires an interaction between ‘the general and the concrete’; it requires ‘consideration, judg­ ment and choice’; and, more than anything else, it requires experience (Flyvbjerg, 2001: 57; see also Berlin, 1997: 41, 45; Kronman, 1987: 850; Nussbaum, 2001: 290, 305 and 1990: 68, 74). Thus context and experience are central to the ideal of phronesis. With phronesis, the particular and the situationally dependent are emphasised over the universal and over rules. The concrete and the practical are, Flyvbjerg states, emphasised over the theoretical (ibid 58; 2006: 373; see also Nussbaum, 2001: 300–01). As Coulter and Wiens explain, meaning is found primarily in the particular and not the universal; for example, no universal concept of beauty can be applied to a work of art to judge its worth. Connecting the particular to the general cannot be accomplished by applying principles, rules, laws, virtues or knowledge; instead, the link is made by using the imagination (Coulter and Wiens, ibid 16). In sentencing, for example, Davis argues that in assessing the seriousness of crimes there is no externally observable, objective source of ‘real’ seriousness that we can use to stand­ ardise assessments made by different people (2002: 344). The exercise is qualitative rather than quantitative; the sentencer must focus on what is specific rather than what is general (Berlin, ibid 46). Davis maintains that our assessments of seriousness are more akin to the kinds of aesthetic or moral assessments we make which are judgements or conclusions drawn about events and people; the seriousness of an offence is thus a subjective quality. In the case of concepts such as temperature, we have external, objective physical sources of standardisation: temperature can be measured, for example, by the use of a thermometer. Similarly, we can compare measured units of sugar in ice cream with perceptions of its sweetness. The subjective nature of offence seriousness does not, however, lend itself to such measurement (Davis, ibid 344–45). Davis suggests that the seriousness of any given offence is not like the sweetness of an ice cream or the heat of a day, but is more like the ‘deliciousness’ of an ice cream or the ‘friendliness’ of an act (ibid 345). If there is only ‘perceived’ seriousness, Davis argues, then the process of sentencing cannot be seen as the scientific discovery of the seriousness in a crime and the allocation of a number to it which is then converted into a punishment unit by applying a rate of exchange (ibid). It is, as Berlin puts it, as if we were to look for ‘a theory of tea-tasting, a science of architecture’: the factors to be evaluated are too numer­ ous and everything depends on the sentencer’s skill in identifying and integrating them (­Berlin, ibid 50). A sentencer’s knowledge of the seriousness of a particular crime (or, for that matter, of the prospects of an offender’s rehabilitation or the need for a denunciatory sentence) is knowledge in the sense in which ‘trainers know their animals … or conductors their orchestras, as opposed to that in which chemists know the contents of their test tubes, or mathematicians know the rules that their symbols obey’ (Berlin, ibid 47). Thus, phronesis is flexible, practical and the product of a shared understanding concerned with particular, concrete human beings (Ruderman, 1997: 410). Flyvbjerg notes how the centrality of experience in the decision making process has been glossed over by modern science. For Flyvbjerg, the concrete, the practical and the ethical have, despite their importance, been neglected because (contrary to the ideal of phronesis) current thinking— such as that behind the present form of sentencing guidelines in England and Wales36—dictates that the sciences must concern themselves precisely with the explication of universals (2001: 59). 36 

Discussed in Chapter 6.

From Novice to Expert—the Dreyfus Model of the Human Learning Process

 127

Phronesis, however, is concerned with particulars, and not universals (Nussbaum, ibid 303). Social scientists, criminologists and policy makers have spent too much time focusing on ­episteme and techne to the neglect of phronesis (Landman, 2012: 27).

From Novice to Expert—the Dreyfus Model of the Human Learning Process To demonstrate that human action can be best understood by reference to context and judgement, Flyvbjerg draws on the work of Hubert and Stuart Dreyfus (1986) in an attempt, firstly, to understand the link between knowledge and context and, secondly, to address the question of whether knowledge about human activity can ever be context-independent (Flyvbjerg, 2001: 9). The ideal of phronesis, with its emphasis on cases and context in the understanding of human behaviour, is at the very centre of the model of the human learn­ ing process proposed by Dreyfus and Dreyfus. Flyvbjerg notes that detailed phenomenological studies of human learning indicate that people pass through several phases or levels in the learning of all skills, ranging from the technical (such as building a house) to the intellectual (such as analysing a text) (ibid 10). The Dreyfus model operates with five levels: (i) novice; (ii) advanced beginner; (iii) compe­ tent performer; (iv) proficient performer; and (v) expert. The five levels consist of recognis­ able, qualitatively different ways of acting and performing in the process of learning a given skill. Although individuals at a given level do better than individuals at the previous level, not all people achieve the highest level in a given field (ibid 10–11).

The Novice Actor A novice experiences a given problem and a given situation in a given task area for the first time. Through instruction in the particular task, the novice learns the objective facts and characteristics of the situation that are relevant for the performance of the skill. The novice also learns rules for action by recognising the relevant facts and characteristics as and when they appear (Flyvbjerg, 2001: 11). As Flyvbjerg explains: Facts, characteristics, and rules are defined so clearly and objectively for the novice that they can be recognized without reference to the concrete situation in which they occur. On the contrary, the rules can be generalized to all similar situations, which the novice might conceivably confront. At the novice level, facts, characteristics, and rules are not dependent on context: they are contextindependent (ibid).

A novice will judge his skills by evaluating how well he has followed the rules he has learned. Once the novice has learned a number of rules for a given skill, however, continued improvement of performance is impeded because performance of the skill has, by this time, become more complex and more demanding of concentration. Thus, the first rules are necessary for gaining initial experiences; however, these rules soon become a barrier to the learning process and must be put aside for the novice to advance to the level of advanced beginner (ibid).

128 

The Practical Wisdom of Sentencing

The Advanced Beginner Context is the key factor in the progression from novice to advanced beginner. Whereas the focus at the novice stage is often on deliberative and protected learning situations, it is through achieving real life experience that the second stage of advanced beginner is attained (Flyvbjerg, 2001: 11). The advanced beginner will then learn to recognise relevant elements in relevant situations; such recognition occurs because of the actor’s perceptions of similarity in relation to prior examples of the same situation: Gaining experience consists in a cumulative recognition of similarities. Recognition is concrete and dependent on context, and it is precisely context which plays the decisive role, for it is context which becomes increasingly more important as one proceeds up the levels of the learning process (ibid 12).

At this stage, the actor’s personal experience obtained via trial and error is more important than context-independent, explicit, verbally formulated facts and rules (ibid).

The Competent Performer If the attainment of the rank of advanced beginner is dependent on context, then progres­ sion to the level of competent performer depends on the actor’s ability to interpret and to judge. As the advanced beginner gains experience, he is overwhelmed by the number of recognisable elements he sees in a concrete situation. He lacks a feeling of what elements are important or, in other words, he is unable to prioritise. Flyvbjerg explains how the actor addresses this problem: At this stage, individuals learn from themselves and from others to apply a hierarchical, ­prioritizing procedure for decision-making. By first choosing a goal and a plan with which to organize the information about the concrete situation, and then processing only those factors relevant to achieving the goal and plan, the individual can simplify his or her task and obtain improved results (Flyvbjerg, 2001: 12).

In order to progress, the actor is required to move from the initial, rule-based levels in the learning process to the kind of prioritising behaviour and overview which characterises competence (ibid). An important corollary is that the actor will now be personally involved in their actions—what Flyvbjerg calls ‘a relationship of involvement’ (ibid 13). The novice and the advanced beginner have only limited responsibility for the consequences of their actions, the actions themselves being predetermined by learned elements and fixed learned rules. This means that, unless the novice or beginner makes a gross error, a bad result will appear as having been caused by inadequately specified elements and rules. Flyvbjerg explains that actions and results thus stand in an external relation to the beginner. They can be justified and given a rational explanation in relation to objective facts and the abstract rules which the beginner has been required to employ (ibid). As Flyvbjerg demonstrates, however, the relationship of involvement means that the position is very different for those at the level of competent performer and above: The competent performer, after having struggled with the problem of selecting a plan, feels responsible for the consequences of the choice precisely because selecting a plan cannot be done

From Novice to Expert—the Dreyfus Model of the Human Learning Process

 129

­ bjectively, but must nevertheless be carried out in order to be able to act competently. Hence, the o actions of the competent performer comprise an element of interpretation and judgment … [T]he ability to make these judgments becomes crucial at the upper levels of the learning process. It is this ability … which constitutes the core of true human expertise (ibid—emphasis added).

The Proficient Performer Experience, resulting in continuous and non-sequential decision making, is the hallmark of the actor who has advanced to the next stage in the learning process: that of proficient performer. Flyvbjerg states that proficient performers tend to be deeply involved in their actions; they have evolved their perspective on the basis of prior actions and experiences (Flyvbjerg, 2001: 16). When a proficient performer is faced with a particular situation, cer­ tain key features will stand out, whilst others recede into the background. The predominant features to which the proficient performer has regard, along with his plans and expecta­ tions, are in due course changed by new actions and experiences. This in turn changes the actions of the proficient performer (ibid). As Flyvbjerg explains: No objective choice or conscious evaluation of appropriateness takes place, which is the case in selecting elements, rules, and plans. The choice is simply made … And this seems to happen because the proficient performer has experienced similar situations earlier. Via spontaneous interpretation and intuitive judgment the memory of these situations generates plans corresponding to plans which have worked before. Similarly, memory of earlier situations releases expectations about actions, which correspond to those actions carried out in similar situations earlier” (ibid— emphasis added).

Whilst proficient performance is marked by intuitive understanding and organisation of tasks, Flyvbjerg also stresses the importance of analytical reflection on the perceived outcome: Elements and plans from the performer’s experiences, which appear as intuitively important, are evaluated and combined analytically with the help of rules for reaching decisions about the most appropriate actions. Deep intuitive involvement in performance thus interacts with analytical deci­ sion-making (ibid).

The Expert Actor The final stage in the learning process is that of expert. Flyvbjerg explains the key ­characteristics of the expert actor as follows: In contrast to the competent performer, genuine human experts exhibit thinking and behavior that is rapid, intuitive, holistic, interpretive, and visual and which has no immediate similarity to the slow, analytical reasoning which characterizes rational problem-solving and the first three levels of the learning process. On the contrary, it seems that there is a fundamental and qualitative jump from analytical problem-solving to genuine, human expertise. This jump must be made in order for someone to be really adept at performing a given skill (Flyvbjerg, 2001: 14).

How is this jump made? Flyvbjerg considers that, due to analytical rationality’s slow ­reasoning and the emphasis which it places on rules, principles and universal solutions, an

130 

The Practical Wisdom of Sentencing

exclusive use of analytical rationality tends to impede further improvement in performance. Further, factors including speed and an intimate knowledge of what Flyvbjerg describes as concrete cases in the form of good examples are prerequisites for true expertise (ibid 15). Flyvbjerg elaborates: The proficient performer gradually achieves intimate experience from different situations, all of which touch upon the same goal and the same perspective, but which demand different tactical decisions. The proficient performer then perhaps achieves a level in which it is not only situations, which are recognized intuitively, but also—synchronically and holistically—the relevant decisions, strategies, and actions … [T]his is the level of genuine, human expertise … (ibid 17).

Flyvbjerg notes that in normal, familiar situations, real experts do not solve problems and do not make decisions. They just do what ‘works’. Flyvbjerg stresses, however, that this does not mean that the expert never thinks consciously, nor that he always does the right thing: ‘When there is time, and when much is at stake, experts will also deliberate before they act. Their deliberation, however, is not based on calculated problem solving but on critical reflection over the intuition, which the expert applies’ (ibid). Although the expert will carefully reflect and deliberate when faced with a set of cir­ cumstances where much is at stake, situations remain where his or her decisions do not work (ibid). Nevertheless, experts operate from a mature, holistic, well-tried understanding which is exercised intuitively and without conscious deliberation: their skills have become part of themselves (ibid 18–19). Experts can thus readily discern important distinctions between and amongst narrative accounts in ways that draw out the important elements of any inquiry (Landman, 2012: 36). They can grasp the unique combination of charac­ teristics that constitute a particular situation—that situation and no other (Berlin, 2013a: 489–90, 1997: 45; see also Kronman, 1987: 848; and Nussbaum, 1990: 94). Such actors pos­ sess, above all: ‘… a capacity for integrating a vast amalgam of constantly changing, multicoloured, evanescent, perpetually overlapping data, too many, too swift, too intermingled to be caught and pinned down and labelled like so many individual butterflies’ (Berlin, 1997: 46). The intuition which is such an important—perhaps the most important—part of the expert’s decision making is not, however, a kind of guesswork. It is not a form of irrational decision making, nor is it a form of decision making guided by supernatural inspiration (Berlin, ibid 45–46). Flyvbjerg explains that intuition is a property which each individual uses in everyday life. Flyvbjerg gives the example of riding a bicycle. If you are able to ride a bicycle then you will not have formulated a set of rules which, if followed, could teach somebody else to ride a bicycle. One cannot ‘teach’ the difference between nearly falling and the need to lean over to turn corners. How can one explain the best response to being off balance? A cyclist can ride a bicycle because he has what Flyvbjerg describes as the ‘­necessary know-how, achieved via practical experiences’. Thus, for Flyvbjerg, experience cannot necessarily be verbalised, intellectualised, and made into rules (Flyvbjerg, ibid 19; see also Berlin, ibid). Although everyone uses intuition, sentencers assume a special role and face situations that are distinct from the circumstances faced by the rest of us (Solum, 2003: 189). Whilst we all use intuition, sentencers employ their intuition over a larger field, against a wider horizon of possible courses of action and with far greater power (see Berlin, ibid 47–48).

From Novice to Expert—the Dreyfus Model of the Human Learning Process

 131

Intuitive Decision Making versus Sentencing Information Systems: The Sentencer as an Expert Actor Flyvbjerg illustrates his thesis through a discussion of ‘artificial intelligence’ and ‘expert systems’ which seek to replicate the decisions of professionals such as doctors, geologists, chemists, pharmacologists, stockbrokers and—it is suggested—judges through computer programming. Flyvbjerg notes that when designers of such expert systems (for example, sentencing information systems)37 seek to replicate the decisions of the professional groups, the professionals encounter difficulty in explaining to the system designers what they are doing in terms of specific procedures and rules (Flyvbjerg, 2001: 19–20). Flyvbjerg cites the following description of the problem provided by Schön: When [the professional practitioner] tries, on rare occasions, to say what he knows—when he tries to put his knowing into the form of knowledge—his formulations of principles, theories, maxims, and rules of thumb are often incongruent with the understanding and know-how implicit in his pattern of practice (Schön, 1987: 252, cited in Flyvbjerg, ibid 20).

The knowledge and skills required of a sentencer cannot be taught a priori but grow from the bottom up, emerging out of practice (Flyvbjerg et al, 2012a: 2). As Schram observes, such knowledge and skill ‘[forgoes] the hubris of seeking claims to a decontextualized uni­ versal rationality stated in abstract terms of false precision’ (Schram, 2012: 17). As a phronimos, a person of practical wisdom, the sentencer must be able to select and focus on those aspects and features of the case before him which are most relevant to his aims in crafting an appropriate disposal—a task for which judicial discretion is essential (Aas, 2005: 35). This conception does not reside in a set of rules or precepts; in no case will there be a rule to which the sentencer can simply appeal to tell him exactly what to do (Wiggins, 1975: 48). Through their professional experience, sentencers develop a capacity for taking in the total pattern of a human situation—of the way things ‘hang together’—a talent to which the power of abstraction and analysis (central to the drafting of, for example, prescrip­ tive sentencing guidelines) seems alien, if not positively hostile (Berlin, 1997: 50; see also ­Nussbaum, 2001: 305). The sentencer must therefore develop what Hyde, in a different context, describes as a form of ‘rudderless intelligence’: he or she must respond to cases as they arise (Hyde, 1998: 158, cited in Frank, 2012: 51). The effective sentencer needs what Gardiner describes as a developed capacity for a ‘non-generalising assessment of specific situations’; a finely attuned sensitivity to the shifting contours or levels of social existence, and, combined with this, an intuitive ‘feel’ for what is empirically feasible in the ‘intricate and frequently recalci­ trant sphere of particular facts or circumstances’ (Gardiner, 1997: xvii). In the context of sentencing, law is not exclusively a formal discourse but also a social practice and performance that is woven into a constantly changing social fabric (Aas, ibid; Hogarth, 1971; and Hutton, 2013a: 90). Thus the sentencer is not a detached observer as in the epistemic model, but someone who must intuitively address the given social context (Simmons, 2012: 255). 37  On the use of sentencing information systems generally, see Aas, 2005: 31–37; Hutton, 2013b: 111; Miller and Wright, 2005: 371–72; O’Malley, 2013a: 232–33 and 2006: 65– 66; and Tata et al, 1996.

132 

The Practical Wisdom of Sentencing

In a similar vein, Posner explains that in undertaking any judicial task, the costs of ­ rocessing information are often so great as to make intuition—what he describes as p ‘a ­telescoped form of thinking’—a more rational method of arriving at a decision than ­logical, step by step reasoning (2008: 37). Posner compares judicial decision making and judgment writing to the exercise of writing a novel. Both judges and novel writers are, to a great extent, intuitive reasoners, in the sense that much of their creative thinking is uncon­ scious. A novelist, for example, will write a passage one way rather than another because it feels right, although he may be unable to explain why it feels right. Similarly, a sentencer will often have a strong sense of what disposal to impose, but when he tries to explain the deci­ sion in a sentencing statement or, less frequently at first instance, in a written opinion, the explanation will often turn out to be a rationalisation of the result reached on inarticulate grounds (ibid 63). As Flyvbjerg notes, the Dreyfus model of learning allows us to understand why this is so. The expert recognises thousands of cases directly, holistically, and intuitively on the basis of his experience (Flyvbjerg, 2001: 20). Through his experience, the expert: ‘… holisti­ cally discriminates among classes of situations and associates with these classes appropriate responses’ (Dreyfus and Dreyfus, 1986: xii, emphasis per original; cited in Flyvbjerg, ibid; see also Schram, ibid 18). Flyvbjerg explains that because heuristic expert systems (such as sentencing information systems) are rule based, they are unable to go further than level three in the learning pro­ cess (ibid). The Dreyfus model contains a qualitative jump from the first three levels in the learning process (novice, advanced beginner and competent performer) to the fourth and fifth levels (proficient performer and expert). As Flyvbjerg explains, context and intuition are the two factors that characterise the actions of proficient performers and experts: The jump [from the first three to the final two levels] implies an abandonment of rule-based think­ ing as the most important basis for action, and its replacement by context and intuition. Logically based action is replaced by experientially based action (ibid 21).

Expert systems cannot, however, make the qualitative jump to levels four and five of the Dreyfus model and so will never become as skilful as human experts. In expert systems we see the primacy of knowledge itself, rather than the primacy of the thinking and judg­ ing subject; action is tied to knowledge and there is little space or need left for reflection (Aas, ibid 87). The language of such systems is coded and dissociated from the cultural context from which it emerges (ibid 136). In the context of sentencing, each offence exists as an objective category, without a context (ibid). As Frank explains, when intelligence sys­ tems dependent on principle confront unforeseen situations to which their principles are not readily applicable, their circuits jam. Phronesis, however, continues to function partly because such intelligence does not try to make present circumstances conform to long term plans: ‘Unencumbered by obligation to follow a predetermined plan, phronesis can respond to whatever requires response, seeing exactly what is at hand and using the resources at hand’ (Frank, 2012: 52). Flyvbjerg thus considers the term ‘expert system’ a misnomer: in terms of the Dreyfus model they can never be more than ‘competent systems’ (ibid 20). The result is that for any professional, be they a judge, teacher, doctor or psychologist, their best performance requires expertise based on experience, judgement and intuition. Posner also refers to the major role played by intuition in judicial decision making. For Posner, the faculty of intuition that enables a judge (or, for that matter, a businessman

Judicial Wisdom—Phronesis and Sentencing

 133

or an army commander), to make a quick judgement without a conscious weighting and comparison of the pros and cons of the possible courses of action is best understood as a capability for reaching down into a subconscious repository of knowledge acquired from one’s education and particularly one’s experience (Posner, 2008: 107). Insofar as sentencing is concerned, intuition in this sense is related to judgement. Experienced individuals tend to have good judgement. Although they may have largely forgotten many of their profes­ sional experiences, they remain accessible repositories of buried knowledge for coping with challenges that, despite being new, are not novel because they resemble previous challenges: their experiences ‘nourish their intuitions’ (ibid 107, 113; see also Kronman, 1987: 861–63). At first blush, such reliance on ‘intuition’ by a sentencer would appear to sit uneasily with the requirements of the rule of law. Flyvbjerg, however, notes the following: Intuition is the ability to draw directly on one’s own experience—bodily, emotional, intellectual— and to recognize similarities between these experiences and new situations. Intuition is internal­ ized; it is part of the individual. Existing research provides no evidence that intuition and judgment can be externalized into rules and explanations, which, if followed, lead to the same result as intuitive behavior (ibid 21—emphasis added).

Thus, context is of central importance in the development of any professional’s knowl­ edge and skills. Expertise in any given field is attained through the qualitative jump from the Dreyfus model’s first three stages to the last two stages. In so doing, the actor moves from rule-based, context independent behaviour to experience-based, situational behav­ iour (ibid 22). The sentencer’s practical wisdom has three aspects: its content, its personal quality, and its action (Frank, ibid 48). As content, phronesis is a resource—a stock of experien­ tial knowledge. As a personal quality, it is what enables acquisition and appropriate use of that knowledge—a capacity. Finally, as action, phronesis necessarily involves imposing sentences—a practice in which experiential knowledge is both used and gained. Having phronesis, Frank observes, is iteratively dependent on practising phronesis (ibid; see also Flyvbjerg et al, 2012b: 286).

Judicial Wisdom—Phronesis and Sentencing Flyvbjerg’s thesis seeks to develop an ethos and method of social science based on phronesis. In the context of sentencing, this approach is grounded in a search for practical knowledge. It can be compared with Edney and Bagaric’s (2007) ideal of a predictive science38 (episteme) that seeks greater social and technical control of judicial discretion in sentencing (see also Griggs and Howarth, 2012: 169). Flyvbjerg observes that, compared to rational decision making, intuitive decision making (such as sentencing) has been neglected as an object for scientific study (2001: 17). Indeed, with regard to sentencing, we have seen how judicial intuitive decision making in the form of Markarian analysis has been subject to scathing academic criticism. As Flyvbjerg notes, whilst ultimately it is a question of what constitutes science and whether it is possible to

38 

See the discussion in Chapter 4.

134 

The Practical Wisdom of Sentencing

study phenomena such as intuition and synchronicity scientifically, we are all familiar with most of these phenomena in their non-scientific form (seeing what needs to be done in an instant) when we perform in a craft, sport or music (ibid; see also Nussbaum, 2001: 299). It is suggested, therefore, that there is no reason why the intuitive decision making that ­constitutes discharge of the sentencing task—the ‘instinctive synthesis’—cannot be e­ xamined in these terms. Flyvbjerg cites the work of Pierre Bourdieu (1990: 9), noting that what he calls the ‘feel for the game’ is central to all human action of any complexity. Flyvbjerg continues: ‘[I]t ena­ bles an infinite number of “moves” to be made, adapted to the infinite number of possible situations which no rule, however complex, can foresee. Therefore, the judgment, which is central to phronesis and praxis, is always context-dependent’ (2001: 136; see also Flyvbjerg, 2006: 379 and Nussbaum, 1993b: 258). The more structured, predictable and less discriminating view of sentencing advocated by critics such as Hutton (2006a) and Edney and Bagaric (2007) is an excellent example of what Flyvbjerg describes as the problem of the dominance of rule-based rationality over practical experience (Flyvbjerg, 2001: 23). These theorists rely on a rule-based, rational mode of conceiving judicial sentencing which collapses when confronted with the Dreyfus phenomenology (ibid 22). The rational mode of thinking is, it is suggested, inadequate for comprehending the sentencing task. Hutton, Edney and Bagaric focus on properties such as consistency, foreseeability and certainty. These are properties by which humans most resemble machines: rule-based deliberation based on formal logic (ibid). It is an example of what Flyvbjerg terms the ‘rational fallacy’. Insofar as sentencing is concerned, analysis and rationality are raised into the most important mode of operation. Analysis and rationality are allowed to dominate our view of the sentencing task to the extent that other, equally important, modes of understanding and behaviour—the facts and circumstances of the specific offence, the personal circumstances of the offender, the harm sustained by the ­victim, in short, context—are made invisible (ibid 23). Flyvbjerg explains that the Dreyfus model does not present a situation of ‘either rational­ ity or intuition’; rather, it presents both of them in their proper context (ibid). The posi­ tion of intuition is not beyond rationality, states Flyvbjerg, but is complementary to it and, insofar as we speak of experts, above rationality: The model specifies that what is needed in order to transcend the insufficient rational perspective is explicit integration of those properties characteristic of the higher levels in the learning process which can supplement and take over from analysis and rationality. These properties include con­ text, judgment, practice, trial and error, experience, common sense, [and] intuition … (ibid).

Although Flyvbjerg notes that the pervasiveness of the rational paradigm to the near exclu­ sion of others is a problem in practical fields such as engineering, policy analysis and plan­ ning, he considers that law is an exception. The practice of law, states Flyvbjerg, cannot be de-contextualised to the same degree as other disciplines; he maintains that, as a result, it has never been made ‘scientific’ to the same extent (ibid 23–24). True as this may be for the practice of the substantive law, de-contextualisation of sentencing is exactly what is advo­ cated by theorists such as Hutton, Edney and Bagaric. For Flyvbjerg, the rule-based, rational mode of thinking (adopted by those who would limit or remove judicial discretion in sentencing) generally constitutes an obstacle to good results because the rational perspective has been elevated from being necessary to being

Judicial Wisdom—Phronesis and Sentencing

 135

s­ufficient, or even exclusive (ibid 24). Flyvbjerg explains the result as follows: ‘This has caused people and entire scholarly disciplines to become blind to context, experience, and intuition, even though these phenomena and ways of being are at least as important and necessary for good results as are analysis, rationality, and rules’ (ibid). Thus, rationality may endanger sensitivity to context, experience and intuition. In order to achieve the most effective results in any given field, Flyvbjerg explains that one must incorporate knowledge and make it instinctive. Context, judgement and practical knowl­ edge are key (ibid). To demand or preach mechanical precision, even in principle, in a field incapable of it such as sentencing is, as Berlin tells us, to be blind and to mislead others (Berlin, 1997: 53). The five steps in the human learning process set out in the Dreyfus model emphasise the importance of gaining concrete experience as a precondition for the qualitative leap from what Flyvbjerg describes as the rule-governed, analytical rationality of the first three levels to the intuitive, holistic and synchronous performance of tacit skills of the last two levels (Flyvbjerg, ibid 71). Sentencers obtain this ‘concrete experience’ through their years of prac­ tice at the Bar and, later, on the Bench. Context-dependent knowledge and expertise is at the very heart of expert activity since experts operate on the basis of intimate knowledge of several thousand concrete cases in their areas of expertise (ibid). Thus it is experience with concrete cases that allows an individual to move from level three in the Dreyfus model to levels four and five: If people are exclusively trained in context-independent knowledge and rules, that is, the kind of knowledge which forms the basis of textbooks and computers, they will remain at the first levels of the learning process. This is the limitation of analytical rationality: it is inadequate for the best results in the exercise of a profession, as student, researcher, or practitioner (ibid).

Levels four and five in the Dreyfus model can be reached only through the individual’s own experiences as a practitioner of the relevant skills. To illustrate the consequences of this, Flyvbjerg cites the following passage from Aristotle’s The Nicomachean Ethics39 wherein, recognising the important association between the particular case and experience, Aristotle criticised the subjects of geometry and mathematics: [A]lthough [people] develop ability in geometry and mathematics and become wise in such mat­ ters, they are not thought to develop prudence [phronesis]. The reason for this is that prudence also involves knowledge of particular facts, which become known from experience … We should therefore pay no less attention to the unproved assertions and opinions of experienced and older people (or of prudent people) than to demonstrations of fact; because they have an insight from their experience which enables them to see correctly (cited in Flyvbjerg, ibid 72).

Thus, human behaviour (in our case, the imposition of just, proportionate, and effective sentences) cannot be meaningfully understood as simply the rule-governed acts found at the lowest levels of the learning process (Flyvbjerg, ibid). Such an approach would likely result in the consistent and predictable sentences desired by Edney, Bagaric and Hutton; however, such sentences would not be imposed as a result of judicial wisdom and would take insufficient account of the relevant context: the facts and circumstances of the offender, the victim(s) and the interests of society as a whole. Applying the Dreyfus model, we see

39 

Aristotle, 1976: 1142a12, 1143a32-b17.

136 

The Practical Wisdom of Sentencing

that the judge’s experience and the context of the individual case are of central importance in undertaking the sentencing task.

A Reconceptualisation of Sentencing as ‘Art’: The ‘Phronetic Synthesis’ The ‘Art’ of Sentencing It is almost trite to state that sentencing is a complex task which calls for careful judgment. It is a task based on the sentencer’s experience and instinct derived after taking into account all the facts and circumstances of the particular case.40 We have seen how the sentencer must navigate between the particular circumstances of a case and the universal legal rules, constructing coherent sentencing narratives in order to sentence individuals as opposed to legal categories of offenders (Aas, 2005: 78–80, 86–87; Tombs, 2008: 84–85). Aas (ibid 28) suggests that because the nature of judges’ work is to solve concrete cases, they are not actors who are strong in terms of having models and concepts (see also LloydBostock, 1988: 64). It is perhaps for this reason that, rather than justifying the need for their wide sentencing discretion in terms of Solum’s equity, Berlinian value pluralism or Aristo­ telian phronesis, sentencers in both Scotland and in other common law jurisdictions have tended to claim that sentencing is an art and not a science. It is typically observed by the judiciary that effective sentencing requires that the disposal be tailored to the circumstances of the particular offence and the characteristics of the offender (Carloway, 2013: 15 and 23–24; Findlay et al, 2014: 246–47; Krasnostein, 2015a: 16–17; Krasnostein and Freiberg, 2013: 267; O’Malley, 2008a: 9 and 2006: 66–67; and Warner, 2002: 6). Thus sentencing is a process which cannot be reduced to the mechanical application of rigid formulae— professional experience, intuition and subjective judgment are central (Aas, ibid 25; Cana­ dian Sentencing Commission, 1987: 281; Cardwell Hughes, 1999: 20, cited in Aas, ibid, and the empirical studies by Ashworth et al, 1984: 50; Tombs, 2004: 44—see also Jacobson and Hough, 2007: 48—and Tombs, 2008: 106). The Court of Appeal (Criminal Division) has conceptualised the sentencing task in terms of art rather than science on many occasions.41 In New Zealand the courts also subscribe to the notion that sentencing is not an exact science,42 the inference being that the flexibility of the sentencing discretion must be preserved (Hall, 2007: 5). In Canada, the Supreme Court recently expressed very similar sentiments in R v LM, where LeBel J, writing for the major­ ity, prefaced his discussion of the fundamental principles of sentencing with the following observation: ‘Far from being an exact science or an inflexible predetermined procedure, sentencing is primarily a matter for the judge’s competence and expertise. The trial judge enjoys considerable discretion because of the individualized nature of the process’.43 This view of the sentencing process was later approved by Lord Justice Clerk Gill (as he then was) in delivering the leading opinion of the Appeal Court in Gemmell v HMA.44 It is 40 

Markarian v The Queen [2005] HCA 25 at [69]. See, for example, Attorney-General’s Reference (No 24 of 2009) (sub nom R v Bradbury) [2010] 1 Cr App R (S) 23 at [23]; and R v McCormack [2011] EWCA Crim 1601 at [15]; see also Judge, 2007: 1. 42  Gould v The Queen [2012] NZCA 284 at [34]. 43  R v LM [2008] 2 SCR 163 at [17]. 44  Gemmell (n 11) at [59]. 41 

Judicial Wisdom—Phronesis and Sentencing

 137

in Australia, however, where the notion of sentencing as ‘art’ has most often found expres­ sion in the judgments of the appellate courts (Mackenzie, 2005: 15–16, 2001: 133–36). In one of the most notable examples, the Supreme Court of New South Wales in R v Jurisic45 issued certain sentencing guidelines for offences in which death or grievous bodily harm was caused by dangerous driving. In the course of their judgments, two of the five judges expressed the view that sentencing was an art. Spigelman CJ referred to ‘the inexact art of sentencing’46 whilst Sully J spoke of ‘the art of sentencing’47 and stated that: ‘as this Court has pointed out on many occasions, sentencing is an art and not a science; and it is nei­ ther practical nor desirable to impose procedural strait-jackets upon primary sentencing Judges’.48 Mackenzie (2001: 133) notes that in the earlier decision of R v Astill (No 2),49 Sully J elaborated on the notion of the art of sentencing when he stated that it is neither possible nor desirable for a sentencer to approach the task as though it were little more than ‘an arid mathematical exercise’. His Honour also contrasted this conceptualisation with the notion of a science in the unreported decision of R v Howland:50 If it be true to say that sentencing is an art and not a science—and there is clear authority in this Court for that proposition—then it is equally the case that sentencing is not an exercise in either creative forensic mathematics or creative forensic semantics.

In what would subsequently become a much cited dicta in Australian sentencing appeals, Crisp J of the Tasmanian Court of Criminal Appeal explained the typical judicial approach in R v Wise51 as follows: Sentencing is an art and not a science. By that I mean that if in any given case it were possible to arrive at the exact measure of the punishment to be awarded by the application of logical rules which proceeded with mathematical inevitability to a determinate conclusion, then indeed the judge’s responsibility would be a different one from that which we know. Notwithstanding this fundamental fact, this court is constantly faced with attempts, based on some system of analysis, to expound principles of precision in relation to punishments awarded from time to time by different judges in respect of crimes of the same generic character no doubt, but which differ individually and infinitely both in degree and circumstance.

This view was later affirmed in Attorney-General (Tas) v Blackler.52 Having approved the dicta of Crisp J in Wise, the Court acknowledged that whilst there is a general requirement for consistency in sentencing, in almost all cases it is impossible to arrive at the appropriate sentence as if it was an inevitable conclusion following a comparison of the particular case with other cases and a strict and logical application of sentencing rules.53

45 

(1998) 45 NSWLR 209. Ibid at 215. 47  Ibid at 250. 48  Ibid at 249. 49  (1992) 64 A Crim R 289 at 301. 50  New South Wales Court of Criminal Appeal, 18 December 1997. 51  [1965] Tas SR 196 at 200. 52  [2001] TASSC 27. 53  Ibid at [10]. 46 

138 

The Practical Wisdom of Sentencing

The view that sentencing is an ‘art’ is criticised by Walker who considers that when s­ entencers claim that their task is an ‘art’, they usually mean this literally. This, he states, is both sententious and pretentious: The aims of sentencing are neither aesthetic nor original. If all that is meant is that [sentencing] cannot be taught by mere rules, that is true, at least of modern British sentencing; but it is also true of skills which we would not seriously call arts, such as playing bridge (Walker, 1999: xii; see also Walker, 2003: 149–50 and Warner, 2002: 66).

As Fox and Freiberg note, the proposition that sentencing is an ‘art’ carries with it the impli­ cation that it is an instinctive or intuitive skill rather than a learned one (1999: 195). Even if capable of being learned, it is an ability acquired only by years of experience at the criminal Bar and on the Bench (Connolly, 2006: 2; Fox, 2006: 15 and 1987: 226; Walker, 1969: 163– 64). Thus, unlike so much of the law which has its genesis in the notion of ‘common sense’, sentencing is driven by a sense of the uncommon variety (Connolly, ibid). The description of sentencing as an ‘art’ is one which—like the practice of equity, pluralist decision making and the application of phronesis—gives primacy to the role of the sentencer (Mackenzie, 2003: 289). Indeed, what the appellate courts across the Commonwealth mean when they refer to the ‘art’ of sentencing is the intuitive, holistic and interpretive form of decision making involved in phronesis (Flyvbjerg, 2001: 14). Whilst critics such as Edney and Bagaric (2007) and pro­ ponents of prescriptive, presumptively binding sentencing guidelines seek to impose formal rationality, rules and consistency in sentencing through a form of decision making that is system-orientated, abstract and objective, the sentencer (as an expert actor and phronimos) focuses on the context of the individual case and draws on his or her experience in employ­ ing a form of decision making that is case-orientated, concrete and subjective (Flyvbjerg, ibid 21; Aas, ibid 26). In analysing the instinctive synthesis approach to sentencing, Leader-Elliot considers that, rather than conceptualising the sentencing task in such terms, it might have been pref­ erable for the courts to risk an accusation of what he describes as ‘overweening arrogance’ and defend the exercise of sentencing discretion on the ground that it permits the expres­ sion of judicial wisdom (2002: 8; see also Spigelman, 2008: 460). I suggest, however, that it is judicial practical wisdom—or phronesis—which best describes the exercise of discretion in sentencing.

A Balancing Exercise In the present study, 11 of the 25 judicial officers interviewed (nine sheriffs and two judges) conceptualised the sentencing task in terms of balancing: In general, I think there has to be discretion. That’s justice—weighing the scales, it’s not just a table to follow. You have to weigh up all the factors (Sheriff 15). [F]irst and foremost is the offence itself: the gravity of that offence and all the circumstances which surround the offence—the mitigation, the aggravation. But there’s also looking at the offender and that offender’s personal circumstances. One is seeking to b­ alance those two aspects. And I suppose, at a very basic level, that is the underlying tension (Sheriff 10). You have to reconcile a variety of conflicting aims: punishment, protection of the public, deterrence … You have to try to balance these aims to do justice in the whole circumstances of the case—the circumstances of the offence and of the offender himself (Judge 4).

Judicial Wisdom—Phronesis and Sentencing

 139

These findings are consistent with Mackenzie’s study of the Queensland judiciary. Over one third of Mackenzie’s respondents (12 of the 31 judges) described the process in terms of balancing (Mackenzie, 2005: 14–15, 2001: 127–29). The factors to be balanced depended on the circumstances of the particular case, varying from a balancing of the principles which favoured a heavy penalty or a lighter one, to the interests of all parties in the process—the accused, the accused’s family, the victim, his or her family and the wider community (ibid). These interests can be hard to reconcile (Mackenzie, 2006a: 1; Spigelman, 2008: 450). The sentencer has to determine which of the many facts, factors and features are rel­ evant to the sentence and then, in striving for the attainment of all the purposes of pun­ ishment, decide what weight to give each of them (Ashworth, 1984: 529; Cooke, 1987: 57; O’Malley, 2003: 134; Terblanche, 2007: 114; Terblanche and Roberts, 2005: 189 and 193). In an interview given in 2008, for example, the Scottish judge Lady Dorrian (as she then was) explained: You’re always conscious of the consequences of the decision that you’re making and the conse­ quences to the individual person. But at the same time you have to weigh up the effects on the victims of the crime, and what has been done. So sending people to prison is never something that anyone would do lightly, but it’s a balancing exercise involving a number of considerations, and you have to take all those considerations into account and then make the decision that you think best meets the requirement for punishment, deterrent [sic], protection of the public (see Goring, 2008: 29).

This is not, however, balance in the ordinary sense of the word. Any mitigating factors may be completely outweighed by the seriousness of the offence; in such circumstances this cannot amount to‘balance’ since the scales would be heavily weighed against the offender (Terblanche, ibid 147; see also Maslen and Roberts, 2013: 138). The issue was considered by the then Appel­ late Division of the Supreme Court of South Africa in S v Zinn54 where Rumpff JA held that in imposing sentence, the judge must consider a ‘triad’ consisting of ‘the crime, the offender and the interests of society’.55 Justice Rumpff ’s ‘sentencing triad’ would later become a much quoted passage in South African sentencing appeals (Terblanche, 2011: 263, 266–67). The principles set out in Zinn are commended by Terblanche and Roberts as they allow the court to make an accurate assessment of the seriousness of the offence, to take the personal circumstances of the offender into consideration, and to have regard to the interests of society (Terblanche and Roberts, ibid 198–99; see also Terblanche, 2008: 123). Rather than viewing the balancing exercise required of the sentencer in its ordinary sense, Terblanche thus considers it more accurate to state, as in S v De Kock,56 that the three factors of the Zinn triad have to be considered in conjunction with one another and that each should be afforded a certain weight depending on the facts of the case (Terblanche, 2007: 147, 186). This approach is more nuanced and sophisticated than the traditional, bald statement of courts throughout the Commonwealth that all factors relating to offence and offender require to be balanced.57 It is an approach that has found favour with the 54 

1969 (2) SA 537 at 540. also Viljoen Commission, 1976: paras 5.1.3.33–34; S v Toms; S v Bruce 1990 (2) SA 802 (A) at 806; S v Kibido 1998 (2) SACR 213 at 216; and S v Serabo 2002 (1) SACR 391 at 397. 56  1997 (2) SACR 171. 57  See, for example, Ryan v The Queen [2001] HCA 21 at [133] and [157]; R v M (CA) [1996] 1 SCR 500 at [91]; and Ashraf v HMA 2010 SCCR 826 at [12]–[14]. 55  See

140 

The Practical Wisdom of Sentencing

South African courts. For example, in S v Banda58 it was acknowledged that the elements of the Zinn triad ‘contain an equilibrium and a tension’ which means that the sentencer should counterbalance them in order to ensure that one element is not emphasised at the expense or exclusion of another (see Terblanche, ibid 147).

The Phronetic Synthesis Phronesis as practical wisdom is a capacity for synthesis rather than analysis (Berlin, 1997: 47). It is phronesis which allows judicial recourse to equity in sentencing, which ensures that the judge affords the appropriate weight to each factor in the Zinn triad and which is thus the most compelling argument in favour of a wide sentencing discretion. Whilst it is still common for the judiciary to describe sentencing as an ‘art’ or as a ­‘balancing exercise’, the ‘instinctive synthesis’ developed since the mid-1970s by the Aus­ tralian courts (and recently approved by the Appeal Court in Scotland) remains one of the most accurate shorthand descriptions of the sentencing task. I suggest, however, that when one considers the importance of judicial recourse to equity in sentencing and the practice of phronesis, judicial sentencing methodology—to the extent that it relies on intuition and experience—is better conceptualised in terms of a phronetic synthesis of the relevant facts and circumstances. This is not to claim—pace Field, 2002: 275—that the sentencing process is akin to ‘the alchemist’s potion, compiled from exotic ingredients combined by means of a formula known only to adepts’. Rather, the phronetic synthesis conceptualises the sentenc­ ing task as a form of case-orientated, concrete and intuitive decision making that achieves individualisation through judicial recognition of the profoundly contextualised nature of the process.

58 

1991 (2) SA 352 at 355.

6 Structuring the Sentencer’s Discretion Introduction The discretionary nature of sentencing does not mean that the discretion afforded to judges should be untrammelled. The central problem in framing just sentencing laws is how to provide sentencers with genuine guidelines which ­nevertheless are flexible enough to be adaptable to meet the special circumstances of a particular case (Bottoms, 1998: 69). It is, however, an unavoidable consequence of any attempt to structure sentencing discre­ tion that it must necessarily involve circumscribing the powers of the judiciary (Duff, 2005: 1165; Krasnostein, 2015a: 164 and 2015b: 52; Krasnostein and Freiberg, 2014: 4; Roberts and Baker, 2008: 564). Providing a principled structure for sentencing entails constraining the sentencer, yet the exercise of discretion is crucial if justice is to be done in the individual case (Easton and Piper, 2016: 37–38). If sentencing is to be structured, decisions must be taken on the content of guidance; on the source of guidance; on the authority by which it should be laid down; and on the style in which it should be formulated, with consideration also being given to the mechanics of putting the guidance into practice ­(Ashworth, 1998a: 213; Roberts, 2009: 230). Insofar as the content itself is concerned, ­Ashworth suggests that any guideline system ought to deal with at least three aspects of the sentencing decision: (i) general principles of approach, aggravation and mitigation; (ii) specific offences; and (iii) the use of particular types of sentence (Ashworth, 2006: 9). Both Roberts and Ashworth consider that widespread judicial discretion has led to sen­ tencing being influenced by a variety of principles and policies, largely according to the incli­ nations of the individual sentencer (Ashworth, 1998a: 214; Roberts, ibid 231). For Roberts and Ashworth, one of the aims of structuring discretion should be to ensure that it is exer­ cised in a principled manner (Ashworth, ibid; Roberts, ibid). An essential step in achieving this is to decide upon a leading aim or rationale for sentencing, with a choice being made between deterrence, rehabilitation, incapacitation, reparation or desert as the leading aim of the system. Having chosen the leading aim, a decision should be taken about whether it is permissible for sentencers to be influenced by any other aim or aims, to what degree, and in what types of case. Roberts and Ashworth both consider that unless decisions of principle are taken on priorities and spheres of application of two or more sentencing aims, uncer­ tainty will result which would be ‘a recipe for disparity’ (Ashworth, ibid; Roberts, ibid). In considering the source of sentencing guidance, its style and the authority by which it should be promulgated, the traditions and legal culture of the jurisdiction will be a signifi­ cant factor (Ashworth, ibid 215–16; Freed, 1992: 1683; O’Malley, 2013a: 226–27; Padfield, 2013: 32; Roberts, ibid 232). One possible approach is for the basic principles to be set out in

142 

Structuring the Sentencer’s Discretion

primary legislation, with the judiciary being left to develop the detailed guidance through appellate judgments. This approach does, however, depend for its success on the senior judiciary’s willingness to take the principles seriously and to develop them sympathetically (Ashworth, ibid; Roberts, ibid). Different styles of sentencing guidance vary in the extent to which they reduce judicial discretion but, as both Roberts and Ashworth note, it would be wrong to assume that the pursuit of principled sentencing means that the most constraining approach is necessarily the best: [T]here will inevitably be questions of detailed application that can be answered differently by dif­ ferent sentencers, so that, even if all sentencers were conscientiously pursuing the same aim or set of aims, inconsistencies could result … Much will depend, in practice, on what is deemed appro­ priate in the context of the legal and political culture of the jurisdiction (Ashworth, ibid 216–17; Roberts, ibid 233).

The Pursuit of Consistency in Sentencing— Alternatives to the Phronetic Synthesis A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines. With consistency a great soul has simply nothing to do (Ralph Waldo Emerson, Self Reliance, cited in Spigelman, 2008: 450).

The English Approach to Structuring Discretion through Sentencing Guidelines The English judiciary has traditionally insisted that rigid rules are incapable of taking account of the complex combination of facts that arise in individual cases and that a degree of judicial discretion is essential for the just application of legal rules and principles (Player, 2012: 248). This has resulted in a definition and pursuit of the notion of consistent sentenc­ ing in relation to a conformity of approach, rather than a conformity of outcome (Ash­ worth, 1987a: 34; Player, ibid; Wasik, 2008: 259; see also the dicta of Lane CJ in R v Bibi).1 The concept of ­consistency of approach aims to strike an equilibrium between the supposed dangers of too much and too little judicial discretion in sentencing (Krasnostein, 2015a: 77–78). The guideline movement in England and Wales has, however, evolved rapidly in recent years (Ashworth and Roberts, 2012: 891). Its development can be seen, broadly, as consisting of four ‘phases’ (ibid 881–82).

The First ‘Phase’—Appellate Guidance In the first phase, the Court of Appeal began to lay down various general principles of sentencing in its judgments in appeals against sentence, a practice which became more 1 

(1980) 2 Cr App R (S) 177 at 179.

The Pursuit of Consistency in Sentencing

 143

frequent from the 1960s onwards (Ashworth, 2009: 243 and 2013b: 16; Ashworth and Roberts, 2013: 3). The Court took a more active role in issuing sentencing guideline judg­ ments in the 1980s and 1990s: the Lord Chief Justice would select a particular appeal case in which to deliver such a judgment, setting out the parameters of sentencing for a range of variations of the crime in question (Ashworth, ibid; Ashworth, 2006: 2; Bingham, 2000b: 46–47; Fox, 1987: 227; Harvey and Pease, 1987: 96; Henham, 1995a: 220–21; Kras­ nostein and Freiberg, 2014: 13; Lovegrove, 1987: 134; O’Malley, 2016: 13–14; O’Malley, 2006: 58; Pease, 1987: 41; Samuels, 1987: 69; and Wasik, 2008: 253). Such judgments often indicated one or more starting points and set out various aggravating and mitigating factors (Ashworth, 2001: 73; Easton and Piper, 2016: 48). They began to provide a frame­ work for sentencing in the contexts to which they related, around which sentencers could analyse the circumstances of the individual case (Thomas, 1987: 19). The former Lord Chief Justice of England and Wales, Lord Judge, reports that the deci­ sion to issue offence-specific sentencing guidelines was a direct result of the creation of the Judicial Studies Board (‘JSB’, now the Judicial College). As more Crown Court judges met at JSB events, it became clear that there were differences of approach amongst the judges to specific sentencing questions. Judge reports that disagreements about the broad approach to these and similar questions were drawn to the attention of the then Lord Chief Justice, Lord Lane, by the Chairmen of the JSB. This led to the issue of guideline judgments designed to offer authoritative guidance on controversial problems (Judge, 2014: 679–80).

The Second ‘Phase’—the Sentencing Advisory Panel English sentencing guidelines entered their second phase in 1998 when a Sentencing Advi­ sory Panel (‘SAP’) was introduced under sections 80 and 81 of the Crime and Disorder Act 1998. The SAP started England upon the journey towards more structured sentencing (Roberts, 2012a: 333). The SAP prepared draft guidelines on various offences, consulted widely on them and sent written advice to the Court of Appeal proposing guidelines for the offence in question. The Court retained the power to accept, modify or reject the Panel’s advice but in practice it usually accepted it, subsequently incorporating the proposals (with or without modifications) into a guideline judgment (Ashworth, 2013b: 16, 2010b: 389, 2009: 244, 2008: 112–13, 2006: 2–3; Easton and Piper, 2016: 49; Krasnostein and Freiberg, 2014: 14; O’Malley, 2006: 59; and Roberts and Ashworth, 2016: 334).

The Third ‘Phase’—the Sentencing Guidelines Council In the third phase, the system was altered again in 2003 with the creation of the Sentencing Guidelines Council (‘SGC’).2 The Council had the power to issue ‘definitive guidelines’ to which judges were under a statutory obligation to ‘regard’,3 reasons having to be given by judges if they departed from the guidelines4 (Ashworth, 2013b: 23; Gerry et al, 2014: 601– 02; Krasnostein and Freiberg, 2014: 14; O’Malley, 2006: 59–60; Roberts, 2011a: 1001; and Roberts and Baker, 2008: 568–69). This change accelerated the development of definitive sentencing guidelines in England and Wales (Ashworth and Roberts, 2013: 5). 2 

For a discussion of the background to the creation of the SGC, see Ashworth and Player, 2005: 827–29. Section 172(1)(a) of the Criminal Justice Act 2003. 4  Section 174 of the 2003 Act. 3 

144 

Structuring the Sentencer’s Discretion

R v Oosthuizen5 was the first decision of the Court of Appeal to be concerned with the treatment of definitive guidelines published by the SGC (Thomas, 2005b: 981). In deliver­ ing the judgment of the Court, Rose LJ noted that although judges were directed by statute to have regard to such guidelines, it did not necessarily follow that they fell to be applied in every case: ‘having regard’ to a relevant guideline did not mean that it had to be followed.6 In this connection, his Lordship quoted the dicta of Judge LJ in R v Peters,7 that: ‘Guidelines, whether resulting from cases decided in this Court, or produced by the Sentencing Guide­ lines Council, are guidelines: no more, no less’.8 The Court of Appeal would frequently emphasise that guidelines were no more than an aid to the judge in exercising judicial discretion and that the ultimate responsibility for determining a just sentence rested with the judge (Ashworth, ibid 23; Thomas, 2008: 160). For example, having delivered its judgment in Oosthuizen, the Court was later to stress in R v Martin9 that sentencing is not a mathematical exercise; that the statutory requirement to have regard to the guidelines fell well short of ‘a demand for robotic adherence’;10 and that guidelines do not provide fixed sentences: every case is different and the point of hav­ ing a judge is to ensure that each case is assessed individually.11 The importance of doing ‘justice in the circumstances of an individual case’, meanwhile, was stressed by the Court in Attorney-General’s References (Nos 7 to 9 of 2009);12 it was noted that a judge was entitled to disregard a guideline if an injustice would result from following it.13 In this third phase, the SAP performed much the same function as before, but now advised the SGC rather than the Court of Appeal (Wasik, 2008: 253). This was a period of considerable activity in England. A substantial corpus of sentencing guidelines was created with most cases in the lower courts being covered by the Magistrates’ Court Sentencing Guidelines and most cases in the Crown Court also falling within guidelines (Ashworth, 2010b: 389, 2006: 4–8; Ashworth and Roberts, 2012: 882; Pina-Sánchez, 2015: 77; see also Easton and Piper, 2016: 49–50 and Roberts and Ashworth, 2016: 334). This style of guidance depended to a considerable extent on narrative elements of the guideline documents with the actual guidelines taking the form of tables; it was a style developed from the two models that had previously been used in England: the narrative approach of the Court of Appeal guidelines and the more structured approach of the spe­ cial guidelines for magistrates’ courts (Ashworth, 2009: 245, 248). The aim of these guide­ lines was to structure rather than eliminate proper decision making by sentencers (Wasik, ibid 255). In a valuable critique of the system, the barrister John Cooper QC cautiously wel­ comed certain aspects of the SGC’s work, but highlighted areas of concern that had hith­ erto received little attention from academic commentators (Cooper, 2008). Although the

5 

[2006] 1 Cr App R (S) 73. Ibid at [15], citing R v Last [2005] EWCA Crim 106 at [16]. 7  [2005] 2 Cr App R (S) 101. 8  Oosthuizen (n 5) at [15], citing Peters (n 7) at [3]. 9  [2007] 1 Cr App R (S) 3 at [21]. 10  R v Matthews [2005] EWCA Crim 2768 at [9]. 11  Attorney-General’s Reference (Nos 32, 33 and 34 of 2007) [2008] 1 Cr App R (S) 35 at [16]; R v Ismail [2005] 2 Cr App R (S) 88 at [15]. 12  [2010] 1 Cr App R (S) 67 at [39]. 13  Ibid; see also Attorney-General’s Reference (No 8 of 2007) [2008] 1 Cr App R (S) 1 at [16]. 6 

The Pursuit of Consistency in Sentencing

 145

principle of guidelines both enables relevant and defined argument to take place in court and permits counsel to manage the lay client’s expectations of the sentencing outcome (ibid 278), Cooper stressed the importance of retaining judicial discretion: Whilst predictability and consistency is crucial to the sentencing regime, so too is judicial discre­ tion and the preservation of judicial confidence to be flexible when necessary. Providing that the guidelines handed down by the Council do not put the judiciary into a straitjacket, then, from a practical point of view there is much to be gained from this initiative. No case is ever identical to another, however, and each sentencing exercise is as particular to itself as the individual being dealt with. There is a significant and oft-repeated concern within the legal profession that the guidelines are stifling judicial discretion (ibid 279).

Whilst acknowledging that the guidelines referred to the need for flexibility and variability, Cooper reported that the majority of both the judiciary and the legal profession perceived the guidelines to be if not mandatory then heavily prescriptive. Many practitioners perceived the sentencing process to have become a ‘box ticking’ exercise at the expense of a flexible sen­ tencing regime with the layout of the SGC’s various guidelines lending itself to a ‘multiple choice’ approach. Cooper reported that the guidelines were being used not as helpful sign­ posts to a proper disposal, but rather as ‘fireproofing’ for any subsequent appeal. The use of guidelines could prevent sentencers from ‘taking a chance’ on a particular offender, result­ ing in imprisonment when a more imaginative disposal might have been better. This, noted Cooper, was all to the detriment of flexible, sensible and just sentencing (see also Blackett, 2013: xv). The guidelines-inspired ‘box ticking’ approach tended to replace a careful con­ sideration of both the individual and the pertinent facts of the case (Cooper, ibid 279–80; see also Lovegrove, 2010: 921–22 and Tiarks, 2016: 71–72; cf Ashworth, 2008: 118). This is deeply concerning since sentencing is not, and should never be, a rigid and formulaic process (Selfe, 2012: 4). A perception was found amongst the criminal Bar that guidelines for some offences, such as robbery, resulted in sentences being driven up, a danger acknowledged by Martin Wasik, a staunch supporter of the English guidelines movement (Wasik ibid 260; 2003: 250–52). Cooper suggested that this was due to the structure of the guidelines which presented pre­ scriptive lists of aggravating features, doing nothing to assist either counsel in their speech in mitigation or the court itself from structuring a sentence which embraced the individu­ ality of a case (Cooper, ibid 283; cf Ashworth, 2006: 21–22). Similar findings are reported by Hough et al (2003): the authors’ interviews with judges led to the suggestion that guide­ line judgments, in combination with the possibility of prosecutorial appeal, served to draw lenient judges’ decisions up to the guideline level, whilst leaving those of tougher judges unchanged (ibid 25; see also Millie et al, 2007: 249–50; and Tombs, 2004: 35). Curiously, these findings are dismissed by ­Ashworth as ‘purely anecdotal’ (2008: 120, 2006: 18), despite the fact that the study by Hough et al involved one to one semi-structured interviews with 48 Crown Court judges, recorders and district judges, along with five members of the senior judiciary (Hough et al, ibid 4–5).14 The sentencing regime requires sophistication rather than prescription (Cooper, ibid 284). The approach of the SGC prioritised sentencing the offence, rather than the offender, 14  The oft-cited Crown Court study by Ashworth et al, by comparison, involved interviews with 25 judges (Ashworth et al, 1984: 9–10). Quaere whether this means that the findings of that study can similarly be dismissed as ‘purely anecdotal’.

146 

Structuring the Sentencer’s Discretion

thus making a constructive disposal highly unlikely. From the practitioner’s point of view, there exists a fundamental conflict between the need for conformity and certainty in sen­ tencing and the reality that each case will inevitably depend upon its own facts (ibid 285). Inherent within this problem is the erosion of judicial discretion to deal with each case individually and, ultimately, fairly (ibid 286).15 There is a danger that the guidelines system in England and Wales excluded—and continues to exclude—the humanitarian concerns which sentencers rightly have for those appearing before them (Padfield, 2011: 97; see also Dhami, 2013a: 293–302 and 2013b: 175–81).

The Fourth ‘Phase’—Presumptively Binding Guidelines and the Sentencing Council The fourth and final phase of the development of English sentencing guidelines occurred in April 2010 with the commencement of the Coroners and Justice Act 2009.16 The 2009 Act revamped the previous arrangements by making two key changes: the amendment of the courts’ duty to comply with guidelines, and the establishment of the Sentencing Council of England and Wales which replaced the SAP and the SGC. Whilst under the previous regime, statute dictated that courts ‘must have regard’ to any relevant guidelines, section 125(1) of the 2009 Act introduced the requirement that courts must follow the relevant guidelines unless satisfied that it would be contrary to the interests of justice to do so (Ashworth, 2012: 93–94; Ashworth and Roberts, 2012: 882; Easton and Piper, 2016: 51–52; Edwards, 2012: 344–345; Hutton, 2013a: 92; Krasnostein and Freiberg, 2014: 14; Pina-Sánchez, 2015: 77; Roberts, 2013a: 12–13, 2012b: 273–74, 2011a: 1010; Rob­ erts and Ashworth, 2016: 336–37). The interpretation and application of this part of the section by the Court of Appeal will, as Thomas notes, be a matter of great interest (Thomas, 2010: 9). Cooper suggests that the tension between the directive to follow the guidelines and the ‘interests of justice’ test will continue to ‘undermine any aspiration for consistency in sentencing’ (2013: 158). It has been suggested that the interests of justice must relate to the facts of the par­ ticular case, not to disagreement with the approach taken in any guideline issued by the Sentencing Council (Ashworth, 2010b: 395, a view later affirmed by the Court of Appeal (Criminal Division) in R v Healey17). Further suggestions for good practice include, firstly, a requirement for a court that departs from the guidelines to both articulate its reasons for so doing and explain the extent of the departure involved: stepping outside a guideline should not mean ­‘stepping into an entirely unstructured realm where proper justifica­ tions are absent’ (­ Ashworth, 2012: 95); secondly, a requirement for a court to give rea­ sons for adopting a starting point lying outside the offence range even if it ­subsequently imposes a sentence that comes within the offence range (for example having made a reduction for a guilty plea); and for these two principles to apply equally to the Court of Appeal (Ashworth, ibid). 15  See also Cooper’s criticism of the Sentencing Council’s new format guidelines (Cooper, 2013), discussed later in the chapter. 16  For the background to the 2009 Act and the changes to the sentencing environment it introduced, see Ash­ worth, 2013b: 24–25; 2010b: 389–90; Ashworth and Roberts, 2013: 5; 2012: 882; Easton and Piper, 2016: 50–51; Roberts, 2013a: 3; 2012a: 334–35; 2012b: 270–71; 2011a: 998–99 and 1006–09; and Roberts and Ashworth, 2016: 334–35. 17  [2013] 1 Cr App R (S) 33 at [5]. For commentary, see Thomas, 2012b.

The Pursuit of Consistency in Sentencing

 147

In R v Blackshaw,18 where the Court of Appeal reviewed a number of sentences for offences committed during the riots of August 2011 (see generally Roberts and Hough, 2013: 236), the Lord Chief Justice (Judge) stated that the statutory requirement to ‘fol­ low’ guidelines does not require ‘slavish adherence’, not least because of the presence of the ‘interests of justice’ exception. His Lordship continued: The court should approach the sentencing decision by reference to any relevant guidelines … This provides the starting point and it produces the desirable consistency of approach to sentencing decisions up and down the country without sacrificing the obligation to do justice in the individual and specific case. The often quoted aphorism, that sentencing guidelines are guidelines not tram­ lines, continues to be fully reflected in the present legislative framework.19

The Lord Chief Justice then referred to the decision in R v Height20 in which the Court of Appeal observed that, with regard to the statutory provisions governing sentencing in cases of murder, such provisions do not require an inflexible approach: even if they appeared to be detailed and comprehensive, the overriding obligation of the sentencer is to ‘achieve a just result’. His Lordship concluded that the legislation relating to Sentencing Council guidelines could not ‘impose a more rigid system than that which applies to the statutory sentencing framework created for sentencing in murder cases’.21 Thus, whilst it is important that sentencers do have regard to the relevant guidelines, they are merely guidelines: the different levels and starting points in them should not be viewed in a purely mechanical or mechanistic way (see Cooper, 2013: 158; Edwards, ibid 345 and 2013: 80; and Woolf, 2008a: 296). Notwithstanding the statutory duty to follow the guidelines, the Court of Appeal has continued to stress the importance of doing justice in the individual case. In AttorneyGeneral’s Reference (Nos 15, 16 and 17 of 2012),22 for example, the Court stated that the Sentencing Council’s Definitive Guideline on drug offences (Sentencing Council, 2012a) was ‘not a statute requiring strict statutory construction’ and continued: ‘… the Guideline does not provide a series of boxes into which every offender and every offence must be squeezed with no exceptions permitted. A judge must obviously explain any departure from the ranges provided but departures are possible …’. In the course of considering the application of the Sentencing Council’s Definitive Guideline on burglary offences (Sentencing Council, 2012b) in an ­unsuccessful appeal by a recidivist burglar, the Court in R v Brooke23 observed that ‘the Definitive Guideline does not remove from a sentencing judge the discretion to do justice’. Similarly, in R v Marcantonio24 (another unsuccessful appeal against sentence by a ‘career burglar’), it was observed that there will be cases where the offender’s record, combined with his inability or unwilling­ ness to rehabilitate himself and respond to non-custodial options, will entitle the sentencer ‘not just to depart from the relevant sentencing guidelines, but to depart radically from

18 

[2012] 1 Cr App R (S) 114. Ibid at [14]; see also the earlier decisions to the same effect in R v Rendell [2011] EWCA Crim 1438 at [12]; R v Millberry [2003] 2 Cr App R (S) 31 at [34]; and R v Peters (n 7) at [3]. 20  [2009] 1 Cr App R (S) 656 at [29]. 21  R v Blackshaw (n 18) at [15]; cf Ashworth, 2013b: 25 and 2012: 81–82, 94–95. 22  [2013] 1 Cr App R (S) 52 at [17]. 23  [2012] EWCA Crim 1642 at [19]. 24  [2012] EWCA Crim 1279 at [12]. 19 

148 

Structuring the Sentencer’s Discretion

those guidelines so that there can be no doubt that the protection of the public is being considered first and foremost’. Thus, whilst courts must follow the approach set out in the guidelines, the resulting sentence must reflect the fact-specific circumstances of the offence—an exercise which involves the exercise of judicial discretion (Banks and Harris, 2013: 1–670). The second key change introduced by the 2009 Act was the replacement of the SAP and the SGC by a single authority, the Sentencing Council of England and Wales. The creation of a single guidelines authority was intended to promote more effective development and dissemination of guidelines (Roberts, 2013a: 3). The guidelines issued by the SGC are to remain in force until such time as the Sentencing Council revises and re-issues them, a process expected to take several years (Ashworth and Roberts, 2012: 883 and 2013: 5, 11; Roberts and Ashworth, ibid 335; Maslen and Roberts, 2013: 126).25 The Sentencing Council began to issue its own guidelines in 2011.26 They follow a dif­ ferent structure from guidelines issued by the SGC by setting out a multi-staged process for determining sentence; in so doing, they seek to promote ‘uniformity’ and consistency in sentencing (Roberts, ibid 5, 22). Section 120 of the 2009 Act requires the Council to prepare sentencing guidelines which may be general in nature or limited to a particular offence, category of offence, or particular category of offender. Wasik explains that under section 121, the guidelines should specify the ‘offence range’ appropriate for a court to impose on an offender convicted of that offence and, if the guidelines describe different categories of case, they should specify for each category a ‘category range’ within the offence range (Wasik, 2014: 48–49). The guidelines should also specify the ‘starting point’ within the offence range or within each category range (Wasik, ibid 49; see also Maslen, 2015: 173). The guideline for assault, for example, came into effect on 13 June 2011 (Sentencing Council, 2011). Roberts reports that it will serve as a model for all future guidelines issued by the Council (ibid 4). The guideline identifies three overlapping ranges of sentence length, each range relating to a separate category of seriousness. Sentencers are directed to under­ take a nine step process in arriving at sentence, of which the first two are the most critical (Ashworth and Roberts, 2013: 6; Hutton, 2013a: 94–98; Maslen and Roberts, 2013: 131; Roberts, 2012b: 275–78; 2011a: 1011; Roberts and Rafferty, 2011: 682). The process commences with a determination of which of the three levels of serious­ ness is appropriate. Once the sentencer has determined which category range is appropri­ ate, he or she will use the starting point sentence within the range as a point of departure (step one). The sentencer is then required to ‘fine tune’ the sentence within the chosen range by considering other factors relating to the seriousness of the crime as well as per­ sonal mitigation (step two), before following a series of seven further steps to determine the final sentence (Ashworth, 2010b: 391; Ashworth and Padfield, 2015: 657; Ashworth and Roberts, 2012: 883–84 and 2013: 5–9; Dhami, 2013b: 174; Irwin-Rogers and Perry, 2015: 194–95; Krasnostein and Freiberg, 2014: 15; Padfield, 2013: 38; Roberts, 2013a: 8–10; Rob­ erts and Bebbington, 2013: 335–36; Roberts and Rafferty, 2011: 682–86; and Wasik, ibid

25  On the operation of the transitional provisions, see Thomas, 2010: 7–8. On the Court of Appeal’s power to issue interim guidance, see Ashworth, 2013b: 25–26. 26  For a discussion of the Council’s other statutory duties, including monitoring the operation and effect of guidelines, see Ashworth and Roberts, 2013: 10–11; Pina-Sánchez and Linacre, 2013: 1119; Roberts, 2013a: 3–4, 2013b: 105–06; and Roberts and Ashworth, 2016: 343.

The Pursuit of Consistency in Sentencing

 149

51–53). These remaining seven steps—described by Hutton (2013a: 96) as being required mainly for ‘technical/legal reasons’—are as follows: consideration of whether the sentence should be reduced to reflect assistance provided to the prosecution or police; consideration of whether to reduce the sentence for a guilty plea; consideration of whether to impose an extended sentence; consideration of the totality principle and whether the sentence is proportionate; consideration of whether to make a compensation or other ancillary order; explanation of the effect of the sentence; and consideration of any time served on remand (for a full discussion, see Roberts, ibid 10–11 and Ashworth and Roberts, 2013: 8–9). The drafting of section 125 of the 2009 Act has, however, been criticised as nothing within the legislation imposes on the court a separate duty to impose a sentence which is within the category range. The duty to follow guidelines is tethered to the offence range, which is usually very wide, and fails to require a court to explain why it has moved out of the applicable category range (­ Ashworth, 2010b: 395–96; see also Roberts and Ashworth, 2016: 337). Thus Thomas contends that the duty to follow guidelines has been ‘diluted … to the point of meaninglessness’ (2010: 7), whilst Ashworth criticises the wording of the 2009 Act as ‘pitifully loose’ (ibid 395)—a view shared by Hutton (ibid 93) and Young and King (2013: 213). The format of the guidelines and their operation has also been criticised by academics and practitioners. Lyndon Harris and Felicity Gerry QC (both experienced criminal bar­ risters) criticise the Council’s adoption of what they term ‘a template-style approach to the production of guidelines’ (Harris and Gerry, 2013: 240). The authors note—by reference to previous sets of guidelines issued by the SGC—that ‘it cannot be thought that the sentenc­ ing of actual bodily harm, failure to control a dog, importation of cocaine and controlling a brothel for prostitution should be approached in the same way’ (ibid 240–41). Harris and Gerry consider that in issuing such guidelines, the Sentencing Council has adopted a mechanistic and restrictive approach that fails to appreciate the often complex nature of offending; such a ‘broad brush’ methodology has the potential to result in serious injustice (ibid 239; see also 241). The authors continue: We consider that the ranges and starting points system is an artificial method of calculating sen­ tence. It places undue emphasis on a mathematical approach in that, once the category is selected, movement up or down within the range is based on reference to a list of factors. There is no assis­ tance given to the sentencer when considering the weight to attribute to each factor that is present … The focus is placed on the presence or absence of the factors, as opposed to identifying the real feature of the offending … What is required at sentence is a true examination of the offending behaviour in order to ensure the correct sentence is imposed (ibid 241–42; see also Harris, 2013: 847 and 2016: 73).

Nicola Padfield (an academic and barrister who also sits as a part-time recorder) considers the Council’s guidelines ‘less easy to apply’ and ‘more difficult to use’ than the guidelines issued by the SGC (Padfield, 2013: 35; see also Ashworth and Padfield, 2015: 658 and Harris, 2015: 552). That the guidelines issued by the Sentencing Council are becoming ‘increasingly complex’ is acknowledged by Professor Julian Roberts, himself a member of the Council (see Roberts and Hough, 2015: 13). Experience in the criminal courts has shown that the problem with what may be termed the mechanistic approach of the Council’s guidelines is that, in many cases, sentencing hearings have become ‘unseemly exercises in trading figures’ that are not conducive to

150 

Structuring the Sentencer’s Discretion

public understanding of sentencing (Harris and Gerry, ibid 240). In many situations, use of the Council’s guidelines hinders the sentencer’s ability to do justice (ibid 242). Harris and Gerry conclude that the present format of sentencing guidelines has ‘led to an over-depend­ ence on mathematics and undue criticism of judges who take a fact-based approach’ (ibid). David Thomas QC criticised the nine step process required by the guidelines as ‘increas­ ingly mechanistic’ and as ‘time-consuming and not often productive’ (Thomas, 2013: 1000). For Thomas, the new system of guidelines has the potential to turn judges into ‘mere technicians oiling the wheels’ (ibid; cf Treacy, 2014: 298). The new format was also criticised by the Court of Appeal (Criminal Division) in R v Perkins and others27 as being ‘something of a distraction’ and ‘in some ways academic’. John Cooper QC persuasively argues that the new guidelines format focuses on offence severity and offender culpability at the expense of personal mitigation (Cooper, 2013: 158), a view shared by Harris and Gerry (ibid 240; see also Harris, 2013: 848). The Council’s approach is, on the whole, victim-centric (Cooper, ibid). Whilst the approach of the SGC was to provide guidance by means of specific fact-based scenarios, the court is now encour­ aged to weigh harm and culpability equally in order to reach a specific category of offence within the guidelines (ibid 159). As seriousness is the overwhelming determinant of sen­ tencing severity, the influence of personal mitigation has been marginalised and reduced to the extent that it is hardly mentioned in any of the Council’s guidelines (ibid). Cooper considers that the Council interprets personal mitigation very narrowly (ibid 160). At step two of the process in the definitive guideline for assault (see above), for exam­ ple, eight factors are listed which are relevant to personal mitigation whilst the detailed list of aggravating factors extends to 20 (Cooper, ibid). Thus, the focus is clearly on aggravating circumstances (ibid). Personal mitigation is, however, a more complex con­ sideration than the Council seems to acknowledge. The focus on aggravating factors at the expense of mitigation clearly affects the ability of sentencers to respond to the charac­ teristics of the individual offender, particularly if the offender’s needs are best addressed though community orders (ibid). Cooper stresses that the pursuit of fairness and justice must include a central role for personal mitigation (ibid 164). In a scathing criticism of the Sentencing Council’s approach, Cooper concludes: The sentencing of offenders is a complex and individual exercise; there is no blue-print or a ‘one size fits all’ solution. In the Council’s quest for consistency the complex and idiosyncratic nature of offenders has been shelved in place of a process which, it is argued, can be easily understood by the public. This approach is superficial, and perhaps worse, it is patronizing. Sometimes, sentenc­ ing cannot be reduced to this two dimensional approach … [I]t is time to acknowledge that the sentencing exercise is often difficult, sometimes complex, and always individualized (ibid).

The current system of English sentencing guidelines arguably seeks consistency at the expense of individualised justice. As O’Malley argues, however, in echoing the views of Cooper (ibid): ‘Justice is always preferable to consistency, and the more courts are obliged to comply with formal sentencing norms, the greater the risk of injustice in individual cases’ (O’Malley, 2011: 253). In earlier discussions of the guidelines system under the SAP and SGC, Ashworth notes that the intention was to strive for a formulation that gave ‘sufficient structure without 27 

[2013] 2 Cr App R (S) 72 at [52].

The Pursuit of Consistency in Sentencing

 151

being overly detailed’ (2008: 118; 2006: 16). Similarly, Roberts et al caution that guidelines should not be ‘heavily deterministic or prescriptive’ (Roberts et al, 2011: 528). How can a system which obliges the court to follow a series of nine steps in arriving at the sentence be described as anything other than overtly detailed, deterministic and prescriptive? Is the system a ‘straitjacket’ for sentencers (Charleton and Scott, 2013: 13)? It is to these questions that the discussion now turns.

The Continued Importance of Individualised Sentencing The new scheme introduced by the Coroners and Justice Act 2009 has been lauded by leading academic commentators in England as an ideal model, an exemplar for other jurisdictions to follow or at least monitor before introducing their own schemes to structure judicial discretion in sentencing (Ashworth and ­Roberts, 2012: 892 and 2013: 11; Roberts, 2013b: 120–21, 2012a: 320 and 343, 2012b: 267–68, 281 and 283; Roberts and Ashworth, 2016: 334; Roberts and Bebbington, 2013: 337–38; Roberts and Pei, 2016; and Roberts and Rafferty, 2011: 688).28 As the findings of the current study demonstrate, however, academic confidence that other jurisdictions will either take up or adopt versions of the English system of guidelines is misplaced, at least insofar as Scottish judges and sher­ iffs are concerned. It is contended that—pace the realistic assessment of the English guide­ line schemes under the SGC and the Sentencing Council by Cooper (2013, 2008), Harris and Gerry (2013) and Thomas (2010)—much academic writing on the subject has been, as Ashworth states in a different context, ‘too doctrinaire and too remote from the myriad considerations pressed upon courts in individual cases’ (Ashworth, 2001: 86). For Roberts, the provisions focus a court’s attention on the relevance of the guidelines, yet also allow significant judicial discretion to impose a fit sentence (Roberts, 2012a: 337). Ashworth also maintains that there is no insoluble conflict between guidelines and the appropriate use of judgment. For Ashworth, the key is process: sentencers should fol­ low a decision making process that incorporates all the relevant steps and which brings transparency to the decision ­(Ashworth, 2010b: 399). Whilst Roberts and Rafferty main­ tain that presumptively binding numerical guidelines provide courts with guidance as to the appropriate d ­ isposals for most cases (ibid 687), such guidelines singularly fail to serve the courts, the legal profession and the accused in cases which do not fit any stand­ ard pattern, and some would argue that these form the majority of criminal cases.29 As O’Malley observes: ‘No statute or guideline system, no matter how finely tuned, can cater in advance for the unique circumstances of every offender who will come before the courts for sentence’ (O’Malley, 2006: ix; see also ­Pomerance, 2013: 315). The staged approach to sentencing is inimical to traditional approaches that view sen­ tencing as a ‘human process’ (Cooke, 1987: 58; Hogarth, 1971). Despite the empirical studies already considered and the various dicta from the Australian and Canadian courts which all demonstrate, firstly, that sentencing cannot be reduced to the mechanical application of rigid 28  It may, however, be that a comparative examination of Scottish and Commonwealth sentencing jurispru­ dence has a contribution to make to the development of English sentencing law and practice—see Brown, 2013d and Lord Gill, 2016: 19–20. 29  Cooper, 2008: 280; R v Charles [2012] 1 Cr App R (S) 74 at [26].

152 

Structuring the Sentencer’s Discretion

formulae and, secondly, that the key factors are the sentencer’s professional experience, ­intuition and subjective judgment,30 the current system of English guidelines structures the decision making process by effectively creating an ‘algorithm’ for sentencers to follow ­(Roberts, 2011a: 1011; see also Hutton, 2013a: 95)—the judgment of experienced sentencers counts for less and less (Thomas, 2013: 999). The result is a detailed and relatively prescrip­ tive scheme (Roberts, 2012b: 283) completely alien to the approach of the courts in Scotland, Canada, South Africa and Australian jurisdictions. To date, England is the only jurisdic­ tion outside of the United States to develop such formal, presumptively binding numerical guidelines ­(Ashworth and Roberts, 2013: 2; Gazal-Ayal, 2013: i–ii; Krasnostein, 2015a: 165; Roberts, 2013a: 1, 2012b: 267; Roberts and Ashworth, ibid 310, fn 3; and Wiles, 2015: xi). Within the common law world, such guidelines remain exceptional (O’Malley, 2013a: 219–20). The current system of English guidelines is arguably an example of what P ­ osner (2013) terms ‘internal complexity’. Posner distinguishes between two kinds of complexity: external and internal. Whilst the law regulates complexity that is external to the legal system, it does not create it (ibid 4). With regard to sentencing law, this external complexity comprises what Lord Justice Clerk Carloway (as he then was) described in Ferguson v HMA31 as the ‘complex matrix of factual and legal material’ which the sentencer must consider in making his decision (see also Posner, ibid 15). For example, does the offender have any previous convictions; are his convictions analogous to the present offence; how old is he; is he in employment; does he have any dependents; was he in a position of trust when the offence was committed; what risk does he pose to the public; what harm was sustained by the vic­ tim? In addition, the legal system generates its own complexity—‘internal complexity’— which makes it more, rather than less difficult for judges to cope with external complexity: Abetted by lawyers, law professors … and legislators, judges ‘complexify’ the legal process need­ lessly—and do so in part to avoid the struggle to understand the complex real-world environment that generates much of the business of a modern court … [Judges] escape from complexity into complexity (Posner, ibid 4, 14).

That legal responses to the external complexity of the real world can take the form of internal complexity can be illustrated by the current system of English sentencing guidelines. The law of sentencing in England and Wales is largely, in Posner’s terms, ‘a compendium of texts’ (in the form of Definitive Guidelines issued by the Sentencing Council). The task of the judge is to ‘discern and apply the internal logic of the compendium’ (Posner, ibid 4–5). The judge is, as Posner explains, ‘an interpreter, indifferent or nearly so to the consequences of his interpretations in the real world’; the judge who takes into account the consequences of alternative interpretations—in the sense of disregarding the guidelines to achieve indi­ vidualised justice—is, on this view, stepping outside the law (ibid 5). Posner decries such responses as ‘gratuitously creating internal complexity’ (ibid 65); they direct judges to ‘master and apply a baffling and ultimately fruitless system for avoiding 30  Discussed under the heading ‘The “art” of sentencing’, in Chapter 5. It should also be noted that an empirical study of judges’ sentencing practices during the ‘second phase’ of English guidelines under the SAP found that there were gaps in the guidelines and that judges were still required to ‘use their experience and knowledge to relate guidelines to all the facts’ (Davies and Tyrer, 2003: 263). Such judicial knowledge and experience will, I suggest, remain important features of the current English guidelines system (see the discussion of the operation of sections 124(8) and 125(1) of the Coroners and Justice Act 2009 later in the chapter). 31  2014 SCCR 244 at [104].

The Pursuit of Consistency in Sentencing

 153

engagement with reality’ (ibid 235). As Yoshino explains in an observation that is particu­ larly germane to the system of English guidelines in light of the practitioners’ criticisms already considered, legal responses involving internal complexity are ‘a neurotic reaction to external complexity; if you cannot control what is important, you make important what you can control’ (Yoshino, 2013: 13, emphasis added). Posner also considers that the more internal complexity grows, the greater its costs become relative to the benefits that it generates (ibid 96). For critics such as Cooper (2013, 2008), Harris and Gerry (2013) and Thomas (2010), the main cost of the internal com­ plexity generated by the English system of guidelines is that consistency in sentencing is achieved at the expense of justice. The present English system of sentencing guidelines differs from Posner’s model in one important respect. Posner’s criticism is aimed at the decision making of US judges and thus, for Posner, internal complexity is created by judges themselves along with academic lawyers, practising lawyers and the legislature (ibid 4, 59). The English sentencing guideline system, however, is a creature of statute and so the internal complexity of English sentenc­ ing guidelines is politically, not judicially, generated. In any event, it is interesting to note that under section 124(8) of the Coroners and Jus­ tice Act 2009, the power of the Court of Appeal to issue sentencing guidance is explicitly preserved. Thus, the Court may still issue its own guideline judgments as it did in the first ‘phase’ of the English guideline movement. As Lord Judge CJ observed in R v Blackshaw:32 [N]othing in the 2009 Act has diminished the jurisdiction of this court, where necessary, to prom­ ulgate judgments relating to the principles and approach to be taken in sentencing decisions. They bind sentencing courts. The relationship between this court and the Sentencing Council proceeds on the basis of mutual respect and comity.33

The jurisdiction of the Court of Appeal to issue a guideline judgment on its own initiative enables sentencing guidance to be promulgated when there is an urgent need for it, or when the Court believes that the Sentencing Council will be unable to consider a particular issue due to pressure of business (Judge, 2014: 680–81). Banks and Harris note that the fact that the Court of Appeal can issue its own guidelines runs counter to the belief of many in Par­ liament and elsewhere that the 2009 Act was intended to replace the Court’s role in issuing sentencing guidelines (2013: 1-675). The Court’s re-affirmation of its power to issue such guidelines is, however, commended by the authors as ‘exceptionally welcome’ (ibid). Such guidance rests, as continues to be the case in Scotland, on the authority of precedent rather than on any definitive guidelines which are binding by virtue of statute (Ashworth, 2013b: 27, 2010b: 393). In addition to the multi-staged, statutory form of guidelines, it seems likely that judicial self-regulation through appellate review—‘guidance by words’ (Ashworth and Roberts, 2012: 881; Duff, 2005: 1179; Roberts, 2012a: 339; and Roberts and Baker, 2008: 566)—will continue to have an important role in the English sentencing environment (Cooper, 2013: 157; Thomas, 2011b: 418–19)34 despite Roberts’ criticism that this approach ‘leaves too much room for individual judges’ (Roberts, ibid).35 32 

Blackshaw (n 18) at [15]. See also R v Thornley [2011] 2 Cr App R (S) 62 at [14]; Attorney-General’s Reference (Nos 3, 73 and 75 of 2010) [2011] 2 Cr App R (S) 100 at [5]; and R v Dyer [2013] EWCA Crim 2114 at [13]–[15]. 34  See also R v Caley [2013] 2 Cr App R (S) 47 and R v Kahar and others [2016] EWCA Crim 568. 35  Thomas has also questioned whether formal sentencing guidelines are the best way to produce consistency in sentencing given the broadly similar facts but ‘spectacularly different outcomes’ of the decisions in R v Charles (n 33 

154 

Structuring the Sentencer’s Discretion

As Thomas astutely observes, whatever changes section 125 of the 2009 Act may make to the practice of sentencing, the ‘judicial discretion escape clause’36 in ­section 125(1)—that allows the court to decline to follow a guideline if it is satisfied that it would be contrary to the interests of justice to follow it—preserves the principle that has always been at the core of sentencing, namely: [T]hat the determination of the sentence to be imposed in any particular case is ultimately a matter for the judgment of the sentencing judge, in the exercise of a discretion informed by principle and judicial experience. Whatever sentence may appear to be indicated by the guidelines, the final duty of the sentencing judge is to consider whether that sentence is compatible with what were recently described as ‘the ancient principles of justice and mercy’ [R v Hussain and Hussain, [2010] EWCA Crim 94, per Lord Judge CJ at [44]]. If it is not, it is the duty of the judge to ignore the guideline and pass a sentence which is (Thomas, 2010: 9; see also Banks and Harris, 2013: 1–668 and Thomas, 2013: 1000).

But how does a judge ascertain whether a sentence generated by the nine step process mandated by a Sentencing Council guideline is indeed compatible with justice and with mercy? Arguably this decision can only be made through phronesis and by having recourse to equity, for as Lord Justice Clerk Carloway (as he then was) noted in Ferguson v HMA: ‘Sentencing is “a delicate art based on competence and expertise” … rather than an exact science. A judge should not have to go through a formal check list of procedures before arriving at the appropriate and proportionate sentence’.37 For all the dizzying complexity of the nine stage approach to sentencing in modern Eng­ lish sentencing guidelines, the inclusion of the section 125(1) ‘escape clause’ preserves the importance of individualisation, the importance of judicial discretion and the importance of the phronetic synthesis in sentencing.

A Blending of the Phronetic Synthesis with Sentencing Guidance—Principled Discretion in Scottish Sentencing The approach of the Scottish judiciary to sentencing is similar to that of the Australian courts: sentencers in both jurisdictions are firmly of the view, firstly, that the retention of a wide judicial discretion is necessary and, secondly, that individual justice is possibly more important than some more abstract notion of systemic fairness (Freiberg, 2014: 228, 2001: 35). ‘Individualism’ is the dominant sentencing paradigm both in Australia (Freiberg and Krasnostein, 2011: 76; Krasnostein, 2015a: 18) and in Scotland (Hutton, 1999: 173; Tata, 2010a: 197). Judicial discretion is seen as a crucial component of fair sentencing as it

29) and Attorney-General’s References Nos 11 and 12 of 2012 [2013] 1 Cr App R (S) 43—viz 12-month suspended sentences and seven years’ imprisonment respectively for offences of rape of a child under 13 (see Thomas, 2012c: 726). 36 

The term is used by Roberts, 2011a: 1006. Ferguson (n 31) at [103] citing Lord Justice Clerk Gill (as he then was) in Gemmell v HMA 2012 JC 223 at [59]. Whilst Lord Carloway’s remarks were made in the context of judicial consideration of risk assessment reports and the imposition of orders for lifelong restriction, they apply to the task of sentencing more generally. 37 

Principled Discretion in Scottish Sentencing

 155

enables abstract legal rules to be applied to real life offences (Krasnostein and Freiberg, 2013: 267; see also Krasnostein, ibid 21). The courts view the exercise of broad judicial discretion, the principle of individualised justice, and the attainment of fair sentencing outcomes as being closely, if not directly, related (Freiberg, 2014: 19; Krasnostein, ibid 18; Krasnostein and Freiberg, ibid). The Scottish courts thus favour individualised justice over consistency in sentencing. Judges and sheriffs in Scotland have long enjoyed a wide sentencing discretion, restricted only by appellate review, a very small number of mandatory sentences38 and, since 2003, a requirement to consider the granting of a discount in sentence following the tendering of a guilty plea.39 The principal means by which the Appeal Court in Scotland has in recent years sought to ensure a measure of consistency in sentencing is through appellate review and the use of non-binding guideline judgments.

Appellate Review All common law jurisdictions rely to some extent on appellate guidance as a means of structuring sentencing discretion (O’Malley, 2006: 57). Appellate review assists in the inter­ pretation of guidelines and sentencing legislation, as well as developing a body of case law derived from judicial experience in individual cases (Dhami, 2013a: 304). It is a process that can promote consistency, firstly by acting as a check on individual sentences in the lower courts, and secondly by correcting the course of sentencing practices by providing guidance to sentencers via statements of policy or principle, often in the form of guideline judgments (Krasnostein and Freiberg, 2013: 274). As Chanenson explains: Appellate courts should be key players in the consultative and interactive process of sentencing guidance and communication. Appellate review ought to be the fulcrum around which guided sentencing systems revolve. With their dual focus on establishing broad principles of sentencing law and evaluating individual cases, appellate courts can bring a distinctive voice to the sentencing discussion (Chanenson, 2005: 177).

In the absence of a body to promulgate sentencing guidelines in Scotland, appellate review has traditionally been the primary method of ensuring consistency. Exclusive reliance on appellate review can, however, result in a very fragmented system of sentencing guidance (Crowe, 2013a: 26; O’Malley, ibid 63). As O’Malley notes, simple appellate review must be supplemented by a structure that allows for a more comprehensive review of sentencing practices (ibid). The means by which this has been attempted in Scotland is by the issuing of guideline judgments.

Guideline Judgments Guideline judgments are judgments of appellate courts which go beyond the facts of the particular case before the court by articulating sentencing principles (Findlay et al, 2014: 269; Freiberg, 2014: 977–78). This may be done by considering numerous variations of a 38  39 

See Chapter 4. See Chapter 7.

156 

Structuring the Sentencer’s Discretion

specific offence, the importance of aggravating factors and of factors commonly raised in mitigation, and by discussing the relevance of different sanctions to the particular offence or by suggesting appropriate sentencing ranges (Dingwall, 1997: 144; Edney and Bagaric, 2007: 37; Krasnostein, 2015a: 163 and 2015b: 42; Krasnostein and Freiberg, 2013: 281; and Warner, 2003: 9). As in other jurisdictions which favour individualised sentencing over consistency and in which guideline judgments have been issued by the appellate courts, the principal objective in issuing such judgments has been to structure, but not restrict or abolish, judicial discretion (see Bagaric and Edney, 2016: 48–49; Colvin, 2003: 86; Findlay et al, ibid; Mackenzie and Stobbs, 2010: 32; Morgan and Murray, 1999: 94–95, 106; Shroff, 2003: 316; and Spears, 1999: 300). In issuing guideline judgments, the appellate courts thus intend to redress patterns of inconsistency, undue leniency or severity by setting appropriate standards to guide the exer­ cise of the sentencing discretion (Freiberg, ibid 978–79; ­Krasnostein, 2015b: 42). Guideline judgments also promote the rule of law, transparency in the decision making process, and public confidence in the criminal justice system (Krasnostein, ibid). As the Supreme Court of Victoria recently observed in the course of delivering its first guideline judgment: The provision of a guideline judgment can promote consistency and public confidence in the sen­ tencing process by articulating elements that must be taken into account in a particular sentencing context, and by giving guidance as to a unified approach. It can also facilitate the development of coherent sentencing practice by way of unified application of principle and, in turn, assist the iden­ tification of relevant similarities and differences between cases.40

In Scotland, primary legislation—sections 118(7) and 189(7) of the ­Criminal Procedure (Scotland) Act 1995, relating to solemn and summary appeals r­ espectively—provides that, in disposing of an appeal, the Appeal Court may pronounce an opinion on the sentence or other disposal which is appropriate in any similar case. Sentencers must ‘have regard’ to such opinions when issued (section 197 of the 1995 Act). Thus, where the Appeal Court issues such a guideline judgment in terms of the legislation, the lower courts are obliged to have regard to it. This can be contrasted with the system in England and Wales from the 1980s onwards where the Court of Appeal’s guideline judgments, although generally followed by the lower courts, were, in strict legal terms, obiter dicta and therefore not binding (Ashworth, 1998b: 229; see also O’Malley, 2016: 20). Thus, whilst the current Scottish system of sentencing guidance is modelled on the early English approach, it also incorporates the statutory provision directing lower courts to have regard to the appellate guidance which was introduced in England along with the establishment of the Sentenc­ ing Advisory Panel. The Scottish system of sentencing guidance is therefore essentially one of judicial selfregulation through appellate review and judicially imposed guidelines. It is a technique that works well in a small jurisdiction such as Scotland where the Appeal Court is experienced in delivering principled judgments (Ashworth, ibid 236; see also Anderson, 2004: 152–53; and Krasnostein and Freiberg, ibid 273, fn 76). The system is not nearly as intricate as the procedure adopted in jurisdictions such as Sweden or Finland, where the aims of sentenc­ ing and detailed principles, including guidance on how to resolve conflicts of principle, are

40 

Boulton v The Queen [2014] VSCA 342 at [40].

Principled Discretion in Scottish Sentencing

 157

set out in legislation (see Ashworth, 2009: 248–49; Duff, 2005: 1168–69; Jareborg, 1995; von Hirsch and Jareborg, 2009 and 1989). What the Scottish approach does, however, is establish a system of sentencing guidance through appellate judgments. It does so in a manner broadly consistent with the ‘first phase’ of the English approach discussed above. The courts regulate their own sentencing prac­ tices, within the boundaries set by the legislature (Ashworth, ibid 243), in an area in which sensitive, individual calculations—which necessarily involve elements of discretion—are key (Cooper, 2008: 278; Muller, 1993: 305). This is an approach which has the advantage of providing judges with a framework within which they can locate the individual case, without depriving them of the discretion to deal differently with a case which has unusual features (Ashworth, ibid; see also Ashworth, 1998b: 228). Used in this way, guideline judgments produce a discretion that is underpinned by prin­ ciples rather than hemmed in by rules (O’Donnell, 2012: 256). They provide guidance to sentencers in the exercise of their sentencing ­discretion, information for defence agents and counsel representing accused persons, and education for any interested observers (see O’Donnell, ibid). This allows sentencers to retain sufficient flexibility to keep justice to the fore whilst injecting enough consistency to make sentencing outcomes more predictable (ibid 257; see also Stobbs et al, 2014: 158). The very different approaches to the use of sentencing guidelines in ­England and Scot­ land can thus be seen as a dichotomy between the two different methods of judicial decision making in sentencing. The English approach is system-­orientated, abstract and objective, whilst the Scottish approach is more case-orientated, ­concrete and intuitive (see Aas, 2005: 26; Brown, 2014a: 5–6 and 2013a: 7–8; and Flyvbjerg, 2001: 21).

Judicial Receptiveness to Guideline Judgments An important feature of the Scottish approach is that, rather than presenting the lower courts with a stark table of numbers, a grid, or a sentencing matrix, the appellate guidance is narrative and has the familiar form of an appellate court judgment (see Ashworth, 2009: 243 and 1998b: 228). Bottoms noted almost 20 years ago that narrative sentencing guide­ lines are to be preferred over numerical guidelines (Bottoms, 1998: 70). We are looking for sentencers to engage in the trained application of reasons of law to individual cases and this, Bottoms maintains, is more likely to be achieved from a set of narrative guidelines than from numerical guidelines (ibid; see also O’Malley, 2011: 13). The reasons for the English judiciary’s acceptance of early narrative sentencing guidelines of the kind currently operating in Scotland are explained by Ashworth: ‘Guideline judgments were welcomed by most English judges, largely because they were constructed by judges for judges, and also because they were seen as providing a com­ mon framework while preserving flexibility for individual cases’ (Ashworth, 2009: 244; see also Ashworth, 1998b: 229; Anderson, 2004: 151; and Krasnostein, 2015a: 194, 165 and 2015b: 53). As discussed below, respondents in the present study broadly supported the provision of sentencing guidance from the Appeal Court for the same reasons. Such appellate guidance is not subjected to the suspicion and hostility that sentencers sometimes show towards leg­ islative attempts to structure sentencing (Ashworth, 1998b: 229).

158 

Structuring the Sentencer’s Discretion

The Shortcomings of Guideline Judgments The disadvantages of this approach as it formerly operated in England, however, were that guideline judgments tended to be sporadic and developed in a piecemeal fashion, the Court of Appeal having to wait for a suitable case to arise on appeal before a guideline judgment could be issued (Ashworth, 1987b: 545, cited in Anderson, 2004: 154; see also O’Malley, 2016: 18; and Woolf, 2008a: 304 and 2008b: 289). There was no overall strategy, many fre­ quently occurring types of offence were hardly covered at all, and the guidance was framed with input only from judges (Anderson, 2004: 152; Ashworth, 2009: 244; 2008: 112; 2001: 74; ­Dingwall, 2008: 404; Henham, 1995a: 219; Koffman, 2006: 282–83; Tonry, 2002: 76). The system was also reactive, with the Court of Appeal making statements of principle in response to the issues raised by the particular case under consideration (Ashworth, 1998b: 228; see also O’Malley, ibid). As a means of developing sentencing principles, appellate review was thus ‘a slow, haphazard and time-consuming process’ (Fox, 1987: 226–27; see also Krasnostein, 2015a: 102–03 and 192–93). These criticisms can equally be made of the present system of sentencing guidance in Scotland. The fact that a guideline judgment will be issued only when an appropriate case arises on appeal is a particular drawback (see Brown, 2014d: 5). At present, sentence appeals may be identified as suitable for a guideline judgment in one of four ways. Firstly, a suit­ able case may be identified at the sifting stage when leave to appeal is granted (see generally Carloway, 2012: 15–17). A number of cases which raise important points of principle may be continued and heard at a conjoined hearing.41 Secondly, in lodging an appeal against sentence on the grounds of undue leniency, the Crown may invite the Appeal Court to issue sentencing guidance on the offence forming the subject of the appeal. This was done in HMA v Boyle42 where the Crown invited the Court to give guidance on punishment parts in life sentences and in HMA v Graham,43 where the Court was invited to issue updated sentencing guidelines for offences involving child pornography. Thirdly, the Appeal Court comprised of two judges may remit an appeal to a larger Bench ex proprio motu. This can be done if the court considers that the appeal raises questions of principle and that sentencing guidance could be issued in disposing of the appeal.44 Finally, cases may be identified as potentially suitable for guideline judgments by the Lord Justice Clerk’s Judicial Assistants when preparing reports—or ‘Bench memoranda’ as they are referred to in other jurisdictions—on sentence appeals for the Appeal Court. The Judicial Assistants are essentially professional support lawyers to the senior judiciary. They are legally qualified and work under the direction of the Lord Justice Clerk.45 They provide research assistance to the Appeal Court and assist in the preparation of opinions. All of

41 

As with the decisions in Du Plooy v HMA 2005 1 JC 1 and Gemmell v HMA (n 37). 2010 JC 66. 43  2011 JC 1. 44  See, for example, Ogilvie v HMA 2002 JC 74. 45  On the role of Judicial Assistants (otherwise known as Law Clerks, Research Assistants or Judicial Associates) in various common law jurisdictions, see Coonan, 2006; on the role of Judicial Researchers in Ireland, see Charle­ ton and Scott, 2013: 15 et seq, and on the role of Judicial Assistants in the UK Supreme Court, see Darbyshire, 2011: 383–84; Nesterchuk, 2013; and Paterson, 2013: 247–57. 42 

Principled Discretion in Scottish Sentencing

 159

these methods, however, are dependent on the lodging of an appropriate appeal against sentence either by a convicted offender or by the Crown. No guideline judgment can be issued until such an appeal is identified. As O’Malley observes, sentencing systems which operate with wide judicial ­discretion and which rely exclusively on appellate review place the onus on counsel as much as on judges to develop principled sentencing through a dynamic reliance upon existing prec­ edents and the fashioning of new ones (O’Malley, 2016: 802–04 and 2001: 136; see also Ashworth, 1987a: 33). The extent to which the Scottish Bar has discharged this task is ques­ tionable. Whilst in Du Plooy,46 the Appeal Court commended counsel for the four appel­ lants and the Lord Advocate for their assistance (their submissions having examined the position in Scotland regarding guilty plea discounting along with a comparative analysis of practice in England and Australia), in recent years the Court has not always received such comprehensive submissions. In Graham,47 for example, where the Crown had specifically requested sentencing guid­ ance, the Court criticised the Advocate Depute’s inadequate preparation of the appeal. Lord Gill listed the deficiencies in the Crown’s case and noted that, in issuing the guidance requested, the Court had relied to a great extent on insights offered by the solicitor-advo­ cate for the respondent and on the research resources available to it.48 Similarly, in Murray v HMA, Lord Gill noted that counsel had not been instructed in good time, nor had he been properly briefed by his instructing solicitors. His Lordship observed that, ‘something has gone seriously wrong’ and reminded counsel of their duty to adequately prepare in advance of sentence appeals.49 Reference can also be made to Lord Gill’s criticism of counsels’ submissions in Gemmell to the effect that accused persons were ‘entitled’ to a discount in sentence by reason of their early plea of guilty. Whilst this was ‘a careless use of language’, it also demonstrated a mis­ understanding of the relevant statutory provisions and of the substantial jurisprudence on the operation of guilty plea discounts.50 A further disadvantage of the present Scottish system is that appellants, or rather their counsel, may be unwilling to have their case used as the vehicle for issuing a guideline judgment—a problem identified with guideline judgments in other jurisdictions (see, for example, Cowdery, 2006: 16–17). In July 2010, for example, the Appeal Court refused the appeal by Iain Munro against his conviction at the High Court at Aberdeen for indecently assaulting a four year old child.51 The Court continued the appellant’s appeal against the sentence of two and a half years’ detention imposed by the trial judge. The appellant, a first offender aged 16 at the time of the offence, suffered from cystic fibrosis. The sentencing judge had been provided with medical reports detailing the appellant’s condition: it was 46 

Du Plooy (n 41) at [1]. Graham (n 43). Graham (n 43) at [60]. See also HMA v Collins [2016] HCJAC 102 where, in a Crown appeal against the cumulo sentence of six years’ imprisonment imposed on the respondent following his conviction for a series of historic sexual offences including rape, neither the Crown nor the solicitor advocate for the respondent cited any domestic authority. In setting out broad ranges of sentence for sexual offences committed whilst in a position of trust and/or authority, the Court noted that it had been required to undertake its own researches into the matter (at [30]). 49  Murray v HMA 2013 SCCR 88 at [30]–[31]. 50  Gemmell (n 37) at [30]; see also [42]. The issue of guilty plea discounts is discussed in Chapter 7. 51  Munro v HMA [2010] HCJAC 78. 47  48 

160 

Structuring the Sentencer’s Discretion

treatable only in hospital and, in addition to an extensive drug regime and special dietary requirements, the appellant was required to spend approximately two out of every six to eight weeks as an inpatient. It appeared that Polmont Young Offenders Institution (where all young offenders in Scotland are detained) had neither the facilities nor the expertise to care for the appellant. The appellant’s consultant was strongly of the view that the psycho­ logical and physical stresses of a custodial sentence would lead to a worsening of the appel­ lant’s condition and reduce his life expectancy. The sentencing judge nevertheless decided that, given the severity of the offence, the public interest demanded a custodial sentence. In light of the appellant’s health difficulties, however, the judge restricted the sentence to two and a half years. Mindful that the Court of Appeal (Criminal Division) had issued sentencing guid­ ance on the impact of a sentence on an offender in R v Bernard,52 I considered the appeal suitable for a guideline judgment and discussed the matter with the Lord Justice Clerk. His Lordship agreed that the appeal would be suitable for a guideline judgment on the issue of impact mitigation, the only previous Scottish authority on the issue being HMA v McColl.53 The appeals manager and clerk of court continued Mr Munro’s sentence appeal to a hearing before a Bench to be chaired by the Lord Justice Clerk. When the appellant’s counsel and instructing solicitors were informed of this, however, they aban­ doned the appeal against sentence and the issue of impact mitigation never came before the Appeal Court. Counsel’s decision to abandon the appeal against sentence may have been due to the decision of the Court in Graham.54 This guideline judgment had been delivered only nine weeks before the Court refused Mr Munro’s appeal against conviction and resulted in the respondent Graham’s sentence being increased more than 12-fold.55 With the decision in Graham fresh in their minds, it is possible that Mr Munro’s agents and counsel viewed the prospect of a guideline judgment as being synonymous with an increased sentence, leading them to abandon the appeal. As a result, it may be that the majority of guideline judgments will arise from Crown appeals against sentence.56

The Scottish Sentencing Council These difficulties will perhaps be addressed when the Scottish Sentencing Council (‘the SSC’) begins to issue its own guidelines for submission to the Appeal Court. The SSC was set up under Part I of the Criminal Justice and Licensing (Scotland) Act 2010. The 2010 Act was passed by the Scottish Parliament on 30 June 2010 and received Royal Assent on 6 August 2010. Part I of the 2010 Act, however, lay dormant for some five years, primar­ ily due to the cost of establishing the SSC (Carloway, 2013: 25; Charleton and Scott, 2013: 14; Crowe, 2013a: 26; and Tata, 2013: 238, 254). The SSC was eventually established on 19 October 2015 and was formally launched on 17 November 2015 (Nicholson, 2015: 16).

52 

[1997] 1 Cr App R (S) 135. 1996 SCCR 523. 54  Graham (n 43). 55  Graham (n 43) at [59]. 56  As in Graham (n 43) and Boyle (n 42). 53 

Principled Discretion in Scottish Sentencing

 161

The foundations for the establishment of the SSC can be traced back to the Sentencing Commission for Scotland’s report on The Scope to Improve Consistency in Sentencing (Sen­ tencing Commission for Scotland, 2006). The Commission had originally recommended the creation of an ‘Advisory Panel on Sentencing in Scotland’ (APSS). It was envisaged that the APSS would be of similar composition and have similar powers to those of the Sentencing Advisory Panel in England and Wales thereby ensuring that, whilst sentencing would con­ tinue to be a judicial function, the Appeal Court would have the benefit of advice from a body drawing on a range of relevant skills and experience (ibid 46). This was seen as a preferable approach to the creation either of a sentencing commission or of a body such as the Sentencing Guidelines Council in England as the Commission considered that the Appeal Court should be allowed to remain in overall control of sentencing policy (ibid; see also Tata, ibid 249). The Commission recommended, inter alia, that the APSS should comprise around 10 members, including at least one sentencer from each level of the judiciary (ibid); that the APSS should prepare draft sentencing guidelines for the consideration of the Appeal Court on general topics relating to sentencing and new sentencing disposals (ibid 47); and that the Appeal Court should be able to approve the draft guidelines, remit them back to the APSS, or decline to approve them (ibid 48). Sentencers would then be required to have regard to such guidelines approved by the Appeal Court (ibid). The Commission also recommended that the Appeal Court, the Scottish Ministers and the Lord Advocate should all have the power to require the APSS to draft sentencing guidelines on specific topics (ibid 50). These proposals would result in sentencing guidelines being promulgated incrementally by the Appeal Court (ibid). The Commission’s proposals were brought forward as Part I of the 2010 Act. Section 1 of the 2010 Act establishes the SSC. The SSC is chaired by the Lord Justice Clerk. It has six judicial members, three legal members (a prosecutor, an advocate and a solicitor) and three lay members (including a police officer), the inclusion of non-judicial representation hav­ ing been reduced during the original Bill’s passage through Parliament (Hutton and Tata, 2010: 276). The SSC is tasked with promoting consistency in sentencing practice, assisting the d ­ evelopment of policy in relation to sentencing, and promoting greater awareness and understanding of sentencing policy and practice (section 2 of the 2010 Act). The Council is to prepare sentencing guidelines which, following a period of consultation, will be sub­ mitted to the Appeal Court for approval (sections 3 and 4). These guidelines can relate not only to particular types of offence or offender but also to sentencing levels generally and to the principles and purposes of sentencing (section 3). The Appeal Court may then either approve or reject the proposed guidelines in whole or in part, with or without modifica­ tions (section 5). Thus the SSC is in essence an advisory body (Hutton, 2016b: 10). In the event that guidelines are approved, all courts will then be required to ‘have regard’ to them and to state their reasons if the guidelines are not followed (section 6). As the Lord Justice General explains, the guidelines will thus not have binding status (Carloway, 2016: 14). This ensures that the principle of judicial autonomy in relation to sentencing is retained (Welsh, 2010: 810; see also Nicholson, ibid). Under section 8 of the 2010 Act, the Appeal Court is also given the power to require the Council to prepare guidelines or review any existing guidelines. These provisions have been criticised both as being too conservative (by simply per­ petuating the ‘cherished and energetically defended values of unfettered discretion’

162 

Structuring the Sentencer’s Discretion

through a ‘fudged and judicially dominated’ Council—Munro et al, 2010: 273) and as being unnecessary: Most of us are naïve enough to think that such a body [setting the parameters of sentencing] existed called the Appeal Court but it appears that some body with a majority of its members being appointed from Scotland’s good and the great with suitably liberal credentials is now necessary (Aitken, 2009: 767).

In any event, the Lord Justice General appears to have high expectations for the SSC: I have little doubt that, once the Council is established, it will take Scotland into a new era of sen­ tencing: one which will attempt to create a more principled approach and will define, upon the basis of concrete research, what we are trying to achieve and how it can be maintained. It will not eradicate crime but it will advance Scotland into a more civilised era where retribution, other than in relation to the most serious of crimes, will have a smaller plate at the sentencing table (Carloway, 2013: 25–26).

The SSC also has a remit to publish information about sentencing matters, provide general advice and guidance, and conduct research (section 11 of the 2010 Act). At the time of writ­ ing, the SSC had neither issued any proposed guidelines for consultation nor issued any other material relating to sentencing, with one Council member noting that the question of how the SSC’s responsibilities will be carried out ‘is still to be determined’ as it ‘takes its first steps to lay the foundation for future work’ (Hutton, ibid). In a recent address, how­ ever, the Lord Justice General reported that during his own brief chairmanship of the SSC as Lord J­ ustice Clerk, he had suggested looking first at the general principles of sentencing and to those applicable to women and young offenders, rather than ‘engaging in an exercise of selecting gridline penalties’ (Carloway, 2016: 14). Similarly, in an interview given some five months after her appointment as Lord Justice Clerk, Lady Dorrian noted of the work of the SSC that: We’ve been looking at sentencing councils or commissions elsewhere and the kind of guidelines they have. We don’t want just to incorporate some other jurisdiction’s guidelines and put a kilt on them. We want to develop guidelines that are appropriate for the Scottish system and the people it serves (see Nicholson, 2016: 14).

It is notable, however, that in terms arguably redolent of the English Sentencing Council’s emphasis on consistency and uniformity in sentencing, two non-judicial members of the SSC have stressed the importance of ‘enhancing consistency’ in sentencing (Hutton, ibid 11) and of ‘encouraging more consistent approaches’ to sentencing (O’Rourke, 2016: 22), with consistency being seen, by at least one non-judicial member, as the ‘key’ factor in the work of the SSC (O’Rourke, ibid).

The Merits of Judicial Self-Regulation Although judicial self-regulation in sentencing has been criticised as perhaps being an unsuitable vehicle to deliver either consistency of approach or effective change, it has also been acknowledged that self-regulation as a means of delivering sentencing guidance offers an excellent basis for the development of principles that are highly sensitive to the practi­ cal problems of sentencers (Ashworth, 2009: 254). Guideline judgments in England were a

Principled Discretion in Scottish Sentencing

 163

significant innovation and an important tool for the development of consistent sentencing (Ashworth, 1998b: 229; Bingham, 2000b: 46–47). They demonstrated that guidance could be fashioned in a judge-friendly way, based on experience, to shape discretion without unduly constraining it (Ashworth, 2001: 74). Whilst a system such as that currently operating in Scotland which is reliant on appellate review of sentences may be expected to contribute to the practical enforcement of sentenc­ ing guidance, the extent of its contribution will depend on three points: (i) the need for the appellate court to take a strong line against unwarranted deviations from the guid­ ance by lower courts; (ii) the need for appellate judgments to serve as precedents for later cases, rather than merely deciding the particular case; and (iii) a willingness by the appellate court to address issues of principle in its judgments (Ashworth, 1998a: 217; Roberts, 2009: 233–34). To what extent have these aspects been followed by the Appeal Court in Scotland?

Principled Discretion The Appeal Court in Scotland was traditionally slow to issue sentencing guidelines (AN Brown, 2010a: 162; Hutton and Tata, 1993: 231; Maher, 1998: 861–63; Sentencing Commis­ sion for Scotland, 2006: 15; and Tata, 2013: 238). This was partly because it was customary for judges hearing sentence appeals to say no more than was strictly necessary for the dis­ posal of the case, and partly because the Appeal Court adopted what Nicholson describes as ‘a pragmatic and individualised approach to questions of sentence’, deciding cases on their own facts and circumstances rather than on the basis of any declared principles (Nicholson, 1992: 177; see also Hutton and Tata, 2010: 272; and Kelly, 1993: 1). Two important develop­ ments, however, have taken place over the past 10 to 15 years: firstly, an increased use by the Appeal Court of its statutory power to issue sentencing guidelines (as recommended by the Sentencing Commission for Scotland, 2006: 45) and, secondly, an increased willing­ ness to address and articulate questions of ­principle in sentencing, often through a com­ parative examination of sentencing practice in jurisdictions such as England and Wales and Australia. The first case to make explicit use of the guideline provisions in section 118(7) of the 1995 Act was Du Plooy v HMA on the operation of sentence discounts for guilty pleas.57 The Court examined practice in England58 and in Australia59 before issuing sentencing guid­ ance for the Scottish courts. The guidelines on sentence discounting were later refined in Spence v HMA60 and Gemmell v HMA61 with the relevant principles reiterated in Murray v HMA.62 In the earlier decision of Ogilvie v HMA,63 however, the appeal against sentence was remitted to a larger court so that ‘guidelines’ could be given on the appropriate sentence for cases involving downloading child pornography. The Appeal Court subsequently issued

57 

Du Plooy (n 41). See also Chapter 7. Du Plooy (n 41) at [5]–[7], [10] and [17]. Du Plooy (n 41) at [11] and [12]. 60  2008 JC 174. 61  Gemmell (n 37). 62  Murray (n 49). 63  Ogilvie (n 44). 58  59 

164 

Structuring the Sentencer’s Discretion

sentencing guidance without any explicit reference to section 118(7) (see Leverick, 2008a: 44 and 2008b: 308). Updated sentencing guidance for child pornography offences was later provided in HMA v Graham, which was specifically stated to be a guideline judgment in terms of sec­ tion 118(7) of the 1995 Act.64 Having made extensive reference to sentencing practice in England and Wales,65 guidelines were issued by the Appeal Court which focused on rehabilitation for offenders who possessed small amounts of material, with considera­ tions of denunciation and deterrence being appropriate for offenders who possessed and/ or distributed large amounts of material.66 Further sentencing guidelines were issued for the production of cannabis,67 where English practice was also considered; social security fraud,68 with English practice again being considered; and the imposition of punishment parts in life sentences.69 As in jurisdictions such as New South Wales, the Appeal Court has approached the devel­ opment of guideline judgments on a selective and principled basis with guidelines promul­ gated where the Court considered it necessary either to direct sentencers ‘on to the path of consistency in approach’ or to promote equity in sentencing outcomes proportionate to the nature and seriousness of the offending (Anderson, 2004: 144–45). In Spence, for example, the Appeal Court refined and clarified the guidelines on guilty plea discounts because of inconsistencies in the implementation of the earlier Du Plooy guidance in the High Court and the sheriff courts.70 Guidelines for cannabis cultivation were considered necessary in Lin partly because of the degree of disparity in the sentences imposed in the High Court on persons convicted of relatively minor involvement in such ­activity.71 Similarly, the social security fraud sentencing guidelines in Gill were issued due to ‘divergences in view’ between differently constituted sentencing courts.72 Such guidelines can lead to more consistent and appropriate sentences moulded by reference to known criteria (Cowdery, 2006: 14). They may also result in fewer Crown appeals against sentence, whilst allowing defence agents and counsel to predict more accurately the sentence which an offender is likely to receive (ibid). Of equal importance is the emergence of appellate decisions in which questions of prin­ ciple are considered in disposing of sentence appeals. Whilst such cases are not formal guideline judgments in terms of sections 118(7) or 189(7) of the 1995 Act, they are notable both for their comparative consideration of sentencing authority and for the guidance they provide to the lower courts. The number of such cases has increased dramatically in recent years. Guidance has been provided in a series of cases establishing a policy that denunciation and deterrence should be paramount in cases involving the supply of Class A drugs, with custodial sentences

64 

Graham (n 43) at [60]. Graham (n 43) at [17]–[20] and [26]–[27]. 66  Graham (n 43) at [26]–[43]. 67  Lin v HMA 2008 JC 142; see also Du v HMA 2009 SCCR 779 and Shen v HMA High Court of Justiciary on appeal, 14 October 2015, unreported (see the case note by Brown, 2015a). 68  Gill v Thomson, 2012 JC 137; see also Ryan v Murphy 2015 SCL 67. 69  Boyle (n 42). 70  Spence (n 60) at [14]; see also Orr, 2007. 71  Lin (n 67) at [10]. 72  Gill (n 68) at [11] and [19]. 65 

Principled Discretion in Scottish Sentencing

 165

generally being appropriate for such offences.73 A custodial sentence, however, is not inevi­ table following conviction for such an offence, particularly in the case of young offend­ ers. In Smart v HMA,74 the Appeal Court noted that although those convicted of such offences can generally expect to receive a significant custodial sentence in all but excep­ tional circumstances, it is always necessary to consider the individual context in which the youth of an offender would be a very important consideration. Even a very serious offence, which in an adult might require a custodial sentence, might not do so in relation to a young person.

Considerations relating to the sentencing of sex offences were discussed in a n ­ umber of cases.75 The majority of the Court in HMA v Cooperwhite76 expressed an intention to issue sentencing guidelines for rape at a future date.77 Certain guidance on sentencing rape was subsequently issued by the Appeal Court in HMA v AB, HMA v SSK and HMA v Collins.78 In SSK the Court also made certain observations on the levels of sentence appropriate for offenders convicted of lewd, indecent and libidinous practices and behaviour towards chil­ dren where the offending involved penetrative sexual abuse, whilst in Collins broad ranges of sentence were set out for sexual offences committed whilst in a position of trust and/or authority.79 The correct approach to sentencing for firearms offences was set out in Price v HMA,80 with the operation of the statutory minimum sentence for such offences being considered in Lowe v HMA, HMA v McGovern, and Cochrane v HMA.81 Sentencing in cases involv­ ing death by dangerous driving was considered in HMA v Stalker; HMA v Macpherson; HMA v Roulston; Wright v HMA; HMA v Noche; Neill v HMA; Geddes v HMA; and HMA v McKeever;82 causing death by careless driving was considered in HMA v McKay and HMA v McCourt;83 sentencing for the offence of causing death by driving whilst under the influ­ ence of alcohol was considered in Grant v HMA and Fleming v HMA,84 with guidance on levels of disqualification for drink driving being provided in Russell v Troup and Jenkins v Harvie.85 73  HMA v McPhee 1994 SCCR 830; Carlin v HMA 1994 SCCR 763; HMA v Lee 1996 SCCR 205; Slaven v Spiers 2005 SCCR 308; Irvine v PF, Dornoch [2005] HCJAC 53; Bunton v HMA High Court of Justiciary on appeal, 1 November 2005, unreported; Mitchell v PF, Aberdeen High Court of Justiciary on appeal, 27 April 2010, unre­ ported; and Creggy v HMA High Court of Justiciary on appeal, 28 October 2011, unreported. 74  [2016] HCJAC 73 at [14]. 75  Robertson v HMA 2004 JC 155; HMA v Currie 2009 SCCR 48 and HMA v KH 2014 JC 195 (for commentary, see Brown, 2014e). 76  2013 SCCR 461. 77  For commentary, see Brown, 2013e and Crowe, 2013a: 26. 78  HMA v AB 2016 SCCR 47, HMA v SSK 2016 SCCR 74 and HMA v Collins (n 48). The decision in SSK has, however, been criticised as failing to provide sentencing guidelines which bring Scottish sentencing practice into line with practice in other common law jurisdictions such as New Zealand (see Brown, 2016a: 6–7). 79  SSK (n 78) at [24], by reference to the decisions in HMA v Malley 2005 GWD 12-207 and HMA v Briody 2000 GWD 29-1138; Collins (n 48) at [40]–[41]. 80  Price v HMA [2007] HCJAC 44. 81  Lowe v HMA 2008 SCCR 760; HMA v McGovern 2007 JC 145; and Cochrane v HMA 2011 SCCR 63. 82  HMA v Stalker 2003 SCCR 734; HMA v Macpherson 2004 SCCR 579; HMA v Roulston 2006 JC 1; Wright v HMA 2007 JC 119; HMA v Noche 2012 SCL 329; Neill v HMA [2014] HCJAC 67; Geddes v HMA 2015 SCCR 230; and HMA v McKeever 2016 SCL 564. 83  HMA v McKay 2011 SCL 253 and HMA v McCourt 2014 JC 94. 84  Grant v HMA 2013 SCCR 201 and Fleming v HMA 2013 SCL 386. 85  Russell v Troup 2003 SCCR 753 and Jenkins v Harvie 2016 SCCR 268.

166 

Structuring the Sentencer’s Discretion

Offences involving the breach of health and safety legislation in which death or serious injury occurs were considered in HMA v Munro & Sons (Highland) Ltd; HMA v Discovery Homes Ltd; Scottish Sea Farms Ltd v HMA; Dundee Cold Stores Ltd v HMA; and Scottish Power Generation Ltd v HMA.86 Sentencing in cases involving false identity documents was considered in Kunle v HMA.87 The need to deal seriously with those convicted of assaulting public service workers (especially police officers) in the course of their duties was discussed in Townsley v HMA, the Court noting that imprisonment will often be appropriate.88 It was confirmed in McGill v HMA89 that an accused should not generally be given any reduction in sentence to reflect time spent on pre-trial curfew. Issues relating to the sentencing of children were considered in H v HMA (sub nom Hibbard v HMA) and McCormick v HMA.90 Finally, a series of recent decisions has addressed questions of principle in the sentencing of young offenders, holding that rehabilitation rather than denunciation should be the primary aim in such cases.91 The Appeal Court has thus developed a significant corpus of sentencing jurisprudence in recent years. There is a great deal more case law—both in the form of guideline judg­ ments and general discussions of principle—than academic commentators such as Hutton (2008: 144) appear to believe. Pace Lord Judge (Judge, 2014: 678), Scotland does not ‘remain guideline free’. The Appeal Court’s traditional antipathy towards sentencing guidance may have been due to a fear that formal guidelines would inhibit some of the ­flexibility which is a feature of Scottish sentencing (Carloway, 2013: 10; Nicholson, ibid 17892). This concern has been addressed by repeated statements from the Appeal Court emphasising the discretionary nature of guidelines. In HMA v Graham, for example, the Appeal Court stated that sentenc­ ing guidelines provide a structure for, but do not remove, judicial discretion93 and cautioned against an unduly rigid application of guidelines that leads to a mechanistic approach: The responsibility for fixing the sentence in every case rests on the sentencer alone. Sentenc­ ing therefore should always involve the sentencer’s judgment and discretion, which he must in every case exercise by making due allowance for the particular circumstances of the case … [G]uidelines are intended to help sentencers. They are not a straitjacket from which they cannot escape.94

86  HMA v Munro & Sons (Highland) Ltd 2009 SCCR 265; HMA v Discovery Homes Ltd 2010 SCCR 765; Scottish Sea Farms Ltd v HMA 2012 SCL 440; Dundee Cold Stores Ltd v HMA 2012 SCL 1008; and Scottish Power Generation Ltd v HMA [2016] HCJAC 99. 87  2007 SCL 92. 88  Townsley v HMA [2016] HCJAC 87 at [3]. 89  2014 SCCR 46 at [16]. 90  H v HMA (sub nom Hibbard v HMA) 2011 JC 149 and McCormick v HMA 2016 SCL 651. 91  Kane v HMA 2003 SCCR 749; Devine v PF, Stirling High Court of Justiciary on appeal, 12 September 2008, unreported; Graham v PF, Dumbarton High Court of Justiciary on appeal, 9 September 2009, unreported; Lynn v PF, Ayr High Court of Justiciary on appeal, 22 January 2010, unreported; Ashraf v HMA 2010 SCCR 826; Cuthill v HMA 2011 SCL 690; King v HMA High Court of Justiciary on appeal, 7 June 2016, unreported; Ross v HMA [2016] HCJAC 54; Smart (n 74); Finnegan v HMA [2016] HCJAC 88; and Lord Gill’s dissenting opinion in Mitchell v HMA 2012 JC 13. 92  See also HMA v May 1995 SCCR 375 at 378 (Lord Justice Clerk Ross) and, more recently, Lord Eassie’s opin­ ion in HMA v Cooperwhite (n 76) at [28]. 93  Graham (n 43) at [21]. 94  Graham (n 43) at [22], references omitted.

Principled Discretion in Scottish Sentencing

 167

Similarly, in Masson v HMA, the Appeal Court noted that: ‘“guidelines” are just that: guide­ lines … They are not prescriptive’.95 In Jakovlev v HMA, the Appeal Court stated that: While the sentencing judge must have regard to sentencing guidelines and other guidance provided by the appeal court to assist in the exercise of his or her discretion, the sentencing judge alone retains the responsibility for determining the appropriate sentence in any case.96

As Lord Gill observed in Mitchell v HMA,97 sentencing guidelines are always subject to the discretion of the sentencer and, on appeal, to the discretion of the Appeal Court. Experi­ ence demonstrates that there are frequently special factors in individual cases that are not expressly dealt with in guidelines. In HMA v Roulston, an earlier decision considering the application of English sentencing guidelines in the Scottish courts, Lord Gill said this: It is helpful, particularly in offences under UK legislation, to look at the guidelines applied by the English courts and to consider, to the extent that they are relevant, the specific factors on which those guidelines are based. But in doing so the court should not lose sight of its overall duty to assess the sentence that in all the circumstances of the case most justly reflects the culpability of the accused and the mitigating factors, if any, that are found to exist.98

These sentiments were echoed in a series of later decisions by Lord Gill’s successor as Lord Justice Clerk, Lord Carloway. His Lordship explained that English sentencing guidelines often provide a useful cross-check for sentences in Scotland (especially where the offences are regulated by a UK statute and where the sentencing maxima are identical); however, his Lordship cautioned that such guidelines should not be applied in Scotland in a rigid or mechanistic fashion given the differences in sentencing purposes, practices and regimes between the two jurisdictions.99 Thus guideline judgments in Scotland are merely directory and, unlike the system in Eng­ land and Wales, they are not presumptively binding. As in certain Commonwealth jurisdic­ tions such as Canada, New Zealand, New South Wales and, most recently, Victoria, guideline judgments in Scotland are expressed so as not to confine the exercise of the sentencing dis­ cretion (Warner, 2003: 15, 21). They structure judicial discretion by seeking only to assist and guide s­entencers without imposing prescriptive requirements on them (Anderson, 2004: 145; O’Malley, 2016: 19; Sentencing Advisory Council (Victoria), 2016: 22, 130–31). Guideline judgments, as they operate within these jurisdictions, still allow for individualised justice. They allow sentencers to retain sufficient discretion to take each case on its merits and to impose an appropriate sentence that takes account of all the circumstances of the offence and of the offender (Crilly, 2005: 54–56). As Berger JA of the Alberta Court of Appeal noted in R v Lee: ‘The pursuit of justice demands flexibility and the exercise of judgment … [I]t is the essence of the judicial role to depart from a guideline when justice demands it’.100 95 

Masson v HMA, High Court of Justiciary on appeal, 10 August 2012, unreported at [1]. Jakovlev v HMA 2012 JC 120 at [11]; see also Rippon v HMA 2012 SCCR 699 at [3]. 97  Mitchell (n 91) at [15]. 98  Roulston (n 82) at [17]. 99 See Sutherland v HMA 2016 SCCR 41 at [20]; HMA v AB (n 78) at [13]; HMA v SSK (n 78) at [23]; and Scottish Power Generation Ltd v HMA (n 86) at [35] and [38]. Similar remarks were made by differently constituted Benches in HMA v McKeever (n 82) at [16] and HMA v Collins (n 48) at [43]. 100  R v Lee 2012 ABCA 17 at [29], see also [32] and [36] and the unanimous judgment of the Supreme Court of Canada in R v Nasogaluak [2010] 1 SCR 206 at [44]. 96 

168 

Structuring the Sentencer’s Discretion

Guideline judgments are to be taken into account as a ‘check’, a sounding board or a ‘guide’, but not as a ‘rule’ or ‘presumption’.101 Such judgments are not rules of universal application. They can be departed from if, given the particular facts and circumstances, jus­ tice so demands,102 thereby allowing flexibility to ensure that justice and equity are done in the individual case (Anderson, ibid 156; Ashworth, 1984: 523–24; Crilly, ibid; Warner, 2003: 22; and Young, 1999: 13). Guideline judgments thus establish ‘guidelines not “tramlines’’’.103 The current position regarding sentencing guidance in Scotland is akin to what O’Malley (2011, 2001), in his discussion of the sentencing jurisprudence of the Irish courts, refers to as ‘principled discretion’. O’Malley argues that sentencing, in order to comply with the demands of justice, must remain discretionary and so the selection of sentence in specific cases must remain exclusively a judicial task (2011: 5, 253). In order to comply with other important values such as procedural fairness, equality and the rule of law, however, O’Malley considers that sentencing must reform in order to produce consistency of approach, to reduce (if not eliminate) unwarranted disparity and to develop rational criteria for deciding on the nature and severity of appropriate sanctions (O’Malley, ibid 5, 13, 253; 2013b: 532, 2006: x; see also Denham, 2011: vii). How is this to be achieved? O’Malley explains: The challenge is to achieve consensus on acceptable methodologies for assessing offence gravity and other ethically relevant factors, on the weight to be attributed to those factors, and on the kind and quantum of punishment appropriate for various categories of offender. This is the essence of principled discretion (O’Malley, 2011: 5, emphasis in original).

For example, although judges in the Republic of Ireland have a wide sentencing discre­ tion and, like their counterparts in Scotland, have traditionally resisted the introduction of sentencing guidelines (Bacik, 2002: 351; Campbell, 2008: 291–92), the Court of Criminal Appeal recently began the process of moving away from this approach by issuing guide­ line judgments (Davey, 2014: 31). In DPP v Ryan, the Court set out ranges of sentence for offences involving the possession of firearms104 and identified the factors which were likely to influence the placing of a particular offence along the relevant scale.105 The Court was, however, careful to stress the importance of judicial discretion in applying such guidance: It clearly remains a matter for the sentencing judge to form a judgment, on all of the relevant facts, as to where on that range the offence for which the accused is to be sentenced lies. It is also clearly a matter for the sentencing judge to decide on the extent to which any aggravating or mitigating factors identified ought to increase or decrease the sentence to be imposed. Thus, any such range provides broad guidance but does not seek to impose any form of standardisation of penalty.106

Under O’Malley’s system of principled discretion, sentencers retain the discretion they already enjoy, but exercise it in accordance with settled principles. Departure from these 101  R v Whyte [2002] NSWCCA 343 at [87]; see also R v Ngui (2000) 1 VR 579 at [13]; Murray v The Queen [2013] NSWCCA 12 at [44] and Spigelman, 2008: 455. 102 See R v Henry [1999] NSWCCA 111 at [29] and Spigelman, ibid. 103  Bagaric and Edney, 2016: 49, citing Hardcastle v The Queen [2011] NSWCCA 87 at [37]; see also O’Malley, ibid 19–20. 104  DPP v Ryan [2014] IECCA 11 at 7.16. 105  Ibid at 7.15. 106  Ibid at 2.3 (emphasis added). See also the decision in DPP v Fitzgibbon, [2014] IECCA 12 in which the Court of Criminal Appeal set out broad guidelines for offences of assault causing serious harm whilst again stressing the importance of judicial discretion in their application (see paragraphs 8.10 to 8.14). For a full discussion, see O’Malley, 2016: 15–19.

Principled Discretion in Scottish Sentencing

 169

principles is permissible when a novel or exceptional aspect of a particular case so requires (O’Malley, 2001: 135; see also O’Malley, 2008a: 11). Individualised justice remains possible precisely because the principles operate at a higher level of generality than rules and are suf­ ficiently flexible to permit departure and variation when the particular circumstances of a case so demand (O’Malley, 2013b: 533). The principles ‘guide courts in navigating their way through the facts in order to arrive at an acceptable sentence’ (ibid). In this way a kind of ‘sentencing canon’ develops, consisting of leading appellate deci­ sions which are authoritative but not inflexible.107 Sentencers are expected to consider the appellate decisions and to follow them unless a different approach is justified by the facts of the case (O’Malley, 2011: 136). In a similar vein, Thomas stresses that sentencers must accept the obligation to account for their decisions, firstly by explaining the reasoning process by which they have been reached and, secondly, by relating the decision in the particular instance to a general body of doctrine and practice (Thomas, 1987: 22; see also Ashworth, 1984: 529 and Green, 1996: 120). In Scotland, this is routinely done although the volume of cases prosecuted on sum­ mary complaint in the sheriff courts means that practice does vary between solemn and summary procedure. In the summary courts, sheriffs will often give an account of the reasons for their sen­ tences. The volume of cases in the summary courts dictates that such accounts are often necessarily brief; however, attendance at a sentencing diet in any sheriff court demonstrates that reference is typically made to factors such as the offender’s record, the seriousness of the offence, and the effect on the victim(s). There is a statutory presumption against sen­ tences of three months or less: such a sentence must not be imposed unless the sentencer considers that no other method of dealing with the person is appropriate (section 204(3A) of the Criminal Procedure (Scotland) Act 1995). Importantly, where a sentencer imposes such a sentence, he or she must state their reasons for doing so (section 204(3B)). Simi­ larly, sentencers may not imprison offenders who have never previously served a sentence of imprisonment or detention unless they consider that ‘no other method of dealing with him is appropriate’ (section 204(2), the one exception being where a first offender is con­ victed of murder—section 204(5)). Where such an offender is imprisoned, the sentencer is under a statutory obligation to give his reasons for imposing the sentence (section 204(3)). In reporting on any subsequent appeal against a first custodial sentence, the Appeal Court has stressed the importance of the sentencer clearly stating that he or she was aware of these statutory provisions and of explaining why they thought that no other sentence was appropriate.108 With regard to more serious offences prosecuted on indictment in either the sheriff court or the High Court, it has become increasingly common for the sentencing judge or sheriff to issue a sentencing statement in which the reasons for sentence are provided, often in some detail. Sentencing statements are often issued in high profile cases, where the disposal may be complicated, or where the sentence could be seen as controversial. They are made

107  See Charleton and Scott, 2013: 21–22; O’Malley, 2006: 66–67 and 1994: 9–10; and, in Australia, Spigelman, 2008: 451; Street J in R v Oliver (1980) 7 A Crim R 174 at 177; and Gleeson CJ in Wong v The Queen [2001] HCA 64 at [7]–[12], all of whom advocate the same approach; see also Krasnostein, 2015a: 83 and 100–01. 108  Keogh v Watt 2000 SCCR 443.

170 

Structuring the Sentencer’s Discretion

available as press releases and are posted to the Judiciary of Scotland website.109 Whilst the availability of such sentencing statements should not detract from the need for the sen­ tencer to give cogent reasons in every case (see O’Malley, 2016: 781–86 and 2013b: 542–43), the Appeal Court commended their use in RG v HMA: The use of sentencing statements … is to the benefit of the administration of justice. It facilitates the parties to criminal proceedings, and others who may be present in court in the public benches, for whatever reason, hearing a clear explanation as to why the sentencing judge had chosen the particular sentence imposed. Equally importantly, it allows an accurate record of what was said in court at the point of sentence to be made available to members of the media. That assists the press and broadcasters in accurately reporting what has happened in court.110

Similarly, a system of principled discretion requires the appellate court to discharge the task of formulating and disseminating policies and principles, using the individual case as a vehicle for the analysis and exposition of general sentencing principles (Thomas, ibid; see also Roberts et al, 2011: 530). Thomas elaborates: There is no conflict between the duty of an appellate court to remedy injustice in the individual instance and the duty of the court to develop general principles and offer guidance to the trial judiciary: they are two aspects of the same function, and remedying injustice in the individual case provides the best illustration and reinforcement of the general principle (Thomas, ibid).

Principled discretion calls for informed judgment (O’Malley, 2011: 5). In the S­ cottish con­ text, this is arguably achieved through the phronetic synthesis tempered both by non-bind­ ing sentencing guidelines and by appellate sentencing guidance. That sentencing guidance can, and should, operate in tandem with the phronetic synthesis is demonstrated by Dhami’s 2010 study of judicial attitudes towards, and experiences of, the English sentencing guide­ lines system. Whilst the 89 Crown Court judges who participated in the study generally welcomed guidance on sentencing for substantive offences, they typically disagreed with proposals that would enable the new system of English guidelines to guide their actual deci­ sion making process (Dhami, 2013a: 301–02). Such proposals included specifying the way in which the various aggravating and mitigating factors should be weighed in sentencing, and how culpability, harm and seriousness ought to be judged. Dhami reports that whilst sentencers generally wanted more information and to have that information presented in an easy to use way, they did not want to be instructed on how to use that information (ibid 302; see also Dhami, 2013b: 180 and Padfield, 2013: 50). These findings are indicative of the continued importance of the phronetic synthesis in sentencing, even in jurisdictions where judicial ­discretion is circumscribed by presumptively binding guidelines. Krasnostein and Freiberg (2013: 269) note that the instinctive synthesis approach to sen­ tencing—and hence the phronetic synthesis—is theoretically incompatible with sentencing tariffs, ranges or starting points, other than where they might play a role in ‘informing’ the instinctive synthesis. As the authors explain: This is because, broadly speaking, the only ‘starting point’ should be all the particular circum­ stances of the case before the court. To begin elsewhere—for example, with the presumption of a

109 

See www.scotland-judiciary.org.uk/8/0/Sentencing-Statements (accessed 22 September 2016). RG v HMA, High Court of Justiciary on appeal, 27 October 2010, unreported at [10] (Lord Mackay of Drumadoon); see the case note by Brown, 2011a. 110 

Principled Discretion in Scottish Sentencing

 171

certain range—is to ignore the proper ‘instinctive synthesis’ in a way that ‘distorts the already dif­ ficult balancing exercise which the judge must perform’.111

Thus, it must be appreciated that ‘[t]he method of instinctive synthesis will by definition produce outcomes upon which reasonable minds will differ’.112 To put it another way, the judgment inherent in principled discretion comes at the price of tolerating some element of disparity in sentencing outcomes (O’Malley, 2011: 5). Such principled discretion is best achieved through a wise blending of sentencing guidance with the phronetic synthesis.

Ripping up ‘the Telephone Book’: Respondents’ Desire for Guidance—not Guidelines—in Sentencing In an observation with particular resonance for the sentencing judge, Solum notes that one of the main reasons in favour of a virtue-centred theory of judging and of doing particular­ ised justice is that the infinite variety and complexity of particular fact situations outruns our capacity to formulate general rules (Solum, 2003: 206; see also Solum, 2008: 147). What is to be done? Solum says this: The solution is not to attempt to write the ultimate code, with particular provisions to handle every possible factual variation. No matter how long and detailed, no matter how many exceptions, and exceptions to exceptions, the code could not be long enough. Rather, the solution is to entrust deci­ sion to virtuous judges who can craft a decision to fit the particular case … [N]o set of rules can do justice in every case (Solum, 2003: 206; see also Tadros, 2006: 181; Tamanaha, 2006: 229; and Nussbaum, 2001: 301 and 1990: 71).

Sentencing guidelines are based on a belief that one’s responsibility should be determined only by a particular criminal act and not by wider social circumstances (Aas, 2005: 24). How­ ever, since, in reality, there are no identical criminal acts that can be punished in a standard way (Aas, ibid), judicial discretion remains of paramount importance. It is precisely for this reason that the Appeal Court in Scotland has stressed the importance of judicial discre­ tion in sentencing even in the application of sentencing guidelines. The current English approach—where, as we have seen, presumptively binding guidelines are set out in tabu­ lar form—was criticised by respondents in the present study as too rigid and prescriptive. As one very experienced sheriff put it: I do cling very strongly to the idea that whilst broad guidance—and in some specific crimes, spe­ cific guidance in the form of guidelines—[is welcome] (and obedience above all to what we’re told from on high), I wouldn’t like us to get to the English situation where they more or less get a telephone book of instructions that tells you how to sentence and it becomes a matter of looking it up! I don’t think we need that in Scotland. There must always be room for compassion in the individual case. There must always be room to say that fairness and justice does not require this to be done or that fairness and justice does require this to be done (Sheriff 9).

111  Krasnostein and Freiberg, 2013: 285–86, citing Wong v The Queen (n 107) at [76]; see also Freiberg, 2016: 433; Freiberg and Krasnostein, 2011: 74–75; and Krasnostein, 2015a: 43 and 130. 112  Hudson v The Queen [2010] VCA 332 at [27], cited in Krasnostein and Freiberg, ibid 269; see also Russell v The Queen [2011] VSCA 147 at [57]–[58] and McPhee v The Queen [2014] VSCA 156 at [8].

172 

Structuring the Sentencer’s Discretion

The English approach to guidelines did not prove popular amongst respondents. Although two judges (Judges 1 and 6) stated that they found it useful to consult English guidelines for certain offences under UK legislation,113 Judge 1 cautioned against a blind acceptance of English guidelines such as the Definitive G ­ uideline on the Sexual Offences Act 2003 (Sentencing Guidelines Council, 2007a)114 because: [T]he policy which informed that guidance is not entirely consistent with the policy which informs the sentencing in our courts. The consequence is that the guidance seems to result in higher sen­ tences than I would have expected to be pronounced in our courts. So, [the English guidelines] are of interest, but I’m not sure it’s guidance that one could draw on without qualification (Judge 1).

Certain decisions of the Appeal Court evidence a judicial reluctance to apply the English guidelines. In HMA v McManus,115 for example, in which the Crown unsuccessfully appealed against the sentences imposed on the respondents on charges of assault to severe injury, permanent impairment and danger of life, the Appeal Court specifically stated that it did not find reference to the English guidelines on Assault and other Offences Against the Person (Sentencing Guidelines Council, 2008) to be of assistance as English law catalogues offences against the person in different ways to the common law of assault in Scotland.116 In Lees v HMA,117 (an appeal against the sentence imposed on a charge of opening lockfast places, the appellant having broken into a church to steal its organ), the Appeal Court made no mention of the English Sentencing Council’s Definitive Guideline on bur­ glary offences which had been issued nine months previously (Sentencing Council, 2012b). This was despite Mr Lees’ appeal having been remitted to a Bench of three judges and despite the Definitive Guideline containing provision for non-domestic burglary (ibid 11–14). This was a notable omission given the Appeal Court’s earlier reference to English guidelines issued either by the SGC or the Court of Appeal in cases such as Roulston (causing death by dangerous driving), Graham (child pornography offences) and Price (firearms offences).118 The Definitive Guideline on burglary offences was brought to the Appeal Court’s attention in Lees during submissions. The Court’s refusal to even consider them may have been due firstly to the new, prescriptive nature of English sentencing guidelines and, secondly, to the tension between the guidelines’ staged approach and the instinctive synthesis approach to sentencing approved in Gemmell v HMA119 and Ferguson v HMA120 (Brown, 2012b: 8).

113  A practice approved—with certain qualifications—by the Appeal Court in HMA v Roulston (n 82) at [17] and later reiterated in McKeever (n 82) at [16]–[18]; Geddes (n 82) at [18]; Milligan v HMA 2015 SCL 984 at [5]; Sutherland (n 99) at [20]; HMA v AB (n 78) at [13]; HMA v SSK (n 78) at [23]; and Scottish Power Generation Ltd v HMA (n 86) at [35]. 114  This guideline was in force in England and Wales at the time of the interview with the respondent. On 12 December 2013 the Sentencing Council published its new Sexual Offences Definitive Guideline (Sentencing Council, 2013) which replaced the existing sexual offences guideline on 1 April 2014. For criticism of the new guideline, calling for courts to take a flexible approach to ensure that disproportionate and unjust sentences are not imposed, see Harris, 2013. 115  2010 JC 84. 116  Ibid at [15]. See also S v HMA 2015 SCL 829 at [46]—for commentary, see Brown, 2015b—and Gill v Thomson (n 68) at [20]. 117  [2012] HCJAC 143. 118  Roulston (n 82), Graham (n 43) and Price (n 80). 119  Gemmell (n 37). 120  Ferguson (n 31).

Principled Discretion in Scottish Sentencing

 173

Similarly, in Hughes v HMA (a case involving mortgage fraud decided in July 2014), the Appeal Court refused to consider the English sentencing guidelines on fraud which had been published by the Sentencing Council some six weeks previously (Sentencing Council, 2014). This was despite the sifting judge having identified Mr Hughes’ appeal as being suit­ able for a sentencing guideline judgment on fraud and the appeal having been remitted to a Bench of three judges chaired by the Lord Justice Clerk (see Brown, 2014f: 6; see also Brown, 2015c: 5).121 The English form of prescriptive guidelines was deprecated by other respondents in the present study: You’ve got to consider not only the seriousness of the offence but the personal circumstances of the accused and the circumstances surrounding the commission of the offence. In England, they have this notion of a tariff for just about every offence. Some of my colleagues operate on a tariff system as well. I think it’s a dreadful approach. It’s a shocking attitude. You’re not just looking at the offence. You have to consider all of the circumstances of the offender as well as the offence (Judge 5). I think [the English approach] reduces sentencing to box ticking. It can never be a box ticking exercise because of the complexities of the individuals concerned and the different permutations that can arise (Judge 2). I just don’t think [the English approach] is a judicial process at all … In general, I think there has to be discretion. That’s justice—weighing the scales. It’s not just a table to follow (Sheriff 15).

The English guidelines on sexual offences came in for particular criticism from one very experienced judge (Judge 3); his Lordship considered them too ‘rigid’ and preferred the ‘much greater flexibility’ of the Scottish approach: Guidelines are only guidelines … but the English are desperate to force everything into boxes … To say: ‘A four year starting point for this and a four and a half year starting point for that’; it’s the way English law has always gone. They love to put everything into compartments … [and] treat these compartments as if they were watertight … My general feeling is that I’ll look very carefully at them but these people are liable to have got it wrong because they’ve come from a legal training which essentially encourages them to get things wrong. Very often—more often than not—I’ve found myself criticising and not following English authority that’s been handed up as if it’s something that we should just go along with (Judge 3).

This accords with the view of the High Court of Australia in Wong v The Queen, where numerical guideline judgments were criticised as being incompatible with the proper appli­ cation of sentencing principles and the exercise of judicial discretion: To focus on the result of the sentencing task, to the exclusion of the reasons which support the result, is to depart from fundamental principles of equal justice. Equal justice requires identity of outcome in cases that are relevantly identical. It requires different outcomes in cases that are differ­ ent in some relevant respect. Publishing a table of predicted or intended outcomes masks the task of identifying what are relevant differences.122

121  Cf also the decisions in Murray v HMA High Court of Justiciary on appeal, 17 February 2015, unreported (see the case note by Brown, 2015d) with the decision in Sutherland v HMA (n 99) (for discussion, see Brown, 2016b). 122  Wong v The Queen (n 107) at [65] (emphasis in original); see also Barbaro v The Queen [2014] HCA 2 at [34].

174 

Structuring the Sentencer’s Discretion

Aas criticises presumptively binding sentencing guidelines as the sentencer can no longer ‘look the accused straight in the eye’, having instead to use the lens of two-dimensional guidelines: Formalisation in sentencing has been achieved at the expense of richness of language and nar­ ratives and … human dimensions as well. Inherent in the guideline approach is a large degree of impersonality. The only information permitted for consideration [are] standards prescribed beforehand … The factors that are taken into consideration are objective factors that do not require an understanding of complex subjective rationalities (Aas, ibid 21).

Perry notes that whilst judges have traditionally been required to consider the life circum­ stances of ‘a living, breathing, all-too-human being’ in imposing a sentence that does justice to the crime, to the offender and to society, the introduction of presumptively binding guidelines and numerical sentencing grids have constrained judicial discretion to the extent of making a ‘mockery’ of the judge’s historic stature as an ‘oracle of justice’ (Perry, 2007: 422). As O’Donnell observes, when judicial discretion in sentencing is curtailed and ‘the individual fades from sight behind a thicket of rules’, there is a danger that the scope for doing justice will be unduly restricted (O’Donnell, 2012: 255). Aas also notes how such guidelines place a much greater emphasis on criminal acts than on the offender’s biographi­ cal and social narratives; they focus on ‘offence over life [and] category over narrative’ in concentrating on the act and not on the offender (Aas, ibid 110). Such approaches dehumanise the sentencing process and are deprecated by Laws LJ as ‘rule-book justice’: Rule-book justice is barbarous. It treats the criminal not as an individual, but merely as a member of a class, to be dealt with according to the rules set to govern the class. If the State systematically looks at its citizens, even the most flawed among them, in that grim light, it looks at them as things not people (Laws, 2004: 67).

In Scotland, by contrast, while the sentencer must have regard to guidelines and other guid­ ance provided by the Appeal Court to assist in the exercise of his or her discretion, it is the sentencer who retains the responsibility for determining the appropriate sentence in every case.123 Solum’s argument that no set of rules can do justice in every case is well illustrated by the way in which one judge in the present study dealt with a case involving child pornogra­ phy offences which he had sentenced immediately before the interview took place. Whilst the Appeal Court had recently issued sentencing guidelines for such offences,124 the judge explained that the present circumstances were different from the typical offences envisaged in the guidelines. In the case sentenced that morning the accused, in addition to possess­ ing indecent images of children, had sexually abused young persons whilst in a position of trust; he had also photographed several of the acts committed. The judge explained his decision to sentence outwith the guidelines thus: I took it away over the weekend because I wanted to fit it in with the guideline judgment and show why I was distinguishing it … I suppose there was an element of instinct. I knew that it was a seri­ ous matter and didn’t fit in too well with the guideline categories (Judge 8).

123  124 

Jakovlev (n 96) at [11]. Graham (n 43).

Principled Discretion in Scottish Sentencing

 175

The judge explained that whilst counsel was ‘trying to shoe [him] into particular catego­ ries’, he did not accede to such submissions. The combination of factors such as the contact abuse committed whilst in a position of trust, the photographing of the abuse and the subsequent possession of the resulting indecent images called, in the judge’s view, for a sentence beyond that suggested by the guidelines. Similarly, a number of sheriffs indicated that they found the Appeal Court guidelines useful as long as they remained guidelines from which they were free to depart—in essence, to do equity—should this be warranted by the circumstances of the particular case: I don’t have a difficulty—save that one’s got to master the wretched guidelines—but I don’t have a great difficulty with guidelines as long as they are just that. As long as I can say, ‘Well, that’s the guideline; I know why I’m not following it’ (Sheriff 4). So far as I’m concerned, I think the most influential advice that a judge can get is any guideline judgment from the Appeal Court. I’ve always been in favour of getting a guideline judgment which gives you a range of sentences (Sheriff 16). If I thought that to apply the guidelines would, in the particular circumstances of this offender, be really quite unfair and counter-productive perhaps, then I would probably just depart from the guidelines and say that’s why I’m doing so—because it would be unfair (Sheriff 9). I for one have no difficulty with general guidelines. If it gives one a general idea, I don’t think there’s any particular difficulty with that. But as long as they are merely guidelines and general, because otherwise one might end up with what one might think is an unjust result, because the Appeal Court can never foresee all of the circumstances which may arise (Sheriff 10).

These findings match those of previous studies of the Scottish judiciary. Millie et al (2007), for example, report that whilst most sentencers would welcome more guidance from the Appeal Court, they were on the whole against the use of presumptively binding guidelines (ibid 251; see also O’Malley, 2013a: 234). The approach of respondents to sentencing guide­ lines in the present study also accords with the following observations of Sir Igor Judge (as he then was) in an address delivered at Lincoln’s Inn in October 2007: You can have your guidelines. You can have the endless—is it now 17, or maybe more?—factors which the judge is under a statutory obligation to address. In the end you may be able to sentence a piece of paper by a process which you can colourably describe as scientific. But sentencing a fellow human being is indeed an art, a human skill, a skill in humanity, not a science, and it is this skill, and its application, that is embodied in the possibly pompous sounding phrase, ‘judicial discretion’ (Judge, 2007: 1).

The account presented thus far is not mere Scottish parochialism, an obstinate refusal to follow the English model simply because it emanates from south of the Tweed. The findings both of previous research and of the present study demonstrate that presumptively bind­ ing numerical sentencing guidelines applied through a staged approach—as exemplified in the nine stage ‘sentencing machine’125 devised by the Sentencing Council for England and Wales—are genuinely unsuited to this particular jurisdiction.

125  This term is taken from Tonry’s discussion of the US Federal Sentencing Guidelines (‘the forty-three-level “sentencing machine”’)—see Tonry, 1996: 98.

176 

Structuring the Sentencer’s Discretion

Hedgehogs and Foxes on the Bench The English Hedgehog The examination of the current use of formal, prescriptive and presumptively binding sen­ tencing guidelines in England and Wales presented thus far demonstrates that the guidelines issued by the Sentencing Council can be conceptualised as a form of the monist decision making considered in Chapter 5. We have seen, for example, that one of the Sentencing Council’s own members explains how its guidelines seek to promote ‘uniformity’ and con­ sistency in sentencing ­(Roberts, 2013a: 5, 22). We have also seen that the same Council member describes the nine step decision making procedure as a ‘detailed and structured methodology for courts to follow’ and as ‘an algorithm’ which should achieve consistency in sentencing (Roberts, 2011a: 1011; see also Roberts, 2012b: 283). This ‘algorithm’ is required because the previous system of appellate sentencing guidance apparently left ‘too much room for individual judges’ (Roberts, 2012a: 339).126 The ­Sentencing Council’s nine step decision making process is thus a paradigm e­ xample of a monistic decision-procedure; one that is devised, formulated and administered by experts and which, in principle, Berlin’s ‘slide-rule’ could perform (see Berlin, 2013b: 241 and Crowder, 2004: 115). The sentencing methodology introduced by the Sentencing Council is arguably one based on techne rather than phronesis. As a form of technical knowledge or know-how, the Coun­ cil’s methodology concentrates on similarities, not differences; it aims to be a standardised form of decision making, one that omits whatever is not relevant to answering the severely limited question that it sets itself to ask: how to achieve consistency in sentencing offend­ ers (see Berlin, 2013d: 40). It is a form of decision making characterised by Nussbaum as ‘pseudo-scientific’, one that seeks to capture rational choice in a system of general rules or principles which can then simply be applied to each new case (Nussbaum, 1990: 66). As a monist form of decision making, guidelines issued by the Sentencing Council require not that the aims of sentencing be compatible—we have seen that this cannot be the case127—but that conflicts among them have a single correct resolution indicated by a single correct formula; namely, the Council’s nine step approach.128 In directing that this approach is the only way in which sentences can be imposed, the Sentencing Council has effectively constructed a law-governed or systematic theory that fits into a unified scheme the multiplicity and variety of heterogeneous elements to which sentencers must have regard (see Gardiner, 1997: xvi). The Council brings forth its sentencing algorithm and uses it ‘simply to govern the indeterminacies of the new, impressing their order upon it’ (Nussbaum, 2001: 315). On Nussbaum’s account, the Sentencing Council’s nine step approach is an attempt to extract from that which the courts do and have done some ‘more elegant and simple procedure’ that can now be normative for what the courts do (Nussbaum, 1990: 73). 126  See the earlier discussion under the sections entitled ‘The Fourth “Phase”—Presumptively Binding Guide­ lines and the Sentencing Council’ and ‘The Continued Importance of Individualised Sentencing’. 127  See the discussion in Chapter 5 under the section entitled, ‘Value Pluralism in Sentencing’. 128  See the earlier discussion under the section entitled, ‘The Fourth “Phase”—Presumptively Binding Guide­ lines and the Sentencing Council’.

Hedgehogs and Foxes on the Bench

 177

The Sentencing Council’s formula in turn has recourse to the common currency of consistency—or perhaps, as Roberts (2013a: 5) states, uniformity—in sentencing. The attraction of such a monist decision making process is that it offers—indeed, imposes upon—judges in England and Wales a system from which they can then derive a single correct answer to any sentencing problem. To paraphrase Richardson, things work out very simply indeed for English and Welsh judges when there is at the bottom but one kind of intrinsic value—consistency—in terms of which all sentencing values are to be measured together, or commensurated. The sentencing decision then becomes a matter of the judge concerning himself with maximising the commensurating value or good of consistency (see Richardson, 1997: 15, cited in Crowder, 2002: 50). On Roberts’ account (2013a: 5, 22, 2012b: 283, 2011a: 1011), it appears that the Sentenc­ ing Council holds the monist belief that consistency in sentencing—the ‘mantra of sentenc­ ing reform’ (O’Malley, 2016: 7)—overrides all other sentencing purposes. This, as Crowder explains, is to hold that, ultimately, consistency is the only intrinsic value, the one thing in sentencing that is valuable for its own sake (Crowder, ibid 48). In English sentencing practice, the Sentencing Council’s guidelines are predicated on the monist belief that the single end or principle of consistency resolves value conflicts in every case regardless of cir­ cumstances (see Crowder, ibid 88). Whilst the Sentencing Council cannot ignore the will of Parliament (Treacy, 2016: 489) and must therefore allow that the purposes of sentencing set out in section 142 of the Criminal Justice Act 2003129 are genuine values, the Council nev­ ertheless appears committed to the monist belief that the statutory purposes of sentencing must be ‘instrumental or subordinate’ to the super value of consistency (see Crowder, ibid). In this way, ‘neatness’ and consistency are achieved at the price of individualised justice in a manner that is ultimately false and even corrupting (see Nussbaum, 1990: 60 and 73–74). Thus in Berlin’s terms, English and Welsh sentencers have ‘been brought by ­others to obey [an] ultimate principle before the bar of which all problems can be brought’ (Berlin, 2002a: 47). On applying Berlin’s account to contemporary ­English sentencing practice, we see that judges in this jurisdiction have been coerced by a ‘specialist body’ (O’Malley, ibid 14) of experts, in the name of consistency, to follow a decision making process which is inimical to the traditional English focus on judicial discretion as essential for the just application of legal rules and principles (Player, 2012: 248); a decision making process which, in the absence of the Sentencing Council, sentencers themselves would be unlikely to pursue (see Berlin, 2013b: 204). Through an application of Nussbaum’s Aristotelian account of decision making we see how, by following the Council’s guidelines, English sentencers ‘escape from the burden of the intractable and unexpected’: by accepting that there is only a single item (consistency) in terms of which all values are commensurable, the sentencer comes to each new case prepared to see only those items (consistency and uniformity) about which he or she already knows how to deliberate (see Nussbaum, 1990: 67). Berlin’s own discussion of how individuals can subjugate their decision making pow­ ers to a ready-made monist decision-procedure has particular resonance for contempo­ rary English sentencing practice, in particular, the English judiciary’s use of the Sentencing Council’s guidelines: Happy are those who live under a discipline which they accept without question, who freely obey the orders of leaders, spiritual or temporal, whose word is fully accepted as unbreakable law … I 129 

See the discussion in Chapter 5 under the section entitled, ‘Value Pluralism in Sentencing’.

178 

Structuring the Sentencer’s Discretion

can only say that those who rest on such comfortable beds of dogma are victims of forms of selfinduced myopia, blinkers that may make for contentment, but not for understanding of what it is to be human (Berlin, 2013c: 11).

The present system of English sentencing guidelines makes possible a seemingly perfected system of punishment in which there is universal agreement on a single method of imposing such punishment (see Crowder, 2004: 128). Yet conflicts among incommensurables, includ­ ing the aims of sentencing as set out in section 142 of the Criminal Justice Act 2003,130 can­ not be decided for good reason merely by the mechanical application of a standard monist rule. Such monist rules are unable to capture the relevant differences between individual cases and fail to do substantive justice to, or between, individual offenders (Duff, 2005: 1173–74). We have seen that the pluralist sentencer cannot rely on a ready-made monist procedure to resolve deep moral conflicts; rather he must go behind such perspectives to weigh the values they embody for himself (see Crowder, ibid 167). As in all courts, English and Welsh judges are on a daily basis faced with the question, ‘What sentence should I impose on this offender?’ This question has no clear answer if it is conceived, as Crowder (2002: 5) puts it, as ‘a demand for a general judgement, one that mediates decisively across a generality of cases’, between the purposes of retribution, general deterrence, specific deterrence, rehabilitation, incapacitation, and denunciation. Although each make their own distinctive claims, and neither outranks the other in all, or most, cases (Crowder, ibid), the Sentencing Council nevertheless seeks to commensurate all such sentencing purposes, or values, by subordination to the summum bonum of consist­ ency in sentencing, independent of context. On this view, English judges are—or, rather, have been forced to become—Berlinian ‘hedgehogs’ who ‘know one big thing’: the formula by which to achieve consistency in sentencing (see Berlin, 2013a: 436). The Sentencing Council’s monist approach implausibly supposes that the super value of consistency can override other sentencing aims or values in most, if not every, context. Yet as Crowder (ibid 70) asks, is it really possible to embrace the immense diversity of human values within a single end or formula? Crowder continues: It is hard to avoid the conclusion that in this ambition monists are trying to square the circle: to reduce the entire field of values to a single dimension while at the same time claiming that nothing has been lost or distorted. The past failures of monism are unlikely to be accidental. They are more likely to flow from the fact that no single commensurating medium or super-value or general rank­ ing can ‘do justice to all the different types of values there are’, or adequately ‘represent all of the considerations pertaining to some choice in terms of some single dimension’ (Crowder, ibid, citing Kekes, 1993: 58 and Richardson, 1997: 104).

The members of the Sentencing Council for England and Wales, however, place ‘their faith in a given pattern’ (Berlin, 2013d: 58) ahead of concrete human complexity. As Crowder observes, once we have forced our understanding of human experience into ‘conceptual boxes’—such as those involved at steps one and two of the nine step process devised by the Sentencing Council131—it becomes easier to ignore the real effects of our actions and ‘to force actual people into [metaphorical] wooden boxes’ (Crowder, 2004: 56). As Nussbaum

130 

See the discussion in Chapter 5 under the section entitled, ‘Value Pluralism in Sentencing’. the earlier discussion under the section entitled, ‘The Fourth “Phase”—Presumptively Binding Guidelines and the Sentencing Council’. 131 See

Hedgehogs and Foxes on the Bench

 179

cautions us, we should beware of losing sight of the human reality behind conflicts of val­ ues, and of allowing ‘numbers and dots’ to take the place of real people (Nussbaum, 1990: 101; see also 1993b: 258–59). Annan reminds us that people are not an undifferentiated mass to be organised as effi­ ciently as possible (Annan, 2013: xxx). Yet under the Sentencing Council’s nine step deci­ sion making process, offenders do not have their own context or their own background beyond the classificatory schema into which they happen to fall. In applying the Council’s guidelines, sentencers simply force offenders into their allotted slot (Hausheer, 2013: xlii; see also Berlin, 2013c: 16), into ‘moulds for which they are not fitted’ (Crowder, 1994: 299). For Berlin, such monist decision-procedures are but ‘snares’ that ‘achieve ­absurdities in their theories and humiliating failures in practice’ (Berlin, 2013a: 488–89, 497). To para­ phrase Annan’s (2013: xxxi) summary of Berlin’s attitude towards those experts who devise, formulate and administer monist decision making procedures, the technocratic Sentencing Council, driving through their prescriptive, presumptively binding sentencing guidelines, ‘sublime in their indifference to the ignorant opposition’ of critics such as Cooper (2013 and 2008), Harris and Gerry (2013) and Thomas (2013 and 2010)132 appal those judges and lawyers in other common law jurisdictions where the sentencing task is discharged not by recourse to a monist sentencing algorithm but by way of phronetic synthesis. For sentenc­ ers in jurisdictions including Scotland, the Republic of Ireland, Canada and Australia, the almost endless plurality and the ‘teeming individuality’ (Berlin, 2013a: 491) of offenders, offences, sentencing aims and purposes, and aggravating and mitigating factors precludes any prescriptive view of how sentencing should be done.

Foxes on the Bench—Value Pluralist Sentencers in other Common Law Jurisdictions Rather than sacrificing individualised justice to the ideal of consistency in sentencing, the appellate courts in Scotland, the Republic of Ireland, Canada and Australian jurisdictions adopt a Berlinian value pluralist approach to sentencing decision making. In issuing their own, more flexible, sentencing guidelines, the appellate courts in these jurisdictions have eschewed the shallow reductionist approach of the Sentencing Council for England and Wales. The Scottish, Irish, Canadian and Australian appellate courts view the Sentencing Council’s system of decision making based on algorithms and abstract rules as being too remote from their own concrete experience of sentencing in the criminal courts (see Berger, 2015: 21 and Nussbaum, 1993b: 243–44). Appellate courts in these jurisdictions have been wary of the abstract ideals and projections of the Sentencing Council and have sought to maintain a check on the real human and financial costs of sentencing (see generally Crow­ der, 2004: 173). This approach to sentencing views the plurality of sentencing values—retribution, gen­ eral and specific deterrence, rehabilitation, incapacitation, d ­ enunciation, and mercy—as a plurality of intrinsically different values or ends, purposes valuable for their own sake

132 See the earlier discussion under the section entitled, ‘The Fourth “Phase”—Presumptively Binding Guidelines and the Sentencing Council’.

180 

Structuring the Sentencer’s Discretion

(Crowder, 2002: 48). In imposing sentence, judges in these jurisdictions do not insist on trying to resolve the situation by rigid application of a general rule, as judges in England and Wales are obliged to do; an approach which loses the ethical value of contextuality and particularity (Nussbaum, 2001: 310). Rather, they ‘look through the smooth and regular outlines’ drawn by the Sentencing Council to ‘the uneven and often chaotic details of actual experience they conceal’ (Gardiner, 1997: xvii). These sentencers put more weight on the concrete choice situation and judge primarily with a view to the demands of that situa­ tion (Nussbaum, 1990: 93). In so doing, the courts attempt to reach a balance between, on the one hand, the general rules of appellate sentencing guidance and non-binding appel­ late sentencing guidelines and, on the other, the particular judgment required to tailor the sentence to the individual facts and circumstances of the particular case (see Crowder, ibid 191). Above all, the value pluralist sentencer is, as Nussbaum explains, devoted to finding precisely the right way of ‘rendering the concrete, putting all the variety, messiness, and indefiniteness of the “matter of the practical” into words that will not debase its value, or simplify its mystery’ (Nussbaum, ibid 104). This examination of contemporary sentencing practice in England and Wales and other common law jurisdictions illustrates the profound distinction between two very differ­ ent approaches towards sentencing and the imposition of punishment. The moral monist approach is exemplified by the English ‘engine room’ or ‘factory floor’ (see Hough et al, 2003: 39). It is here that the nine step sentencing machine is run, and the Moloch of sen­ tencing consistency worshipped, by Berlinian hedgehogs: judges who ‘know [the] one big thing’ (Berlin, 2013a: 436) of seeking consistency and uniformity in sentencing, yet who lose sight of the individual, contextualised offender. The value pluralist approach, however, is exemplified by the sentencing methodology of the courts in Scotland and in other com­ mon law jurisdictions. Here, sentences are imposed by Berlinian foxes—judges who ‘know many things’ (Berlin, ibid) about the offence, the offender, the victim and indeed societal mores in general. It is here that the Berlinian fox imposes sentence by way of phronetic syn­ thesis: an adjudicative balancing of the various aims and purposes of sentencing all medi­ ated by professional and judicial experience.

Judicial Conceptions of the Sentencing Process The Findings of Previous Studies England and Wales The 1981 Crown Court pilot study133 found that only a minority of the judges interviewed regarded sentencing as a matter of principles and reasoned conclusions. The authors of the study report that most judges described sentencing as an intuitive process, using terms such as ‘instinct’, ‘experience’, ‘hunch’ and ‘feeling’ (Ashworth et al, 1984: 50). Similar findings were reported almost 20 years later by Hough et al (2003). When asked about how they

133 

For details of this study, and the others discussed in this section, see Chapter 1.

Judicial Conceptions of the Sentencing Process

 181

make sentencing decisions, the professional judiciary did not consider the process to be highly structured; they were more likely to say that the process was intuitive, at least to some degree (ibid 37; Millie et al, 2007: 249). In the subsequent study by Jacobson and Hough, the intuitive dimension of sentencing was again stressed by a number of respondents, one of whom remarked that sentencing was about ‘a personal and sometimes emotional response to a particular set of circumstances’ (Jacobson and Hough, 2007: 48). Having noted that subjective factors (such as an offend­ er’s remorse) can carry significant weight as mitigation, Jacobson and Hough consider it inevitable that sentencing becomes partly a matter of what their respondents referred to as ‘experience and feeling’ or ‘gut feeling rather than careful calculation’ (ibid). Certain of the judges in Jacobson and Hough’s study favoured a more structured approach to sentencing. This could involve, for example, making notes on any relevant guideline judgments before the sentencing diet; dealing in a structured way with the wide range of aggravating and mitigating factors in the particular case (perhaps asking counsel to address each in turn—on occasion, a ‘very painful process’); and adjourning (if necessary) to review all applicable legal arguments. This was particularly, but not only, the case with less experienced judges (ibid 49). Jacobson and Hough cite the experience of a recorder, newly appointed from the civil Bar, who had developed her own ‘template’ to guide herself through each sentencing decision; however, more experienced judges (especially those who had practised at the criminal Bar) had told her that when it comes to sentencing ‘you just do it’ (ibid).

Scotland In Tombs’ study of the Scottish judiciary, the 40 respondents were asked whether they saw the decision making process as primarily structured or as more ­intuitive and based on experience (Tombs, 2004: 42). Tombs reports that ­sentencers’ descriptions ranged along a continuum of emphasis: they most commonly described what they viewed as a structured decision making process, in terms of stages in the sentencing task, although the 17 sentenc­ ers who conceptualised sentencing in these terms did note that intuition and experience played a part (ibid; see also Millie et al, 2007: 248). The next largest group of sentencers in Tombs’ study (12 respondents) considered that the sentencing process was not particularly structured. They contended that experience and intuition—the ‘feel’ of a case—were more influential in sentencing decision making (Tombs, ibid). These respondents opined that sentencing is ‘more of an art than a science; more intuitive than structured’ and that, as a sentencer, ‘you develop a feel for how things are’ (ibid 44). Almost as many sentencers (11 respondents) said that their decision making process was intuitive and based on experience (ibid 42). One judge, for example, stated that: I’ve been sentencing for [many years]. The decision comes from within. You see the case; hear about the facts from the Crown. In the High Court it’s almost always custody. How long? A number springs to mind. You hear the defence and adjust at the margins (ibid 44).

Similar sentiments were expressed by a sheriff: The feel of the courtroom is a big part in sentencing. The longer you do it the more intuitive it is. I’m against structured sentencing. You’ve got to trust your judge (ibid 45).

182 

Structuring the Sentencer’s Discretion

Thus, Tombs’ findings on the nature of the decision making process reflect the earlier work undertaken in England and Wales by Hough et al (ibid). In both jurisdictions, the majority of respondents in the respective studies (all members of the professional judiciary: judges and sheriffs in Scotland; district judges, recorders and Crown Court judges in England and Wales) acknowledged the importance of experience and intuition in sentencing. Whilst the English magistrates surveyed by Hough et al tended to say that they followed a structured decision making process, neither Tombs’ study nor the present study explored the views of lay justices of the peace. It may be, however, that the lay judiciary in Scotland also follows a structured process in sentencing. The decision in Purvis v Macdonald,134 for example, concerned the use of sentencing guidelines in the then District Court comprising levels of fines for commonly encountered offences. The guidelines had been drawn up by a committee of justices. The guidelines had not been published and the appeals against sen­ tence were remitted to a court of three judges to consider the implications of the practice of justices using guidelines devised for that Court. Although the Appeal Court reserved its opinion as to whether the guidelines were unlawful, it disapproved the use of such guide­ lines. The Court noted that as the guidelines had not been ­published, their use was not in accordance with natural j­ustice.135 More importantly, however, the Court criticised the guidelines as being vague, unduly restrictive and confusing.136 Whilst it was clear that the Appeal Court did not consider it appropriate that a committee of lay justices of the peace should promulgate sentencing guidelines, the fact that the four divisions of the then Scot­ tish Borders District Court convened to draft such guidelines perhaps indicates a preference for a structured approach to sentencing amongst the lay j­udiciary in Scotland.

Australia In the course of her study on the use of suspended sentences by the Tasmanian judiciary, Bartels asked respondents how they approached the task of imposing this particular dis­ posal (Bartels, 2008: 99). Bartels reports that although this did not necessarily elicit a clear explanation of what she acknowledges is ‘a largely unexamined and intuitive process’, reli­ ance upon the instinctive synthesis was the most common approach followed by respond­ ents (ibid 99–100). In a discussion of the desirability of giving one’s reasons for suspending a sentence, for example, one of Bartels’ respondents (a Judge of the Supreme Court) said this: Generally speaking, I’m not a great one for elaborate reasons for passing sentence because it is after all an intuitive judgment … I think that if we go into more elaborate reasons, pages of it, it just ends up with more appellate work and more scrutiny and upset when it is an intuitive process anyway (ibid 120, emphasis added).

Judicial reliance on an instinctive approach to sentencing was also evidenced by the follow­ ing comments from, respectively, a Judge of the Supreme Court and a magistrate: I never engage in [the thought process] until I consider an actual case. The case will dictate it for me (ibid 100). Sometimes you go through an intellectual process, but mostly I think it just hits you. You hear the facts, the circumstances of the offence, and the offender, and it just hits you (ibid). 134  135  136 

2006 JC 155. Ibid at [7]. Ibid at [11].

Judicial Conceptions of the Sentencing Process

 183

In her study of the Queensland judiciary, Mackenzie found that although six of the 31 judges interviewed saw aspects of sentencing as either an intuitive or instinctive process, none described the act of sentencing as an instinctive or intuitive synthesis (Mackenzie, 2001: 142). This was due perhaps to the fact that the term was not traditionally used by the Queensland Court of Appeal (ibid 140; 2006: 2). Judicial comments reported by Mackenzie in which respondents indicate their view of sentencing as instinctive or intuitive include: I always listen to what is being submitted to me, and I work to some degree on a feel for things. I like to think that it is based on experience. I suspect I am not often wrong in my perception of something in this area. Most experienced judges will do that. Sentencing is an educated response to a set of facts. Sentencing is an intuitive process. Sentencing really becomes a gut reaction, dependent on the individual case (Mackenzie, 2001: 143).

Although the term ‘instinctive synthesis’ was not used by judges in Mackenzie’s study, intuition, gut reaction and perception were all cited as useful approaches to sentencing by respondents (ibid).

Ireland In her study of consistency in Irish sentencing, Maguire reports that the majority137 of the 15 District Court judges interviewed described the sentencing process in terms of the indi­ viduality of each case, stressing that it was not only the crime that was being punished but the criminal also (Maguire, 2010: 33). Respondents were unable to explain how they actu­ ally weighed the various factors to determine the final sentence other than in very general terms. As one of Maguire’s respondents explains in terms redolent of equity and phronesis: It’s the totality, the total overview of the situation, to try and distil out. It’s very hard … And it’s not fly by your pants either. You have to trust your instincts as well and as I say the overriding thing is to be fair (ibid 34).

Thus, although based on a more limited sample than any of the other comparable studies, the Irish judiciary’s descriptions of sentencing also appear consistent with the instinctive synthesis approach (ibid).

Phronetic Synthesisers and Intuitive Over-Riders: The Findings of the Present Study Given the findings of Tombs’ study (2004) and the way in which, as Hammond J of the New Zealand Court of Appeal has put it in a less than flattering analogy, judges cling ‘limpetlike’ to what they regard as ‘their’ sentencing discretion (Hammond, 2007: 223; see also Hutton, 2013b: 119, 2003: 319, 323; and Hutton and Tata, 2010: 273), it was envisaged that most judges when questioned about their sentencing methodology would conceptualise the process in terms of instinctive synthesis by emphasising the importance of experience and intuition. As with Tombs’ study, however, responses were found to range along a continuum of emphasis. 137 

The exact number is not provided by the author.

184 

Structuring the Sentencer’s Discretion

Structured Sentencing Only two respondents (Judge 7 and Sheriff 12) regarded sentencing as a purely structured process. Both Judge 7 and Sheriff 12 were experienced sentencers. Unlike the other seven judges in the study, Judge 7 appeared to operate a tariff system for most offences—an approach which, whilst prevalent in England and Wales, is the very antithesis of the indi­ vidualised approach to sentencing traditionally adopted by the Appeal Court in Scotland: I don’t think sentencing is instinctive. It is essentially a structured exercise in the sense that for particular crimes in particular circumstances there are what are generally regarded to be the appro­ priate sentences … I don’t think you could class it as instinctive. If, for example, you decide that someone’s to get five years for a robbery, you don’t form that view on the basis of some innate instinct as to what is appropriate; you do it on the basis that that is what society generally regards as appropriate for that particular crime. I think the judge has to learn what are the appropriate sen­ tences for the appropriate circumstances. So it’s structured. It’s learned. It’s almost a science in that sense (Judge 7).

Judge 7 was the only respondent to describe sentencing in terms of a science and, along with Sheriff 12, was one of only two respondents critical of a wide discretion in sentencing: I’m not an enthusiast for the use of discretion in the sentencing context … I’m not a great one for believing that there’s a range of penalties between one year and three years, within which the sentencer has a discretion to select what he or she wants. Or, in the context of a fine, that there’s a range of ‘reasonableness’ like what we’d get in civil cases … I know a lot of my colleagues don’t approach it in that way—they consider sentencing to be much more of a discretionary thing. I’m not an enthusiast for that (Judge 7).

Similarly, Sheriff 12 advocated a far more structured approach to sentencing similar to the English regime: I do feel that some approach in the future whereby certain elements are isolated in particular cases and then you have a three or a five box matrix to assist the sentencing judge in determining which level the particular case fits would, I think, give greater consistency, transparency and would avoid the uncertainty, judicial shopping and the lottery that goes on at the moment where it’s largely unregulated. I think often we all apply a seat of the pants approach to sentencing and some try to call it art—an art form (Sheriff 12).

The Instinctive Approach Eight respondents (two judges and six sheriffs) adopted an exclusively instinctive approach to sentencing. The two judges who followed this approach were amongst the most experi­ enced; between them they had sat in the High Court for over 25 years. Judge 8 remarked that: I think there’s a lot of instinct in my case. I get a feeling as to what I think is the right sentence … It’s [down to] experience … Instinctively you feel that a sentence is right (Judge 8).

The second judge to follow a purely instinctive approach justified the need for a wide judi­ cial discretion in sentencing in language consistent with the ideal of phronesis: You can ask why judges should have the sole discretion to decide the appropriate ­sentence. Why can’t it be argued that the job would be done better by sociologists, by the social work department, or by former police officers? It’s because of the vast reservoir of experience that judges and sheriffs

Judicial Conceptions of the Sentencing Process

 185

bring to the task of sentencing … We bring this experience to bear when sentencing and that is why judicial discretion is so important (Judge 5).138

This respondent proved particularly receptive to the notion of sentencing as an instinctive synthesis. After I had explained the concept of instinctive synthesis by reference to the deci­ sions of the Victorian Court of Criminal Appeal in R v Williscroft139 and the High Court of Australia in Markarian v The Queen,140 the respondent said this: An instinctive synthesis? Yes, I like that idea. You don’t go through all these stages when you sen­ tence. You just … well, you just do it (Judge 5).

Similarly, one sheriff explained how he favoured the instinctive synthesis approach by making reference (without any prompting or prior discussion from me) to the concept as applied in Australian sentencing jurisprudence: There was, for a while, a quote I carried round from a decision in a minority judgment in a case … and the dissenting judgment I think made reference to a quote from an Australian court about sentencing as [an instinctive] distillation of various things. And I thought that that put it very well (Sheriff 1).141

As with the judges, it was found that the more experienced sheriffs described their sen­ tencing as instinctive. This was to be expected for, as Lloyd-Bostock observes, the decision becomes more automatic as one’s skill at sentencing develops. New cases are more likely to be treated as falling into types and categories that the sentencer has seen before (LloydBostock, 1988: 80). The six sheriffs who followed an instinctive approach to sentencing had all sat in the sheriff court for between nine and 17 years. One sheriff (who had sat for over 10 years) explained his approach as follows: I have to say that I see it as an instinctive process. In the vast majority of cases that I have to deal with, I can usually tell without too much difficulty what the general type of sentence ought to be … There will be cases where it is almost immediately obvious on a narration of the circumstances that that is a case where someone is going to have to go to prison. There will be other cases where it becomes apparent again from as soon as you hear the facts that prison is not an appropriate dis­ posal … I think generally I would find it a fairly straightforward exercise whether someone ought to go to prison or not (Sheriff 17).

The six sheriffs considered the instinctive approach to be particularly appropriate when sentencing in the summary court: An instinctive approach particularly makes sense in the summary context of sentencing, where by definition you can take bigger risks because, by definition, you’re dealing with less serious criminals … At summary level, I think you can afford to be more instinctive (Sheriff 3).

138  See also the remarks of respondent ‘L’ (a practising barrister) reported by Fitz-Gibbon in her study into judicial and practitioner attitudes to the mandatory life sentence for murder: ‘The sort of judges that are dealing with a murder case are so experienced. He has a lifetime of experience and he knows the right tariff, he knows what the offence deserves and I think to try and straightjacket [sic—read “straitjacket”] the judge is wrong because you cannot cater for every situation … I just think you should have discretion. You’ve got to trust your judge and I don’t think they get it wrong’ (Fitz-Gibbon, 2013: 520; see also Fitz-Gibbon, 2016: 60). 139  [1975] VR 292. 140  [2005] HCA 25. 141  The respondent is here referring to the dissenting opinion of Lord Philip in RB v HMA 2004 SCCR 443 at [27] where his Lordship approved the concept of instinctive synthesis by reference to the decision of the New

186 

Structuring the Sentencer’s Discretion

Two of the most experienced sheriffs took the view that, due to the volume of business, one can really only sentence by way of instinctive synthesis in the s­ heriff court. Sheriff 5, who had spent most of his professional life as a prosecutor, considered that anyone who came to the Bench and who did not sentence instinctively would simply ‘make a rod for their own back’: Sentencing is instinctive … If you sit and try and work it out—‘well, he gets three months because it’s this, but then you’ve got to do this and then do that’—you’ll be there forever, particularly in summary sentencing. You can’t do it like that (Sheriff 5).

Sheriff 16 noted that: If you’re in a remand court where you’re having to deal with maybe 30 to 40 cases, you’re not going to be able to give the same amount of time to those cases as you do in, say, a High Court case where you’re maybe only sentencing one case that morning and have a lot more time to think about it (Sheriff 16).

One sheriff considered that his sentencing by way of instinctive synthesis was hampered, to some extent, by the volume of business in the sheriff courts: I think that an instinctive approach is important, provided you know you’ve mastered the facts. I’m not sure everyone does that. Indeed, it is very difficult to do that given the weight of work some of us have (Sheriff 4).

In referring to the demanding nature of the sentencing task and the high volume of cases with which many sheriffs have to deal, Sheriff 5 contrasted the positions in the High Court and sheriff courts: Sentencing is a more complex, time consuming and stressful part of the sheriff ’s job. Every Thurs­ day we have a solemn court [where] you can be expected to sentence up to 17 indictments. I’d be fascinated to know how many indictments their Lordships [in the High Court] ever sentence in one day (Sheriff 5).

Sheriff 3 was even more vociferous: There’s probably not a sheriff in the country who wouldn’t say to you that you shouldn’t be allowed to sit in the Appeal Court until you’ve sat in a remand court and watched the sentencing process in action in summary crime or in solemn crime. We routinely have 14 or 15 cases in our solemn sentencing court … And a High Court judge has, what? One, maybe two? They don’t understand us. We are not understood! (Sheriff 3).

Similar sentiments were expressed by one of the district judges interviewed by Hough et al who commented that he would like to see the senior judiciary visit a magistrates’ court: … which is the engine room? The Crown Court isn’t the engine room, the Court of Appeal isn’t the engine room, the House of Lords isn’t the engine room—this is the factory floor (Hough et al, 2003: 39).142

South Wales Court of Appeal in R v Thomson [2000] NSWCCA 309. This aspect of Lord Philip’s dissenting opinion was subsequently approved by a majority of the Court in Gemmell v HMA (n 37) at [80]. 142  Similarly, in his study of the Canadian judiciary, Hogarth discussed the workload of Canadian magistrates and noted that most routine sentencing decisions have to be made quickly with little time for reflection (Hogarth, 1971: 217–18). See also Cooke, 1987: 60 and Pennington and Lloyd-Bostock, 1987: 3.

Judicial Conceptions of the Sentencing Process

 187

As with the judges, the sheriffs who favoured an instinctive approach to sentencing all emphasised the importance of judicial experience. This group of respondents typically made comments such as: The more experienced you become the more aware you are of the issues that are very relevant in sentencing. So, I suppose an observer might think it’s almost intuitive, whereas actually it’s based on years of experience (Sheriff 16).

This accords with Flyvbjerg’s emphasis on the necessity of an in-depth understanding of context and the ability to make political judgments within those contexts (see Simmons, 2012: 246). Consonant with the ideal of phronesis, the last named respondent continued: I think your experience allows you to structure the decision making process possibly a little more quickly … It’s like when you’ve done something for years and years and years and it’s only when you actually sit down and actually put pen to paper as to how you should do it, you realise that’s probably how you’ve been doing it for years but you haven’t gone through that mental process each time you’ve undertaken the exercise … The more experienced you get, the easier I think it is to do (Sheriff 16).

Intuitive Over-Riding: The Mixed Approach Five judges and 10 sheriffs in the present study regarded sentencing to be a mixture of intui­ tive and structured decision making. One of the judges (Judge 2) considered that, whilst judicial instinct in sentencing is still important, the process has become more structured than it used to be. As offenders are now increasingly appearing in the High Court either as first offenders or with a limited criminal history,143 the judge noted that a lot more back­ ground information is required in the form of social enquiry and risk assessment reports. The availability of different sentencing options such as extended sentences created what Judge 2 described as ‘the opportunity for a much more structured approach to sentencing’: You have a wider variety of options to consider … I think that means that you do now approach it on a more structured basis, looking at what could be served or achieved by any of these differ­ ent options, and trying to fit it to the precise circumstances and facts of the case and the offender (Judge 2).

One of the 10 sheriffs who considered his sentencing methodology to be a mixture of the intuitive and structured approaches agreed that sentencing had become more structured in recent years: Sentencing has become very much more structured. The reason for that, of course, has been this drive for consistency on the one hand and also a public failure to appreciate that people are differ­ ent (Sheriff 9).

Whilst the stages of ordering social enquiry and other reports and the consideration of sentencing guidelines were seen to impose a form of structure on sentencing, it is the latter part of the sentencing task—tailoring the disposal to the precise facts and circumstances— in which judicial experience is crucial. Judge 4, who also considered sentencing to be a

143 

See the discussion in Chapter 4 in the context of statutory minimum sentences.

188 

Structuring the Sentencer’s Discretion

mixture of intuitive and structured decision making, explained the process he followed in sentencing: I look at previous cases and any relevant sentencing guidelines but it still has to be a judgment about trying to reconcile a variety of competing aims. You must objectively rationalise and justify the sentence by looking at the circumstances of the offence, the effect on the victim and by using the right sentencing tools … You have to exercise your judgment and your discretion … There is no grid system, no matrix, that will work out the sentence for you. You just have to act in good faith to reach a just decision, a proper decision, the right decision (Judge 4).

What respondents described as a mixed approach to sentencing most closely resembles the form of judicial decision making described by Guthrie et al (2007) as the ‘intuitiveoverride’ approach. This approach comprises a blend of the formalist model of judging, in which judges are seen as ‘highly skilled mechanics’ (Neuborne, 1992: 421) who apply the governing law to the facts of a case in a logical, mechanical and deliberative way,144 and the realist model, in which judges follow an intuitive process to reach conclusions which they only later rationalise with deliberative reasoning145 (Guthrie et al, ibid 2–3). The intuitiveoverride model posits that judges generally make intuitive decisions but sometimes over­ ride their intuition with deliberation (ibid 3, 8–9). Guthrie et al describe the model as both ‘realist’ in the sense that it recognises the important role of judicial intuition and ‘formalist’ in the sense that it recognises the importance of deliberation in constraining the inevitable influence of such intuition (ibid 3). They explain that the intuitive-override model: … views judges neither as the purely deductive decision makers envisioned by the formalists nor as the intuitive rationalizers envisioned by the early realists. Rather, it views judges as ordinary people who tend to make intuitive … decisions, but who can override their intuitive reactions with com­ plex, deliberative thought (ibid 9).

In their study of US trial judges, Guthrie et al administered a five item questionnaire to 295 circuit court judges (ibid 13 et seq). The questionnaire included Frederick’s ‘Cognitive Reflection Test’, or ‘CRT’; a three item test designed to distinguish intuitive from delib­ erative processing (Frederick, 2005: 26–28). The CRT measures ‘cognitive reflection’, which Frederick describes as ‘the ability or disposition to resist reporting the response that first comes to mind’ (ibid 35). Guthrie et al found that judges are predominantly intuitive decision makers but some­ times override that intuition with deductive reasoning (ibid 13, 17–19, 27–29). They consider that eliminating all intuition from judicial decision making is both impossible and undesirable because, firstly, it is an essential part of how the human brain functions

144  As Posner explains, formalist approaches to law—which he strongly rejects—are ‘premised on a belief that all legal issues can be resolved by logic, text, or precedent, without a judge’s personality, values, ideological leanings, background and culture, or real-world experience playing any role’ (Posner, 2013: 1). For Posner, legal formalism ‘retards the profession’s ability to solve … novel legal problems’ (ibid 8); it ‘constantly [veers] into absurdity’ (ibid 116) and ‘trades in obscurantism’ (ibid 357). 145  Posner elaborates: ‘The core of a defensible legal realism is the idea that in many cases, and those the most important, the judge will have to settle for a reasonable, a sensible, result, rather than being able to come up with a result that is demonstrably, irrefutably, ‘logically’ correct … What is reasonable or sensible will often depend on moral feelings, common sense, sympathies, and other ingredients of thought and feeling that can’t readily be translated into a weighing of measurable consequences’ (Posner, 2013: 6). For Posner, legal decisions are—or ought to be—fact-driven, not theory-driven (ibid 7).

Judicial Conceptions of the Sentencing Process

 189

(Hogarth, 2001: 66 cited in Guthrie et al, ibid 5) and, secondly, intuitive judgments are often quite accurate (ibid 29—see also Tversky and Kahneman, 1974: 1124, cited in Guthrie et al, ibid 29–30). As Gigerenzer and Todd demonstrate, intuitive thinking can, with ‘a minimum of time, knowledge, and computation’ enable decision makers ‘to make adaptive choices in real environments’ in some circumstances (Gigerenzer and Todd, 1999: 14, cited in Guthrie et al, ibid 30). Experts can, and do, use intuitive thinking successfully. In a study of chess grandmasters, for example, Ross notes that such players who have attained this level of expertise rou­ tinely use intuitive rather than deliberative strategies to great effect (Ross, 2006). In terms very similar to the importance attached by Judge 5 to ‘the vast reservoir of experience that judges and sheriffs bring to the task of sentencing’, mentioned above, Ross notes that a chess grandmaster relies not so much on an intrinsically stronger power of analysis as on ‘a store of structured knowledge’ that enables him to see the correct move immediately, without consciously analysing anything at all (Ross, ibid 67, cited in Guthrie et al, ibid 30). It is this conversion of deliberative judgment into intuitive judgment that can be regarded as the hallmark of expertise (Guthrie et al, ibid) and explains why it was the most experi­ enced judges and sheriffs in the present study who described their sentencing methodology as instinctive or intuitive.146 What of the respondents in the present study who professed to follow a mixed approach to sentencing? The following hypothetical example illustrates the operation of Guthrie et al’s intuitive-override model in sentencing. An accused (a first offender) pleads guilty in the sheriff court to six historic charges of lewd, indecent and libidinous practices and behaviour. The offences were committed some 30 years ago against three young girls (daughters of a family friend) over a period of several years. The abuse began when the offender himself was in his mid-teens and finished when he was in his early 20s. In such a case the sheriff, who both represented and prosecuted several such offenders during his career at the Bar and who has since dealt with a num­ ber of similar offences since his appointment to the Bench, notes the striking similarity between this case and others of a similar type: the abuse followed a typical pattern of the offender sexually abusing the eldest sibling until the onset of puberty when he moved on to abuse each of her younger sisters.147 The sheriff sees the resemblance to previous cases without having to expend mental effort. He may decide at this stage that a custodial sen­ tence of several years is warranted and may even decide on an initial figure—say, four years’ imprisonment. This is the sheriff ’s tacit system of decision making in action. The sheriff must then, how­ ever, ‘go deeper’ before proceeding to sentence (Christie, 2005; see also Berlin, 2013a: 496; and Posner, 2013: 131–32), moving from surface to depth in examining the offender’s biog­ raphy and circumstances (see Berlin, ibid 498; and Garland, 1999b: 524) in order to achieve a deep and internal understanding of the particular, concrete case (see Nussbaum, 2001: 300–01, 1993a: 99, 1993b: 257). The sentencing exercise is not conducted in a vacuum, but follows upon the receipt of a diverse range of information from a wide variety of sources (Hall, 2007: 249; Hogarth, 1971: 279; Hutton, 1999: 168–69; Laws, 2004: 65; O’Malley,

146  Cf the only two respondents who described their methodology as purely structured, both of whom also had many years’ experience on the Bench. 147  The Appeal Court recently referred to such patterns of sexual offending in R v HMA 2013 JC 212 at [17].

190 

Structuring the Sentencer’s Discretion

2016: 139–41; Spohn, 2009: 122–24; see also Royer v The State of Western Australia).148 An in-depth understanding of the facts and circumstances is required and—pace presump­ tively binding sentencing guidelines—working on the surface (in the sense of being able to get things done without needing to understand them) is not appropriate in the sentencing of the offender as an individual (Aas, 2005: 5). Thus, the sheriff may call for a criminal justice social work report (and will be required to order such a report if the offender has not previously served a term of imprisonment); he may order other reports such as a psychiatric report or a risk assessment; he will consider any victim statements and will hear the plea in mitigation, all of which assist the sheriff in understanding the person with whom he is dealing: Slowly, the judge builds a story of what has happened. To the judge, and hopefully also to others, this is a comprehensible story; in the best case, a creation of a piece of art. Through that story, the judge intuitively comes to a conclusion—the final sentence (Christie, ibid)

As the present example involves historic offences, some of which were committed whilst the offender himself was a juvenile, the sheriff will also be required to have regard to the sen­ tencing principles expressed in Greig v HMA:149 account must be taken both of the offend­ er’s age, and hence relative immaturity, at the time of the offences and of the time which has elapsed since the date of the offending, as it indicates that the risk of his re-offending is low (see also Brown, 2012a: 3–4 and von Hirsch and Ashworth, 2005: 178). This second process—the process of going deeper by finding out more about the offender, the circumstances of the offences and the impact on the victims, and also by applying sen­ tencing principles set out by the Appeal Court—is deliberative. It involves recalling details of codified legal knowledge and then applying that knowledge. It involves attention and mental effort. It is, as Hogarth notes, the deliberative system at work (Hogarth, 2001: 22). Such use of the intuitive-override model may (through an application of the principle of equity in sentencing) ultimately result in a lesser custodial sentence than one would nor­ mally expect (JW v HMA;150 Sharp v HMA;151 HMA v Brand and HMA v Gordon152); it may result in a non-custodial disposal for offences that would normally attract a prison sen­ tence;153 or—exceptionally—it may result in no punishment being imposed at all despite the seriousness of the offence.154 Of the 15 respondents in the present study who considered their sentencing methodol­ ogy to be a mixture of an intuitive and structured approach, two thirds (three judges and seven sheriffs) articulated their use of the mixed approach in terms that accorded precisely

148 

[2009] WASCA 139 at [28]. 2013 JC 115 at [10]–[11]. 150  High Court of Justiciary on appeal, 1 October 2014, unreported. See the case note by Brown, 2015e. 151  High Court of Justiciary on appeal, 21 February 2013, unreported. See the case note by Brown, 2013b. 152  HMA v Brand 1998 SCCR 71; HMA v Gordon, 1996 SCCR 274. 153  Smart v HMA (n 74); Goldie v HMA [2016] HCJAC 69; HMA v Graydon 2016 SCL 418; HMA v Smith 2014 SCCR 39; Muirhead v HMA 2014 SCL 702; HMA v Q 2013 SCL 440; Macintosh v HMA High Court of Justiciary on appeal, 11 December 2013, unreported; HMA v Currie (n 75); Campbell v HMA 2005 GWD 35-662; HMA v Gilmour 2004 SCCR 117; Kane v HMA (n 91); Robson v HMA High Court of Justiciary on appeal, 26 November 2003, unreported; Paton v HMA 2002 SCCR 57. See, however, Lady Paton’s dissenting opinion in Muirhead, ibid at [1]–[4]. For commentary, see Brown, 2014g, the case note by Brown, 2014c: 8, and the case note by Phillips, 2004. See also, Lord Gill’s dissenting opinion in HMA v Gilmour, ibid at [1]–[4]. 154  HMA v M (sub nom HMA v McArthur) 2010 SCL 1309; HMA v Edge 2005 GWD 20-360. 149 

Conclusion

 191

with the intuitive-override model. Judge 6, for example, explained how, having considered the nature of the crime and the offender’s record, he arrived—instinctively—at a particular sentence. The respondent then applied a ‘cross-check’ to the figure by considering previous reported decisions and applying any relevant guidelines. This allowed him to see whether his instinctive reactions squared with ‘a more logical, rational thought process’. Judge 3 explained that he would always have a range within which to exercise his discretion. The range in the individual case was dictated by his own experience, his researches into dispos­ als in similar cases (including guideline judgments), and the ‘many, many hours’ he spent in deliberation. Judge 3 explained that, within this range, the final sentence may be reduced by defence counsel’s plea in mitigation. Having heard the plea in mitigation, Judge 3 would then ‘tailor the sentence to the specifics of the case’. Similarly, Sheriff 9 explained that he formed a ‘provisional view’ as to sentence by read­ ing the case papers the night before the sentencing diet. This ‘provisional view’ could be altered in light of the plea in mitigation; the respondent reported that he had departed from his initial assessment on many occasions as a result of a particularly strong plea in mitiga­ tion. Sheriff 15, meanwhile, spoke to ‘weighing all the relevant factors’ to determine ‘the appropriate range’ for the offence, before ‘fine tuning’ the sentence by consulting previous decisions and any guideline judgments. Sheriff 7 described his approach to sentencing in very similar terms: I think you have an instinctive reaction when you look at someone’s record, coupled with the charge that they’re facing. I tend to look initially at the charge and the accused’s record. I may then form a view that it crosses a particular line … and takes him into the arena of a potential custodial disposal. If that’s the case, I then move on to look at the background and the circumstances … Occasionally the context as presented in the social enquiry report, or the plea in mitigation, might bring you back across the line which you think you’ve crossed (Sheriff 7; see also Hough et al, 2003: 37 and Tombs, 2004: 42–45).

For 15 respondents in the present study, cogent sentencing decisions meld first impressions— gestalt—borne of their ‘vast reservoirs of experience’ with deliberate analysis of the facts and circumstances of the case, of statutory provisions and of sentencing guidance and principles laid down by the Appeal Court. The deliberation in which these respondents engage prior to passing sentence is based on critical reflection over their intuition, thereby allowing them to impose a sanction which, in their expert view, is deemed most likely to ‘work’ (Berlin, 1997: 47; Flyvbjerg, 2001: 17).

Conclusion The English style of sentencing guidelines can be seen as a form of epistemic knowledge. The Sentencing Council has both strived for and sought to legitimate English sentencing practice in terms of this ideal. Judicial application of English guidelines involves the appli­ cation of knowledge and skill (techne) to accomplish the goal of uniform, consistent, fore­ seeable and certain sentencing decided by the exercise of theoretical wisdom (or sophia) expressed within the ‘general truths’ (Flyvbjerg, 2001) or the universal rules of presump­ tively binding, numerical guidelines. These guidelines are themselves articulated in terms of

192 

Structuring the Sentencer’s Discretion

relatively prescriptive theoretical axioms rather than centring upon the practical wisdom of the judge. In the English guidelines we see punishment being imposed by monist decision makers, or Berlinian ‘hedgehogs’. In English sentencing we see the primacy of knowledge itself, rather than the primacy of the thinking and judging sentencer. As Aas (2005) argues, action is tied to knowledge and there is little space or need left for reflection. By comparison, the Scottish courts, like the courts in the Republic of Ireland, Canada and Australia, have always taken the view that the retention of a wide judicial discretion in sentencing is necessary and that individual justice is more important than either consist­ ency or an abstract notion of systemic fairness. On this basis, whilst the English guidelines are no more than a sentencing algorithm—or a form of knowledge—the Scottish approach follows that of the Australian and Canadian courts in that appellate sentencing guidance in this jurisdiction comprises an amalgam of knowledge, virtue and reasons that enables the value pluralist judge (a Berlinian ‘fox’) to craft a sentence appropriate to the facts and circumstances of the individual case. Thus, the tasks facing the sentencer in Scotland and in England and Wales are very dif­ ferent. Whilst English judges employ techne through application of the theoretical wisdom contained in the Sentencing Council’s guidelines—arguably an example both of the domi­ nance of rule-based rationality over practical experience described by Flyvbjerg (ibid) and of the legal system’s internal complexity described by Posner (2013)—a different form of wisdom is required of judges and sheriffs in Scotland. Sentencers in this jurisdiction must acquire the requisite knowledge and virtue in the spheres of criminal law and sentencing and then match that knowledge and virtue to particular factual situations. In so doing, judges and sheriffs must analyse the values and interests of the offender, the victim(s) and of society. Rather than applying universal rules set out in presumptively binding guidelines through techne (as in England and Wales), sentencers in Scotland make value judgments about specific cases in which context is a key consideration. Whilst analytical reasoning is central to the decision making process in the English courts, in Scotland the judge or sheriff ’s decision making is intuitive, holistic, and interpretative; in short, sentencers in this jurisdiction practice phronesis. The Scottish approach is more conducive to achieving justice in the individual case. Sen­ tencing in Scotland focuses on the particular and the situationally dependent rather than on the universal and on rules. Scottish sentencing practice is thus a paradigm example of the operation of phronesis: the concrete and the practical are emphasised over the theoretical (Flyvbjerg, ibid). Whilst English sentencing under the Sentencing Council’s guidelines is con­ cerned with producing uniform and consistent sentences, judges and sheriffs in Scotland are concerned with individualised justice—imposing a sentence that takes account of all the ­circumstances of the case. This is achieved not by judicial recourse to a presumptively bind­ ing sentencing guideline largely divorced from context, but by consideration, judgment and choice exercised in accordance with the sentencer’s experience (Flyvbjerg, ibid). The introduction of prescriptive, presumptively binding sentencing guidelines has transformed the normative discourse of sentencing in England and Wales: an individual­ ist approach to the just allocation of punishment has been challenged and replaced by a discourse which privileges an abstract and de-personalised approach to justice (see Aas, 2005; Hutton, 2006b). English policy makers and the Sentencing Council have arguably spent too much time focusing on episteme (analytical, scientific and context-independent knowledge) and, in drafting the present form of guidelines, have neglected the ideal of

Conclusion

 193

p­ hronesis. In so doing, the rule-based, rational mode of thinking has been elevated in English sentencing to being the only legitimate approach. This approach is subject to increasing criticism from experienced practitioners. Those critics who routinely appear as counsel in the English courts argue in favour of a return to a more discretionary based approach to sentencing (see Cooper, 2013 and 2008, and Harris and Gerry, 2013; see also Thomas, 2013 and 2010). Nevertheless, the result of the Sentencing Council’s guidelines is that, in the main, sentencing decisions in England and Wales are rarely more than the relatively simple, rule-governed acts found at the lowest levels of Dreyfus and Dreyfus’s learning process considered in Chapter 5. The experience and intuition which are such important parts of a judge or sheriff ’s deci­ sion making are not, however, a kind of guesswork. The knowledge and skills required of a sentencer cannot be taught a priori but emerge out of practice. As a phronimos—a person of practical wisdom—the sentencer must be able to select and focus on those aspects and features of the case before him which are most relevant to his aims in crafting an appro­ priate sentence. This is a task for which a high degree of judicial discretion is essential. As sentencing is a social practice and performance which is woven into a constantly changing social fabric, the sentencer is not a detached observer as in the epistemic model, but some­ one who must intuitively address the given social context. This requires expertise based on experience, judgment and intuition (Flyvbjerg, ibid). As sentencing is a task which cannot be reduced to the mechanical application of rigid formulae but is one that involves the making of an overall judgment from a consideration of numerous factors based on judicial experience and instinct, judicial sentencing methodology is best conceptualised, not in terms of the ­Australian ‘instinctive synthesis’, but as a phronetic synthesis of the facts and circumstances of the individual case. Yet the wide judicial dis­ cretion exercised by way of the phronetic synthesis is not completely unfettered. The last 10 years have seen an increased use by the Appeal Court of its statutory power to issue guideline judgments along with an increased willingness to address and articulate ques­ tions of principle in sentencing. Such guidelines and appellate guidance are very different to the English approach: as the Appeal Court has repeatedly stated, guidelines in Scot­ land are intended to structure, but not remove, judicial discretion and thus still allow for individualised justice. The Scottish approach essentially comprises judicial self-regulation through appellate review and judicially imposed sentencing guidance. The modern Scottish approach has resulted in a significant corpus of sentencing jurisprudence—what O’Malley, in the Irish context, describes as a ‘sentencing canon’ of leading appellate decisions which are authoritative but not inflexible (O’Malley, 2006, 2001). The Scottish system is far from perfect, a major drawback being the fact that the system is reactive. Whilst the Appeal Court can at present only make statements of principle in response to the issues raised by the particular case under consideration, it is to be hoped that this issue will be addressed when the Scottish Sentencing Council, an advisory body, starts to submit its own guidelines to the Appeal Court for consideration. Nevertheless, the current approach of the Appeal Court is an effective basis for the devel­ opment of principles that are highly sensitive to the practical problems of sentencers (Ash­ worth, 2009). It provides sentencers with a framework within which they can locate the individual case, without depriving them of the discretion to deal differently with a case which has unusual features. It produces a discretion underpinned by principles, rather than hemmed in by rules (O’Donnell, 2012; O’Malley, 2011, 2001).

194 

Structuring the Sentencer’s Discretion

The present study found widespread support for this approach amongst S­ cottish judges and sheriffs. All but two of the 25 respondents were of the view that sentencing must remain discretionary to comply with the demands of justice. Whilst the study found that most respondents welcomed guidance from the Appeal Court, systems of prescriptive and pre­ sumptively binding sentencing guidelines which promote a rule-based rationality over practical experience did not find favour. ‘Principled discretion’ (O’Malley, 2011, 2001) is, in the Scottish context, achieved through a blending of the phronetic synthesis with both non-binding sentencing guidelines and appellate sentencing guidance. The present study found that principled discretion was exercised by 23 of the 25 respond­ ents in one of two ways. The approach of the most experienced sentencers (eight respondents) was purely instinctive: these respondents conceptualised sentencing in terms of an exercise of intuition and judgment in which the key factor was their experience gained whilst in practice at the Bar and since appointment to the Bench. Intuition was still a crucial factor in the sentencing practice of 15 sentencers in the study; however, these respondents reached their decisions by combining their first impressions of a case with deliberative analysis of the facts and circumstances, of the relevant statutory provisions, and of any sentencing guidance—in short, through critical reflection over their intuition. Both these approaches are characterised not by the staged, rule-based approach of the current English guidelines system, but by the practical wisdom of the sentencer. As alluded to earlier,155 one of the few ways in which the judge or sheriff ’s sentencing dis­ cretion can truly be said to be constrained is through the statutory requirement to consider the granting of a discount in sentence following the tendering of a guilty plea—a statutory provision which has generated a substantial volume of case law. The next chapter proceeds to examine the operation of the guilty plea discount in terms of the core themes of discre­ tion, practical wisdom and individualised justice discussed thus far. In considering and evaluating judicial concerns about the operation of the discount principle, a unique insight is provided into the operation of the legislation and of these aspects of sentencing practice.

155  See the earlier discussion under the section entitled, ‘A Blending of the Phronetic Synthesis with Sentencing Guidance—Principled Discretion in Scottish Sentencing’.

7 A ‘Seedy Little Bargain with Criminals’? Judicial Discretion and the Guilty Plea Discount Introduction It has been argued that, to the extent that judicial sentencing methodology relies on the sen­ tencer’s intuition and experience, the sentencing task is best viewed in terms of a phronetic synthesis of the individual facts and circumstances of the particular case. Whilst, to use the terms employed by Flyvbjerg (2001: 21) and Aas (2005: 26), sentencing in Scotland is a form of decision making that is case-orientated, concrete and subjective, there is one aspect of the decision making process in this and in other common law jurisdictions that is system-orientated, abstract and objective, viz the statutory duty to take into account an offender’s plea of guilty and the practice of allowing such an offender a discount in sentence to reflect his guilty plea. This chapter examines the practice of guilty plea discounting1 in Scotland and uses it as a lens to throw into sharp relief the importance of judicial discretion, equity and individualised justice discussed in Chapters 3 to 6. The chapter begins by examining the law and practice of guilty plea discounting in Scotland; the t­heoretical and policy bases on which such discounts are awarded are then considered. Through a discussion of the empirical findings of the present study, the practical risks of guilty plea discounting are then examined along with the ways in which, despite the statutory duty to consider the granting of a discount, Scottish s­entencers continue to do equity in individual cases to achieve ­particularised j­ustice. In prioritising economic efficiency and cost savings within the ­criminal justice system over the traditionally accepted need to impose sentences ­proportionate to an offender’s wrongdoing, the practice of guilty plea discounting ignores the ­contextualised nature of the process and conflicts with the instinctive, or phronetic, synthesis approach to sentencing.

1  The terms ‘guilty plea discounting’ and ‘sentence discounting’ are used synonymously: whilst the former term is strictly more accurate, the latter is widely used both by practitioners and by sentencers themselves.

196 

Judicial Discretion and Guilty Plea Discount

Sentence Discounting for Guilty Pleas: Law and Practice in Scotland The Traditional Approach to Guilty Plea Discounting There was traditionally something of an antipathy towards the practice of ­sentence discount­ ing in Scotland. It was viewed by the Appeal Court as an ‘objectionable practice’2—a form of plea bargaining that offended against the presumption of innocence which disabled sen­ tencers from exercising their discretion fully and freely in a particular case (see Leverick, 2010: 148, 2004: 361, 2003: 267; M ­ cConville and Marsh, 2014: 193–95). Permissive leg­ islation on sentence ­discounting s­ubsequently came into effect in 1995 (Leverick, 2004: 361), removing the restrictions placed on sentencers by the decision in Strawhorn v McLeod (Bonomy, 2002: 37–41; Gordon et al, 2016: 337).3

The Statutory Provisions The statutory provisions relating to sentence discounting are contained in section 196 of the Criminal Procedure (Scotland) Act 1995. Section 196(1) provides that in sentencing an offender, the court ‘shall take into account’ (emphasis added) the stage in the proceedings at which the offender indicated his intention to plead guilty and the circumstances in which the indication was given. Sentencers are placed under a duty to state whether a discount has been granted (section 196(1A)(a)) and to specify the amount of the discount (Practice Note (No 1 of 2008)—Recording of Sentencing Discount). Reasons must be given if no such discount is granted to an accused (section 196(1A)(b)). Section 196 says ­nothing about the scale or magnitude of discounts (Leverick, 2014: 338, 2013: 260), but implies that the discount should be greater the earlier the plea is tendered.4 It was not until the decision in Du Plooy v HMA that guidelines on the practice of s­ entence discounting emerged.5

Guidelines on Guilty Plea Discounting—the Decisions in Du Plooy and Spence In Du Plooy the Appeal Court laid down guidance for sentencers ‘as to the basis of, and scope for, an allowance in the sentencing of an accused in respect of the fact that he has pled guilty, and the form which such an allowance might take’.6 The Court held that section 196(1) of 2 

Strawhorn v McLeod, 1987 SCCR 413 at 415. ibid. For a full discussion of the development of sentence discounting in Scotland, see Leverick, 2012: 233, 2010: 148, 2004: 360–69, and 2003: 267; McConville and Marsh, 2014: Ch 7; and Tata, 2010a: 203–04. 4  Gemmell v HMA 2012 JC 223 at [14]. 5  Du Plooy v HMA, 2005 1 JC 1. Although reported in the 2005 volume of Justiciary Cases, the Appeal Court issued its opinion in Du Plooy on 3 October 2003. 6  Du Plooy, ibid at [1]. 3 Strawhorn,

Sentence Discounting for Guilty Pleas

 197

the 1995 Act enabled the sentencer to make allowance, according to the circumstances of the case, for the fact that the tendering of a guilty plea was likely to save public money and court time, and in general to avoid inconvenience to witnesses; or, in certain types of cases, to avoid the additional distress that could be caused by complainers and witnesses having to give evidence.7 The Appeal Court adopted the language of Spigelman CJ of the Court of Appeal of New South Wales in R v Thomson8 and of Kirby J of the High Court of Australia in Cameron v The Queen in holding that these specific considerations gave ‘utilitarian value’ to the plea.9 The Court in Du Plooy laid particular stress on the discretion available to the sentencer and for that reason declined to lay down any maximum or minimum discount, whilst at the same time expressing the view that a discount should not normally exceed a third of the sentence which would otherwise have been imposed.10 Subsequent deci­ sions of the Appeal Court continued to stress the efficiency justification for the sentence discount (see the discussion by Leverick, 2010: 151–53); however, the Du Plooy guidance was not applied consistently by the courts (see Orr, 2007: 144 and Brown, 2011b: 4–6). As one sheriff in the present study remarked: I can’t make head nor tail of the decisions in relation to Du Plooy! The Appeal Court’s problem is that you can get three people one day who say one thing and three people the next day who say another thing. It’s impossible to work out what seamless thread there is supposed to be running through this (Sheriff 8).

The manner in which discounts were applied led the Appeal Court in Spence v HMA11 to give further guidance on the appropriate levels of discount in ­solemn proceedings, ­similar guidance in respect of summary proceedings having been given some three months ear­ lier by the Appeal Court in Leonard v Houston.12 In Spence, the Court set out a sliding scale of sentence discounts of one third (or, in ‘exceptional circumstances’, possibly more), one quarter, and a maximum of one tenth for guilty pleas tendered by way of section 76 ­procedure (where the accused can request an early hearing specifically to plead guilty), at a preliminary hearing, and at the trial diet respectively.13 The scale in Spence was described as comprising ‘broad figures … intended for guidance only’.14 The Court emphasised that the figures were not prescriptive and that the amount of the discount (if any) in a particular case was dependent on the individual circumstances.15 7 

Ibid at [16]. [2000] NSWCCA 309 at [122]. 9  Cameron v The Queen [2002] HCA 6 at [65], cited in Du Plooy (n 5) at [11] and [12]; see also [14], [19] and [21]. It should be noted, however, that Kirby J delivered a separate, partially dissenting judgment in Cameron; the majority judgment (Gaudron, Gummow and Callinan JJ) justified the practice of sentence discounting not on the ground of pure administrative convenience, but on the ground that a guilty plea can be seen, subjectively, as ‘the willingness of the offender to facilitate the course of justice’ (Cameron, ibid at [13]; see also [14] and [22]). On this view, a guilty plea will not be recognised on the sole basis that it saved the community the expense of a trial (Findlay et al, 2014: 253). The decision of the majority in Cameron places the individual offender’s subjective mental state and motivation for tending the guilty plea ahead of the objective utilitarian value of the plea (ibid)—a fact which was arguably overlooked by the Appeal Court in Du Plooy (n 5). 10  Du Plooy (n 5) at [26]. See also Orr, 2007: 143. 11  2008 JC 174. 12  2008 JC 92. 13  Spence (n 11) at [11] and [14], the substance of the guidance in Leonard, ibid being broadly the same—see Leverick, 2008a: 44. 14  Spence (n 11) at [15]. 15  Spence (n 11) at [15]. 8 

198 

Judicial Discretion and Guilty Plea Discount

The decision in Spence firmly entrenched efficiency considerations as the ­primary j­ustification for sentence discounting—the ‘utilitarian value’ of an early plea was the only rationale mentioned by the Court (Leverick, 2008a: 45).16 ­Moreover, the sliding scale in Spence was effectively identical to the English ­Sentencing Guidelines Council’s Definitive Guideline on Reduction in Sentence for a Guilty Plea (SGC, 2007b; see Leverick, ibid 44 and 2008b: 311). By stressing the ‘utilitarian value’ of a guilty plea, the purpose of the Du Plooy and Spence guidelines was, as in England and Wales, to dispense with trials that were deemed unnec­ essary, ensuring that all those who are guilty plead guilty, and do so at the earliest stage of the process (Ashworth and Zedner, 2008: 34; Tata, 2010a: 203). As Sheriff Chris Shead notes, the principles articulated in Du Plooy were not universally popular with sentencers (Shead, 2013: 94; see also McConville and Marsh, 2014: 190). Practitioners became used to pronouncements from the Bench at sentencing diets to the effect that the judge or sheriff was ‘obliged’ or ‘bound’ by statute and by case law to consider discounting an offender’s sentence to reflect his guilty plea. At appellate level, however, it appeared that the Appeal Court was reluctant to grant discounts in sentence to recidivist offenders;17 the Court was similarly disinclined to grant discounts to offenders whose guilty pleas could be described as ‘tactical’;18 and there was a tendency to restrict the level of discount in cases where the Appeal Court itself had determined to impose a custodial sentence, as opposed to its ­reviewing a custodial sentence imposed by a lower court.19

The Discretionary Nature of Guilty Plea Discounting—the Decisions in Gemmell and Murray Discounting as a Discretionary Process—Gemmell v HMA In Gemmell v HMA,20 the practice of sentence discounting was again considered, this time by a Full Bench in a conjoined hearing in seven appeals against sentence. The Court took the opportunity to examine the rationale and justification for discounting. The opinion of the chairman, Lord Justice Clerk Gill (as he then was), is particularly important in this regard: it was Lord Gill who undertook the task of clearing the jurisprudential Augean Sta­ ble created by Du Plooy, Spence and subsequent cases. The Court in Gemmell was divided on the main issue common to the appeals, that of whether a discount should apply to any public protection element of a sentence (a majority of three to two holding that it should). For present purposes, however, the import of the decision lies in the observations of Lord Gill on the discretionary nature of guilty plea discounting. Before issuing its opinion in Du Plooy, the Appeal Court had repeatedly emphasised that the decisions of whether to allow a discount and, if so, what discount to allow remained

16 

See also Balgowan v HMA 2012 JC 5 at [3]. See, for example, Coyle v HMA 2008 JC 107 and Horribine v Thomson 2008 JC 306. 18 See Cocozza v HMA High Court of Justiciary on appeal, 4 December 2009, unreported. 19  See, for example, HMA v Booth 2005 SCCR 6 and HMA v McMillan [2005] HCJAC 119, discussed by Orr, 2007. 20  Gemmell (n 4). 17 

Sentence Discounting for Guilty Pleas

 199

matters for the discretion of the sentencer.21 This trend continued following the principles articulated in Du Plooy with the Court routinely stating that the question of discount and its calculation remained within the discretion of the sentencer, having regard to all the circumstances of the particular case.22 This aspect of the practice of discounting was again considered by Lord Gill in Gemmell. His Lordship noted that: At the hearing in these appeals, it was said more than once that by reason of an early plea an accused was ‘entitled to’ a discount. That is a careless use of language; but it reflects an approach that is prevalent in certain courts of first instance. From the many appeals that we hear in this court, and from a wealth of anecdotal evidence, it is apparent that in some courts discounts of one-third are being allowed for early pleas as a matter of routine.23

Lord Gill noted that a superficial reading of the decision in Du Plooy may have left ­sentencers in the lower courts with the impression that discounts were there for the asking, a prob­ lem compounded by the decision in Spence which, his L ­ ordship observed, may have cre­ ated ‘a climate of expectation among practitioners’.24 It appears that the Spence guidelines had crystalised into rules—a problem with guidelines generally which O’Malley (2006: 61) identifies. The apparent ubiquity of the view amongst the Scottish Bar that a guilty plea would automatically result in a reduced sentence led Lord Gill to reiterate that an accused is not entitled to any particular discount in return for a plea of guilty. The level of discount, if any, is—and must always be—a matter for the discretion of the sentencer;25 it is, his Lordship stated, a discretion that should be exercised ‘sparingly and only for c­ onvincing reasons’.26 The sentence discounting process was described by Lord Gill as comprising three stages: (i) the sentencer decides what the sentence should be if no question of a discount arose (described as ‘the headline sentence’); (ii) a decision is then made as to whether there should be a discount; and (iii) if so, a decision is taken on the quantum of the discount.27 In deciding upon the headline sentence, Lord Gill considered that a sentencer should apply what he described as ‘the normal, well-established principles of sentencing’, having regard to such factors as the circumstances and gravity of the offence, the accused’s record and the need to protect the public.28 His Lordship elaborated: The assessment of the headline sentence is not a matter of precise arithmetical ­calculation. It involves the making of an overall judgment from a consideration of numerous factors based on judicial experience. It is a process described in the Australian courts as that of ‘instinctive s­ ynthesis’ (cf Markarian v R, [2005] HCA 25) and seen in the ­Canadian courts as a delicate art based on ­competence and expertise (R v M (C A), [1996] 1 SCR 500; R v LM, [2008] 2 SCR 163).29

21  HMA v Forrest 1998 SCCR 153 at 158; Docherty v McGlennan 1999 GWD 4-176; Tennie v Munro 1999 SCCR 70 at 71. 22 See Will v HMA [2010] HCJAC 113 at [13] and RB v HMA 2004 SCCR 443 at [17] and [22]. 23  Gemmell (n 4) at [30]. 24  Gemmell (n 4) at [31]. 25  Gemmell (n 4) at [31]. 26  Gemmell (n 4) at [77]. 27  Gemmell (n 4) at [27]. 28  Gemmell (n 4) at [28]. 29  Gemmell (n 4) at [59]. These comments were later endorsed by Lord Justice Clerk Carloway (as he then was) in McGill v HMA 2014 SCCR 46 at [13] and Ferguson v HMA 2014 SCCR 244 at [103].

200 

Judicial Discretion and Guilty Plea Discount

Lord Gill considered that ‘the same process applies to the assessment of a sentence discount’30 and explained: [T]he sentencer should apply section 196 by deciding what sentence is appropriate in the circum­ stances of the case and then simply applying the discount, if any. In this way the sentencer is saved the effort of making complex calculations, particularly in a busy sheriff court, and the resulting sentence represents a true exercise of judgment w ­ ithout a spurious appearance of arithmetical exactitude.31

Lord Gill considered that the claim that an appellant had received an insufficient discount following his guilty plea was heard too frequently in the Appeal Court.32 Approving the dissenting remarks of Lord Philip in RB v HMA,33 his Lordship considered that where a sentencer has given cogent reasons for allowing the d ­ iscount in question or for declining to apply a discount at all, it would only be in exceptional circumstances that the Appeal Court should interfere.34 Whilst Lord Gill did not explicitly disapprove the sliding scale in Spence, his Lordship noted that it carried ‘a risk of rigidity’; his Lordship emphasised that the Court in Spence had expressed the reservation that the discount in the individual case remained a matter for the sentencer.35 Lord Gill was reluctant to specify the appropriate level of dis­ count, except to note, ‘the broad principle that, in general, the discount will be the greater the earlier the plea is probably a sufficient statement of guidance for most purposes’.36 There was, however, disagreement amongst the majority in Gemmell over the desirability of the sliding scale approach. Whilst Lord Eassie agreed that the a­ llowance of a discount was, ‘in its essential nature, an exercise of discretion’, his Lordship advocated recourse to the Spence scale in all but exceptional circumstances.37 As none of the other members of the Court addressed the issue, sentencers were left to choose between Lord Gill’s discretionary approach and the more formulaic approach advocated by Lord Eassie (see Leverick, 2012: 235 and McConville and Marsh, 2014: 207–08). The discretionary approach favoured by Lord Gill is criticised by Leverick as being expressed in vague terms that took the law back to where it stood before Spence when there were what she describes as ‘considerable inequalities in the application of discounts’ ­(Leverick, ibid; see also Leverick, 2013: 261–62). As McConville and Marsh observe, how­ ever, Lord Eassie’s approach would involve Scottish judges adopting a policy of sentence discounting which simply cannot be reconciled with the historic commitment to the sen­ tencer having discretion in this regard (McConville and Marsh, ibid 208). The discretionary approach was, however, later affirmed in emphatic terms by a Bench of three judges in Murray v HMA.38 Lord Gill concluded his opinion in Gemmell by firing a warning shot across the bows of those counsel and defence agents who had hitherto regarded the granting of discounts as

30 

Gemmell (n 4) at [59]. Gemmell (n 4) at [59]. 32  Gemmell (n 4) at [80]. 33  RB v HMA (n 22) at [22]. 34  Gemmell (n 4) at [81]. 35  Gemmell (n 4) at [78]. 36  Gemmell (n 4) at [78]. 37  Gemmell (n 4) at [145]. 38  2013 SCCR 88. 31 

Sentence Discounting for Guilty Pleas

 201

virtually automatic: ‘Those representing an accused who has tendered an early plea should bear this [viz. the discretionary nature of the discount] in mind when considering whether to lodge an appeal based solely on the amount of the discount’.39 It appears, however, that the climate of expectation engendered by the decision in Spence was so entrenched amongst the Bar (and, indeed, amongst certain judges of the Appeal Court) that Lord Gill’s directions on how the practice of discounting was to operate did not fully filter through to certain practitioners and sentencers (Brown, 2013a: 7). The Appeal Court on occasion continued to grant discounts in a less than discriminating manner and continued to apply the sliding scale in Spence, apparently without regard to Lord Gill’s dicta in Gemmell.40 Thus in ­Murray,41 the Appeal Court sought to clarify and cement the princi­ ples laid down in Gemmell, in particular the discretionary nature of the guilty plea discount (Duff, 2013: 1; Leverick, ibid 262).

The Discretionary Nature of Discounting Affirmed—Murray v HMA The appellant in Murray pleaded guilty by way of section 76 procedure to various sexual offences, including rape, committed against elderly residents of the care home in which he worked. The appellant had deliberately targeted residents with severe dementia who were unable to communicate. The sentencing judge, applying the decision in Gemmell, under­ stood the level of discount to be a matter for his discretion and imposed a cumulo sentence of seven years’ imprisonment, discounted from a headline sentence of nine years. Despite Lord Gill’s remarks in Gemmell that the Appeal Court would be disinclined to entertain appeals against sentence based solely on the ground of a supposedly insufficient discount having been awarded, an appeal was lodged on the ground that, as the appellant’s guilty plea had been tendered under section 76 procedure, the sentencing judge had erred in restricting the discount and that a discount of one third ought to have been allowed. As the appeal raised an important point of principle, it was referred to a Bench of three judges comprising, inter alios, Lord Gill (by this stage Lord Justice G ­ eneral) and Lord Jus­ tice Clerk Carloway (as he then was). This appears to have been the first occasion in which Scotland’s two most senior judges sat together in a three judge sentence appeal (Gordon, 2013: 95). In delivering the leading opinion, Lord Gill rehearsed the salient points of his opinion in Gemmell, noting that an accused is not entitled to any particular discount in return for a guilty plea, that the discount will be greater the earlier the plea was tendered, and that in order to maintain public confidence in the justice system and the credibility of sentences, the court’s discretion to allow a discount should be exercised sparingly and only for convincing reasons.42 His Lordship reiterated that the decision in Spence did not lay down a mandatory sliding scale: it reserved the decision to the discretion of the sentencer in every case.43 The present appeal was, his Lordship observed, misconceived as it was founded

39 

Gemmell (n 4) at [81]. See, for example, Docherty v HMA [2012] HCJAC 106 and Lees v HMA [2012] HCJAC 143—decisions which were, with respect, difficult to reconcile with Lord Gill’s opinion in Gemmell (n 4); see Brown, 2013a: 7; 2012b: 8 and 2012d: 7; and McConville and Marsh, 2014: 208–10. 41  Murray (n 38). 42  Murray (n 38) at [16]–[17]. 43  Murray (n 38) at [24]. 40 

202 

Judicial Discretion and Guilty Plea Discount

on the notion, discredited in Gemmell, that an early plea entitles every accused to a discount of one third. This, Lord Gill noted, was not the case.44 Thus, the appeal was ill-advised: [The appeal] was presented in the face of the principle recognised by this court that it is only in exceptional cases that the court will interfere with a discretionary decision on discounting for which the sentencer has given cogent reasons.45

In disposing of the appeal, the Court held the headline sentence of nine years to be unduly lenient. The appellant’s sentence was increased to nine years and six months’ imprisonment, comprising a headline sentence of 12 years discounted by approximately a fifth.46

Discounts and the Nature of Sentencing Guidelines Sheriff Alistair Duff observes that the decision in Murray underlines in an emphatic ­manner the scope of the discretion of the first instance sentencer to consider the ­question of ­discount.47 As McConville and Marsh explain, ‘discretion triumphed over policy’ ­(McConville and Marsh, 2014: 212); or, in Flyvbjerg’s terms, the concrete and the practical were emphasised over the theoretical (Flyvbjerg, 2006: 373, 2001: 58). In giving pre-­eminence to the discretion of the sentencer, the Appeal Court underlined the ‘unique character and individuality of every case’ (McConville and Marsh, ibid 249). The contrast between, on the one hand, Lord Gill’s opinions in Gemmell and Murray in which the discretionary nature of ­discounting is emphasised and, on the other, counsel’s submissions in the ­Murray appeal in which—pace Gemmell—discounting was seen in structured, ­formulaic and mechanistic terms, highlights the different approaches to sentencing guidelines ­discussed in Chapter 6. Counsel’s submissions in Murray—particularly on the supposed effect of a ­section 76 plea—were akin to the formal, prescriptive, heavily structured, ­numerical and presump­ tively binding approach of the English courts, as dictated by s­ ection 125(1) of the Coroners and Justice Act 2009 and the guidelines issued by the Sentencing Council.48 The Court in Murray, however, continued to favour the traditional Scottish approach towards sentencing guidelines. This approach, as we have seen,49 maintains that whilst sentencers are obliged by statute to have regard to guidelines, they retain the responsibility for determining the appro­ priate sentence. It is a system in which sentencing guidelines provide a structure for, but do not remove, judicial discretion and in which guidelines should never be interpreted in such a way as to lead to a mechanistic approach (Brown, 2013a: 7–8 and 2014a: 5–6; cf Roberts,

44 

Murray (n 38) at [25]. Murray (n 38) at [26]. 46  Murray (n 38) at [27]–[29], Lord Justice Clerk Carloway (as he then was) and Lady Paton ­concurring. Thus, as Sheriff Welsh QC observes: ‘… when it comes to the ever growing jurisprudence s­ urrounding “discounts”, it may be more prudent not to open what could be a Pandora’s Box in the appeal court. Discount is more art than science, quieta non movere’ (Welsh, 2013a: 251–52). 47  Duff, 2013: 6; see also the later decisions in Archer v HMA 2014 SCL 150 at [18]; Forrester v HMA High Court of Justiciary on appeal, 27 January 2015, unreported at [6] (see the case note by Brown, 2015f); Geddes v HMA 2015 SCCR 230 at [21]; McInally v Dunn 2016 SCCR 243 at [8]–[9]; Findlay v HMA 2016 SCL 731 at [11]; and Gilchrist v PF, Dumbarton [2016] SAC (Crim) 28 at [3]–[4]. 48  See the discussion in Chapter 6 under the section entitled, ‘The Fourth “Phase”—Presumptively Binding Guidelines and the Sentencing Council’. 49  See the discussion in Chapter 6 under the section entitled, ‘Principled Discretion’. 45 

Sentence Discounting for Guilty Pleas

 203

2012b: 278–79 who discusses operation of the English sentencing guidelines on guilty plea discounts—SGC, 2007b50—in terms of an accused’s ‘entitlement’ [sic] thereto). The decision in Murray encapsulates the essential difference between judicial use of discretionary sentencing guidelines in jurisdictions such as Scotland, the Republic of Ire­ land and much of Australia and, alternatively, judicial recourse to prescriptive and pre­ sumptively binding numerical guidelines in England and Wales. These very different approaches can be seen as a dichotomy between the two different methods of judicial decision making in sentencing considered in Chapters 5 and 6. The English approach— and indeed the earlier guidance of the Appeal Court in laying down the sliding scale of discounts in Spence—is s­ ystem-orientated, abstract and objective. This approach creates a clear conflict between the pursuit of consistency and the value of individualised sentencing (see O’Malley, 2006: 61). For McConville and Marsh, it creates ‘a non-­individuated assemblyline production process’ (McConville and Marsh, ibid 85–86). The ­Scottish approach, how­ ever—exemplified in Gemmell and reiterated in Murray—is more case-­orientated, concrete and subjective (Aas, 2005: 26; Brown, 2013a: 7–8; and F ­ lyvbjerg, 2001: 21). As Lord Gill observed in Gemmell, the assessment of an accused’s sentence and the deci­ sions relating to any discount for a guilty plea (viz whether to grant a discount and, if so, the level of discount) are all discretionary judgments based on judicial experience.51 We have seen how, in discussing the importance of judicial experience in Gemmell, Lord Gill described both the process of sentencing and of assessing the discount in terms of the Markarian instinctive synthesis.52 Lord Gill’s opinion in Gemmell emphasises the primacy of the sentencer as an expert actor and phronimos. It is the sentencer’s vast reservoir of practical experience that equips him or her to achieve equity in the sense of particularised justice (Solum, 2008: 145, 2003: 205, and 1994: 120 and 123; see also Berlin, 2013a: 467 and Scott, 2005: 27)—equity both in setting the headline sentence and in addressing the ques­ tion of discount. As demonstrated by the decisions in Gemmell and Murray, the essence of the Scottish approach to sentencing and to the use of guidelines is thus the intuitive, holistic and interpretive form of decision making involved in phronesis (Flyvbjerg, 2001: 14).

Discretion in Discounting—the Findings of the Present Study Although the judicial interviews for the present study were completed some 13 months before the decision in Gemmell was issued, the concerns which Lord Gill raised in Gemmell and in Murray regarding, firstly, counsels’ perception of the supposedly automatic ‘right’ to a guilty plea discount and, secondly, the prescriptive nature of the practice following 50  On 11 February 2016 the Sentencing Council for England and Wales issued a consultation paper on a pro­ posed new guideline for reduction in sentence for a guilty plea (Sentencing Council, 2016). For discussion, see Roberts and Harris, 2016. The Council’s proposals have been criticised by practitioners as prioritising consistency at the expense of defendants (Cohen-Lask, 2016: 9); as writing out the role and value of defence solicitors in assess­ ing the evidence in any given case (R Harris, 2016: 15); as placing pressure on certain defendants to plead guilty (Padfield, 2016: 302); and as circumscribing judicial discretion, depriving judges of the discretion to do justice in the individual case (Stanbury, 2016: 6–7); cf Treacy, 2016. 51  Gemmell (n 4) at [59]. 52  Gemmell (n 4) at [59].

204 

Judicial Discretion and Guilty Plea Discount

Spence were echoed by 15 respondents (6 of the 8 judges and 9 of the 17 sheriffs) in the study. The six judges expressed concern that the guidance in Du Plooy and Spence had resulted in discounting becoming too mechanistic. The nine sheriffs also voiced concern about this aspect of the practice, whilst certain sheriffs—although not necessarily approv­ ing of the practice of discounting—explained how the Appeal Court guidance had resulted in an expectation amongst local defence agents that sentences would always be discounted to reflect guilty pleas. One respondent (a very experienced judge) criticised counsel for maintaining that ‘their client is always entitled to a particular discount’ (Judge 8). As the respondent noted, how­ ever, ‘the guidelines are just guidelines and it’s really still up to the sentencing judge to make up his mind’ (ibid). Another experienced judge was extremely critical of the ease with which guilty plea discounts were routinely granted in the lower courts: The problem is that sheriffs have been dishing out discounts like sweeties. They give a third off just like that. But I do understand that they are subject to different pressures than we are in the High Court. They are very busy. Often when they sentence in the sheriff court it’s just ‘bang, bang, bang’. It has to be, otherwise they’d be there until 10 o’clock at night (Judge 5).

A third judge was particularly critical both of the guidance in Spence and the way in which the guidance had been interpreted by the Appeal Court: I’m generally unhappy about the way that the discount arrangements work at the moment … I think we should be refusing to allow discounts more often. I very often tell people that you’re not getting any discount: you’ve done such and such and it’s entirely your own fault. The trouble is that when sentences are appealed against, there are those in the Appeal Court who will try to apply mathematical, clockwork discounts … Some people will just automatically give a third off just because it’s a kind of box ticking exercise. Whereas, if possible, I won’t give anything like that level of discount because I don’t think that’s the way it should really operate … Some judges—the metronomic clockwork men—are taking automatic percentages in certain situations irrespective of the merits of the case, whereas others among us (myself included) are not doing that (Judge 3).

This respondent deprecated the Appeal Court’s adoption in Spence of the same sliding scale applied by the English courts: We’ve got no independent thought of our own going into it! That’s not discretion. It’s a kind of mechanical, clockwork mentality that produces that sort of stuff. It just doesn’t work (Judge 3).

These views were shared by another judge who, in discussing the Spence guidance, said this: I think that that sort of tariff, box ticking approach is one which is not conducive to doing justice to the complexities of a case … I think that reduces sentencing to box ticking. It can never be a box ticking exercise because of the complexities of the individuals concerned and the different permuta­ tions that can arise … It’s farcical, just farcical, because it can result in a sentence that’s completely disproportionate to the circumstances (Judge 2).

A more recent appointee to the High Court Bench concurred: I have concerns about whether [the guidance in Du Plooy and Spence] imposes a straitjacket on the sentencer. I’ve often felt that there were circumstances in which to follow the guidance in its strict form appeared to result in a sentence which was inadequate. I have myself on some occa­ sions declined to provide a full one third discount in section 76 cases; I’ve also declined to provide the guideline discount of 20 or 25 per cent in the circumstances identified by Spence because it seemed to me that to do so would just result in an outcome which was disproportionate … Courts shouldn’t take a mechanistic approach to sentencing in section 76 cases. They shouldn’t just simply

Sentence Discounting for Guilty Pleas

 205

assume that such discounts—or any discounts—have to be afforded in every single case. There is an element of discretion available to the sentencing judge according to the circumstances of the case … I think there is a general feeling—certainly I feel it—that the sort of approach in Spence just isn’t consistent with justice (Judge 1).

Even one of the few judges who considered that guilty plea discounting could be justified purely on efficiency grounds expressed concern at how discounting ­operated in practice: There are problems with Du Plooy in the sense that it can be slightly too arithmetical an approach in certain cases … I don’t think we should encourage appeals which are based purely on a suppos­ edly insufficient discount having been allowed … I think we’re in danger of doing that. Looking at it from an appellate point of view, we should be looking at the overall result (Judge 7).

A similar view was taken by an experienced sheriff who, whilst noting the tendency amongst certain sheriffs to grant one third discounts as a matter of routine, explained that he did not personally follow such an approach: It’s got to the extent now that agents are inclined to spend ages right at the start of their plea in mitigation explaining why it was that this plea of guilty was only tendered at such and such a stage. I sometimes cut them short by saying that I don’t think the correct approach is to go into all of that (Sheriff 17).

The structured, formulaic and mathematical approach exemplified by the sliding scale in Spence was also criticised by sheriffs: Discounting is a purely arithmetical, mechanical exercise … The Spence guidance? Human nature being what it is, there is a tendency once a guideline is issued for it to become a tramline. It’s as ­simple as that … Uppermost in my mind is what is a fair disposal at the end of the day? And dis­ counts simply do not fit … To my mind, discounts are extremely unhelpful and simply complicate an already very complicated picture (Sheriff 10). From my chats amongst colleagues, here and elsewhere, there’s quite a wide divergence. There are people who do the arithmetic, for God’s sake! They divide by three, work it out in decimal points, work it out in days. What a piece of nonsense! That’s not the object of the exercise at all (Sheriff 6). It’s getting ridiculous … I think the whole [sentence discount] principle is overdone. I suspect every sentencer of first instance thinks it’s overdone (Sheriff 3). The Appeal Court may have seen [the Du Plooy and Spence guidance] as a way of ­reducing ­everyone’s workload, but I think it was pragmatism going too far … I’d just tear up Du Plooy and start again (Sheriff 4). It’s almost formulaic now—you now have such a rigid, formulaic system to be gone through. You must give a discount and you know that inevitably there will be an appeal if you don’t give a discount … Because it’s so rigid, it’s difficult to exercise discretion without jumping outside the rigidity of this formula and leaving the accused a ground of appeal … Where you have rigidity, discretion is eroded (Sheriff 14).

A further three sheriffs explained how the ‘climate of expectation’ surrounding sentence discounts amongst defence agents had resulted in their feeling pressured to award discounts for guilty pleas. As one sheriff explained: I think we generally feel that 99 per cent of the time we should be giving a discount of some sort, unless [the accused] pleads guilty during the trial. But even if they plead guilty at the trial diet they’re going to get 10 per cent off. And it may be that people feel that unless they give something, they’ll be appealed. And we get fed up writing these bloody reports [for the Appeal Court]! I think

206 

Judicial Discretion and Guilty Plea Discount

because of the decisions in Du Plooy and Spence there’s a tendency to give a discount of some sort (Sheriff 2).

The second sheriff ’s reasons for granting discounts as a matter of routine was a perceived lack of support from the Appeal Court were he to grant a discount that did not accord with the sliding scale in Spence: It’s maybe just a lack of courage on my part on the basis that the High Court might turn around and say ‘What’s this clown doing? He pled guilty at the first available opportunity; he should have got a third, he only got a quarter, so—bang—it’s a third’. So I just err on the side of what I would regard as caution and just give them a third no matter what (Sheriff 8).

Similarly, the third sheriff—who fundamentally disapproved of the practice of guilty plea discounting—often felt constrained to grant larger discounts than he thought were war­ ranted due to a combination of defence agents’ expectations (which, if not met, would inevitably lead to an appeal) and a perceived lack of appellate support: Defence agents expect discounts and expect, for example, a third off for a plea by way of section 76 procedure. I wouldn’t have any confidence that I would be upheld if I didn’t give the full Du Plooy or Spence figure (Sheriff 5).

It was also found—as with the respondent Judge 7, mentioned earlier—that one sheriff who regarded guilty plea discounting as a practical necessity nevertheless expressed con­ cerns about the mechanical nature of the process and stressed the need for discretion in fixing the level of discount: I think it’s important that it’s not prescriptive to the extent that there is absolutely no discretion. The importance of judicial discretion is still there (Sheriff 7).

Can the Guilty Plea Discount be Justified? The Rationale for the Guilty Plea Discount The practice of discounting a sentence for a guilty plea has traditionally been ­justified in three main ways: on the grounds that it indicates the offender’s remorse;53 on the basis that it spares victims and witnesses the ordeal of giving evidence in court; and on the grounds of efficiency (Leverick, 2014: 339–40, 2013: 259–60, 2012: 234, 2010: 149, 2008a: 45, and 2004: 370; McConville and Marsh, 2014: 221–29; O’Malley, 2009: 877–78; and Rauxloh, 2012: 35–36). The Court in Gemmell identified the efficiency rationale as the most convincing.54 As Lord Gill observed: The euphemism ‘utilitarian value’ may be thought to give the principle of ­discounting some ­ethical content; but sentence discounting is not an exercise in ­Benthamite ­philosophy. It is not based on 53  Traditionally a major justification in England and Wales (see Ashworth, 2015: 181; Bingham, 2000a: 307; McConville, 1998: 653–54; and McConville and Marsh, 2016: 114 and 2014: 72). 54  Gemmell (n 4) at [34] and [137]. On the primacy of the efficiency rationale, see also Leverick, 2010: 149; Mack and Roach Anleu, 1998: 274; McConville, 1998: 564; McConville and Marsh, 2014: 222–26; and Thaman, 2010: xvii.

Can the Guilty Plea Discount be Justified?

 207

any high moral principle relating to the offence, the offender or the victim … It is a statutory encouragement of early pleas … [T]he primary benefit that is realised in every case is the saving of administrative costs and the reduction in the court’s workload.55

In the present study, 11 respondents (three judges and eight sheriffs) regarded the discount principle as essentially an administrative necessity and cited it as either the primary justi­ fication for the practice of guilty plea discounting or as one of the two main justifications (in conjunction with the rationale of saving complainers and witnesses from the ordeal of giving evidence): Pragmatically and realistically the utilitarian value is undoubtedly important … for the criminal justice system to operate efficiently and effectively and as speedily as possible (Sheriff 11). It’s the only way of getting through the business! And trials of course nowadays are much more com­ plex affairs … so I suppose the system has to find ways of getting through the business (Sheriff 17). The utilitarian value is the big one—it’s just part of the system (Sheriff 3). There’s a big problem with the Crown not being ready and with not preparing cases properly. So I can see why lots of pleas are being accepted. I’ve just recently been in the summary trials court and it’s a shambles, frankly. Cases being deserted simpliciter because the Crown haven’t cited their witnesses after a third adjournment … It’s a complete and utter shambles. Things are improving, so they say, but I haven’t seen it yet. But I think that’s why pleas are being accepted, because there is that utilitarian effect and, standing Du Plooy, I suppose we’ve got to reduce the sentence. There’s not a lot you can do about that (Sheriff 16). I feel that the discount set up has to be part and parcel of a modern, pressure driven system and therefore in most instances some discount will be afforded … In my view the tendering of a guilty plea almost invariably involves some sort of discount. I think we need that to keep afloat … I think that sentencers should know about the health and well-being of the system and the fullness or otherwise of the prisons that we may send people to. I think you have to have that in mind … I still feel as an ex civil servant [the respondent had spent most of his career as a prosecutor prior to his appointment to the Bench] that I’m sentencing in the real world. Resources are finite and I have to take that into consideration (Sheriff 12).

It was also accepted by the Court in Gemmell that sparing complainers and other witnesses can be a relevant sentencing consideration.56 Sentencers were, however, urged to bear in mind that such persons will not always be vulnerable; that many complainers will wish the trial to proceed;57 that in many cases (particularly those prosecuted on summary complaint) the witnesses will be police officers and that in such cases a guilty plea will attract ‘at most a token discount’;58 and, finally, that in some cases (such as breaches of preventative statutes or social security frauds) the idea of sparing the victim from an ordeal will be irrelevant.59 The notion of sparing victims from the ordeal of giving evidence had traditionally found considerable support in appellate decisions both in Scotland60 and in England and Wales.61

55 

Gemmell (n 4) at [34]. Gemmell (n 4) at [45]. 57  Gemmell (n 4) at [45]. 58  Gemmell (n 4) at [46]. 59  Gemmell (n 4) at [47]. 60 See Khaliq v HMA 1984 SCCR 212; Sweeney v HMA 1990 GWD 25-1385 and especially Du Plooy (n 5) at [16]. 61 See R v Millberry [2003] 2 Cr App R (S) 31 at [27]–[28]; see also Leverick, 2004: 372. 56 

208 

Judicial Discretion and Guilty Plea Discount

For the reasons outlined by Lord Gill in Gemmell, however, it is now regarded as an uncon­ vincing justification for the practice of sentence discounting in the vast majority of cases (see also Leverick, 2004: 374). Some victims might actually wish to go through the ordeal of giving evidence because it gives them the opportunity to be heard within the criminal justice system (Leverick, ibid 373; McConville and Marsh, 2016: 115); because it may lead to their learning more about the offence and the offender (Sanders et al, 2010: 494–95); or because they feel that, given the choice, they would rather undergo the pains of giving evidence if it means that the offender receives no discount and thus a longer sentence (Ashworth, 2015: 181; Dawes et al, 2011: 19; Fenwick, 1997: 327; Henham, 2012: 157, 2002: 393, 2001: 66, and 1999: 537; and ­Mackenzie, 2006b: 7–8;). McConville and Marsh consider that ‘the symbolism of “victims” … is being exploited to give credence to State-induced guilty pleas and to other marketdriven policies which infuse the criminal justice system’ (2014: 227). The authors conclude that ‘when the process of State-induced guilty pleas needs ideological cleansing, abstract ­“victims” are shamelessly invoked’ (ibid 228). Nevertheless, the rationale of sparing victims and witnesses the ordeal of giving evidence was regarded as particularly important by respondents in the present study for certain types of offence, notably sexual offences and offences involving children. Eighteen respondents (four judges and 14 sheriffs) considered it to be either the primary justification for the prac­ tice of guilty plea discounting or one of the two main justifications (in conjunction with the efficiency rationale later approved by the Court in Gemmell): Obviously the big one is rape, which I don’t do, but I do attempted rape and lewd and libidinous and some of it is pretty nasty stuff. I do think that, obviously, the complainer there will be delighted when the accused pleads. I think they’re delighted not only because they don’t have to give evidence but because it’s a public affirmation that he did something … Anyone who’s actually required to give evidence in something unpleasant such as rape or serious assault that’s happened to them are going to have a bad experience. They’re not going to enjoy it. So for discounting, to me, it’s a big factor (Sheriff 9). Anything involving vulnerable complainers would lead me to the view that the guilty plea was particularly helpful (Sheriff 7). To save people having to re-live sexual abuse is important. I had one about a year or so ago— historic child sexual abuse—two 25 year olds giving evidence in respect of abuse by their father when they were young … The case was very well conducted for the defence by an experienced counsel who did it very, very fairly but these people were so distressed—the things that had begun to come back to them (Sheriff 3).

Two sheriffs, however, cautioned against assuming that it is only in cases of s­ exual abuse that complainers are saved from an ordeal by virtue of the accused’s guilty plea. Sheriff 16 gave the example of a case of dangerous driving where the elderly witnesses became extremely distressed when it was put to them in c­ ross-examination that they had fabricated their account. Another respondent (Sheriff 1) discussed a case of assault to severe injury and permanent disfigurement in which the complainer’s face had been slashed from his ear to his mouth, leaving a permanent scar. During the trial, the complainer was shown—for the first time—a CCTV recording of the incident. As the respondent explained, this was an experience which the complainer found extremely traumatic.

Can the Guilty Plea Discount be Justified?

 209

In views consistent with both the academic literature and the later opinion of Lord Gill in Gemmell, however, four respondents noted that many complainers would rather give evidence than see the offender receive a discounted sentence: Some complainers would rather go through the trial if it meant that the accused would get ‘the full bhuna’. I don’t see why people should assume that a complainer doesn’t want to give evidence. It’s been suggested that it can be quite cathartic to have your pain and anger recognised in public. I quite agree with that. I don’t think it’s for the courts to assume one way or the other (Sheriff 5). Yes, the victim’s been spared giving evidence—and that’s a big factor in many cases—but there may be cases where they think, ‘I want to go to court and give my evidence. Why should that person get something off his sentence just for at last admitting it?’ The injury to the victim is the same … I’m sure that the complainer must feel a great sense of injustice … That’s the problem with policy, in a sense, because a hard line will always create hard cases (Sheriff 15). Many complainers would be much happier to have that person away for a very long time if they gave evidence (Judge 3). The idea that you’ve saved the complainer the ordeal of giving evidence is pretty threadbare. In, say, a rape case or other sexual assault, the complainer might want her side of the story to come out. She might want to give evidence, to say how the crime has affected her. To say that discounts in sentence are justified because they save ‘wee Jeanie’ from distress just isn’t right. It’s just bollocks (Judge 5).

The remorse rationale was rejected in Gemmell.62 Indeed, remorse is not seen as a credible justification either by academic commentators (Chasse, 2009: 57; Darbyshire, 2000: 901; Field, 2002: 264–67; Leverick, 2004: 370; Rauxloh, 2012: 36–37; and Wasik, 2006: 146;) or by the public (Dawes et al, 2011: 31). As far back as 1967, Arthur Rosett (a former lawyer and Assistant US Attorney) noted how, in the US context: One may doubt whether many of the defendants who ‘cop a plea’ on any given day are motivated by this sort of spiritual awakening. In many courts, the guilty-plea process looks more like the purchase of a rug in a Lebanese bazaar than like the confrontation between a man and his soul (Rosett, 1967: 75).

Similarly, in its response to the Sentencing Council’s consultation paper on the proposed new guideline for reductions in sentence for a guilty plea in England and Wales (Sentencing Council, 2016), the Criminal Bar Association observed that: ‘It is not entirely cynical to point out that only a very few remorseful defendants ‘wish’ to plead guilty; almost all do so out of a realistic assessment that they will probably be convicted in any event’ (CBA, 2016: 8). In Gemmell, Lord Gill considered that where section 196 of the 1995 Act applies, it applies regardless of whether the accused has shown remorse.63 Whilst there is seldom any sure crite­ rion for assessing whether an accused is truly remorseful—as opposed to his being ‘forensically resourceful’ (Posner, 1999b: 319) in seeking simply to avoid a greater degree of punishment (see also Ashworth, 2015: 181 and Nicholson, 1981: 218)64—where there is convincing evidence of 62 

Gemmell (n 4) at [51]. Gemmell (n 4) at [51]. That the accused will inevitably be informed by his solicitor about the likelihood of a guilty plea resulting in a discounted sentence, is noted by Dawes et al, 2011: 33; Roach Anleu and Mack, 2009: 4; Tata, 2007: 494–95 and 515; Tata and Stephen, 2006: 736–37; and Tata et al, 2004: 129–31. Those accused who habitually appear before the courts, meanwhile, are well aware of how the discount provisions can operate to their advantage (Brown, 2012d: 7–8; and Thomson, 2008: 131; see also Dawes et al, ibid 21–22 and 32 and Zander, 1993: 85). 63  64 

210 

Judicial Discretion and Guilty Plea Discount

remorse then a sentencer may make allowance for it, as an aspect of mitigation, in deciding on the headline sentence.65 This approach corresponds with both the position in ­England and Wales (Sentencing Guidelines Council, 2007b: ii and paragraph 2.4—see also Sentencing Council, 2016: 13; Maslen and Roberts, 2013: 127; and Roberts and Bradford, 2015: 192) and with the views of respondents in the present study: whilst all 25 respondents acknowl­ edged the relevance of genuine remorse in sentencing, none cited remorse as a justification, or rationale, for the practice of guilty plea discounting. Respondents who discussed remorse in detail viewed this factor as qualitatively different from the question of whether to allow a discount in sentence. Thus, as later set out by the Court in Gemmell, genuine remorse was seen as an aspect of mitigation and not as featuring in the assessment of guilty plea discounts: I think that remorse is something that should be reflected in mitigation rather than as an element of the discount. I don’t think a plea necessarily indicates remorse, that’s the bottom line of it. I think therefore remorse is something which truly should be reflected in mitigation rather than as part of an automatic discount for a plea of guilty. People plead guilty for all sorts of reasons. Remorse may be one of them, in certain cases, and it can properly be reflected in the mitigation of sentence before you give a discount. But I think the idea that a plea automatically means remorse and therefore you are entitled to a discount for that is a flawed concept (Judge 2).

The Practical Risks of Guilty Plea Discounting In the course of his opinion in Gemmell, Lord Gill identified what he described as two ‘significant risks’ in sentence discounting: that the discount may act as an incentive for an accused with a stateable defence to plead guilty66 and that the allowance of substantial discounts may cause the sentencing decisions of the criminal courts to lose credibility, thus eroding the authority of the courts ­generally.67 To these concerns we may add a further five problems: (i) that the sentence discount penalises those who exercise their right to trial and who are then found guilty; (ii) that it can result in the development of a ‘guilty plea culture’ amongst defence agents and counsel; (iii) that the interaction of charge or fact bargain­ ing with sentence discounting can result in the imposition of inadequate or inappropriate sentences; (iv) that—related to point (iii)—the guilty plea can ensure that the full details of the offence are not disclosed to the sentencer, thus resulting in an ‘inherent’ discount; and (v) that the sentence discount principle is thus irreconcilable both with the practice of equity and the traditional concern of the Scottish courts with the principle of individual­ ised justice.68

65  Gemmell (n 4) at [51]. See also the earlier judgments of McHugh and Kirby JJ to the same effect in Cameron v The Queen (n 9) at [39] and [65](4) respectively. One has to ask why this aspect of Kirby J’s judgment was not considered by the Appeal Court in Du Plooy (n 5). 66  Gemmell (n 4) at [73]. 67  Gemmell (n 4) at [74]; see also Carloway, 2013: 12–13. 68  Points (iii) to (v) are, of course, allied to the problem of the credibility of sentences identified in Gemmell (n 4).

Can the Guilty Plea Discount be Justified?

 211

The Incentive on the Innocent to Plead Guilty A system of formalised sentence discounting places pressure on some innocent accused to plead guilty: the fear of an enhanced sentence on conviction after trial runs the risk of induc­ ing guilty pleas from those who are or may be innocent, and to this extent the presump­ tion of innocence is compromised (Ashworth, 2015: 186–87; Ashworth and ­Redmayne, 2010: 299; Ashworth and Zedner, 2008: 34–35; Bagaric and Brebner, 2002: 52; Baldwin and ­McConville, 1977: 59–82; Darbyshire, 2000: 902–04; Henham, 2001: 1; L ­ everick, 2014: 340–41 and 348, 2013: 260, 2010: 151 and 2008a: 45; McConville, 2002: 369–71, 1998: 565; Mulcahy, 1994: 413; O ­ dgers and Yeo, 2005: 70; Roberts and Bradford, 2015: 187; Stanbury, 2016: 7; Tonry, 2002: 97; and Willis, 1995: 65). As McConville and Bridges note, this aspect of the practice of guilty plea discounting undermines the principled nature of the system, fosters cynicism towards our institutions, and rewards hypocrisy (1993: 161; see also ­McConville and Marsh, 2016: 104 and 2014: 216; and Welsh, 2009: 661). As Lord Gill observed in Gemmell, the greater the potential discount, the greater the risk of an innocent accused (or at least an accused with a stateable defence) deciding to play safe and plead guilty for the sake of the discount.69 This issue was acknowledged by one judge and two sheriffs in the present study: It’s an unwholesome possibility that someone who has a stateable defence may be advised to play the percentage game and plead guilty. It’s a thoroughly unwholesome possibility (Judge 5). A guilty plea does save the complainer from having to give evidence but the logical ­problem with that is what if the complainer’s telling lies and really ought to be subjected to cross examination? … I think [Du Plooy] was a political decision. It sits extremely ­uneasily with the fundamental ethos of our system, which is that anyone accused is ­entitled to have the case proved (Sheriff 5). What concerns me would be if a solicitor’s advice to a client that he or she could still c­ hallenge the Crown case is somehow overshadowed or unduly influenced by the Du Plooy set up … There’s a case I know of where the accused has pleaded guilty to drugs charges. I had a look at the list of Crown witnesses and they’re all people who have been through this court on numerous occasions. I suspect that if these people were ever put into the witness box to give evidence, their stories would be all over the place, and this person might very well ‘get off ’ … I just wonder if, in this particular situa­ tion, perhaps a factor that has resulted in a guilty plea has been the Du Plooy discount (Sheriff 8).

The last named respondent referred to another case in which the accused had expressed a strong desire to plead guilty because, the respondent strongly suspected, of the likelihood of his receiving a discounted sentence. After ‘tremendous difficulty’—in which the respondent had advised the accused that ‘if I were in your shoes, I’d be wanting a lawyer’—the accused agreed to seek legal advice. The respondent continued: The solicitor got him off because the Crown case was mince [a Scottish colloquialism for ‘rubbish’]! So it’s a worrying feature of Du Plooy, I have to say’ (Sheriff 8).

The incentive to plead guilty will be particularly strong in cases where the guilty plea will make the difference between a custodial and a non-custodial sentence (Mack and Roach Anleu, 1997: 135). In England and Wales, the Definitive Guideline on Reduction in ­Sentence 69 

Gemmell (n 4) at [73]; see also Mack and Roach Anleu, 1997: 135.

212 

Judicial Discretion and Guilty Plea Discount

for a Guilty Plea indicates that a guilty plea can have the effect of reducing a custodial sentence to a community sentence (SGC, 2007b: paragraph 2.3—see also Ashworth, 2015: 183–84; Roberts and Bradford, 2015: 192, fn 6; and Sentencing Council, 2016: 21).70 In such circumstances sentencers are not required to apply a further reduction on account of the plea (Ashworth, ibid; Sentencing Council, ibid; and SGC, ibid). There are no comparable guidelines in Scotland, nor has the issue been addressed by the Appeal Court (see the dis­ cussion by Brown, 2014e: 6–7). Fourteen of the 17 sheriffs in the present study, however, confirmed their practice to be that, in appropriate cases,71 the discount could be reflected in their decision to impose a non-custodial sentence. In such circumstances, as with the guid­ ance in England, sheriffs would not make a further discount in, for example, the number of hours of community service: I do it … Your discount is that you’re getting community service, essentially. I’ve never been appealed on that. I think it would be wrong in principle if you couldn’t do that … I’m very firmly of the view that if you get community service in that situation, you don’t get a fur­ ther discount. I do impose the maximum of 300 hours in that kind of case (Sheriff 9). Yes, no question. I think that has to be implicit in the statutory provision. If you can give a discount on the sentence, it must involve discounting from one type of sentence to another. But that’s what the discount should be (Sheriff 3). It’s a relatively regular occurrence. The agents know that, so they’ll say: ‘The discount could per­ haps be that you’ll look away from custody and look to the maximum end of community service’. That’s a legitimate use of Du Plooy and sometimes entirely appropriate … But if you discount in this way you’d get the maximum, or close to the maximum, number of hours of community service (Sheriff 1).

Whilst this practice may very well be an effective and pragmatic use of shrieval discretion and is, as Ashworth notes, a logical application of the discount, it results in a ‘crucial and momentous’ decision for the accused who is advised by his solicitor or counsel that his case is on the cusp of custody (Ashworth, ibid 184). It is likely to increase the pressure on accused persons in cases on the borderline between custodial and community sentences to plead guilty (Ashworth, ibid; Mack and Roach Anleu, ibid; see also the discussion by Chasse, 2009: 62–64). Thus guilty plea discounting is, potentially, a practice that ‘betrays the truly innocent on the altar of supposed practicality’ (Lynch, 1994: 133). As Alschuler notes, ‘[t]o turn a substantial portion of a defendant’s punishment on a single tactical decision is … to assign to the defendant a responsibility that he cannot fairly be required to bear’ (Alschuler, 1981: 668, cited in Lynch, ibid 133; see also Ashworth and Zedner, 2008: 47, and JUSTICE, 1993: 9).

Public Confidence in the Criminal Justice System and the Credibility of Sentences When punishment is varied substantially for administrative reasons of cost and efficiency, confidence in the criminal justice system can be lost, with the process coming to be seen 70  The same approach is followed in Australia (Mackenzie, 2006b: 4). Freiberg explains of Victorian sentencing practice that whilst the guilty plea may change a custodial sentence to a non-custodial one, the resulting sentence should not again be reduced by reason of the discount (Freiberg, 2014: 383; see also R v Lo [2003] NSWCCA 313). 71  Namely ‘borderline’ or ‘marginal’ cases, on the cusp between a custodial and non-custodial ­sentence (for which see Laws, 2004: 64 and Tombs, 2004: 47–51).

Can the Guilty Plea Discount be Justified?

 213

simply as a bargain unrelated to the accused or his conduct (Mack and Roach Anleu, 1997: 132–33; see also Leverick, 2014: 341 and 348). In Australia the problem was identified by McHugh J in the course of his judgment in R v Everett: [I]nadequate sentences … give rise to a sense of injustice, not only in those who are victims of the crimes in question but also the general public. Inadequate sentences are also likely to undermine public confidence in the ability of the courts to play their part in deterring the commission of crimes.72

In research undertaken on behalf of the English Sentencing Council into public attitudes to the sentencing of sexual offences, Nicholls et al report that there was little support for the practice of discounting either from victims or from the public (Nicholls et al, 2012: 55). The issue had previously been explored in more depth by Clarke et al (2002) in the course of research undertaken on behalf of the ­English Sentencing Advisory Panel on pub­ lic attitudes to so-called ‘date rape’ and to relationship rape. Clarke et al conducted their research by holding 28 discussion groups and through interviews with 62 members of the public (including 25 victims of rape) (ibid 14–18). The study demonstrated the extent of public scepticism about the justification for the sentence discount principle. It was reported that the idea of substantial mitigation on the grounds of a guilty plea in cases of rape was regarded as ‘totally unacceptable’ by respondents (ibid 9), who strongly condemned the practice (ibid 34). Although respondents accepted that a guilty plea obviated the need for the complainant to give evidence in court, it was felt that any reduction in sentence meant that the offender was not getting the full punishment he deserved (ibid). Guilty plea discounts were seen as counter-productive, serving only to create opportunities for the offender to manipulate the judicial system to his own advantage (ibid 55). Clarke et al reported that the problem with the policy of sentence discounting was seen in three different ways by respondents: firstly, a lack of equivalence if the offender received a reduced sentence for pleading guilty in relation to the severity of the crime and the effect on the victim; secondly, a perception that the principle is concerned with saving court time and costs, and prison costs, rather than sparing the victim from an ordeal; and thirdly, a perception that an accused would use the sentence discount principle as an opportunity to play the system (ibid 55–56). Some respondents in the study by Clarke et al suggested that the amount by which a sentence is reduced for an early guilty plea should be a lesser percentage than a third (ibid 56). Research on public attitudes towards reductions in sentence for guilty pleas was also commissioned by the Sentencing Council (Dawes et al, 2011). The study comprised a survey administered to 987 members of the public, 35 in-depth interviews with victims and witnesses, and 15 in-depth interviews with convicted offenders (ibid 8–11). The study found that respondents’ initial reactions towards the principle of discounting were largely negative, with just one in five approving of the practice (ibid 16). Whilst the most com­ mon view was that pleading guilty should result in an offender receiving a more lenient sentence in some cases (45 per cent of respondents), 29 per cent of respondents considered that pleading guilty should never result in a more lenient sentence. Thus, whilst there is limited public support for guilty plea discounts, there is clear discomfort at the universal ­application of such discounts (ibid). 72 

R v Everett (1994) 124 ALR 529 at 537.

214 

Judicial Discretion and Guilty Plea Discount

These views were found by Dawes et al to shift slightly as the principle of ­discounting was examined in more detail with respondents in discussion groups. Whilst some respondents remained firmly opposed to the practice of sentence discounting throughout the discus­ sions, as deliberation progressed others d ­ emonstrated some recognition of possible justifi­ cations for, and the benefits of, the practice (ibid 23). With regard to the level of discount, participants in the discussion groups considered that discounts in the order of 10 per cent should be allowed for early guilty pleas, with the current practice of awarding a third being seen as ‘ridiculous’; however, victims and witnesses were reported as being more amenable to the prospect of discounts larger than a tenth, especially if they were to be given in con­ junction with a wider focus upon rehabilitation (ibid 26). The Sentencing Council’s study also found that whilst there was not universal sup­ port for guilty plea discounts amongst victims and witnesses, perceptions of the practice amongst these groups were, on the whole, more positive than the general public (ibid 18). Whether or not victims and witnesses felt a sentence discount was acceptable in their own case tended to be linked to their attitudes towards the prospect of having to attend court to give evidence (ibid 19). Whilst most respondents were not in favour of allowing discounts in sentence to repeat offenders, first offenders and those who had assisted the authorities in bringing others to justice were seen as more deserving of discounts (ibid 30–31). There was, however, ‘a strong level of agreement’ amongst most respondents that no discount in ­sentence should ever be allowed for offences which caused significant physical or mental harm, such as murder and sexual offences. In such cases, respondents considered that the seriousness of the offence overrode any possible benefits that may follow from the tender­ ing of a guilty plea (ibid 29–31). Recent research in England and Wales suggests that similar views may in fact be held by sentencers. In their study of guilty plea discounting in England and Wales in which they analysed data from the Crown Court Sentencing Survey,73 Roberts and Bradford report that although courts appeared to generally comply with the guidelines, there was also a sig­ nificant degree of departure (Roberts and Bradford, 2015: 203). In particular, the authors’ findings suggest that courts have evolved different tariffs of reduction for different offences and that the presence of aggravating factors tends to suppress the level of reduction, even though the relevant guideline (SGC, 2007b) states that these matters should be considered independently (Roberts and Bradford, ibid 201–02). The authors conclude that this ‘may reflect courts’ recognition that the more serious categories of crimes should attract a more modest sentence reduction, for reasons related to community reaction or some other rea­ son’ (ibid 208). Eight respondents in the present study (two judges and six sheriffs) considered the ­practice of allowing discounts in sentence in return for guilty pleas to have had a deleteri­ ous effect on the credibility of sentences. One judge and five sheriffs considered the very practice itself to be objectionable: What does [the practice] say about the criminal justice system? Because of the sentencing discount, the system is losing credibility in the eyes of the public (Judge 5).

73  For a discussion of the Crown Court Sentencing Survey, see Roberts, 2013b and Roberts and ­Ashworth, 2016: 342–43. In 2015 the Sentencing Council made the decision to terminate the ­survey, taking the view that its statu­ tory duty to monitor the effect of sentencing guidelines could be more efficiently achieved ‘by time-limited and offense-specific data collection’ (Roberts and Ashworth, ibid 343).

Can the Guilty Plea Discount be Justified?

 215

I disapprove fundamentally of Du Plooy. Always have done (Sheriff 5). I don’t think [the discount principle] particularly helps to improve the perception of the sentenc­ ing process in the eyes of the man on the Clapham omnibus, if I can put it that way (Sheriff 14).

One judge and one sheriff directed their criticism towards the levels of discount allowed by the Appeal Court, particularly in cases where the accused had no choice but to plead guilty: I’m just unhappy, as I think a lot of my colleagues are, with Du Plooy and its derivatives. I’m mainly unhappy because I think that, in a kind of rather casual and cavalier way, we’re doing the public down for the benefit of people who don’t deserve it … I feel that the whole thing is flawed because the public interest is being put right down in the scale of importance … I mean, it’s utterly absurd. It just demeans the process (Judge 3). My worry about the level of discount being a third, particularly where the crime is so serious, is that it loses credibility amongst some of the public. That’s certainly my impression from speaking to people—they don’t like this at all. I think it’s the top level of discount that’s doing that damage. I’m not saying we shouldn’t give a discount … But I just think a third is frankly too high (Sheriff 16).

Two sheriffs expressed particular concern about the discount principle affecting the ­credibility of fines imposed in the summary courts: With fines, it’s quite difficult sometimes to find a way that doesn’t sound like some bargain ­basement figure—where ‘discount’ sounds like a discount. Three hundred pounds but to you, two hundred. I mean, really! Find a form of words which is satisfactory somewhat! It’s like an ­auction or something! (Sheriff 10). It’s a piece of absolute nonsense to be discounting a fine of £400 to £350 because he pled guilty at the intermediate diet when you’re calculating sentences on the basis of the fact that he only gets £96 per fortnight in Jobseeker’s Allowance. When you have a big court—I had 50-odd in front of me this morning—to have to say, ‘the fine would have been £300 but it will be discounted to £200 because you pled guilty at the earliest opportunity’, well, I find it to be quite embarrassing, frankly. Du Plooy was a High Court decision and, with the greatest respect, they don’t deal with 50 cases at summary level (Sheriff 5).

The Penalising of Accused who Exercise their Right to Trial The principle of guilty plea discounting is also criticised as ‘undeniably’ punishing those who exercise their right to trial and who are then found guilty (Darbyshire, 2000: 901; see also Ashworth, 2015: 187, 1983: 312; Bagaric and Brebner, 2002: 58; Henham, 2002: 393, 1995b: 241; JUSTICE, 1993: 20; Leverick, 2014: 340, 2013: 260, 2010: 151 and 2008a: 45; Mack and Roach Anleu, 1997: 141; McConville, 2002: 368–69; Mulcahy, 1994: 413; Pincus, 1987: 477; and Willis, 1995: 65, 1985: 139). On this view, the discount principle provides a penalty for asserting the basic human right to demand a trial (Henham, 2001: 1; Tonry, 2002: 97). Whilst Sheriff Principal Nicholson QC observes that it is unacceptable that a person found guilty after trial is punished more severely simply because he did not plead guilty (Nicholson, 1981: 219; see also Freiberg, 2014: 381; McConville and Marsh, 2014: 80; and O’Malley, 2016: 163, 2006: 123),74 respondents in the present study did not cite this as 74  See further the dissenting remarks of Cox J in R v Shannon (1979) 21 SASR 442 at 458–59; the remarks of Kirby J (partially dissenting) in Cameron v The Queen (n 9) at [65](3); the dissenting remarks of McHugh J in Cameron (n 9) at [41], and the majority judgment in Cameron (Gaudron, Gummow and Callinan JJ) (n 9) at [12]–[14].

216 

Judicial Discretion and Guilty Plea Discount

a problem with the practice of guilty plea discounting. Although, as we have seen, many respondents were strongly opposed to the practice of discounting, their concerns focused on issues such as the discount acting as an incentive for accused with a stateable defence to plead guilty and, particularly, the effect of the discount principle on the credibility of sentences.

The Development of ‘a Guilty Plea Culture’ amongst Defence Counsel McConville (1998) suggests that the sentence discount principle contributes towards what he terms ‘a guilty plea culture’ amongst lawyers (see also Chasse, 2009: 57; and McConville and Marsh, 2016: 111, 2014: Chapter 8). Within a ­system of limited resources, a case may no longer reflect the quality of justice but rather each participant’s priorities (Seifman and Freiberg, 2001: 65). Thus, for M ­ cConville, the defence will commonly follow the prosecu­ tion in adopting a strategy of case disposal through the guilty plea: Defence lawyers approach their work on the basis of standardized case theories and stereotypes of the kind of people who become involved in criminal events; images of clients as feckless or dishon­ est are allowed to structure the way their cases are handled from the outset; the views of clients are given little weight and their accounts not investigated; and the case proceeds on the basis that the lawyer knows best in a context in which all incentives point towards a guilty plea (McConville, ibid 572; see also JUSTICE, 1993: 3; McConville and Marsh, 2016: 107–08, 112 and 114; Mulcahy, 1994: 423; and Rauxloh, 2012: 50–51).

The existence of a ‘guilty plea culture’ amongst certain defence lawyers in Scotland is well illustrated by the decisions in Gallagher v HMA75 and Blockley v Cameron.76 In both these cases the appellants’ convictions were quashed due to the pressure to plead guilty, exerted by their respective defence agents in circumstances where, in fact, both appellants had state­ able defences. Lord Gill’s observations in Gemmell that defence counsel considered an accused, by rea­ son of an early plea, to be ‘entitled’ to a discount77 and that the decision in Spence had cre­ ated ‘a climate of expectation’ amongst counsel78—observations confirmed by the present study—suggest that the formal system of discounting introduced by Du Plooy and Spence may have resulted in McConville’s ‘guilty plea culture’ having taken root amongst certain members of the criminal Bar in ­Scotland. The extent to which defence lawyers in Scotland have become immersed in the guilty plea culture must, however, be a matter of conjec­ ture—there will of course be exceptions (Ashworth and Redmayne, 2010: 297)—and would benefit from further research.79

‘Double Dipping’—the Interaction of the Discount with Charge/Fact Bargaining Guilty plea discounting in Scotland sits alongside a system of informal charge and/or fact bargaining (Leverick, 2013: 259, 2010: 145, 2004: 360). Charge bargaining is the practice 75 

2010 JC 240 at [10]–[16]. 2013 SCCR 181 at [29]–[30]. 77  Gemmell (n 4) at [30]. 78  Gemmell (n 4) at [31]. 79  Previous studies have focused on practice in England. The sub-title of the study by McConville et al, 1994— ‘The Organisation and Practices of Criminal Defence Lawyers in Britain’ (emphasis added)—is an egregious misnomer. 76 

Can the Guilty Plea Discount be Justified?

 217

whereby the prosecutor accepts a plea of guilty in exchange for the reduction or deletion of a charge on the indictment or complaint (Ashworth and Redmayne, 2010: 296–97; Freiberg, 2014: 124; Henham, 2001: 98; JUSTICE, 1993: 2; Leverick, 2004: 360; McConville et al, 1994: 194; McPherson, 2013: 145; Mulcahy, 1994: 419; Seifman and Freiberg, 2001: 69, 70; and Torre and Wraith, 2013: 193). Fact bargaining, meanwhile, refers to the dele­ tion or amendment of the narrative contained in the charge (Ashworth and Redmayne, ibid 300–01; Henham, ibid; JUSTICE, ibid; Leverick, ibid; and McPherson, ibid), a practice which is dependent on the negotiating skills of the accused’s legal representative (Leverick, 2014: 346). As Seifman and Freiberg observe of the practice as it operates in Australia, the object of the exercise is to conceal certain aspects of the accused’s conduct from the sentencer, restricting him to meting out a sentence based on a limited set of agreed facts: it is a prac­ tice designed ‘to exclude some of the more inflammatory aspects of a case’ (Seifman and Freiberg, ibid 71) or ‘eradicate aggravating features’ of the offence (McPherson, ibid). In both charge bargaining and fact bargaining, the hope of the accused is that the changes to the charges will be reflected in a lesser sentence (Leverick, 2013: 259, 2010: 145, 2004: 360). This practice has been criticised by Australian commentators as it weakens the concept of the adversarial system that allows the sentencer to base the sentence on an accurate under­ standing of the facts behind the charges (Fox and Freiberg, 1999: 76, Gerber, 2003; see also Brook et al, 2016: 1184). Charge and fact bargaining can result in the submission of an agreed narrative of the facts by the prosecution and the defence that does not prop­ erly reflect the true gravity of the criminal conduct (Fox and Freiberg, ibid; King, 2005: 295–97). Gerber cites the example of a notorious and high profile case of gang rape in Sydney in which a ‘collusive bargain’ was struck between the Crown and the defence: in return for the three accused youths tendering pleas of guilty, the Crown agreed ‘to with­ hold some aggravating facts from the court’, resulting in lighter sentences (Gerber, ibid 212). The actions of the Crown and the defence in this case are deprecated by Gerber: he considers the actions of the defence in ‘horse trading’ a plea of guilty by ­obtaining the ­prosecution’s agreement to withhold aggravating facts from the court to be ‘nothing less than a corruption of the legal process’. For Gerber, it ‘constitutes an attempt to pervert the course of justice, denying the court knowledge of the true circumstances in which the plea was made’ (ibid 212, 214). Gerber concludes: [I]n sentencing an accused who pleads ‘guilty’ to an offence, the trial judge must be informed of all the facts surrounding the offence, both those which are mitigating and those which render the offence more serious, not a sanitised version in order to reduce the appropriate punishment (ibid 214).

This is also a particular problem in Scotland where the process of charge and fact bar­ gaining is entirely dependent on informal negotiations between the prosecution and the defence. The practice has no statutory basis and the judge or sheriff has no involvement (Leverick, 2010: 145–46; McPherson, ibid 144 and 147; Moody and Tombs, 1982: 106–27; Stephen et al, 2008: 214; cf Ashworth and Redmayne, 2010: 300). An accused may thus receive a double benefit. By pleading guilty to what may be a greatly reduced charge, the accused ensures that a lesser sentence will be imposed for, as the Appeal Court noted in HMA v Cooperwhite,80 a sentence must be ­justified by the conviction 80 

2013 SCCR 461 at [15].

218 

Judicial Discretion and Guilty Plea Discount

which, in turn, depends upon the extent of the libel (see also O’Malley, 2006: 568–73).81 The guilty plea itself may then attract a f­ urther, formal Du Plooy discount under section 196 of the 1995 Act (see the o ­ pinion of Lord Hamilton, as he then was, in RB v HMA).82 This is a ­practice referred to as ‘double dipping’ by Australian commentators (Mack and Roach Anleu, 1997: 128; Willis, 1995: 54; 66). Judges in the present study were alive to the problems posed by the interplay of charge bargaining and guilty plea discounts.83 That sentencers in the High Court are frequently being called upon to impose discounted sentences on charges which have already been drastically reduced by the Crown was a matter of real concern amongst judges. As Judge 5 explained: Now, this [the sentence discount procedure] all assumes that the accused pleads guilty to the charge or charges as libelled. It takes no account of the charge bargaining that goes on with the Crown (Judge 5).84

Judges 3 and 5 illustrated the problem by reference to the Crown accepting reduced pleas in sexual offences, murder, and drugs offences. Judge 5 noted that, where the accused is charged with rape, the Crown may well accept a plea of guilty to indecent assault or attempted rape. Judge 3 stated that he had recently been due to preside over a murder trial, the preparation for which had involved his reading over three and a half thousand pages of material including transcripts of police interviews, witness statements and medical reports. The respondent explained that in the week before the trial was due to start, the Crown had accepted a plea of guilty to the lesser crime of culpable homicide. The respondent was criti­ cal of the Crown’s decision to accept the reduced plea: Now, having read all that stuff and having seen the photographs, the transcripts and the post mortem [report], having seen the hospital records, you’ve got a fair idea. And then you see the Crown’s narrative and you know that it doesn’t contain all kinds of stuff which had been admit­ ted in a police interview but you can’t take any of that into account—you’ve got to sentence on the basis of the narrative. So there’s quite often a big difference. A lot of my colleagues feel the same. There were major incentives already to plead guilty for other reasons, not the discount (Judge 3).

Both Judge 3 and Judge 5 further illustrated the problem of the Crown accepting reduced pleas that did not fully reflect the accused’s culpability—which may then attract a further

81  See, for example, Thomson v HMA High Court of Justiciary on appeal, 16 October 2003, ­unreported in which the appellant was charged in the sheriff court with assault with intent to rape. His plea of guilty to indecent assault under deletion both of the allegation of intent to rape and of the allegation that he struck the complainer on the head and body was accepted by the Crown. The sheriff nevertheless considered that the offence to which the appellant had pleaded guilty was ‘not so very far removed from the crime of attempted rape’ and imposed an extended sentence of two and a half years, comprising a custodial term of 18 months and an extension period of 12 months. On appeal, it was held that the sheriff had misdirected himself and that the view he took of the case was not one which was justified by the appellant’s plea. The extended sentence was quashed and a probation order of two years was substituted. 82  RB v HMA (n 22) at [12], Lord Abernethy concurring. 83  None of the 17 sheriffs interviewed for the present study discussed this aspect of the practice of sentence discounting. This may be due to the more pronounced effect of charge bargaining in the High Court. 84  In this regard, see also the dissenting opinion of Lord Philip in RB v HMA (n 22) at [27].

Can the Guilty Plea Discount be Justified?

 219

Du Plooy discount—by reference to charges under section 4(3)(b) of the Misuse of Drugs Act 1971 (being concerned in the supply of drugs): They’ve been charged with being concerned in the supply of drugs over a period of weeks and the Crown accept a plea to being concerned in the supply on a single day, despite the police raiding their house and findings scales, cling film, tick lists, bundles of cash and so on. They’ve been at it for weeks, if not months. But you’ve got to sentence on the libel. It’s libelled as supply on a single day. That’s the basis on which you’ve got to sentence them. So they’ve already received a discount in the sentence by virtue of pleading to a much reduced charge. Then they’re getting a Du Plooy discount on top of that. It’s like a double discount. This has no place in the criminal justice system. It’s scandalous. Giving a third off the sentence is just ridiculous (Judge 5). The offence to which people plead guilty is sometimes amended down in the plea ­process to an extent that takes out almost all the substance of what the person’s done. Very often you get people who are charged with drugs offences over a period of three months and obviously they’re dealing, they’ve been at it—the house is full of stuff, there’s dust everywhere, there’s scales, people are com­ ing to the door, there’s telephone lists. They’ve been at it for ages but they plead guilty to having done it on a particular day only. [The plea] bears no relation to what they’ve actually done but this is part of the m ­ itigation. The defence say it was on a single day only. So, ‘they’ve never done it before’ and it’s just most unfortunate that the police just happened to come to the house on the only day that they’ve ever done it … Come on! It’s rubbish! (Judge 3).

Similar findings were made by Leverick and her colleagues in the course of interviews con­ ducted with the judiciary as part of a project to evaluate the impact of reforms to High Court procedure in Scotland (see Chalmers et al, 2007). Although not forming part of the final report, Leverick reports that at least two judges questioned the appropriateness of the practice of sentence discounting (Leverick, 2010: 153). As in the present study, judicial concern focused on the fact that the extent of the reduction in sentence as a result of the interaction between formal sentence discounting and charge and fact bargaining can be enormous: not only can an accused benefit from a discount for his guilty plea, but he may also have pled guilty to a reduced charge as part of a charge bargain thereby ensuring, as one judge put it, that ‘the gory details’ will not emerge, that any particularly aggravating features of the offence will go unheard, and that the pre-discount headline sentence will thus be lower (ibid 153–54; see also McConville, 2002: 368; McConville et al, 1994: 261–62; and Mulcahy, 1994: 420). Charge or fact bargaining may thus combine with the practice of sentence discounting to result in a sentence that reflects the outcome of negotiations rather than the harm and culpability of the offender’s criminal conduct (Ashworth, 1998c: 292; Moody and Tombs, ibid).

The ‘Inherent’ or ‘Unseen’ Sentence Discount Mack and Roach Anleu distinguish between what they term ‘inherent’ and ­‘formal’ sen­ tence discounts (Mack and Roach Anleu, 1997: 130). If an accused pleads not guilty and is convicted after trial, then the court will have heard significantly more detail about the offence, with the sentencer perhaps being more inclined to take a harsher view of the seri­ ousness of the accused’s criminal conduct. For this reason, sentences imposed after a guilty plea may be inherently less than those imposed after conviction at trial. This ‘implicitly discounted’ sentence is, as Mack and Roach Anleu explain, further reduced by the explicit,

220 

Judicial Discretion and Guilty Plea Discount

formal ­discount. The cumulative effect of the inherent and formal discounts can create a substantial difference between the likely sentence after conviction at trial and the poten­ tially far lower sentence handed down after a guilty plea (ibid). This issue was acknowledged as a particular problem by the respondent Judge 3— a very experienced judge—who considered that the agreed narrative85 read to the court by the prosecutor very often did not give ‘anything like the flavour, the full horror, of what they’ve actually done’. Accused persons were, the respondent noted, well aware of this— particularly in the sheriff court where the preparation of an agreed narrative is usually done ‘in a great tearing hurry’. In the view of Judge 3, this was an important reason why many accused persons plead guilty: So what they’re getting away with is a fairly bland narrative, instead of the horror of several days of evidence of what they’ve actually done … Generally speaking, what people will expect (and I think it does happen), is that the Crown’s narrative doesn’t give anything like the full horror of what they’ve actually done … They know that they’ll get away with the Crown’s narrative, rather than the evidence; the court perhaps won’t see the evidence, won’t see the CCTV footage of what they’ve done; the court won’t hear the evidence of eyewitnesses … You’re sentencing in the dark to some extent when people plead guilty to things … That’s a major benefit as well, so they’re gaining all kinds of things by pleading guilty … It just goes against the grain that where they’ve gained a number of benefits you’re having to give them something else [viz the formal Du Plooy discount] … I don’t personally see why we should be giving them significant discounts at all (Judge 3).

These comments accord with the findings of Roach Anleu and Mack’s study of plea nego­ tiation in which defence lawyers, prosecutors and members of the judiciary from across Australia were interviewed (Roach Anleu and Mack, 2001: 159–60). The study found that accused persons were not disadvantaged by pleading guilty to an agreed set of facts. As two legal aid solicitors observed: I’ve never negotiated my client out of a position. I’ve always negotiated them into one. And so, from my point of view, my client’s always going to be better off if he gets his view accepted without having a trial. [T]here’s probably a lot of … cases where defendants end up with a much, much better deal than they’d have got going to trial. Particularly, where you [the defence] negotiated facts down; I mean, you can negotiate a lot of aggravating facts out of it” (Roach Anleu and Mack, ibid 167).

That the practice of guilty plea discounting takes insufficient recognition of the preparation of agreed narratives was stressed by another judge in the present study: It ignores the fact that the plea is a negotiated plea and therefore the plea may well be less than the Crown think they could achieve if they went to trial … Every case now has an agreed narrative on a plea; there’s never an issue or dispute between one side and the other about what actually happened requiring to be resolved in some way … There is an advantage, generally speaking, for an accused person in obtaining an agreed narrative because they have a certain degree of control over the facts that are put to the court (Judge 2).

85  On the use of agreed narratives in Scottish criminal procedure generally, see Chalmers et al, 2007: 62–63 and McPherson, 2013: 141–47. In Sutherland v HMA 2016 SCCR 41 at [2], the Appeal Court remarked on the poor quality of many agreed narratives, noting that they are often ‘bedevilled by irrelevant evidential material’. The Court considered that the agreed narrative in Sutherland had, no doubt for reasons of expediency, ‘the air of being cut and pasted from the Crown Office precognition report’ (ibid).

Can the Guilty Plea Discount be Justified?

 221

The problem was even acknowledged by one of the three shrieval respondents who ­positively approved of the practice of guilty plea discounting on the grounds of efficiency: The whole plea process can be reduced to bland statements. All of these things militate towards a certain sentence, whereas if there’s a very long and protracted trial and really no defence is forth­ coming and the accused shows his or her true colours, then obviously it’s more likely there will be a condign sentence at the end of that procedure (Sheriff 12).

The last named respondent continued: Often you get an inadequate version of the events, very little is said in mitigation because the agent realises that a bowdlerised version of events has been given and so actually there’s an unseen dis­ count … where you’re dealing with a kind of abstract offence, a f­aceless offence, compared with where you actually see the victim face to face and they may give you valuable information about the sequelae of the offence (Sheriff 12).

The combination of charge and/or fact bargaining and the discount inherent in pleading guilty with the formal, Du Plooy discount under section 196 of the 1995 Act arguably pro­ duces ‘double’ or even ‘triple’ discounts. Where, for example, an accused charged with rape pleads guilty to sexual assault86 he is likely to receive the first discount in sentence by virtue of the reduced libel (sexual assault almost invariably being sentenced less severely than rape); a second discount by virtue of the fact that no evidence will be given by the com­ plainer or led in court from medical witnesses concerning the physical, emotional and psy­ chological trauma87 caused by the attack (thus ensuring that the ‘full horror’ of the offence is not ­disclosed to the court);88 and a third, formal discount in terms of section 196 of the 1995 Act. As Schulhofer suggests of plea bargaining generally, ‘double’ or ‘triple’ discounting is a practice that ‘compromises our aspirations to justice at the same time that it undermines the effective punishment of serious offenders’ (Schulhofer, 1994: 135). Whilst, as we have seen, the Scottish courts seek to impose individualised sentences through what the Cana­ dian courts have termed ‘a wise blending of penal aims’,89 the inherent sentence discount and the resultant possibility of double or triple discounts in sentence for offenders who plead guilty introduces what some may consider to be a distinctly unpalatable ingredient into this blend. The result is often what is arguably a disproportionately lenient sentence: one which reflects neither the offender’s culpability nor the harm caused and which is thus irreconcilable with the demands of individualised justice.

Discounting and Individualised Justice In his opinion in Gemmell, Lord Gill considered the most fundamental problem with regard to public confidence in the criminal justice system to be the ­possible perception of injustice 86  Either at common law (in which case the lesser offence is indecent assault) or under sections 1 (rape) and 3 (sexual assault) of the Sexual Offences (Scotland) Act 2009. 87  On the possible effects of such trauma, see Jina and Thomas, 2013: 18; Mason and Murphy, 2016; and ­Payne-James et al, 2011: 132–33. 88  See the comments of the respondent Judge 3, mentioned earlier. Whilst the complainer would be entitled to submit a victim statement under section 14 of the Criminal Justice (Scotland) Act 2003, one respondent in the present study (Judge 1) noted that such victim statements are not often tendered by the Crown. See, for example, HMA v AB 2016 SCCR 47 at [14]. 89  R v Willaert (1953) 105 CCC 172 at 176.

222 

Judicial Discretion and Guilty Plea Discount

arising from the operation of the discount principle, ­particularly in cases where severe sen­ tences are deserved.90 If in such a case there are two accused, then the accused who pleads guilty at the earliest opportunity may receive a sentence that is less by a matter of years than that imposed on the accused who is found guilty after trial.91 As Darbyshire notes in a pas­ sage cited by Lord Gill,92 if a person is genuinely guilty, why should they be rewarded for ­pleading guilty (Darbyshire, 2000: 901; see also Vogel, 2007: 5)? The present study has demonstrated the importance of individualised justice in Scottish sentencing. A major feature of sentencing in this and in other Commonwealth jurisdictions such as Canada, South Africa and Australia, as well as in the Republic of Ireland, is the con­ sistent emphasis on judicial discretion in order to tailor the sentence both to the facts of the particular case and to the circumstances of the individual offender.93 There is, however, a tension between judicial discretion as a means of individualising sentences and the sentence discount as an administrative tool which prioritises consistency and predictability (Mack and Roach Anleu, 1997: 133; Willis, 1995: 55, 71). Willis notes that the sentence discount involves the use of a vital component of the crimi­ nal justice system—sentencing—for administrative purposes (Willis, ibid 70). Instead of the traditional conception of the sentencer as a phronimos, using his or her experience to craft a disposal appropriate for the particular offender convicted of the particular offence through a process we have identified as that of the phronetic synthesis,94 the sentence dis­ count principle turns the sentencer into ‘an administrator seeking to expedite case-flow’ (Willis, ibid; see also Seifman, 1982: 72). Counsel for the appellant in Spence v HMA,95 for example, reports that during debate, the presiding judge asked the Advocate Depute whether he could ‘estimate the cost’ of the trial (Carroll, 2007). Not only does a regime of guilty plea discounting detract from ­sentencing ­flexibility, it also appears to give administrative goals unwarranted i­mportance (Willis, ibid 71). It is a blunt, not a precision, instrument, introduced as a ­bureaucratic rather than a principled response to the needs of criminal justice (McConville, 1998: 585). Rigid quantification of sentence (as in the sliding scale approach to discounts in Spence) conflicts with what we have already identified as the ultimate goal of individualised sentencing, in which judicial discretion is paramount (Mack and Roach Anleu, ibid 134).96 At the heart of the tension between individualised justice and the practice of sentence discounting is the fact there is no obvious principled basis for punishing offenders who plead guilty less severely than those who elect to proceed to trial (Bagaric and Brebner, 2002: 51; Edney and Bagaric, 2007: 203). The crux of the problem was identified more than 30 years ago by Willis: ‘[T]he final product after allowing for the guilty plea is not the appro­ priate sentence according to traditional penological criteria, but an alloy of uncertain value’ (Willis, 1985: 143; see also Darbyshire, 2000: 902 and Willis, 1995: 65) . Whilst the actual sentence imposed on an offender should be based on his conduct, his personal characteristics, the harm caused, and the wider interests of s­ ociety, a discount for 90 

Gemmell (n 4) at [76]. Gemmell (n 4) at [76]. Gemmell (n 4) at [76]. 93  See Chapters 3 to 6. 94  See Chapter 5. 95  Spence (n 11). 96  See Chapters 4 to 6. 91  92 

Can the Guilty Plea Discount be Justified?

 223

a guilty plea on the basis that it saves time and expense conflicts with these principles. The sentencing outcome is determined not by the sentencing principles or the sentencing canon developed by the Appeal Court through its exercise of principled discretion,97 rather it is determined by administrative considerations (Mack and Roach Anleu, 1997: 132; Roberts, 2011c: 8; and Willis, 1995: 65). Such administrative considerations are unrelated to either the offence or the offender (Mack and Roach Anleu, ibid 141),98 yet the individual, contex­ tualised offender and the offence which he has committed are the key considerations if one is to sentence by way of phronetic synthesis. This issue was identified by nine respondents (three judges and six sheriffs) in the present study. As one experienced judge put it in the context of discussing the ‘utilitarian value’ of a guilty plea: Stripped of the euphemisms, what it means is that the court enters into a pact with criminals: if the accused is obliging enough to admit his guilt, the court will give him a sentence that is less than he deserves and, therefore, a sentence that is less than justice demands. There’s nothing noble about this. It’s just a seedy little bargain that we enter into with criminals … It’s just not an ethically justifiable stance … Utilitarianism is a theory based on the greatest good for the greatest number of people. There is nothing utilitarian about giving people less than they deserve to help clear the backlog of cases in the system and to make savings in the justice budget (Judge 5).

Whilst another two judges and five sheriffs expressed concern that the discount principle resulted in disproportionately lenient sentences, the problem was c­ onsidered in some depth by the respondent Sheriff 10: There is something which makes me uncomfortable with the idea that a sentence should be reduced because it is administratively convenient to do so. If an offence has been c­ ommitted and if the approach to discounting requires one to start with what is an appropriate sentence, if that is correct then one has decided what the sentence ought to be. One is then reducing the sentence because of administrative reasons and one has to ask the question, is that just? Does that accord with a proper concept of justice? Because, ex hypothesi, one has started with the correct sentence … Administrative convenience does not, and on one view should not feature. Should I reduce what would otherwise be considered a just sentence for administrative convenience? I confess to feeling a degree of unease about that (Sheriff 10).

The present study found there to be a strong degree of judicial resistance to the practice of guilty plea discounting. The study demonstrates that, as noted by ­Australian commentators, systems of quantified discounts may be seen to conflict with the instinctive synthesis—or the phronetic synthesis—approach to s­ entencing (see Freiberg, 2014: 233–34 and 385; and Odgers and Yeo, 2005: 69; cf Markarian v The Queen).99 The traditional Scottish focus on the individualised sentencing of contextualised offend­ ers is such that seven respondents either expressly stated that they paid only lip service to 97 

See Chapter 6. Hence McHugh J’s observation in Markarian v The Queen [2005] HCA 25 at [74] that the discount is ‘offered as an incentive for specific outcomes in the administration of criminal justice and is not related to sentencing purposes’. See also the opinion of Lord Gill in Gemmell (n 4) at [59] and Roberts et al, 2011: 528 who note that the sentence discount is unrelated to the purposes and principles of sentencing. This has long been recognised by the Australian courts: see the dissenting judgment of Cox J in R v Shannon (n 74) at 457, the judgment of Asche CJ in R v Jabaltjari (1989) 64 NTR 1 at 15 and the decision of the New South Wales Court of Appeal in R v Thomson (n 8) at [115]. 99  Markarian v The Queen, ibid at [24]. 98 

224 

Judicial Discretion and Guilty Plea Discount

the provisions of section 196 of the 1995 Act or made comments which strongly suggested that this was the case. In particular, one respondent very candidly admitted that in certain cases he increased his starting point (or ‘headline sentence’ as described in Gemmell) to ensure that the accused who pleaded guilty still received what he considered an appropriate sentence: Judges do look at how long the accused will serve. How long will he be inside? If, as a result of the Du Plooy discount, the end result is too short then of course you’ll reassess the starting point. You’d have to be a saint not to reassess your sentence in that situation (Judge 4).

Three other respondents, whilst not following this practice, were aware that it occurred: Because of Du Plooy I think some people are thinking, well, what sentence would I be giving and then grossing it up (Judge 8). There are problems with [the practice of discounting] in the sense that you’ve got to watch that people are not actually selecting a sentence which is really much higher than they would impose and they’re then going through an exercise, so they’re effectively grossing up what they think is appropriate. There are cases where that’s clearly what is being done (Judge 7). One has to be frankly dishonest in picking a starting point that will allow you to get to the point which you want to get to. Now, one shouldn’t be in that position as a sentencer—that’s just non­ sense. It’s intellectually dishonest and it’s quite wrong. At the end of the day, the overall sentence has to be just. It’s got to be the right disposal, as far as the ­sentencer’s concerned, and in my view that’s where the discount principle gets in the way (Sheriff 10).100

The last named respondent was one of seven respondents in the present study who empha­ sised the importance of individualised sentencing over the application of the discount prin­ ciple. In explaining that the discount principle did not necessarily affect their ability to do justice in the individual case, the respondents explained: I have to confess, I have just done what I think is right, and so far I haven’t been appealed. I’m not approaching it from the point of view of the discount, because I can’t do it—it just doesn’t lend itself to that kind of disposal … I have never been applying a discount. I may have been deciding what is an appropriate disposal, but that is not the same as applying the discount … One has to look at what is the correct disposal (Sheriff 10). It’s just something we can have a grump about over a cup of coffee. But we can deal with it. Because whatever [the politicians] say, we’ll go and—individually—sentence our cases appropriately (Sher­ iff 3). You find a way out. The most important thing is justice (Sheriff 6). It’s something one has to have regard to, but I don’t think it’s something that would affect our deci­ sion if in all the circumstances we didn’t want to give effect to it, as it were (Sheriff 2). You have to deal with each case very much on its own facts and circumstances. And the discount is just another factor in it (Sheriff 15).

100  On this point, see also Leverick, 2010: 146. Sheriff Thomas Welsh QC notes that ‘under-­discounting’ also occurs, particularly in the summary criminal courts in cases where sheriffs consider their sentencing powers of 12 months to be inadequate, due to the seriousness of the offence, the offender’s record, or a combination of the two (see the decision in Jones v Nisbet 2013 SCCR 282 and Sheriff Welsh’s commentary thereto in the Scottish Criminal Law report—Welsh, 2013b: 538).

Conclusion

 225

I have never given anybody a third off anything. I have never done the mathematical thing … In most cases, provided the end result sentence is broadly within tolerable lines, and you give them [the accused and his counsel] the line ‘But for your plea of guilty it would have been up to X per cent higher’ or, for example, ‘up to 18 months higher’, ‘up to a year more’, something like that, [they won’t appeal]. So I’m satisfied that the level of discount I’m giving is just a modest amount … I don’t think [discounting] works, but I can’t say that in open court. So I go through the rigmarole but I make sure that the difference is not more than a manageable amount of time in real terms (Judge 3).

Conclusion Whilst the practice of guilty plea discounting was originally viewed as ­objectionable by the Appeal Court, the introduction of permissive legislation and its ­subsequent interpretation by the Court resulted in the offender’s guilty plea becoming an established sentencing con­ sideration. Discounts in sentence came to be ­routinely awarded for early guilty pleas. There has, however, been a marked shift in appellate attitudes towards the practice in recent years: whilst the Appeal Court’s decisions in Du Plooy and Spence mirrored English sentencing jurisprudence and practice in setting specific ‘tariffs’ or ‘going rates’ of discount in the form of a sliding scale with a concomitant focus on efficiency considerations as the primary justification for the practice, recent decisions such as Gemmell and Murray have reverted to the traditional Scottish focus on individualised justice in stressing the discretionary nature of guilty plea discounting. The contrast between the current, discretionary approach to guilty plea discounting in Scotland and the previous, tariff-based approach modelled on English practice encapsu­ lates the essential difference between judicial use of discretionary sentencing guidelines in jurisdictions such as Scotland, the Republic of Ireland, Australia and Canada and judicial recourse to presumptively binding numerical guidelines in England and Wales. The Eng­ lish approach, founded upon by counsel for certain of the appellants in Gemmell and by counsel for the appellant in Murray, is formal, prescriptive, heavily structured, numerical and presumptively binding. Modern Scottish practice, however, is illustrative both of the rejection of a mathematical approach to sentencing and of the continued importance of the traditional approach to guidelines in this jurisdiction: whilst appellate decisions provide a structure for discretion, it is the individual judge or sheriff who retains responsibility for determining both the appropriate sentence and the appropriate discount (if any), given the particular circumstances of the individual case. The two approaches can be seen as a dichotomy between the two different methods of judicial decision making in sentencing considered in Chapters 5 and 6 (see Aas, 2005 and ­Flyvbjerg, 2001). Whilst the English approach is system-orientated, abstract and objective— an approach which creates a clear conflict between the pursuit of consistency and the value of individualised sentencing—the Scottish approach is more case-orientated, concrete and subjective. Lord Gill’s opinions in Gemmell and Murray stress that the assessment of an accused’s sentence and the subsequent decisions of whether to grant a discount for his early plea and, if so, the level of that discount, are all discretionary judgments based on judicial ­experience—a form of decision making described by Lord Gill in terms of the Markarian

226 

Judicial Discretion and Guilty Plea Discount

instinctive ­synthesis. The decision of the majority in Gemmell thus emphasises the pri­ macy of the sentencer as an expert actor and phronimos, or a person of practical wisdom. As ­illustrated by Gemmell, the sentencer achieves equity, in the sense of particularised j­ ustice, by calling upon his vast reservoir of practical experience both to set the headline sentence and to address the question of discount. This is a paradigm example of the ­intuitive, holistic and interpretive form of decision making involved in phronesis. Respondents in the present study were deeply concerned at the way in which the ­statutory provisions on guilty plea discounting and early appellate jurisprudence (particu­ larly Du Plooy and Spence) constrained their ability to do justice in the individual case. Respondents’ concerns regarding the practical risks of guilty plea discounting focused on (i) the incentive that the practice provides for those accused who have a stateable defence to plead guilty; (ii) the deleterious effect that the practice has on the credibility of sentences; (iii) ‘double dipping’, ie the interaction of the discount with earlier charge and/or fact bar­ gaining between the Crown and the defence which can potentially result in accused persons receiving a double benefit; (iv) the fact that a guilty plea may itself result in an ‘inherent’ or ‘unseen’ discount in that the court will not have heard the full evidence and will instead have to proceed on the basis of what may be a bland and uninformative agreed narrative; and (v) the tension between judicial discretion (as a means of individualising sentences) and the discount principle as an administrative tool (which both prioritises consistency and conceptualises the sentencer not as a phronimos whose task it is to craft an appropriate sentence, but as an administrator tasked with expediting case flow).101 Respondents’ concern to achieve particularised justice through the practice of equity, through recourse to a value pluralist form of decision making and by exercise of the instinc­ tive—or phronetic—synthesis was such that the discounting regime prior to Gemmell was deeply unpopular. Particularly with cases prosecuted on indictment, some respondents paid only lip service to the statutory provisions and appellate jurisprudence, whilst oth­ ers ignored it altogether either by inflating their headline sentences and/or by continuing to focus exclusively on the individual case. Several respondents stressed the importance of individualised sentencing over the application of the statutory discount provisions in explaining that the requirement to consider the allowance of a discount did not neces­ sarily affect their ability to do justice in the individual case. By emphasising, for exam­ ple, that ‘the important thing is justice’ and that ‘whatever [the politicians] say we’ll go and—­individually—sentence our cases appropriately’,102 respondents demonstrated that, as discussed in Chapter 4,103 they continue to sentence according to values that privilege their own professional knowledge and experience as judges, rather than the policy goals of politicians.

101  The fact that the practice may also in effect penalise an accused who exercises his right to trial (an issue much debated by academic writers) proved not to be an issue of concern to respondents in the present study. 102  The comments of Sheriffs 6 and 3 respectively. 103  Under the section entitled, ‘Individualised Notions of Justice’.

8 The Phronimos and the Metronomic Clockwork Man In my role as a professional support lawyer to the senior judiciary in Scotland, I have been afforded a rare insight into judicial perceptions of, and attitudes towards, sentencing; and yet I have always been acutely aware of the dearth of empirical studies, particularly in the United Kingdom, on judicial views of the sentencing process. Having thus identified a gap in the empirical research, I devised and undertook the present study in an attempt to ascer­ tain what judges and sheriffs in Scotland think about sentencing and how they approach the task. The resulting study comprises the first qualitative empirical examination of j­udicial sentencing in Scotland in over a decade. The study offers a unique insight into judicial atti­ tudes and perceptions of the sentencing process in Scotland and of how the Scottish judi­ ciary go about making their sentencing decisions. It enables us to learn from those whose lives represent the wisdom of accumulated practice (Frank, 2012: 51), producing intimate knowledge of judicial understandings of the sentencing process (see Schram, 2012: 17). I was fortunate to secure the support of the Lord Justice General, the Lord Justice Clerk and the six Sheriffs Principal. Having been granted permission to approach sentencers with an invitation to participate in the study, I was able to secure interviews with eight judges and 17 sheriffs across Scotland. The level of judicial interest in the study was such that I was able to interview considerably more than the 15 sentencers I had originally intended. Almost 70 per cent of those judicial officers approached agreed to participate in the study: this figure compares well with Mackenzie’s 2001 study of the Queensland judiciary and far exceeds the response rate of recent empirical studies of the Irish and English judiciaries (see, respectively, Maguire, 2010 and Dhami, 2013a). The overall impression gained of the interviewing process was of engagement and commitment by respondents, all of whom appeared pleased to have been afforded the opportunity to discuss the issues. The study demonstrates the importance of an individualised and d ­ iscretionary approach to sentencing, one in which the inherent flexibility of the process ensures that the sentence is precisely tailored both to meet the needs of the individual offender and to reflect the gravity of the particular offence. My claim has been that the balancing exercise involved in such discretionary approaches relies on the sentencer’s experience, intuition and judgment. Thus, the sentencing task is best conceptualised in terms of a phronetic synthesis of the rel­ evant facts and ­circumstances of the individual case. The study also highlights the significant disjuncture between, on the one hand, the ­political desire to restrict and control judicial discretion in sentencing (through the intro­ duction of mandatory minimum sentences, presumptively binding, numerical sentencing guidelines and the requirement to consider granting discounts for guilty pleas) and, on the other, the needs of judges and sheriffs to retain a wide discretion in order to achieve

228 

The Phronimos and the Metronomic Clockwork Man

individualised justice. It has been demonstrated that the statutory requirement to take an offender’s guilty plea into account in sentencing, along with the way in which the pro­ vision was originally interpreted by the Appeal Court, did not generally prove popular amongst respondents. Based on the analysis presented here, it is suggested that any moves by P ­ arliament to control the sentencing discretion of judges and sheriffs by introducing mandatory minimum sentences, or especially by introducing a system of presumptively binding sentencing guidelines modelled on the current English approach, would not only meet with resistance from the Scottish judiciary but, more importantly, would be ineffec­ tive. The study confirms the findings by Tombs (2008: 97) that judges and sheriffs in this jurisdiction will always continue to sentence according to values that privilege their own professional knowledge and experience as judges.

How Judges Sentence The Phronetic Synthesis I had ten years as a practising advocate [and] thirty-six years sitting on the Sheriff Court Bench … in courts all over Scotland. If it is correct to say that the proper study of m ­ ankind is Man and, of course, Woman, what I have seen and heard in these courts in these forty-six years convinces me that this admittedly proper study is one where the ­student will never exhaust the subject. There is early realisation that the doings of humanity are infinitely varied, and as infinitely varied are the explanations that h ­ umanity ­frequently offers to explain these doings … [T]he law and the courts are concerned in acute and p ­ ersonal detail with the infinite variety of the passions, challenges, ambitions and i­ nterests of humanity (Smith, 2011: 298, 309). [S]entencing is an art, not a science. It could, of course, be made into a science, as it appears to be in some other countries—a precise tariff is prescribed by law, including specific additions for aggravating factors and detailed deductions for mitigating factors: the Sheriff would then simply press the appropriate keys on his computer, and its penal cash register would ring up the net total of the sentence. Such a system would have the advantage of certainty, but it might well add up to injustice in many cases. It is quite contrary to the Scottish system, which, within prescribed l­imits, leaves sentencing as a matter for the judge’s discretion. That, I believe, is as it ought to be, for the circumstances of one crime and one criminal are never the same as another … The art of the sentencer lies in determining where the emphasis should be placed in each case—on retribution, deterrence, or rehabilitation (Thomson, 2001: 128, 130). In our way of life, it is judges and judges alone who determine the appropriate sentences … [J]udges should not be dictated to by politicians on the subject of ­sentencing in an attempt to find a solution to a problem that they themselves should be solving (Wheatley, 1987: 217).

These observations on the nature of sentencing by three judges (Sheriff Irvine Smith QC, Sheriff Nigel Thomson and the former Lord Justice Clerk, Lord ­Wheatley), who between them amassed a total of 97 years on the Bench, illustrate the central argument advanced by this study: that as individualised justice and the practical wisdom of judges are the key factors in sentencing, it is a process best viewed not necessarily in terms of the ‘instinc­ tive synthesis’ methodology preferred by the Australian courts and subjected to much academic criticism, but rather in terms of a phronetic synthesis of the relevant facts and

How Judges Sentence

 229

circumstances of the individual case. My claim is that in order to comply with the demands of justice, s­entencing must be conceptualised as a form of case-orientated, concrete and ­intuitive d ­ ecision making. It is a form of decision making that must remain an essen­ tially ­discretionary process, structured by appellate guidance, to allow judges to exercise principled discretion. In applying the concept of the phronetic synthesis, I have sought to advance a view of ­sentencing methodology that incorporates both the discretionary based ­‘instinctive ­synthesis’ approach of the Australian courts—in which the sentencer aggregates and assesses the factors bearing on a sentencing decision in a single, global process of ­reasoning1— and the intuitive, holistic and interpretive form of decision making involved in the ­Aristotelian concept of phronesis, or ‘practical wisdom’. The study has demonstrated how phronesis is used by judges and sheriffs to deliberate in relation to substantive issues of ­sentencing law, practice and policy (Flyvbjerg et al, 2012a: 3). In employing such a phronetic perspective, the study has not attempted to produce what Flyvbjerg et al describe as ‘the unrealizable perfection of expert knowledge’ such as that which comes from abstract ­models; rather, the intent has been to strive for the ‘adequation’ of what works for judges and sheriffs in their discharge of the sentencing task (Flyvbjerg et al, 2012b: 286). The result is a form of sentencing methodology that achieves individualisation through judicial recogni­ tion of the profoundly contextualised nature of the process. This study has examined alternatives to the instinctive, or phronetic, synthesis ­methodology of sentencing. In Chapter 3, the ‘staged’ approach was considered, whereby the s­entencer begins by postulating a sentence appropriate to the gravity of the offence and then proceeds to make adjustments for relevant mitigating and aggravating factors. The staged approach is favoured by commentators such as Edney and Bagaric (2007) who criticise the instinctive synthesis as insufficiently transparent and as contrary to the rule of law. H ­ utton, meanwhile, criticises wide judicial discretion in sentencing more generally, viewing sentencing decisions as naked expressions of value preference which are ‘hardly recognisable as legal decisions at all’ (Hutton, 2006a: 169). Chapter 6 examined the devel­ opment of sentencing guidelines in England and Wales—a process that has culminated in the introduction of prescriptive and presumptively binding numerical guidelines issued by the Sentencing Council. This study has argued against both approaches. There is a tension between, on the one hand, effective and just outcomes in sentencing and, on the other, certain outcomes in sen­ tencing that legalist critics of judicial discretion who decry uncertainty overlook (­ Posner, 2008: 374). The staged approach to sentencing puts a premium on the transparency, con­ sistency and standardisation of penal decisions at the expense of individualised justice. Critics such as Edney, Bagaric and Hutton are clearly opposed to a sentencing approach that ­prioritises judicial discretion and in which consistency is seen as neither desirable nor achievable. As Stobbs notes, however, it is telling that although the views of the judiciary are of paramount importance in this area, Edney and Bagaric fail to consider any research which examines the personal perspectives of the judiciary in relation to their roles in c­ rafting appropriate sentences (Stobbs, 2008: 465). This arguably betrays a lack of understanding of how cases are actually decided and of how judges actually sentence (see generally Posner, ibid 219, 2010: 18 and 2016: 306; see also Herz, 2012: 12). 1 

Markarian v The Queen [2005] HCA 25; R v Young [1990] VR 951; R v Williscroft [1975] VR 292.

230 

The Phronimos and the Metronomic Clockwork Man

This study has, in particular, argued against the system of guidelines ­currently o ­ perating in England and Wales where discretion is structured by the use of a sentencing a­ lgorithm (Roberts, 2011a: 1011). This ensures that sentencing o ­utcomes are consistent and ­predictable. Such consistency is achieved, however, at the expense of individualised justice: judicial discretion is replaced with a ‘one size fits all’, ‘box ticking’ approach to ­sentencing (Cooper, 2008: 279–80 and 2013: 164) whilst individual, contextualised ­offenders are replaced with ‘two dimensional crime-and-criminal history amalgams’ (Tonry, 1996: 20; see also Cooper, ibid 158–64). In Scotland, the continued emphasis on judicial discretion in sentencing means that, far from being uniform, regimented and linear, sentencing o ­ utcomes are ‘wobbled through the prism of personality’2—the personality, history and characteristics of both the offender and the victim, along with the professional experience of the sentencer as a phronimos, or person of practical wisdom. As the circumstances of particular offences and particular offenders are ‘infinitely v­ arious’3 and ‘infinitely complicated’ (Padfield, 2011: 99), they out­ run our ­capacity to ­formulate general rules (Solum, 2008: 147 and 2003: 206). The English system fails to acknowledge that sentencers are often required to choose between rendering substantive justice in the case at hand and maintaining the law’s certainty and predictability (see Posner, 1999a: 242). Thus, the English system of sentencing guidelines presupposes a uniformity of crime and criminals which simply does not exist. Justice is always preferable to consistency and no guideline system can cater for the unique circumstances of every case.

The Importance of (Principled) Judicial Discretion The conclusions of the present study confirm the findings of Mackenzie (2001: 462) that there remains a divide between the judiciary, who are defensive of the maintenance of a wide sentencing discretion, and commentators—most recently Edney and Bagaric (2007), Hutton (2006a) and Roberts (2012a, 2012b)—who see a need for a reduction in dispar­ ity and an increase in consistency which, they argue, structuring judicial discretion would bring. The present study has demonstrated, however, that in the years since Mackenzie con­ ducted her research with the Queensland judiciary, support for a wide judicial discretion in sentencing—or at least for the exercise of such discretion in accordance with principles laid down by the appellate courts—has increasingly been voiced by academics (for example Aas, 2005 and Tombs, 2008, 2004) and especially by practitioners (for example O’Malley, 2011 and Brown, 2014a, 2014f and 2013a). Such renewed calls for a wide judicial discretion have often been made in response to the guideline system now operating in England and Wales (see, for example, Cooper, 2013 and 2008; Harris, 2013; Harris and Gerry, 2013; and Thomas, 2013). Differentiation must be made between inappropriate restriction of judicial discretion in sentencing, and appropriate regulation (Mackenzie, ibid 463). In order to comply with the demands of justice, sentencing must remain discretionary and so the selection of sentence

2  The phrase is taken from Martin Amis’ comparison of chess and poker: ‘On a chessboard, the properties and powers of a bishop are permanently fixed. In poker, it’s all wobbled through the prism of personality’ (reported in Coren, 2009: 178). 3  R v Young (n 1) at 955.

How Judges Sentence

 231

in specific cases must remain exclusively a task for the judge or sheriff. In order to produce consistency of approach and to reduce unwarranted disparity in sentencing, however, it is necessary for the Appeal Court to develop rational criteria for deciding on the nature and severity of sanctions. The challenge, as O’Malley (2011: 5) notes, is to achieve consensus on acceptable methodologies for assessing the gravity of offences and the impact of other ethically relevant factors such as an offender’s youth, deprived background, old age, or his caring responsibilities for dependent children. Consensus must also be achieved on the weight to be attributed to such ethically relevant factors and on both the kind and quantum of punishment appropriate for various categories of offender (O’Malley, ibid). The best way to achieve such goals in a manner consistent with the application of the sentencer’s practical wisdom is by recourse to O’Malley’s notion of ‘principled discretion’. Sentencers are to retain the discretion they already enjoy, but must exercise it in accordance with settled principles which can, nevertheless, be departed from when justice so requires (O’Malley, 2001: 35). Such principled discretion results in informed judgment (O’Malley, 2011: 5). This, in turn, leads to the development of a ‘sentencing canon’ of leading appel­ late decisions—decisions which are authoritative yet not inflexible (O’Malley, 2006: 66–67). Based on a thorough analysis of Scottish sentencing jurisprudence, it has been shown that in recent years the Appeal Court has embraced the notion of principled discretion— although never articulating it in such terms—both through its increasing tendency to address questions of principle in its judgments and through an increased use of its statutory power to issue guideline judgments. The Appeal Court has, however, repeatedly emphasised that such guideline judgments are not prescriptive: they provide a structure for, but do not remove, judicial discretion.4 They provide important and appropriate guidance to judges and sheriffs whilst also continuing to allow flexibility.5 In short, guideline judgments in Scotland facilitate the exercise of principled discretion and the attainment of individualised justice in sentencing through a wise blending of appellate sentencing guidance with the phronetic synthesis.

Phronetic Synthesis and Principled Discretion in Practice— Sentencers’ Views Far from seeing the instinctive, or phronetic, synthesis as an ‘inscrutable’ notion (Ashworth, 2015: 81; 2010a: 77; 2005: 73), 23 of the 25 respondents in the present study employed this methodology to effectively balance the competing sentencing aims in the particular case. Whilst the eight most experienced sentencers conceptualised sentencing purely in terms of an exercise of intuition and judgment in which the key factor was their experience, intuition remained a crucial factor in the sentencing practice of the 15 respondents who undertook the task through critical reflection over their intuition. Both approaches are characterised by the practical wisdom of the sentencer rather than the staged, rule based approach of the present English system of sentencing guidelines. 4 See the discussion in Chapter 6 under the section entitled, ‘A Blending of the Phronetic Synthesis with ­Sentencing Guidance—Principled Discretion in Scottish Sentencing’. 5  See also Mackenzie, 2001: 476, who concludes her study by advocating the use of guideline j­udgments in Queensland.

232 

The Phronimos and the Metronomic Clockwork Man

Twenty three of the 25 respondents saw sentencing as a process where judicial discretion should be maintained for reasons of fairness and equity. These findings are consistent with the only previous comparable study of the Scottish judiciary’s sentencing practices (Tombs, 2004). In sentencing by way of the phronetic synthesis, 23 respondents in the present study approached the task in the same way as their English counterparts have traditionally done.6 This approach to sentencing is also followed by judges in Australian jurisdictions, by the Canadian judiciary, the South African judiciary, and by sentencers in the Republic of Ireland.7 Judges in the present study were not in favour of legislation which restricted their sen­ tencing discretion8 whilst sheriffs were similarly critical of moves to constrain their discre­ tion in sentencing.9 As with Mackenzie’s study of the Q ­ ueensland judiciary (Mackenzie, 2001: 462), the present study found a solid attachment on the part of judges and sheriffs to the view that discretion in sentencing ought to remain with the sentencer. Respondents were generally dismissive of moves in England and Wales to structure sentencing discretion through the use of presumptively binding guidelines. Respondents shared the concerns of English commentators and practitioners such as Cooper (2013, 2008) in viewing the English approach as a ‘box ticking’ exercise, irreconcilable with their usual careful consideration of the individual offender and the pertinent facts of the indi­ vidual case. The present study suggests that there would likely be considerable resistance on the part of judges and sheriffs towards any attempts to structure discretion through the introduction of prescriptive and presumptively binding, numerical sentencing guidelines applied through a staged approach, as exemplified by the new format of guidelines issued by the Sentencing Council for England and Wales.10 As with the study by Millie et al, however, it was found that whilst 23 of the 25 respond­ ents were against the use of presumptively binding sentencing guidelines, they generally welcomed sentencing guidance from the Appeal Court (see Millie et al, 2007: 251). Thus, the Appeal Court should continue to address and articulate questions of principle in ­sentencing by issuing sentencing guideline judgments. As O’Malley observes, every academic author hopes to avoid the kind of censure meted out by Lawton LJ in R v Shelton11 where his Lordship described the case as ‘a striking exam­ ple of what can happen if counsel, after consulting the commentaries of academic writers, develop arguments which have the allure of legal logic but which, if taken too far, affront common sense’ (O’Malley, 2013b: vii). Whilst the approach of critics such as Edney and Bagaric (2007) and Hutton (2006a) may have the allure of legal—or, more accurately, legalist—logic, their position, whilst not necessarily affronting common sense, certainly affronts the sense of the uncommon variety that drives sentencing in Scotland as well as in ­Australian jurisdictions (see Connolly, 2006: 2). The arguments of the legalist critics in favour of certainty and consistency ignore what this study has shown to be the centrality of

6 

See the discussion in Chapter 6 under the section entitled, ‘Judicial Conceptions of the Sentencing Process’. Ibid. See the discussion in Chapter 4 under the section entitled, ‘Equity in the Sentencing Practice of the Scottish Courts’. 9  Ibid. 10  See Chapter 6. 11  (1986) 83 Cr App R 379. 7 

8 

Towards a New System of Guilty Plea Discounting

 233

the judge’s practical wisdom to the task of sentencing. The very different approaches to the use of sentencing guidelines in England and Scotland can be seen as a dichotomy between the two methods of judicial decision ­making. Whilst, in terms of the model described by Aas (2005), the English approach is system-­orientated, abstract and objective, sentencing in Scotland remains, and should continue to remain, a discretionary process—one in which judicial ­decision ­making is concrete, case-orientated and subjective.

Towards a New System of Guilty Plea Discounting—Suggestions for Reform A number of findings have emerged from the analysis presented in this study of judicial approaches to the operation of section 196 of the Criminal Procedure (Scotland) Act 1995 on the guilty plea discount. It may be useful to note these findings in the hope that they inform further debate. As Mackenzie (2001: 474) notes of certain of her own findings, some of these matters are issues for legislative change, whilst others fall to be considered by the Appeal Court. The present study demonstrates that, particularly in the High Court, the practice of guilty plea discounting is not easily reconciled with the traditional S­ cottish focus on indi­ vidualised justice in sentencing. Based on the sample of judges interviewed, it would appear to be a practice that is subject to widespread judicial resistance. In light of the comments made by respondents in the present study, it is argued that consideration should be given to reforming the system of guilty plea discounting. For sentencing to remain a discretionary task sensitive to the facts of individual cases, a number of options are required to be con­ sidered by legal policy makers and by the Appeal Court itself: they include abolition of the guilty plea discount; abolition of the sliding scale of discounts set out in Spence v HMA;12 ‘capping’ the discount at a lower level; and restricting discounts in cases where the accused has entered into plea and/or charge bargains with the Crown. In Australia, abolition of the guilty plea discount has been advocated by Mack and Roach Anleu (Mack and Roach Anleu, 1998: 274, 1997: 124). The authors consider that the guilty plea discount places an inappropriate burden on the accused’s choice to plead guilty; that it undermines proper sentencing principles; risks inducing pleas of guilty from the innocent; and undermines judicial neutrality and independence (ibid). Abolition of the guilty plea discount has in the past been favoured by Ashworth (Ashworth, 1998c: 293–97, especially at 294–95, 1983: 314). Ashworth argues convincingly that a complete reappraisal of the system of discounting is required and suggests that either complete abolition or major changes in criminal procedure could produce a fairer system for both victims and witnesses (ibid). Ashworth suggests that: To support the discount is … to endorse a dilution, for bureaucratic reasons, of rights which the criminal process is supposed to protect—the right to require the prosecution to prove guilt, the right to trial in open court, and most fundamentally the right not to be subjected to unfair ­pressure in determining how to plead to the charges (Ashworth, 1983: 314). 12 

2008 JC 174.

234 

The Phronimos and the Metronomic Clockwork Man

As the present study has shown, it can also be argued that discounting can result in ­sentences that do not adequately reflect the seriousness of the particular offender’s conduct. Abolition of the system of discounting is an attractive option given that discounting ‘undermines our efforts to do justice, protect the innocent from conviction, and appropri­ ately punish the guilty’ (Schulhofer, 1994: 139). It is necessary, however, to balance the many criticisms that can be made of the guilty plea discount in principle against its effectiveness as a practical mechanism (Brook et al, 2016: 1221–22; Freiberg, 2014: 376; Mack and Roach Anleu, 1997: 138). As Schulhofer notes, any court that successfully eliminated all guilty plea concessions would undoubtedly face a substantial increase in its trial rate (Schulhofer, ibid 142). Although written in the American context, there is reason to believe that a similar result would ensue in Scotland if formal guilty plea discounts were abolished. Statistics from the Crown Office and Procurator Fiscal Service (COPFS), for example, disclose that in 2015/2016 a total of 29,665 cases in the justice of the peace court were resolved by way of a guilty plea whilst 1,789 such cases went to trial; 47,057 cases prosecuted on summary complaint in the sheriff court were resolved by way of a guilty plea whilst 5,616 such cases went to trial. With regard to cases prosecuted on indictment, 4,398 sheriff and jury cases were resolved by way of a guilty plea whilst 1,115 such cases went to trial. In the High Court, meanwhile, 234 cases were resolved by way of a guilty plea whilst 283 cases went to trial (COPFS, 2016). These figures have remained reasonably constant over the preceding five years (ibid). Thus, whilst the proportion of guilty pleas is highest at the lowest levels of court, a substantial proportion of even the most serious cases indicted in the High Court are similarly resolved by way of a guilty plea (see also Leverick, 2010: 148). If guilty plea discounts were abolished it is likely that more cases—particularly those pros­ ecuted on summary complaint—would be taken to trial, with attendant resource implica­ tions for the criminal justice system (see Crowe, 2013b: 26 and Leverick, 2014: 345). It is for this reason that guilty plea discounting can perhaps be seen as ‘a necessary, efficient, but unprincipled evil’ (Mack and Roach Anleu, 1995: 238). This is reflected in the fact that, whilst not necessarily approving of the practice of discounting, eight sheriffs in the present study regarded the guilty plea discount as a practical necessity. It appears that, as in Australian jurisdictions, the guilty plea discount has become part of the sentencing landscape (see Cow­ dery, 2016: 123 and Willis, 1995: 72) and, as such, complete abolition is simply not feasible. Turning to the operation of discounts in practice, despite Lord Gill’s ­emphasis on the discretionary nature of discounting in both Gemmell v HMA13 andMurray v HMA14 and his Lordship’s clear preference for judicial restraint in g­ ranting guilty plea discounts, the decisions in neither Du Plooy v HMA15 nor Spence v HMA16 have been overturned or disap­ proved (Shead, 2012: 95; cf ­Gordon, 2012: 221). In light of the views expressed by certain respondents in the present study, however, when an appropriate case arises on appeal, a Full Bench could be convened to overrule the decision in Spence, thus leaving the quantifica­ tion of the discount exclusively a matter for the unfettered discretion of the sentencer. This would result in a more flexible and less prescriptive approach, one that is consistent with the Appeal Court’s general rejection of a mathematical approach to sentencing. 13 

2012 JC 223. 2013 SCCR 88. 15  2005 1 JC 1. 16  Spence (n 12). 14 

Towards a New System of Guilty Plea Discounting

 235

Abolition of the sliding scale in Spence would no doubt be criticised by c­ ommentators such as Leverick who value the supposed consistency resulting from the guidelines (see Leverick, 2014: 343–45 and 348–49, 2012: 234–35 and 237–38). Such a course would, however, be entirely consistent with the phronetic synthesis approach to sentencing. The present study has shown that rigid guidelines such as the sliding scale in Spence are anath­ ema to a system which prioritises judicial discretion to achieve particularised justice. Rather than blindly adopting the English Sentencing Guidelines Council’s sliding scale of discounts,17 as was done in Spence, a preferable course is to examine practice in other ­Commonwealth jurisdictions. Few common law jurisdictions, if any, traditionally made formal provision for the amount of discount merited by a guilty plea, preferring to allow courts to assess the appropriate sen­ tence reduction in light of the specific circumstances (O’Malley, 2006: 126). A sliding scale of discounts is not mandated by any of the sentencing statutes in force across Australia, nor have the appellate courts adopted such guidelines of their own volition. Indeed, there has been a clear reluctance amongst appellate courts in Australia to set a general tariff for guilty plea discounts (Brook et al, ibid 1213–14; Edney and Bagaric, 2007: 207; Freiberg, ibid 382; Mackenzie, 2006b: 2; Mackenzie and Stobbs, 2010: 91; and Mack and Roach Anleu, 1997: 126–27).18 As Kirby J noted in Cameron v The Queen:19 ‘No such discount can be reduced to a set formula. Elements of intuition and judgment remain to be given weight in arriving at the aggregate sentence finally imposed’. Similarly, in R v Harman, Andrews CJ of the Queens­ land Court of Criminal Appeal said this: … I can see no reason why sentencing authorities should not emphasise that there is r­ elevance in timely co-operation by persons guilty of offences provided that care in expression is taken to ensure the understanding that there is no system of discounting whether precise or otherwise in force and that there are numerous circumstances to be taken into account in the exercise of what is, after all, a necessarily wide discretion.20

Thus, as Mandie JA of the Supreme Court of Victoria observed in Giordano v The Queen,21 the appropriate discount for a plea of guilty in any given case involves the exercise of the instinctive, or intuitive, synthesis of all relevant factors as much as any other aspect of the exercise of the sentencing discretion.

17  Or indeed adopting the new guidelines on reduction in sentence for a guilty plea when they are issued by the Sentencing Council for England and Wales, following its consultation (see Sentencing Council, 2016). 18  See also the decisions in Bell v Wesley [2007] WASC 264 at [50]; Moody v French [2008] WASCA 67 at [33]; and Bahar v The Queen [2011] WASCA 249 at [43]. 19  [2002] HCA 6 at [71]. 20  R v Harman [1989] 1 Qd R 414 at 419. See also the remarks of Asche CJ of the Northern ­Territory Court of Criminal Appeal in R v Jabaltjari (1989) 64 NTR 1 at 15 in which his Honour urged a greater concentration on the individual offender when determining whether to grant a guilty plea discount. The judgment of Asche CJ in Jabaltjari is also notable for his Honour’s comparative analysis of the practice of guilty plea discounting in England and Wales and, particularly, in Scotland. In this ­latter regard, having noted that guilty plea discounts were routinely awarded by the English courts (at 13–14), Asche CJ observed that: ‘This approach has not found favour with the more logical Scots’ (at 14). His Honour then proceeded to consider certain excerpts from Nicholson’s The Law and Practice of Sentencing in Scotland (Nicholson, 1981); a statement of the law which has, of course, since been superseded by the decisions in Du Plooy and Spence. 21  [2010] VSCA 101 at [43].

236 

The Phronimos and the Metronomic Clockwork Man

In his study of Victorian County and Supreme Court judges, Seifman (1982) reports that the Victorian judiciary did not favour the introduction of guidelines to regulate the practice of guilty plea discounting. The majority of the 38 members of the judiciary inter­ viewed for the study believed such guidelines to be unnecessary (ibid 83). Only one judge in ­Seifman’s study expressed the view that guidelines in this area would be helpful, compared to 12 who stated that such guidelines were not practicable or could be misinterpreted (ibid). More recently, in its report on ‘Sentence Indication and Specified Sentence Discounts’, the Sentencing Advisory Council for the State of Victoria advocated the retention of a wholly discretionary approach to guilty plea discounting: The Council believes that Victorian law should not require all offenders to receive a reduction in sentence for a guilty plea automatically … Cases will arise in which it would be inappropriate for an offender to receive any reduction in sentence, notwithstanding their guilty plea. It would be undesirable to have a situation where, due to a compulsory reduction in sentence for a guilty plea, an offender received a sentence that was incongruous with the offence committed (Sentencing Advisory Council (Victoria), 2007: 57).

Similarly, in its report on the sentencing of federal offenders, the Australian Law Reform Commission (ALRC) did not support legislative prescription of the quantum of a discount, whether in the form of a fixed percentage, a range of percentages, or a maximum percent­ age. The ALRC considered that judicial discretion would be unduly fettered by such an approach (ALRC, 2006: 316). It recommended that the extent of the discount, if any, should be left to the sentencer’s discretion (ibid 317). In New Zealand, the Supreme Court recently examined the operation of guilty plea ­discounts and, in disapproving the sliding scale approach, similarly affirmed the discretion­ ary nature of the process. In Hessell v The Queen,22 the Supreme Court overruled an earlier guideline judgment of the Court of Appeal23 in which the Court had set out a sliding scale of discounts similar to that in Spence.24 In so doing, the Court of Appeal had acknowledged that the English approach to guilty plea discounting had been ‘a major influence’ upon its decision.25 The Supreme Court considered the Court of Appeal’s sliding scale to be objectionable for two reasons: firstly, it was too heavily structured and involved an inappropriate departure from the duty placed on sentencers to evaluate ‘the full circumstances of each individual case’;26 and secondly, it carried the unacceptable risk of pressuring innocent persons to plead guilty.27 The Supreme Court held that any allowance for a guilty plea should be the result of the sentencer’s evaluation of all the circumstances in which the plea is tendered.28 As the Court explained: The only way in which the many variable circumstances of individual cases which are relevant to a guilty plea can properly be identified is by requiring their evaluation by the sentencing judge, and

22 

[2011] 1 NZLR 607. Hessell v The Queen [2010] 2 NZLR 298. Spence (n 12) at [14]–[18]. 25  Hessell v The Queen (n 23) at [3] and [14]. 26  Hessell v The Queen [2011] 1 NZLR 607 at [72]. 27  Ibid. 28  Ibid at [74]. 23  24 

Towards a New System of Guilty Plea Discounting

 237

allowing that judge scope in light of the conclusion he or she reaches to give the most appropriate recognition of the guilty plea in fixing the sentence.29

Thus, whilst the New Zealand Court of Appeal had been persuaded by the prescriptive and structured approach of the Sentencing Guidelines Council and the English courts to sentence discounting, the Supreme Court considered such an approach inappropriate as it restricted the capacity of judges to determine sentences that were considered to fit all the circumstances of the case (L ­Richardson, 2011: 164). In contrast to what Shiels et al (2016: 504) describe as the ‘rigid or formalised scale of sentencing reduction’ that formerly operated in Scotland through the Spence guidelines, and which continues to operate in England and Wales through the Sentencing Guidelines Council’s Definitive Guideline,30 the approach of the Supreme Court of New Zealand is more principled, focusing on the indi­ vidual offender and the circumstances in which he tendered his guilty plea (J Richardson, 2011: 127). Similarly, the Irish courts have never quantified the discount merited by a guilty plea (see O’Malley, 2016: 164–65, 2013a: 226, 2013b: 547, 2006: 128). In the Republic of Ireland there is no formal provision for the amount of reduction merited by a plea of guilty; in particular, there is ‘no mathematical formula by which any such reduction may be arrived at’.31 Insofar as discounts for guilty pleas are concerned, the Irish courts prefer ‘to uphold judicial discre­ tion, to be exercised in light of the specific circumstances’.32 Sentencers are thus required to adopt ‘a contextual approach’ in deciding upon the appropriate discount (if any) for a guilty plea (see O’Malley, 2016: 166). The need for flexibility and discretion in fixing the level of discount was recognised in England and Wales prior to the establishment of the Sentencing Advisory Panel and Sen­ tencing Guidelines Council. For example, whilst the organisation JUSTICE accepted that there were sound public policy reasons for a plea of guilty to reduce the sentence in most cases, it was not convinced that it was either necessary or desirable to move towards a sys­ tem of fixed and graded discounts (JUSTICE, 1993: 18). The organisation believed such a system to be too rigid: to be compatible with the principle of a just sentencing system, it considered that the discount regime should remain flexible (ibid). JUSTICE was also aware of the concerns subsequently expressed by respondents in the present study regarding the way in which discounts interact with charge and fact bargaining. The problem was articu­ lated as follows: Whatever the aims of sentencing, a minimum requirement in any rational system is that there is some degree of correspondence between the facts of the crime and the sentence. To sentence on the rigid basis of fixed and graded discounts … starts to undermine this requirement (ibid 20).

More recently, the research commissioned by the Sentencing Council for England and Wales to examine public attitudes towards the practice of sentence discounting found that the general public tends to be less in favour of universal, standardised approaches to sentence discounts and instead favours a more flexible approach (Dawes et al, 2011: 6–7).

29 

Ibid at [62]. also the Sentencing Council’s consultation paper on the proposed new guideline on reduction in sen­ tence for a guilty plea (Sentencing Council, 2016: 17–19). 31  People (DPP) v Daly [2012] 1 IR 476 at [100]. 32  Ibid. 30  See

238 

The Phronimos and the Metronomic Clockwork Man

Respondents thought that the question of discount should be considered on ‘a case by case basis’ (ibid 29; see also Roberts et al, 2009: 779 and 781), although there was more support for a scaled approach to discounting amongst victims of, and witnesses to, serious offences (Dawes et al, ibid 26–28). The weight of judicial opinion in Australia, New Zealand and the Republic of Ireland, as well as of public opinion in England and Wales, thus favours a flexible and discretionary approach to the granting of guilty plea discounts.33 In Scotland, the primacy of judicial discretion in the granting of discounts—and the concomitant importance of individual­ ised sentencing and particularised justice—could be further enhanced by reconsidering the decision in Spence. Another possibility would be to ‘cap’ the maximum permissible discount at a lower level. As Roberts notes in his response to the Australian Law Reform Commission’s Issues Paper on the sentencing of federal offenders, the purpose of placing a ‘cap’ on the amount of any discount is to discourage guilty pleas from the innocent and to discourage community criticism of the sentencing process (Australian Law Reform Commission, 2006: 316). In the course of his opinion in Gemmell, Lord Gill suggested that the Appeal Court ‘may have to give further consideration to the suitability of discounts of a third or more contemplated in Du Plooy and in Spence’.34 His Lordship continued: To my mind, discounts of that size, particularly in light of the early release provisions … could in certain cases raise a question of public confidence in the ability of our criminal justice system to deal with offenders fairly and resolutely.35

These observations are supported by the findings of the present study. It is suggested that, following the decision of the majority of the Supreme Court of South Australia in R v Shannon,36 the level of discount should be such as to amount to no more than a ­‘moderate encouragement to plead’.37 A maximum discount of 10 per cent might be sufficient to pro­ vide a small incentive for an accused to plead guilty without imposing undue pressure to do so (Ashworth, 1983: 314; see also Ashworth and Zedner, 2008: 47; JUSTICE, 1993: 11; and Mack and Roach Anleu, 1997: 142). It would also have the advantage of addressing the issue of public confidence considered by Lord Gill in Gemmell. The present study has shown that respondents consider the formal system of guilty plea discounting that operated in Scotland prior to the decisions in ­Gemmell and Murray to have had a deleterious effect on the imposition of proportionate sentences, particularly in cases prosecuted on indictment. Respondents also considered the problem to be c­ ompounded by the interaction of the Du Plooy discount with, firstly, charge and fact bargaining and, ­secondly, the inherent discount obtained by pleading guilty and entering into an agreed narrative with the Crown. Thus, the factors that serve to reduce sentences extend well

33  No guidelines on sentence discounting exist in Canada—neither the Supreme Court nor the ­ provincial courts of appeal have offered any guidance on the issue (see Roberts, 2013b: 120, fn 66 and Roberts and Ashworth, 2016: 345). 34  Gemmell (n 13) at [79]. 35  Gemmell (n 13) at [79]. 36  (1979) 21 SASR 442. King CJ, Zelling, Wells and Mohr JJ; Cox J dissenting. Shannon was one of first Austral­ ian decisions to examine the principles of sentence discounting and remained the locus classicus until the decision of the High Court in Cameron v The Queen(n 19) (see Willis, 1995: 42). 37  R v Shannon ibid at 449 (King CJ).

Towards a New System of Guilty Plea Discounting

 239

beyond the guilty plea discount. Individually and collectively, these three factors have a potentially corrosive effect on the sentencer’s ability to do equity in the individual case through the imposition of a sentence that is tailored to the specific circumstances of the offence, the offender, the victim, and the wider community. They are inimical to the judicial approach to sentencing in jurisdictions such as Scotland where individual justice is seen as more important than some more abstract notion of systemic fairness and where particular­ ised justice is achieved through judicial recourse to the phronetic synthesis. This issue could be addressed through allowing the sentencer—in the exercise of his or her discretion in deciding whether to allow a discount—to restrict the amount of any dis­ count granted in cases where, through charge and/or fact bargaining, the accused has nego­ tiated a plea with the Crown. This approach appeared to find favour with Lord Philip in his dissenting opinion in RB v HMA.38 Thus, the discount would be restricted on the basis of the reduced charge to which the accused has pleaded guilty and not on the basis of the seri­ ousness of the offence per se. Treating offence seriousness as a relevant factor in withholding or reducing a discount is criticised by Leverick as sentence discounting ‘is not a practice that lends itself to justification on intrinsic moral grounds’ (Leverick, 2014: 348). Leverick also argues that ‘“seriousness” is not an easily quantifiable concept … and restricting discounts on this basis risks inconsistency’ (ibid). Arguably, however, it is the sentencer as a phronimos who is ideally placed to assess the seriousness of a particular offence.39 Although this approach would rely heavily on the provision of the details of any plea negotiation by both the Crown and the defence, the parties are already under a duty to pro­ vide the court with detailed information as to the form and timing of the plea.40 A precedent for this approach exists in Australian sentencing jurisprudence. The major­ ity view in R v Shannon was that a discount need not necessarily be granted if the guilty plea resulted from a plea bargain where the accused had pleaded guilty to a lesser charge in return for the prosecution agreeing not to proceed with a more serious charge.41 Similarly, a majority of the Victoria Court of Appeal held in R v Gray42 that a discount in sentence need not be granted in cases where the accused’s plea was ‘tainted overmuch by self-interest’.43 Even in England and Wales where, as we have seen, sentencers are constrained by presump­ tively b ­ inding, numerical guidelines, certain dicta of the Court of Appeal clearly indicates that judges retain discretion to withhold all credit for a guilty plea (Roberts, 2013b: 119).44 In R v T,45 for example, it was said that: ‘We do not consider that the sentencing guideline precludes the withholding of any reduction for a guilty plea in a proper case’.46 Thus, whilst a negotiated plea is still likely to result in a lower headline sentence, the suggested approach of restricting any discount in such cases would curb the worst excesses 38 

2004 SCCR 443 at [27]. See the discussion in Chapter 5 under the section entitled, ‘The Ideals of Episteme, Techne and Phronesis’. 40  Shiels et al, 2016: 504; Murray (n 14) at [31]; HMA v Simpson 2009 SCCR 554 at [15]; Mills v HMA 2005 SCCR 1 at [3]; and Smith (CC) v HMA 2004 SCCR 521 at [18]. 41  R v Shannon (n 36) at 452–53; see also Willis, 1995: 43–44, 1985: 135. 42  [1977] VR 225 at 232. 43  See also Lane, 1980: 57 and the decisions in R v Spiller (1969) 4 CCC 211 and R v Wisniewski (1975) 29 CRNS 342. 44  Cf Roberts’ earlier reference to an accused’s ‘entitlement’ to a discount for a guilty plea (Roberts, 2012b: 278–79). 45  [2011] EWCA Crim 2345 at [18]. 46  See also R v March [2002] 2 Cr App R (S) 98 at [22](4). 39 

240 

The Phronimos and the Metronomic Clockwork Man

of the present system whereby disproportionately lenient sentences are often imposed on indictment as a result of the combined effect of negotiated pleas and the formal Du Plooy discount. Such an approach would, it is suggested, go a significant way towards reconciling sentencers’ concern with the case-orientated focus on individualised justice and the systemorientated, economic demands of the modern criminal justice system.

The Phronimos on the Bench The sentencing task may be approached in a number of different ways. Offenders may be sentenced by judges who are under a statutory duty to follow prescriptive, presumptively binding, numerical sentencing guidelines, as in England and Wales. Alternatively, judges can arrive at their decisions by exercising intuition and judgment borne of their ‘vast reser­ voirs of experience’.47 A third approach involves sentencers melding their first impressions of the individual case with deliberative analysis of the facts and circumstances, of statutory provisions, and of sentencing guidance laid down by the Appeal Court—deliberation based on critical reflection over their intuition. This study of how judges do sentence in Scotland suggests that individualised justice is best achieved by either of the latter two approaches, both of which comprise an application of phronetic synthesis. Equitable, effective and proportionate sentencing is more difficult to achieve for judges who have in many cases been rendered ‘metronomic clockwork men’48 (and women) by virtue of a regime of presumptively binding guidelines which they are obliged to apply in a series of nine stages. It is the judge as a phronimos who is best placed to achieve particularised justice in sentencing—those solicitors, advocates and barristers who, after many years’ experience at the Bar, now preside over the criminal courts and who (in jurisdictions such as Australia and Scotland) discharge the sentencing task by way of instinctive, or phronetic, synthesis.

47 

See the comments of Judge 5 in Chapter 6 under the section entitled, ‘The Instinctive Approach’. See the comments of Judge 3 in Chapter 7 under the section entitled, ‘Discretion in ­Discounting—the Findings of the Present Study’. 48 

BIBLIOGRAPHY

Aas, KF (2005) Sentencing in the Age of Information: From Faust to Macintosh (London, The Glasshouse Press). Abadee, AR (2006) ‘The Role of Sentencing Advisory Councils’, Paper delivered at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra. Abbs, R (2008) ‘Markarian v The Queen’ (2006–08) 10 The Newcastle Law Review 135–56. Adelman, L and Deitrich, J (2008) ‘Marvin Frankel’s Mistakes and the Need to Rethink Federal Sentencing’ 13 Berkeley Journal of Criminal Law 239–60. Aitken, B (2009) ‘Criminal Sentencing’ Scottish Criminal Law 766–68. Alison, A (1832) Principles of the Criminal Law of Scotland (1989 Reprint) (Edinburgh, Law Society of Scotland/Butterworths). Alschuler, AW (1981) ‘The Changing Plea Bargaining Debate’ 69 California Law Review 652–730. Anderson, JL (2006) ‘Standard Minimum Sentencing and Guideline Judgments: An Uneasy Alliance in the Way of the Future’ 30 Criminal Law Journal 203–23. —— (2004) ‘“Leading Steps Aright”: Judicial Guidelines Judgments in New South Wales’ 16(2) Current Issues in Criminal Justice 140–59. Annan, N (2013) ‘Prologue’ in H Hardy and R Hausheer (eds), The Proper Study of Mankind—An Anthology of Essays, 2nd edn (London, Vintage Books). Aristotle (1976) The Ethics of Aristotle: The Nicomachean Ethics (translated by JAK Thomson, revised by H Tredennick) (London, Penguin). Ashworth, A (2015) Sentencing and Criminal Justice, 6th edn (Cambridge, Cambridge University Press). —— (2013a) ‘Editorial—‘Unavoidable’ Prison Sentences?’ Criminal Law Review 621–22. —— (2013b) ‘The Struggle for Supremacy in Sentencing’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2012) ‘Departures from the Sentencing Guidelines’ Criminal Law Review 81–96. —— (2011) ‘Re-Evaluating the Justifications for Aggravation and Mitigation at Sentencing’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2010a) Sentencing and Criminal Justice, 5th edn (Cambridge, Cambridge University Press). —— (2010b) ‘Coroners and Justice Act 2009: Sentencing Guidelines and the Sentencing Council’ Criminal Law Review 389–401. —— (2009) ‘Techniques for Reducing Sentence Disparity’ in A von Hirsch, A Ashworth and JV Roberts (eds), Principled Sentencing—Readings on Theory and Policy, 3rd edn (Oxford, Hart Publishing). —— (2008) ‘English Sentencing Guidelines in their Public and Political Context’ in A Freiberg and K Gelb (eds), Penal Populism, Sentencing Councils and Sentencing Policy (Cullompton, Willan Publishing). —— (2006) ‘The Sentencing Guideline System in England and Wales’ 19 South African Journal of Criminal Justice 1–22. —— (2005) Sentencing and Criminal Justice, 4th edn (Cambridge, Cambridge University Press). —— (2004) ‘Criminal Justice Act 2003—Criminal Justice Reform: Principles, Human Rights and Public Protection’ Criminal Law Review 516–32.

242 

Bibliography

—— (2003) ‘Sentencing and Sensitivity: A Challenge for Criminological Research’ in L Zedner and A Ashworth (eds), The Criminological Foundations of Penal Policy—Essays in Honour of Roger Hood (Oxford, Oxford University Press). —— (2002) ‘European Sentencing Traditions’ in C Tata and N Hutton (eds), Sentencing and Society— International Perspectives (Aldershot, Ashgate). —— (2001) ‘The Decline of English Sentencing and Other Stories’ in M Tonry and RS Frase (eds), Sentencing and Sanctions in Western Countries (Oxford, Oxford University Press). —— (2000) Sentencing and Criminal Justice, 3rd edn (London, Butterworths LexisNexis). —— (1998a) ‘Structuring Sentencing Discretion’ in A von Hirsch and A Ashworth (eds), Principled Sentencing—Readings on Theory and Policy, 2nd edn (Oxford, Hart Publishing). —— (1998b) ‘Four Techniques for Reducing Sentence Disparity’ in von Hirsch and Ashworth, ibid. —— (1998c) The Criminal Process—An Evaluative Study (Oxford, Oxford University Press). —— (1995a) ‘Reflections on the Role of the Sentencing Scholar’ in C Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Oxford, Clarendon Press). —— (1995b) Sentencing and Criminal Justice, 2nd edn (London, Butterworths). —— (1989) ‘Criminal Justice and Deserved Sentences’ Criminal Law Review 340–55. —— (1987a) ‘Disentangling Disparity’ in DC Pennington and S Lloyd-Bostock (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies). —— (1987b) ‘Closing Summary’ in I Potas (ed), Sentencing in Australia: Issues, Policy and Reform (Canberra, Australian Institute of Criminology). —— (1984) ‘Techniques of Guidance on Sentencing’ Criminal Law Review 519–30. —— (1983) Sentencing and Penal Policy (London, Weidenfeld and Nicolson). Ashworth, A, Genders, E, Mansfield, G, Peay, J and Player, E (1984) Sentencing in the Crown Court, Occasional Paper Number 10 (Oxford, Centre for Criminological Research, University of Oxford). Ashworth, A and Padfield, N (2015) ‘Editorial—Five Years of the Sentencing Council’ Criminal Law Review 657–58. Ashworth, A and Player, E (2005) ‘Criminal Justice Act 2003: The Sentencing Provisions’ 68(5) Modern Law Review 822–38. Ashworth, A and Redmayne, M (2010) The Criminal Process, 4th edn (Oxford, Oxford University Press). Ashworth, A and Roberts, JV (2013) ‘The Origins and Nature of the Sentencing Guidelines in England and Wales’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). Ashworth, A and Roberts, JV (2012) ‘Sentencing: Theory, Principle, and Practice’ in M Maguire, R Morgan and R Reiner (eds), The Oxford Handbook of Criminology, 5th edn (Oxford, Oxford University Press). Ashworth, A and Zedner, L (2008) ‘Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure, and Sanctions’ 2 Criminal Law and Philosophy 21–51. Atkinson, P and Coffey, A (2011) ‘Analysing Documentary Realities’ in D Silverman (ed), Qualitative Research—Issues of Theory, Method and Practice, 3rd edn (London, Sage). Australian Law Reform Commission (2006) Same Crime, Same Time—Sentencing of Federal Offenders (Report Number 103, April 2006), Sydney, Australian Law Reform Commission. —— (1988) Sentencing (Report Number 44), Canberra, Australian Government Publishing Service. Bacik, I (2002) ‘The Practice of Sentencing in the Irish Courts’ in P O’Mahony (ed), Criminal Justice in Ireland (Dublin, Institute of Public Administration). Bagaric, M (2016) ‘Consistency in the Application of Sentencing Principles: Journey without a Knowable or Testable Destination’ 40(1) Criminal Law Journal 3–5. —— (2015) ‘Sentencing: From Vagueness to Arbitrariness: The Need to Abolish the Stain that is the Instinctive Synthesis’ 38(1) University of New South Wales Law Journal 76–113.

Bibliography

 243

—— (2011) ‘The High Court on Crime in 2010: Analysis and Jurisprudence’ 35 Criminal Law Journal 5. —— (2002) ‘What Sort of Mandatory Penalties Should We Have?’ 23 Adelaide Law Review 113–40. —— (2001) Punishment & Sentencing: A Rational Approach (London, Cavendish Publishing). —— (2000) ‘Consistency and Fairness in Sentencing: The Splendour of Fixed Penalties’ 2 California Criminal Law Review 1. Bagaric, M and Alexander, T (2015) ‘Addressing the Curious Blackspot that is the Separation between the Principle of Legality and Sentencing’ 41(3) Monash University Law Review 515–51. Bagaric, M and Alexander, T (2013) ‘Rehabilitating Totality in Sentencing: From Obscurity to Principle’ 36(1) University of New South Wales Law Journal 139–67. Bagaric, M and Brebner, J (2002) ‘The Solution to the Dilemma Presented by the Guilty Plea Discount: The Qualified Guilty Plea—I’m Pleading Guilty Only Because of the Discount …’ 30 International Journal of the Sociology of Law 51–74. Bagaric, M and Edney, R (2016), Sentencing in Australia, 3rd edn (Pyrmont, Thomson Reuters). Bagaric, M and Edney, R (2003) ‘What’s Instinct Got to Do with It? A Blueprint for a Coherent Approach to Punishing Criminals’ 27 Criminal Law Journal 119–41. Bagaric, M and Pathinayake, A (2013) ‘The Paradox of Parity in Sentencing in Australia: The Pursuit of Equal Justice that Highlights the Futility of Consistency in Sentencing’ 77(5) The Journal of Criminal Law 399–416. Baldwin, J (2008) ‘Research on the Criminal Courts’ in RD King and E Wincup (eds), Doing Research on Crime and Justice, 2nd edn (Oxford, Oxford University Press). Baldwin, J and McConville, M (1977) Negotiated Justice—Pressures to Plead Guilty (London, Martin Robertson). Banks, R and Harris, L (2013) Banks on Sentence (Vol 1) (Etchingham, Robert Banks). Barnes, R (2012) ‘The Advantages of Transcribing your Own Interviews’ in A Bryman, Social Research Methods, 4th edn (Oxford, Oxford University Press). Bartels, L (2009) ‘Suspended Sentences—A Judicial Perspective’ 9(1) Queensland University of Technology Law and Justice Journal 44–63. —— (2008) Sword or Feather—The Use and Utility of Suspended Sentences in Tasmania (PhD thesis, Hobart, University of Tasmania). Bell, A (1999) For a Shilling of Reward—A Glimpse at the Law in Scotland (Edinburgh, AA Bell). Bennett, H and Broe, GA (2007) ‘Judicial Neurobiology, Markarian Synthesis and Emotion: How Can the Human Brain Make Sentencing Decisions?’ 31 Criminal Law Journal 75–90. Berger, BL (2015) ‘Sentencing and the Salience of Pain and Hope’ Osgoode Legal Studies Research Paper Series, Paper 97. Berlin, I (2013a) ‘The Hedgehog and the Fox—An Essay on Tolstoy’s View of History’ in H Hardy and R Hausheer (eds), The Proper Study of Mankind—An Anthology of Essays, 2nd edn (London, Vintage Books). —— (2013b) ‘Two Concepts of Liberty’ in Hardy and Hausheer, ibid. —— (2013c) ‘The Pursuit of the Ideal’ in Hardy and Hausheer, ibid. —— (2013d) ‘The Concept of Scientific History’ in Hardy and Hausheer, ibid. —— (2002a) ‘Five Essays on Liberty—Introduction’ in H Hardy (ed), Liberty (Oxford, Oxford University Press). —— (2002b) ‘Political Ideas in the Twentieth Century’ in Hardy, ibid. —— (2000) ‘My Intellectual Path’ in H Hardy (ed), The Power of Ideas (London, Chatto & Windus). —— (1998) Personal Impressions, 2nd edn (edited by H Hardy) (London, Pimlico). —— (1997) ‘Political Judgement’ in H Hardy (ed), The Sense of Reality—Studies in Ideas and their History (London, Pimlico). —— (1975) ‘The End of the Ideal of a Perfect Society’ Radio broadcast transcribed in H Hardy (ed), The Isaiah Berlin Virtual Library.

244 

Bibliography

Berlin, I and Williams, B (1994) ‘Pluralism and Liberalism: A Reply’ 42(2) Political Studies 306–09. Bingham, Lord (2010) The Rule of Law (London, Allen Lane). —— (2000a) ‘The Sentence of the Court’ (lecture delivered to the Police Federation, Merchant Taylors’ Hall, London, 10 July 1997) in The Business of Judging—Selected Essays and Speeches, 1985–1999 (Oxford, Oxford University Press). —— (2000b) ‘The Discretion of the Judge’ (the Royal Bank of Scotland Lecture, Oxford University, 17 May 1990) in ibid. Blackett, HH Judge J (2013) ‘A View from the Court Martial’ in R Banks and L Harris, Banks on Sentence, 8th edn (Etchingham, Robert Banks). Bloos, MR and Plaxton, MC (2003) ‘Starting-Point Sentencing and the Application of Laberge in Unlawful Act Manslaughter Cases’ Criminal Reports (Articles) (6th Series) 346. Bloos, MR and Renke, WN (1997) ‘Case Comment: Stopping Starting Points: R v McDonnell’ 35(3) Alberta Law Review 795–812. Bonomy, Lord (2002) Improving Practice—2002 Review of the Practices and Procedures of the High Court of Justiciary (Edinburgh, Scottish Executive). Bottoms, A (1998) ‘Five Puzzles in von Hirsch’s Theory of Punishment’ in A Ashworth and M Wasik (eds), Fundamentals of Sentencing Theory (Oxford, Oxford University Press). —— (1995) ‘The Philosophy and Politics of Punishment and Sentencing’ in C Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Oxford, Clarendon Press). Bourdieu, P (1990) In Other Words: Essays Towards a Reflexive Sociology (Cambridge, Polity Press). —— (1977) Outline of a Theory of Practice (Cambridge, Cambridge University Press). Braithwaite, J and Pettit, P (1990) Not Just Deserts (Oxford, Oxford University Press). Brennan, G (2006) ‘The High Court and the Sentencing Environment’, Paper delivered at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra. Brett, N (1992) ‘Mercy and Criminal Justice: A Plea for Mercy’ 5(1) Canadian Journal of Law and Jurisprudence 81–94. Brien, A (1998) ‘Mercy Within Legal Justice’ 24(1) Social Theory and Practice 83–110. Brook, CA, Fiannaca, Justice B, Harvey, Judge D, Marcus, P, McEwan, J and Pomerance, Justice R (2016) ‘A Comparative Look at Plea Bargaining in Australia, Canada, England, New Zealand, and the United States’ 57(4) William & Mary Law Review 1147–1224. Brown, Sheriff AN (2015) Greens Annotated Acts—Sexual Offences (Scotland) Act 2009, 2nd edn (Edinburgh, W Green). —— (2012) ‘Legislation’ Criminal Law Bulletin 118, 4–5. —— (2010a) Criminal Evidence and Procedure, 3rd edn (Edinburgh, Avizandum Publishing). —— (2010b) ‘Case Notes—Assault by a Schoolteacher (and Over-Reaction by the Crown?)’ Criminal Law Bulletin 103, 2–3. Brown, G (2016a) ‘Sentencing Guidance for Rape: The Crown Sentence Appeals in K and B’ Criminal Law Bulletin 127, 2–7. —— (2016b) ‘Sentencing for Large Scale Drugs Supply: Sutherland v H.M. Advocate, 2016 SLT 93’ Criminal Law Bulletin 140, 4–7. —— (2015a) ‘Sentencing Cases—Being Concerned in the Supply of Cannabis—Extension of Lin Guidelines to Offences under Section 4(3)(b) of the Misuse of Drugs Act 1971’ Criminal Law Bulletin 138, 7. —— (2015b) ‘The Appeal Court’s Rejection of English Sentencing Authority on Assault: SS v H.M. Advocate, [2015] HCJAC 63’ Criminal Law Bulletin 137, 5–7. —— (2015c) ‘The Proper Use of English Sentencing Guidelines and Further Observations on the Guilty Plea Discount—Geddes v H.M. Advocate, [2015] HCJAC 45’ Criminal Law Bulletin 136, 3–6. —— (2015d) ‘Sentencing Cases—Supply of Heroin—Reference to English Guidelines Disapproved’ Criminal Law Bulletin 134, 6–7.

Bibliography

 245

—— (2015e) ‘Sentencing Cases—Rape of a Young Child and Communicating Indecently with a Young Child by 15-Year-Old Offender’ Criminal Law Bulletin 133, 7–8. —— (2015f) ‘Sentencing Cases—Discretionary Nature of the Guilty Plea Discount’ Criminal Law Bulletin 137, 7–8. —— (2014a) ‘Curfew Credit, “Instinctive Synthesis” and the Role of the Appeal Court in Sentence Appeals—the Decision in McGill v H.M. Advocate’ Criminal Law Bulletin 127, 3–6. —— (2014b) ‘Order for Lifelong Restriction and the Fixing of Punishment Parts (Yet Again)—The Decisions in Ferguson and Chalmers’ Criminal Law Bulletin 129, 3–6. —— (2014c) ‘Sentencing Cases—Culpable and Reckless Conduct—Irresponsible Prank with Serious Consequences for the Complainer’ Criminal Law Bulletin 130, 8. —— (2014d) ‘Impact Mitigation—Is the Effect of Custody on an Offender’s Child a Legitimate Sentencing Consideration?’ Criminal Law Bulletin 128, 2–5. —— (2014e) ‘Sentencing Teenaged Sex Offenders, Absolute Discharges and the Notification Requirements—H.M. Advocate v KH’ Criminal Law Bulletin 130, 4–7. —— (2014f) ‘The (Non) Applicability of English Sentencing Guidelines—Neill v H.M. Advocate and Hughes v H.M. Advocate’ Criminal Law Bulletin 131, 5–7. —— (2014g) ‘Case Comment—Muirhead v H.M. Advocate [2014] HCJAC 90’ Criminal Law Bulletin 132, 5–7. —— (2013a) ‘The Discretionary Nature of the Sentence Discount Affirmed—Murray v H.M. Advocate’ Criminal Law Bulletin 121, 6–8. —— (2013b) ‘Sentencing Cases—Intercourse with Girl Under 16’ Criminal Law Bulletin 123, 7–8. —— (2013c) ‘Sentencing Cases—Supply of Heroin—Mitigating Factors Justifying a ­Non-Custodial Disposal’ Criminal Law Bulletin 121, 5–6. —— (2013d) ‘Sentence Discounting in England and Scotland—Some Observations on the Use of Comparative Authority in Sentence Appeals’ Criminal Law Review 674–77. —— (2013e) ‘Towards Sentencing Guidelines for Rape’ Criminal Law Bulletin 125, 5–8. —— (2012a) ‘Sentencing Historic Cases—Greig v H.M. Advocate, Criminal Law Bulletin 120, 1–4. —— (2012b) ‘Sentencing Cases—Theft by Opening Lockfast Places—Theft of Metal from Church’ Criminal Law Bulletin 120, 7–8. —— (2012c) ‘Sentencing Cases—Crown Appeal Against Sentence—Assault to Severe Injury and Permanent Disfigurement’ Criminal Law Bulletin 120, 6–7. —— (2012d) ‘Sentencing Cases—Assault to Severe Injury and Permanent Disfigurement—Level of Discount’ Criminal Law Bulletin 119, 7–8. —— (2011a) ‘Sentencing Cases—Lewd and Libidinous Practices—Judicial Use of Sentencing Statements’ Criminal Law Bulletin 109, 5–6. —— (2011b) ‘Sentencing Cases—Assault and Robbery—Assistance Provided to Authorities on Another Matter—Discount in Sentence of 50 Per Cent’ Criminal Law Bulletin 112, 4–6. —— (2010a) ‘Sentencing Cases—Robertson and Patterson v P.F., Aberdeen’ Criminal Law Bulletin 105, 6–8. —— (2010b) ‘Sentencing Cases—Social Supply of Class A Drugs—Previous Sentencing Policy Affirmed’ Criminal Law Bulletin 106, 7–8. Bryman, A (2012) Social Research Methods, 4th edn (Oxford, Oxford University Press). Campbell, L (2008) ‘Reconstituting Sentencing Policy in the Republic of Ireland’ 59(3) Northern Ireland Legal Quarterly 291–303. Canadian Sentencing Commission (1987) Sentencing Reform: A Canadian Approach (Ottawa, Canadian Sentencing Commission). Cardwell Hughes (1999) Examining Our Commissions & Sentencing Policy, Report from the National Association of Sentencing Commissions 1999 Annual Conference, Salt Lake City, Utah. Carroll, J (2007) ‘Discounting Justice’ Journal of the Law Society of Scotland Online, available at: www.journalonline.co.uk/Magazine/52-12/1004818.aspx (accessed 22 September 2016).

246 

Bibliography

Carloway, Lord (2016) ‘Imprisonment in Scotland: Towards a Penological Post-Modernism?’ Address delivered at the Scottish Universities Insight Institute, 15 March 2016, Dovecot Studios, Edinburgh. —— (2013) ‘Sentencing: Beyond Punishment and Deterrence’ 2013 SACRO lecture, Playfair Library, University of Edinburgh, Edinburgh. —— (2012) The Sifting Process and the Gatekeeping Function of the Judge, Address delivered at the Senators’ Training Course, 6 June 2012, Mackenzie Building, Edinburgh. Carnie, J (1990) Sentencers’ Perceptions of Community Service by Offenders, Scottish Office Central Research Unit Papers, Edinburgh, Scottish Office. Chalmers, J, Duff, P, Leverick, F and Melvin, Y (2007) An Evaluation of the High Court Reforms Arising from the Criminal Procedure (Amendment) (Scotland) Act 2004 (Edinburgh, Scottish Executive Social Research). Chanenson, SL (2005) ‘Guidance from Above and Beyond’ 58 Stanford Law Review 175–94. Charleton, the Hon Mr Justice and Scott, L (2013) ‘Throw Away the Key: Public and Judicial Approaches to Sentencing—Towards Reconciliation’ the 2013 Martin Tansey Memorial Lecture, Association for Criminal Justice Research and Development, Dublin. Charmaz, K (2006) Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis (London, Sage). —— (1994) ‘Grounded Theory: Objectivist and Constructivist Methods’ in N Denzin and Y Lincoln (eds), Handbook of Qualitative Research, 2nd edn (Thousand Oaks, Sage). —— (1983) ‘The Grounded Theory Method: An Explication and Interpretation’ in RM Emerson (ed), Contemporary Field Research: A Collection of Readings (Boston, Little Brown). Chasse, K (2010) ‘Untruth In Sentencing Credit for Pre-Sentence Custody’ 15(1) Canadian Criminal Law Review 75–109. —— (2009) ‘Plea Bargaining Is Sentencing’ 14(1) Canadian Criminal Law Review 55–77. Christie, MGA (ed) (2000) The Criminal Law of Scotland, 3rd edn Volume II (Edinburgh, W Green). Christie, N (2005) ‘Preface’ in KF Aas, Sentencing in the Age of Information—From Faust to Macintosh (London, The Glasshouse Press). Clarke, A, Moran-Ellis, J and Sleny, J (2002) Attitudes to Date Rape and Relationship Rape (Guilford, Department of Sociology, University of Surrey). Cohen-Lask, M (2016) ‘Unbalancing the Burden of Proof ’ 113(8) Law Society Gazette 9. Cole, D and Manson, A (1990) Release from Imprisonment: The Law of Sentencing, Parole and Judicial Review (Toronto, Carswell). Colvin, E (2003) ‘Sentencing Principles in the High Court and the PSA’ 3(1) Queensland University of Technology Law and Justice Journal 86–103. Connolly, Justice T (2006) ‘Human Rights Aspects of Sentencing’ Paper delivered at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra. Cooke, Judge RK (1987) ‘The Practical Problems of the Sentencer’ in Pennington, DC and Lloyd-Bostock, S (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies). Coonan, G (2006) ‘The Role of Judicial Research Assistants in Supporting the Decision-Making Role of the Irish Judiciary’ 6(1) Judicial Studies Institute Journal 171–96. Cooper, G (2008) ‘Conceptualising Social Life’ in N Gilbert (ed), Researching Social Life, 3rd edn (London, Sage). Cooper, J (2013) ‘Nothing Personal—The Impact of Personal Mitigation at Sentencing Since Creation of the Council’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— ‘The Sentencing Guidelines Council—A Practical Perspective’ 4 Criminal Law Review 277–86. Coren, V (2009) For Richer For Poorer—A Love Affair with Poker (Edinburgh, Canongate). Corns, C (1990) ‘Destructuring Sentencing Decision-Making in Victoria’ 23 Australian & New Zealand Journal of Criminology 145–57.

Bibliography

 247

Coulter, D and Wiens, JR (2002) ‘Educational Judgment: Linking the Actor and the Spectator’ 31(4) Educational Researcher 15–25. Cowdery, N (2016) ‘Negotiating Justice with Integrity in New South Wales’ in Hunter, J, Roberts, P, Young, SNM and Dixon, D (eds), The Integrity of Criminal Process—From Theory into Practice (Oxford, Hart Publishing). —— (2006) ‘Guideline Judgments: It Seemed Like A Good Idea At The Time’ Paper presented at the International Society for the Reform of Criminal Law, 2– 6 July 2006, Brisbane. Crilly, B (2005) ‘Guideline Judgments in Victoria: An Examination of the Issues’ 31(1) Monash University Law Review 37–63. Criminal Bar Association of England and Wales (2016) ‘A Sentencing Council Consultation Response: On the Reduction in Sentence for a Guilty Plea Guidelines’ Current Sentencing Practice News 2, 7–9. Croall, H, Mooney, G and Munro, M (2010) ‘Criminal Justice in Contemporary Scotland: Themes, Issues and Questions’ in H Croall, G Mooney and M Munro (eds), Criminal Justice in Scotland (Abingdon, Willan Publishing). Crowder, G (2007) ‘Value Pluralism and Liberalism—Berlin and Beyond’ in G Crowder and H Hardy (eds), The One and the Many—Reading Isaiah Berlin (Amherst, New York, Prometheus Books). —— (2004) Isaiah Berlin—Liberty and Pluralism (Cambridge, Polity Press). —— (2003) ‘Review Article—Hedgehog and Fox’ 38(2) Australian Journal of Political Science 333–39. —— (2002) Liberalism and Value Pluralism (London, Continuum). —— (1994) ‘Pluralism and Liberalism’ 42(2) Political Studies 293–305. Crowe, Sheriff FR (2013a) ‘A Serious Matter’ 58(8) Journal of the Law Society of Scotland 26–27. —— (2013b) ‘Getting in Quick’ 58(2) Journal of the Law Society of Scotland 26–27. Crown Office and Procurator Fiscal Service (2016) Crown Office and Procurator Fiscal Service Statistics on Case Processing—Last Five Years (Edinburgh, Crown Office and Procurator Fiscal Service). Cuneen, C (2002) ‘Mandatory Sentencing and Human Rights’ 13(3) Current Issues in Criminal Justice 322–27. Darbyshire, P (2011) Sitting in Judgment—The Working Lives of Judges (Oxford, Hart Publishing). —— (2000) ‘The Mischief of Plea Bargaining and Sentencing Rewards’ Criminal Law Review 895–910. Davey, E (2014) ‘Sentence Structure’ 108(5) Gazette of the Law Society of Ireland 30–33. Davies, M and Tyrer, J (2003) ‘“Filling in the Gaps”—A Study of Judicial Culture’ Criminal Law Review 243–65. Davis, J (2002) ‘The Science of Sentencing: Measurement Theory and von Hirsch’s New Scales of Justice’ in C Tata and N Hutton (eds), Sentencing and Society—International Perspectives (Aldershot, Ashgate). Dawes, W, Harvey, P, McIntosh, B, Nunney, F and Phillips, A (2011) Attitudes to Guilty Plea Sentence Reductions (Sentencing Council Research Series 02/11) (London, Sentencing Council). Denham, Chief Justice S (2011) ‘Foreword’ in T O’Malley, Sentencing—Towards a Coherent System (Dublin, Round Hall). Dhami, MK (2013a) ‘Sentencing Guidelines in England and Wales: Missed Opportunities?’ 76(1) Law and Contemporary Problems 289–307. —— (2013b), ‘A “Decision Science” Perspective on the Old and New Sentencing Guidelines in England and Wales’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). Dickson, DJ and Vannet, A (2008) Stair Memorial Encyclopaedia, 2nd reissue Criminal Procedure (Edinburgh, Law Society of Scotland/LexisNexis). Dingwall, G (2008) ‘Deserting Desert? Locating the Present Role of Retributivism in the Sentencing of Adult Offenders’ 47(4) The Howard Journal 400–10. —— (1997) ‘The Court of Appeal and “Guideline” Judgments’ 48(2) Northern Ireland Legal Quarterly 143–51.

248 

Bibliography

Doob, AN (2011) ‘The Unfinished Work of the Canadian Sentencing Commission’ 53(5) Canadian Journal of Criminology and Criminal Justice 279–97. Doob, AN and Beaulieu, LA (1992) ‘Variation in the Exercise of Judicial Discretion with Young Offenders’ Canadian Journal of Criminology 35–50. Doob, AN and Webster, CM (2016) ‘Weathering the Storm? Testing Long-Standing Canadian Sentencing Policy in the Twenty-First Century’ 45(1) Crime and Justice 359–418. Douglas, R (1980) ‘Verdict’ in La Trobe University Legal Studies Department, Guilty, Your Worship—A Study of Victoria’s Magistrates’ Courts (Occasional Monograph No 1) (Melbourne, La Trobe University). Dreyfus, H and Dreyfus, S (1986) Mind over Machine: The Power of Human Intuition and Expertise in the Era of the Computer (New York, Free Press, rev 1988). Drewry, G, Blom-Cooper, L and Blake, C (2007) The Court of Appeal (Oxford, Hart Publishing). Duff, Sheriff A (2013) ‘Briefing Note: High Court of Justiciary Appeal Court Decision—Stephen Murray v HM Advocate’ (Edinburgh, Judicial Institute). Duff, RA (2005) ‘Guidance and Guidelines’ 105 Columbia Law Review 1162–89. Duncan, AGM (1992) Green’s Glossary of Scottish Legal Terms, 3rd edn (Edinburgh, W Green). Easton, S and Piper, C (2016) Sentencing and Punishment—The Quest for Justice, 4th edn (Oxford, Oxford University Press). Edney, R (2005a) ‘In Spite of Itself? The High Court and the Development of Australian Sentencing Principles’ 2 University of New England Law Journal 1–29. —— (2005b) ‘Still Plucking Figures Out of the Air?: Markarian and the Affirmation of the Instinctive Synthesis’ 1(2) High Court Quarterly Review 50–57. Edney, R and Bagaric, M (2007) Australian Sentencing—Principles and Practice (Port Melbourne, Cambridge University Press). Edwards, I (2013) ‘Victims, Sentencing Guidelines and the Sentencing Council’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2012) ‘Sentencing Councils and Victims’ 75(3) Modern Law Review 324–46. Farmer, L (2010) ‘The Idea of Principle in Scots Criminal Law’ in J Chalmers, F Leverick and L Farmer (eds), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh, Edinburgh University Press). Farrelly, C and Solum, LB (2008) ‘An Introduction to Aretaic Theories of Law’ in C Farrelly and LB Solum (eds), Virtue Jurisprudence (Basingstoke, Palgrave Macmillan). Fenwick, H (1997) ‘Procedural ‘Rights’ of Victims of Crime: Public or Private Ordering of the Criminal Justice Process?’ 60(3) Modern Law Review 317–33. Ferguson, PR and McDiarmid, C (2014) Scots Criminal Law—A Critical Analysis, 2nd edn (Edinburgh, Edinburgh University Press). Field, D (2002) ‘Plead Early and Convincingly to Avoid Disappointment’ 14(2) Bond Law Review 251–84. Fielding, NG (2011) ‘Judges and Their Work’ 20(1) Social & Legal Studies 97–115. Fielding, N and Thomas, H (2008) ‘Qualitative Interviewing’ in N Gilbert (ed), Researching Social Life, 3rd edn (London, Sage). Findlay, M, Odgers, S and Yeo, S (2014) Australian Criminal Justice, 5th edn (Melbourne, Oxford University Press). Findlay, M, Odgers, S and Yeo, S (2009) Australian Criminal Justice, 4th edn (Melbourne, Oxford University Press). Finnis, J (1980) Natural Law and Natural Rights (Oxford, Clarendon Press). Fitz-Gibbon, K (2016) ‘Minimum Sentencing for Murder in England and Wales: A Critical Examination 10 Years After the Criminal Justice Act 2003’ 18(1) Punishment & Society 47–67.

Bibliography

 249

—— (2014) ‘Overcoming Barriers in the Criminal Justice System: Examining the Value and Challenges of Interviewing Legal Practitioners’ in K Lumsden and A Winter (eds), Reflexivity in Criminological Research—Experiences with the Powerful and the Powerless (Basingstoke, Palgrave MacMillan). —— (2013) ‘The Mandatory Life Sentence for Murder: An Argument for Judicial Discretion in England’ 13(5) Criminology & Criminal Justice 506–25. Flynn, A and Fitz-Gibbon, K (2011) ‘Bargaining with Defensive Homicide: Examining Victoria’s Secretive Plea Bargaining System Post-Law Reform’ 35(3) Melbourne University Law Review 905–32. Flyvbjerg, B (2006) ‘Making Organization Research Matter: Power, Values and Phronesis’ in SR Clegg, C Hardy, TB Lawrence and WR Nord (eds), The Sage Handbook of Organization Studies, 2nd edn (London, Sage). —— (2001) Making Social Science Matter—Why Social Inquiry Fails and How it Can Succeed Again (Cambridge, Cambridge University Press). Flyvbjerg, B, Landman, T and Schram, S (2012a) ‘Introduction: New Directions in Social Science’ in B Flyvbjerg, T Landman and S Schram (eds), Real Social Science—Applied Phronesis (Cambridge, Cambridge University Press). Flyvbjerg, B, Landman, T and Schram, S (2012b) ‘Important Next Steps in Phronetic Social Science’ in Flyvbjerg, Landman and Schram, ibid. Foddy, W (1993) Constructing Questions for Interviews and Questionnaires (Cambridge, Cambridge University Press). Fox, RG (2006) ‘Sentencing in the Garden of Eden’ 32(1) Monash University Law Review 4–27. —— (1999) ‘When Justice Sheds a Tear: The Place of Mercy in Sentencing’ 25 Monash University Law Review 1–28. —— (1987) ‘Controlling Sentencers’ 20(4) Australian and New Zealand Journal of Criminology 218–46. Fox, RG and Freiberg, A (1999) Sentencing—State and Federal Law in Victoria, 2nd edn (Melbourne, Oxford University Press). Frank, AW (2012) ‘The Feel for Power Games: Everyday Phronesis and Social Theory’ in Flyvbjerg, Landman and Schram, ibid. Frankel, M (1972), ‘Lawlessness in Sentencing’ 41(1) Cincinnati Law Review in A von Hirsch, A Ashworth and JV Roberts (eds), Principled Sentencing—Readings on Theory and Policy, 3rd edn (Oxford, Hart Publishing). Frederick, S (2005) ‘Cognitive Reflection and Decision Making’ 19(4) The Journal of Economic Perspectives 25–42. Freed, D (1992) ‘Federal Sentencing in the Wake of Guidelines: Unacceptable Limits on the Discretion of Sentencers’ 101(8) Yale Law Journal 1681–1754. Freer, E (2013) ‘First Time Lucky? Exploring Whether First-time Offenders Should be Sentenced More Leniently’ 77(2) The Journal of Criminal Law 163–71. Freiberg, A (2016) ‘The Road Well Traveled in Australia: Ignoring the Past, Condemning the Future’ 45(1) Crime and Justice 419–57. —— (2014) Fox & Freiberg’s Sentencing—State and Federal Law in Victoria, 3rd edn (Pyrmont, Thomson Reuters). —— (2008) ‘The Victorian Sentencing Advisory Council: Incorporating Community Views into the Sentencing Process’ in A Freiberg and K Gelb (eds), Penal Populism, Sentencing Councils and Sentencing Policy (Cullompton, Willan Publishing). —— (2006) ‘Twenty Years of Changes in the Sentencing Environment and Courts’ Responses’ Paper delivered at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra. —— (2003) ‘The Four Pillars of Justice: A Review Essay’ 36(2) Australian and New Zealand Journal of Criminology 223–30.

250 

Bibliography

—— (2001) ‘Three Strikes and You’re Out—It’s Not Cricket: Colonization and Resistance in Australian Sentencing’ in M Tonry and RS Frase (eds), Sentencing and Sanctions in Western Countries (Oxford, Oxford University Press). —— (1995) ‘Sentencing Reform in Victoria: A Case-Study’ in CMV Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Oxford, Clarendon Press). Freiberg, A and Krasnostein, S (2011) ‘Statistics, Damn Statistics, and Sentencing’ 21(2) Journal of Judicial Administration 73–92. Freiberg, A and Ross, S (1999) Sentencing Reform and Penal Change—The Victorian Experience (Sydney, The Federation Press). Fuller, LL (1969) The Morality of Law, revised edn (New Haven, Yale University Press). Galston, W (1999) ‘Expressive Liberty, Moral Pluralism, Political Pluralism: Three Sources of Liberal Theory’ 40 William & Mary Law Review 869–907. Gardiner, P (1997) ‘Introduction’ in H Hardy (ed), The Sense of Reality—Studies in Ideas and their History (London, Pimlico). Garland, D (1999a) ‘Preface’ in P Duff and N Hutton (eds), Criminal Justice in Scotland (Aldershot, Dartmouth). —— (1999b) ‘Penal Modernism and Postmodernism’ in R Matthews (ed), Imprisonment (Aldershot, Dartmouth). Gazal-Ayal, O (2013) ‘Foreword—A Global Perspective on Sentencing Reforms’ Law and Contemporary Problems 1, i–viii. Gerber, P (2003) ‘When Is Plea Bargaining Justified? 3(1) Queensland University of Technology Law and Justice Journal 210–15. Gerry, F, Sjölin, C and Harris, L (2014) Sexual Offences Handbook—Law, Practice and Procedure, 2nd edn (London, Wildy, Simmonds & Hill Publishing). Giddens, A (1984) The Constitution of Society—Outline of the Theory of Structuration (Oxford, Polity Press). Gifford, T (1986) Where’s the Justice?—A Manifesto for Law Reform (Harmondsworth, Penguin). Gigerenzer, G and Todd, PM (1999) ‘Fast and Frugal Heuristics: The Adaptive Toolbox’ in G Gigerenzer, PM Todd and the ABC Research Group (eds), Simple Heuristics That Make Us Smart (New York, Oxford University Press). Gill, Lord (2016) ‘The Scottish Legal System—What Next?’ The 2016 Judicial Institute for Scotland Lecture, 14 July 2016, Parliament Hall, Edinburgh. Glaser, BG and Strauss, AL (1967) The Discovery of Grounded Theory: Strategies for Qualitative Research (Chicago, Aldine). Gordon, Sir GH (2013) ‘Commentary to Murray v H.M. Advocate, 2013 SCCR 88’ Scottish Criminal Case Reports, 95. —— (2012) ‘Commentary to Gemmell v H.M. Advocate, 2012 SCCR 176’ Scottish Criminal Case Reports, 220–21. —— (2011a) ‘Commentary to Petch v H.M. Advocate, 2011 SCCR 360’ Scottish Criminal Case Reports, 368. —— (2011b) ‘Commentary to Petch and Foye v H.M. Advocate, 2011 SCCR 199’ Scottish Criminal Case Reports, 239–40. —— (2000) Valedictory Lucky Dip, Address delivered to the Sheriffs’ Association, 5 March 2000, Peebles. Gordon, Sir GH, Gane, C and Chalmers, J (2016), Criminal Procedure According to the Law of Scotland, 6th edn (Edinburgh, W Green). Goring, R (2008) ‘Life and Liberty’ (University of Aberdeen Magazine (December)) 26–29. Green, Sir G (1996) ‘The Concept of Uniformity in Sentencing’ 70 Australian Law Journal 112–24. Griggs, S and Howarth, D (2012), ‘Phronesis and Critical Policy Analysis: Heathrow’s “Third Runway” and the Politics of Sustainable Aviation in the United Kingdom’ in Flyvbjerg, B, Landman, T and Schram, S (eds), Real Social Science—Applied Phronesis (Cambridge, Cambridge University Press).

Bibliography

 251

Guthrie, C, Rachlinski, JJ and Wistrich, AJ (2007) ‘Blinking on the Bench: How Judges Decide Cases’ 93(1) Cornell Law Review 1–43. Hall, G (2007) ‘Sentencing’ in J Tolmie and W Brookbanks (eds), Criminal Justice in New Zealand (Wellington, LexisNexis New Zealand). Hammond, The Hon Justice G (2007) ‘Sentencing: Intuitive Synthesis or Structured Discretion?’ New Zealand Law Review 211–34. Hampton, J (1988) ‘The Retributive Idea’ in JG Murphy and J Hampton J, Forgiveness and Mercy (New York, Cambridge University Press). Harris, L (2016) ‘Commentary to Attorney General’s Reference (Nos. 49 and 50 of 2015) (R v Bakht)’ Criminal Law Review 72, 73. —— (2015) ‘Commentary to R v Buchanan’ Criminal Law Review 551, 552–53. —— (2013) ‘The New Sentencing Council Sexual Offences Definitive Guideline’ 177(52) Criminal Law & Justice Weekly 847–48. Harris, L and Gerry, F (2013) ‘A Behaviour-Based Approach to Sentencing for Sexual Offences’ 177(15) Criminal Law & Justice Weekly 239–42. Harris, R (2016) ‘Could the Guilty Plea Write Out the Role of the Defence?’ 160 Solicitors Journal 15. Harrison, R (1992) ‘The Equality of Mercy’ H in Gross and R Harrison (eds), Jurisprudence— Cambridge Essays (Oxford, Clarendon Press). Harvey, L and Pease, K (1987) ‘Guideline Judgments and Proportionality in Sentencing’ Criminal Law Review 96–104. Hausheer, R (2013) ‘Introduction’ in H Hardy and R Hausheer (eds), The Proper Study of Mankind— An Anthology of Essays, 2nd edn (London, Vintage Books). Healy, P (2013) ‘Sentencing from There to Here and from Then to Now’ 17(3) Canadian Criminal Law Review 291–304. —— (2005) ‘The Effect of Pre-Sentence Custody on Eligibility for a Conditional Sentence’ 9(2) Canadian Criminal Law Review 261–71. Henham, R (2012) Sentencing and the Legitimacy of Trial Justice (Abingdon, Routledge. —— (2002) ‘Sentencing Policy and Guilty Plea Discounts’ in C Tata and N Hutton (eds), Sentencing and Society—International Perspectives (Aldershot, Ashgate). —— (2001) Sentence Discounts and the Criminal Process (Aldershot, Ashgate). —— (1999) ‘Bargain Justice or Justice Denied? Sentence Discounts and the Criminal Process’ 62 Modern Law Review 515–38. —— (1995a) ‘Sentencing Policy and the Role of the Court of Appeal’ 34(3) The Howard Journal 218–27. —— (1995b) ‘Due Process, Procedural Justice and Sentencing Policy’ 23 International Journal of the Sociology of Law 233–52. Herz, R (2012) The Art of Justice—The Judge’s Perspective (Oxford, Hart Publishing). Hewton, T (2010) ‘Instinctive Synthesis, Structured Reasoning, and Punishment Guidelines: Judicial Discretion in the Modern Sentencing Process’ 31(1) The Adelaide Law Review 79–93. Hodkinson, P (2008) ‘Grounded Theory and Inductive Research’ in N Gilbert (ed), Researching Social Life, 3rd edn (London, Sage). Hogarth, J (1971) Sentencing as a Human Process (Toronto, University of Toronto Press (in association with the Centre of Criminology, University of Toronto)). Hogarth, RM (2001) Educating Intuition, 2nd edn (Chicago, University of Chicago Press). Hood, R (1992) Race and Sentencing (Oxford, Clarendon Press). Hooper, Lord Justice (2012) ‘A Valedictory Summing Up’, Lecture delivered at the Honourable Society of the Inner Temple, 18 June 2012, the Honourable Society of the Inner Temple, London. Hope of Craighead, Lord (2010) ‘Foreword to the Second Edition’ in K Norrie (ed), 100 Cases That Every Scots Law Student Needs to Know, 2nd edn (Edinburgh, W Green). Hough, M, Jacobson, J and Millie, A (2003) The Decision to Imprison—Sentencing and the Prison Population (London, Prison Reform Trust).

252 

Bibliography

Hudson, B (1995) ‘Beyond Proportionate Punishment: Difficult Cases and the 1991 Criminal Justice Act’ 22 Crime, Law & Social Change 59–78. Hume, D (1844) Commentaries on the Law of Scotland Respecting Crimes (with a Supplement by Benjamin Robert Bell, Esq) Vol I (1986 Reprint) (Edinburgh, Law Society of Scotland). Hussain, N and Sarat, A (2007) ‘Toward New Theoretical Perspectives on Forgiveness, Mercy, and Clemency: An Introduction’ in N Hussain and A Sarat (eds), Forgiveness, Mercy, and Clemency (Stanford, Stanford University Press). Hutton, N (2016a) ‘Visible and Invisible Sentencing’ in A Hondeghem, X Rousseaux and F Schoenaers (eds), Modernisation of the Criminal Justice Chain and the Judicial System—New Insights on Trust, Cooperation and Human Capital (Ius Gentium, Comparative Perspectives on Law and Justice, Volume 50) (Cham, Switzerland, Springer International). —— (2016b) ‘Scottish Sentencing Council: A New Era’ Current Sentencing Practice—News 1, 10–11. —— (2013a) ‘The Definitive Guideline on Assault Offences—The Performance of Justice’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2013b) ‘From Intuition to Database: Translating Justice’ 17(1) Theoretical Criminology 109–28. —— (2008) ‘The Sentencing Commission for Scotland’ in A Freiberg and K Gelb (eds), Penal Populism, Sentencing Councils and Sentencing Policy (Cullompton, Willan Publishing). —— (2006a) ‘Sentencing as a Social Practice’ in S Armstrong and L McAra (eds), Perspectives on Punishment—The Contours of Control (Oxford, Oxford University Press). —— (2006b) ‘Book Review—Katja Franko Aas, Sentencing in the Age of Information: From Faust to Macintosh’ 6(2) Criminology & Criminal Justice 259–60. —— (2003) ‘Towards a Sentencing Policy for the Use of Short Prison Sentences in Scotland’ Juridical Review 313–29. —— (2002) ‘Sentencing, Inequality and Justice’ in C Tata and N Hutton (eds), Sentencing and Society—International Perspectives (Aldershot, Ashgate). —— (1999) ‘Sentencing in Scotland’ in P Duff and N Hutton (eds), Criminal Justice in Scotland (Aldershot, Dartmouth). —— (1995) ‘Sentencing, Rationality, and Computer Technology’ 22(4) Journal of Law and Society 549–70. Hutton, N and McNeill, F (2010) Criminal Justice and Licensing (Scotland) Bill—Stage 2: Written Submission (Centre for Sentencing Research, University of Strathclyde and Scottish Centre for Crime and Justice Research, Glasgow, University of Glasgow). Hutton, N and Tata, C (2010) ‘A Sentencing Exception? Changing Sentencing Policy in Scotland’ 22(4) Federal Sentencing Reporter 272–78. —— and —— (2000) ‘The Judicial Role in the “Balance” Between Two Visions of Justice in Sentencing’ in S Doran and JD Jackson (eds), The Judicial Role in Criminal Proceedings (Oxford, Hart Publishing). —— and —— (1993), ‘Reform of Sentencing Practice’ 38(4) Journal of the Law Society of Scotland 230–32. Hyde, M (1998) Trickster Makes This World (New York, North Point Press). Irwin-Rogers, K and Perry, TW (2015) ‘Exploring the Impact of Sentencing Factors on Sentencing Domestic Burglary’ in JV Roberts (ed), Exploring Sentencing Practice in England and Wales (Basingstoke, Palgrave Macmillan). Jacobson, J and Hough, M (2011) ‘Personal Mitigation: an Empirical Analysis in England’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— and —— (2007) Mitigation: The Role of Personal Factors in Sentencing (London, Prison Reform Trust).

Bibliography

 253

Jareborg, N (1995) ‘The Swedish Sentencing Reform’ in CMV Clarkson and R Morgan (eds), The Politics of Sentencing Reform (Oxford, Clarendon Press). Jina, R and Thomas, LS (2013) ‘Health Consequences of Sexual Violence Against Women’ 27 Best Practice & Research Clinical Obstetrics and Gynaecology 15–26. Judge, Lord I (2015) ‘The Sentencing Decision’ in I Judge, The Safest Shield—Lectures, Speeches and Essays (Oxford, Hart Publishing). —— (2014) ‘Publication Review—Sentencing Guidelines: Exploring the English Model’ The Law Quarterly Review 130, 678–81. Judge, Sir I (2007) Current Sentencing Issues, Address delivered at Lincoln’s Inn, 29 October 2007, London. Judicial Institute for Scotland (2016) Annual Report 2015/16 (Edinburgh, Judicial Institute for Scotland). JUSTICE (1993) Negotiated Justice—A Closer Look at the Implications of Plea Bargains (London, JUSTICE). Karlton, Judge LK (1991) ‘Letter to the Editors of the Federal Sentencing Reporter’ (20 November 1991) 4(2) Federal Sentencing Reporter 186–87. Kekes, J (1993) The Morality of Pluralism (Princeton, Princeton University Press). Kelly, D (1993) Criminal Sentences (Edinburgh, T & T Clark). Kennedy, C (1994) ‘Annotation to R v Ollenberger, 29 Criminal Reports (4th Series) 166. King, NJ (2005) ‘Judicial Oversight of Negotiated Sentences In A World of Bargained Punishment’ 58 Stanford Law Review 293–308. King, RD and Wincup, E (2008) ‘The Process of Criminological Research’ in RD King and E Wincup (eds), Doing Research on Crime and Justice, 2nd edn (Oxford, Oxford University Press). Kirby, Justice M (1999) ‘Judging: Reflections on the Moment of Decision’ 18 Australian Bar Review 4. Koffman, L (2006) ‘The Rise and Fall of Proportionality: the Failure of the Criminal Justice Act 1991’ Criminal Law Review 281–99. Korthagen, FAJ and Kessels, JPAM (1999) ‘Linking Theory and Practice: Changing the Pedagogy of Teacher Education’ 28(4) Educational Researcher 4–17. Krasnostein, S (2015a) Pursuing Consistency—The Effect of Different Reforms on Unjustified Disparity in Individualised Sentencing Frameworks (PhD thesis, Melbourne, Monash University). —— (2015b) ‘Boulton v The Queen: The Resurrection of Guideline Judgments in Australia?’ 27(1) Current Issues in Criminal Justice 41–58. Krasnostein, S and Freiberg, A (2014) ‘Sentencing Guideline Schemes Across the United States and Beyond’ in Oxford Handbooks Online: Criminology and Criminal Justice (Oxford, Oxford University Press). —— and —— (2013) ‘Pursuing Consistency in an Individualistic Sentencing Framework: If You Know Where You’re Going, How Do You Know When You’ve Got There?’ 76(1) Law and Contemporary Problems 265–88. Kronman, AT (1987) ‘Living in the Law’ 54 University of Chicago Law Review 835–76. Lacelle, L (1996) ‘Curing Disparity: The Placebo Approach of Starting-Point Sentencing’ 47 Criminal Reports (Articles) (4th Series) 204. Lacey, M (2002) ‘Justice, Humanity and Mercy’ in D Ward, J Scott and M Lacey (eds), Probation— Working for Justice, 2nd edn (Oxford, Oxford University Press). Lacey, N (1998) Unspeakable Subjects—Feminist Essays in Legal and Social Theory (Oxford, Hart Publishing). Landman, T (2012) ‘Phronesis and Narrative Analysis’ in B Flyvbjerg, T Landman and S Schram (eds), Real Social Science—Applied Phronesis (Cambridge, Cambridge University Press). Lane, D (1980) ‘Case Notes—R v Shannon: What Every Defendant Should Know’ 7(1) Monash University Law Review 52–59.

254 

Bibliography

Laws, Lord Justice (2004) ‘The Future of Sentencing: A Perspective from the Judiciary’ in P Sedgwick (ed), Rethinking Sentencing—A Contribution to the Debate (London, Church House Publishing). Leader-Elliot, I (2002) ‘Instinctive Synthesisers in the High Court’ 26(5) Criminal Law Journal 5–9. Leverick, F (2014) ‘Sentence Discounting for Guilty Pleas: An Argument for Certainty Over Discretion’ Criminal Law Review 338–49. —— (2013) ‘The Rise and Fall of the Sentence Discount’ Scots Law Times (News) 38, 259–64. —— (2012) ‘Sentence Discounting for Guilty Pleas: A Question of Guidelines’ 16(2) Edinburgh Law Review 233–38. —— (2010) ‘Plea Bargaining in Scotland: The Rise of Managerialism and the Fall of Due Process’ in SC Thaman (ed), World Plea Bargaining—Consensual Procedures and the Avoidance of the Full Criminal Trial (Durham, Carolina Academic Press). —— (2008a), ‘Sentence Following a Guilty Plea: Spence v HM Advocate and Leonard v Houston’, Scots Law Times (News) 43–47. —— (2008b) ‘Sentencing Guidelines under Section 118(7): Lin v HM Advocate and Spence v HM Advocate’ 12(2) Edinburgh Law Review 307–12. —— (2004) ‘Tensions and Balances, Costs and Rewards: The Sentence Discount in Scotland’ 8 Edinburgh Law Review 360–88. —— (2003) ‘Making Sense of Sentence Discounting: Du Plooy v HM Advocate’ Scots Law Times (News) 267–71. Lloyd-Bostock, S (1988) Law in Practice: Applications of Psychology to Legal Decision Making and Legal Skills (Leicester, British Psychological Society in association with Routledge). Lofland, J and Lofland, LH (1995) Analyzing Social Settings: A Guide to Qualitative Observation and Analysis, 3rd edn (London, Wadsworth). Lothian, A (2006) ‘Guiding Hand’ 51(11) Journal of the Law Society of Scotland 18–19. —— (2004) ‘On Sentencing’ 49(2) Journal of the Law Society of Scotland 44–45. Lovegrove, A (2011) ‘The Pernicious Impact of Perceived Public Opinion on Sentencing: Findings from an Empirical Study of the Public’s Approach to Personal Mitigation’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2010) ‘The Sentencing Council, the Public’s Sense of Justice and Personal Mitigation’ Criminal Law Review 906–23. —— (2002) ‘Intuition, Structure and Sentencing: An Evaluation of Guideline Judgments’ 14(2) Current Issues in Criminal Justice 182–204. —— (1987) ‘An Evaluation of Judicial Models for Sentencing Guidelines’ in DC Pennington and S Lloyd-Bostock (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies). Lynch, D (1994) ‘The Impropriety of Plea Agreements: A Tale of Two Countries’ 19 Law and Social Inquiry 115–33. Mack, K and Roach Anleu, S (1998) ‘Reform of Pre-Trial Criminal Procedure: Guilty Pleas’ 22 Criminal Law Journal 263–76. —— and —— (1997) ‘Sentence Discount for a Guilty Plea: Time for a New Look’ 1(2) Flinders Journal of Law Reform 123–43. —— and —— (1995) ‘Balancing Principle and Pragmatism: Guilty Pleas’ 4 Journal of Judicial Administration 232–39. Mackenzie, G (2006a) ‘Judges’ Attitudes and Perceptions Toward the Sentencing Process’, paper presented at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra. —— (2006b) ‘Consistency in Sentencing and Discounts for Guilty Pleas’, paper presented at the International Conference of the International Society for the Reform of Criminal Law, 2–6 July 2006, Brisbane.

Bibliography

 255

—— (2005) How Judges Sentence (Sydney, The Federation Press). —— (2003) ‘The Art of Balancing—Queensland Judges and the Sentencing Process’ 28(6) Alternative Law Journal 288–91. —— (2002) ‘Achieving Consistency in Sentencing: Moving to Best Practice?’ 22 University of Queensland Law Journal 74–90. —— (2001) A Question of Balance: A Study of Judicial Methodology, Perceptions and Attitudes in Sentencing (PhD thesis, Sydney, University of New South Wales). Mackenzie, G and Stobbs, N (2010) Principles of Sentencing (Sydney, The Federation Press). Maguire, N (2010) ‘Consistency in Sentencing’ Judicial Studies Institute Journal 2, 14–54. Maher, G (1998) ‘Crown Appeals Against Sentence in Scotland’ Criminal Law Review 854–63. —— (1988), ‘Compensation Orders in the Scottish Criminal Courts’ in M Adler and A Millar (eds), Socio-Legal Research in the Scottish Courts: Papers Presented at a Conference Held on 23 March 1987 at the University of Edinburgh, Scottish Office Central Research Unit Papers, Edinburgh: Scottish Office. Maher, G, Docherty, CJ, McLean, SAM and Campbell, TD (1988) Compensation Orders in the Scottish Criminal Courts, Scottish Office Central Research Unit Papers, Edinburgh: Scottish Office. Malleson, K (1999) The New Judiciary—The Effects of Expansion and Activism (Aldershot, Ashgate). Manson, A (2004) ‘Pre-Sentence Custody and the Determination of a Sentence (Or How to Make a Mole Hill Out of a Mountain)’ 49 Criminal Law Quarterly 292–350. —— (2001) The Law of Sentencing (Toronto, Irwin Law). —— (1997) ‘McDonnell and the Methodology of Sentencing’ Criminal Reports (5th Series) 277–92. Manson, A, Healy, P, Trotter, G, Roberts, J and Ives, D (2016) Sentencing and Penal Policy in Canada— Cases, Materials and Commentary, 3rd edn (Toronto, Emond Montgomery Publications). Manson, A, Healy, P, Trotter, G, Roberts, J and Ives, D (2008) Sentencing and Penal Policy in Canada— Cases, Materials and Commentary, 2nd edn (Toronto, Emond Montgomery Publications). Marr, A (2013) ‘Foreword’ in H Hardy and R Hausheer (eds), The Proper Study of Mankind—An Anthology of Essays, 2nd edn (London, Vintage Books). Maslen, H (2015) ‘Penitence and Persistence: How Should Sentencing Factors Interact?’ in JV Roberts (ed), Exploring Sentencing Practice in England and Wales (Basingstoke, Palgrave Macmillan). Maslen, H and Roberts, JV (2013) ‘Remorse and Sentencing—An Analysis of Sentencing Guidelines and Sentencing Practice’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). Mason, F and Murphy, P (2016) ‘Psychological Effects of Rape and Serious Sexual Assault’ in Judge P Rook and R Ward, Rook and Ward on Sexual Offences Law and Practice, 5th edn (London, Sweet and Maxwell). McAra, L (2008) ‘Crime, Criminology and Criminal Justice in Scotland’ 5(4) European Journal of Criminology 481–504. McCallum, F (2011) ‘SPICe Briefing—Knife Crime’ (Edinburgh, Scottish Parliament Information Centre). McClellan, Justice P (2012) ‘Sentencing in the 21st Century’, paper presented at the Crown Prosecutors’ Conference, 10 April 2012, Pokolbin, Hunter Valley. McCluskey, Lord and McBride, P (2000) Criminal Appeals, 2nd edn (Edinburgh, Butterworths). McConville, M (2002) ‘Plea Bargaining’ in M McConville and G Wilson (eds), The Handbook of The Criminal Justice Process (Oxford, Oxford University Press). —— (1998) ‘Plea Bargaining: Ethics and Politics’ 25(4) Journal of Law and Society 562–87. McConville, M and Bridges, L (1993) ‘Pleading Guilty Whilst Maintaining Innocence’ New Law Journal 160–61. McConville, M, Hodgson, J, Bridges, L and Pavlovic, A (1994) Standing Accused—The Organisation and Practices of Criminal Defence Lawyers in Britain (Oxford, Clarendon Press).

256 

Bibliography

McConville, M and Marsh, L (2016) ‘Factory Farming and State-Induced Guilty Pleas’ in J Hunter, P Roberts, SNM Young and D Dixon (eds), The Integrity of Criminal Process—From Theory into Practice (Oxford, Hart Publishing). —— and —— (2014) Criminal Judges—Legitimacy, Courts and State-Induced Guilty Pleas in Britain (Cheltenham, Edward Elgar). McCormick, P and Greene, I (1990) Judges and Judging (Toronto, James Lorimer and Company). McNeill, F, Burns, N, Halliday, S, Hutton, N and Tata, C (2009) ‘Risk, Responsibility and Reconfiguration’ 11(4) Punishment & Society 419–42. McPherson, R (2013) ‘The Rise of Agreed Narratives in Scottish Criminal Procedure’ 2 Juridical Review 141–56. Meyer, LR (2007) ‘The Merciful State’ in N Hussain and A Sarat (eds), Forgiveness, Mercy, and Clemency (Stanford, Stanford University Press). Mildren, Justice D (2006) ‘Intuitive Synthesis or the Structured Approach?’ 2(1) International Journal of Punishment and Sentencing 1–7 (text of the paper delivered by his Honour at the conference Sentencing—Principles, Perspectives & Possibilities, 10–12 February 2006, Canberra). Miller, ML and Wright, RF (2005) ‘“The Wisdom We Have Lost”: Sentencing Information and its Uses’ 58 Stanford Law Review 361–80. Millie, A, Tombs, J, and Hough, M (2007) ‘Borderline Sentencing: A Comparison of Sentencers’ Decision Making in England and Wales, and Scotland’ 7(3) Criminology and Criminal Justice 243–67. Moody, S (1995) ‘Images of Women: Sentencing in Sexual Assault Cases in Scotland’ in J Bridgeman and S Millns (eds), Law and Body Politics—Regulating the Female Body (Aldershot, Dartmouth). Moody, S and Tombs, J (1982) Prosecution in the Public Interest (Edinburgh, Scottish Academic Press). Morgan, N (2002a) ‘Why We Should Not Have Mandatory Penalties: Theoretical Structures and Political Realities’ 23 Adelaide Law Review 141–54. —— (2002b) ‘Going Overboard? Debates and Developments in Mandatory Sentencing, June 2000 to June 2002’ 26 Criminal Law Journal 293–311. —— (2000) ‘Mandatory Sentences in Australia: Where Have We Been and Where Are We Going?’ 24 Criminal Law Journal 164–83. —— (1999) ‘Capturing Crims or Capturing Votes? The Aims and Effects of Mandatories’ 22 University of New South Wales Law Journal 267–79. —— (1994) ‘Case and Comment: Sentences of Imprisonment—Instinctive Synthesis or Two-tiered Approach—Punch’ 18 Criminal Law Journal 53–55. Morgan, N and Murray, B (1999) ‘What’s in a Name? Guideline Judgments in Australia’ 23 Criminal Law Journal 90–107. Mulcahy, A (1994) ‘The Justifications of ‘Justice’—Legal Practitioners’ Accounts of Negotiated Case Settlements in Magistrates’ Courts’ 34(4) British Journal of Criminology 411–30. Muller, EL (1993) ‘The Virtue of Mercy in Criminal Sentencing’ 24 Seton Hall Law Review 288–346. Munro, M, Mooney, G and Croall, H (2010) ‘Criminal Justice in Scotland: Overview and Prospects’ in H Croall, G Mooney and M Munro (eds), Criminal Justice in Scotland (Abingdon, Willan Publishing). Murphy, JG (1988) ‘Mercy and Legal Justice’ in JG Murphy and J Hampton, Forgiveness and Mercy (New York, Cambridge University Press). Murray, L (2002) The Pleader (Edinburgh, Mainstream Publishing). Nesterchuk, T (2013) ‘The View from Behind the Bench: The Role of Judicial Assistants in the UK Supreme Court’ in A Burrows, D Johnston and R Zimmermann (eds), Judge and Jurist—Essays in Memory of Lord Rodger of Earlsferry (Oxford, Oxford University Press). Neuborne, B (1992) ‘Of Sausage Factories and Syllogism Machines: Formalism, Realism, and Exclusionary Selection Techniques’ 67 New York University Law Review 419–49.

Bibliography

 257

Nicholls, C, Mitchell, M, Simpson, I, Webster, S and Hester, M (2012) Attitudes to Sentencing Sexual Offences (Sentencing Council Research Series 01/12) (London, Sentencing Council). Nicholson, CGB (1992) Sentencing—Law and Practice in Scotland, 2nd edn (Edinburgh, W Green). —— (1981) The Law and Practice of Sentencing in Scotland (Edinburgh, W Green). Nicholson, Judge J (2010) ‘Sentencing in So-Called “Technical Rape” Cases’, Paper presented at the National Judicial College of Australia’s Sentencing Conference, 6–7 February 2010, Canberra. Nicholson, P (2016) ‘Changing Face of the Courts’ 61(9) Journal of the Law Society of Scotland 12–14. —— (2015) ‘Sentences by the Book’ 60(12) Journal of the Law Society of Scotland 16–18. Nussbaum, MC (2001) The Fragility of Goodness—Luck and Ethics in Greek Tragedy and Philosophy (Cambridge, Cambridge University Press). —— (1993a) ‘Equity and Mercy’ 22(2) Philosophy & Public Affairs 83–125. —— (1993b) ‘Non-Relative Virtues: An Aristotelian Approach’ in MC Nussbaum and A Sen (eds), The Quality of Life (Oxford, Oxford University Press). —— (1990) ‘The Discernment of Perception: An Aristotelian Conception of Private and Public Rationality’ in MC Nussbaum, Love’s Knowledge—Essays on Philosophy and Literature (New York, Oxford University Press). O’Donnell, I (2012) ‘Publication Review—Sentencing: Towards a Coherent System (Thomas O’Malley)’ 47 Irish Jurist 255–57. O’Malley, T (2016) Sentencing Law and Practice, 3rd edn (Dublin, Round Hall). —— (2013a) ‘Living Without Guidelines’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2013b) Sexual Offences, 2nd edn (Dublin, Round Hall). —— (2011) Sentencing—Towards a Coherent System (Dublin, Round Hall). —— (2008a) ‘Principles of Sentencing: Towards a European Conversation’, Paper delivered at the conference ‘The Limits of the Criminal Law’, 23 January 2008, Leiden University, Leiden. —— (2008b) ‘Sentencing Recidivist Sex Offenders: A Challenge for Proportionality’, Paper delivered at the conference ‘Recent Developments in Criminal Law’, December 2008, Trinity College, Dublin. —— (2009) The Criminal Process (Dublin, Round Hall). —— (2006) Sentencing Law and Practice, 2nd edn (Dublin, Round Hall). —— (2003) ‘Sentencing Values and Sentencing Structures’ 3(1) Judicial Studies Institute Journal 130–55. —— (2001) ‘Principled Discretion: Towards the Development of a Sentencing Canon’ 7(3) The Bar Review 135–39; 156–59. —— (2000) Sentencing Law and Practice (Dublin, Round Hall). —— (1994) ‘Resisting the Temptation of Elegance: Sentencing Discretion Re-affirmed’ 4(1) Irish Criminal Law Journal 1–30. O’Rourke, S (2016) ‘The Launch of Scotland’s New Sentencing Council’ Scottish Legal News Annual Review 2016, 22. —— (2014) Glossary of Legal Terms, 6th edn (Edinburgh, W Green). Odgers, SJ (2012) ‘The High Court On Crime in 2011: Analysis and Jurisprudence’ 36 Criminal Law Journal 5–18. Odgers, SJ and Yeo, SMH (2005) ‘Editorial—Sentence Discounts for Pleas of Guilty’ 29 Criminal Law Journal 69–70. Orr, N (2007) ‘Du Plooy Deployed’ Scots Law Times (News) 21, 143–50. Padfield, N (2016) ‘Editorial—Guidelines Galore’ Criminal Law Review 301–02. —— (2013) ‘Exploring the Success of Sentencing Guidelines’, in A Ashworth, and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2011) ‘Intoxication as a Sentencing Factor: Mitigation or Aggravation?’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press).

258 

Bibliography

—— (2003) ‘Sentencing Fundamentals and the Impact of Europe’ 9 The Scottish Journal of Criminal Justice Studies 30–40. Pannick, D (1987) Judges (Oxford, Oxford University Press). Partington, M (2005) ‘Research’ in B Opeskin and D Weisbrot (eds), The Promise of Law Reform (Sydney, The Federation Press). Paterson, A (2013) Final Judgment—The Last Law Lords and the Supreme Court (Oxford, Hart Publishing). Payne-James, J, Jones, R, Karch, SB, and Manlove, J (2011) Simpson’s Forensic Medicine, 13th edn (London, Hodder Arnold). Pease, K (1987) ‘Psychology and Sentencing’ in DC Pennington, and S Lloyd-Bostock (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies). Pennington, DC and Lloyd-Bostock, S (1987) ‘Introduction: Issues and Problems in the Psychology of Sentencing’ in DC Pennington and S Lloyd-Bostock (eds), The Psychology of Sentencing— Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies). Perry, R (2007) ‘Book Review: Sentencing in the Age of Information: From Faust to Macintosh’ 9(4) Punishment and Society 421–24. Phillips, K (2004) ‘Sentencing Cases—Breach of the Peace—Sexual Offences—Teacher—Robson v H.M. Advocate’ Criminal Law Bulletin 67 (February) 6. Pina-Sánchez, J (2015) ‘Defining and Measuring Consistency in Sentencing’ in JV Roberts (ed), Exploring Sentencing Practice in England and Wales (Basingstoke, Palgrave Macmillan). Pina-Sánchez, J and Linacre, R (2013) ‘Sentence Consistency in England and Wales—Evidence from the Crown Court Sentencing Survey’ 53 British Journal of Criminology 1118–38. Pincus, Mr Justice (1987) ‘Court Involvement in Pre-Trial Procedures’ 61 Australian Law Journal 471–78. Piper, C (2007) ‘Should Impact Constitute Mitigation? Structured Discretion versus Mercy’ Criminal Law Review 141–55. Plato (1988) The Laws of Plato (translated by TL Pangle) (Chicago, University of Chicago Press). Player, E (2012) ‘Sentencing Women: Towards Gender Equality’ in L Zedner and JV Roberts (eds), Principles and Values in Criminal Law and Criminal Justice: Essays in Honour of Andrew Ashworth (Oxford, Oxford University Press). Pomerance, Justice RM (2013) ‘The New Approach to Sentencing in Canada: Reflections of a Trial Judge’ 17(3) Canadian Criminal Law Review 305–26. Posner, RA (2016) Divergent Paths—The Academy and the Judiciary (Cambridge, Massachusetts, Harvard University Press). —— (2013) Reflections on Judging (Cambridge, Massachusetts, Harvard University Press). —— (2010) ‘The Judiciary and the Academy: A Fraught Relationship’ 29(1) University of Queensland Law Journal 13–18. —— (2008) How Judges Think (Cambridge, Massachusetts, Harvard University Press). —— (1999a) The Problematics of Moral and Legal Theory (Cambridge, Massachusetts, Harvard University Press). —— (1999b) ‘Emotion versus Emotionalism in Law’ in SA Bandes (ed), The Passions of Law (New York, New York University Press). —— (1993) The Problems of Jurisprudence (Cambridge, Massachusetts, Harvard University Press). Quigley, T (2011) Annotation to R v Arcand, Westlaw International Transcript, 4–6. —— (1999) ‘Has the Role of Judges in Sentencing Changed … Or Should It?’, Paper delivered at the conference Changing Punishment at the Turn of the Century: Finding a Common Ground, 26–29 September 1999, Saskatoon, Saskatchewan. Rauxloh, R (2012) Plea Bargaining in National and International Law (Abingdon, Routledge). Raz, J (2009) The Authority of Law, 2nd edn (Oxford, Oxford University Press).

Bibliography

 259

Reed, Lord Robert (2014) ‘Human Rights and Domestic Legal Traditions’ in RG Anderson, J Chalmers and J MacLeod (eds), Glasgow Tercentenary Essays—300 Years of the School of Law (Edinburgh, Avizandum Publishing). Renaud, the Hon Mr Justice G (2012) ‘Book Review—Mitigation and Aggravation at Sentencing’ 36(5) Criminal Law Journal 317–22. Richardson, HS (1997) Practical Reasoning about Final Ends (Cambridge, Cambridge University Press). Richardson, J (2011) ‘A “Just” Outcome: Losing Sight of the Purpose of Criminal Procedure’ Journal of Commonwealth Criminal Law 105–31. Richardson, L (2011) ‘Reduction in Sentence for a Guilty Plea: The New Zealand Supreme Court Rejects a Sliding-Scale Approach’ Journal of Commonwealth Criminal Law 161–64. Roach Anleu and Mack (2010) ‘Trial Courts and Adjudication’ in P Cane and HH Kritzer (eds), The Oxford Handbook of Empirical Legal Research (New York, Oxford University Press). —— and —— (2009) ‘Intersections between In-Court Procedures and the Production of Guilty Pleas’ 42(1) Australian and New Zealand Journal of Criminology 1–23. —— and —— (2001) ‘Pleading Guilty and Professional Relations in Australia’ 22 Justice System Journal 155–84. Roberts, JV (2013a) ‘Sentencing Guidelines in England and Wales: Recent Developments and Emerging Issues’ 76(1) Law and Contemporary Problems 1–25. —— (2013b) ‘Complying with Sentencing Guidelines—Latest Findings from the Crown Court Sentencing Survey’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2012a) ‘Structuring Sentencing in Canada, England and Wales: A Tale of Two Jurisdictions’ 23 Criminal Law Forum 319–45. —— (2012b) ‘Structured Sentencing: Lessons from England and Wales for Common Law Jurisdictions’ 14(3) Punishment & Society 267–88. —— (2011a) ‘Sentencing Guidelines and Judicial Discretion—Evolution of the Duty of Courts to Comply in England and Wales’ 51(6) British Journal of Criminology 997–1013. —— (2011b) Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2011c) ‘Punishing, More or Less: Exploring Aggravation and Mitigation at Sentencing’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2010) ‘First-Offender Sentencing Discounts: Exploring the Justifications’ in JV Roberts and A von Hirsch (eds), Previous Convictions at Sentencing—Theoretical and Applied Perspectives (Oxford, Hart Publishing). —— (2009) ‘Structuring Sentencing Discretion’ in A von Hirsch, A Ashworth and JV Roberts (eds), Principled Sentencing—Readings on Theory and Policy, 3rd edn (Oxford, Hart Publishing). —— (2008) Punishing Persistent Offenders—Exploring Community and Offender Perspectives (Oxford, Oxford University Press). Roberts, JV and Ashworth, A (2016), ‘The Evolution of Sentencing Policy and Practice in England and Wales, 2003–2015’ 45(1) Crime and Justice 307–58. Roberts, JV and Baker, E (2008) ‘Sentencing Structure and Reform in Common Law Jurisdictions’ in SG Shoham, O Beck and M Kett (eds), International Handbook of Penology and Criminal Justice (Boca Raton, CRC Press). Roberts, JV and Bebbington, HH (2013) ‘Sentencing Reform in Canada: Promoting a Return to Principles and Evidence-based Policy’ 17(3) Canadian Criminal Law Review 327–47. Roberts, JV and Bradford, B (2015) ‘Sentence Reductions for a Guilty Plea in England and Wales: Exploring New Empirical Trends’ 12(2) Journal of Empirical Legal Studies 187–210. Roberts, JV and Cole, DP (1999) ‘Introduction to Sentencing and Parole’ in JV Roberts and DP Cole (eds), Making Sense of Sentencing (Toronto, University of Toronto Press).

260 

Bibliography

Roberts, JV, Doob, AN and Marinos, V (with the assistance of C La Prairie, Judge D Cole and T Perry) (2000) Judicial Attitudes to Conditional Terms of Imprisonment: Results of a National Survey, Report 2000-10e (Ottawa, Department of Justice). Roberts, JV and Harris, L (2016) ‘Editorial’ Current Sentencing Practice—News, 1, 1. Roberts, JV and Hough, M (2015) ‘Empirical Sentencing Research: Options and Opportunities’ in JV Roberts (ed), Exploring Sentencing Practice in England and Wales (Basingstoke, Palgrave Macmillan). —— and —— (2013) ‘Sentencing Riot-Related Offending—Where Do the Public Stand?’ 53 British Journal of Criminology 234–56. Roberts, JV, Hough, M and Ashworth, A (2011) ‘Personal Mitigation, Public Opinion and Sentencing Guidelines in England and Wales’ Criminal Law Review 524–30. Roberts, JV, Hough, M, Jacobson, J and Moon, N (2009) ‘Public Attitudes to Sentencing Purposes and Sentencing Factors: An Empirical Analysis’ Criminal Law Review 771–82. Roberts, JV and La Prairie (2000) Conditional Sentencing in Canada: An Overview of Research Findings, Report 2000-6e (Ottawa, Department of Justice). Roberts, JV and Pei, W (2016) ‘Structuring Judicial Discretion in China: Exploring the 2014 Sentencing Guidelines’ 27(1) Criminal Law Forum 3–33. Roberts, JV and Rafferty, A (2011) ‘Sentencing Guidelines in England and Wales: Exploring the New Format’ 9 Criminal Law Review 681–89. Roberts, JV, Stalans, LJ, Indermaur, D and Hough, M (2003) Penal Populism and Public Opinion: Lessons from Five Countries (New York, Oxford University Press). Robertson, Judge J (2007) ‘The Sentence of the Court’, Paper delivered at the National Judicial College conference on ‘Communication in the Courtroom’, 10 November 2007, Sydney. Rock, P (1993) The Social World of an English Crown Court—Witnesses and Professionals in the Crown Court Centre at Wood Green (Oxford, Clarendon Press). Rodger of Earlsferry, Lord (2010) ‘Foreword’ in J Chalmers, F Leverick and L Farmer (eds), Essays in Criminal Law in Honour of Sir Gerald Gordon (Edinburgh, Edinburgh University Press). Rook, Judge P and Ward, R (2016) Rook and Ward on Sexual Offences Law and Practice, 5th edn (London, Sweet and Maxwell). —— and —— (2014) Rook and Ward on Sexual Offences Law and Practice—First Supplement to the Fourth Edition (London, Sweet and Maxwell). —— and —— (2010) Rook and Ward on Sexual Offences Law and Practice, 4th edn (London, Sweet and Maxwell). Ross, PE (2006) ‘The Expert Mind’ 295(2) Scientific American 64–71. Rosett, A (1967) ‘The Negotiated Guilty Plea’ 374 Annals of the American Academy of Political and Social Science 70–81. Ruby, CC (1999) ‘Foreword’ in JV Roberts and DP Cole (eds), Making Sense of Sentencing (Toronto, University of Toronto Press). —— (1986) ‘Range of Sentence’ 28 Criminal Law Quarterly 447–53. Ruby, CC, Davies, B, Doucette, D, Loosemore, S, Orkin, JR and Wawzonek, C (2008) Sentencing, 7th edn (Markham, LexisNexis Canada). Ruderman, RS (1997) ‘Aristotle and the Recovery of Political Judgment’ 91(2) American Political Science Review 409–20. Rudin, J (2011) ‘Eyes Wide Shut: The Alberta Court of Appeal’s Decision in R v Arcand and Aboriginal Offenders’ 48(4) Alberta Law Review 987–1008. Ryberg, J (2010) ‘Recidivism, Retributivism, and the Lapse Theory of Previous Convictions’ in JV Roberts and A von Hirsch (eds), Previous Convictions at Sentencing—Theoretical and Applied Perspectives (Oxford, Hart Publishing). Samuels, A (1987) ‘Consistency in Sentencing’ in DC Pennington and S Lloyd-Bostock (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies).

Bibliography

 261

Sanders, A, Young, R and Burton, M (2010) Criminal Justice, 4th edn (Oxford, Oxford University Press). Santow, Justice GFK (2000) ‘Mandatory Sentencing: A Matter for the High Court?’ 74 The Australian Law Journal 298–305. Schön, DA (1987) ‘The Crisis of Professional Knowledge and the Pursuit of an Epistemology of Practice’ in CR Christensen and AJ Hansen (eds), Teaching and the Case Method (Boston, Harvard Business School Press). Schram, S (2012) ‘Phronetic Social Science: An Idea Whose Time has Come’ in B Flyvbjerg, T Landman and S Schram (eds), Real Social Science—Applied Phronesis (Cambridge, Cambridge University Press). Schulhofer, SJ (1994) ‘A Wake-Up Call from the Plea-Bargaining Trenches’ Law and Social Inquiry 135–44. Scott, C (2010) Criminal Justice and Licensing (Scotland) Bill—Stage 2: Written Submission (Glasgow, Association of Chief Police Officers in Scotland). Scott, Sheriff R (2005) ‘Reflections on Over a Quarter of a Century as a Sheriff ’ 11 Scottish Journal of Criminal Justice Studies 22–29. Scottish Government (2016) Criminal Proceedings in Scotland, 2014–15 (Edinburgh, Scottish Government). Seale, C and Filmer, P (2004) ‘Doing Social Surveys’ in C Seale (ed), Researching Society and Culture, 2nd edn (London, Sage). Sedley, S (2011) ‘Big Lawyers and Little Lawyers’ in Ashes and Sparks—Essays on Law and Justice (Cambridge, Cambridge University Press). Seifman, R (1982) ‘Plea Bargaining in Victoria—Getting the Judges’ Views’ 6 Criminal Law Journal 69–88. Seifman, RD and Freiberg, A (2001) ‘Plea Bargaining in Victoria: The Role of Counsel’ 25 Criminal Law Journal 64–74. Selfe, D (2012) ‘Sentencing for Rape of a Child under Thirteen’ 207 Criminal Lawyer 3–5. —— (2011) ‘Sentencing for the Rape of a Child under Thirteen Years of Age’ 204 Criminal Lawyer 5–6. Semmens, N (2011) ‘Methodological Approaches to Criminological Research’ in P Davies, P Francis and V Jupp (eds), Doing Criminological Research, 2nd edn (London, Sage). Sentencing Advisory Council (Victoria) (2016) Sentencing Guidance in Victoria (Melbourne, Sentencing Advisory Council). —— (2007) Sentence Indication and Specified Sentence Discounts: Final Report (Melbourne, Sentencing Advisory Council). Sentencing Commission for Scotland (2006) The Scope to Improve Consistency in Sentencing (Edinburgh, Sentencing Commission for Scotland). Sentencing Council for England and Wales (2016) Reduction in Sentence for a Guilty Plea Guideline— Consultation (London, Sentencing Council). —— (2014) Fraud, Bribery and Money Laundering Offences—Definitive Guideline (London, Sentencing Council). —— (2013) Sexual Offences—Definitive Guideline (London, Sentencing Council). —— (2012a) Drug Offences—Definitive Guideline (London, Sentencing Council). —— (2012b) Burglary Offences—Definitive Guideline (London, Sentencing Council). —— (2011) Assault—Definitive Guideline (London, Sentencing Council). Sentencing Guidelines Council (2008) Assault and other Offences against the Person—Definitive Guideline (London, Sentencing Guidelines Secretariat). —— (2008) Causing Death by Driving—Definitive Guideline (London, Sentencing Guidelines Secretariat). —— (2007a) Sexual Offences Act 2003—Definitive Guideline (London, Sentencing Guidelines Secretariat).

262 

Bibliography

—— (2007b) Reduction in Sentence for a Guilty Plea—Definitive Guideline (Revised 2007) (London, Sentencing Guidelines Secretariat). —— (2004) Overarching Principles: Seriousness—Guideline (London, Sentencing Guidelines Secretariat). Shapiro, I (1994) ‘Introduction’ in I Shapiro (ed), The Rule of Law (NOMOS XXXVI) (New York, New York University Press). Shapland, J (1981) Between Conviction and Sentence (London, Routledge and Kegan Paul). Shead, Sheriff C (2012) ‘The Decision in Murray v H.M. Advocate’ Scottish Criminal Law 93–95. Sheriffs’ Association (2009) Submission to the Justice Committee of the Scottish Parliament Considering the Criminal Justice and Licensing (Scotland) Bill 2009 (Edinburgh, Sheriffs’ Association). Sherman, N (1989) The Fabric of Character: Aristotle’s Theory of Virtue (Oxford, Oxford University Press). Shiels, R (2013) Offensive Weapons (Edinburgh, W Green). Shiels, R, Bradley, I, Ferguson, PW and Brown, AN (2016) Greens Annotated Acts—Criminal Procedure (Scotland) Act 1995, 15th edn (Edinburgh, W Green). Shklar, (1987) ‘Political Theory and the Rule of Law’ in A Hutchinson and P Monahan (eds), The Rule of Law: Ideal or Ideology (Toronto, Carswell). Shroff, D (2003) ‘The Future of Guideline Judgments’ 14(3) Current Issues in Criminal Justice 316–23. Simmons, WP (2012) ‘Making the Teaching of Social Justice Matter’ in B Flyvbjerg, T Landman and S Schram (eds), Real Social Science—Applied Phronesis (Cambridge, Cambridge University Press). Sitze, A (2007) ‘Keeping the Peace’ in N Hussain and A Sarat (eds), Forgiveness, Mercy, and Clemency (Stanford, Stanford University Press). Sloth-Nielsen, J and Ehlers, L (2005) A Pyrrhic Victory? Mandatory and Minimum Sentences in South Africa (Pretoria, Institute for Security Studies). Smith, I (2011) Law, Life and Laughter—A Personal Verdict (Edinburgh, Black & White Publishing). Solum, LB (2008) ‘A Virtue-Centered Account of Equity and the Rule of Law’ in C Farrelly and LB Solum (eds), Virtue Jurisprudence (Basingstoke, Palgrave Macmillan). —— (2003) ‘Virtue Jurisprudence—A Virtue Centred Theory of Judging’ 34(1–2) Metaphilosophy 178–213. —— (1994) ‘Equity and the Rule of Law’ in I Shapiro (ed), The Rule of Law (NOMOS XXXVI) (New York, New York University Press). Spears, D (1999) ‘Structuring Discretion: Sentencing in the Jurisic Age’ 22(1) University of New South Wales Law Journal 295–301. Spigelman, the Hon JJ (2008) ‘Consistency and Sentencing’ 82 Australian Law Journal 450–60. —— (2006) ‘Address to the Parole Authorities Conference’, 10 May 2006, cited in N Cowdery, ‘Guideline Judgments: It Seemed Like A Good Idea At The Time’, paper presented at the International Conference of the International Society for the Reform of Criminal Law, 2–6 July 2006, Brisbane. Spohn, C (2009) How Do Judges Decide? The Search for Fairness and Justice in Punishment, 2nd edn (Thousand Oaks, Sage). Stanbury, M (2016) ‘Reduction in Sentence for a Guilty Plea: Comments on the Proposed Reforms’ Current Sentencing Practice—News 1, 6–7. Stanley, L and Wise, S (1990) ‘Method, Methodology and Epistemology in Feminist Research Processes’ in L Stanley (ed) Feminist Praxis: Research, Theory and Epistemology in Feminist Sociology (London, Routledge). Stark, F (2012) ‘A Perfectly Unreasonable Decision: Jenkins v HM Advocate’ 16(1) Edinburgh Law Review 86–92. Steele, C (2010) Criminal Justice and Licensing (Scotland) Bill—Stage 2: Written Submission (Glasgow, Scottish Police Federation).

Bibliography

 263

Steiker, CS (2007) ‘Tempering or Tampering? Mercy and the Administration of Criminal Justice’ in N Hussain and A Sarat (eds), Forgiveness, Mercy, and Clemency (Stanford, Stanford University Press). Steinberger, PJ (1993) The Concept of Political Judgment (Chicago, University of Chicago Press). Stephen, FH, Fazio, G and Tata, C (2008) ‘Incentives, Criminal Defence Lawyers and Plea Bargaining’ 28 International Review of Law and Economics 212–19. Stewart, AL (1997) The Scottish Criminal Courts in Action, 2nd edn (Edinburgh, Butterworths). Stith, K and Cabranes, JA (1997) ‘Judging under the Federal Sentencing Guidelines’ 91(4) Northwestern University Law Review 1247–83. Stobbs, N (2008) ‘Book Review—Richard Edney and Mirko Bagaric, Australian Sentencing: Principles and Practice (Cambridge University Press, 2007) 8(2) Queensland University of Technology Law and Justice Journal 475–77. Stobbs, N, Kleinau, L and Kolstad, S (2014) ‘“Structuring” Judicial Sentencing Discretion— Consistency, Guidance or Pandering to the Punitive?’ 39(3) Alternative Law Journal 156–58. Stoddart, Sheriff C (2011) ‘No Second Chance’ 56(4) Journal of the Law Society of Scotland 42–43. Strauss, A (1987) Qualitative Analysis for Social Scientists (New York, Cambridge University Press). Strauss, A and Corbin, JM (1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory (Thousand Oaks, Sage). —— and —— (1990) Basics of Qualitative Research: Grounded Theory Procedures and Techniques (Newbury Park, Sage). Tadros, V (2006) ‘Two Particular Truths: A Response to Neil MacCormick’s “Particulars and Universals”’ in Z Bańkowski and J MacLean (eds), The Universal and the Particular in Legal Reasoning (Aldershot, Ashgate). Tamanaha, BZ (2006) Law as a Means to an End—Threat to the Rule of Law (New York, Cambridge University Press). —— (2004) On the Rule of Law—History, Politics, Theory (Cambridge, Cambridge University Press). —— (1997) Realistic Socio-Legal Theory—Pragmatism and a Social Theory of Law (Oxford, Clarendon Press). Tarling, R (2006) ‘Sentencing Practice in Magistrates’ Courts Revisited’ 45 Howard Journal of Criminal Justice 29–41. Tarnas, R (1991) The Passion of the Western Mind: Understanding the Ideas that have Shaped our World View (New York, Harmony Books). Tasioulas, J (2006) ‘Punishment and Repentance’ 81(316) Philosophy 279–322. Tata, C (2013) ‘The Struggle for Sentencing Reform—Will the English Sentencing Guidelines Model Spread?’ in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). —— (2010a) ‘Sentencing and Penal Decision-Making: Is Scotland Losing its Distinctive Character?’ in H Croall, G Mooney and M Munro (eds), Criminal Justice in Scotland (Abingdon, Willan Publishing). —— (2010b) ‘A Sense of Justice—The Role of pre-Sentence Reports in the Production (and Disruption) of Guilt and Guilty Pleas’ 12(3) Punishment & Society 239–61. —— (2010c) Criminal Justice and Licensing (Scotland) Bill—Stage 2: Written Submission (Glasgow, Centre for Sentencing Research, University of Strathclyde). —— (2007) ‘In the Interests of Clients or Commerce? Legal Aid, Supply, Demand and “Ethical Indeterminacy” in Criminal Defence Work’ 34 Journal of Law and Society 489–519. —— (2002) ‘Accountability for the Sentencing Decision Process—Towards a New Understanding’ in C Tata and N Hutton (eds), Sentencing and Society—International Perspectives (Aldershot, Ashgate). Tata, C, Burns, N, Halliday, S, Hutton, N and McNeill, F (2008) ‘Assisting and Advising the Sentencing Decision Process—The Pursuit of “Quality” in Pre-Sentence Reports’ 48 British Journal of Criminology 835–55.

264 

Bibliography

Tata, C, Goriely, T, McCrone, P, Duff, P, Knapp, M, Henry, A, Lancaster, B and Sherr, A (2004) ‘Does Mode of Delivery Make a Difference to Criminal Case Outcomes and Clients’ Satisfaction? The Public Defence Solicitor Experiment’ Criminal Law Review 120–35. Tata, C and Hutton, N (1998) ‘What “Rules” in Sentencing? Consistency and Disparity in the Absence of “Rules”’ 26 International Journal of the Sociology of Law 339–64. Tata, C and Stephen, F (2006) ‘“Swings and Roundabouts”: Do Changes to the Structure of Legal Aid Remuneration Make a Real Difference to Criminal Case Management and Case Outcomes?’ Criminal Law Review 722–41. Tata, C, Wilson, JN and Hutton, N (1996) ‘Representations of Knowledge and Discretionary DecisionMaking by Decision-Support Systems: The Case of Judicial Sentencing’ Journal of Information, Law & Technology 2. Terblanche, SS (2011) ‘The Discretionary Effect of Mitigating and Aggravating Factors: A South African Case Study’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2009) ‘Case Reviews: Sentencing’ 22 South African Journal of Criminal Justice 276–86. —— (2008) ‘Case Reviews: Sentencing’ 21 South African Journal of Criminal Justice 119–34. —— (2007) A Guide to Sentencing in South Africa, 2nd edn (Durban, LexisNexis South Africa). —— (2001) ‘Aspects of Minimum Sentence Legislation: Judicial Comment and the Courts’ Jurisdiction’ 14 South African Journal of Criminal Justice 1–19. Terblanche, SS and Mackenzie, G (2008) ‘Mandatory Sentences in South Africa: Lessons for Australia?’ 41(3) Australian and New Zealand Journal of Criminology 402–20. Terblanche, SS and Roberts, JV (2005) ‘Sentencing in South Africa: Lacking in Principle but Delivering Justice?’ 18 South African Journal of Criminal Justice 187–202. Thaman, SC (2010) ‘Introduction’ in SC Thaman (ed), World Plea Bargaining—Consensual Procedures and the Avoidance of the Full Criminal Trial (Durham, Carolina Academic Press). Thomas, DA (2013) ‘Case Comment—R v Karam (Mawawe Ibrahim)—Rape—Sentencing Guideline—Discretion of Judge—Victim Impact’ Criminal Law Review 998–1000. —— (2012a) ‘Case Comment—R v Brereton: Sentencing—Required Minimum Sentence—Firearms Act 1968 s.51A’ Criminal Law Review 396–99. —— (2012b) ‘Case Comment—R v Healey: Sentencing Guidelines—Production of Cannabis’ Criminal Law Review 640–44. —— (2012c) ‘Case Comment—R v Charles (Ashley Dwane): Sentencing—Rape of Child—Application of Sentencing Guidelines Council Guidelines’ Criminal Law Review 723–26. —— (2011a) ‘Case Comment—Sentencing: Firearms Act 1968 s.51A—Required Minimum Sentence’ Criminal Law Review 169–71. —— (2011b) ‘Case Comment—Sentencing: Manslaughter by Reason of Provocation—Relevance of Sentencing Guidelines Council Guidelines’ Criminal Law Review 415–19. —— (2010) ‘Sentencing Guidelines—The New Law’ 4 Archbold Review 7–9. —— (2008) ‘Sentencing: Rape of Child—Sexual Offences Act 2003 s. 5—Definitive Guideline’ Criminal Law Review 155–60. —— (2007) ‘Firearms Offences: More Mandatory Sentences’ 5 Archbold News 4–5. —— (2006) ‘Case Comment—Sentencing: Required Minimum Sentence—Firearms Act 1968, s.51A’ Criminal Law Review 654–57. —— (2005a) ‘Case Comment—Sentencing: Firearms Act 1968—s.51A—Required Minimum Sentence—“Exceptional Circumstances”’ Criminal Law Review 878–82. —— (2005b) ‘Case Comment—Sentencing: Sentencing Guidelines—Definitive Guidelines Published by the Sentencing Guidelines Council’ Criminal Law Review 979–82. —— (1987) ‘Sentencing: Some Current Questions’ in DC Pennington and S Lloyd-Bostock (eds), The Psychology of Sentencing—Approaches to Consistency and Disparity (Oxford, Centre for Socio-Legal Studies).

Bibliography

 265

Thomson, D (2013) Prisons, Prisoners and Parole, 2nd edn (Edinburgh, W Green). —— (2012) ‘The Scottish Ministers Try to Resolve the Issue of Punishment Parts in Discretionary Life Sentence Cases—But do they Succeed?’ Scots Law Times (News) 5 23–28. —— (2011) ‘The Punishment Part of a Discretionary Life Sentence—A Difficulty Remains’ Scots Law Times (News) 17, 105–10. —— (2008) ‘Sentencing Discounts under Section 196—the Unintended Role of the Prosecutor’ Scots Law Times (News) 129–31. Thomson, N (2001) In and Out of Court—The Legal Life and Musical Times of Sheriff Nigel Thomson (Edinburgh, Dunedin Academic Press). Tiarks, EJ (2016) Can Restorative Justice Provide a Solution to the Problem of Incoherence in Sentencing? (PhD thesis, Durham, University of Durham). Tombs, J (2008) ‘Telling Sentencing Stories’ in P Carlen (ed), Imaginary Penalties (Cullompton, Willan Publishing). —— (2005) Reducing the Prison Population—Penal Policy and Social Choices (Edinburgh, Scottish Consortium on Crime & Criminal Justice). —— (2004) A Unique Punishment—Sentencing and the Prison Population in Scotland (Edinburgh, Scottish Consortium on Crime & Criminal Justice). Tombs, J and Jagger, E (2006) ‘Denying Responsibility—Sentencers’ Accounts of their Decisions to Imprison’ 46(65) British Journal of Criminology 803–21. Tonry, M (2007) ‘Crime Does not Cause Punishment: The Impact of Sentencing Policy on Levels of Crime’ 20 SA Crime Quarterly 13–20. —— (2002) ‘Setting Sentencing Policy through Guidelines’ in S Rex and M Tonry (eds), Reform and Punishment—The Future of Sentencing (Cullompton, Willan Publishing). —— (1996) Sentencing Matters (New York, Oxford University Press). —— (1992) ‘Judges and Sentencing Policy: The American Experience’ in C Munro and M Wasik (eds), Sentencing, Judicial Discretion and Training (London, Sweet and Maxwell). Torre, A and Wraith, D (2013) ‘The Demand for Sentence Discounts: Some Empirical Evidence’ 37(3) Criminal Law Journal 193–200. Traynor, S and Potas, I (2002) ‘Sentencing Methodology: Two-tiered or Instinctive Synthesis’ 25 Sentencing Trends and Issues (December) (Sydney, Judicial Commission of New South Wales). Treacy, Lord Justice (2016) ‘Letter to the Editor’ Criminal Law Review 489–90. —— (2014) ‘Letter to the Editor’ Criminal Law Review 298–99. Tversky, A and Kahneman, D (1974) ‘Judgment under Uncertainty: Heuristics and Biases’ 185(4157) Science 1124–31. van Zyl Smit, D (2000) ‘Mandatory Sentences: A Conundrum for the New South Africa?’ 2(2) Punishment & Society 197–12. Viljoen Commission (1976) Report of the Commission of Inquiry into the Penal System of the Republic of South Africa, RP 78 (Pretoria, Government Printer). Vogel, ME (2007) Coercion to Compromise—Plea Bargaining, The Courts, and the Making of Political Authority (New York, Oxford University Press). von Hirsch, A (2010) ‘Proportionality and the Progressive Loss of Mitigation: Some Further Reflections’ in JV Roberts and A von Hirsch (eds), Previous Convictions at Sentencing—Theoretical and Applied Perspectives (Oxford, Hart Publishing). —— (1986) Doing Justice: The Choice of Punishments—Report of the Committee for the Study of Incarceration (Boston, Northeastern University Press). von Hirsch, A and Ashworth, A (2005) Proportionate Sentencing—Exploring the Principles (Oxford, Oxford University Press). von Hirsch, A, Ashworth, A and Roberts, JV (eds) (2009) Principled Sentencing—Readings on Theory and Policy, 3rd edn (Oxford, Hart Publishing).

266 

Bibliography

von Hirsch, A and Jareborg, N (2009) ‘The Swedish Sentencing Law’ in von Hirsch, Ashworth and Roberts, ibid. —— and —— (1989) ‘Sweden’s Sentencing Statute Enacted’ Criminal Law Review 275–81. von Hirsch, A and Roberts, JV (2004) ‘Legislating Sentencing Principles: the Provisions of the Criminal Justice Act 2003 Relating to Sentencing Purposes and the Role of Previous Convictions’ Criminal Law Review 639–52. Wakefield, A (2011) ‘Undertaking a Criminological Literature Review’ in P Davies, P Francis and V Jupp (eds) Doing Criminological Research, 2nd edn (London, Sage). Waldron, J (1990) The Law (Oxford, Routledge). Walker, DM (2001) The Scottish Legal System—An Introduction to the Study of Scots Law, 8th edn (Edinburgh, W Green). Walker, N (2003) A Man Without Loyalties—A Penologist’s Afterthoughts (Chichester, Barry Rose Law Publishers). —— (1999) Aggravation, Mitigation and Mercy in English Criminal Justice (London, Blackstone Press). —— (1969) Sentencing in a Rational Society (London, Allen Lane). Walker, N and Padfield, N (1996) Sentencing—Theory, Law and Practice, 2nd edn (London, Butterworths). Warner, K (2014) ‘Sentencing Review 2013–2014’ 38(6) Criminal Law Journal 364–79. —— (2011) ‘Equality before the Law: Racial and Social Background Factors as Sources of Mitigation at Sentencing’ in JV Roberts (ed), Mitigation and Aggravation at Sentencing (Cambridge, Cambridge University Press). —— (2007a) ‘Sentencing Scholarship in Australia’ (2006–2007) 18(2) Current Issues in Criminal Justice 241–65. —— (2007b) ‘Mandatory Sentencing and the Role of the Academic’ 18 Criminal Law Forum 321–47. —— (2005) ‘Sentencing Review 2004–2005’ 29 Criminal Law Journal 355. —— (2003) ‘The Role of Guideline Judgments in the Law and Order Debate in Australia’ 27 Criminal Law Journal 8–22. —— (2002) Sentencing in Tasmania, 2nd edn (Sydney, The Federation Press). Wasik, M (2014) A Practical Approach to Sentencing, 5th edn (Oxford, Oxford University Press). —— (2008) ‘Sentencing Guidelines in England and Wales—State of the Art?’ 4 Criminal Law Review 253–63. —— (2006) ‘Publication Review—Sentencing: Law and Practice (Thomas O’Malley)’ 41 Irish Jurist 146–47. —— (2003) ‘Sentencing Guidelines—Past, Present and Future’ 56 Current Legal Problems 239–64. Webb, J, Schirato, T and Danaher, G (2002) Understanding Bourdieu (Crows Nest, Allen and Unwin). Welsh, Sheriff T (2013a) ‘Commentary to Murray v H.M. Advocate, 2013 SCL 243’ Scottish Criminal Law 251–52. —— (2013b) ‘Commentary to Jones v Nisbet, 2013 SCL 535’ Scottish Criminal Law 538. —— (2010) ‘Commentary to H.M. Advocate v Graham, 2010 SCL 789’ Scottish Criminal Law 808–10. —— (2009) ‘Commentary to Allan v H.M. Advocate, 2009 SCL 653’ Scottish Criminal Law 661–62. Wheatley, Lord (1987) One Man’s Judgment—An Autobiography (London, Butterworths). Wheeler, S, Mann, K and Sarat, A (1988) Sitting in Judgment—The Sentencing of White-Collar Criminals (New Haven, Yale University Press). White, RM and Willock, ID (2006) The Scottish Legal System, 4th edn (Haywards Heath, Tottel Publishing). Wiggins, D (1975) ‘Deliberation and Practical Reason’ 76 (1975–1976) Proceedings of the Aristotelian Society, New Series, 29–51. Wiles, P (2015) ‘Foreword’ in JV Roberts (ed), Exploring Sentencing Practice in England and Wales (Basingstoke, Palgrave Macmillan).

Bibliography

 267

Williams, B (1978) ‘Introduction’ in H Hardy (ed), Concepts and Categories—Philosophical Essays (London, The Hogarth Press). Willis, JE (1995) ‘New Wine in Old Bottles: The Sentencing Discount for Pleading Guilty’ 13(2) Law in Context 39–78. —— (1985) ‘The Sentencing Discount for Guilty Pleas—Are We Paying Too Much for Efficiency?’ 18(2) Australian and New Zealand Journal of Criminology 131–46. Woolf, Lord (2008a) ‘Making Sense of Sentencing’ in C Campbell-Holt (ed), The Pursuit of Justice (Oxford, Oxford University Press). —— (2008b) ‘Do We Need a New Approach to Penal Policy?’ in Campbell-Holt, ibid. —— (2007) ‘Foreword’ in J Jacobson and M Hough, Mitigation: The Role of Personal Factors in Sentencing (London, Prison Reform Trust). Yoshino, K (2013) ‘Holding Court’ (Book review of Reflections on Judging, RA Posner, Harvard University Press, 2013) The New York Times Book Review (10 November 2013) (New York, New York Times Company) 13. Young, A (1988) The Role of an Appellate Court in Developing Sentencing Guidelines (Ottawa, Department of Justice). Young, Mr Justice PW (2014) ‘Current Issues—Respect for the Judiciary’ 88 Australian Law Journal 155–56. —— (1999) ‘Current Issues—Guideline Judgments in Criminal Appeals’ 73 Australian Law Journal 13–14. Young, W and King, A (2013) ‘The Origins and Evolution of Sentencing Guidelines—A Comparison of England and Wales and New Zealand’, in A Ashworth and JV Roberts (eds), Sentencing Guidelines—Exploring the English Model (Oxford, Oxford University Press). Zander, M (1993) ‘The “Innocent” (?) Who Plead Guilty’ New Law Journal 85–88; 95–96. Zimring, FE (1977) ‘Making the Punishment Fit the Crime: A Consumer’s Guide to Sentencing Reform’ in RA Duff and D Garland (eds) (1994), A Reader on Punishment (Oxford, Oxford University Press).

268 

INDEX

A Aas, KF  136, 174, 192, 195 AB v The Queen  36, 44 Abbs, R  35 Adam J  30–31 aggravating factors agreed narratives in guilty plea discounting  220 consideration, generally  116 cumulative sentencing model  89n Definitive Guideline  78 and equity  88–93, 104 flat-rate sentencing model  89n, 90 instinctive synthesis model  102 judicial conceptions of sentencing process  181 progressive loss of mitigation theory  89n, 90 self-aggravating offences  90 sentencing guidelines  141 staged (two-tiered) sentencing  30, 31 statutory  88 agreed narratives guilty plea discounts  220–21, 220n Aitken, Bill  66 Alschuler, AW  212 Anderson, JL  34 Annan, N  179 antisocial behaviour orders  86 Archilochus The Hedgehog and the Fox  108–09 Aristotle episteme  124–25, 127, 133 equity (epieikeia)  58–59, 94 eudaimonia  125 Lesbian Rule  61 Nicomachean Ethics  58, 135 phronesis  1, 11, 124–27, 135, 136, 229 poiesis  125 praxis  125 sophia  125 techne  124–25, 127, 176 armed robbery  31, 90, 91 Asche CJ  235n Ashley JA  108 Ashworth, Andrew  7, 22 aggravating factors  90 decision making process  151 guideline judgments  157 guilty plea discounts  233–34 instinctive synthesis model  51, 56 judicial discretion  55, 56, 98, 141 mercy in sentencing  81–82, 83

purposes of sentencing  114, 115, 116 Sentencing in the Crown Court  3–4 sentencing guidelines  141–42, 145, 146, 149, 150–51, 212 sentencing research  1–2, 15, 16 statutory minimum sentences  66 assault see also violence serious  73–75 sexual see sexual offences attentiveness as pluralist virtue  120 Attorney-General (Tas) v Blackler  137 Attorney-General’s Reference (No 16 of 2005)  80 Attorney-General’s Reference (No 8 of 2007)  80 Attorney-General’s Reference (Nos 32, 33 and 34 of 2007)  80 Attorney-General’s Reference (Nos 7 to 9 of 2009)  144 Attorney-General’s Reference (Nos 15, 16 and 17 of 2012)  147 Australia charge and/or fact bargaining  217–18 ‘double dipping’  217–18 guideline judgments  156 guilty plea discounts  197, 203, 212n, 219–20, 223, 233, 234, 235–36, 238, 239 individualised sentencing  36–37, 49, 64, 102, 107–08, 137, 139, 154 instinctive synthesis model  29–46, 49, 50, 102, 140, 182–83, 223, 228–29 judicial conceptions of sentencing process  182–83 judicial discretion  25, 30–31, 32–33, 34, 35–46, 192, 225 Law Reform Commission  71–72 mercy in sentencing  81 phronetic synthesis  232 research into judicial decision making  8, 9–10, 182–83 sentencing guidelines  45–46, 121, 167, 173, 203 sentencing jurisprudence generally  21–22 sentencing methodology  29–46, 152 staged (two-tiered) sentencing  30–37 statutory minimum sentences  69, 70, 71–72, 103 value pluralism  118, 179

270 

Index

B Bagaric, M  54 bail, offending on  88 Baillie v HMA  77 Baker, Richard  66 Baldwin, J  4 Banks, R and Harris, L  153 Barbaro v The Queen  45–46 Barile v Griffiths  87–88 Barile v PF, Dundee  87–88 Bartels, L  9, 182 Basten JA  52 Berger JA  39 Berlin, Isaiah instinctive synthesis model  53 moral monism  112–13, 177–79, 192 phronesis  125, 126 The Hedgehog and the Fox  108–09, 112, 118, 176, 192 value pluralism  1, 11, 107–13, 108n, 116–17, 124, 135 Berlin, Isaiah and Williams, B pluralist/monist dichotomy  111 Bingham of Cornhill, Lord definition of the rule of law  54–55 Blockley v Cameron  216 Bloos, MR and Plaxton, MC  38 Bottoms, A  157 populist punitiveness  66 rational exercise of mercy  83 Bourdieu, Pierre habitus  63 intuitive decision making  134 Braithwaite, J and Pettit, P  116n breach of trust  88 Brennan J  43, 102 Brett, N  82, 88 Brien, A  82, 87, 88 Brodeur, Jean-Paul  27 Bryman, A  23 C Calem v HMA  74 Callinan J  33–34 Cameron v The Queen  197, 197n, 235 Canada individualised sentencing  25–29, 49, 53, 64, 101, 136 judicial discretion  25, 26–27, 192, 225 phronetic synthesis  232 proportionality  25, 28–29 purposes of sentencing  29 research into judicial decision making  8–9 sentencing guidelines  167 sentencing methodology  25–29, 152 starting-point method  37–39 statutory minimum sentences  66n, 70 value pluralism  118, 179 Canadian Sentencing Commission sentencing as an art  52 survey of judges’ opinions  9

Carloway, Lord  47, 48, 98, 152, 154, 167, 201 Carnie, J  7 case law, analysis  21–23 causing death by dangerous driving aggravating factors  91 leniency in sentencing  84–85 character references consideration of  70–80 charge bargaining ‘double dipping’  216–19, 226 Charmaz, K  19–20 Clarke, A, et al  213 Cobiac v Liddy  81 Cochrane v HMA  165 Cole, D and Manson, A  27, 38 common law offences equity  72–79 common sense standards rule of law doctrine  57 compassion mercy and mitigation  83 condign punishment equity  88–93, 103–04 consistency and uniformity in sentencing  46, 51, 56, 97–102, 103, 105 alternatives to phronetic synthesis  142–54 England and Wales  142–54, 176–77, 180, 191–93 equity  97–102, 105 generally  230–31, 232–33 guideline judgments  156 instinctive synthesis model  43, 45, 46, 51 and judicial discretion  142–54, 230–31 monism  176, 180 public concern  3 sentencing guidelines  148, 150, 153, 161, 162, 176–77, 180, 191–93 sentencing purposes  114, 115 starting-point method  37–38 conviction definition  64 Cooper, Lord  72 Cooper, G  18 Cooper, J  144–46, 150, 151, 153, 179 Coroners and Justice Act (2009)  146, 151, 153, 154, 202 Coulter, D and Wiens, JR  125, 126 Court of Appeal see also guideline judgments appellate review  42–43, 92, 94–96, 103, 155 questions of principle  164–65 Scottish Sentencing Council  160–62 sentencing guidelines  146, 232 crime rates affecting sentence  62 reduction as purpose of sentencing  113, 116 and statutory minimum sentences  66–67, 71 Criminal Cases (Punishment and Review) (Scotland) Act (2012)  48 Criminal Justice Act (2003)  116, 178 purposes of sentencing  113, 114, 115

Index Criminal Justice and Licensing (Scotland) Act (2010)  160–61 Criminal Justice and Licensing (Scotland) Bill  66, 70, 101 Criminal Law (Consolidation) (Scotland) Act (1995)  67, 77 Criminal Procedure (Scotland) Act (1995)  47, 156, 169, 196, 197, 233 Criminal Proceedings (Scotland) Act (1993)  47 Crisp J  137 Crockett J  30–31 Crowder, G sentencing guidelines  121, 177, 178 value pluralism  108, 108n, 111–12, 117, 118, 119, 120–24 culpable homicide equity in sentencing  72–73, 84 cumulative sentencing model  89n custodial sentence factors leading to  96–97 D Dalgarno v PF, Elgin  77n Darbyshire, P  5–6, 222 Davis, J  126 Dawes, W, et al  213–14 deliberative system  190, 191, 194 denunciation as a purpose of sentencing  29, 98, 116, 117, 126, 179 statutory minimum sentences  70 desert in sentencing  116 deterrence certainty of arrest  71 generally  91–92, 93 and judicial discretion  58 knife possession  91–92 as a purpose of sentencing  26, 29, 52, 98, 113, 114, 116, 179 sentence length  71 staged (two-tiered) sentencing  37–39 statutory minimum sentences  70, 71, 103 Dhami, MK  5, 15, 170 Dingwall, G  115 discounts see guilty plea discounts discretion see judicial discretion Doherty J  28–29 Doob, AN  27 Doob, AN and Beaulieu, LA  9 Dorrian, Lady  139, 162 ‘double dipping’ guilty plea discounts  210, 216–19, 226 Douglas, R  114 DPP (Cth) v De La Rosa  45 DPP v Drought  88 Drewry, G, Blom-Cooper, L and Blake, C  5 Dreyfus model of learning advanced beginner  128 competent performer  128–29

 271

expertise  129–33 generally  107, 127, 193 interpretation and judgment  129 intuitive understanding  129, 130–33, 135 novice actor  127 phronesis  127 proficient performer  129 rationality  134–35 relationship of involvement  128–29 drugs offences aggravating factors  90, 91 cannabis farming  91 offender’s motivation  65–66 social supply  91 statutory minimum sentences  64–66 trafficking  91 Du Plooy v HMA  46, 159, 163, 164, 196–99, 204–06, 211–12, 225–26, 234, 238 Duff, RA  118 Duff, Sheriff A  202 Duffy v HMA  93 Dundee Cold Stores Ltd v HMA  166 E Eassie, Lord  200 Edney, R and Bagaric, M consistency in sentencing  230 episteme  133 equity in sentencing  97 instinctive synthesis model  34–35, 40, 41, 54, 55–56, 57, 102, 104–05, 229, 232 rule of law doctrine  57–58, 97–98 rule-based deliberation  134, 135, 138 staged (two-tiered) sentencing  41 Edward & Sons v McKinnon  72 Elias v The Queen  53 Emslie, Lord  48 England and Wales consistency in sentencing  142–54, 176, 180, 191–93 guideline judgments  143, 156–57, 158 guilty plea discounts  198, 202, 203, 203n, 211–12, 214–15, 225, 237–38, 239 intuitive decision making  180–81 judicial conceptions of sentencing process  180–81 judicial discretion  142, 202, 230 Judicial Studies Board  143 mercy in sentencing  81–82 monism  176–79, 180, 192 phronetic synthesis  232 research into judicial decision making  2–6, 180–81 Sentencing Advisory Panel  143, 144, 146, 148 Sentencing Council  146–51 sentencing guidelines  123, 142–54, 167, 172–79, 191–93, 202, 203, 229–30, 232 sexual offences  78–79 SGC see Sentencing Guidelines Council Enlightenment monism  109

272 

Index

epieikeia see equity episteme  124–25, 127, 133, 192 equity aggravating factors  88–93, 104 Aristotelian theory  58–59 common law offences  72–79 condign punishment  88–93, 103–04 consistency and uniformity  97–102, 105 culpable homicide  72–73, 84 ethical flexibility  61 guilty plea discounts  210, 226 individualised sentencing  64, 103 instinctive synthesis model  57–60, 102–03 judicial attitude to  97 and justice  58–59 justice as fairness  51, 58–59, 102–03 legal justice rectified  58–59 mitigating factors  88, 93 non-virtuous sentencers  94–96 Plato’s Laws  88 rule of law doctrine  57–64, 94, 97–98 Scotland  51, 64–79 serious assault  73–75 sexual offences  76–79 Solum’s theory  51, 58, 97, 102–03, 136 statutory minimum sentences  69–72 virtue-centred theory  58–60 virtuous judge  60, 61 eudaimonia  125 experience context-dependent  134–35 habitus  63, 103 importance  47, 52, 58 inexperienced judges  94–96 instinctive synthesis model  26, 54, 94–96, 102–03, 194 intuitive decision making  129–33, 181–82, 189–91 judicial conceptions of sentencing process  183, 184–85 judicial discretion  49–50 phronesis  95–96, 103, 126–27, 129–33 and rationality  135 expertise see also knowledge Dreyfus model of learning  129–33, 193 and intuitive thinking  189 F fact bargaining ‘double dipping’  216–19, 226 fairness equity  51, 58–59, 102–03 non-virtuous sentencers  94–96 rule of law doctrine  58 statutory minimum sentences  68–69, 70, 103 Ferguson, PR and McDiarmid, C  77 Ferguson v HMA  47, 152, 154, 172 Field, D  140

Finland sentencing guidance  156–57 firearms offences statutory minimum sentences  64, 70 Fitz-Gibbon, K  5, 185n flat-rate sentencing model  89n, 90 Fleming v HMA  165 flexibility ethical  61 guilty plea discounts  222, 237–38 judicial discretion  25, 227 as pluralist virtue  120–21 sentencing guidelines  141, 145 Flyvbjerg, Bent  202 definition of intuition  133 Dreyfus model  127–33 phronesis  11, 59, 107, 124–35, 195, 229 phronesis and sentencing  133–35, 187, 192–93 rule-based rationality  192 Fox, RG mercy in sentencing  83 Fox, RG and Freiberg, A  138 Frank, AW  132, 133 Frankel, Judge Marvin  56, 57 Frederick’s ‘cognitive reflection test’  188 Freiberg, A  69 Fuller, LL  57 G Gallagher v HMA  216 Galston, W  119 Gardiner, P  131 Garling J  101 Geddes v HMA  165 Gemmell v HMA  47, 159, 163, 172, 198–203, 206–11, 216, 221–22, 225–26, 234, 238 generosity as pluralist virtue  120 Gerber, P  217 Gifford, Lord  3 Gigerenzer, G and Todd, PM  189 Gill, Lord  15, 47, 136–37, 159, 167 guilty plea discounts  198–203, 206–11, 216, 221–22, 225–26, 234, 238 Gill v Thomson  164 Giordano v The Queen  235 Glaser, BG and Strauss, AL grounded theory  17–19, 20 Gleeson CJ  33–34, 53 Gordon, Sir Gerald  48–49, 69 Grant v HMA  165 Greig v HMA  190 grounded theory  17–19, 20 guideline judgments Appeal Court  42–43, 64, 121, 142–43, 153 Australia  156 consistency in sentencing  156 England and Wales  143, 156–57, 158 generally  155–56 identifying suitable cases  158 and instinctive synthesis model  42–43 judicial discretion  167–69

Index narrative form  157 Scotland  155–60 shortcomings  158–60 guidelines see also guideline judgments; sentencing guidelines Definitive  78–79, 143–46 guilty plea discounts agreed narratives  220–21, 220n amount of discount  196, 197–98, 199, 201 Australia  197, 203, 212n, 219–20, 223, 233, 234, 235–36, 238, 239 capping  238, 239–40 development of guilty plea culture  205, 210, 216 disclosure/recording  196 discretionary nature  198–206, 237–40 ‘double dipping’  210, 216–19, 226 England and Wales  198, 202, 203, 203n, 211–12, 214–15, 225, 237–38, 239 equity  210, 226 fines, discounting  215 flexibility  222, 237–38 generally  1, 58, 194, 195 guidelines  149, 196–98, 200, 202–06, 235–37 incentive to plead guilty  210, 211–12 individualised sentencing  202, 210, 221–25, 233, 237–40 inherent (unseen)  210, 219–21, 226 instinctive synthesis model  195, 223 judicial discretion  221–25, 233–40 justification  197, 198, 206–10, 221, 223, 234 New Zealand  236–37, 238 Practice Note (No 1 of 2008)  196 public confidence in justice system  210, 212–15, 238 recidivist offenders  198 reducing flexibility  222 Republic of Ireland  237, 238 risks associated with  210–26, 234, 238–39 Scotland  46, 47, 48, 194, 195–226, 238–39 sliding scale  197–201, 203, 204, 205, 233, 235, 236–37 staged (two-tiered) approach  47, 48 statutory provisions  196 suggested reforms  233–40 summary courts  215 tactical pleas  198, 213 under-discounting  224n where no discount granted  196 Gummow J  33–34 Guthrie, C, et al  188–89 H H v HMA (sub nom Hibbard v HMA)  166 habitus, concept of  63, 103 Hamilton, Lord  15, 48 Hammond J  183 Hampton, J  83 Hantzis v The Queen  52n Harris, L and Gerry, F  149, 150, 151, 153, 179

 273

Harrison, R mercy in sentencing  82 Hayne J  33–34 Hessell v The Queen  236–37 High Court of Justiciary analysis of case law  21–23 jurisdiction  13–14 reporting decisions of  22 sentencing powers  14 Hili v The Queen  45–46 HMA v AB  116, 165 HMA v Bell  80, 94 HMA v Boyle  91, 118–19, 158 HMA v Brand  190 HMA v Brough  91 HMA v Campbell  94, 95–96 HMA v Clark  94, 95–96 HMA v Collins  159n, 165 HMA v Cooperwhite  165, 217–18 HMA v Currie  76–77, 97 HMA v Discovery Homes Ltd  166 HMA v Edge  73, 97 HMA v Fallan  91 HMA v Gordon  190 HMA v Graham  158, 159, 160, 164, 166, 172 HMA v Jamieson  73, 92 HMA v M (sub nom HMA v McArthur)  22, 75, 97 HMA v McColl  160 HMA v McCourt  165 HMA v McGovern  165 HMA v McKay  81, 165 HMA v McKeever  165 HMA v McManus  172 HMA v Macpherson  165 HMA v Munro & Sons (Highland) Ltd  166 HMA v Noche  165 HMA v Roulston  165, 167, 172 HMA v SSK  94, 95–96, 165 HMA v Stalker  165 HMA v Stopper  91 HMA v Trainer  94–96 HMA v V  118–19, 121 Hodkinson, P grounded theory  18 Hogarth, J  8–9 Hogarth, RM  190 Hood, R  4 Hope, Lord  81 Hope of Craighead, Lord  21 Hough, M, et al  5, 80, 145, 180–81, 182 housebreaking, theft by  90 Hudson, B  116n Hughes LJ  80 Hughes v HMA  173 Hunter v HMA  77n Hutton, N  63, 149 consistency in sentencing  230 judicial discretion decried  56–57, 102, 104–05, 229, 232 rule-based deliberation  134, 135

274 

Index

Hutton, N and McNeill, F  66 Hyde, M  131 I incapacitation as a purpose of sentencing  29, 116, 179 individualised sentencing see also judicial discretion aggravating factors  78, 88–93 Australia  36–46, 49, 64, 102, 107–08, 137, 139, 154 balancing relevant factors  52–54 Canada  25–29, 49, 53, 64, 101, 136 consistency and uniformity in sentencing  46, 97–102, 103, 105, 191 criticisms of  56–57, 104–05 culpable homicide  72–73 equity  64, 103, 104–05 generally  69, 104–05 guilty plea discounts  202, 210, 221–25, 233, 237–40 importance  36–37, 44–45, 51, 54, 151–54 instinctive synthesis model  36–37, 51, 53–54, 102 leniency  79–80, 84–88 mercy  80–88 mitigating factors see mitigation New Zealand  136 phronesis  60, 103, 136–40, 179, 183 police view  67 Republic of Ireland  183 rule of law doctrine  55 Scotland  53, 64, 154–55, 192–94, 227–28, 233 sentencing as an art  136–40 sentencing guidelines  144 serious assault  73–75 sexual offences  76–79 sophia  60, 103 value pluralism see value pluralism virtue-centred theory  60 inflationary drift statutory minimum sentences  69, 103 instinctive synthesis model see also individualised sentencing; judicial discretion alternatives to  229 Australia  29–46, 49, 50, 102, 140, 182–83, 223, 228–29 balancing relevant factors  52–54, 102 Berlin, I  53 Canada  140 consistency in sentencing  43, 45, 46 criticisms of  43–45, 50, 51, 54, 55–57, 133, 229, 232 discretionary sentencing  51–54, 229 equity  57–60, 102–03 guideline judgments  45–46 guilty plea discounts  195, 223 inconsistent decisions  56 indeterminacy  55 individualised sentencing  36–37, 51, 53–54, 102

intuitive decision making  133–40 judicial conceptions of sentencing process  180–83, 184–91 knowledge and experience  54, 94–96, 102–03, 194 McHugh’s judgment  40–43 meaning  51–54 phronesis (practical wisdom)  59–60, 103, 129–33, 138, 140, 170–71, 179, 228–30 proportionality  43, 46 rule of law doctrine  51–57, 102–03 Scotland  46–49, 50, 81–82, 102, 140, 225–26, 228–30 sentencing as an art  52 sentencing guidelines  170–71 transparency of decisions  34–35, 41, 43–44, 102 unprincipled decisions  56 use of term  33 internal complexity  152–53, 192 intoxication, voluntary  88, 104 intuitive decision making accuracy  189–90 context-dependent  134–35 Dreyfus model of learning  129–30, 135 and experience  129–33, 181–82, 189–91 Frederick’s ‘cognitive reflection test’  188 intuition defined  133 judicial conceptions of sentencing process  180–83, 184–91 Markarian analysis  35–43, 133, 185 mixed intuitive-structured approach  187–91 phronesis  129–40, 179, 229, 240 and rationality  134–35 sentencing information systems compared  131–33 Irish Republic see Republic of Ireland J Jacobson, J and Hough, M  5, 181 Jakovlev v HMA  167 Jenkins v Harvie  165 Judge, Lord  5, 93, 114, 143, 144, 153, 166, 175 judicial discretion see also individualised sentencing; instinctive synthesis model appellate review  92, 94–96, 103 Australia  25, 30–31, 32–33, 34, 35–46, 192, 225 balancing relevant factors  52–54 Canada  25–26, 192, 225 consistency and uniformity in sentencing  46, 97–102, 105, 142–54, 191, 230–31 criticisms of  55–57, 102, 104–05, 229, 232 culpable homicide  72–73 England and Wales  142, 202, 230 experience  49–50 flexibility  25, 227 generally  1, 49, 51, 53, 104–05, 107–08, 183, 232–33 guideline judgments  167–69

Index guilty plea discounts  198–206, 221–25, 233–34, 237–40 importance  44–45, 230–31 leniency  79–80, 84–88 mercy  80–88 non-virtuous sentencers  94–96 phronesis  136–40, 179, 240 principled  141–42, 168–70, 194, 230–33 Republic of Ireland  168, 192, 225 rule of law doctrine  55–58 Scotland  25, 154–75, 192–94, 227–40 sentencing as an art  136–40 sentencing guidelines see sentencing guidelines serious assault  73–75 sexual offences  76–79 structuring  141–42 to select sentencing rationale  56 value pluralism see value pluralism virtue-centred theory  58–60 judicial intelligence  59 judicial self-regulation  162–63 judicial virtue  60 non-virtuous sentencers  94–96 judicial wisdom  59, 94–96 justice definition  61 and equity  58–59 and mercy  82 guilty plea discounts  237 JW v HMA  190 K Kable v DPP (NSW)  54 Kane v HMA  63 Kaye AJA  29 Kirby, Justice M  2, 43–45, 48, 50, 197, 197n, 235 knife possession deterrence  91–92 proposed minimum sentences  66–69 knowledge context-dependent  134–35 Dreyfus model see Dreyfus model of learning episteme  124–25, 127, 192 importance  47, 52 instinctive synthesis model  54, 102 intuitive understanding  129 relationship of involvement  128–29 techne  124–25, 127, 176, 191, 192 Korthagen, FAJ and Kessels, JPAM  125 Krasnostein, S and Freiberg, A  170–71 Kunle v HMA  166 L Lacey, N  116n Lamer CJ  27 Lane, Lord  3–4, 80, 143 Laws LJ  174  Lawton LJ  90, 232  Leader-Elliot, I  138 LeBel J  26–27, 136

 275

Lees v HMA  172 legal profession, structure  94 legal tradition, importance  62–63 leniency in sentencing appellate review  92, 94–96, 103 conscious leniency  80 generally  79–80, 103–04 judicial views on  84–88 summary offences  85 undue leniency  80 Leonard v Houston  197 Lesbian Rule, Aristotle’s metaphor  61 Leverick, F  200, 219, 235, 239 life sentences  5, 47–49, 50, 158, 164, 185n Lin v HMA  164 Lovegrove, A  80 Lowe v HMA  165 M McClellan J  94 McCluskey v HMA  49 McConville, M  216 McConville, M and Marsh, L  200, 202, 203, 208, 211 McCormick, P and Greene, I  9 McCormick v HMA  166 McGill v HMA  47, 166 McHugh J  213 judgment in Markarian  30–31, 35–43, 223n Mack, K and Roach Anleu, S  219–20, 222, 233 Mackay J  26 Mackenzie, G  9–10, 15, 17, 20–21, 137, 139, 183, 230, 232, 233 Mackenzie, G and Stobbs, N purposes of sentencing  113 McLachlin CJ  70 McNeilage v Griffiths  92 Maguire, N  15, 183 Maher, G, et al Compensation Orders in the Scottish Criminal Courts  6–7 Mahoney JA  54 Malleson, K  14 mandatory sentences see life sentences; statutory minimum sentences Mandie JA  235 Manson, Allan  28, 37–38, 99 Manson, Allan, et al  27, 29 Markarian v The Queen  33–43, 46, 47, 185, 203, 223, 225–26 criticism of Markarian analysis  133 Kirby’s dissent  43–45, 48, 50 McHugh’s judgment  35–43, 223n Masson v HMA  167 maximum sentences sheriff court  58 mental health of offender  85–86 mercy in sentencing appellate review  92, 94–96, 103 Australia  81 criteria for application  83–84 definition  83

276  England and Wales  81–82 generally  80–83, 103–04 judicial views on  84–88 mercy as justice  82 mitigation distinguished  83, 87 moral monism  117 phronesis  84 value pluralism  117, 179 Meyer, LR on justice  58, 103 Millie, A, et al  8, 96, 104, 175, 232 minimum sentences see statutory minimum sentences Mitchell v HMA  167 mitigation consideration in sentencing  52, 93, 116, 119, 190, 191 Definitive Guideline  78 and equity  88, 93 individualised sentencing  69, 88, 93 instinctive synthesis model  102 judicial conceptions of sentencing process  181 mercy distinguished  83, 87 mercy in sentencing  81–82, 87 offender’s remorse  209–10 progressive loss theory  89n, 90 public attitude to  80 research into  5 sentencing guidelines  141, 150 staged (two-tiered) sentencing  30, 31 statutory minimum sentences  66 value pluralism  119 monism Berlin’s view  109, 112–13, 177–79 consistency in sentencing  176, 180 England and Wales  176–79, 180, 192 Enlightenment view  109 generally  108–09, 117–18, 180 incommensurable values  109–10 sentencing guidelines  117, 176–79, 192 subordination to a summum bonum  109 and totalitarianism  112, 112n value pluralism compared  107, 108–09, 117–18 Morgan, N statutory minimum sentences  70, 72 Morgan, N and Murray, B on consistency in sentencing  99 Mulato v The Queen  41 Muller, EL  82, 88 Munro, M, et al  162 Munro v HMA  159–60 murder guilty plea discounts  214, 218 High Court jurisdiction  13–14 statutory minimum sentence  64 Murphy, JG mercy in sentencing  82 Murray v HMA  163, 200–203, 225–26, 234

Index N natural justice rule of law doctrine  55 Neill v HMA  165 New Zealand guilty plea discounts  236–37, 238 sentencing discretion  136 sentencing guidelines  167 Nicholls, C et al  213 Nicholson, CGB  163, 215 Norbis v Norbis  107 Nussbaum, MC  178–79 equity  60, 61, 88 sentencing guidelines  121, 176, 177 value pluralism  110, 111 O objective circumstances of offence  32 objectivity  62–63, 99 O’Donnell, I  174 offender age  190 alcohol or drug dependency  86 disadvantaged  85–86 mental health problems  85–86 personal circumstances  134–35, 139, 170 personal responsibility  29, 58, 170 protection by rule of law  58 remorse  84 risk assessment reports  190 sentencing triad  139–40 Ogilvie v HMA  163–64 O’Malley, T  1, 66, 150, 151, 155, 159, 168–69, 199, 231, 232, 237 open mindedness as pluralist virtue  120 Orders for Lifelong Restriction Scotland  47, 49 Osborne, Lord  76–77 P Padfield, N  115, 149 Pannick, D  2 parsimony consideration in sentencing  103, 116 Paterson v HMA  46–47 Paton v HMA  77n Perry, R  174 Petch and Foye v HMA  47–49 Philip, Lord  46, 185–86n, 200, 218n, 239 Phillips, Lord  80 phronesis (practical wisdom) alternatives to phronetic synthesis  142–54 Dreyfus model of learning  127 episteme and techne compared  124–25, 127, 133 experience  95–96, 103, 126–27 generally  1, 11, 59–60, 124–27, 229 instinctive synthesis model  59–60, 103, 138, 140, 203

Index intuitive decision making  129–31, 183, 229, 240 judicial experience  95–96, 102, 103, 203 phronetic synthesis  11, 107, 170–71, 179–80, 193–95, 222–23, 227–33, 239–40 principled sentencing  170–71, 194 Scotland  170, 192–93, 227–40 sentencing, generally  133–40, 179, 187 sentencing as an art  136–40 value pluralism  123–24 virtuous judge  60, 61 Plato Laws  88 pluralism see value pluralism poiesis  125 Pomerance, Justice Renee  79–80 Posner, Judge Richard  58, 61 internal complexity  152–53, 192 intuitive decision making  132–33 judicial decision making  96–97, 101 postmodernism  62–63 practical wisdom see phronesis praxis  125 premeditation  88 previous convictions as aggravating factor  88–89, 89n consideration in sentencing  52, 96–97 recidivist sentencing premium  89, 104 previous good character consideration in sentencing  75, 119 Price v HMA  165, 172 prison overcrowding statutory minimum sentences  69, 103 Prisoners and Criminal Proceedings (Scotland) Act (1993)  47 proportionality Canada  25, 28–29 instinctive synthesis model  43, 46 sentencing guidelines  149 staged (two-tiered) sentencing  31–32, 40 statutory minimum sentences  70–71, 103 protection of society as a purpose of sentencing  52, 98, 113, 116 sentencing triad  139–40 psychiatric reports  190 public interest, sentencing in  97 punishment as a purpose of sentencing  26, 113 sentencing guidelines  118–19 staged (two-tiered) sentencing  37–39 Purvis v Macdonald  182 R R v Astill (No 2)  137 R v B  79 R v Bernard  83, 160 R v Bibi  3 R v Blackshaw  147, 153 R v Brooke  147 R v Charles  79 R v Cleverley  79 R v Corran  78

 277

R v Engert  53 R v Everett  213 R v Geddes  31 R v Gray  239 R v Hamilton  28–29 R v Harman  235 R v Healey  146 R v Height  147 R v Howland  137 R v Ipeelee  26–27 R v Jabaltjari  235n R v Jurisic  137 R v Kane  84 R v Lacasse  26–27 R v Larcombe  78 R v Lee  167 R v Lloyd  67n R v LM  136 R v M (CA)  26–27 R v McDonnell  38–39 R v McDougall and Collas  41 R v Marcantonio  147–48 R v Martin  144 R v Mooney  79 R v Nasogaluak  115 R v Nur  70 R v Oosthuizen  144 R v Peters  144 R v Rushby  53 R v Shannon  239 R v Shelton  232 R v T  239 R v Thomson  46, 197 R v Turner  90 R v Whyte  94 R v Willaert  26 R v Williscroft  30–33, 44, 115, 185 R v Wise  137 R v Young  31–33, 44 racial aggravation  88 rape aggravating factors  90–91 equity in sentencing  76–79 guilty plea discounts  208–09, 213, 214, 217 High Court jurisdiction  13–14 Raz, J  57–58, 97–98 RB v HMA  46, 185–86n, 200, 218n, 239 realism as pluralist virtue  120 recidivism as aggravating factor  88–89, 89n guilty plea discounts  198 sentencing premium  89, 104 Rees v R  101 rehabilitation and judicial discretion  58 moral monism  117 prospects of  126 as a purpose of sentencing  26, 29, 52, 98, 113, 114, 116, 179 testimony and character references  70–80

278 

Index

undue emphasis on  31 value pluralism  117 relativism, pluralism compared  111 religious prejudice, aggravation by  88 remorse guilty plea discounts  209–10 lack of  88 offender exhibiting  84 reparation or restitution as a purpose of sentencing  29, 113, 116, 179 repeat offences  88 Republic of Ireland  15 guilty plea discounts  237, 238 judicial conceptions of sentencing process  183 judicial discretion  168, 192, 225 phronetic synthesis  232 sentencing guidelines  121, 168, 203 value pluralism  179 research into judicial decision making Australia  8, 9–10, 182–83 Canada  8–9 case law  21–23 England and Wales  2–6, 180–81 generally  3–11 interviewing sentencers  13, 14–17, 20 judicial conceptions of sentencing process  180–94 judicial reluctance  14–15 Republic of Ireland  183 Scotland  6–8, 14–17, 181–82 retribution and judicial discretion  58 moral monism  117 as a purpose of sentencing  26, 52, 93, 114, 116 staged (two-tiered) sentencing  37–39 statutory minimum sentences  70 value pluralism  117 RG v HMA  170 Richardson, HS  177 risk assessment reports  190 Roberts, JV  149, 151, 153, 177 consistency in sentencing  230 sentencing discretion  141–42 Roberts, JV, et al  21, 22, 151 Roberts, JV and Bradford, B  214 Roberts, JV and Rafferty, A  151 Rose LJ  144 Rosett, Arthur  209 Ross, PE  189 Royer v The State of Western Australia  190 rule of law doctrine common sense  57 conceptions of  57–58 definition  54–55 entitlement to protection of  58 equity  57–64, 94, 97–98 fairness  58 formal and substantive branches  54n individualised sentencing  55, 102–03 instinctive synthesis model  51–57, 102–03 judicial discretion  55–58 natural justice  55

principles deriving from  55, 57 sentencing rationales  56 statutory minimum sentences  71, 103 Rumpff, JA  139 Russell v The Queen  29 Russell v Troup  165 Ryan v The Queen  32 S S v Banda  140 S v De Kock  139 S v Zinn  139–40 Sanchet v DPP (Cth)  52 Schön, DA  131 Schram, S  131 Schulhofer, SJ  221, 234 Scotland appellate review  155 charge and/or fact bargaining  216–19 consistency in sentencing  161, 162 discretionary life sentences  47–49 ‘double dipping’  216–19, 226 equity  51, 64–79 guideline judgments  155–60 guilty plea discounts  46, 47, 48, 194, 195–226, 233, 238–39 High Court of Justiciary  13–14 individualised sentencing  53, 154–55, 192–94, 227–28 instinctive synthesis model  46–49, 50, 102, 140, 225–26, 228–30 intuitive decision making  181–82, 229–30 judicial conceptions of sentencing process  181–82 judicial discretion  25, 154–75, 192–94, 227–40 judiciary, generally  13–14 Orders for Lifelong Restriction  47, 49 phronetic synthesis  170, 192–93, 227, 228–33, 239–40 purposes of sentencing  116 reactive sentencing system  193 research into judicial decision making  6–8, 181–82 sentencing guidelines  121, 123, 161–75, 192–93, 203, 232 Sentencing information System  7 sentencing jurisprudence generally  21 sentencing methodology  152 sentencing statements  169–70 sheriff courts  14 staged (two-tiered) approach  46, 47–49 statutory minimum sentences  64, 70, 72 submissions to the Appeal Court  160–62 summary courts  169 transparency in sentencing  169–70 value pluralism  118, 123, 179–80 Scottish Power Generation Ltd v HMA  166 Scottish Sea Farms Ltd v HMA  166 Scottish Sentencing Council  160–62, 193 Seifman, R  9, 236 Seifman, RD and Freiberg, A  217

Index sentencing as an art  26, 52, 80, 136–40 decision making process  151 deliberative system  190, 191, 194 discounts see guilty plea discounts discretionary nature  107–08, 141–42 guidelines see guideline judgments; sentencing guidelines judicial conceptions of sentencing process  180–94 matters influencing  114–15 phronesis see phronesis principled  141–42, 168–71, 194, 230–33 purposes  1, 29, 52, 93, 113–18, 179–80 sentencing triad  139–40 studies of judicial decision making  1–10 young offenders  113–14 Sentencing Commission for Scotland The Scope to Improve Consistency in Sentencing  161 Sentencing Council of England and Wales  146–51, 237–38 sentencing guidelines aggravating and mitigating features  78, 140, 141 aim  141, 144 appellate see guideline judgments Australia  45–46, 121, 167, 173, 203 Canada  167 category ranges  148–49 consistency in sentencing  148, 150, 153, 161, 162, 176–77, 191–93 Court of Appeal  146 criticisms of  145–46, 149–54 Crown Courts  144 departure from  143–44, 146, 154–55 drafting  131, 148 duty to comply  146–51, 167–69, 173–74 England and Wales  123, 142–54, 167, 172–79, 191–93, 203, 229–30, 232 flexibility  141, 145 generally  36, 141–42, 171–75 guilty plea discounts  149, 196–98, 202–03, 235–37 individualised sentencing  144 internal complexity  152–53, 192 and judicial discretion  144, 145, 146–51, 167–69, 174–75, 230 magistrates’ courts  144 moral monism  176–79, 192 New Zealand  167 offence, not offender prioritised  145–46 offence-specific  141, 143, 148–49 proportionality  149 punishment  118–19 Republic of Ireland  121, 168, 203 Scotland  121, 123, 161–75, 192–93, 203, 232 self-regulation  162–63 sentence-specific  141 source  141–42 value pluralism  121–22, 123 virtue-centred theory  171–72

 279

Sentencing Guidelines Council Definitive Guidelines  78–79, 143–46 guilty plea discounts  198, 211–12 monism  117, 175, 176–79 ‘Overarching Principles: Seriousness’  115 powers  143–44 replacement  146, 148 sentencing information systems intuitive decision making compared  131–35 sentencing rationales judicial discretion to select  56 rule of law doctrine  56 seriousness of crime assessment  126 sexual offences see also rape aggravating factors  90–91 Definitive Guideline  78 England and Wales  78 equity in sentencing  76–79 guilty plea discounts  208–09, 213, 214, 217, 218 violence  90–92 Shapland, J  114 Sharp v HMA  190 Shead, Sheriff C  198 sheriff courts jurisdiction  14 reporting decisions of  22 sentencing powers  14 shrieval appointments  14 Shklar, Judith N  61 Simpson J  45 Smart v HMA  165 Smith, Lady  73 Smith, Sheriff Irvine  228 social theories purpose of sentencing  116, 116n social work reports  52, 74, 84, 190 Solum, Lawrence equity  51, 58, 97, 102–03, 136 judicial guidelines  174 judicial intelligence  59 judicial wisdom  59, 95–96 over-inclusive legal rules  64 virtue-centred theory  10–11, 51, 58–61, 94, 171 sophia  60, 103, 125, 191 Sopinka J  38 South Africa phronetic synthesis  232 sentencing methodology  152 sentencing triad  139–40 statutory minimum sentences  69–70, 71, 103 Spence v HMA  163, 164, 197–200, 203, 204–06, 222, 225–26, 233–35, 237, 238 Spigelman CJ  41, 46, 72, 99, 137, 197

280  staged (tiered) sentencing aggravating and mitigating features  30, 31 Australia  30–37 generally  229 guilty plea discounts  47, 48 illusory nature  39–40 judicial conceptions of sentencing process  184 judicial discretion  32–33 mixed intuitive-structured approach  187–91 proportionality  31–32, 40 risks associated with  35–43, 48–49, 50, 151–52 Scotland  46, 47–49 sentencing guidelines  148, 151–52, 154, 175, 176–79 transparency in sentencing  35, 39–40, 41, 229 stare decisis principle  62 starting-point method  37–39 statutory minimum sentences Australia  69, 70, 71–72, 103 Canada  66n, 70 and crime rates  66–67, 71 deterrent effect  70, 71, 103 drugs offences  64–66 equity  69–72 fairness  68–69, 70, 103 firearms offences  64, 70 generally  51, 64 inflationary drift  69, 103 iniquity  69–72 knife possession  66–69, 70 mitigating factors  66 murder  64 police view  67 prison overcrowding  69, 103 proportionality  70–71, 103 rule of law doctrine  71, 103 Scotland  64, 70, 72 South Africa  69–70, 71, 103 Steiker, CS  82–83, 88 Stobbs, N  229 Strauss, A and Corbin, JM  18 Strawhorn v McLeod  196 Street CJ  53 structured sentencing see staged sentencing subjectivism and pluralism compared  111 suicide, attempted  85 Sully J  137 summary courts guilty plea discounts  215 leniency in sentencing  85 summum bonum moral monism  109 Sutherland v HMA  220n Sweden sentencing guidance  156–57 T Tamanaha, BZ  57, 59, 63 tariff Scottish courts  64 unofficial  62

Index Tarling, R  115 Tata, C  7, 36 Tata, C, et al  7 Tata, C and Hutton, N  7–8 techne  124–25, 127, 176, 191, 192 Terblanche, SS and Mackenzie, G  69–70, 71 testimonies, consideration in sentencing  70–80 Thomas, DA  146, 149, 150, 151, 153, 154, 169, 170, 179 Thomas and Waddell v HMA  78 Thomson, D  48 Thomson, Sheriff Nigel  228 Thomson v HMA  218n Tombs, J  8, 64, 80, 181–82, 183, 228 Tombs, J and Jagger, E  8, 96, 104 Tonry, M  36, 71, 104 totalitarianism and monism  112, 112n Townsley v HMA  166 Trajkovski v The Queen  41 transparency in sentencing generally  39–45, 54, 102, 104–05, 151, 156, 169, 184 instinctive synthesis model  34–35, 41, 43–44, 102 Scotland  169–70 sentencing statements  169–70 staged (two-tiered) sentencing  35, 39–40, 41, 229 Turnbull, Lord  75 U uniformity see consistency and uniformity in sentencing United States judicial conceptions of sentencing process  188–89 sentencing grids  36, 40, 152, 153 V value judgments in sentencing  52, 107–08 value pluralism Australia  179 Berlin’s theory of  1, 11, 107–13, 108n, 116–17, 124, 135 Canada  179 incommensurability of values  109–10, 120–23 judicial discretion  107–13, 117 monism compared  107, 108–09, 117–18 phronesis  123–24 pluralist virtues  120–22 pluralist/monist dichotomy  111 relativism compared  111 Republic of Ireland  179 Scotland  118, 123, 179–80 in sentencing, generally  113–22, 179–80 and sentencing guidelines  121–22, 123 subjectivism compared  111 value ranking  122 Veen v The Queen (No 2)  41, 52, 53 victim statements  190 violence as aggravating factor  88–93 armed robbery  91

Index guilty plea discounts  214 sexual attacks  90–92 virtue-centred theory equity  51, 58–61 generally  171–75 instinctive synthesis model  10–11 judicial discretion  58–60 judicial virtue  60 non-virtuous sentencers  94–96 sentencing guidelines  171–72 virtuous decision  60 virtuous judge  60, 61, 103 virtues, pluralist  120–22 von Hirsch, A  83, 107 W Wagner J  27 Waldron, J  55 Walker, N  83, 137 Walker, N and Padfield, N  98 Walsh v HMA  73 Warner, K  45–46 Wasik, M  145, 148 Weinberg JA  41 Weininger v The Queen  54

Welsh, Sheriff T  224n Wheatley, Lord  228 Williams, B  111, 112 Willis, JE  222 Windeyer J  81 wisdom Dreyfus model see Dreyfus model of learning judicial  59, 94–96 practical see phronesis theoretical (sophia)  60, 103, 125, 191 Wong v The Queen  33, 34–35, 39, 44, 45–46, 53, 173 Woolf, Lord  79 Wright v HMA  165 Y Yoshino, K  153 Young, W and King, A  149 young offenders  113–14 Z Zimring, FE  71

 281

282