Tribute to a great sentencing conference – 21 years on

Writing in a very recent post about Professor Cyrus Tata’s fantastic new book, Sentencing: A Social Process reminded me of a great sentencing conference he and his colleague Professor Neil Hutton organised exactly 21 years ago at Strathclyde University in Glasgow. It was, to the best of my knowledge, the first major international conference on sentencing, in recent history at least. How many attendees they initially hoped to attract I don’t know, but they must have been gratified when more than 120 turned up from all over the world. People came from several continental European countries, the United States, Australia, Canada, China and, of course, from many parts of Britain itself.

During that three-day conference in June 1999, we had an extraordinary array of lectures, presentations, seminars and discussions on all aspects of sentencing, many of them given or led by top international scholars in the field. A selection of the papers delivered was later published in book form: Tata and Hutton (eds), Sentencing and Society: International Perspectives (Ashgate, 2002). It is still available from Amazon in kindle, hardback and paperback forms, and well worth having because it has so many contributions of enduring value. Of course, there was a social side as well. I recall one particularly enjoyable evening at Babbity Bowster, a great Glasgow institution I would never fail to revisit when given the opportunity. Throughout the conference, we were treated with great hospitality, and the general atmosphere made it a tremendously enjoyable as well as a highly instructive event.

The conference had far-reaching results in some important ways. Before that, most of us had probably heard of one another and, as already mentioned, the attendees included some leading scholars with whose work we would all have been familiar. But the conference gave us an opportunity to get to know one another. Many of us have remained in contact ever since. There was a second such conference in 2002, also in Strathclyde, and that, if I recall correctly, was even better attended. However, as a result of the 1999 conference, what started as a small group of us began to meet for informal seminars once or twice a year in different locations throughout Europe. We now meet mainly as a specialised group of the European Society of Criminology, and our number has grown to more than 100. Over the years, we have had great meetings at Strathclyde itself, Leiden, Oxford, Dublin, Barcelona, Como, Reims and elsewhere.

Let me just mention two further aspects of the 1999 conference, one with an Irish dimension, the other of more universal application.

On the first afternoon of the conference we had a presentation on the Scottish Sentencing Information System which had just then been established. Neil and Cyrus, working in close collaboration with senior members of the Scottish judiciary, had played a leading role in its development. The idea was to provide judges, by means of an easily searchable computerised data base, with reliable information on existing sentencing practice for the more commonly prosecuted offences. It occurred to me that something similar might be useful in Ireland and so, on my return, I began to canvass the views of judges, government officials and others. It took a long time to get anywhere with it, but eventually, thanks largely to the energy and enthusiasm of Judge Susan Denham, later Chief Justice, a steering committee was put together. We managed to get some funds, very limited it must be said, but better than nothing and, to cut a long story short, the Irish Sentencing Information System was established. I was rather chuffed at being able to suggest that it should be called ISIS. She, after all, was the ancient goddess of life and magic but, for reasons readers will understand, that abbreviation had to be quietly dropped in later years. Cyrus Tata gave us great help in establishing the system and came to Dublin for meetings on a few occasions. For an excellent account of the system while it was operating, see Brian Conroy and Paul G Gunning, “The Irish Sentencing Information System (ISIS): A Practical Guide to a Practical Tool” [2009] Judicial Studies Institute Journal 37-53. The Judicial Council Act 2019 now provides for the establishment of a Sentencing Guidelines and Information Committee. So, it took 20 years, but when God made time he made lots of it.

Something else I took away from the 1999 conference was a renewed appreciation and understanding of the value of comparative law. Most of the attendees had already published on sentencing in their own jurisdictions. Reading some of that work, especially from faraway countries, one might be inclined to conclude that it had little relevance to one’s own system. However, when a group of people from different parts of the world are gathered together in a seminar room or a pub, or huddled together during a coffee beak, it quickly becomes apparent that, despite structural, procedural, linguistic and terminological differences, we are all grappling with the same fundamental questions: Why punish offenders? What range of sentencing options should be available to the courts? How much discretion should judges have? What are the most essential distributive principles of sentencing? Is there a role for formal guidelines? Issues like these may not be common to all mankind (in the words of Justinian’s Digest), but they are of central importance in most countries. This, of course, is the key insight of the functional theory of comparative law which eschews comparison of formal legal texts in favour of examining how similar practical problems are addressed in different countries.

This brings me to what was meant to have been a retirement project, except that retirement did not happen quite as planned (I am still around NUI Galway like a bad penny). However, the project remains live and it is a book to be entitled A Common Law of Sentencing. Courts throughout the common-law world, and often beyond it, apply the same general sentencing principles, although they may differ in emphasis and scope of application. What I aim to do in the book is bring together key extracts from 30 to 40 leading cases from different jurisdictions illustrating those principles, with a commentary on each. The criteria for selection are that the case should provide a clear and authoritative statement of the relevant principle, preferably accompanied by a critical analysis, and that it should crystallise the law rather than merely provide a litany of earlier precedents. The project is at an early stage of development. Right now, Canada is leading the field in terms of selected cases. There is one sure Irish candidate, Deaton v Attorney General [1963] I.R. 170, but there will certainly be others. Once my old hunting ground, the IALS on Russell Square in London, re-opens I will be back prowling among its superb collection of printed law reports looking for inspiration. However, it would be great to have some material (which can be translated) from continental European countries as well, or even from further afield. Suggestions about cases that might be included, from common-law jurisdictions and elsewhere, will always be welcome.

So, if big, if belated, thank you to Neil and Cyrus for having organised such a wonderful conference that had such enduring results, and to all who attended back in those days before we ever heard of social distancing.

Finally, you might ask if we could have a similar conference in Ireland. Well, on that score, there is bad news and good news. The bad news is that I had been planning a conference, admittedly on a much smaller scale, in Galway in spring of this year. It was to have concentrated on sentencing guidelines, because of the aforementioned Judicial Council Act, and would have had some superb contributors who had kindly agreed to partcipate. However, it had to be cancelled for obvious reasons. The good news is that it will take place, hopefully in the not-too-distant future, once life is back to normal, or something close to it. In my undergraduate days (roughly around the time of the Beatles’ last LP), the Dean of the Arts Faculty used to comfort us before examinations with the advice that “there is no such thing as failure, only success deferred.” So, roll on a vaccine!

More on jury trial during pandemic

There have been a few earlier posts here on jury trial and how it might function during the present pandemic. However, readers will be interested to know that Mark Coen (our leading expert on all matters connected with juries and jury trial) has a great blog post “Trial by Jury in a Pandemic” which deals with several legal and practical aspects of the matter. It can be accessed at Incidentally, the blog on which he published it is part of a TCD initiative entitled “Covid-19 Law and Human Rights Observatory” which is always worth consulting from time to time because it has many very good and interesting contributions.

A new book on sentencing by Cyrus Tata

David Downes once described criminology as a rendezvous subject, meaning that it is a crossroads at which several different disciplines meet. The same holds true of sentencing – which some might see as a sub-discipline of criminology. Sentencing scholarship, viewed in its totality, consists of contributions by sociologists, philosophers, lawyers, psychologists, political scientists, economists, historians and, of course, criminologists (though many of them will have started life in one of the other disciplines). I have often been inclined to think of sentencing as an estuary into which many different rivers – law, sociology and other disciplines – flow. On reflection, however, this fluvial metaphor is not entirely apt because, while all these disciplines make great contributions to sentencing scholarship, they seldom mingle very well, and sometimes not at all. The sheer volume of research being produced within the various disciplines makes it almost inevitable that each of us must, reluctantly or otherwise, stay largely within the confines of our own. Yet, if we are serious about what we do, we must at times be willing to cross boundaries and explore other disciplinary and intellectual perspectives. After all, every once in a while, a work appears that mounts a robust challenge to some of our most fundamental assumptions about sentencing. Cyrus Tata’s new book, Sentencing: A Social Process. Re-thinking Research and Policy (Palgrave Macmillan, 2020) is one such work.

At one level, this book may be situated within a body of scholarship that treats sentencing primarily as a process or a social practice. John Hogarth’s Sentencing as a Human Process (University of Toronto Press, 1971) was a major, pioneering work that investigated the sentencing behaviour of magistrates, largely through a series of in-depth interviews with a selection of them. It devoted several chapters to the manner in which judges (magistrates) interpret information furnished to them as part of the sentencing process. More recently, a much-cited book chapter by Neil Hutton (a colleague of Cyrus at Strathclyde Law School) adopted Pierre Bourdieu’s notion of habitus as a conceptual framework for the purpose of “understanding the sociological distinctiveness of sentencing as a form of legal decision making” (Hutton, “Sentencing as a Social Practice” in Armstrong and McAra (eds), Perspectives on Punishment: The Contours of Control (Oxford University Press, 2006), chap. 9). Viewed in the context of more conventional legal theory, Cyrus’s book may be interpreted as offering an “external” critique (of sentencing decision-making in this instance) in contradistinction to the “internal” point of view championed by H.L.A. Hart in The Concept of Law which is essentially about trying to understand the law from the perspective of those who participate in the practice of it, and who accept the practice. See, for example, Grant Lamond’s essay “Methodology” in the even more recently published Cambridge Companion to the Philosophy of Law (2020), p. 21. Having said this, Cyrus is also, to a significant degree, offering an internal perspective, something he is admirably qualified to do because of the considerable amount of empirical research he has conducted during the past 25 years on several aspects of the criminal justice system including sentencing (in particular), legal aid and judicial support systems.

Sentencing: A Social Process is a fairly short book (about 170 pages) but it is a rich and intricately argued piece of work. All I can do here (and this is not intended as a review, just a description) is to dwell on a few of its more salient features. The book does not set out to provide a detailed blueprint for an ideal sentencing system. Rather its purpose is to offer a sustained critical analysis of how sentencing decision making is conceived and represented. Cyrus starts by identifying two major trends in modern sentencing scholarship – what he calls the legal-rational tradition and the judicial defensive tradition. The main concern of the legal-rational approach is to promote a sentencing system imbued with values such as fairness, equality and consistency. This, in turn, calls for judicial sentencing powers to be limited and structured through, for example, the adoption of guidelines. The judicial defensive tradition treats sentencing as an exercise in practical wisdom which judges, by virtue of their training, experience and expertise, bring to bear on the facts of each specific case. Proponents of this view typically believe that sentencing judges should be conferred with fairly extensive discretion and allowed to get on with the job. Cyrus argues, quite convincingly, that both of these traditions share a common premise or assumption which he calls autonomous individualism.

This notion of autonomous individualism and its deficiencies as an explanatory tool form a connecting theme throughout the book. As used in the book, the term applies in at least three different contexts. First there is the idea of the sentencing judge as an autonomous individual, and we shall return to that. Secondly, other professionals, such as lawyers and probation officers, who deal with defendants at various stages of the process usually work in isolation from one another and, to this extent, see themselves as autonomous individuals. Thirdly, the facts of a case tend to be atomised (although he does not use that term) in the sense of being presented as discrete autonomous factors to which rules and principles can be applied in order to arrive at an appropriate sentence. Instead, Cyrus argues (p. 4) that sentencing is “interpretive, processual, relational and performative.” He is certainly right to stress that it is “processual” because all too often it is imagined as solely a judicial task, something that begins and ends with the judge, again an autonomous agent. Yet, as he writes (p. 59):

“The sentencing agenda is to a considerable degree pre-determined and shaped by non-judicial actors. If it is the case that since sentencing (rather than the determination of guilt) is the key decision of the court which all actors build towards, then it means that we should see the reality of sentencing work as a multi-professional, multi-disciplinay collaborative process.”

This is an argument that some of us have long been making, but Cyrus presses it further than anyone has done before. To phrase the matter in simple terms, sentencing is one point along a continuum (the criminal process) that is punctuated by key decisional moments or episodes most of which are discretionary in nature, and the outcomes of which will strongly influence, if not determine, what sentence, if any, is ultimately imposed. Earlier decisions on, for example, prosecution, charge and mode of trial will shape the task and the decision of the sentencing judge. Further, in performing that task, the judge will often be provided with information or advice, which may (or may not) prove influential, in the form of professional reports from probation officers, psychologists and others. And then, of course, there is the input of counsel, and especially defence counsel who will make a plea in mitigation. Nor is sentencing always, or even commonly, the end point of the continuum. Others must take responsibility for implementing the sentence and they, too, are often empowered to exercise discretion. Thus understood, sentencing is, as Cyrus reiterates throughout this book, a truly collaborative exercise.

The various decisions that influence or determine the sentence ultimately imposed (and perhaps the implementation of that sentence) are made by a range of professionals including lawyers, judges, probation officers and expert witnesses. Those of us who describe ourselves as professionals seldom pause to reflect on what exactly a “profession” is. In fact, there is an extensive academic literature on the topic and Cyrus has mastered it very thoroughly, as reflected in Chaps 4 and 5 of the book. (These chapters will be of great interest to any lawyer, irrespective of their specialism, wanting to reflect on the social and ethical dimensions of their calling). Two conceptions of “profession” are identified, the trait model and the proprietorial control model. The trait model focuses on the characteristics of professionalism such as educational and training requirements, competitive entry, being subject to regulation, and expertise in applying abstract or specialised knowledge to concrete cases or factual matrices. The proprietorial control model, as the name suggests, is more concerned with “owning” an occupation or area of work. Viewed from this perspective, professionalism also involves controlling the relationship between facts and rules. Chapter 5 of this book, entitled “The Humanising Work of the Sentencing Professions: Individualising and Normalising” has many important insights into how the various professions work to present, as far as they can, an ideal defendant for sentencing. I found particularly interesting what Cyrus had to say about remorse. As he notes, much of the academic literature on this topic concentrates on how to identify true remorse. Yet the presence, or even the expression, of remorse can also operate to validate or legitimate the imposition of punishment. The sentencer can take comfort, as it were, from the defendant’s remorse as a justification for inflicting the suffering or deprivation that the sentence entails.

This book, for the most part, is written in a very condensed style that demands concentration, but the effort is always well worthwhile because of the paradoxes exposed and the insights offered. However, Chap. 6 is written in an entirely different register and that, I suspect, is because it was written with some passion. The chapter is entitled “The Rise of Technology and the Demise of the Sentencing Professions?”. Two pieces of background information are necessary to understand the argument made in this chapter. First, back in 2005, the sentencing world was taken by storm with a book by Norwegian criminologist, Katja Franko Aas, entitled Sentencing in the Age of Information: From Faust to Macintosh (London, 2005). Her argument was broadly within the new penology and actuarial justice schools of thought. Essentially, she was lamenting the decline of individualised justice and judicial discretion and their replacement with various technology-based strategies that were preoccupied with segmented data. Jacqueline Tombs took up this theme is another important contribution (“Telling sentencing stories” in Carlen (ed), Imaginary Penalties (Willan Publishing, 2008). The second background issue relates to the development of the Scottish Sentencing Information System (SIS) which was undertaken on the initiative of senior members of the Scottish judiciary in the early 1990s (and described by Cyrus, who had a key role in it, in this book and elsewhere). The SIS rightly attracted wide international attention and admiration and was, indeed, the model for the Irish Sentencing Information System while it lasted. Its value lay, first, in the large number (many thousands) of previous sentencing decisions that were analysed in depth and, secondly, in the various levels at which the system could be interrogated by judges. Rather than merely providing raw data on, for example, the number of sentences of a specified level of severity previously imposed for a given offence (though it did that too), it allowed users to dig deeper and get a full account of each case. This was vitally important. Knowing the ultimate sentence imposed in any case is of little value unless one also knows, for example, if the offender was being sentenced simultaneously for several offences, if he or she had a criminal record and, if so, for what, and other relevant matters.

The arguments made by Franko Aas, Tombs and others were, of course, very good ones as far as they went. Indeed, they are probably even more relevant today with the growing use of algorithm-based decision making within the criminal justice system, about which there is now a voluminous academic literature. However, both Franko Aas and Tombs had pointed to the SIS as an instance of the phenomenon they were describing and criticising. In a powerfully written chapter (6) of this book, Cyrus rebuts this argument with tremendous clarity and persuasiveness. First, the SIS was intended to be descriptive rather than normative in nature. Secondly, and most importantly for the purpose of his argument, the SIS presented “whole case” accounts rather than a set of fragmented, decontextualized data. This chapter should be required reading for anyone engaged in the development of sentencing guidelines in this jurisdiction.

I will just pick out a few other important points made throughout in this book, including in the final chapter which recommends new directions in policy and research. It observes that the neat distinctions we tend to draw between discretion and rules, and rules and facts, are often illusory. Rules often phrased in an open-ended way that invites the exercise of discretion by including terms such as “reasonable” or “dangerous”. As to the distinction between rules and facts, Cyrus writes (p. 76):

“The rule depends on the context in which it is implemented: the specific case facts. So naturally one turns to the case “facts”, only to find that what count as “the relevant facts” depend on “the legal rule“. Thus, the scholar of decision-making is required continually to shuttle between the two (the rules and the facts), unable to apprehend what determines the decision.”

The many valuable recommendations made in the final chapter include a plea for much greater concentration on “typified whole case stories”; the need to rethink the concept of “efficiency” given that the output of the justice system is, by definition, justice and this should take precedence over case processing and management; the need for a deeper empirical understanding of decision-making; taking the concept of imprisonment as a last resort seriously which means that the seriousness of the current offence should be the sole criterion; rethinking legal aid in a way that would reward defence lawyers for helping offenders to exit the criminal justice system, mainly by referring them to agencies that can assist with the social and personal problems that have contributed to the offending behaviour.

Meanwhile, those of us who are “old and grey and full of sleep, and nodding by the fire” will probably continue to pursue the legal analytical approach, concentrating on sentencing law and practice, and the policies underpinning it. But the good news is that there is a growing cohort of young scholars, in Britain in particular, who are now doing and will continue to do fantastic work in bringing sentencing research to a new level. Then there are initiatives such as the Sentencing Academy ( which is dedicated to informing public debate about sentencing and creating a bridge between sentencing experts, policy makers and the public in England and Wales. Back in those halcyon days before Covid-19, I attended a great seminar hosted one evening by the Academy at University College London on “Sentencing Reform in the New Parliament.” Actually, it was only last January, but as Louis MacNeice might say, if he were still around, it now seems “so unimaginably different and all so long ago.”

These younger scholars will find much to inspire them in Cyrus’s book. But so will everyone working in the field. All too often, in our enthusiasm to suggest how the sentencing system should be “fixed”, we neglect to consider the true inner nature and collaborative quality of the decision-making process we are aiming to improve. Sentencing: A Social Process is a great book which should, and doubtless will, encourage all of us to reflect more deeply on this fundamental issue.

UK Supreme Court dismisses appeal in “paedophile hunter” case

In Sutherland v Her Majesty’s Advocate (Scotland) [2020] UKSC 32, the UK Supreme Court dismissed an appeal by a convicted sex offender who claimed that the manner in which evidence of his conduct had been obtained by a paedophile hunter group and then used by the public prosecutor violated his rights under Article 8 of the European Convention on Human Rights (ECHR). The judgment is important for what it has to say about the scope of Article 8 and the relevance of Article 17 of the ECHR. The issue in the case is neatly summarised in Para. (1) of the judgment:

“This case concerns the use in a criminal trial of evidence obtained by members of the public acting as so-called “paedophile hunter(“PH”) groups, and whether this is compatible with the accused person’s rights under article 8 of the [ECHR]. PH groups impersonate children online to lure persons into making inappropriate or sexualised communications with them over the internet, and then provide the material generated by such contact to the police.”

The sequence of events forming the background to the case followed a very familiar, indeed now almost universal, pattern. The appellant (S) went on to an internet site (Grindr in this instance) and made contact with a person who claimed to be a 13-year-old boy although, in fact, the person was an adult decoy. S sent a sexually explicit photograph of himself to this person, and eventually arranged to meet. When he arrived at the appointed place (believing he was to meet a 13-year-old), he was confronted and detained by members of the PH group, and handed over to the police. The confrontation between members of the PH group and S was broadcast live on Facebook and later posted on other social media.

The offences of which S was convicted included “meeting a child following certain preliminary contact” contrary to s. 1 of the Protection of Children and Prevention of Sexual Offences (Scotland) Act 2005 which is very similar to the offence created by s. 7 of our Criminal Law (Sexual Offences) Act 2017. In respect of each offence in S’s case, the charge was one of attempt because he believed, wrongly as it transpired, that the person with whom he was in contact was a child. As noted in a previous post on this blog “Meeting a child for the purpose of sexual exploitation” (May 5, 2020), the same would probably apply here. I watched most of the Supreme Court hearing of this appeal on the internet and one interesting point to emerge was that in Scotland there has been an extraordinary high number of prosecutions resulting from detection by PH groups.

Counsel for S, if I remember rightly, had accepted that there was no entrapment in this case (presumably because the detection had not resulted from police action) but he did claim that it amounted to impermissible “virtue testing.” The Supreme Court held categorically (at [16]) that entrapment was not an issue in this case. It made no reference whatever to virtue testing (a topic that was analysed by the Canadian Supreme Court in R v Barnes [1991] 1 S.C.R. 449).

The appeal to the Supreme Court in this case was based on two compatibility issues as certified by the Scottish High Court. These were:

1. Whether, in respect of the type of communications used by the appellant and the PH group, article 8 [of the ECHR] rights may be interfered with by their use as evidence in a public prosecution of the appellant for a relevant offence; and

2. the extent to which the obligation on the state, to provide adequate protection for article 8 rights, is incompatible with the use by a public prosecutor of material supplied by PH groups in investigating and prosecuting crime.”

Article 8(1) of the ECHR states:

Everyone has the right to respect for his private and family life, his home and his correspondence.

Article 8(2) then provides that “[t]here shall be no interference by a public authority with the exercise of this right” unless the interference is in accordance with law and necessary in the interests of national security, the prevention of crime, “the protection of the rights and freedoms of others” and so forth,

Was Art. 8(1) of the ECHR engaged by the facts of this case?

Whenever a person claims to have suffered a violation of Art. 8 (or, for that matter, of Art. 9, 10 or 11 each which has the same structure as Art. 8), a court, whether a national court or the European Court of Human Rights (ECtHR), will begin by asking if the matter complained of comes within the purview of Art. 8. If satisfied that there appears to have been a breach of the Article, the court then proceeds to consider if the impugned act or omission was nonetheless justified, or at least permitted, by virtue of Art. 8(2). It is not uncommon for the answer to both questions to be “yes”, in which case no violation will be found. Sutherland, however, was one of those cases where the Supreme Court did not accept that there was any interference with S’s rights under Art. 8(1) at any stage of the criminal proceedings or, indeed, before they began. The Court stressed that what Art. 8 guarantees is the right to respect for private and family life, etc. S’s conduct did not merit such respect. Lord Sales, with whose judgment all other members of the Court agreed, said (at [31]):

“In my judgment, there are two reasons why the appellant’s rights under article 8(1) in relation to respect for private life and respect for his correspondence were not interfered with in the circumstances of this case: (i) the nature of the communications from the appellant to the decoy, whom he believed to be a child, was not such as was capable of making them worthy of respect for the purposes of the application of the ECHR; and (ii) the appellant had no reasonable expectation of privacy in relation to the communications, with the result that he enjoyed no relevant protection under article 8(1) as regards their disclosure to and use by the [prosecutor] and the other public authorities referred to above.”

The Court then dwelt on the positive obligation imposed by the ECHR on states parties to have effective measures in place for the prevention and criminalisation of sexual violence, abuse and exploitation. This is reflected in leading ECtHR judgments such as X and Y v The Netherlands (1986) 8 EHRR 235. In fact, it is the rights of victims and potential victims under Art. 8 (and also perhaps Art. 3) that mandate such positive measures. In this case, S’s conduct amounted to a clear violation of a Convention right of an intended victim and, as such, it did not merit protection. In this connection, the Court referred to Art. 17 of the ECHR, an article we seldom hear very much about, but which provides:

” Nothing in this Convention may be interpreted as implying for any State, group or person any right to engage in any activity or perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the Convention.”

In Lawless v Ireland (No.3) (1979-1980) 1 EHRR 15, the ECtHR had said:

“… the purpose of Article 17, in so far as it refers to groups or individuals, is to make it impossible for them to derive from the Convention a right to engage in any activity, or perform any act aimed at destroying any of the rights and freedoms set forth in the Convention .”

(See Emmerson, Ashworth and Macdonald, Human Rights and Criminal Justice, 3rd ed. (London, 2012), pp 147-148 et passim, which has a good discussion of Art. 17).

Art. 17 seems designed to deal with precisely the kind of situation that arose in Sutherland where a person claims Convention protection in respect of conduct that clearly violated, or had the potential to violate, the rights of others.

In Sutherland, Lord Sales concluded (at [40]) that “the reprehensible nature of the communications [between S and the decoy] is such that they do not attract protection under article 8(1). They do not involve the expression of an aspect of private life or an aspect of correspondence which is capable of respect within the scheme of values inherent in the ECHR.”

In light of this conclusion, there was obviously no need to consider Art. 8(2) which deals with permissible limitations on the right to respect for private life where a prima facie violation of Art. 8 has been found.

Although the Supreme Court did not draw this comparison, the idea that certain communications are inherently unworthy of respect or protection under Art. 8 of the ECHR bears some similarity to decisions of the US Supreme Court to the effect that certain kinds of expression, by virtue of their content, do not merit protection under the First Amendment which provides, in unqualified terms, that no law shall be enacted abridging freedom of speech or of the press. The US Supreme Court has so held in relation to child pornography, among other things. See, for example, New York v Ferber 458 U.S. 747 (1982).

Did S have a reasonable expectation of privacy?

Both the UK Supreme Court and the ECtHR have accepted that in interpreting the concept of private life in specific contexts, it is often relevant to ask if the applicant had a reasonable expectation of privacy. See, for example, Benedik v Slovenia (Application No. 62357/14. ECtHR 24 April 2018). In Sutherland, the Supreme Court had no hesitation in finding that S had no such expectation: “He could not reasonably expect that, where his messages constituted evidence of criminal conduct on his part, the recipient would not pass them on to the police.” [58].

Use by public prosecutor of material supplied by PH group

The second certified question related to the extent to which the use by a public prosecutor of material furnished by a PH was compatible with Article 8. The Supreme Court said that it could deal briefly with this issue, as indeed it did. Essentially, it held that since S’s Art. 8 rights were not violated in the first place, he could scarcely complain about the use of the material in the investigation and prosecution of the offences with which he was charged. The Court’s reasoning is encapsulated in these two passages:

“Clearly, therefore, in this case the state had no supervening positive obligation arising from article 8 to protect the appellant’s interests which would impede the [prosecutor] in any way in making use of the evidence about his communications with the decoy to investigate or prosecute in respect of the crimes he was alleged to have committed. On the contrary, in so far as positive obligations under art. 8 were engaged, the relevant positive obligation on the [prosecutor], as a public authority, was to ensure that the criminal law could be applied effectively so as to deter sexual offences against children. Contrary to the appellant’s argument, article 8 has the effect that the [prosecutor] should be entitled to, and indeed might be obliged to, make use of the evidence of the communications with the decoy in bringing a prosecution against him.” [64].

“Further, even where article 8 is applicable, a contracting state has a margin of appreciation as to how to strike a fair balance between the competing interests which are in issue. Since, in the present context, the state has a positive obligation to operate an effective criminal law regime to deter and punish persons who threaten to harm young children, there is no doubt that the use by the [prosecutor] of the evidence provided by the decoy for the purposes of the prosecution of the appellant under that regime involved no breach of any positive obligation owed to the appellant” [67].

Other issues that arose in the Scottish High Court

The Supreme Court agreed with two other findings by the Scottish High Court, even though they did not fall be determined in this appeal. First, even if there had been a breach of Art. 8, the use of the evidence provided by the decoy would have been justified under Art. 8(2). Secondly, even if a violation of Art. 8 had been found, it would not follow that the conviction should be quashed. Evidence obtained in breach of Art. 8 may still be used in criminal proceedings without any consequent breach of Art. 6 which guarantees the right to a fair trial. As the ECtHR has repeated held, when a breach of Art. 6 is alleged, the question is whether the proceedings, viewed in their totality, were fair. This is something we are apt to forget. It is not a question of holding, as might be true under a domestic constitutional regime such as our own, that once evidence is obtained in a certain way it should be excluded, and failure to exclude may justify quashing a conviction. Under the ECHR, the use of evidence obtained through torture, contrary to Art. 3, would render a trial unfair, but that would not necessarily (or even predominantly) be true of evidence obtained in breach of Art. 8. The convicted person might be entitled to a remedy, typically monetary compensation, for the breach of Art. 8, but that would not imply a breach of Art. 6.


The outcome of this appeal was scarcely surprising. What is, however, remarkable is the absence of any reservations on the part of the Supreme Court about the activities of PH groups, despite their prevalence throughout the United Kingdom (as the Court noted) and elsewhere. At no point did the Court expressly validate or endorse their activities but, on the other hand, it left no reason to doubt the entitlement of police and prosecutors to avail themselves of evidence collected by PH groups in investigating and prosecuting sexual offences and attempted sexual offences against children. One need only read the final sentence of Para. 67 (quoted above) where the Court unambiguously held that the use by prosecutors of materials provided by decoys involved no breach of any positive obligation owed by the State to accused persons. In fact, Para. 64, also quoted above, suggests that they might be obliged to use the evidence once it was furnished to them.

An important judgment on the right to silence

The rules governing the conduct of criminal trials must attempt to strike a balance between truth and justice. There is a view, mostly closely associated with Jeremy Bentham, that the sole purpose of any legal proceeding, civil or criminal, should be to discover the truth and that no competing considerations should be allowed to frustrate or impede the attainment of that objective. Bentham, especially in A Rationale of Judicial Evidence (1827), claimed that the essential purpose of any trial was “rectitude of decision”, another way of saying that it should, as far as possible, aim to discover the truth. However, even he accepted that this principle had to be subject to some constraints and, as befitted a utilitarian, he identified the avoidance of expense, delay and vexation as permissible limitations. But he would abolish any rules, especially exclusionary rules, that might hinder the quest for truth.

Of course, Bentham and others in the rationalist tradition such as Thayer and Wigmore had to accept that a trial is usually an exercise in reconstruction. The event to which it relates, say a murder, a serious assault or a burglary, will have happened sometime in the past. Nowadays, the typical interval between the event and the trial is probably longer than ever before. Further, there may be no independent witnesses of the event. The trial, therefore, entails an effort to construct an evidence-based narrative that corresponds as closely as possible to the event which had an independent existence of its own. William Twining summarises this correspondence theory of the truth as follows:

“Present knowledge about past events is typically based upon incomplete evidence; it follows from this that establishing the truth about alleged past events is typically a matter of probabilities or likelihoods falling short of complete certainty.”

(Twining, Theories of Evidence: Bentham & Wigmore (London, 1985) 13. See also John Jackson, “Theories of truth finding in criminal procedure: An evolutionary approach” (1988) 10 Cardozo L.R. 475, one of the great pieces of writing on evidence theory during the past half-century or so).

Neither in Ireland nor in any other country with a similar legal culture is the Benthamite doctrine fully applied. We hold that discovering the truth is a vitally important aim of the trial but that it is not the only value that must be observed. Legality, fidelity to the rule of law, procedural fairness and promoting “equality of arms” between defence and prosecution are among the other values that must, by our lights, inform the trial process. What Art. 38.1 of the Constitution of Ireland guarantees to every accused person is a trial in due course of law, not a trial that has the discovery of the truth as its sole and exclusive purpose. In this connection, however, it should be recalled that not all exclusionary rules are inimical to discovering the truth. Excluding evidence that is unreliable or that might seriously prejudice or distract a fact-finder may enhance rather than diminish the prospects of arriving at the truth.

It is seldom that appeal courts devote much attention, expressly at least, to the complex relationship between truth and justice, which is why the recent judgment of the Court of Appeal People (DPP) v S.M. [2020] IECA 170 (per Donnelly J) is particularly interesting. The appellant (M) was convicted of five counts of indecent assault committed against one of his sisters and one count of indecent assault committed against another sister. All the offences were committed during the period 1981 to 1987. He was given an effective sentence of three-and-a-half years’ imprisonment which was reduced somewhat on appeal, and that will be the subject of a future post. While M was detained by the GardaĆ­, he answered many questions at great length; in fact, the memoranda of interviews left to the jury ran to 78 pages. However, there were 13 questions he did not answer and these are conveniently described as “no comment” answers. The memoranda were suitably edited before being given to the jury so as to exclude all questions to which there had been no comment. This was done in accordance with the standard practice adopted in the wake of People (DPP) v Finnerty [1999] 4 I.R. 364. As the Canadian Supreme Court said R v Chambers [1990] 2 S.C.R. 1293, “it would be a snare and a delusion to caution the accused that he need not say anything in response to a police officer’s questions but nevertheless put in evidence that the accused clearly exercised his right and remained silent in the face of a question that suggested his guilt.”

In People (DPP) v S.M., defence counsel, when cross-examining the interviewing Garda, asked certain questions that might be interpreted as suggesting that M had answered all questions put to him while in detention. The following provides a flavour of the relevant part of the cross-examination (and here I abbreviate substantially, but the relevant part of the trial transcript is reproduced in the Court of Appeal judgment, pp 3 to 5):

“Q. So in essence what the gardai are saying to somebody is you have the right to silence, if you wish to exercise it; isn’t that correct?

A. Yes, that’s correct.

Q. And I think that I’m sure in your experience and you have considerable experience, there would have been occasions when a person has been detained and you would have arrested them and that particular person in light of the caution they have received and the advice and notice they received may have decided to exercise their right to silence, and opt not to answer any questions asked by An Garda Siochana?

A. Yes, that’s correct.

Q. But obviously in this case, it’s readily apparent from the lengthy interviews we’ve listened to, there were lengthy questions and lengthy answers given by [M]?

A. Yes, that’s correct.

Q. And he answered questions given by – by yourself?

A. He did, he answered questions freely.”

Earlier in the same cross-examination, there was the following exchange:

Q. Yes, okay, and I think to strip it all down and to boil it down, when [M] was asked about the allegations that were made individually by [the complainants] and what his attitude in response to those was he denied the allegations that were made by each of them, isn’t that correct?

A. That’s correct.”

At the conclusion of the cross-examination, counsel for the prosecution applied to re-examine the Garda so as to bring to the attention of the jury that there were further questions to which there had been “no comment” answers. The judge permitted this to be done. M was convicted, and the sole ground of appeal was that the trial judge had erred in admitting into evidence the “no comment” answers given by M to the interviewing Garda.

To be fair to defence counsel, at no point, as far as I can see, did he or she directly ask the Garda to confirm that M had answered every question put to him or that he had not exercised his right to silence in respect of any question. Indeed, the Court of Appeal made clear that it was not saying that defence counsel had set out to deceive in a deliberate manner. However, having analysed the relevant portion of the transcript, the Court concluded that there was “no doubt that this line of questioning was designed to emphasise to the jury that this was an accused person who did not rely on his right to silence in contrast to other persons who do so” (para. 17).

The principal question then for the Court of Appeal was whether M had suffered a violation of his constitutional right to silence as a result of the entire content of the interviews, including the “no comment” questions being made known to the jury. In normal circumstances, as already noted, revealing those questions and answers would be a breach of the right. However, the Court stressed that different considerations apply where an accused has led to the jury to believe that he did not exercise his right to silence. As the Court of Appeal phrased the matter (para. 21): “The issue that arises here is whether a misrepresentation in respect of the exercise of a constitutional right can or should be corrected”

It was submitted in effect on behalf of M that it should never be possible to cross-examine in relation to the refusal of a defendant to answer questions during interview. However, the Court rejected this argument, saying (para. 24): “That would completely trammel the purpose of a criminal trial: the search for the truth bound by rules of evidence. It would amount to a travesty of justice.” It continued (para. 25):

“In our view, both the common law and the statutory encroachments on the rules of evidence permit factually incorrect but relevant matters to be countered through cross-examination, by calling evidence in rebuttal or re-examination as appropriate.”

The Court went on to say that the constitutional right to a fair trial does not extend to the right to misrepresent the facts, and the constitutional right to silence does not confer an immunity from all appropriate references being made to the times the accused exercised it. Misrepresenting the exercise of the right to silence is akin to a waiver of the right to the extent necessary to rectify the misrepresentation:

“When an accused goes behind the agreed (or ruled on) editing of the interviews, then he has in effect waived his right to silence to the extent needed to correct the record. He cannot gain an advantage simply because he sought that advantage. The concept of a trial in search of the truth must permit the prosecution to correct an incorrect impression that has been conveyed to the jury.” (Para. 31).

Finally, and importantly, the Court stressed that in a case such as this, the measure adopted by a trial judge to correct an erroneous impression that may have been conveyed to the jury must be proportionate in light of the overall circumstances. Here there were only 13 questions, out of a very large number, to which there had been “no comment” answers. The Court said that in a situation like this, prosecution counsel must tread warily because it “is not the situation that the door once opened permits a flood.” The Court concluded (paras 37 and 38):

“We make it clear that in each case if a jury is misled as to the exercise of the right to silence, there must be an individual assessment of the proportionate interference with the right to silence by way of remedy. Not every case will require or permit the full questions and answers to be read to the jury. To adduce evidence of the fact that he did not answer every question will be sufficient in many, if not most, cases.

In the present case, the trial judge held in his decision on the discharge application that it was also relevant that the jury be told of the specific questions or otherwise they might wonder why those particular allegations were not put to him. We are not convinced that this on its own would be a relevant basis for admitting the evidence. However, we are satisfied that it was just and proportionate to have admitted the questions in light of the totality of the circumstances in which that evidence was given in this case and where the jury were told as to the reason for the admission and were warned not to draw an adverse inference.”

The appeal against conviction was therefore dismissed.

I am not here concerned with the merits of this decision. My knowledge of the case derives solely from what appears in the judgment. However, if the Court was right in concluding, in the passage quoted earlier from para. 17 of the judgment, that the line of questioning by the defence during cross-examination was designed to convey to the jury that M did not exercise the right to silence, then the legal conclusions drawn by the Court are surely logical. A party cannot, as the Court said, “approbate and reprobate”. (Henchy J. used similar language in Corrigan v Irish Land Commission [1977] I.R. 317 at 326, only there he had warmed to the theme: “The complainant cannot blow hot and blow cold; he cannot approbate and then reprobate; he cannot have it both ways”).

Here, there was no genuine countervailing right or value to justify a constraint on establishing the truth in relation to the particular matter at issue. There would have been such a right (to silence) had it not been waived (assuming, again, that the Court was correct in deciding that it had been). As the Court of Appeal said (para. 29), the right to silence “does not sit above” the right to a fair trial but is part of the right to a fair trial. But the right to a fair trial does not extend to a right to misrepresent facts, even where this has been done inadvertently. Any such misrepresentation or error can be, and usually is, easily corrected.

A rather similar situation arose in People (DPP) v Almasi (Supreme Court, 26 June 2020), about which there will be a later post, except that there the trial court was held to have erred in permitting the prosecution to edit out of the memoranda of interviews certain statements made by the interviewing Garda as part of the questioning. The application had been made and granted in the trial court on the basis that it would be unfair to the prosecution to allow those statements to be made known to the jury. The Supreme Court dealt at length with the matter but one way of interpreting its decision is that permitting such editing impeded the discovery of the truth by the fact finder in relation to the matter in question.

How many members need a jury have?

In a recent interview with Joshua Rozenberg on “Law in Action” (BBC radio), the Lord Chief Justice of England and Wales, Lord Burnett, mentioned two strategies being considered to address the backlog of jury trials while restrictions necessitated by Covid-19 remain in place. One is to allow “either way” offences (which tend to be at the lower end of the spectrum of gravity) to be tried by a judge and two magistrates without a jury. In England and Wales where the doctrine of parliamentary sovereignty prevails, such an arrangement could doubtless be introduced by statute, but it would scarcely be constitutionally permissible in this jurisdiction. The other strategy is to reduce the number of jurors, possibly to seven as was done during the Second World War (under the Administration of Justice (Emergency Provisions) Act 1939). In fact, similar legislation was introduced at the time in Scotland and Northern Ireland and, in Scotland with its tradition of a majority verdicts, a person could be convicted by a majority of five to two.

This prompts the question as to whether it would it be constitutionally permissible to reduce the number of jurors in criminal trials here in Ireland? As noted in an earlier post, the Constitution (Art. 38.5) requires jury trial for serious offences, but it does not define the term “jury.” The courts have been left identify the essential characteristics of a jury, and the most important of these is that it should consist of a representative cross-section of the community. Any numerical reduction that rendered a jury less than representative in this respect would therefore be unconstitutional. We are concerned here solely with the essential compositional qualities of a jury. Obviously jurors, once sworn, must also be objective, impartial and faithful to their oath to give a true verdict according to the evidence. Under the Juries Act 1976, a person must be a citizen to qualify for jury service which means that our juries are representative of the citizenry, though not necessarily of the entire community.

Even as matters stand, it is not strictly correct to say that a jury must have 12 members. That is the number that must be sworn in at the outset, with the possibility nowadays of three further “stand by” jurors being sworn at the beginning of a lengthy criminal trial. However, a trial may proceed with 10 or 11 jurors where, for some reason, one or two have to drop out. In England and Wales, a trial may proceed with nine jurors (Juries Act 1974, s. 16). Nobody seems certain as to why the composition of a petit jury came to fixed at 12. United States Supreme Court judgments are often a rich source of information on legal history and, especially, early English legal history. In Williams v Florida 399 U.S. 78 (1970), of which more later, the Court surveyed the various historical sources and concluded that fixing the number at 12 was “a historical accident” (p. 102).

Some early writers on the topic explained the adoption of the number twelve by reference to its biblical significance. Giles Duncombe, in his Trials per Pais, or the law of England Concerning Juries by Nisi Prius, &c, wrote:

“This number is no less esteemed by our own law than by holy writ. If the twelve apostles on their twelve thrones must try us in our eternal state, good reason hath the law to appoint twelve to try us in our temporal. The tribes of Israel were twelve, the patriarchs were twelve, and Solomon’s officers were twelve.”

This book, first published in 1766, was apparently “the most widely read treatise in colonial America concerning evidence and courtroom procedure” and, yes, it is available from Amazon. Duncombe seems to have forgotten about Judas or else he had forgiven him. He might also have noted the religious significance of other numbers such as seven (e.g. the deadly sins) or ten (e.g. the Commandments or the parable of the ten virgins in Matthew 25, though sadly only five of them were wise).

In Scotland, a criminal trial jury has 15 members, though a trial may continue with as few as twelve, if necessary. In any case, for a valid conviction, at least 8 jurors must be satisfied of the guilt of the accused. A Scottish Government Report (The Modern Scottish Jury in Criminal Trials) published in 2008 stated that it was not known how the number originally came to be fixed at 15.

Throughout the common-law world, a 12-member jury is the norm in criminal trials, at least in the sense that there should be 12 at the outset. Florida is a notable exception where a criminal trial jury has only six members, although 12 are required in a capital murder case. The constitutionality of this arrangement was unsuccessfully challenged in Williams v Florida 399 U.S. 78 (1970), even though in Thompson v Utah 170 U.S. 343 (1898), the US Supreme Court had held that under the Sixth Amendment, a federal jury meant “a jury constituted, as it was at common law, of twelve persons, neither more nor less.” However, by a majority of seven to one (only eight Justices took part), the Court in Williams held that a state law providing for a jury of six in a criminal trial did not violate the Fourteenth Amendment. Delivering the opinion of the Court, White J said (p. 100):

“,,, the essential feature of a jury obviously lies in the interposition between the accused and his accuser of the common sense judgment of a group of laymen, and in the community participation and shared responsibility that results from the group’s determination of guilt or innocence. The performance of this role is not a function of the particular number of the body that makes up the jury. To be sure, the number should probably be large enough to promote group deliberation, free from outside attempts at intimidation, and to provide a fair possibility for obtaining a representative cross-section of the community. But we find little reason to think that these goals are in any meaningful sense less likely to be achieved when the jury numbers six, than when it number 12 – particularly if the requirement of unanimity is retained. And, certainly the reliability of the jury as a factfinder hardly seems likely to be a function of its size.”

Marshall J. dissented, being of the view that the Fourteenth Amendment guaranteed a jury of 12 members.

Once Florida got away with a jury of six, Georgia decided to take a chance on five, but this time the Supreme Court was having none of it. Williams had been heavily criticised, with many academic articles, some of them empirically based, pointing out the superiority of a larger number of jurors. Much of this research was expressly taken on board by the Supreme Court in Ballew v Georgia 435 U.S. 223 (1978) where it unanimously held that a five-member jury deprived the petitioner of his constitutional right to jury trial. The Court gave five main reasons for so holding, the first being this:

First, recent empirical data suggest that progressively smaller juries are less likely to foster effective group deliberation. At some point, this decline leads to inaccurate fact-finding and incorrect application of the common sense of the community to the facts. Generally, a positive correlation exists between group size and the quality of both group performance and group productivity. A variety of explanations have been offered for this conclusion. Several are particularly applicable in the jury setting. The smaller the group, the less likely are members to make critical contributions necessary for the solution of a given problem. Because most juries are not permitted to take notes, memory is important for accurate jury deliberations. As juries decrease in size, then, they are less likely to have members who remember each of the important pieces of evidence or argument. Furthermore, the smaller the group, the less likely it is to overcome the biases of its members to obtain an accurate result. When individual and group decisionmaking were compared, it was seen that groups performed better because prejudices of individuals were frequently counterbalanced, and objectivity resulted. Groups also exhibited increased motivation and self-criticism. All of these advantages, except, perhaps, self-motivation, tend to diminish as the size of the group diminishes. Because juries frequently face complex problems laden with value choices, the benefits are important and should be retained. In particular, the counterbalancing of various biases is critical to the accurate application of the common sense of the community to the facts of any given case.”

The Court also noted that research data had cast doubt on the accuracy of results achieved by progressively smaller panels, and that the risk of convicting the innocent rose as the size of the jury diminished. It said that it adhered to its decision in Williams, while “readily admit[ting] that we do not pretend to discern a clear line between six members and five.” But it was satisfied that anything less than six would be constitutionally unacceptable. There had been some change in the Court’s membership between Williams and Ballew (Black, Harlan and Douglas JJ had been replaced by Powell, Rehnquist and Stevens JJ). One cannot help getting the impression that the Ballew Court might, with the benefit of hindsight, have decided Williams differently.

There may be nothing magic about the number twelve, or ten for that matter. But, for all the reasons identified by the US Supreme Court in Ballew, the lower the number falls the less reliable and legitimate a verdict may become. If a lower number, say seven, were to be adopted, even temporarily, several further questions would arise. Would this be acceptable in all cases, however serious the charge and however severe the likely punishment in the event of conviction? Would majority verdicts be allowed and, if so, by what margin? Would it be permissible for a trial to proceed if one or more members of the jury had to be discharged before the trial concluded?

Perhaps the most important question, to return to where we began, is whether a reduced number of jurors would satisfy the requirement that a jury should represent a fair cross-section of the community. In O Maicin v Eire [2014] 4 I.R. 477 and 583 (see Addendum below), MacMenamin J stated at 668-689:

“A jury then must be selected from a pool genuinely representative in character and reflecting the diffuse nature of the community. Among the necessary characteristics are that it be, and perceived to be, constituted with the necessary assurance of diffused impartiality; randomly and indiscriminately drawn from a pool broadly representative of the community, inclusive of large distinctive groups within that community. What is necessary is the “stamp of responsibility” or involvement as part of the community as a whole. The pool must be representative of every category of the public in the area in which the trial is to be conducted; were it not so, it would not be constitutionally compliant. There is, too, a natural justice dimension to the requirements.”

This is an authoritative a statement of the necessity for a broadly representative jury pool. However, it must surely follow that the jury actually sworn in to serve in a given case should also be broadly representative or, at least, stand a good chance of being so. Allowing the initial number of jurors to fall below 12 might risk diminishing the representative quality of the jury, and this in turn would raise a serious question about the constitutionality of the arrangement. Add to this what we may describe as the Ballew factors and the case against reducing the number becomes even stronger.

Addendum to earlier post “Can jury trial be waived?”

In an earlier post on the question of whether jury trial under Art. 38.5 of the Constitution should be interpreted as a right or an imperative, I overlooked one recent, and very relevant, authority. Ironically, it was a case in which I appeared myself, O Maicin v Eire [2014] 4 I.R. 477 and 583 where it was held that there was no constitutional right to an Irish speaking jury. Clarke J (as he then was), a member of the Supreme Court majority, having referred to Art. 38.5 and the circumstances in which a person may be tried without a jury for a minor offence, said (p. 646):

These measures are in recognition of the fact that trial by jury goes beyond the rights of the parties and involves a constitutional imperative. Given that imperative and the absence of any definition of what a trial with a jury might mean, there has, understandably, been debate over the extent to which it is open to the Oireachtas to regulate both the composition of a jury and the manner in which a jury trial can properly be conducted in accordance with the constitutional imperative.”

In the same spirit, MacMenamin J said (p. 669) that “the proper administration of criminal justice is also a duty owed to the public at large. It is not just the prerogative of the accused person.”

Virtual Jury Trials: An Interesting English Experiment

The resumption of jury trials for serious criminal offences is probably the biggest difficulty facing the courts system as we make our way out of lockdown. Social distancing will remain critical for controlling the spread of the Corona virus, but if there is one place on earth where social distancing is not possible in normal times, it is the jury box. Many other participants and observers also find themselves seated in close proximity to one another throughout a trial.

Several competing interests must be balanced when deciding if jury trials should be resumed and, if so, the form they should take. The health of individual jurors must not be placed at risk. Courts must therefore be sympathetic to persons called for jury service who are apprehensive about serving. Social distancing must be maintained, and this probably means seating jurors in the body of the court rather than in the jury box. At the same time, jurors must have ready access to any documentation or exhibits that would normally be given to them and, crucially, they must be able to deliberate in private without feeling any pressure to reach a verdict more quickly than in normal times. Obviously, persons other than jurors must also be present at trial ( leaving aside for the moment members of the public) and this, in turn, may necessitate allocating the various participants and observers to different courtrooms.

One possible solution would be to adjourn all jury trials until they can safely be resumed in the normal way. But this would scarcely be viable – certainly not the optimal solution – when other factors are considered. Accused persons on remand in custody remain entitled to have their trials held with reasonable expedition and in accordance with their constitutional rights to due process. Victims too have a right to expect that trials will proceed without undue delay, especially where a long interval has already elapsed between the date of the alleged offence and the projected trial date, or where there have already been one or more adjournments. Further, a wholesale adjournment of jury trials would lead to a “pile up” in the future, and possibly a deluge of applications to have trials prohibited because of prejudice or hardship resulting from delay.

JUSTICE, a British organisation that does tremendously valuable work in many areas of law, recently conducted four experiments involving mock jury trials where all participants joined the court by video link. The fourth is the most interesting and, I gather, in their estimation the most successful. That involved bringing a jury together in a physical hub where all necessary safety measures were in place. The judge (a retired judge in this instance), registrar, counsel, defendant and witnesses were in other (different) locations. All were visible and audible to each other. There was also a virtual public gallery. Apart from a few very minor hitches, the trial went remarkably well from a technical point of view. It was a fairly short trial involving one defendant charged with an offence of wounding contrary to s. 20 of the Offences Against the Person Act 1861 (still in force in England and Wales). Experts who evaluated the experiment and those who served on the jury were very positive about the entire exercise. Some remarked that they could hear and see the various participants even better than in an ordinary trial. I can vouch for that, having watched it online. Everyone could be very clearly heard and, as someone who can be hard of hearing at times, I can assure you that this is not something that can always be said about Irish courts where microphones are usually available but, from what I can see (or not hear), they are seldom used. Supreme Court, in particular, please take note! In any event, a full account of the experiment can be found on the JUSTICE website ( which has a link to the trial. The experiment was developed in collaboration with a firm of solicitors and an audio visual solutions company (AVMI) to whose website you will be directed to view the trial which lasts almost four hours.

The trial in the JUSTICE experiment was by any standards a straightforward one. The charge was equivalent to assault causing harm under our law, the allegation being that the accused had committed an unprovoked attack on the complainant outside a pub. There were four witnesses (including the defendant) and there was nothing complicated about either the facts or the law. There was no CCTV footage to be shown, but photographs and documents could be displayed on the screen, alongside the trial proceedings, for the benefit of those in the public gallery. There was even a very short trial within a trial in the absence of the jury for a ruling on the admissibility of a particular piece of evidence. The trial could therefore be completed well within a day (though because of time constraints, it ended without the jury being able to agree on a verdict). A virtual trial might not, as the researchers concede, be suitable where, for example, one or more witnesses qualified for special measures as would sometimes be true of sexual offence trials. Still, there may be some trials that could be run in this way.

Secondly, this was a mock jury trial, though very carefully organised so that it resembled a real trial in every essential respect, apart from the virtual element. This can be a very valuable research strategy for investigating and evaluating a wide range of jury-related matters. Such experiments have recently been described by Professor Fiona Leverick as follows:

“Mock juror studies simulate the experience of sitting on a jury by asking participants to read, listen to or watch trial materials. The trial materials used are generally fictional and significantly abbreviated in comparison with a real criminal trial. Studies vary greatly in terms of the extent of their realism and this in turn affects generalisability – how far their findings are likely to apply to real juries, deliberating in actual criminal trials.”

(Leverick, “What do we know about rape myths and juror decision making?” (2020) International Journal of Evidence & Proof (forthcoming)).

In those jurisdictions, including Ireland, where the questioning of jurors about their deliberations is strictly prohibited, simulated jury research is often the only realistic means of investigating the impact that certain factors may have on jury decision-making. One example that comes to mind is the study conducted by Professor Sally Lloyd-Bostock, a leading scholar in this area, on the effect of informing jurors about a defendant’s previous convictions. See Lloyd-Bostock, “The effect on juries of hearing about the defendant’s previous criminal record: A simulation study” [2000] Crim. L.R. 734. Simulated jury trials can be expensive to run and require meticulous planning and implementation, but in the right hands they can yield valuable results.

A third general comment is that we must be careful to place the present pandemic and the measures it has necessitated in proper historical perspective. Radical permanent changes to well-established legal structures and practices would scarcely be wise in order to deal with what will hopefully be a fairly short-term problem. There is, to be sure, a school of thought among historians of infectious diseases and pandemics to the effect that, far from politics and law merely responding to such emergencies, many political policies have been shaped in far-reaching ways by the imperative of preventing contagion. See, for example, Peter Baldwin, Contagion and the State in Europe, 1830-1930 (Cambridge University Press, 1999). Be that as it may, it is fair to predict that at least some of the measures we have had to adopt in response to the Covid-19 pandemic may endure, in one form or another, long after we have defeated the virus. Remote working is one such possibility. We must think carefully about which aspects, if any, of the legal system we would be willing to see permanently changed as a result of our experiences during the pandemic. Jury trial is scarcely one them, but that does not preclude the possibility that this form of trial might be conducted in a somewhat different way while the pandemic lasts.

Finally, there is one further measure that might help to reduce the backlog of criminal trials, though it should be adopted only with great caution and circumspection, and that is to offer an enhanced sentence discount for a guilty plea with the assurance that this will be additional to any credit due for other mitigating factors. Plea incentives must always be treated with caution. After all, a guilty plea effectively amounts to an acceptance that the prosecution has proved the charge beyond a reasonable doubt and it is also, of course, a waiver of the defendant’s constitutional right to trial in due course of law. Incentives to plead must always be surrounded by safeguards designed to ensure that defendants, especially vulnerable defendants, do not feel any pressure to plead to an offence of which they may not, in fact or in law, be guilty. A person convicted following a trial that was delayed for some appreciable length of time because of the Covid-19 restrictions may also be entitled to some reduction in sentence on that account, but that is a different matter.

Policing, procedural justice and remedial justice

The death of George Floyd in such horrific circumstances at the hands of the police in Minneapolis on 25 May sparked off civil unrest throughout the United States and public protests in many other parts of the world. Sadly, this was not an isolated event. Writing in The Observer (31 May 2020), Professor Paul Butler of Georgetown Law School, a leading scholar of race and criminal justice, noted that since 2005 about 15,000 people have been killed by US law enforcement officers, while less than 150 officers have been charged with homicide. Of those charged, a majority were found not guilty or had the charges dropped. As it happens, an academic article published a few weeks before the tragic events in Minneapolis gives some reason to hope that standards of policing as well as relations between police and community can be improved. The article in question, “Procedural justice training reduces police use of force and complaints against officers” by George Wood, Tom R Tyler and Andrew V. Papachristos, was published in the Proceedings of the National Academy of Sciences ( It reports on the outcome of a programme in Chicago where 8,480 police officers were encouraged to adopt procedural justice policing strategies which, in the words of the report, “emphasise respect, neutrality, and transparency in the exercise of authority, while providing opportunities for civilians to explain their side of events.” It was found that over a two-year period, the adoption of such strategies resulted in a 10% reduction in complaints against the police and a reduction of 6.4% in the use of force against civilians. These results may seem modest but, when the raw numbers are considered, the programme appears to have effected a significant improvement in relations between the police and the community.

Procedural justice in this context is always associated with the pioneering work of one co-author of that report, Tom Tyler. He is a social psychologist by background and now a Professor of Law and Psychology at Yale Law School. He is best known for his book, Why People Obey the Law (Yale University Press, 1990, with a further edition in 2006). For this and other leading works, he was awarded the Harry Kalven Prize in 2000 for “paradigm shifting scholarship in the study of law and society.” He drew a distinction between instrumental and normative reasons why people obey the law. The instrumental explanation reflects a utilitarian theory of human action which assumes that deterrence, especially in the form of criminal punishment, is necessary to secure compliance with the law. The normative perspective, on the other hand, focuses on other reasons why people obey the law – because of their own moral principles, for example, or because of a belief in the legitimacy of the law enforcement authorities. Another distinction, and an important one in the present context, is that the instrumental perspective assumes people assess procedural fairness according to the favourability of the outcome (e.g. I got acquitted, therefore the system was fair). The normative perspective is not just concerned with outcome but also with “neutrality, lack of bias, honesty, efforts to be fair, politeness, and respect for citizens’ rights” (Why People Obey the Law, p. 7). Consistent adherence by law enforcement personnel to these values should, according to Tyler’s argument, lead to a reduction in criminal behaviour.

Central to the normative perspective is the concept of legitimacy. People comply with the law, Tyler claimed, principally because they feel that law enforcement authorities are legitimate and that their actions are generally fair. Largely as a result of Tyler’s work, legitimacy has become an important theme in criminology and criminal justice scholarship during the past 20 years or so. It has featured in some important empirical and analytical work on prisons as well as policing (see, for example, Ben Crewe, The Prisoner Society: Power, Adaptation and Social Life in an English Prison (Oxford, 2009) and Alison Liebling, Prisons and their Moral Performance (Oxford, 2004)). During the past decade Anthony Bottoms and Justice Tankebe, both of Cambridge University, have developed a theory of “dialogic legitimacy.” See Bottoms and Tankebe, “Beyond Procedural Justice: A Dialogic Approach to Legitimacy in Criminal Justice” (2012) 102 Journal of Criminal Law and Criminology 119. The essence of this theory is that various power-holders, such as the police, make claims to exercise authority and those claims are made to different audiences (e.g. different racial or socio-economic groups). The responses of these various audiences may cause the power-holders to adjust their claims. This, if done effectively, should presumably increase the power-holders’ legitimacy. For a critique of the Bottoms-Tankebe theory, see Richard Martin and Ben Bradford “The Anatomy of Police Legitimacy: Dialogue, Power and Procedural Justice” (forthcoming, 2020) Theoretical Criminology, and available on SSRN. Incidentally, Richard Martin has another superb article “Ethno-National Narratives of Human Rights: The Northern Ireland Policing Board” (2020) 83(1) Modern Law Review 91. This includes an in-depth analysis, based on an empirical investigation, of the role of the Northern Ireland Policing Board in monitoring the performance of the police in complying with the Human Rights Act 1998. (This is one of the Board’s functions under the Police (Northern Ireland) Act 2000, s. 3). After a great introductory section on the “social life” of human rights, the article reveals, for example, marked attitudinal differences between Nationalist and Unionist members of the Board in terms of how they value and interpret human rights in a policing context.

It so happens that, back in 2015, on the day the Supreme Court delivered judgment in People (DPP) v J.C. [2017] 1 I.R. 417, mitigating the rigours of the exclusionary rule (assuming that’s a reasonable description of what the Court did), I was making a presentation to a conference of senior Garda officers. As the news about J.C. broke, I told them that they should find the decision regrettable and then, pausing momentarily to give them an opportunity to eject me from the conference hall (which they didn’t), I explained that I was proposing in any event to talk about procedural justice and Tyler’s work in particular. The essential message I had for them, and which I still believe to be valid, is that a steadfast adherence to procedural justice, even to what some might see as the point of absurdity, can pay dividends for the police themselves. I would not necessarily share Tyler’s optimism about the capacity of procedural justice to reduce crime, though I would not discount that possibility. Procedural justice, I suggest, is more likely to influence the behaviour of the law-abiding rather than the law breakers. But that of itself can bring great practical benefits. The ability of the police to detect certain kinds of crime (gangland crime being a prime example) and to amass sufficient evidence to justify a prosecution depends crucially on the willingness of those who have relevant information to share it with the police. Their willingness to do so will often be determined by their perception of the legitimacy of the police and their actions. They are more likely to accept that legitimacy if they know that the police response will be marked by the qualities identified by Tyler – objectively, lack of bias, strict adherence to the law, politeness, fairness and so forth.

The issue of remedial justice, mentioned in the title of this post, relates to a legal controversy in the United States which long predated the death of George Floyd but which, as a result of that death, is now very much to the fore again. This is the issue of “qualified immunity” and the story is very briefly as follows. United States law provides (42 USC s. 1983) that a person whose constitutional or other federal rights have been violated by state officials can sue for damages:

“Every person who, under colour of any statute, ordinance, regulation, custom or usage of any State or Territory or the District of Columbia, subjects, or causes to be subjected any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges or immunities secured by the Constitution and laws, shall be liable to the injured party….”

That seems very clear indeed. However, the Supreme Court has grafted on to it a defence known as qualified immunity which was expressed as follows in Harlow v Fitzgerald 457 U.S. 800 at 818 (1982):

“We therefore hold that government officials performing discretionary functions, generally are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.”

The Court explained that the introduction of the element of objective reasonableness “should avoid excessive disruption of government and permit the resolution of many insubstantial claims on summary judgment.” The requirement that an action be a “clearly established” rights violation had robbed the section of much of its effectiveness. However, the qualified immunity doctrine has come in for intense criticism. In fact, it was criticised by Justice Thomas in Ziglar v Abbasi 582 U.S. – (2017). Its impact can be seen in a case such as Surratt v McClaran 851 F.3d 389, decided by the Court of Appeals for the Fifth Circuit in 2017. Police officers had effectively choked a woman in the back of a patrol car while trying to dislodge a packet of drugs which she was hiding in her mouth. She became unresponsive and was brought to hospital where she died about two weeks later. A section 1983 claim by her sister failed in both the District Court and the Court of Appeals. The latter court approved a statement of the former: “previous case law has provided no guidance regarding what is precisely reasonable and what is unreasonable regarding the use of force to an individual’s throat where the individual appears to be concealing something in their mouth.” The Supreme Court later declined to hear a further appeal.

Last week, Representative Justin Amash (the sole Libertarian Party member serving in Congress) introduced legislation proposing to end qualified immunity. It has already received some cross-party support.

European Court of Human Rights upholds expulsion of “foreign” offender

The punishment endured by a person convicted of a criminal offence may extend well beyond the primary, judicially imposed penalty which will typically be a fine, a term of imprisonment or a community order. Ancillary penalties in the form of disqualification, notification requirements and the like have proliferated during the past few decades. Not all of these are formally recognised as “penalties” (for the purpose of Art. 7 of the ECHR, for example), but they can still have a severe punitive impact. A criminal conviction may have further collateral consequences, whether arising by operation of law (such as long-term disenfranchisement which is still common in parts of the United States) or from executive action. Deportation is one such consequence and, for some offenders, it is the most severe, disruptive and enduring punishment of all.

The severity and potential arbitrariness of deportation or expulsion in response to a criminal conviction is illustrated by Azerkane v The Netherlands, a judgment of the European Court of Human Rights handed down on 2 June 2020. Azerkane (“A”) was a second generation immigrant born in The Netherlands in 1993 to Moroccan parents with whom he lived all his life. His parents and five of his seven siblings have Dutch nationality. However, he remains a Moroccan national who held temporary renewable residence permits in The Netherlands. When one such permit expired on 17 January 2005, it was not renewed until 15 February 2005, and this short gap, which was apparently due to an oversight on the part of A’s father, proved very significant for the purpose of the present case. Between 2006 (when he was still a juvenile) and 2011, he committed several offences, the most serious being armed robbery of which he was convicted in 2012 and for which he was sentenced to three years’ imprisonment. In October 2013, his residence permit was revoked and he was made subject to a 10-year re-entry ban. This decision was made primarily by setting the total of 41 months custody which he had served for his various offences against the uninterrupted period of six to seven years during which he had been lawfully resident in The Netherlands before committing the robbery. (In reality, of course, he had at that point been a resident for 20 years). All his appeals against this decision were dismissed. In January 2016, he lodged an application with the European Court of Human Rights claiming that the revocation of his residence permit and the ten-year re-entry ban amounted to a violation of his rights under Article 8 of the ECHR. This Article guarantees the right to respect for private and family life, though it may be restricted “in accordance with law and [as] is necessary in a democratic society” in the interests of promoting certain social goals including national security, public safety, and the prevention of crime or disorder. The European Court (Fourth Section) accepted that the decision of the Dutch authorities interfered with A’s right to respect for both his family and his private life, but ultimately found no violation of Article 8 because the expulsion “was proportionate to the aims pursued and therefore necessary in a democratic society” ([84]).

This is the latest in a long line of cases in which the Strasbourg court has had to determine if the decision of a State Party to deport or expel a foreign offender is compatible with Article 8. The applicants are generally non-EU nationals who cannot rely on the freedom of movement provisions of the TFEU or on national regulations giving effect to the relevant Directive. (On the application of this aspect of EU law Ireland see, for example, P.R. v Minister for Justice [2015] IEHC 201). The Strasbourg court has developed a set of criteria for evaluating the compatibility of expulsion decisions with Article 8 of the ECHR. These are often known as the Boultif factors, after Boultif v Switzerland (2001) 33 EHRR 50 where they were first articulated though they were later set out in more tabular form in Uner v The Netherlands (2007) 45 EHRR 14 at [54], a judgment of the Grand Chamber. They include the nature and seriousness of the offence, the length of the applicant’s stay in the country, the time that has elapsed since the commission of the offence and the applicant’s conduct during that period, the nationalities of the persons concerned, and the applicant’s family circumstances. In Uner, the Court added two further factors, while accepting that they may already have been implicit, namely, the best interests and well-being of children and “the solidity of social, cultural and family ties with the host country and the country of destination.”

The ECHR does not prohibit the deportation of foreign nationals. In one of its many tropes or set pieces, the Strasbourg court says:

The Court reaffirms at the outset that a state is entitled, as a matter of international law and subject to its treaty obligations, to control the entry of aliens into its territory and their residence there…. The Convention does not guarantee the right of an alien to enter or reside in a particular country and, in pursuance of their task of maintaining public order, contracting states have the power to expel an alien convicted of criminal offences. However, their decisions in this field must, in so far as they may interfere with a right protected under para. 1 of Art. 8, be in accordance with the law and necessary in a democratic society, that is to say justified by a pressing social need and, in particular, proportionate to the legitimate aim pursued….”

(See, for example, Uner at [54]). Expulsion of offenders has occasionally been found to violate Article 8, as in A.A. v United Kingdom (20 September 2011) and Maslov v Austria (23 June 2008). In both cases, the offences were committed while the applicants were minors, although in A.A. the offence consisted of the gang rape of 13-year-old girl when the applicant was about 15 years.

The applicant in Azerkane did, to be sure, have a bad criminal record though much of it consisted of offences committed while he was a juvenile. It was probably no worse than that of many Dutch nationals or, for that matter, nationals of any other member state. He was a Dutch national in everything but name, having been born and lived all his life in The Netherlands. He had very slender ties (though he had some) to Morocco to which he was, in effect, being deported, though it seems that he spoke at least some Berber. The court frankly acknowledged (at [84]) that the expulsion would have a severe impact on the applicant’s family and private life and it acknowledged his limited ties to what it called “his country of origin”. This presumably referred to Morocco although his true country of origin was The Netherlands. There was some evidence that he had a mild mental disability. However, one point that certainly counted against him is that he had continued to reoffend after the exclusion order was made and even after lodging his application with the Strasbourg court. In this respect the case differed from A.A. v United Kingdom (20 September 2011) where the offence was very serious but the applicant’s conduct afterwards had been exemplary.

One problem with cases such as Azerkane is that the Strasbourg court views them solely through the lens of Article 8 of the ECHR, rather than focusing more on the punitive dimension of deportation, or indeed on the consequent discriminatory impact. The happenstance of being an immigrant, even a second-generation immigrant, can greatly increase the overall punishment endured for an offence. Granted, the applications are usually based on Article 8 alone and this is presumably because the Court lacks a well developed proportionality jurisprudence, at least in relation to sentencing. (It has an extensive jurisprudence on what we would call Heaney proportionality for the purpose of assessing the permissibility of State-imposed limitations on the rights guaranteed by Articles 8 to 11). The more fundamental problem is that the European Convention itself has no express requirement that punishment be proportionate to the crime. It is to be contrasted in this respect with the EU Charter of Fundamental Rights (Art. 49(3)). However, there are indications from cases such as Weeks v United Kingdom (1988) 10 EHRR 293 that a grossly disproportionate penalty may fall foul of Article 3 of the ECHR which outlaws torture and inhuman or degrading treatment or punishment. If the ECHR were more dynamically interpreted by the Strasbourg court to require proportionality of punishment, the court might be more inclined to consider the cumulative impact of all the consequences, direct and collateral, of a criminal conviction. There is, of course, a general reluctance to read into treaties such as the ECHR additional provisions, unless they can readily be inferred from the express terms of the text. Yet, the Court had no difficulty in doing just that in Scoppola v Italy (No.2) (2010) 51 EHRR 12 where it interpreted Article 7(1) as providing for lex mitior, even though the text clearly did not, and may not have been intended to.

The Court cannot be unaware of punitive impact of deportation following conviction. After all, it has been pointedly reminded of it in certain dissenting opinions, notably that of Judges Costa, Zupancic and Turmen in Uner v The Netherlands (2007) 45 EHRR 14. In that case, a 14-member majority of the Grand Chamber found no violation of Article 8 in circumstances where the applicant’s residence permit was withdrawn following his conviction for manslaughter. He was a Turkish national who had lived in the Netherlands since he was 12 years of age. By the time the permit was withdrawn he was in his late twenties and the father of two children born and living in The Netherlands. The joint dissenting opinion stated in part (at O-II17):

“The principle is that of “double punishment”, or rather the discriminatory punishment imposed on a foreign national in addition to what would have been imposed on a national for the same offence. We do not agree with the assertion in [56] that the applicant’s exclusion was to be seen as preventive rather than punitive in nature. Whether the decision is taken by means of an administrative measure, as in this case, or by a criminal court, it is our view that a measure of this kind, which can shatter a life or lives – even where, as in this case, it is valid, at least in theory, for only ten years (quite a long time, incidentally) – constitutes as severe a penalty as a term of imprisonment, if not more severe. This is true even where the prison sentence is longer but is not accompanied by an exclusion order or expulsion. That is why some states do not have penalties of this kind specific to foreign nationals, while others have formally abolished them in recent times….”

A similar point was made by Judge Morenilla in his dissenting opinion in Boujlifa v France (21 October 1997) where he described the deportation as “an aggravation of the criminal penalty imposed on Mr Boujlifa in relation to those imposed on French nationals, so that it is discriminatory.” A joint dissenting opinion by Judges Baka and Van Dijk in Boujlifa stressed that the applicant had already served the term of imprisonment imposed upon him and that the sentence in question may be assumed to have been proportionate to the offence committed. Why, then, they asked should the applicant be treated differently from a French national, especially when there was no serious reason to believe that he would commit further offences.

In this respect, it is interesting to compare Azerkane and similar Strasbourg decisions with the United States Supreme Court judgment in Padilla v Kentucky 559 U.S. 356 (2010). Padilla, a native of Honduras, had been lawfully resident in the United States for 40 years during which time he served in Vietnam, when he pleaded guilty to a serious drugs distribution charge. As a result of this he was made subject to deportation proceedings. His lawyers had failed to advise him of this likely consequence before he decided on a plea, and his argument to the Supreme Court was that, as a result of this failure, he was denied effective assistance of counsel under the Sixth Amendment, a right recognised by the Court in Strickland v Washington 466 U.S. 668 (1984). A majority of the Supreme Court decided the case in his favour. The Supreme Court of Kentucky had held that the collateral consequences of a conviction were “outside the scope of representation required by the Sixth Amendment.” Stevens J., delivering the opinion of the U.S. Supreme Court, disagreed:

“Although removal proceedings are civil in nature,… deportation is nevertheless intimately related to the criminal process. Deportation as a consequence of a criminal conviction is, because of its close connection to the criminal process, uniquely difficult to classify as either a direct or collateral consequence. The collateral versus direct distinction is thus ill-suited to evaluating a Strickland claim concerning the specific risk of deportation. We conclude that advice regarding deportation is not categorically removed from the ambit of the Sixth Amendment right to counsel. Strickland applies to Padilla’s claim…”

Padilla was not concerned with the constitutionality of the deportation provisions themselves, but solely with the obligation of defence counsel to advise their clients properly as to the entire consequences (collateral and otherwise) of a conviction. Nonetheless, the Court majority was clearly sensitive to the punitive impact of deportation and its close relationship with any other penalty imposed. To be fair, the Strasbourg Court in Azderkane did accept that the applicant would suffer as a result of deportation, but it was still prepared to uphold the deportation decision.

Another disappointing aspect of Azerkane is that the Court did not seem particularly troubled that the applicant was a second-generation immigrant who was born and had lived all his life in the country from which he was being expelled. Dutch citizenship law is based on the concept of ius sanguinis which means that citizenship is not automatically acquired by being born in the country. The same holds true of other European countries, including nowadays Ireland. However, in the case of a person such as A who is a citizen in everything but name, who could probably have become naturalised without too much difficulty and who is being sent to a country with which he or she has little or no familiarity and few connections, expulsion should scarcely be countenanced. It is obviously different where the offender was born and raised elsewhere and had spent only a limited period of time in the country where the offences were committed. In Uner (above), decided in 2006, the Court noted that a majority of member states of the Council of Europe permitted the deportation of second-generation immigrants, whereas eight member states did not. The court in Azerkane acknowledged in passing (at [71]) that “very serious reasons” were required to justify the expulsion a settled migrant. However, it seems from the joint dissenting opinion of Judges Baka and Van Dijk in Boujlifa v France (21 October 1997) that the court has not always been united on the deportation of second generation immigrants. Those dissenting judges urged the court to “abandon its casuistic approach to the matter and take a clear position on the question of whether and to what extent so-called “second generation” immigrants constitute a special category for whose deportation very serious reasons have to be advanced to make it justifiable under the second paragraph of Article 8.”

It is time for the Grand Chamber of the Court to undertake a radical reassessment of the compatibility of this practice with the Convention.

Canadian Supreme Court on Sentencing Child Sexual Abuse

Ever so often, the Canadian Supreme Court delivers a major sentencing judgment addressing policies and principles that are also recognised and applied in other jurisdictions. As such they are of great comparative interest, and especially perhaps in Ireland because of certain constitutional and systemic features the two countries have in common. Four of those features are worthy of note. First, Canada, like Ireland (for now at least), is a non-guideline jurisdiction. Most Canadian provinces have judicially-developed starting points or sentence ranges for certain offences, as do we, but there are no formal guidelines like those in England and Wales or in parts of the United States. Secondly, proportionality is the overarching distributive sentencing principle in both jurisdictions. In Canada, that principle is now enshrined in the Criminal Code (s. 718.1) whereas here it has been judicially developed, as it originally was in Canada. Thirdly, both countries have written constitutions, providing for judicial review of legislation and affording strong protection to personal rights and procedural fairness. In both jurisdictions, legislation may be struck down if found to be in conflict with the Constitution (which, in Canada, for this purpose is the Charter of Rights and Freedoms adopted in 1982). See R v Nur [2015] 1 S.C.R. 773 (Canada) and Ellis v Minister for Justice [2019] IESC 30 (Ireland). The legislatures in both countries are therefore subject to certain constraints in terms of the punishments they can specify for offences and in enacting more general sentencing rules. Fourthly, both countries share a broadly similar sentencing culture, accepting that “culture”, when used in this sense, can be something of a contested concept. Canada, like Ireland, remains committed to an essentially discretionary sentencing system or, more precisely, a system of principled discretion. Further, both countries, by international standards and certainly when compared with the United States, can fairly be said to adhere to a moderate penalty system. One point of difference is that Canada has a much higher number of minimum sentences than we do, though many of them are quite short and the Canadian Supreme Court has been willing to strike down some of them on constitutional grounds.

Major Canadian Supreme Court sentencing judgments from the fairly recent past include R v Nasogaluak [2010] 1 S.C.R. 206 (serious police misconduct at the time of arrest a mitigating factor at sentencing); R v Lacasse [2015] 3 S.C.R. 1089 (circumstances in which an appeal court should interfere with sentence, and consideration of local prevalence of offence); R v Nur [2015] 1 S.C.R. 773 (striking down a mandatory minimum sentence for firearms offences); R v Suter [2018] 2 S.C.R. 496 (punishment beating of defendant and mistake of law as mitigating factors). Now, in R v Friesen 2020 SCC 9 (April 2, 2020), the Court has dealt at great length with the sentencing of child sexual abuse offences.


Friesen (F.) met a woman (we shall call her Ms A) through a dating website. One evening in 2016, they went back to her home which she shared with her four-year-old daughter (B) and her one-year-old son. There they had consensual sexual intercourse after which F told Ms A to bring B into the bedroom so that they could engage in sexual acts with her. This was done although Ms A audio recorded the conversation on her mobile phone. B was understandably terrified at the sexual acts to which she subjected by F and started to scream. Fortunately, a friend of Ms A’s who was babysitting was also staying in the house and, on hearing B’s screams, she came and rescued B from the room. F wanted Ms A to bring B back to the room and threatened her that, unless she did so, he would tell her friend, the babysitter, that Ms A had sexually abused her one-year-old son. This formed the basis of a separate extortion charge to which F pleaded guilty. He also pleaded guilty to sexual interference with a child and it is the sentence for the latter offence that is the subject of the Supreme Court judgment. In the Provincial Court of Manitoba the trial judge imposed a sentence of six years’ imprisonment for the sexual interference offence and a concurrent six-year sentence for the attempted extortion. The Manitoba Court of Appeal reduced these sentences to four-and-a-half years and 18 months respectively, to run concurrently. Following a prosecution appeal, the Canadian Supreme Court restored the six-year sentence imposed by the trial judge for the sex offence.

The judgment

The joint judgment was delivered by Wagner C.J. and Rowe J. with the remaining seven judges concurring. Effectively, therefore, it was the unanimous judgment of a nine-judge court and, as such, holds considerable authority. Moreover, it is a lengthy and detailed judgment running to 183 paragraphs and over 100 pages of typescript. Most of it is devoted to a close analysis of the nature of child sexual abuse, the harm it causes and the principles according to which it should be sentenced. The Court did not establish any general sentencing guidelines in the form of starting points or ranges. It usually prefers to leave that to the provincial appeal courts, rather than lay down guidelines itself. As might be expected from the length of the judgment, there is a great deal in it and what follows provides nothing more than a brief and incomplete summary.

Starting points

In Canada, criminal law may be created only by the federal parliament, and most criminal offences are contained in the Criminal Code. However, the criminal law is administered in the individual provinces each of which has its own court system. Over the years, several provincial appeal courts have delivered judgments indicating appropriate starting points for the sentencing of particular offences. Under this system, an appeal court identifies the appropriate sentence for a typical case; this is a starting point to which adjustments may be made in light of the particular circumstances of actual cases. The practice may be traced back to R v Sandercock 1985 ABCA 218 (CanLII), a judgment of the Alberta Court of Appeal dealing with sexual assault. There the Court said:

“The sentencing process now adopted by this Court is to state typical categories with precision, and to acknowledge at the same time that each actual case presents differences from the archetypical case. These differences might mitigate or aggravate. Nevertheless, the idea of a typical case affords a starting point for sentencing because one can state a precise sentence for that precise category. An actual sentence in a real case will vary upwards or downwards from that depending upon the balance of the factors presented in the actual case.”

In Sandercock itself which, it must be recalled, was decided 35 years ago, the Court identified three years’ imprisonment as the starting point for major sexual assault. The staring point approach has strong supporters as well as fierce critics, the latter variously claiming that it interferes with judicial discretion and that it creates artificial categories of offences in a way that Parliament did not intend. However, the approach was strongly reaffirmed by the Alberta Court of Appeal in R v Arcand 2010 ABCA 363 (CanLII). Some provincial appeal courts have identified sentence ranges rather than starting points. That, of course, is the approach favoured here in Ireland in the line of appeal court judgments from People (DPP) v Ryan [2014] 2 ILRM 98 onwards. The Canadian Supreme Court has so far tolerated starting point approaches except that it has firmly held, in a series of cases starting with R v McDonnell [1997] 1 S.C.R. 948, that failure on the part of a trial judge to place a particular offence within a judicially created category could never, of itself, amount to an error that would justify an appeal court interfering with the sentence which might still be “fit” in the circumstances.

The reason I mention all of this in the present context is that in Friesen some interveners, notably the Legal Aid Society of Alberta and the Criminal Trial Lawyers Association, seem to have launched an all-out attack on the starting point system. Having noted this, the Supreme Court said ([41]): “While we have determined that this case does not provide an appropriate opportunity to assess the merits of these concerns, they raise an issue of importance that should be resolved in an appropriate case.” This might be seen as a rather ominous statement, especially in light of the policy of deference on which the Supreme Court has strongly and repeatedly insisted over many years. A trial judge’s sentencing decision should be respected by an appeal court unless the sentence is demonstrably unfit or is based on an error of principle (which may include an error of law) that has had an impact on the sentence. A “harmless error”, to use an American expression, will not justify appellate intervention. This was stressed in R v Lacasse [2015] 3 S.C.R. 1089 and again, in Friesen. In truth, Irish and Canadian law are largely the same in this respect, save that we don’t use the language of deference quite as much. We must therefore wait and see what the present Canadian Supreme Court has to say about starting point sentencing when the matter eventually comes up for consideration, and that may happen before too long. However, the Court will probably (or should) be alert to the possibility that if it condemns the whole idea of starting point sentencing, Parliament may respond by introducing formal guidelines that would place even more constraints on judicial discretion although, to complicate matters further, those too might be subject to constitutional challenge. Meanwhile, as noted below, the Court in Friesen expressly disapproved of the content of some provincial sentencing guidelines for child sexual abuse.

For anyone wanting to read up on Canadian criminal law and sentencing, Kent Roach’s Criminal Law, 7th ed. (Irwin Law, 2018) is fairly concise (about 600 pages) but tremendously good. On sentencing, there is a great casebook which is quite up to date: Manson, Healy, Trotter, Roberts and Ives, Sentencing and Penal Policy in Canada: Cases, Materials, and Commentary 3rd ed. (Emond Montgomery Publications, Toronto, 2016). Allan Manson’s The Law of Sentencing (Irwin Law, 2001) may be almost 20 years old but it has deservedly become something of a classic for its critical and perceptive discussion of general sentencing principles. It is often cited by the Supreme Court (as indeed it was in Friesen). On starting point sentencing (and for a strong defence thereof), see Paul Moreau, “In defence of starting point sentencing” (2016) 63 Criminal Law Quarterly 345.

Sentencing child sexual abuse

The offence to which F. pleaded guilty was sexual interference with a child which, under s. 151 of the Criminal Code, carries a maximum sentence of 14 years’ imprisonment following conviction on indictment. However, the Supreme Court made clear that its general observations were equally applicable to related offences such as invitation to sexual touching, incest, sexual assault, child luring and sexual exploitation. (Incidentally, our offence of invitation to sexual touching under s. 4 of the Criminal Law (Sexual Offences) Act 2017 seems to be based directly on s. 152 of the Canadian Criminal Code).

In Friesen, the Supreme Court begins with a lengthy account of the harm caused by child sexual abuse, including forms of abuse that are facilitated by new technologies ([46]-[74]). It recognises that females are much more likely to be victims of such abuse but also acknowledges that it may have a particularly detrimental impact on children of indigenous communities, children in care, children with disabilities and LGBT2Q+ youth. Child sexual abuse can have a ripple effect by damaging children in their social relationships and by destroying the trust of parents and caregivers in family, friends and social institutions. The Court also placed some emphasis on the inherent wrongfulness of child sexual abuse as a violation of the child’s right to bodily integrity which, as it had previously said in R v Ewanchuk [1999] 1 S.C.R. 330, “lies at the core of human dignity and autonomy.” When assessing harm in cases where the victims are still children, sentencing courts should consider not only the immediate or observable harm but also the reasonably foreseeable potential harm that victims may experience well into adulthood. In this connection also the Court stressed the importance of taking account of any increases in maximum sentences introduced by legislation.

The Court did, of course, acknowledge that an offender’s culpability must also be considered and it accepted, in particular, that a mental disability imposing serious cognitive constraints is a mitigating factor. At the same time, it expressly approved Parliament’s decision, as expressed in the Criminal Code, to prioritise denunciation and deterrence in sentencing child sexual abuse offences.

The Court expressed serious concern about the content of guidelines, in the form of sentence ranges or starting points, adopted by some provincial appeal courts, which provide that the upper limit may be exceeded only in exceptional circumstances. It said ([112]) that it is “inappropriate to artificially constrain sentencing judges’ ability to impose a proportionate sentence in this manner.” It went even further, at [117], by directing provincial appellate courts to revise and rationalise sentence ranges and starting points where they treat sexual violence against children and sexual violence against adults similarly. Sexual violence against children is generally more serious. This may be a foretaste of what is to come when the Court gets an opportunity to deal more fully and directly with the legitimacy of sentence starting points and ranges.

Having said all of this, the Court then set out the following “significant factors” that should determine a fit sentence for a child sexual abuse offence:

(a) likelihood of reoffending;

(b) abuse of a position of trust or authority;

(c) duration and frequency of offending (“Each further instance of sexual violence traumatizes the child victim anew and increases the likelihood that the risks of long-term harm will materialize” ([133]).

(d) the age of the victim

(e) the degree of physical interference. However, in this regard, the Court cautioned against attributing too much significance to the form which the abuse took, and in particular against assuming that non-penetrative sexual acts merit less punishment. Courts “should not assume that there is any clear correlation between the type of physical act and the harm to the victim” ([142]). All kinds of physical acts can cause emotional and psychological harm to child victims. Still, increases in the degree of physical interference increases the wrongfulness of the sexual violence ([145]).

(f) “victim participation.” The Court went to great pains to stress that it is an error of law to treat “de facto” consent as a mitigating factor. It said: “Adults, not children, are responsible for preventing sexual activity between children and adults.” Granted, the Court was here assuming that the offence involved abuse of a child by an adult, and in Canada the age of consent is 16 years. Although it did not say so, it may have accepted that consent could be a relevant factor when dealing with sexual activity between young persons of the same age, assuming of course, that there was actual consent.

Finally, the Court said that co-ordinated sexual assault, meaning an assault which, as in this case, involved a number of participants, was all the more serious for that reason.


The Supreme Court found that it was the Manitoba Court of Appeal that fell into error and not the trial judge. In fact, it didn’t have a good word to say about the Court of Appeal judgment, but it expressed strong support for the trial judge’s approach to the case. As noted, it restored the sentence he had imposed . In this respect, the judgment is reminiscent of the majority judgment in R v Lacasse [2015] 3 S.C.R. 1089 which was strongly critical of the Quebec Court of Appeal on that occasion. However, it must be said that the dissenting judgment in Lacasse was far more convincing on the circumstances in which and the conditions under which a court should take the local prevalence of an offence into account at sentencing.


Child sexual abuse is, unfortunately, a universal phenomenon. although it is only in fairly recent times that it has begun to be prosecuted with any degree of frequency. All actors within the criminal justice system, including the courts, have had to embark on a fairly steep learning curve in order to appreciate its prevalence, the circumstances in which it occurs and, above all else, the harm (both short-term and long-term) it causes to victims. The judgment in Friesen will doubtless be of great value to courts and sentencing bodies elsewhere in devising principles and guidelines for sentencing child sexual abuse.

Parts of the Friesen judgment might make one reflect on the production of knowledge, if that does not sound a bit too “Foucault.” For example, in setting out the harm caused by child sexual abuse, the Court relied mainly on dicta from earlier judgments of its own or of other courts. For instance at [57], the Court says:

“A number of this Court’s decisions provide insight into these forms of harm. In R v L. (D.O.) [1993] 4 S.C.R. 419, L’Heureux-Dube J. emphasised the emotional trauma that the nine-year old complainant experienced from sexual violence (pp. 439-442). Similarly, in McDonnell, McLachlin J. (as she then was) stressed the emotional harm of “the violence of the child victim’s integrity and sense of self-worth and control over her body” that the child victim experienced as a result of being sexually assaulted while sleeping (para. 111). The likely result of the sexual assault would be “shame, embarrassment, unresolved anger, a reduced ability to trust others and fear that… people could and would abuse her and her body” (para. 113).”

Nobody would question the validity of what is said in that passage. And L’Heureux-Dube J. and McLachlin C.J. (as she later became) are both very fine jurists indeed. But the question is whether purely judicial decisions and dicta are the best source of authority for matters that are essentially empirical in nature. (Admittedly, L’Heureux-Dube J. did refer to some reports in the judgment that is cited in Friesen). One might contrast this approach with that of the majority of the United States Supreme Court in the line of cases beginning with Roper v Simmons 543 U.S. 551 (2005) on the reduced culpability of child offenders. There, extensive use was made of social science (and especially psychological) research. The Court was, to be sure, criticised by some for the particular research on which it drew (see Deborah W. Denno, “The scientific shortcomings of Roper v Simmons (2006) 3 Ohio St. Crim. L.J. 379), but I would still suggest that the American court was right to make use of such research rather than rely solely on judicial precedents.