*/
Larger juries are better than smaller juries. But how many jurors does that entail, exactly? A theoretical discussion of jury decision-making and optimal jury size
By Matthew Butt QC and Daniel Butt
The recent proposal by Labour to introduce ‘wartime juries’ with membership reduced from 12 to seven received a ‘cool reception’ from criminal lawyers. Debate about both the number of jurors and the efficacy of the system in general is of course not new and this is a topic which has been debated in Counsel before in the context of the pandemic. Professor Laura Hoyano argued in May 2020 that the backlog could be addressed by allowing defendants to choose a judge only trial, drawing upon the Canadian model, though in the same edition Sailesh Mehta and Mahesh Karu articulate the point feared by many practitioners that measures introduced in extremis during the pandemic will be retained for reasons of economy in the future. Few thought in May 2020, however, that we would still be addressing these problems in 2021.
Conventional courtroom wisdom is that a full complement of jurors returning a unanimous verdict is the ‘gold standard’ for acquitting the innocent and convicting the guilty, to use the language of the Criminal Procedure Rules overriding objective. Any attempt from judges to ether prematurely discharge jurors or give a majority direction ‘too early’ in proceedings is normally resisted by the defence. The reasons for this vary. There is a perception that 12 jurors will give the broadest range of views thus improving the quality of deliberation and the likelihood of a just verdict (sometimes conflated with an acquittal). There is also inevitably the ‘12 Angry Men’ effect. There is some force in observing that the seminal 1957 film would have reached a rather different ending if Henry Fonda (pictured below) had been discharged before the jury retired, perhaps remembering that he had a dental appointment at 5pm that day. Equally, there would have been something of an anti-climax if the jury was called back in for a majority direction and convicted by a majority of 10-2 before the ‘second knife’ was even produced in the jury room. For those who have not seen the film or read David Langwallner’s excellent analysis in Counsel (December 2020), the knife cast doubt on a major plank of the prosecution case (and suggested the judge might have skipped some important parts of his summing up).
To make decisions about jury numbers and majorities by applying such logic is to leave matters rather to chance. From a defendant’s perspective, what if it was ‘Juror 3’ (the prejudiced prosecution minded ‘holdout’ from the film) who was discharged before the jury retired? What if the plot of 12 Angry Men was being played out in reverse in the retiring room, with a biased foreperson brandishing his copy of the Daily Mail and bullying the other jurors into convicting an innocent defendant?
Smaller juries, of course, currently exist in England and Wales. Juries empanelled in civil trials in the county court have eight members in recognition of the fact that the outcome will not impinge upon the liberty of the subject. In inquests, the Coroners and Justice Act 2009 provides that a jury shall have between 7 and 11 members. While inquests are expressly prohibited from determining criminal (on the part of a named person) or civil liability, the importance of inquests such as that into the death of Mark Duggan should not be understated and such inquests commence with 11 jurors in recognition of this.
As the deliberations in a jury room are by law secret there is limited reliable empirical data we can use from this jurisdiction to work out what the optimum number of jurors is (though see Professor Cheryl Thomas’ research paper Are Juries Fair?, Ministry of Justice Research Series 1/10 (2010) as explored by David Wurtzel in Counsel December 2019). There are, however, a plethora of theoretical studies and international examples which can provide insight and possibly a rational way for lawyers and society to decide the optimal number of jurors (a) on a panel and (b) to agree upon a verdict.
One view sees both the traditional 12 person jury and the commitment to unanimity as reflecting tradition rather than efficiency. The thought that one would be able to do more, more quickly, with smaller juries making decisions by, for example, simple majority has obvious appeal. When one looks at studies on jury decision-making, however, things become rather less clear.
The most famous theoretical discussion of jury decision-making is found in the Jury Theorem propounded by the Marquis de Condorcet (pictured below) in 1785, which, broadly speaking, maintains that if each member of the jury has at least a better than 50% chance of getting the right answer, then, under majority decision-making, the greater the number of jury members, the higher the probability of the overall judgment being correct. On this approach, what is needed is a jury large enough to be likely to come to the ‘right’ decision, but small enough to be practical – and a majority, or at least a super-majority, decision is likely to be sufficient.
But the converse of Condorcet’s theorem is also often noted: if each member of the jury is even slightly more likely to get the answer wrong than right, then increasing the size of the jury means that chances of reaching the wrong decision also rise. The challenge for those who would make practical recommendations on the basis of abstract theorising is to incorporate realistic assumptions as to jury behaviour into their models, and this is no easy task. So, for example, it is true that some experimental studies have claimed that decision-making procedures aimed at unanimity can encourage strategic voting and increase the probability of the jury getting the answer wrong (Feddersen, Timothy, and Wolfgang Pesendorfer, ‘Convicting the innocent: The inferiority of unanimous jury verdicts under strategic voting’, American Political Science Review 92 (1998): 23-35).
However, others have argued that the more realistic one makes the experimental models (by, for example, allowing communication between jurors and allowing those in a minority to hold out for a hung jury), the stronger the case becomes for unanimity as a decision-making rule, as jury members are able to vote non-strategically on the basis of their convictions (Coughlan, Peter J., ‘In defense of unanimous jury verdicts: Mistrials, communication, and strategic voting’, American Political Science Review 94 (2000): 375-393).
Things become, unsurprisingly, more complicated if one assumes bad faith on the part of some jurors, stipulating, for example, that they are biased in favour of convicting particular defendants. Larger juries and unanimous decision rules can have the effect of insulating verdicts from the distorting effects that even small numbers of such jurors can have (Mackenzie, Dana, ‘What’s the best jury size?’, Slate 24/4/2013).
From this perspective, larger juries are better than smaller juries. But how many jurors does that entail, exactly? It is striking that 12 seems to be the preferred number of jury members in a wide range of different jurisdictions, despite the number having its historic origins many centuries back, and notwithstanding the slightly larger number of 15 employed in criminal trials in Scotland. Mimicry? Coincidence? Or is something else going on? There is certainly theoretical support available for the traditional number of 12.
For example, a 2020 study by Takamitsu Watanbe entitled ‘A numerical study on efficient jury size’ studied data on jury deliberations and concluded that the efficient jury size is 11.8 (Humanities and Social Sciences Communications 7 (2020): 1-7). It should be admitted that in practice it is hard to obtain four-fifths of a juror, and acknowledged that this estimate has a margin of error of plus or minus 3. Even the downside of this, however, would not go as low as a 7 member jury. Watanbe suggests that practice may have converged on 12 for good reason: ‘the number of jurors may have been implicitly optimised for efficient unanimous decision-making throughout human history’. We may have reached the right answer without necessarily being able to show our workings.
Moving from the classroom to the courtroom, the experience of judicial systems that have in recent years relied upon relatively small juries gives rise to some real concern. We might look in particular at practice in the United States where the question of jury size has been a live debate for years. Although most American criminal juries have 12 members, the state of Florida reduced the number of jurors in non-capital cases from 12 to 6 in 1967. This practice was held to be compatible with the Sixth Amendment by the US Supreme Court in Williams v Florida (1970). The majority judgement, delivered by Justice White, expressed scepticism at the idea that jury size was of critical importance, holding that the point of having a jury was to prevent government oppression, providing ‘an inestimable safeguard against the corrupt or overzealous prosecutor and against the compliant, biased, or eccentric judge’, and arguing that, ‘[t]he performance of this role is not a function of the particular number of the body that makes up the jury’. ‘Certainly,’ he writes, ‘the reliability of the jury as a factfinder hardly seems likely to be a function of its size.’
Scholars since this judgment have not always shared the court’s confidence. In particular, writers have queried the representative nature of six person juries in racially diverse communities. Marc Pearce and Twila Wingrove write that ‘research has consistently shown that six-member juries are not likely to represent the community as effectively as 12-member juries’ (‘Is a jury of six as good as one of 12?’, Monitor on Psychology 40 (2009): 32).
The issue came to the fore of the political agenda in 2013 following the acquittal on second-degree murder charges of George Zimmerman for the fatal shooting of unarmed 17-year-old African-American high school student Trayvon Martin in Sanford, Florida. The verdict was reached by a jury of six, all of whom were female, and five of whom were white, and led to multiple calls for a revisiting of Williams v Florida and a return to 12 person juries. Such calls have persisted. A 2020 article by Patrick Higginbotham, Lee Rosenthal and Steven Gensler, ‘Bringing Back the Twelve-Person Civil Jury’ calls the 12 person jury the ‘gold standard’, arguing that larger juries are more predictable, make better decisions, and are more inclusive and representative of the community (Judicature 104 (2020): 47-57). History shows that the decision to abandon the gold standard is not one that should be taken lightly.
Concerns about diversity in our legal system should also sound a warning to those attracted to reducing jurors or jury trials. In her research on jury decision-making referenced above, Cheryl Thomas found little evidence that juries are not fair, noting in particular that verdicts showed only small differences based on defendant ethnicity. Ultimately, we do not know what the effects would be of reducing juror numbers. It is a sobering fact that in 2021, not a single High Court, Court of Appeal or Supreme Court judge self identifies as Black. As the legal system seeks to urgently address the systemic problems raised in for instance the Lammy Review or the issues eloquently described by Professor Leslie Thomas QC in Counsel July 2020, restricting trial by jury would strike at one of the few parts of our criminal justice system that is truly reflective of society.
Daniel Butt is Fellow and Tutor in Politics at Balliol College, Oxford. He has published on a range of justice-related subjects, and is the author of Rectifying International Injustice (OUP, 2009).
The recent proposal by Labour to introduce ‘wartime juries’ with membership reduced from 12 to seven received a ‘cool reception’ from criminal lawyers. Debate about both the number of jurors and the efficacy of the system in general is of course not new and this is a topic which has been debated in Counsel before in the context of the pandemic. Professor Laura Hoyano argued in May 2020 that the backlog could be addressed by allowing defendants to choose a judge only trial, drawing upon the Canadian model, though in the same edition Sailesh Mehta and Mahesh Karu articulate the point feared by many practitioners that measures introduced in extremis during the pandemic will be retained for reasons of economy in the future. Few thought in May 2020, however, that we would still be addressing these problems in 2021.
Conventional courtroom wisdom is that a full complement of jurors returning a unanimous verdict is the ‘gold standard’ for acquitting the innocent and convicting the guilty, to use the language of the Criminal Procedure Rules overriding objective. Any attempt from judges to ether prematurely discharge jurors or give a majority direction ‘too early’ in proceedings is normally resisted by the defence. The reasons for this vary. There is a perception that 12 jurors will give the broadest range of views thus improving the quality of deliberation and the likelihood of a just verdict (sometimes conflated with an acquittal). There is also inevitably the ‘12 Angry Men’ effect. There is some force in observing that the seminal 1957 film would have reached a rather different ending if Henry Fonda (pictured below) had been discharged before the jury retired, perhaps remembering that he had a dental appointment at 5pm that day. Equally, there would have been something of an anti-climax if the jury was called back in for a majority direction and convicted by a majority of 10-2 before the ‘second knife’ was even produced in the jury room. For those who have not seen the film or read David Langwallner’s excellent analysis in Counsel (December 2020), the knife cast doubt on a major plank of the prosecution case (and suggested the judge might have skipped some important parts of his summing up).
To make decisions about jury numbers and majorities by applying such logic is to leave matters rather to chance. From a defendant’s perspective, what if it was ‘Juror 3’ (the prejudiced prosecution minded ‘holdout’ from the film) who was discharged before the jury retired? What if the plot of 12 Angry Men was being played out in reverse in the retiring room, with a biased foreperson brandishing his copy of the Daily Mail and bullying the other jurors into convicting an innocent defendant?
Smaller juries, of course, currently exist in England and Wales. Juries empanelled in civil trials in the county court have eight members in recognition of the fact that the outcome will not impinge upon the liberty of the subject. In inquests, the Coroners and Justice Act 2009 provides that a jury shall have between 7 and 11 members. While inquests are expressly prohibited from determining criminal (on the part of a named person) or civil liability, the importance of inquests such as that into the death of Mark Duggan should not be understated and such inquests commence with 11 jurors in recognition of this.
As the deliberations in a jury room are by law secret there is limited reliable empirical data we can use from this jurisdiction to work out what the optimum number of jurors is (though see Professor Cheryl Thomas’ research paper Are Juries Fair?, Ministry of Justice Research Series 1/10 (2010) as explored by David Wurtzel in Counsel December 2019). There are, however, a plethora of theoretical studies and international examples which can provide insight and possibly a rational way for lawyers and society to decide the optimal number of jurors (a) on a panel and (b) to agree upon a verdict.
One view sees both the traditional 12 person jury and the commitment to unanimity as reflecting tradition rather than efficiency. The thought that one would be able to do more, more quickly, with smaller juries making decisions by, for example, simple majority has obvious appeal. When one looks at studies on jury decision-making, however, things become rather less clear.
The most famous theoretical discussion of jury decision-making is found in the Jury Theorem propounded by the Marquis de Condorcet (pictured below) in 1785, which, broadly speaking, maintains that if each member of the jury has at least a better than 50% chance of getting the right answer, then, under majority decision-making, the greater the number of jury members, the higher the probability of the overall judgment being correct. On this approach, what is needed is a jury large enough to be likely to come to the ‘right’ decision, but small enough to be practical – and a majority, or at least a super-majority, decision is likely to be sufficient.
But the converse of Condorcet’s theorem is also often noted: if each member of the jury is even slightly more likely to get the answer wrong than right, then increasing the size of the jury means that chances of reaching the wrong decision also rise. The challenge for those who would make practical recommendations on the basis of abstract theorising is to incorporate realistic assumptions as to jury behaviour into their models, and this is no easy task. So, for example, it is true that some experimental studies have claimed that decision-making procedures aimed at unanimity can encourage strategic voting and increase the probability of the jury getting the answer wrong (Feddersen, Timothy, and Wolfgang Pesendorfer, ‘Convicting the innocent: The inferiority of unanimous jury verdicts under strategic voting’, American Political Science Review 92 (1998): 23-35).
However, others have argued that the more realistic one makes the experimental models (by, for example, allowing communication between jurors and allowing those in a minority to hold out for a hung jury), the stronger the case becomes for unanimity as a decision-making rule, as jury members are able to vote non-strategically on the basis of their convictions (Coughlan, Peter J., ‘In defense of unanimous jury verdicts: Mistrials, communication, and strategic voting’, American Political Science Review 94 (2000): 375-393).
Things become, unsurprisingly, more complicated if one assumes bad faith on the part of some jurors, stipulating, for example, that they are biased in favour of convicting particular defendants. Larger juries and unanimous decision rules can have the effect of insulating verdicts from the distorting effects that even small numbers of such jurors can have (Mackenzie, Dana, ‘What’s the best jury size?’, Slate 24/4/2013).
From this perspective, larger juries are better than smaller juries. But how many jurors does that entail, exactly? It is striking that 12 seems to be the preferred number of jury members in a wide range of different jurisdictions, despite the number having its historic origins many centuries back, and notwithstanding the slightly larger number of 15 employed in criminal trials in Scotland. Mimicry? Coincidence? Or is something else going on? There is certainly theoretical support available for the traditional number of 12.
For example, a 2020 study by Takamitsu Watanbe entitled ‘A numerical study on efficient jury size’ studied data on jury deliberations and concluded that the efficient jury size is 11.8 (Humanities and Social Sciences Communications 7 (2020): 1-7). It should be admitted that in practice it is hard to obtain four-fifths of a juror, and acknowledged that this estimate has a margin of error of plus or minus 3. Even the downside of this, however, would not go as low as a 7 member jury. Watanbe suggests that practice may have converged on 12 for good reason: ‘the number of jurors may have been implicitly optimised for efficient unanimous decision-making throughout human history’. We may have reached the right answer without necessarily being able to show our workings.
Moving from the classroom to the courtroom, the experience of judicial systems that have in recent years relied upon relatively small juries gives rise to some real concern. We might look in particular at practice in the United States where the question of jury size has been a live debate for years. Although most American criminal juries have 12 members, the state of Florida reduced the number of jurors in non-capital cases from 12 to 6 in 1967. This practice was held to be compatible with the Sixth Amendment by the US Supreme Court in Williams v Florida (1970). The majority judgement, delivered by Justice White, expressed scepticism at the idea that jury size was of critical importance, holding that the point of having a jury was to prevent government oppression, providing ‘an inestimable safeguard against the corrupt or overzealous prosecutor and against the compliant, biased, or eccentric judge’, and arguing that, ‘[t]he performance of this role is not a function of the particular number of the body that makes up the jury’. ‘Certainly,’ he writes, ‘the reliability of the jury as a factfinder hardly seems likely to be a function of its size.’
Scholars since this judgment have not always shared the court’s confidence. In particular, writers have queried the representative nature of six person juries in racially diverse communities. Marc Pearce and Twila Wingrove write that ‘research has consistently shown that six-member juries are not likely to represent the community as effectively as 12-member juries’ (‘Is a jury of six as good as one of 12?’, Monitor on Psychology 40 (2009): 32).
The issue came to the fore of the political agenda in 2013 following the acquittal on second-degree murder charges of George Zimmerman for the fatal shooting of unarmed 17-year-old African-American high school student Trayvon Martin in Sanford, Florida. The verdict was reached by a jury of six, all of whom were female, and five of whom were white, and led to multiple calls for a revisiting of Williams v Florida and a return to 12 person juries. Such calls have persisted. A 2020 article by Patrick Higginbotham, Lee Rosenthal and Steven Gensler, ‘Bringing Back the Twelve-Person Civil Jury’ calls the 12 person jury the ‘gold standard’, arguing that larger juries are more predictable, make better decisions, and are more inclusive and representative of the community (Judicature 104 (2020): 47-57). History shows that the decision to abandon the gold standard is not one that should be taken lightly.
Concerns about diversity in our legal system should also sound a warning to those attracted to reducing jurors or jury trials. In her research on jury decision-making referenced above, Cheryl Thomas found little evidence that juries are not fair, noting in particular that verdicts showed only small differences based on defendant ethnicity. Ultimately, we do not know what the effects would be of reducing juror numbers. It is a sobering fact that in 2021, not a single High Court, Court of Appeal or Supreme Court judge self identifies as Black. As the legal system seeks to urgently address the systemic problems raised in for instance the Lammy Review or the issues eloquently described by Professor Leslie Thomas QC in Counsel July 2020, restricting trial by jury would strike at one of the few parts of our criminal justice system that is truly reflective of society.
Larger juries are better than smaller juries. But how many jurors does that entail, exactly? A theoretical discussion of jury decision-making and optimal jury size
By Matthew Butt QC and Daniel Butt
The beginning of the legal year offers the opportunity for a renewed commitment to justice and the rule of law both at home and abroad
By Louise Crush of Westgate Wealth Management sets out the key steps to your dream property
A centre of excellence for youth justice, the Youth Justice Legal Centre provides specialist training, an advice line and a membership programme
By Kem Kemal of Henry Dannell
By Ashley Friday of AlphaBiolabs
Providing bespoke mortgage and protection solutions for barristers
Joanna Hardy-Susskind speaks to those walking away from the criminal Bar
Tom Cosgrove KC looks at the government’s radical planning reform and the opportunities and challenges ahead for practitioners
From a traumatic formative education to exceptional criminal silk – Laurie-Anne Power KC talks about her path to the Bar, pursuit of equality and speaking out against discrimination (not just during Black History Month)
James Onalaja concludes his two-part opinion series
Yasmin Ilhan explains the Law Commission’s proposals for a quicker, easier and more effective contempt of court regime
