Jury duty

I recently completed some weeks as a member of the jury in a criminal trial. It was a strange experience. Without wishing to comment on the case itself (at least, not in a way that identifies any of the people involved), I want to talk about it. Jury trials are a signature of our democratic way of life — that is a commonplace opinion in Australia, Canada, the United States and other countries with a British imperial heritage. But juries are rare or non-existent in many other democracies, and the rationale for their existence is something more believed than proven. Indeed, in a justice system that is emphatic about the value of evidence, the evidence for the effectiveness of juries is remarkably thin. Because the privacy of jury members is protected and juries do not give reasons for their decisions, gathering evidence about their effectiveness is difficult. At the heart of our criminal justice system, we have instituted a mystery.


I have never been on a jury before. All New South Wales voters (with some exceptions, most importantly lawyers and criminals) are eligible to be called for jury duty,[1] but as the selection is random, it is possible to go through life without ever being called. I was beginning to think that would happen to me. Sometimes I would encounter someone who would say they had been called for jury duty — usually they would say this with a groan, and explain how they were hoping to get out of it. Inwardly, I would disapprove of this shocking lack of civic responsibility, because I had long accepted that juries were central to the fair administration of justice, and that the duty to be a juror, like the duty to vote in elections, was honourable and not to be shirked. I envied people who had the opportunity. When my turn finally came, there was no way that I was going to miss it.

The system is this: I received a notice that my name had been selected from the electoral roll to be included on the jury roll. Some months later, I received another notice, asking me to attend for jury duty at the Downing Centre (Sydney’s central courthouse) on a certain day. This did not mean I was on a jury, only that I had to show up for possible selection.

I arrived at the appointed time, finding a crowd of potential jurors waiting for admission to the building. Also loitering outside were a few photographers armed with long-lens cameras, looking out for noteworthy crims, lawyers or celebrities who might be attending court that day. Security at the entrance was tight: the first layer intended to exclude people who might have Covid-19; the second (airport-style) to exclude weapons.

The Downing Centre
The Downing Centre, corner of Liverpool and Castlereagh Streets, Sydney. Once the home of the Mark Foys Department Store, The Downing Centre now houses the District Court and other law offices (https://en.wikipedia.org/wiki/Downing_Centre). Image by J Bar; The Downing Centre.JPG, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=1227480

Inside, I was directed to the jury assembly area and given a number. The airport-like experience continued as I sat in a cavernous room along with a few hundred other people. Most of these people were called up at some point during the day; some of them were called only to be told that they were dismissed, their jury duty complete for the time being. I waited. Around lunchtime, some of us were asked if we were willing to return on another day. I said yes: those who did not volunteer were dismissed.

Returning a few days later, I was directed to wait again, but this time, after only an hour or so, my number was called and I became one of a jury panel — this being twenty or thirty people from whom the jury would be selected. We were led into a room adjoining a courtroom and the process of selection began.

First, we were told, in general terms, the nature of the crime to be tried and asked to exclude ourselves if we thought we could not participate impartially. We were shown the judge and the lawyers acting for the Crown (that is, the prosecution)[2] and the defence, and asked to exclude ourselves if we knew them. We were told the names of the complainant, defendant and witnesses, and again asked to exclude ourselves if we knew any of them. We were then asked if we had any other reason to be excluded. This whole process eliminated only one juror from our panel.

Twelve juror numbers were then drawn by lot, and these people (including me) were led into court. Because of distancing requirements in the current pandemic, only half of us were placed in the jury box, and the rest were scattered in part of the public gallery. One by one, we stood to face challenge by the prosecution and defence counsel.

This process, known as ‘peremptory challenge’, involves no questioning of the juror. Counsel for the prosecution and the defence simply look you up and down, and if they don’t like what they see, they say ‘challenge’, and you are led away. Crown and Defence have three peremptory challenges each. We had been warned about this, and told that we should not take offence if challenged. Some people stand there hoping that they will be challenged: it is their last chance to avoid the responsibility. Others who are challenged look quite put out.

With the exception of one ferrety individual, I could not see any rhyme or reason to the challenges made by Crown or Defence. Only when these challenges were exhausted did we have a jury, which was then sworn in. This jury included people of both sexes and a wide range of ages and ethnic backgrounds. As I was to discover over the weeks of the trial, it was a genuinely diverse group and at the same time quite ordinary — just what you might expect from a random process working upon a group of Australian citizens preselected by the process to be law-abiding, literate, at a fixed address, and physically and mentally stable (or at least not floridly insane).

Jury selection is the first and fundamental part of the magic of juries. A jury is supposed to represent the community in the justice system. Although a jury is conscripted to apply the law, which of course is not necessarily in step with community values, there is a sense in which juries are imagined as bringing community values into justice. As the NSW Department of Justice webpage puts it:

Jury service plays an important role in our justice system. Juries are used to ensure that legal verdicts are impartial and in line with community standards of behaviour. [3]

Impartial: as in, not dictated by the State, or imposed by a biased and entrenched judiciary, or open to corruption. In line with community standards of behaviour, inasmuch as an independent jury that represents the community will do what the community thinks is right.

There is so much to consider here. Are randomly selected juries really capable of being impartial? Are they really in line with community standards? Is it even possible to be both impartial and in line with community standards? And can any of these things be done when a jury is asked to work within the strict confines of a legal system that narrowly defines the questions a jury can address, reducing its input to a closed yes or no that always addresses an individual and never the society in which that individual lives? The jury in a criminal trial decides whether someone is guilty or not guilty, but it does not decide what is a crime, who gets charged, how they are investigated or tried, how the guilty are punished, or how the victims and the innocent are compensated. It is possible that the main role of the jury is to shore up community support for a justice system that is otherwise quite isolated from and powerful over the community. The accompanying diagram may suggest to you that jury trials can only play a symbolic role in the administration of justice and the maintenance of a sense of national identity.[4]

Standard arguments for and against juries are readily available, so I won’t repeat them here (see the notes below if you want more).[5]

Studies of the effectiveness of juries are hampered by a central feature of the system, which is that juries deliberate in secret and give no reasons for their decisions. One method of study is to compare a jury decision to the decision that the judge in the case would have made if there had been no jury. In studies of this kind, the agreement of judge and jury ranges from 50% to about 75%, but of course there is no ‘gold standard’ for determining which decisions are factually correct and just. Are the juries off the mark, or are the judges? Even when they do agree, where is the proof that the decision is right? In fact, if we expected that juries would match the decisions of judges, why would we bother with juries at all?[6]

In the case of the jury that I was part of, all jury members seemed to be seriously engaged by the process. The trial went on for many days, and we listened intently even during the most banal presentations of evidence. Some of the matters being presented were confronting, but the legal process wraps everything in a cotton wool of formality, hesitation, delay and repetition that goes a long way to wash the emotional stain out of the facts of the case. In turn, the contestability of evidence is highlighted in the speeches of the counsel and the judge at every turn, so that soon even facts themselves lose all shape and colour.

It is striking how vague witnesses can be, how imprecise and contradictory. In this trial, one witness gave dates separated by months for one event, and was contradicted by other witnesses whose date estimates for the same event ranged over a whole year.

I agree with this observation by researcher Yvette Tinsley:[cited in 6]

Jurors do not piece together information and make sense of it at the conclusion of the trial. Rather, the New Zealand research confirms the “story model” of jury decision-making (Hastie,et al, 1983; Pennington and Hastie, 1990). Jurors actively process the evidence as it emerges, evaluating it and attempting to fit it into a story that makes sense to them. This suggests that the initial frame which jurors adopt in order to construct their “story” is important, because it is this frame that enables jurors to sift and interpret the evidence as it begins to emerge. Although jurors are willing to change the story as new elements are introduced, this is inevitably based on their understanding of the earlier evidence, which in turn is the result of the process of filtering and interpretation that has already taken place. Jurors who fail to construct a coherent story that relates to the emerging evidence risk confusion and misdirection. At best they have difficulty identifying and focusing on the relevant issues; at worst they reach decisions based on theories and interpretations of the evidence which bear little relationship to the disputed issues in the case.

We had at least one juror who was ready to reach a verdict by the end of day one; fortunately, by the end of the trial, the rest of us agreed, or we might have gone into the black hole of a hung jury. Along the way there was much frustration with evidence that did not seem to fit any ‘story’ at all, and much agonising with the meaning of the phrase ‘beyond reasonable doubt’. For some people, ‘reasonable doubt’ means any doubt that they can imagine; taking this approach in a trial would lead inevitably to a ‘not guilty’ verdict, even in cases where the accused was caught in the act. Other people are the opposite: once they have formed a view, any suggestion that there might be a doubt about it they see as a personal affront.

In this trial, as it turned out, the jury found that the only reasonable story which fitted the few facts agreed by both sides was that the accused was guilty. To conclude otherwise would have required us to believe that the complainant was an expert liar who decided to invent her complaint because she knew it would do great harm to her, her family and friends.

Delivering this verdict, unanimous only after two days of deliberation, was the emotional crisis of this jury’s work. I know it reduced some jurors to tears — not because they regretted their decision, but because they suspected how significant the outcome would be for the complainant, the accused and the people close to them — all real people not too different from we jurors ourselves. I think this human element is the best thing about the jury system. The imprecision of the process is a fitting match for the fuzziness of truth. In some ways, the best we can hope is that people will make these sorts of judgements with compassion and honesty.

Rhetorically speaking, that is a good conclusion to come to — but really, how do we know that juries (or judges, or courts, or laws) are a good idea? Some people might say that we know because we observe the weight of history and tradition behind our practices. If so, bring back hanging, I say. The horrible truth is that we do not know how good our legal institutions are; we are culturally attuned to them, and unable to imagine ourselves without them. We are sure things could be worse, but not whether they could be better.

What we really need is an experimental approach to these social institutions: some randomised controlled trials of different ways of holding trials, for instance.

Here’s an example: the hypothesis that any twelve randomly selected people will make a jury that can be trusted to reach a just decision could be tested by mounting a trial in which all the evidence is video-recorded and documented rather than presented live. Exactly the same trial could then be shown to not one, but (say) ten juries. If all ten juries reached the same verdict, this might be evidence that random jury selection was producing consistent results. But what would you think if four juries said guilty, three not-guilty and three were hung juries that could not reach a verdict? If eight out of ten juries agreed on the same verdict, would you say that a failure rate of two out of ten was acceptable?

As any scientist could tell you, a single experiment of this kind would not be conclusive. A statistician could give you an idea of how many experiments might be required to give you evidence beyond reasonable doubt. I suggest, however, that we have never collected that kind of evidence about the reliability of a system supposedly centred upon proving things ‘beyond reasonable doubt’.

A lawyer might scoff at the idea of this kind of experiment, pointing to a range of technical difficulties (which would indeed be considerable) and perhaps saying that the cost could never be justified. But the current processes of the law are expensive, and perhaps we don’t know whether the expense of time and money that goes into the jury system is really worth it. What price should we pay to ensure justice?

I am not saying that the experiment outlined above is the one and only experiment worth doing, so I won’t get into details of how it might be done, or consider the ethical questions involved in making it real. The important point is that an experimental approach is valuable in the social sciences and public policy just as much as it is in the physical and biological sciences. Experiments frequently unlock unexpected findings and lead to new and better ways of doing things. In law, it would be fruitful to test alternative approaches to legal problems and to adopt scientific methods for evaluating changes to law and legal processes — methods such as randomisation, prospective data gathering, reliability testing, double-blind experiments, and so on.

Of course, I am not the first to have thought of this. People have been calling for a bit of scientific thinking in the law for close to a century at least.[7] Some academic organisations and research projects now call for randomised controlled trials and other experimental methods in matters of law and public policy,[8] although it is surprising the degree to which law reform is still something for clever people to argue about, without much resort to the kind of tests that might show whether proposed reforms are able to achieve their stated objectives. You can imagine how chaotic it would be in medicine if drug development was a process of having an idea and persuading people that it was a good one, with no controlled testing of the actual effect of taking the drug. You can imagine, too, how disastrous it would be in medicine if treatment decisions started to be tied to political positions regardless of the experimental evidence on treatment effectiveness.[9] Yet, in relation to the law, having a political position is often both the beginning and the ending of the whole search for justice, with little time for an excursion into open-minded investigation.

© 2020 Craig Bingham

Read something similar:

Towards an ameliorist manifesto [opinion]
The networked car [opinion]

Read something different:

departure [fiction]
Tables [memoir]
What I did on my holidays [opinion]
A fatal case of Art [review of The horse’s mouth]


[1]      I say ‘New South Wales voters’ because it is the New South Wales Jury Act 1977 that sets out the rules for juries where I live (https://www.legislation.nsw.gov.au/#/view/act/1977/18). Other Australian states have similar rules, which I imagine are not too different from the law in Britain or other Commonwealth countries.

[2]      The Crown: how quaint that in Australia, nominally ruled as we are by the Queen of Australia, who is actually the Queen of England, the metonym ‘Crown’ stands for the State and for the prosecutor who represents the State. This is the sort of thing that gives the law its otherworldly sheen, as do legal robes and wigs of a style unfashionable since the eighteenth century. It baffles most people. Some of the trial documents we were given were labelled ‘Regina v [name of defendant]’. ‘Who is Regina?’ asked one jury member. ‘It’s a pseudonym for the complainant,’ another guessed. When I told them that Regina was latin for Queen and another way of saying ‘The Crown’, they could scarcely believe it — what did the Queen have to do with it? When I told them that if we had a King it would say ‘Rex v [defendant]’ they thought I was joking. Rex is a dog’s name.

[3]      Jury service in New South Wales. NSW Department of Justice. http://www.courts.justice.nsw.gov.au/Pages/cats/jury_service/jury_service.aspx

[4]      The diagram refers to a report from the NSW District Court: Joanne Baker, Adrian Allen and Don Weatherburn, Hung juries and aborted trials: An analysis of their prevalence, predictors and effects; Crime and Justice Bulletin 2002, No.66, https://www.bocsar.nsw.gov.au/Publications/CJB/cjb66.pdf.

[5]     Quoting Chloe Watson and Professor Jill Hunter (Why do we have juries? https://newsroom.unsw.edu.au/news/business-law/why-do-we-have-juries):
CW: What are the advantages of a jury?
JH: The earmarking of community standards highlights the strength of the jury system. Juries ensure community representation informs the weighing of evidence and allowing everyday perspectives to be incorporated into judging those accused of serious crimes. They ensure this decision-making is not just the province of elites (like judges) and keeps apace of changing community values.
Jurors in a trial also force transparency into the process by requiring evidence in court to be accessible to the average member of the community.
There are other advantages – in reaching complex decisions, 12 heads are better than one; gender and racial diversity are intrinsic, albeit imperfectly, in a jury mix; and, as jury deliberations require jurors to discuss, explain and deliberate, there is an airing and accounting of contrary views, in a process that reflects democratic principles.
CW: What are the disadvantages?
JH: The jury system is by no means without challenges. Juries do not provide a written judgment, so there is a lack of transparency in their verdict. It can also be difficult for expert witnesses to communicate complex or technical evidence to lay jurors. This is especially a problem if technical jargon obscures rather than enlightens the issues in dispute.
It can be difficult to create a fair representation of the community on juries, let alone a ‘jury of peers’. Originally (and not so long ago), juries were all-male and all white. Even now, Indigenous Australians are under-represented on juries. This a serious issue in light of the over-incarceration of Indigenous people.
The impact of media, including social media, can prove challenging to jury trials. For example, the Victorian County court’s attempt to suppress media reports of the guilty verdicts against George Pell in late 2018 was frustrated by many international media agencies. The court-ordered suppression had sought to insulate a possible second jury trial from the danger of jurors being emotionally influenced by the earlier guilty findings.
For arguments for and against juries, in the context of some research into their operation, see: Sex, Drugs, the Internet and Juries (10 March 2011). https://www.youtube.com/watch?v=c7Uoetud3zc . This is a lunchtime lecture from the UCL’s Jury Project.
Finally, an article in The Conversation gives some of the mechanics, advantages and problems of juries before concluding:
A few overseas studies have asked trial judges what verdict they would have come to in jury trials. A comparison between what the judges said and the real jury verdict reveals a high level of agreement between the two.
While scientifically we cannot confirm that specific jury verdicts are “correct”, the jury system is necessary for Australia’s justice system. Jacqui Horan, All about juries: why do we actually need them and can they get it ‘wrong’? (https://theconversation.com/all-about-juries-why-do-we-actually-need-them-and-can-they-get-it-wrong-112703).

[6]     The article ‘Juror Decision-Making: A Look Inside the Jury Room’ (Yvette Tinsley, in  Papers from the British Society of Criminology Conference, Leicester, July 2000; http://www.britsoccrim.org/volume4/004.pdf) is good on this and also points to other academic studies of the jury system. This article is also my source for some of the statistics quoted.

[7]     See, for instance: Walter W Cook. Scientific method and the law. American Bar Association Journal 1927; 13(6), 303-309. https://www.jstor.org/stable/25707158 . There are plenty of other examples, but one that I thought looked interesting from the 1980s was: Nancy Levit. Listening to Tribal Legends: An Essay on Law and the Scientific Method. Fordham Law Review 1989; 58(3). https://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=2859&context=flr

[8]     For example: Laura Haynes, Owain Service, Ben Goldacre, David Torgerson. Test, Learn, Adapt: Developing Public Policy with Randomised Controlled Trials. UK Cabinet Office and Behavioural Insights Team, 2012. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/62529/TLA-1906126.pdf  . A critical response is in: Warren Pearce, Sujatha Raman. The new randomised controlled trials (RCT) movement in public policy: challenges of epistemic governance. Policy Science 2014; 47: 387–402. https://link.springer.com/content/pdf/10.1007/s11077-014-9208-3.pdf

[9]     You don’t need to imagine: America is now in the grip of a wholly unnecessary pandemic catastrophe simply because politics Trumped science.


  1. Great writing, thoughtful and interesting. I hope you have read the book by Bri Lee, Eggshell Skull. I became concerned at juries from running a series of sexual assault trials in rural settings in the 1990’s. The verdicts followed a predictable pattern – white victim = guilty, black victim = not guilty. On your proposed experiment, I suspect that if you did the same exercise with judges or even panels of judges you would get the same disparity. Look for example at the Pell case. I have conducted literally thousands of sentencing exercises with judicial officers where on the same facts and record, they sentence entirely differently. Imperfect to say the least.


    • Hi David, thank you so much for commenting! I haven’t read Eggshell Skull yet, but look forward to it. I know you have eminent experience in justice issues, so I take sad note of your observations: there’s plenty of room for prejudice in our system, and not much reliability in legal decision-making. I would like to see some experimentation in new methods, or at least some more data gathering to lift the veil on the variability of outcomes under our current process.
      Regarding the Pell case, I imagine that a conservative legal perspective on this would be to say: ‘The system works — this is why we have appeal courts — the final decision was the right one.’ A comforting assertion, but not a demonstrated truth.
      Best wishes,


  2. Around 2011, comparative rates of incarceration (per 100,000) were USA, > 720, Australia 112, Ireland 96. I’m currently reading ‘The End of Policing’ by Prof Alex Vitale, a dreary but important construct which surely suggests that juries are of limited utility if society’s laws render you unnecessarily guilty.


    • Thanks, Kev — good statistics to bear in mind. This is definitely one of my points about juries — they work in a tightly controlled way within a much bigger system that may or may not be fair. I’m sure that law reform has bigger things to work on than the jury system.


Leave a Reply to David Heilpern Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s