Dominic Grieve – 2013 Speech on Trial by Google


Below is the text of the speech made by the Attorney General, Dominic Grieve, at the University of Kent on the 6th February 2013.

As Attorney General I have various roles. I was once described as the “man with two hats”. I have to say, I saw that description and thought – if only it was only two!

Most, if not all, the hats I wear are non-political; that is to say, I act independently of the Government, and certainly do not act in a political manner: legal advice is legal advice, and must not be calibrated to political considerations.

Various aspects of the role of Attorney embody the rather broad notion of being “the guardian of the public interest”. This includes having the ability to refer certain criminal cases to the Court of Appeal on the basis that the sentence imposed was “unduly lenient”, being required to consent to certain criminal prosecutions – for example, terrorist offences which impact on the affairs of another jurisdiction, or prosecutions under the Official Secrets Act.

Being guardian of the public interest also encompasses enforcing the law of contempt. Contempt is, broadly speaking, a jurisdiction to protect the integrity of the judicial system and the courts. Just as the judicial system has many facets, so does the law of contempt.

And, as we shall see, the nature of contemporary contempt is changing, but the purpose of the law remains the same: in this context, it is to protect the right to a fair trial.

Allow me to illustrate the point with the assistance of another jurisdiction.

Late last month, The Economist ran a story about two senior federal prosecutors in Louisiana who resigned in disgrace when it was revealed that they were the source of vitriolic, anonymous blog posts directed at particular Federal judges.

Their resignation was followed by resignation of their boss, the United States Attorney for the Eastern District of Louisiana, a well respected man with a reputation for campaigning against political corruption and white collar crime.

The Economist said this:

“The episode is a cautionary tale about the perils of the internet.”

Although many people think the anonymity that veils their online rants is absolute, plenty of jurisprudence argues otherwise…

The piece continued,

“Naturally, a host of federal targets – including some who have already pleaded guilty or been convicted – are now crying foul, saying the commenting amounted to a campaign to sway public opinion and poison the jury pool…”

It is an interesting, disappointing story; one which I hope is never replayed by prosecutors in this jurisdiction.

But it illustrates vividly an important point: what we do on the internet does matter, and it is not only our jurisdiction which is concerned about the impact of the internet and social media on the right to a fair trial.

Just as defendants in the state of Louisiana are concerned that the jurors who convicted them may have been swayed by improper influence, we too must be careful to ensure that our juries are not improperly influenced, whether through published material they inadvertently encounter, or through conducting their own research – which I have termed “trial by Google” for tonight’s purposes, for such research is usually internet-based.

The way our legal system mitigates those risks is through the law of contempt.

To think about this in more depth, I propose to cover:

Forms of contempt; in particular how the law of contempt protects the integrity of trial by jury

Contempt is a broad jurisdiction:

It is the means by which certain court orders are enforced.

It is the means by which judges regulate proceedings before them.

Contempt encompasses a summary jurisdiction, unique in our legal system, allowing judges to deal with certain matters as contempt in the face of the court.

It is more than a mere summary jurisdiction; as well as regulating what happens during proceedings in court, it covers what can be said about proceedings from outside court.

Some contempts are so serious that rather than dealing with it there and then, the judge refers it to me to consider bringing proceedings in the Divisional Court, part of the High Court.

The law of contempt regulates the behaviour of those involved in proceedings, including but by no means limited to, the jury.

As we shall see, the law relating to publication contempt and that relating to the conduct of jurors go hand in hand.

The Contempt of Court Act 1981 placed on a statutory basis what is known as the “strict liability rule”. That rule provides that a publication – and it must be a publication for the strict liability rule to apply – may be in contempt of court, regardless of intent to do so, for conduct which tends to interfere with the course of justice.

The strict liability rule is limited by section 2(2) of the Act to apply only to,

“a publication which creates a substantial risk that the course of justice in the proceedings will be seriously impeded or prejudiced…”

There is a defence to breaches of the strict liability rule in the following terms,

“a person is not guilty of contempt of court under the strict liability rule in respect of a fair and accurate report of legal proceedings held in public, published contemporaneously and in good faith”.

And finally, the rule only applies when proceedings are “active”; a concept about which, as I am sure you can imagine, much legal ink has been spilled over the past 30 years.

The 1981 Act was enacted following growing uncertainty about the scope of the former common law regime for strict liability contempt, which culminated in considerable criticism from the European Court of Human Rights in 1979 case Sunday Times v UK.

The Strasbourg Court held that an injunction obtained by the then Attorney against the Sunday Times to prohibit publication of an article breached its Article 10 rights.

Article 10, of course, guarantees the right to freedom of speech. It is a broad guarantee, and an extremely important one.

Freedom of speech, and its legal cousin, the open justice principle, feature in the legal system of any jurisdiction which respects the rule of law. Freedom of speech and the rule of law go hand in hand: both are certainly part of our proud common law heritage.

Freedom of speech encompasses not only the right of the media to speak, as it were, but also their right to gather material in order to exercise the right to free speech. It extends to the right of the public to be informed, by the media.

But it is not an unfettered right. Article 10(2) of the Convention provides that the right,

“may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of national security, territorial integrity or public safety…”

The list goes on, and concludes with,

“…maintaining the authority and impartiality of the judiciary.”

And the judiciary, of course, includes the jury in a Crown Court trial. Article 6 of the Convention guarantees the right to a fair trial; again, a matter of heritage for our jurisdiction in any event.

So Parliament, seeking to balance these competing requirements, enacted the Contempt of Court Act 1981.

Far from being a restrictive enactment, the 1981 Act was intended to shift the balance of the law in favour of freedom of speech. It sought to clarify what could and could not be published about legal proceedings.

By clarifying the law, restrictions on publication were defined – and thereby limited – by the Act.

For example, section 4(2) of the Act allows a court to make an order postponing publication of certain reports of the proceedings until some future date or event. This is how the provision was described by Lord Denning in a leading case some time ago,

[The Contempt of Court Act]

“is not a measure for restricting the freedom of the press. It is a measure for liberating it. It is intended to remove the uncertainties which previously troubled editors. It is intended that the court should be able to make an order telling the editors whether the publication would be a contempt or not.”

The paradigm example of this would be where there has been pre-trial legal argument, say to have the case thrown out as an abuse of process. Such argument would take place, of course, in the absence of the jury. It would be highly prejudicial, not to mention nonsensical, for a jury to be sent out of court during that argument, only to read about what happened in report of the proceedings the paper the following morning!

Or there may be several linked trials relating to the same crime: again, it would be highly prejudicial for the jury in one case to read of the evidence adduced in another. Of course, the same evidence may be common to both cases, but the manner in which it is adduced, the full context which will accompany it and the corresponding directions of the judge are vital components of the adversarial trial process.

Take away that context, and the evidence may assume an entirely different meaning. It was once said that if you take text out of context, all you are left with is a con, and I think there is some truth to that.

Underlying the strict liability rule is the recognition that the jury are entitled to – and will – read the papers, watch the news, and listen to the radio, and in doing so, encounter information about their cases, unless the judge directs otherwise. Parliament did not intend that jurors, or witnesses in the case, or even the judge, should be subjected to an automatic media blackout! If that was not the case, we would have a system of wholly secret justice.

Indeed, courts have a healthy realism about the integrity of jurors, their ability to focus on the evidence and to follow judicial directions. In 2006, the Court of Appeal said,

“There is a feature of our trial system which is sometimes overlooked or taken for granted… juries up and down the country have a passionate and profound belief in, and a commitment to, the right of a defendant to be given a fair trial. They know that it is integral to their responsibility. It is, when all is said and done, their birthright…”

We cannot too strongly emphasise that the jury will follow [the judge”s directions], not only because they will loyally abide by the directions of law which they will be given by the judge, but also because the directions themselves will appeal directly to their own instinctive and fundamental belief in the need for the trial process to be fair.

However, it must be true that by framing so carefully what may or may not be said about legal proceedings in the 1981 Act, Parliament recognised that much harm could be done by juries encountering information that falls outside that framework.

Put simply, we are not to have trial by newspaper.

All this poses a question, a rather significant question: How does a legal regime framed when the internet was but a gleam in the eye of Tim Berners-Lee cope when faced with the flow of information that now forms the fabric of our culture?

More specifically, what does the internet mean for our system of trial by jury? Is the trial process equipped, or even able, to regulate the information that jurors receive? How can we be sure that jurors decide their cases on the basis of the evidence they hear – and not what they looked up on their smart phones on the bus on the way to court?

To answer these questions, I will first consider two contempt cases I brought under the strict liability rule, before moving to address juror misconduct under contempt of court at the common law.

One of the first contempt cases I brought since coming to Office was that relating to the trial of a Ryan Ward . It was, the Divisional Court noted, the first time an internet-based contempt had been referred to them.

Mr Ward faced trial for murder in Sheffield Crown Court. The case had received a considerable degree of local publicity. It was the prosecution case that the defendant had murdered the victim following a gallant attempt he made to intervene in an attack by the defendant against a woman. The nature of Mr Ward’s defence, self defence and the absence of murderous intent, meant that the need for the media to abide by their obligations under the strict liability rule was as important as ever.

The jury was addressed by the trial judge in the following terms,

“Also, I would imagine by the nature of this case, and you’ll see there’s obviously press interest in it, there will be some reporting of this case. Again that’s a matter the press are free to report upon but you go on only the evidence you hear in this room, not the view other people may or may not have about it.”

He added another warning:

“Please don’t try and get information from outside this room about this case. Don’t, for example, consult the Internet, if there is anything out there on it. I’m not saying for one moment there is but don’t go there, don’t try and get it from anywhere else…”

During the early evening of the first day of the trial, the Daily Mail published an article under the headline, “Drink-fuelled attack: Ryan Ward was seen boasting about the incident on CCTV, alongside a photograph of the defendant holding a pistol with his finger on the trigger. The photograph remained accessible on the Daily Mail website for just under five hours; it was removed following a request from the police.

In the early hours of the following morning, The Sun published the same photograph on its website and in its print edition. The photograph in the print edition was cropped to conceal the gun; the online version was partially cropped, but that the defendant was holding a gun remained clear from the photograph. The photograph was taken down that evening, again following a request from the police.

When the matter was brought to the trial judge’s attention, he carefully asked whether any of the jury had seen the articles or the photographs.They had not. The case continued and Mr Ward was convicted of murder.

I brought proceedings for contempt. In this case, each defendant publisher conceded that publication of the photograph was wrong, and attributed the mistake to innocent error. But each – unsuccessfully – argued that the photograph did not create a substantial risk of serious prejudice.

The Divisional Court found the case to be proved; there was a substantial risk that a juror trying the case would see the photograph and be prejudiced by it. Each paper was fined £15,000 with £28,000 costs.

Far from highlighting any inability of the law to deal with internet contempt matters the Ward case clarifies, helpfully in my view, how the strict liability rule applies to internet publications, and what the consequential expectations on publishers are.

And although the two publications involved may not have welcomed the ruling, I think the clarity brought by the judgment has been welcomed by the media.

Such clarity was, after all, was one of the reasons which lay behind the enactment of the 1981 Act.

Shortly after bringing that case, I brought proceedings against the publishers of The Sun and The Mirror for their vilification of a man named Chris Jefferies during the investigation into the tragic death of Joanna Yeates in late 2010. It was clear from the outset of the press coverage during the investigation that the media “had their man”. Chris Jefferies was later to say that he became a household name, “for all the wrong reasons”.

There was nothing particularly new with this type of coverage; the media “feeding frenzy” is by no means a modern phenomenon. What was striking about the case was the rigour with which Mr Jefferies was pursued by the media during the period when the strict liability rule in the 1981 Act was supposed to be engaged.

The coverage sought to portray Mr Jefferies as plainly responsible for the death of the victim, associated him with allegations of child abuse, and referred to him as an “oddball”.

A melodramatic side piece titled, “1974 strangler never caught”, declared ominously “Last night police refused to rule out a link between the two killings” (which is hardly surprising: find me a single officer who will categorically rule out a connection between two similar unsolved crimes in the same area!). Another headline read, “The Nutty Professor” above a banner stating, “Bizarre past of Joanna Yeates murder suspect”.

The contempt was proved. What was interesting about the decision of the Divisional Court was that, not only did it consider the residual impact of the extreme publicity on any eventual juror, it also considered that the extent of the vilification may have deterred witnesses on behalf of Mr Jefferies, had he been charged, from coming forward, for fear of being associated with such an obviously guilty man.

Of course, not only was Mr Jefferies never charged, another man altogether was later convicted of the murder and sentenced to life imprisonment with a minimum term of 20 years.

So it is clear that the law of contempt does not permit trial by newspaper, whether that is in the print or online editions.

But neither does the law of contempt permit trial by Google.

(Of course, I say Google, I mean any internet search platform, Bing, Yahoo, Wikipedia, Twitter, Facebook, blogs, the list goes on…)

And this brings me onto common law contempt.

Common law contempt is intentional contempt. It is conduct which tends to undermine the administration of justice, done with the intention of undermining the administration of justice.

All juries are directed in robust terms about the need not to conduct their own research into the case. These robust instructions reflect the gravity of a juror’s task. Indeed, it is hard to think of a more serious or important civil duty that virtually any member of the public may be called upon to conduct.

It has always been necessary to direct jurors not to discuss the case with anyone, not to visit the scene of the crime, not to research the witnesses or defendant details. And now such directions extend to not researching the case on the internet.

To ignore those directions, intentionally, amounts to a contempt of court.

Intention, of course, is different to motive. While you may not desire, for example, to derail a trial, the law considers that by embarking on a course of conduct that is virtually certain to derail a trial, you have intended to bring about that result.

Before getting into the detail, a word about the internet and the law: am I trying to reconcile the irreconcilable?

It is often said that the system of trial by jury was the bulwark of our democracy, a bastion of freedom.

While few would dispute describing trial by jury in such terms, I wonder if for many it would seem more natural to describe the internet in that way: for many, the internet is now the champion of freedom. The connectivity it provides has gone some considerable way towards uniting the world as a global village. The role of social media in the Arab Spring uprisings is well documented.

The internet is surely, the argument goes, an unstoppable force for good?

Taking the argument a little bit further, in contrast to the liberation provided by the internet, we have the law; a rigid framework that is dour, unresponsive, and above all, lacks understanding of the changing role of technology in society.

I do not accept the premise of either assertion.

Certainly the internet has been and is a champion of freedom, and has played an important part in opening up some societies, and helping to achieve social change.

And certainly there have been times when the law has been unable to adapt to modern society, and those responsible for making the law have been unable to see the need for the law so to adapt.

But I want to argue this evening that the law of contempt is both adaptable and resilient in the face of the challenges of technology.

The strict liability rule very deliberately only applies to information which presents a substantial risk of serious prejudice or impedance.

About many criminal trials, there may be all kinds of prejudicial information “out there” on the internet. This could be in the form of archived news reports about the defendant’s previous court appearances, or it could be, to use the words of the old authorities in a modern context, mere chaff and banter about the case on someone’s Facebook page or Twitter feed.

The strict liability rule can be fairly relaxed about such material – it is unlikely to present a substantial risk of serious prejudice because it is a needle buried away in the haystack of the internet. (I say the strict liability rule can be fairly relaxed – it can be, but is not always!)

Indeed, most publishers are very careful not to link reports of live cases to archived news reports about the same defendant.

So the chances of a juror seeing such material are fairly slim, providing they haven’t gone looking for it. So, to use the words of Article 10(2) of the Convention, in view of those risks, attempting to purge the internet of all such material would not be “necessary in a democratic society”.

Trial by Google, however, is different.

The reason is this: I mentioned a moment ago that the internet is a haystack of material, scattered with the odd prejudicial needle, as it were. Trial by Google allows a juror to locate the haystack, find the needle, pull it out and ascribe significance to it that it simply would never have had otherwise.

It takes a minor risk and turns it into a major risk.

In doing so, trial by Google offends some foundational principles of our legal system.

The first principle is that a conviction, or for that matter an acquittal, should be based on evidence adduced in court, in accordance with established rules of evidence, subject to the supervision of the judge.

Let’s say a defendant being tried for grievous bodily harm had previously been tried – and acquitted – of rape. Let’s say the case against the defendant for GBH does not feature details of the rape allegations. And with good reason too: the strict rules of evidence relating to bad character do not allow that kind of highly prejudicial material to be adduced in the circumstances of this case. Even if the rape acquittal was admissible, the judge would have explained the relevance of the bad character evidence in careful terms. The jury trying the defendant are to base their verdict on the evidence adduced before them; the previous acquittal was excluded for good reason. Let’s say the judge has admitted some so-called bad character evidence, but that he was very careful in the way he crafted the directions to the jury about how it is relevant.

Now, let’s say one of the jury decide to take matters into their own hands by looking up archived news reports about the defendant – in defiance of the judge’s directions not to do that very thing.

Before too long, a bit of internet searching reveals that this is not the first time the defendant had been before the courts, and that he had in fact faced trial for rape. The rest of the jury must be told, the juror says to herself! There is information about the defendant that the judge is trying to withhold from us!

From this point onwards, the trial process is undermined.

The jury will no longer be able to deliver a verdict based solely on the evidence adduced before them; the role of the judge has been usurped, the defendant’s right to a fair trial is prejudiced. The press, who had been scrupulous in their reporting of the GBH matter, avoiding all mention of the defendant’s previous convictions, might as well have not bothered. The defendant may not have been tried by newspaper, but he was certainly tried by Google!

Of course, it is often hard to tell if the above research has been carried out, which leads us to the next fundamental objection to trial by Google:

Trial by Google offends the principle of open justice.

It should be clear to the defendant, the public, the victim and the prosecution what the evidence in the case is. If a jury is exposed to prejudicial material which, for whatever reason, is not before the court, the basis on which the defendant is convicted or acquitted will never be known.

The principle of open justice is met by our system of trial by jury through proceedings being in open court, through the adversarial scrutiny of the evidence, and through the judge’s directions to the jury before they retire to consider their verdict. All this is undermined by trial by Google.

A further facet of the principle of open justice is that evidence can be challenged, probed and questioned. Open justice is scrutinised justice. By definition, that is not so with trial by Google; not only is the basis of the jury’s finding unclear, but the parties will have been denied any opportunity to challenge the evidence which the jury itself gathered.

This returns us to our original question: is the law of contempt fit for purpose?

After all, we live in an information age. Searching for information about something we are unsure of is second nature for many; how can the law expect jurors to do something different? Surely only Mr Justice Canute would seek to stem the flow of the tide of information in this way?

Well the law can, and does, expect jurors to show restraint. The principles which underlie this expectation are nothing new.

All that is new is that there is an additional area in which jurors are required to show restraint.

The fundamental principles underlying the need for juror restraint are timeless.

The scenario I described a moment ago involving a juror searching for material about the defendant on the internet was, sadly, not fictitious. The juror’s name was Theodora Dallas and the defendant was called Barry Medlock. The trial was at Luton Crown Court in 2011.

For conducting her searches which revealed the previous acquittal of the defendant for rape, Dr Dallas was found by the Lord Chief Justice to be in contempt of court, and was committed to 6 months” imprisonment. Barry Medlock had to be retried before a fresh jury, and the victim had to give evidence again.

When passing sentence on Dr Dallas, the Lord Chief Justice said;

Jurors who perform their duties on the basis that they can pick and choose which principles governing trial by jury, and which orders made by the judge to ensure the proper process of jury trial they will obey, or who for whatever reason think that the principles do not apply to them, are in effect setting themselves up above the jury system and treating the principles that govern it with contempt…

The Lord Chief Justice went on to underline that the court’s robust approach was not borne out of lack of understanding of the significance and role of the internet.

Judges, no less than anyone else, are well aware of and use modern technology in the course of their work. The internet is a modern means of communication. Modern technology and means of communication are advancing at an ever increasing speed. We are aware that reference to the internet is inculcated as a matter of habit into many members of the community, and no doubt that habit will grow. We must however be entirely unequivocal.

Pausing there for a moment, I think I should highlight that it was our current Lord Chief Justice who, in December 2010, first permitted the use of live, text-based communication from the court room, initially on an interim basis, and later on a settled basis.

The allegation that the judiciary do not understand the internet is simply without merit.

Returning to the Dallas case, the Lord Chief Justice continued,

The problem therefore is not the internet: the potential problems arise from the activities of jurors who disregard the long established principles which underpin the right of every citizen to a fair trial.

I endorse those remarks.

Indeed, the internet has made the commission of many criminal offences much easier. It would be absurd to suggest that such conduct should no longer be criminalised on account of the ease with which such offences can now be committed.

Given the focus of my remarks has been on the need to prevent jurors from accessing prejudicial material, advertently or inadvertently, I have not spent time examining the potential for jurors to use the internet to communicate with defendants, or indeed witnesses, using the internet.

That is not so much trial by Google, but rather trial by Facebook Friend Request. That the law is apt to deal with such irregularities was demonstrated in the case of Frail and Sewart where a juror, Frail, initiated contact with a defendant, Sewart, whom she had just acquitted. The contact was made while the jury was still deliberating the guilt of the remaining defendants in what had been a long, multi-handed organised crime trial.

Details of the jury’s deliberations were revealed by Frail in the course of a Facebook chat she initiated with Sewart. In doing so, Fraill breached the prohibition against that very thing contained in section 8 of the 1981 Act, in addition to breaching the directions of the judge not to go on the internet to research the case.

Both denied the allegations, and a trial took place before the Lord Chief Justice, who, sitting with two High Court Judges, found the case proved.

Sewart was sentenced to two months’ imprisonment, suspended for two years, on account of her young child and the fact that she had already spent 14 months on remand prior to her acquittal. Frail received a term of 8 months’ immediate custody.

I mention the case as I conclude because it further demonstrates the flexibility of the existing legal framework law to this very modern form of offending. Save for a brief discussion at the contempt trial about the true meaning of LOL – opinions vary – there could have been no allegations that there was any lack of appreciation in the court room of the impact of modern technology on the trial process, nor what to do about it.

Avid followers of my contempt caseload will know that I have mentioned only a few of the recent cases I have brought.

I have also not mentioned the Law Commission’s excellent consultation on this very topic, which is due to close on 28 February. It raises important questions about, for example, whether some of the conduct I have outlined this evening should continue to amount to a contempt of court, or whether it should be classified as a criminal offence. I am sure the Commission would be delighted if you were to respond. I am proud to say that my Office has worked closely with the Law Commission during the consultation process, and I look forward to reading the final report next year.

But in what I have said, I hope that I have demonstrated that the legal framework for the jury trial in this jurisdiction starts from the premise that the jury are to be trusted, and establishes a framework in which their vital function is to flourish, and be preserved.

We have never allowed trial by newspaper; and neither do we allow trial by Google.

The Economist article I outlined earlier concluded with the exhortation to the new United States Attorney for the Eastern District of Louisiana to, and I quote, “stay out of the chat rooms”.

I can only conclude by imploring jurors in this jurisdiction to do likewise.