The International Forum for Responsible Media Blog

Libel Reform Debate: Part 1 “Burden of Proof”

In a post last week we suggested that there should be a mature debate on the various issues raised by Libel Reform Campaign.  The first of these issues concerns the burden of proof in libel cases.


In English defamation law, once the claimant establishes that the defendant has published defamatory allegations, the defendant bears the burden of establishing any “defence” which he or she wants to rely on.  In particular, a defendant who wants to rely on the defence of “truth” must prove that the allegations are substantially true.  The claimant does not have to prove either that the allegations are false or that damage has been caused.

The question of whether a claimant should be obliged to prove damage and how this should be done is a complex one which we will reserve for a later post.  In this one we will concentrate on the question of whether the burden of proving the truth of a defamatory allegation should be placed on the claimant or the defendant.

The Reform Argument

The Libel Reform campaign report “Free Speech is not for Sale” summarises the position in this way:

“In English law, the defendant in a libel case is asked to prove the truth of their statement, or that it was a ‘fair comment’, not intended as a statement of fact, or that the allegation, even if false, was made in the public interest. Thus the defendant carries the burden of proof. The English approach to libel therefore suggests that the reputation of the claimant is more important than the free speech of the defendant. This feature is one of the reasons why foreign claimants choose English courts over other jurisdictions that do not presume falsity. It is also an anomaly in English law, where defendants are usually presumed innocent until proven guilty”.

The report goes on to say:

`We also recommend that the claimant be required to provide evidence of falsity or unfairness when they bring a libel action. This reform would reverse the burden of proof, bringing English libel law up to global standards. We recognise that there are cases where it may be impossible for a claimant to provide evidence of the falsity of an allegation and in these instances the defendant may be required to bring evidence supporting the truth of what they have written.

We recommend: Require the claimant to demonstrate damage and falsity

However, Lord Lester’s Defamation Bill leaves the burden on the defendant to establish defences – including truth.  It does, however, require a claimant to show that the publication has caused “substantial harm” – if this is not done the action will be struck out (see clause 12).

Responses to the Proposal

In the report by Professor Mullis and Dr Scott “Something Rotten in the State of English Libel Law?” the authors criticise what they describe as the “appealing rhetoric” being deployed, arguing that

“A presumption of truth can equally be cast as an assumption that the person concerning whom a statement is made is guilty as charged in the trial by media, and that they deserve the lowered reputation. We query whether freedom of expression really demands that the media defendant be entitled to a presumption that a person deserves a bad reputation unless the contrary is established”.

They argue that “the presumption that an impugned statement is false is the preferable starting point” essentially on the basis that it focuses the attention of publishers on the question of whether the allegations being made are true:

“Without such a legal responsibility, speech would become cheap and the proper restraints placed on the media, or indeed any person, when making serious allegations would be undesirably loosened”.

In a speech given on 1 December 2009 “Privacy and the Press: Where are we now?” Mr Justice Eady addressed this issue. He made three important points.  First, he drew attention to the potential effect on journalistic responsibility:

“One can well see a certain attraction in a rule which says that everything a journalist writes is presumed to be true. But the question would need to be asked, “Would that have any effect on journalistic standards of accuracy and, if so, would that effect be in the public interest or not?””

Second, he considered the interesting question as to how much practical difference such a change would make

“What would happen, if the burden were to be reversed? The claimant would, at stage one, set out the words complained of, identify the respects in which they are said to be false and plead the defamatory meanings they are supposed to bear – so nothing new there. The defendant would be required at the next stage to say whether he or she accepted that or not; if not, obviously it would become necessary, in accordance with the rules, to narrow the issues in the case – and, in particular, to say in which respects it was claimed that the words were true. That sounds a bit like particulars of justification to me.  Disclosure of documents would take place in the usual way and then, if the case was one of the very few that actually proceeded to trial, the claimant would go into  the witness box and state that the words were untrue, as happens anyway, and go on to give an account of the relevant facts. There would be the usual opportunity to cross-examine, of course, but you would be rather unwise to count on the claimant breaking down in tears and admitting that, after all, it was all true. The defendant would, therefore, be left rather naked if no evidence was available to be called in rebuttal of the claimant’s case. So it would seem that very little had changed. The legislature might have laboured only to bring forth a mouse”.

Third, he asked the question as to the way in which the jury might be directed

“Suppose then that the jury find themselves in difficulty and ask the judge to give them further assistance. In a criminal context, one is quite used to hearing a judge in such circumstances say to the jury, “If there’s a doubt, then you acquit”. How comfortable would we be, in a libel action about police corruption or sexual abuse by a teacher or nursery nurse, if the judge were required by law to say in comparable circumstances, “Well, if there’s a doubt, members of the jury, just find the claimant guilty of corruption [or sex abuse, as the case may be]”? That is unlikely to arise in practice, since cases in reality do not turn on the burden of proof.”

Finally, he raises a question of principle

“How compatible would the proposal be with Article 6 of the European Convention? If you are accused of murder, terrorism, corruption or child abuse, you are entitled to a fair hearing before an unbiased tribunal. It needs to be carefully considered whether a tribunal could be classified as unbiased if required by law to presume that you are guilty”.

The Strasbourg Cases

The placing of the burden of proving truth on the defendant in defamation cases is compatible with the right to freedom of expression in Article 10 of the European Convention on Human Rights.  In McVicar v United Kingdom (2002) 35 EHRR 22) the Court of Human Rights held that

the requirement that the applicant prove that the allegations made in the article were substantially true on the balance of probabilities constituted a justified restriction on his freedom of expression under Article 10(2) of the Convention in the interests of the protection of the reputation and rights of Mr Christie” [87]

This approach has been followed by the Court on several occasions.


The placing of the burden of proof in relation to truth and damage is not anomalous in English civil law.  In a range of torts such as trespass or false imprisonment, once the defendant is shown to have done the “wrongful act”, damage is presumed and the burden of establishing a defence shifts onto him.   The “innocent until proven guilty” argument cuts both ways as discussed above.  There is also the practical problem of “proving a negative” – as the Libel Reform report appears to recognise in the passage quoted above.  The present system is clear, well understood and does not seem to produce substantial practical problems.

It is noteworthy that in other common law jurisdictions such as Australia, Canada and South Africa, the burden of proving the truth of an allegation is on the defendant.   In civil law jurisdictions such as France the burden is also on the defendant to prove truth.

The arguments in favour of reversing the burden of proof in all cases are not, in our view, convincing.   Although it is difficult to see what practical difference such a reversal would make, it seems to us that someone who makes a defamatory factual allegations should, in an ordinary case, be ready to prove that it is true.   We do not think that the case for reversing the burden has been made out.    As Lord Lester appears to recognise, this is not a reform proposal which should be taken forward.

We would, of course, welcome guest posts from anyone who wants to put forward a different view.

1 Comment

  1. Jonathan Bliss

    I’m curious about your libel reform report findings. I’m not involved in the media at all, just interested.

    My major concern relates to the behaviour of the tabloids, as a keen follower of the tabloidwatch blog I’m concerned about just how often the tabloids appear to lie with very ittle comeback.

    This particular case drew my attention

    I’m concerned that particularly the first two recommendations risk freeing the tabloids to say what they want. Firstly because essentially the claimant has to prove the lie and secondly because even if the tabloid loses, in many cases it may fell that it is money well spent.

    I’m also concerned that given the recent shenanigins with News International possibly being in the pockets of MPs this can only help them.

    Your first recommendation is that the Claimant has to demonstrate the damage and falsity.

    Having seen Simon Singh’s issue with Chiropractors can understand the concerns and the type of issue this is trying to address. However, the obverse is totally unfounded statements.

    If this change is in place what is to stop a tabloid, for instance, stating that Simon Singh is a paedophile? How would a lie like this be demonstrated as a lie? How could the negative be proven? Also, how would damage be proven. Clearly a lie like this, if believed, would result in a catastrophic loss of reputation and may result in the claimant being the victim of physical violence but wouldn’t attributing these in a court of law be very difficult. So what protection are you proposing for this kind of situation?

    The second issue is the £10,000 limit. This is too high for many organisations but in some cases the tabloid newspapers and magazines, who pay huge amounts for some stories may feel that it is worth it. I also don’t think many people are in it for the money, the rich want secrecy and the rest often just want the truth publishing with equal vigour.

    So would not a better system be to introduce punative penalties that DO NOT go to the claimant, either back to the courts or to the exchequer. The claimant may still be able to claim specific damages. These penalties would based on the breadth of publication, the income of the publisher including consideration of the income generated from the specific libel and the degree of maliciousness. They would be decided by a committee. The claimant would have the option to parlay some of this for well publicised correction, something they may well do since the money is not theirs. This would mean a wealth publication could still be hit while an innocent mistake by a parish magazine might amount to a 1 page apology or a £10 fine.

Leave a Reply

© 2023 Inforrm's Blog

Theme by Anders NorénUp ↑

%d bloggers like this: