Harcombe v Associated Newspapers Ltd

This is a libel action that is brought by two individual Claimants. They have separate claims, but for a considerable period now they have been case managed together. The Claimants are Zoe Harcombe PhD and Dr Malcolm Kendrick. They were the subject of articles that were published in the Mail on Sunday and Mail Online in March 2019. In the print edition, there was a main news article and an inside piece. The main news article appeared on p.2 in the Mail on Sunday, on 3 March 2019. The headline was, “Statin deniers are putting patients at risk says Minister.” There was then a separate article that appeared in the Health, Wealth and Holidays section of the newspaper at pp.47 to 50. The supplement article, as I will call it, was trailed in the main article, with a sub-headline, “‘Experts’ behind scare stories and why they are wrong pp.47 to 50”. At p.47 there was a full page spread to begin the supplement article, headed

For the purposes of today’s rulings, I do not need to set out the text of the articles. What I can do, instead, is to set out the meaning that the Claimants attribute to the articles in the Particulars of Claim. There are subtle differences between the meanings that are attributed to individual articles, but there is a degree of commonality. It suffices for me to set out just one of the pleaded meanings to give an overview of the complaints made by the Claimants:

Substantially the same meaning is pleaded in para.6.2 in relation to Dr Kendrick with necessary changes to references to the individual Claimant. As I say, substantively the same meanings are complained of for the online publications as well.

Early in the proceedings, the parties tried to find a way of advancing the case in stages, as is now common in defamation claims. Obvious candidates for early resolution in most claims are the issues of meaning and whether the publication complained of was or contained an allegation of fact or expression of opinion. Resolution of these issues in most defamation claims, even prior to service of a Defence, can offer substantial benefits in terms of case management, see Bokova -v- Associated Newspapers Ltd[2019] QB 861 [10] and Morgan -v- Associated Newspapers Ltd [2018] EWHC 1850 [10].

In this case, there were difficulties in resolving these issues arising from the Defendants’ reliance upon qualified privilege (on several bases) to protect publication of certain paragraphs of the articles. The articles contain a statement by the Secretary of State for Health as well as reference to some scientific studies. The Defendants rely upon defences of qualified under s.15 and Schedule 1 Part 2, para.9(1)(b) Defamation Act 1996 and s.6 Defamation Act 2013.

The complicating factor is the Court of Appeal’s decision in Curistan -v- Times Newspapers Ltd[2009] QB 231. That decision, which predated the Defamation Act 2013, is authority for the position that the court must resolve the extent to which the publication complained of is protected by privilege before the court can determine meaning. In other words, when performing the test of deciding what is the meaning, the natural ordinary meaning of an article, the court must first remove from its consideration such parts of that article as the court finds is protected by qualified privilege.

There has been some criticism of that decision. The authors of Gatley suggest, at para.30.8:

One of the consequences of the Curistanprinciple in relation to this claim is that it stands as a fairly significant impediment to the court determining the natural and ordinary meaning of the article. To do so, the court would have to resolve the question of qualified privilege. The determination of whether the pleas of qualified privilege protect parts of the publications complained of, in turn, would require the court to determine also the plea of malice that the Claimants have advanced. There are other implications of Curistan which have emerged only since the Defamation Act 2013. As the objective meaning of a publication is likely to be an integral part of the assessment of serious harm under s.1 Defamation Act 2013, that means that any dispute as to serious harm to reputation could also only be carried out after the issues regarding qualified privilege have been resolved.

There is no doubt that the impediment that Curistan represents has significant implications for this case and its case management. The court now has the benefit of full statements of case filed by the parties. It is no exaggeration to say that the parameters of this litigation are very substantial. Indeed, this is the most significant piece of defamation litigation that I have seen in a very long time.

It would take too long for me to set out the various defences relied upon by the Defendants. It suffices if I quote from the summary provided in the Defence:

In the Reply, a similar summary has been given of the Claimants’ case:

The parties had tried themselves to fashion a way of enabling the court to resolve the issue of meaning. A consent order was put before Jay J and an order was made directing the trial of the following preliminary issues:

Directions were given for exchange of evidence and the trial was listed for 1½ days on 23 March 2021. Shortly before the preliminary issues were due to be tried, the case was referred to me. I was concerned, initially, about the time estimate which appeared to be a substantial underestimate. More generally, but particularly given the likely scale of the trial, I was concerned that there should be some reconsideration of whether resolution of the preliminary issues remained sensible. In an Order dated 11 March 2021, I vacated the trial and converted hearing on 23 March 2021 to a directions hearing. My concerns were explained in the Order:

Thereafter, a Case Management Conference was fixed for 14 December 2021 (although this had to be adjourned due to the impact of the pandemic). In the meantime, the Claimants gave further consideration as to whether it would be possible to devise alternative preliminary issues that could be resolved that did not have the problems that I had identified. On 11 October 2021, the Claimants issued a further Application Notice seeking orders directing the trial of the following preliminary issues:

Having considered those applications, on 27 October 2021, I refused to delay the service of the Reply, granting only a limited extension of time for that to be filed. I directed that the Claimants’ Application for the revised preliminary issues should be listed to be heard at the CMC in December 2021. The reasons I gave for that order were as follows:

As I have already noted, the CMC could not take place in December and had to be refixed for today’s date. During that period, the Claimants changed tack and decided not to pursue their application for the preliminary issues. Instead, the preliminary issue baton was then picked up by the Defendants who contended that the Court should direct the resolution of a different list of preliminary issues.

With that lengthy introduction, I can now turn to the matter that requires resolution today: whether I should direct now the trial of the preliminary issues in the terms sought by the Defendants. The draft order provided by the Defendants identifies the preliminary issues that the Defendants say the court should direct:

Resolution of issues (i) to (iii) would remove the Curistanobstacle. If issue (iv) was not resolved in the Defendants’ favour, the way would be clear for the Court to go on and resolve, as the final preliminary issues, the issues of natural and ordinary meaning and whether the articles were (or contained) allegations of fact or an expression of opinion. There are subsidiary issues that have been identified and which would have to be resolved by the Court determining the issue of meaning. One of those is whether the news article and the supplement article of the print publication are sufficiently closely connected to be regarded as a single publication. That is important because if they are to be regarded a single publication, they will have a single natural and ordinary meaning. The Defendants contend that they would have been read separately and so the news article and the supplement article would have separate meanings. I understand the argument to be that readers who only read the news article would not understand it to refer to the Claimants. To cater for that, the Claimants rely also on an innuendo based on the class of reader who would have read both articles. That is not a matter that the Defendants contend should be resolved as a preliminary issue in their proposals.

Originally, the Defendants had additionally proposed the inclusion in the first preliminary issue trial, “Trial 1” as it has been referred to in argument, the resolution of the honest opinion defence. An honest opinion defence under s.3 Defamation Act 2013 has several elements and they are familiar and straightforward. The section sets them out very clearly.

It is important to note here because it assumes some importance later that subsection (5) is the statutory formulation of the form of malice that was capable of defeating an honest opinion defence at common law: see Tse Wai Chun Paul -v- Cheng[2001] EMLR 31 For today’s purposes, the important point is that is the only part of the honest opinion defence (excluding subsection (6)) which focuses on the actual state of mind of the relevant defendant. The remaining elements of the defence under the Act are all objective questions.

The Defendants originally included the honest opinion defence in their proposed list of issues to be determined at the preliminary issue trial. They reflected on that and withdrew that issue from their proposal because, I think, they recognised that the factual issues that arise under s.3(4) would have to be tried together with the defence of truth. In other words, the honest opinion defence would have to be resolved in Trial 2 if the result of Trial 1 did not dispose of the entire claim.

The Claimants’ position advanced before me today by Mr Busuttil is that the court not to accede to the application for preliminary issues. They argued that nothing had changed substantially from my decision, set out on 11 March, as to the pitfalls of ordering preliminary issues. In summary, preliminary issues often turn out to be treacherous shortcuts which rather than producing case management benefits, instead achieve only delay and increased costs. By reference to McLoughlin v Junes(quoted above), none of the factors (a) to (d) apply. Nevertheless, and as Mr Busuttil accepts, the court does retain an overall discretion to consider whether a case is suitable for the trial of a preliminary issue. No two cases are the same and that ultimately the court has to make an assessment of the ultimate cost/benefit analysis as it appears to the court at the point where it considers making the direction for the preliminary issues.

Apart from those general points in opposition, there are specific grounds on which the Claimants contended that the court should not direct the preliminary issues as sought by the Defendants. First, the risk of overlapping issues between those matters tried in Trial 1 and those left over to Trial 2. Helpfully, Mr Busuttil provided in his skeleton argument a table which sought to capture the extent of that overlap. Three areas were identified. Firstly, responsibility for publication. This point arises because although Mr Calman was the author of the supplement article, he did not actually write the news article. Nevertheless, the Defendants have accepted that he has caused the publication of that news article. There may be some scope for some interesting questions as to where that dispute will take us ultimately in the proceedings, but I doubt it will have an impact on the overall result of the proceedings. It certainly is not a significant issue, and it is one that on its own would not have deterred me from directing a preliminary issue trial.

The second is more substantial, and it has led to some interesting discussion in submissions this morning. The issue is the overlap between ‘malice’ issues. Malice is pleaded in answer to the qualified privilege defences that have been advanced, but it also arises in the context of the statutory malice under s.3(5) in respect of the honest opinion defence. Those issues will be resolved by determining the state of mind of the Defendants, principally Mr Calman as the author of the supplement article. His state of mind is going to be a matter that is of central importance in Trial 1, if it were ordered, because it is a critical aspect of the public interest defence under s.4. But it also arises in Trial 1, in relation to the malice plea in answer to the qualified privilege defence. Finally, Mr Calman’s state of mind is relevant to the plea of the statutory malice, under s.3(5), in answer to the honest opinion defence.

Mr Busuttil had, in my mind rightly, argued that it would not be satisfactory for Mr Calman to have to give evidence in Trial 1 in relation to the issues relating to his state of mind and then, potentially, face having to come and give evidence in Trial 2 in relation to the issue of honest opinion. I do not understand Ms Evans to have argued that this would be a sensible or desirable outcome. Indeed, she accepted that it would be important that Mr Calman did not give evidence twice, and she also accepted that the Claimants should be permitted to advance their full case against Mr Calman on one occasion.

That opened the way potentially to a solution which Ms Evans had identified in her skeleton argument. She proposed that the honest opinion defence should be tried in Trial 2, but that the issues under s.3(5), relating to Mr Calman’s state of mind, should be tried in Trial 1. Although it is a novel approach, it is not one to which I see any principled objection. It meets the objection that Mr Calman should not be required to give evidence twice potentially, and it meets the objection that the Claimants should not be prevented from deploying their full case on the issues that relate to Mr Calman’s state of mind in Trial 1.

Mr Busuttil considered the position over the luncheon adjournment. It is one that, in fairness, he needed time to consider and reflect upon with his clients. I was perfectly content for that because it is an issue of some significance. After lunch, he confirmed that the Claimants saw no objection in principle with this course.

The final issue that was identified as an overlapping one in relation to the issues that would be involved in Trial 1 and Trial 2 was the issue of the s.4 public interest defence. Mr Busuttil put it this way. The Claimants have raised the issues of whether Mr Calman (a) correctly identified and represented accurately in the articles the claims that the Claimants had in fact made about cholesterol, CBD, and/or statins, and (b) accurately characterised such claim as the Claimants had made as “fake news”. He submits that the same issues will arise in the context of honest opinion under s.3(4)(a) and truth in Trial 2, if it takes place.

In my judgment, however, it is important to be clear about the parameters of, and Court’s approach to, a s.4 defence: see discussion in Hijazi -v- Yaxley-Lennon [2021] EWHC 2008(QB). With a s.4 defence, the issues include identification of the materials that were available to the defendant, the enquiries or investigations that s/he did make, the enquiries he or she could have made, and, critically, the way in which the material that s/he obtained was represented in the publication. Those are all matters which potentially have a bearing ultimately on the question the court has to decide, which is whether or not a belief that the publication was in the public interest was reasonable in all the circumstances.

That is quintessentially an assessment of what was available at the time, or could have been available, and the assessment that is or was made of it by the journalist. Issues of objective truth of that material do not arise under the s.4 defence, so they can and should be kept quite separate from any defence of truth. Equally, the objective matters relied upon in support of an honest opinion defence are quite separate from the subjective matters that arise under s.3(5). I am not persuaded that there is any real overlap of issues. The Court’s task in respect of these defences is very different. I accept that it is likely that in Trial 1, the court will ascertain that which was available to Mr Calman when he was drafting his article, and the court will make decisions as to what he made of that and ultimately will decide whether the s.4 defence is made out.

To the extent the court has therefore captured the material and made findings of what was available to the journalist, that does not trespass impermissibly on the cope of any enquiry that would have to be carried out either for the honest opinion or truth defence. The issues in the ultimate assessment of those are quite different in their respective contexts.

My conclusion is that, with the exception of the s.3(5) issues, which seem to me can be fairly accommodated in and should be resolved at Trial 1, there is no real risk of overlapping issues or unfairness being caused to either party. In my judgment, what I have called the ‘state of mind’ issues can be properly and fairly be tried at Trial 1.

One then comes to consider the benefits and downsides of directing the trial of the preliminary issues on this basis as sought by the Defendants. There is no doubt that a trial of the preliminary issues proposed by the Defendants would be a substantial enterprise. It will involve the resolution of substantial issues of fact. In consequence, it will require disclosure, witness evidence and a trial at which witnesses are cross-examined. There is likely to be substantial cross-examination of the Defendants’ witnesses, particularly in respect of the s.4 defence, but experience tends to show that many s.4 defences are tried in a couple of days. There are likely to be relatively few witnesses. The issues to be resolved would mean that it is very unlikely that the Claimants would have to give evidence. Save in unusual circumstances, in most cases a claimant is unlikely to have any relevant evidence to give in relation to a s.4 defence.

The parties have put in evidence estimates as to the likely costs of the two trials. The Claimants have estimated that the first trial would take eight and a half days and would cost in the region of £565,000. Trial 2 would last for twenty-three and a half days and could cost as much as £1.4 million. The defendants say their first trial would be likely to be resolved in seven days, costing around £360,000 whereas Trial 2 they believe could be conducted or completed within fifteen days at a cost of £875,000. On any view, and it will not come as a surprise given the nature of the issues that arise in these proceedings, this is substantial litigation.

Mr Busuttil makes the point, which I accept, that experience tends to show that splitting trials can lead to an overall increase in costs. It is very difficult to try and make an assessment at this stage as to the likely twists and turns of the litigation. As I expressed in argument, I have a suspicion that if the court directed one single trial of all these issues, that it would be likely to be very resource intensive. That is simply a reflection of the large number of issues that arise and the way in which they interlock. The legal issues raised are complicated. It seems to me to be likely-- I say no more than that-- that the court would have to be actively involved at several stages of the litigation. Each of those issues could produce the risk of an appeal at an interim stage in the proceedings. As I say, it is very difficult to make a reliable assessment of that and how the action would unfold if it were directed to be tried all in one go, but one thing is very clear: it would be a very substantial piece of litigation with many complicated factual and legal matters to be resolved.

The principal benefit that Ms Evans for the Defendants has identified is that, firstly, if the preliminary issues that she seeks are directed, it does potentially offer the possibility that the litigation will be brought to an end. That would be so if the Defendants succeed with their s.4 defence. It is not possible (nor appropriate) at this stage to attempt to assess how likely it is that the s.4 defence will succeed, but I think Ms Evans is entitled to submit that there is a realistic prospect that the preliminary issue Trial 1 would bring an end to the proceedings.

Ms Evans’ second point is that even if it does not itself bring an end to the proceedings, the trial of the preliminary issues she has identified would deliver several identifiable benefits for Trial 2. The first is that the Court will have resolved the issues of meaning and fact or opinion. Experience shows that those are decisions in libel claims the resolution of which are very significant. This is because they are often the issues upon which the parties are divided and unable to resolve themselves by any form of negotiation or alternative dispute resolution. Since it has been possible to resolve meaning and associated issues of fact and opinion early in proceedings, there are many fewer libel actions that go on to a full trial. The experience of the Media and Communications List is that most actions now resolve once the court has determined meaning and fact/opinion. That is not always the case, and there are some examples where the cases do nevertheless go to trial, but those that do are now in a small minority. Those that do proceed, do so with the benefit of key issues having been resolved.

The principal benefits of having resolved meaning and fact/opinion at Trial 1 will be clarity of the issues that remain to be resolved and the case management opportunities thereafter. The interval between Trial 1 and Trial 2 will provide an opportunity for the parties to take stock. It may be possible for them to resolve the remaining issues in dispute. But even if that is not possible, the fact that the court has now set the parameters of meaning and fact or opinion means that the case management of Trial 2 becomes a whole lot easier. Resolution of fact/opinion will clearly set out the extent to which (if at all) the Defendants can rely on defences of honest opinion and truth. At the moment, there is a sharp division between the parties – which is by no means uncommon – as to the natural and ordinary meaning of the articles and whether they make allegations of fact or express opinion. The Defendants do not seek to prove true the meaning complained of by the Claimants. If the Court rules that the articles do bear the Claimants’ meaning and that it is an allegation of fact, two consequences follow. No defence of honest opinion will be available, and the defence of truth – at least in its present form – may not be viable. As the defence of truth is one that is likely to require expert evidence, achieving certainty as to the parameters and viability of any truth defence is particularly important. On the other hand, if, for example, the court decides that the articles contain only defamatory opinion then the truth defence is likely to disappear and with it any need for expert evidence.

For as long as meaning, and the issues of fact and opinion remain live and unresolved, as they would do if the court directed a single trial of all issues, then the proceedings must work on the basis of various hypothetical outcomes. In that eventuality, this case would be like libel trials used to be when we had juries as the ultimate arbiters of fact. That is unsatisfactory because it risks litigation of issues that, in the final analysis, is not relevant. Substantial costs – and Court time – is wasted progressing through the phases of disclosure, witness statements and a trial of a truth defence only for it to be held, following trial, the publication complained of only expressed an opinion. For the reasons expressed in the authorities I have referred to above, there is likely to be a significant benefit in terms of case management and avoiding wasted costs, if the court is able to resolve the issues of meaning and fact or opinion as soon as possible.

One significant downside of having the two trials that I have identified is that there is a very real risk of an appeal that will arise following the determination of the issues at Trial 1. I say that because there has got to be at least a prospect that the Curistan decision, when applied to these proceedings, will mean that here could be two rival meanings: one natural and meaning, arrived at by excluding what the Curistan privilege material, and one in which the Curistan material is taken into account. Curistanis a decision of the Court of Appeal which a first instance Judge would be bound to follow. I have already referred to the fact that Curistan as a decision has received some criticism. There are respectable arguments that the decision profoundly conflicts with a fundamental principle of defamation law; the single natural and ordinary meaning of a publication. That issue alone could lead potentially to an appeal, and there would be several other issues that the parties could well seek to appeal after Trial 1.

The reason why the risk of appeal is such a significant factor against directing substantial preliminary issues is that it may prove, ultimately, to have been a waste of time and costs, if the resolution of the preliminary issue turns out to have been irrelevant to the ultimate outcome. Ordinarily, that is on the basis that the Court has resolved a preliminary issue, that leads to an interim appeal, only for it to turn out, when the full issues in the case in the subsequent trial have been resolved, the claim was decided on another point. That raises the spectre of significant waste of the parties’ costs and the court’s resources, and substantial delay.

Against that, there will probably always be a risk of an appeal when the Court directs trial of any preliminary issue, but rightly, it is one of the factors that looms large when the court is considering whether to direct a preliminary issue trial.

In this case, I think it is fair to say that there is a greater than the average risk of an appeal because of the Curistan point. It is a serious factor that weighs against the granting of the application for the preliminary issues sought for the Defendants. The greatest benefit of trying all the issues together is the finality that that would bring, of course subject to any appeal, but it would mean that the issues would be determined in one go and the parties would know which of those issues had in the end been material and either carried or lost the day. Separating out issues runs the risk that you spend a great deal of time litigating on an issue that if you had tried it with everything else, would not have mattered in the final analysis.

I had started the day today sceptical about the wisdom of ordering a preliminary issue trial principally on the significant risk and downside of an appeal following Trial 1. But I have, nevertheless, been persuaded, in the particular circumstances of this case, that I should direct the preliminary issues sought by the Defendant. This is because I am satisfied that there are significant benefits to be achieved by splitting the trials. There is a realistic prospect the resolution of the issues in Trial 1 could dispose of the whole claim. Even if it does not, resolving the issues in Trial 1 will deliver real benefits (and significant savings in costs and resources) for the management and future conduct of the litigation thereafter. In light of that, I have decided that there is a sufficient upside in ordering the trial of preliminary issues to outweigh what I recognise are significant potential downsides, of which the appeal risk is the greatest. In making this decision, I have considered that the splitting up of the trials may be frustrating for the Claimants. I regret that the complicated nature of this case means that there will inevitably be a period of delay before any Trial 2 can be progressed. For the Claimants, the objective is that the court resolves the issue of meaning and decide whether any defamatory statement made against them can be defended. They seek by these proceedings to vindicate themselves. Against that, I must also properly also consider fairness to the Defendants. They wish to vindicate their journalism and have important rights of freedom of expression. I understand very much those concerns, and it is unfortunate that the complexity of this action and particularly the unusual feature of the Curistan decision has meant that the case management options that would ordinarily be open to the court as a matter of course are not immediately available.

Narrowly, but in the end, I have been convinced, that the court should direct the preliminary issues that have been sought by the Defendants. Trial 1 will also include the resolution of the s.3(5) issues that arise under the honest opinion defence. That is the ruling that I make.