England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM
THE HIGH COURT OF JUSTICE (QUEEN'S BENCH DIVISION)
WARBY J

		Royal Courts of Justice Strand, London, WC2A 2LL
		12/09/2017

Lord Justice Davis:

Introduction

1. This appeal raises a number of issues of law and practice in cases of defamation. All are important to the parties but some have a wider importance, involving a consideration both of the meaning and effect of s.1(1) of the Defamation Act 2013 ("the 2013 Act") and of the practice and procedure to be followed where it is in issue whether a published statement has caused or is likely to cause serious harm to the reputation of the claimant.
2. The matter came before Warby J on a trial of preliminary issues previously directed by orders of Nicola Davies J on 1 April 2015 and Nicol J on 29 June 2015. There was a two day hearing, at which written and oral evidence was presented. By his very full and careful reserved judgment handed down on 30 July 2015 the judge found in favour of the claimant. He ruled, among other things, that various articles published by the defendants by reference to the claimant involved publication of defamatory statements which had caused or were likely to cause serious harm to the claimant: see [2015] EWHC 2242 (QB).
3. The defendants now appeal, by leave granted by Floyd LJ, from that decision. The claimant, while seeking to uphold the judge's decision as to its conclusion, seeks to argue by a Respondent's Notice that the judge could and should have found in his favour by a much shorter route than was adopted, on a proper interpretation and application of s.1(1) of the 2013 Act.
4. Before us, the appellant defendant AOL (UK) Limited appeared by Mr Andrew Caldecott QC, Mr Manuel Barca QC and Miss Hannah Ready. The appellant defendants Independent Print Limited and Evening Standard Limited appeared by Mr David Price QC. The claimant respondent appeared by Miss Adrienne Page QC and Mr Godwin Busuttil. I would pay tribute to the skill and care with which the respective arguments, both written and oral, were advanced.

Background facts

5. The background facts are very fully set out in the judgment under appeal, to which reference can be made. Only a relatively brief summary is needed here.
6. The claimant is a French citizen, born in 1974. By education and employment he acquired very considerable skills in the aerospace and aviation fields. In 2004 he took employment in Dubai with Panasonic Avionics Corporation and has lived in the UAE since then.
7. In 2008 he first met the woman who was to become his wife and who in these proceedings has been called "Afsana". She is a British citizen. A relationship developed and in due course they married, in London, on 26 February 2010. A son, Louis, was born in Dubai (where they lived together) on 4 April 2010.
8. It seems that the marriage very quickly soured. In April 2011 the claimant commenced divorce proceedings in Dubai. At around this time, his employment with Panasonic Avionics Corporation also came to an end. He started looking for a new position, amongst other things using recruitment agencies based in England (there was evidence that England is regarded as an extremely important aviation and aerospace industry hub). In the event, the claimant eventually secured employment in February 2013 as a teacher and instructor at the Military College in Abu Dhabi.
9. The divorce proceedings in the meantime continued in Dubai. They were marked by extreme acrimony. At one stage, it seems that Afsana disappeared with Louis so that the claimant lost all contact with him for a considerable period of time.
10. At the beginning of 2014 the claimant learned through a friend, Mr Macfarlane, of what he was to consider to be a campaign to defame him in the English press and media and which he attributed to Afsana and her eldest son (by a previous marriage). Certain articles were identified. He retained English solicitors, Taylor Hampton, and further articles were in due course identified. It was five such articles which formed the basis of the claims subsequently brought against the defendants.

The articles

11. The first publication was on 20 January 2014 in the Huffington Post (operated by AOL (UK) Limited), the publication being posted online by an individual styling himself as "Director of the Emirates Centre for Human Rights" (with an identified Twitter account). It was entitled: "British Victim of Domestic Abuse Faces Prison in the UAE." It is not necessary to set out in full here all the passages relied on.
12. The article went on to refer to Afsana, naming her as Afsana Lachaux. Among other things it said: "A victim of domestic abuse, Afsana took her baby and bravely left her partner three years ago but has been trapped in Dubai ever since, as her ex-husband has exacted a prolonged campaign of intimidation and harassment against her." It referred to her prospectively appearing in court in Dubai on 21 January 2014 "accused of kidnapping her own child, as she suffers the consequences of the Emirati legal system that affords little protection for victims of domestic violence."
13. The article further went on to say that her "abusive ex-husband" had "used his influence with Emirati authorities to obtain an indefinite travel ban on her and her three year old son". The ex-husband was said to have told her that he would "destroy" her, using the Emirati legal system for that purpose. It was also said: "In October 2013, after more than a year of living in hiding, Afsana's ex-husband snatched their child after finding out where they were living." It was said that "Afsana will appear in court to face unjust charges of kidnapping her own child when it is her ex-husband who should be in court to answer why he has abused this woman and his son in such deplorable ways."
14. There was, and could be, no dispute that this article referred to the claimant. The judge recorded his findings as to the meanings complained of at paragraphs 91 and 92 of his judgment.
15. After subsequent correspondence, this Huffington Post article was removed from its UK website on 29 September 2014. At the same time the Huffington Post published online what has been described (though not so entitled) as an "apology". The apology referred to the post of 20 January 2014 and referred to complaints on behalf of the claimant since received. It then said: "… we accept that the post might fairly be criticised for conveying a one-sided impression of the couple's dispute: it could have been made clearer that Afsana's allegations of domestic abuse were denied by her ex-husband. We are happy to put that right and apologise to him for any embarrassment caused."
16. The second article appeared in the Independent newspaper (published by the defendant Independent Print Limited) on 25 January 2014. It was published both in print and online. That article was subject to a meaning determination by Sir David Eady, sitting as a Judge of the High Court, on 11 March 2015, and from which there was no appeal: [2015] EWHC 620 (QB). The meanings so found are set out in paragraph 93 of the decision of Warby J. Those included (among others) that the claimant had become violent towards Afsana; had himself callously and without justification snatched their son back from his mother's arms; had falsely accused Afsana of kidnapping their son, which, if upheld, could result in her, quite unfairly and wrongly, spending several years in a Dubai jail; was content to use Emirati law, which discriminates against women, to deprive Afsana of custody of and access to their son; was violent, abusive and controlling; and had obtained custody on a false basis and had initiated a prosecution in the UAE founded on a false allegation of abduction.
17. Although there was a dispute as to whether the article had referred to the claimant Warby J ruled that it had.
18. The third article was an abbreviated version of the second article, published in "i", which the judge described as a "sister paper" of the Independent. It was not disputed that it had the same defamatory meanings as found in respect of the second article. The same finding of reference to the claimant was also made by the judge.
19. A further article had been posted online in the Huffington Post on 6 February 2014. Much of it related to an arms deal. But towards its end it referred to Afsana Lachaux by name. It described her moving to Dubai in 2010 "to start a new life with her French husband". It went on: "Sadly, shortly after arriving there, her spouse became violent and abusive. Terrified for her own life, she fled their Dubai apartment with their new baby son, Louis…." Reference to the claimant was conceded. The judge, at paragraph 98, made a finding as to the defamatory meaning to be attributed to that post.
20. The fifth article appeared, in print and online, in the London Evening Standard on 10 February 2014. The article is headed: "Dubai's a small place – he took Louis in an instant." The defamatory meanings in connection with this article had also been determined by Sir David Eady. They are summarised in paragraph 99 of the judgment of Warby J. They are broadly, although not entirely, similar to those found with regard to the article in the Independent. The judge, for like reasons, found reference to the claimant to be established.
21. It was accepted in each case that publication had occurred in England and also (with regard to the Huffington Post, Independent and Evening Standard) online in Dubai. The agreed overall readership figure for the two Huffington Post articles was around 4,800. For the Independent articles, readership figures for the print copies were put between around 154,000 and 232,000 and between 523,000 and 785,000 for i. The Independent article had 5,655 unique visitors online. The Evening Standard readership figures were between 1.67 million and 2.5 million for the print edition, with 1,955 unique visitors online.
22. It might be added by way of postscript that, since the hearing before us, our attention was drawn to the judgment of Mostyn J handed down in the Family Division on 2 March 2017 in proceedings between Afsana and the claimant: [2017] EWHC 385 (Fam). It should be emphasised that none of the defendants in the present cases were parties to those proceedings. In those proceedings, on the evidence adduced before him – and both Afsana and the claimant gave oral evidence – Mostyn J amongst other things rejected Afsana's assertions of domestic violence. He also found that Afsana had herself taken actions, including threatening a Sharia legal action, designed to procure the deportation of the claimant from Dubai. He found that Afsana had herself deliberately removed Louis so that the claimant could not, for a period of 19 months, see his own child (the judge found it "not very surprising" that Afsana was convicted of kidnapping in Dubai on 13 February 2014: indeed the judge indicated that he was more surprised at the leniency of a one month suspended sentence). The judge also rejected Afsana's assertions that the claimant had abducted Louis. He was critical of each of them for accessing the confidential information of the other. It should be noted, however, that leave to appeal has since been granted by Moylan LJ.

The legal context

23. Against that background I turn to the legal context.
24. It was a consequence of the orders as to the preliminary issues to be tried that (in addition to resolving any outstanding issues as to reference and meaning) Warby J had to decide whether, for the purposes of s.1(1) of the 2013 Act, publication of the words complained of in all five articles satisfied the requirements of the statutory provision. In addition, he was required to decide, on the application of the defendant AOL (UK) Limited, whether the claims relating to the Huffington Post articles should be struck out as an abuse of process. As the judge noted at paragraph 9 of his judgment, the main issues of law and fact before him related to the "serious harm" requirement of s.1(1) of the 2013 Act.
25. In my view, it is convenient, before addressing the all-important words of s.1(1) of the 2013 Act, to place the statutory provisions in context: including a brief consideration of the progress of the Bill before it was enacted.
26. The starting point is that the law of defamation is there to protect a person's reputation. As stated by Cave J in Scott v Sampson (1882) 8 QBD 491, to which Miss Page and Mr Busuttil made reference in their written argument: "The law recognises in every man a right to have the estimation in which he stands in the opinion of others unaffected by false statements to his discredit." For this purpose, harm to reputation is capable of occurring on each occasion when a false statement referring to a complainant is communicated to another or others: for it is then that they may be likely to think the worse of the complainant.
27. As I see it, it can be important to distinguish the harm caused to reputation by the publication of falsehoods from the consequences that may flow therefrom. For example, if a written defamatory falsehood is published which has, say, the consequence that a prospective employer withdraws a lucrative job offer to a prospective employee, that may be a consequence of the libel, sounding in additional damages: but it is not coterminous with the harm to the reputation of the individual occasioned by the libel (even though it may be supportive evidence of such reputational harm).
28. In this context, however, and as is borne out by numerous statements in the authorities, there often may in practice be little in the way of available positive evidence to establish the harm to reputation occasioned: see, for example, the observations of Warby J in paragraph 55 of his judgment in Ames v Spamhaus Project Limited [2015] EWHC 127 (QB), [2015] 1 WLR 3409 where he said:

"… but as practitioners in this field are well aware, it is generally impractical for a claimant to seek out witnesses to say that they read the words complained of and thought the worse of the claimant".

Clearly if an identifiable damaging consequence is identified that may be capable of being supportive evidence of harm to reputation: but what if there is none? This sort of consideration surely helps explain why, in cases of libel, it was settled at common law that damage is presumed: the extent of the damage then being left, in the usual way, to assessment at trial if liability on the part of the defendant is established. It is, to my mind, also noticeable that this presumption survived the introduction of the Human Rights Act 1998 and was not displaced by considerations relating to Article 10 of the Convention: see Jameel (Mohammed) v Wall Street Journal Europe Sprl [2006] UKHL 44, [2007] 1 AC 359; Jameel (Yousef) v Dow Jones & Co. Inc. [2005] EWCA Civ 75, [2005] QB 946. In the latter case, Lord Phillips MR, in the course of giving the judgment of the court, said this at paragraph 31:

"There have always been strong pragmatic reasons for proceeding on the premise that a defamatory publication will have caused the victim some damage rather than opening the door to the claimant and the defendant each marshalling witnesses to say that, respectively, they did or did not consider that the article damaged the claimant's reputation."

It was in terms held in that case that the presumption of damage that forms part of the English law of libel is not incompatible with Article 10 of the Convention: see paragraph 41 of the judgment.

29. As far as I can tell, it was really the first instance decision of Tugendhat J in Thornton v Telegraph Media Group Limited [2010] EWHC 1414 (QB), [2011] 1 WLR 1985 which caused there to be a significant development in the law for this purpose.
30. In that case, Tugendhat J, basing himself on the judgment of Neill LJ in Berkoff v Burchill [1996] 4 All ER 1008, listed some of the proposed definitions of the word "defamatory". Tugendhat J was plainly concerned, particularly in the context of Article 10 and proportionality considerations, about the need to exclude trivial claims. He concluded, after considering authorities such as Sim v Stretch [1936] 2 All ER 1237, that there was a requirement for a "threshold of seriousness": see paragraphs 90 and 92. On that basis, and following his review of the authorities, he concluded (at paragraph 96) that the correct formulation was this:

"the publication of which he complains may be defamatory of him because it substantially affects in an adverse manner the attitude of other people towards him or has a tendency to do so."

A threshold of seriousness, phrased in terms of substantiality, was thereby introduced. As to the connection between this approach and the common law presumption of damage Tugendhat J explains it in this way at paragraph 94:

"There is a further point to be noted if my conclusion in paras 90 and 92 is correct. If this is so, then it explains why in libel the law presumes that damage has been suffered by a claimant. If the likelihood of adverse consequences for a claimant is part of the definition of what is defamatory, then the presumption of damage is the logical corollary of what is already included in the definition. And conversely, the fact that in law damage is presumed is itself an argument why an imputation should not be held to be defamatory unless it has a tendency to have adverse effects upon the claimant. It is difficult to justify why there should be a presumption of damage if words can be defamatory while having no likely adverse consequence for the claimant."

Applying the test he had posed at paragraph 96 of his judgment, Tugendhat J went on to give judgment, on the summary judgment application before him, in favour of the defendant: among other things concluding, in that case, that the required threshold of seriousness was not overcome (paragraph 107). This requirement of a threshold of seriousness was subsequently confirmed by the Court of Appeal in Cammish v Hughes [2012] EWCA Civ 1655, [2013] EMLR 13, it being also said, at paragraph 40 of the judgment in that case, that the question of seriousness is "multi-factorial".

31. It is, in my opinion, important in this regard to distinguish between the approach taken in Thornton and the approach taken in Jameel (Yousef) (cited above).
32. In the latter case, the court decided that the defamatory statements which had been made were serious and actionable. However, having decided that the claim had been actionable, the court nevertheless struck it out as an abuse of the process. The claim was struck out because it was no longer serving the purpose of protecting the claimant's reputation: publication within the jurisdiction had, as established by subsequent evidence since obtained, been minimal and did not give rise to a "real and substantial tort". That is different from the case of Thornton. In Thornton the claim was in effect not actionable, as it could not, as found, meet the required test of substantially affecting in an adverse manner the attitude of other people towards the claimant: thus it was not actionably defamatory.
33. These two decisions are of particular importance for present purposes: because they provide important context for the introduction of s.1(1) of the 2013 Act.
34. All parties before us desired us to consider the evolution of the section as it made its way before Parliament. I think, nevertheless, that it would be unwise and inappropriate to attach much, if any, weight to isolated remarks of Ministers culled from the Parliamentary debates. It is, however, interesting to note that the wording in s.1(1) as the Bill progressed evolved from use of the words "substantial harm" – " substantial", it will be recalled, being the word favoured by Tugendhat J – into "serious and substantial harm" and thence, finally, into " serious harm."
35. Perhaps the best extraneous indication of what was contemplated is to be found in paragraphs 10 and 11 of the Explanatory Notes to the 2013 Act (in saying that I well appreciate the formal limitations on the use of Explanatory Notes in interpreting primary legislation). They provide as follows:

"10. Subsection (1) of this section provides that a statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant. The provision extends to situations where publication is likely to cause serious harm in order to cover situations where the harm has not yet occurred at the time the action for defamation is commenced. Subsection (2) indicates that for the purposes of the section, harm to the reputation of a body that trades for profit is not "serious harm" unless it has caused or is likely to cause the body serious financial loss.

11. The section builds on the consideration given by the courts in a series of cases to the question of what is sufficient to establish that a statement is defamatory. A recent example is Thornton v Telegraph Media Group Ltd in which a decision of the House of Lords in Sim v Stretch was identified as authority for the existence of a "threshold of seriousness" in what is defamatory. There is also currently potential for trivial cases to be struck out on the basis that they are an abuse of process because so little is at stake. In Jameel v Dow Jones & Co it was established that there needs to be a real and substantial tort. The section raises the bar for bringing a claim so that only cases involving serious harm to the claimant's reputation can be brought."

36. It can therefore be taken that the broad intention was to "build on" cases such as Thornton and Jameel and to "raise the bar" for bringing a claim in defamation. Ultimately the question arising on this appeal, insofar as it relates to the meaning of s.1(1) of the 2013 Act as enacted, is as to just how far the bar has been so raised.

The 2013 Act

37. Section 1 provides as follows:

"(1) A statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant.

(2) For the purposes of this section, harm to the reputation of a body that trades for profit is not "serious harm" unless it has caused or is likely to cause the body serious financial loss. "

I note that s.1(2) was introduced into the 2013 Act at a very late stage.

38. Section 15 provides definitions of "publish" and "statement" as follows:

"In this Act –

"publish" and "publication", in relation to a statement, have the meaning they have for the purposes of the law of defamation generally;

"statement" means words, pictures, visual images, gestures or any other method of signifying meaning."

By s.16(4) it is provided that nothing in s.1 or s.14 affects any cause of action accrued before the commencement of the section in question. Section 14, in relation to slander, provides that a statement conveying the imputation of a contagious or infectious disease "does not give rise to a cause of action for slander unless the publication causes the person special damage."

39. It is to be noted that the 2013 Act is not designed to codify the law of defamation. In fact, by its short title it is described as an Act to amend the law of defamation. Thus in sections 2, 3 and 4 the common law defences of justification and fair comment and the Reynolds defence are in terms abolished. There are various other very significant changes introduced. But other aspects of the law of defamation self-evidently are not expressed to be abolished or changed.

The meaning of s.1(1)

40. So what, then, is the meaning and effect of s.1(1)?
41. The first point to be made is that the section conspicuously does not purport to offer a definition of what is a defamatory statement. It does not tell you what a defamatory statement is. It tells you what it is not.
42. The second point to be made is that s.1(1) focuses solely on the harm to the reputation of the individual claimant. It does not focus, as s.1(2) does, on whether that harm to reputation has caused or is likely to cause serious financial loss. It does not focus on whether, for example, the published statement has caused injury to feelings. Likewise, in contrast to cases of slander under s.14, s.1(1) imposes no requirement of special damage. Thus the distinction between harm to reputation on the one hand and the consequences of that harm on the other hand is maintained in s.1(1).
43. The third point to be made is that the section is not expressly drafted in terms of actionability. It does not, for example, say "a statement is not actionable as defamatory unless [etc]". It is difficult not to think that such an approach – consistent with the evident approval of Thornton conveyed in the Explanatory Notes – broadly underpins s.1(1): but that is not the actual language used. At all events, one thing is clear: there is an intention to weed out, by means of a threshold of seriousness, trivial claims.
44. The fourth point to consider is what is meant by "serious". It seems to me that that means what it says and requires no further gloss. It can be accepted, however, that it conveys something rather more weighty than "substantial": the word used by Tugendhat J in Thornton.
45. The fifth point to consider is what is meant by the words "or is likely to". On the face of it, these words too are ordinary words of English not requiring any gloss. But on this point matters are, as it seems to me on reflection, not quite so easy as that.
46. It will be recalled that Tugendhat J had, in time honoured language in this field, referred in paragraph 96 of his judgment to a "tendency". But the statutory language "is likely to", which on one view (and as was my own initial view) conveys something rather stronger than that and conveys a meaning of that which is more probable than not. That would also at least be consistent with paragraph 10 of the Explanatory Notes and (if it be admissible, which I doubt) with a statement of the Minister in the course of the Parliamentary debate. Even so, I would, for myself, query if the distinction is here ordinarily going to be very meaningful. If I say of myself as a tennis player that I have a tendency to serve double faults then surely I am in reality saying of myself that I am likely to serve double faults. I have some difficulty in envisaging in practice, for most cases at any rate, any very significant difference between the case of a published statement which has a tendency to cause serious reputational harm and the case of a published statement which is likely to – in the sense of probably will – cause serious reputational harm.
47. That said, what the words "is likely to" mean in a particular statutory context can vary. Certainly there is no settled meaning of "more probable than not" which is applicable in all cases. To take an example – in a context very different from the present – whether the appointment of company administrators (under the then applicable insolvency legislation) was "likely to achieve" one of the identified purposes was held to be satisfied if there was a "real prospect" of one of the identified purposes being achieved; and it was not necessary for the court to conclude that the purpose would more probably than not be achieved: see re Harris Simons Construction Limited [1989] 1 WLR 368. On the other hand, it has been decided, for the purposes of s.12(3) of the Human Rights Act 1998, that whether a claimant was "likely" to establish at trial that publication should not be allowed involved a degree of flexibility in approach, with no single, rigid standard. However, for that purpose the general approach would be that the courts would be very slow to grant interim restraint orders unless the claimant satisfied the court that it would probably ("more likely than not") succeed at trial: see Cream Holdings Limited v Banerjee [2004] UKHL 44, [2005] 1 AC 253.
48. In this context, two things need to be noted. (1) First, in Cream Holdings the House of Lords did not hold that the required approach was simply whether or not the court considered that the claim would probably succeed at trial: the approach was rather more nuanced. As stated at paragraph 20 of the speech of Lord Nicholls: "… "likely" in s.12(3) cannot have been intended to mean "more likely than not" in all situations." (2) Second, in Cream Holdings Lord Nicholls also stated, in paragraph 12 of his speech, that "likely" has several shades of meaning, encompassing varying degrees of likelihood: "its meaning depends upon the context in which it is used."
49. In the present case, Warby J considered that the words "is likely to cause" unequivocally meant more probably than not. He considered that had Parliament intended to maintain the familiar notion of "have a tendency to" it would have said so (paragraph 49 of his judgment). I see the force of that. However, I do not, on consideration, agree. Parliament conspicuously has not said "has caused or will cause". It has used the words "is likely to cause". Parliament must (objectively) be taken to have known of Lord Nicholls' remarks in Cream Holdings at paragraph 12. In the instant case the context is the law of defamation. In that context, likelihood and tendency are words used in effect interchangeably: that is illustrated, for example, by the language of Tugendhat J in the above cited paragraph 94 of Thornton and elsewhere (in paragraph 93, for instance, he had likewise said "a tendency or likelihood is sufficient") and, for example, by the language variously used by Neill LJ in Berkoff v Burchill (cited above). In Andre v Price [2010] EWHC 2572 (QB), Tugendhat J had held that the word "calculated" as used in s.2 of the Defamation Act 1952 meant "likely": accepting that, in that context, that meant something less than "more likely than not".
50. In the circumstances I consider that, to the extent that it may matter, the words "is likely to cause" as used in s.1(1) are to be taken as connoting a tendency to cause.

The decision of the judge and the arguments on s. 1(1)

51. It clearly is convenient to decide first the issue of the meaning and effect of s.1(1) since that could impact on the arguments raised on the substantive appeals. Accordingly, the parties sensibly agreed that Miss Page should address us first on the points raised in the Respondent's Notice.
52. Before the judge, as before us, it was argued on behalf of the claimant that the way in which the bar has been raised by s.1(1) was by way of adopting a threshold of seriousness using the actual language of seriousness (thus on any view hardening up on the test of substantiality proposed by Tugendhat J in Thornton). No other substantive change to the well-established principles of common law was, it was submitted, either foreshadowed or intended. Thus in cases of libel the tort is complete on publication and damage is presumed. Further, the words "is likely to cause" are also wholly apt to cover a threatened publication in circumstances where a quia timet injunction may be sought: whereas the words "has caused" relate to harm caused by actual publication. As to the requirement of "serious harm" that where appropriate can, if in issue, be assessed by the judge at the meaning hearing simply by reference to the words used and, to the extent necessary, having regard to the circumstances in which the publication is made (for example, in a newspaper article). But what is not called for is a lengthy hearing such as occurred in the present case and as has occurred in a number of other recent cases, involving the calling of witnesses, elaborate arguments and hugely costly preliminary hearings lasting some days. Miss Page said that such a radical change in approach and practice could never have been, and was not, contemplated by Parliament and does not reflect the statutory language used.
53. For the defendants, it was contended that, while inferences might sometimes suffice, s.1 (1) nevertheless requires a claimant to prove on the balance of probabilities that he had suffered, or would be likely to suffer, serious reputational harm. That was now a required element of the tort. It was further said that serious reputational harm could be expected to give rise to identifiable adverse consequences: and if no evidence of such consequences was adduced then the court should not be ready to conclude that there was serious reputational harm. Overall, it was said that the claimant's interpretation was too restrictive and failed to give effect to the Parliamentary intention to eliminate claims not meeting the threshold of seriousness.
54. The judge, after assessing the common law context and the language of s.1(1), reached this conclusion at paragraph 45:

"In my judgment this approach leads to the clear conclusion that in enacting s.1(1) Parliament intended to do more than just raise the threshold for defamation from a tendency to cause 'substantial' to 'serious' reputational harm. The intention was that claimants should have to go beyond showing a tendency to harm reputation. It is now necessary to prove as a fact on the balance of probabilities that serious reputational harm has been caused by, or is likely to result in future from, the publication complained of."

55. He went on to say that in his view the claimant's arguments did not sufficiently account for the presence of the words "has caused"; and that s.1(1) had "subsumed all or most of the Jameel jurisdiction into a new and stiffer statutory test requiring consideration of actual harm". He then said this at paragraph 56:

"In arriving at my conclusions I have given careful thought to Ms Page's submissions that they involve imputing to Parliament – contrary to principle – an intention to revolutionise defamation law by implication, and to do so in a way that in practice risks defeating Parliament's intention by making litigation in this area more, not less, expensive and complex. I regard both submissions as alarmist and ill-founded. I acknowledge the presumption against implied repeal of the common law. But I do not accept that my construction of s.1(1) is as radical as is suggested, or that it involves the dramatic consequences attributed to it by the claimant's counsel. To the extent that my construction does involve the implied repeal or amendment of common law principles, an intention to achieve that is, in my judgment, necessarily implicit in Parliament's choice of language."

And he said this at paragraph 60:

"I accept that my construction of s.1(1) means that libel is no longer actionable without proof of damage, and that the legal presumption of damage will cease to play any significant role. These, however, are necessary consequences of what I regard as the natural and ordinary, indeed the obvious meaning of s.1(1). They are, moreover, consequences which had in practice already been brought about by previous developments. The HRA and the emergence of the Jameel jurisdiction which substantially eroded if they did not wholly undermine these common law rules. Since Jameel it has no longer been accurate other than technically to describe libel as actionable without proof of any damage. I cannot see this as a substantial argument against my construction of the statute."

He went on to hold (at paragraph 66) that if a defendant raised an issue that the harm was too slight to justify a claim then it would usually be preferable to try the matter as a preliminary issue (his emphasis).

Disposition on s.1 ground

56. As the arguments wore on before us I became increasingly doubtful that the arguments of the appellants were correct or that the conclusion as to the meaning of s.1(1) reached by the judge on this aspect of the case, as summarised in paragraph 45 of the judgment, was correct. Having further considered the matter, my doubts have only been reinforced. It seems to me that it is the arguments of Miss Page which are in essence correct.
57. The first point is that the interpretation as reached by the judge potentially does involve a substantial change in the law of defamation which was at no stage obviously flagged up in the preceding Parliamentary debates or Explanatory Notes: nor is it evident from the scheme and structure of the 2013 Act itself. As Miss Page neatly put it, and as Mr Price accepted, the judge's approach involves not so much a case of raising the bar: rather it is a case of erecting a further hurdle.
58. A point that particularly troubles me, on the judge's interpretation, is that it effectively removes the presumption of damage which heretofore had always been a concomitant of the tort in cases of libel. But, for the reasons given by Tugendhat J at paragraph 94 of Thornton, the existence of the presumption of damage is compatible with a raised threshold (whether it be "substantial" harm or "serious" harm). The actual language of s.1(1) does not compel a conclusion that the presumption of damage is intended to be abolished: and elsewhere the 2013 Act makes it specific where an aspect of the common law is intended to be abolished.
59. As I see it, the presumption of damage in libel cases (itself no doubt founded on policy grounds as much as on empirical grounds) fits with the notion that what ordinarily causes the reputational harm is precisely the fact of the publication to others. It is at that stage that the harm to reputation will have occurred: even if there may also subsequently be (although not necessarily so) consequential damage.
60. On the defendants' and the judge's approach, as Mr Price and Mr Caldecott accepted and as the judge himself accepted (at paragraph 68 of his judgment), a claimant can drift in and out, as it were, from having an available cause of action at any given moment of time. I appreciate that over the decades defamation cases seem to have acquired their own paraphernalia. Even so I can see no convincing reason why libel cases brought by individuals generally should now (as a result of the 2013 Act) be designed to carry with them their own status of creating some kind of ambient cause of action, drifting in and out of actionability: in contrast with other torts.
61. Moreover, on the defendants' argument, what then is the position on limitation? On Miss Page's approach – which reflects the common law approach – there is no difficulty: it is the date of publication. That is also wholly consistent with the wording of s.8 of the 2013 Act introducing new provisions as to single publication. For by s.8(3) it is provided that for the purposes of s.4A of the Limitation Act 1980 (time limits for actions for defamation) "any cause of action… is to be treated as having accrued on the date of first publication".
62. It is striking that the submissions of Mr Price and Mr Caldecott – two immensely experienced practitioners in this field – differed radically at this stage: although in fairness to them that may reflect the conceptual impenetrability of s.1(1) as drafted. Mr Price, consistently in logic with his argument, maintained that the cause of action only arose when the serious harm was caused (or was likely to be caused). He asserted that that would usually give rise to no problem; and in any case, where that might give rise to difficulties there was always available the court's discretion in an appropriate case to extend the time limits.
63. With all respect, this will not do. Not only would that give rise to potential expense and uncertainty for claimants and their advisers (would the court exercise its discretion?) – expense and uncertainty Parliament is not likely to have intended – even more fundamentally it would mean that s.1(1) has, sub silentio, swept away another well-established common law principle in relation to defamation. That principle is that in defamation the cause of action is complete when the defamatory statement is published to a person by whom it is read or heard: see Grappelli v Derek Block (Holdings) Limited [1981] 1 WLR 822. At p. 825 B-C, for instance, Lord Denning MR said this:

"Upon this point we heard interesting discussion on both sides. I would go by the principle, which is well-established, that in defamation – be it libel or slander – the cause of action is the publication of defamatory words of and concerning the plaintiff. The cause of action arises when those words are published to the person by whom they are read or heard. The cause of action arises then: and not later."

At p. 831 C-D Dunn LJ said this:

"Like Lord Denning M.R., I would prefer to deal with this on principle. I agree that a publication is an essential part of the cause of action; that once there is publication the cause of action is complete, and there is no room for the doctrine that the cause of action can, so to speak, be allowed to be inchoate or lie dormant until such time as some fact emerges which would transform an otherwise innocent statement into a defamatory one…."

64. Mr Caldecott, for his part, was alive to this difficulty. He expressly disclaimed the argument of Mr Price that the principle laid down in Grappelli had been abolished by s.1(1). He also accepted for this purpose that the presumption of damage has not been abolished. He very fairly conceded that any date other than the date of publication would be too uncertain for limitation purposes. But these concessions cannot really fit with the other aspects of his case. A cause of action can only accrue if the words in question are both defamatory and published. If the cause of action accrues for limitation purposes on the date of publication then surely it is not possible to maintain at the same time that the cause of action otherwise may accrue at a later date when serious reputational harm is suffered.
65. We were referred, as had been the judge, to the decision of Bean J in Cooke v MGN Limited [2014] EWHC 2831 (QB), [2015] 1 WLR 895. That case also involved, as a preliminary issue, consideration of whether, by reference to s.1(1) of the 2013 Act, serious harm had been or was likely to be caused. In the course of his judgment Bean J pointed out – in my view unexceptionably – that evidence was not required in every case to satisfy the "serious harm" test. Indeed, as had been understandably emphasised on behalf of the claimant in that case, there is, in contrast with, say, financial damage or physical damage, no generally accepted way of ascertaining the extent of actual or likely reputational damage. Bean J said this at paragraph 43:

"I do not accept that in every case evidence will be required to satisfy the serious harm test. Some statements are so obviously likely to cause serious harm to a person's reputation that this likelihood can be inferred. If a national newspaper with a large circulation wrongly accuses someone of being a terrorist or a paedophile, then in either case (putting to one side for the moment the question of a prompt and prominent apology) the likelihood of serious harm to reputation is plain, even if the individual's family and friends know the allegation to be untrue. In such a case the matter would be taken no further by requiring the claimant to incur the expense of commissioning an opinion poll survey, or to produce a selection of comments from the blogosphere which might in any event be unrepresentative of the population of "right thinking people" generally. …"

66. I certainly would not disagree, any more than did Warby J, with this (although I would not agree at all with any further suggestion that such an inference of serious reputational harm can only be drawn in the rather extreme examples there given: nor did Bean J so state). But where, with respect, I would disagree with Bean J is with what he had earlier said at paragraph 31 of his judgment. He there said this:

"The words "has caused" involve looking backwards in time, the words "or is likely to cause" involve looking forwards. The Act does not make clear the moment which marks the dividing line between past and future. It cannot be the moment of publication, since at that moment no harm "has been caused". The two logical possibilities seem to be the date of issue of the claim and the date of the trial (or of the trial of the preliminary issue of serious harm). Either of these has the curious effect that whether a statement is held to have been defamatory on the day it was published might depend retrospectively on the timing of the issue of proceedings or the timing of the trial."

He went on to conclude that the relevant date was the date of issue.

67. I cannot agree with this for these reasons. First, selection of the relevant date as the date on which the claim is issued seems to me to be entirely arbitrary. It will, to an extent, be happenstance as to when a claim is issued. Second, I cannot in any event accept the proposition that the moment cannot be the moment of publication since at that moment no harm has been caused. On the contrary, and in agreement with Miss Page's submission, I think that ordinarily that is precisely the moment harm will have been caused. How, in terms of damages, the reputational harm is then to be valued and quantified – including also any consequential damage – is then left to assessment at trial in the usual way: using, to the extent appropriate in the quantification of damages at trial, application of the principles of Bwllfa Collieries etc. Limited v Pontypridd Waterworks Co [1903] AC 426.
68. Whilst I can agree with Warby J that, by reason of s.1(1), an individual claimant if he is to succeed is required to show that the published statement has caused or is likely to cause serious (reputational) harm I am not able to agree with him as to the procedural and other consequences which should follow. First, for the reasons given above, I do not accept that the words "is likely to cause" necessarily require it to be proved that it is more probable than not that serious harm will be caused. Second, and perhaps more importantly, I do not agree that the statutory imposition of a threshold of seriousness usually (where serious harm is in issue) will require a substantive "threshold" hearing, in advance of trial, on the issue of serious harm.
69. Where meaning is to be determined (cf. CPR 53 PD para. 4) the generally established approach, as I understand it, is that it is done by selecting a single meaning, by reference to the words used, ascertained without the admission of extrinsic evidence (leaving aside innuendo cases) and on an objective basis: see, for example, Jeynes v News Magazine [2008] EWCA Civ 130. If the meaning so established does not convey a serious defamatory imputation then the claim may, by reason of s.1(1), be vulnerable to being struck out without more ado.
70. If, on the other hand, the meaning so established conveys a serious defamatory imputation – and that does not require cases of the extreme kind referred to in paragraph 43 of Cooke – then an inference of serious reputational harm ordinarily can and should be drawn accordingly. The defendant may seek to rebut or challenge the drawing of such an inference: but that is a different point and may well then, if facts are in issue, be a point suitable for trial. Moreover the drawing of such an inference in such a case (viz. where the words used are assessed to be seriously defamatory) also to an extent accommodates the long acknowledged difficulties for claimants in adducing tangible evidence to support an assertion of harm to reputation (an intangible matter). In this context, I also would reject possible suggestions in the defendants' arguments to the effect that there are no boundaries to the evidence that may be adduced at such a threshold hearing and that evidence of reactions of individual readers, Twitter users etc. is necessarily called for.
71. I should nevertheless observe that I thought that at some stages in the argument before us there was on occasion imprecise conflation between what is a presumption and what is an inference.
72. A presumption, whether rebuttable or irrebuttable, arises before and irrespective of consideration of the evidence. An inference arises after and in consequence of consideration of the evidence. Thus at law, in cases of libel (and some cases of slander) there is a presumption of damage: which presumption has in my view, as will be gathered, not of itself been displaced by the 2013 Act. But there is no presumption, at law, of serious damage in a libel case. Accordingly that, under s.1(1), has to be proved. The point nevertheless remains that serious reputational harm is capable of being proved by a process of inference from the seriousness of the defamatory meaning. Where I part company with Mr Price's submissions is with his apparent suggestion that in libel cases an inference is somehow an inferior and unsatisfactory evidential tool and with his insistence that one should ordinarily, in terms of assessing serious reputational harm, be looking for direct, tangible evidence. I do not agree with the downplaying of an entitlement to draw inferences in an appropriate case. Indeed in criminal trials, in circumstantial evidence cases, juries are routinely instructed as to their entitlement to draw inferences (and, moreover, to the criminal standard) if the evidence so justifies. In my opinion, there is no reason in libel cases for precluding or restricting the drawing of an inference of serious reputational harm derived from an (objective) appraisal of the seriousness of the imputation to be gathered from the words used.
73. As I see it, therefore, if an issue has been raised as to whether serious reputational harm has been caused or is likely and if it is not considered appropriate for that issue to be left to be resolved at trial then it may be that it conveniently can be dealt with at a meaning hearing. The seriousness of the reputational harm is then evaluated having regard to the seriousness of the imputation conveyed by the words used: coupled, where necessary or appropriate, with the context in which the words are used (for example, in a newspaper article or widely accessed blog).
74. It is, I think, possible to discern this as the preferred approach of Judge Moloney QC (sitting as a Judge of the High Court) in the case of Theedom v Nourish Trading [2015] EWHC 3769 (QB). In that case the judge was dealing with preliminary issues both as to the defamatory meaning of the words used and as to whether publication of those words had caused or was likely to cause serious harm pursuant to s.1 (1) of the 2013 Act.
75. Judge Moloney QC loyally sought to follow the approach taken by Bean J in Cooke and by Warby J in the instant case. But amongst other things he said this at paragraph 15:

"(e) Depending on the circumstances of the case, the claimant may be able to satisfy s.1 without calling any evidence, by relying on the inferences of serious harm to reputation properly to be drawn from the level of the defamatory meaning of the words and the nature and extent of their publication.

….

(h) It is important to bear in mind that s.1 is essentially a threshold requirement, intended by Parliament to weed out those undeserving libel claims otherwise technically viable, but which do not involve actual serious harm to reputation or likely serious harm to reputation in the future. Once that threshold has been passed, no useful purpose is served at this early stage of the proceedings by going on to consider evidence which is really material only to the quantum of damage if liability is proved."

76. The judge went on, rather pointedly, to observe in Theedom that the body of evidence, which included oral evidence, adduced before him in terms of whether the case crossed the threshold of serious harm "neither adds nor subtracts very much from the inference one would normally draw from the fact of publication in a case of this kind." He further (rightly) pointed out that under s.1(1) pecuniary loss is not a requirement for an individual claimant. He expressed concern that the present case demonstrated a "further escalation" in the conduct of such hearings, at huge cost. He said this, among other things, at paragraph 37:

"(c) In the result, the hearing of evidence has added little or nothing to the conclusions that an experienced defamation judge would have drawn simply from reading the email and considering the agreed distribution list.

(d) The reason for this is that s.1 sets a threshold test; and the threshold is simply that there shall have been serious harm to reputation. Once that level is passed, further evidence goes to quantum only. Throughout this trial, my sense has been that that distinction was in danger of becoming blurred or lost sight of.

(e) Assuming this action now goes to a final trial, there is a likelihood that there will be a wasteful duplication of evidence and cross-examination already carried out before me and/or that the ultimate trial judge will be vexed with submissions about what has or has not been determined in the course of this phase of the trial."

He concluded by expressing the need for caution in directing preliminary issues under s.1 of the 2013 Act.

77. Judge Moloney QC (as of course is Warby J) is highly experienced in defamation cases. I am not. But, with respect, all these observations of Judge Moloney QC seem to me to make every kind of good sense; and I would endorse them. Such considerations also help prevent a proliferation of complex pre-trial hearings, ostensibly designed to resolve major contested issues. They avoid cost. They avoid potential duplication of evidence at any subsequent trial. They also discourage well-resourced defendants from seeking to batter into submission less well-resourced claimants by use of interlocutory process (I am talking generally, I stress, not by reference to this case). The approach suggested by him would, overall, tend to further the overriding objective: which should be applied to defamation cases just as much as to any other High Court civil litigation.
78. I do not accept that such a conclusion and approach fails to give effect to the wording of s.1(1). I do not accept that such a conclusion and approach would leave defendants at the mercy of trivial claims: precisely the mischief at which s.1 of the 2013 Act was directed. I do not accept that such a conclusion and approach in effect would mean that the labours of Parliament have produced but a mouse. On the contrary, in my judgment Parliament has in effect given statutory status to the decision in Thornton whilst at the same time raising the threshold from one of substantiality to one of seriousness. In my judgment, that is both the extent of and limit to the change in the law made by s.1(1). I also add that the very existence of the section should of itself operate to deter the issuing of trifling and unmeritorious claims in the first place. Even if it does not do so in any given case then the remedy, by reference to s.1(1), is still there with regard to trifling claims.
79. Whether in any given case the imputation is of sufficient gravity as of itself to connote serious reputational harm (quite apart from the question of consequential or special damage) should therefore normally be capable – where the question of serious harm is in issue and is not appropriately to be left to trial – of being relatively speedily assessed at the meaning hearing. If it is, nevertheless, desired by a defendant to put in evidence at an interlocutory stage designed to show that there is no viable claim of serious harm the summary judgment procedure under CPR Part 24 is available if the circumstances so justify. There may, for instance, be cases where the evidence shows that no serious reputational harm has been caused or is likely for reasons unrelated to the meaning conveyed by the defamatory statement complained of. One example could, for instance, perhaps be where the defendant considers that he has irrefutable evidence that the number of publishees was very limited, that there has been no grapevine percolation and that there is firm evidence that no-one thought any the less of the claimant by reason of the publication. Whether such evidence is in truth unanswerable and whether such matters are best resolved on a summary judgment application or best left to trial is then for the court to determine. Alternatively, if subsequent events or evidence show that there has ceased to be a "real and substantial tort" then a strike out application, in accordance with the principles of Jameel, may also be available. At all events, the Jameel procedure, with all respect to the judge who thought otherwise (see paragraph 50 of his judgment), in my view has not been wholly subsumed into s.1 of the 2013 Act, even if there is now a potential degree of overlap.
80. All this is salutary, as I see it, for another reason. If the imputation of the words used is serious, carrying with it the inference that serious reputational harm has been or is likely to be caused, then it is, in my view, not right for a claimant then to have to carry a further burden, at an interlocutory stage, of adducing further evidence to prove serious harm at a preliminary issue hearing. It is surely fairer, once such a case has been properly pleaded and the defamatory meaning is sufficiently grave for an inference of serious harm to be drawn, that it is then for a defendant to seek to show why the claim nevertheless should not be permitted to proceed to trial: whether by making an application under CPR Pt 24 or by a Jameel application. Otherwise, if the facts are contentious the case should be left to go to trial in the usual way – just as in any other tort case.
81. Consequently, with all respect to Warby J and with all respect to those directing the preliminary issues in this case, I think that an unnecessarily elaborate procedure was adopted in the present case. Indeed the hearing before Warby J lasted two very full days (with a resulting judgment of some 190 paragraphs), quite apart from the time needed for preparation and the need for detailed written arguments. The costs were very great. The claimant was himself required by AOL (UK) Limited to attend and give oral evidence. When the claimant's lawyers sensibly and pragmatically indicated on grounds of proportionality that they did not require Mrs Lachaux to attend for cross-examination (how long would the hearing have taken if she had?), they were then met with the cool assertion on the part of the defendants that they were bound by what she had said. And so on. None of this served to advance the overriding objective. I thus consider that where a claimant has advanced a sufficient case on serious reputational harm, by reference to the seriousness of the imputation conveyed by the words used, then ordinarily the case should be left to go to trial: where there can then be finally decided the extent to which there was serious reputational harm and, if it is so established, what the resultant damages – including also recoverable damages for consequential loss (if any) – should be.
82. Drawing the threads together, the position therefore is, as I see it, this:

(1) Section 1(1) of the 2013 Act has the effect of giving statutory status to Thornton, albeit also raising the threshold from one of substantiality to one of seriousness: no less, no more but equally no more, no less. Thornton has thus itself been superseded by statute.

(2) The common law presumption as to damage in cases of libel, the common law principle that the cause of action accrues on the date of publication, the established position as to limitation and the common law objective single meaning rule are all unaffected by s.1 (1).

(3) If there is an issue as to meaning (or any related issue as to reference) that can be resolved at a meaning hearing, applying the usual objective approach in the usual way. If there is a further issue as to serious harm, then there may be cases where such issue can also appropriately be dealt with at the meaning hearing. If the meaning so assessed is evaluated as seriously defamatory it will ordinarily then be proper to draw an inference of serious reputational harm. Once that threshold is reached further evidence will then be likely to be more relevant to quantum and any continuing dispute should ordinarily be left to trial.

(4) Courts should ordinarily be slow to direct a preliminary issue, involving substantial evidence, on a dispute as to whether serious reputational harm has been caused or is likely to be caused by the published statement.

(5) A defendant disputing the existence of serious harm may in an appropriate case, if the circumstances so warrant, issue a Part 24 summary judgment application or issue a Jameel application: the Jameel jurisdiction continuing to be available after the 2013 Act as before (albeit in reality likely only relatively rarely to be appropriately used).

(6) All interlocutory process in such cases should be sought to be managed in a way that is proportionate and cost-effective and actively promotes the overriding objective.

(7) Finally, it may be that in some respects the position with regard to bodies trading for profit, under s.1(2), will be different. I say nothing about that subsection which clearly is designed to operate in a way rather different from s.1(1).

The defendants' appeal

83. Against those considerations, I can take the challenges by the defendants to the judge's actual conclusion (which was in favour of the claimant) relatively shortly.
84. I am in no doubt that, whatever reservations I may have about aspects of his approach to s.1(1), the judge reached a conclusion as to the overall outcome which he was fully justified in reaching.
85. It seems to me that, adopting the approach I have sought to outline above and focusing on the seriousness of the defamatory meanings as found by Sir David Eady and by the judge, the gravity of the imputations derived from the published statements is obvious: and a clear inference is to be drawn that serious harm to the reputation of the claimant has been caused. That seems to me to be clear-cut. Indeed this seems to me to be in fact a case of the kind referred to by Bean J in Cooke at paragraph 43 of his judgment.
86. Just consider the position. The claimant has, among other things, variously been accused of domestic violence and abuse: that connotes criminal acts of assault. He has been accused of child abduction: a criminal act. He has been accused of fabricating false allegations against Afsana with a view to having her imprisoned: a criminal act of attempting to pervert the course of justice. He has been accused of manipulating the Emirate Sharia system so as to discriminate against Afsana and unjustifiably to deprive her of access to her son. I need not go on. It is plain that an inference of serious reputational harm arises. The claimant's pleaded case that the words used were "very seriously defamatory" and had caused or were likely to cause serious harm was entirely justified. The judge's findings that the defamatory meanings conveyed were serious and that serious reputational harm had been caused (save for the second Huffington Post article: as to which finding no challenge by way of cross-appeal has been made) were thus themselves entirely justified.
87. When one adds to that if and to the extent necessary (1) that the defendants' publications have significant numbers of readers, whether in print or online, and that the defendants are of course to be taken as influential and reputable publishers; and (2) that the claimant has in the UK (as found) a standing and reputation both personally and in the aeronautics/aviation field then that conclusion is only confirmed.
88. Against those observations, I turn to the specific grounds of appeal of the defendants.
89. First, it was said that the judge, in assessing whether there was "serious" harm, failed to have regard to the fact that the allegations of Afsana (and other allegations with regard to the claimant) had been made in other media publications. In my view, this submission does not get off the ground, given the inference of serious harm that arises from each publication by each of these defendants of these seriously defamatory statements.
90. In any event, this approach is contrary to that laid down in Dingle v Associated Newspapers Ltd [1964] AC 371, which must be considered to remain good law and not to have been abolished by the 2013 Act. (To the extent that it was again suggested to us in this context that matters might be viewed differently since the passing of the Human Rights Act 1998 then it is sufficient to say that the prominence nowadays to be given to Article 10 is matched by the prominence nowadays to be given to Article 8 and to the fact that an individual's reputation is an important part of his social identity.) The general rule established in Dingle is (subject to the provisions of s. 12 of the Defamation Act 1952, where applicable) that the publication by other persons on other occasions of substantially the same libel is of no relevance on the matter of general damages. The rationale for that perhaps finds its pithiest expression in what was said by Devlin LJ in the Court of Appeal in Dingle [1961] 2 QB 162 at p.189 where he said this:

"If a man reads four newspapers at breakfast and reads substantially the same libel in each, liability does not depend on which paper he opens first. Perhaps one newspaper influences him more than another but unless he can say he disregarded one altogether, then each is a substantial cause of the damage done to the plaintiff in his eyes."

91. As Lord Denning put it in the House of Lords at p.411:

"It does a newspaper no good to say that other newspapers did the same. They must answer for the effect of their own circulations without reference to others…."

That is all consistent with the general principle that each libel, as published, is actionable as causing distinct damage to reputation. Moreover, in the present case the various published articles will doubtless in any event have had substantially different readerships.

92. In my view Warby J was unquestionably right in his lengthy treatment of this point at paragraphs 74-95 of his judgment. I do not propose myself to engage in lengthy further elaboration. As will rather be gathered from what I have previously said, I think it very unfortunate that a point of this kind was ever pursued at the hearing of the preliminary issues.
93. Then it is said by Mr Price that the judge erred in his findings as to reference to the claimant. But the test is an objective one and an inference as to reference could clearly be drawn. It was unnecessary and inappropriate for the claimant to be required to adduce further specific evidence from readers in this regard. The judge's conclusions on reference are unassailable.
94. Complaint was made of the judge's assessment of the evidence which was in fact adduced by the claimant (such as that of Mr Macfarlane) as to the articles that had been read, as to their identification of the claimant as the subject of such articles, as to the existence of reputation in England and so on. In the result the judge made detailed findings, in my view properly open to him, as to the claimant's degree of connection with England; and having made such findings he also said this at paragraph 138:

"… I do not think I would be justified in accepting the defendants' submission that the absence of 'tangible' evidence of adverse responses to publication indicates that the true position lies towards the 'Macfarlane' end of the spectrum, or that those who did read the offending words were unaffected by it. Mr Macfarlane is evidently a good friend of the claimant, who was confident enough to trust him over the publishers on a matter of this kind. Only two other individuals have been named as having read some of the words complained of, and there is no evidence of their reaction to whatever it was they read. But it is only human nature for people with less close relationships whose opinion of a defamed individual has been affected, to shy away from raising the matter with that individual. Sometimes there is an outward display of hostility, or an overt shunning or avoidance of a person. But evidence of that kind has always been rare, for obvious reasons. The advent of social media has notoriously increased public online denunciation by strangers, but there is no evidence that it is common for friends or acquaintances to do this. My conclusion is therefore that there were, on the balance of probabilities, tens of people and possibly more than 100 who know or know of the claimant and read one or more of the articles and identified him, and who thought the worse of him as a result."

Those findings were open to the judge. Also entirely in point, in my view, were his important observations at paragraph 140 where he said this:

"What I think of rather greater significance is that all this discussion tends to leave out of account, as if it was unimportant, the impact of publication on the claimant's reputation in the eyes of people who do not already know the claimant. A person can after all be defamed, and seriously defamed, in the eyes of those who do not know him. He does not need to establish an existing reputation in order to complain, and may be entitled to substantial damages for the harm to his reputation caused by publication to people who have never heard of him. This was acknowledged in Jameel. The matter is discussed in Ames at [41]-[42]."

I can see nothing whatsoever objectionable in any of this. It is, indeed, both relevant and correct in principle.

95. For his part, Mr Caldecott sought to attack the judge's conclusions as to the first Huffington Post article, in particular when the judge said this at paragraph 145:

"The principal grounds for concluding that the publication caused serious harm are (a) that publication on the agreed scale is inherently likely to have reached a significant number of people – by which I mean at the very least a dozen – who know the claimant or know of him, whose opinion of him is likely to have been seriously affected in an adverse way; and (b) that the probability is that the claimant's reputation has been seriously harmed in the eyes of others, whose opinion of him matters. Mr Macfarlane is clearly a person whom this article did reach, who knew who it was about, but whose opinion of the claimant was not so affected. But he was a close friend. For the reasons given above I do not agree that the absence of visible or tangible evidence of adverse reactions from other identifiable publishees undermines my conclusions. As the claimant said in his oral evidence, though in different words, silence is not evidence of the absence of impact."

Here too it is, in my view, sufficient to say that I consider that the judge was perfectly entitled to reach these conclusions.

96. By a supplemental ground of appeal, Mr Caldecott further sought to say that the judge had failed to have sufficient regard to the delay – over 8 months – in the claimant notifying his complaint after the first Huffington Post article had appeared: it being stressed that when a complaint was so notified the posting was then withdrawn and the apology was then published. It was submitted that the claimant had thereby failed to mitigate his damage; and it was said that had only a prompt complaint been made no serious harm would have arisen (particularly in the light of what was said to be the relatively limited readership of the first Huffington Post article).
97. This point was barely debated before the judge, although he alluded to it briefly in his judgment and rejected it. He was right to do so. The reality was, as a matter of inference, that the harm will have been primarily caused on the initial publication: it did not necessarily become the more serious, in terms of harm to reputation, solely because it was not swiftly retracted. There are, to my mind, considerable difficulties generally with the notion that an alleged failure to mitigate can somehow extinguish a cause of action. But be that as it may, more specifically the fact is in any event that this purported "apology" was most unimpressive. I agree with the judge when he said that it did not serve significantly to reduce the reputational harm done by the publication. It did not, for instance, purport to constitute a retraction: it in effect simply said that it "could have been made clearer" that the allegations of domestic abuse were denied by the ex-husband. Nor did the apology even touch on the allegations that the claimant had falsely accused Afsana of child abduction or that he had himself unjustifiably snatched the child back. Yet further, quite how quickly the posting would have been removed had earlier complaint been made may be doubtful; and, moreover, the claimant was entitled to at least a reasonable amount of time to instruct English solicitors and investigate the matter. I need not say more. This point is not sustainable.
98. Equally unsustainable is the further supplemental ground of appeal which Mr Caldecott faintly raised. This was to the effect that by reason of s.12 of the Defamation Act 1952 any damages which might be awarded against AOL (UK) Limited would be "negligible". But that section at this stage – rather as with the Dingle point – gives no answer, of itself, as to the extent of the harm occasioned by the publication or as to what an appropriate award of damages might be in any particular case. In truth, the assessment of damages taking into account s.12 would depend on the assessment at the end of the case. The present contention of AOL (UK) Limited is in reality bare assertion. It cannot realistically be maintained at this stage that the claimant has no prospect of recovering substantial damages against AOL (UK) Limited.
99. I did not understand the appellants to press various other grounds of appeal as formulated in writing. For example, the rather remarkable assertion by AOL (UK) Limited that vindication was not "uppermost" in the claimant's mind cannot really be relevant on the s.1(1) argument; and, though in principle it is capable of being relevant on a Jameel abuse application, it was properly, indeed almost inevitably, rejected on the facts by the judge.
100. The further suggestion that the judge failed properly to evaluate the written evidence of Afsana (who was not cross-examined) was also misplaced. In fact the submission was even made that Afsana's credibility and credit was not challenged: given the entire background, that is astonishing. The reality was that it was on pragmatic and proportionate grounds that Afsana was not required to attend the preliminary issue hearing for cross-examination. To the extent that AOL (UK) Limited chose to cross-examine the claimant on the strength of her evidence that did not, as the judge found, prosper. In any event, her evidence was for the most part peripheral on the issues relevant to the hearing. There is nothing in this point.
101. Challenge was also made by Mr Price in his written grounds to the judge's decision on costs made consequent upon the main hearing. To the extent that was pursued before us it fails. This was pre-eminently a matter for the discretion of the judge. Moreover, to the extent that the challenge was in fact based on the time spent in argument before the judge on s.1(1) of the 2013 Act the outcome of this appeal shows that the stance of the claimant has in truth been substantially vindicated.

Conclusion

102. I would accept the principal argument advanced by the claimant in the Respondent's Notice. I would also and in any event reject all the grounds of appeal variously advanced by the defendants. The judge was, in terms of the outcome even if not in all respects in terms of his approach, correct to rule in favour of the claimant on the preliminary issue by reference to s.1(1) of the 2013 Act. The judge was also correct to reject as he did the Jameel abuse contention which had been made.
103. I would therefore, for my part, dismiss this appeal.

Lady Justice Sharp:

104. I agree.

Lord Justice McFarlane:

105. I also agree.