The public interest defence in section 4 of the Defamation Act 2013
The Court of Appeal was wrong to substitute its determination of the two elements of the s.4 defence for that of Jay J (“the Judge”).
In relation to s.4(1)(a) – “on a matter of public interest”:
These words ought to be given the same interpretation as the same words in the former common law defence of fair comment. (This has been replaced by the statutory defence of honest opinion in s.3 which does not have a public interest requirement). This element of the fair comment defence was always interpreted extremely widely. For example, it included the conduct of those who supply goods and services, even in the absence of any public interest in those to whom they are supplied (see, for example, Spiller v Joseph  AC 852 at ).
The Court of Appeal relied on irrelevant matters and/or conflated matters that were potentially relevant to s.4(1)(b):
At  it stated: “The CJEU most often define "public interest" material as that which contributes to a debate of general interest”. In fact, this concept of the ECHR, not the CJEU, of a contribution to a debate is merely a relevant factor as regards the ultimate question of whether the publisher’s Article 10 right should prevail over the claimant’s Article 8 right, primarily in privacy cases.1 It does not purport to be a definition of “public interest material”, let alone the narrow issue in s.4(1)(a) relating to solely subject matter. Further, the defences of fair comment and Reynolds applied to information that was of legitimate interest to a section of the public, even if not of general interest.2 There is no reason for s.4(1)(a) to be construed more narrowly.
The Court’s analysis at  to , purporting to address s.4(1)(a), was wrongly headed: “Judge wrong to find that the statements complained of were in the public interest” [emphasis added]. The issue is simply whether
1 See the well-known passage in von Hannover v Germany (No 1) (2005) 40 EHRR 1 at .
2 Spiller is an example. There is no general public interest in the reliability of a motown tribute band.
the subject matter is a matter of public interest. This conflation with s.4(1)(b) is evident in , the references to a contribution to a debate in ,  and ,  (the failure to contact) and most glaringly at  (the effect on the Claimant’s reputation and “whether the public needed to know the information at the time the article was published”).
The Court of Appeal was wrong at  to draw a distinction, for the purposes of s.4(1)(a), between “the Claimant's personal life, mores and conduct as a contractor, supplier and volunteer” to a charity and a community organisation and how they “were run as charities”.
In relation to s.4(1)(b) – “the defendant reasonably believed that publishing the statement complained of was in the public interest”:
The Court of Appeal adopted a narrow and inflexible approach, inconsistent with the liberalising concepts of “reasonable belief” and “in the public interest” (addressed further in [3.2.2],  and  below). In this context the Court made fundamental errors of principle:
The defence is not a form of qualified privilege (as stated in ), thereby importing the restrictive “moral duty” test. For example, qualified privilege is not available to a defendant who reasonably, but mistakenly, believes that there is a duty to communicate.3
It is not “confined to the circumstances necessary to protect Article 10 rights” (as stated in ). No authority is cited for this proposition. It is inconsistent with the wording of s.4. It is an inversion of the “necessity” limitation in Article 10(2), which applies to restrictions on the Article 10 right, not the exercise of it.4
The proposition in  that “an individual's Article 8 right to reputation will be breached by the publication of unproven allegations without a remedy” [emphasis added] is a further inversion of the ECHR jurisprudence without authority to support it. The ECHR has relatively recently expanded the Article 8 right to privacy to cover some defamation claims.5 The fact that the Article 8 right may be engaged does not mean
3 See Gatley [14.18].
4 See, for example, Bladet Tromsø v Norway App. No 21980/93, referred to in the Judgment at , at
 to 
5 See Pfeiffer v Austria (2009) 48 EHRR 8, cited in Thornton v Telegraph Media Group Ltd  1 WLR 1985) at .
that it will be breached unless a defendant who cannot truth prove is required to provide a remedy. There is a significant body of ECHR jurisprudence where the Article 10 right has been breached for the opposite reason, namely, the domestic courts have subjected the defendant to a “penalty” (which includes payment of damages and costs) because of an inability to prove truth.6
As Lord Phillips observed in Flood v Times Newspapers Ltd  2 AC 273 at , there is “a need for care when applying to the law of defamation decisions on the tension between article 8 and article 10 in other contexts”. Parliament has resolved the tension in this context by passing s.4. The role of the courts is to apply s.4 dispassionately without gloss or qualification.
There are two principal consequences, in the context of the present case, that flow from the Court of Appeal’s approach:
Reynolds factor 7 – “Whether comment was sought from the plaintiff” – is elevated to a requirement (as it was described at  and ), except where it is impractical for the defendant to do so. This is the corollary of the necessity test referred to in [3.1.2] above. A similar consequence may arise from any perceived journalistic failure. Most failures could be described as unnecessary.
The application of the defence becomes atomised and the bigger picture ignored. Whether publication was in the public interest requires consideration of the bigger picture. A publication can still serve a public interest function, for example, by exposing an exploitative charlatan, even if there are journalistic failings. The Judge was entitled to take the view that “this was a piece that cried out to be published”.7
The Court of Appeal did not demonstrate the requisite restraint before overturning the Judge’s determination of the two limbs of s.4(1):
An appellate court should only interfere with such determinations if the judge made an error of law or they were outside the range of reasonably available alternatives (see Lord Kerr in Stocker v Stocker  UKSC 17 at  to  in the context of defamatory meaning). Where the judge has upheld the s.4 defence at trial, the claimant/appellant faces a dual hurdle: to establish that the
6 Bladet Tromsø referred to above is an example.
7 At , for the reasons primarily referred to at  to .
judge could not have reasonably believed that the defendant reasonably believed that publication was in the public interest. Further, the judge will have had the advantage of seeing the defendant explain, under cross-examination, why he believed that it was in the public interest to publish.
The Court of Appeal’s judgment (“the Judgment) does not identify any error of law made by the Judge. The Court did not agree with his findings that the subject matter was of public interest, the Defendants had made reasonable inquiries prior to publication and the failure to contact the Claimant did not vitiate their reasonable belief. But this is insufficient to overturn them. They were not subjected to a proper perversity test. If s.4 is to be given the appropriate flexibility, consistent with its wording, the Judge’s findings fall well within the bounds of “reasonably available alternatives”.
The Court’s analysis at  to  of the Appellants’ “failures” does not accurately represent the First and Third Appellants’ evidence, conflates the Judge’s findings on truth and s.4 and/or relates to matters on which they were not cross-examined. More fundamentally, an appellate court should only rely on alleged journalistic failures to overturn a judge’s finding in favour of a defendant whether the effect of such failures is that the judge could not have reasonably concluded that the defendant reasonably believed that publication was in the public interest.
The Court of Appeal did not demonstrate the requisite restraint before overturning the Judge’s finding that imputation 13(4) had been proved. It is unprecedented for an appellate court to overturn a finding by the tribunal of fact that all or part of a truth/justification defence has been proved and substitute its finding that it has not. This could only arise where the witnesses found to be truthful were incapable of belief or their evidence (and any other evidence relied on by the tribunal) was incapable of proving the relevant allegation. The Court of Appeal did not make such findings and there would be no basis to do so.
Alternatively, the Judge’s finding that the Respondent’s reputation had been shot to pieces, thereby depriving him of any damages other than nominal, was justified by the other imputations found to be true, including 13(6) which the Judge was entitled to describe as the “most serious”.
Alternatively, following the Court’s finding that the trial was unfair, the Defendants should have been permitted to prove the truth of the imputations complained of other than 13(4) at the retrial and, if necessary, rely on s.2(3) to make good the truth defence.
The Court of Appeal was wrong to hold that in modern civil law practice there is a bar on the judge “descending into the arena” and/or that judicial rudeness can render a trial unfair. A trial can only be rendered unfair by judicial conduct if (1) the judge has prejudged the case; and/or (2) a litigant is deprived of an adequate opportunity properly to advance their case and challenge their opponent’s. The judge “descending into the arena” and/or being rude may support a case in relation to (1) and/or (2), but, of itself, is insufficient.
A civil judge, as opposed to a criminal judge, should be free to adopt a more investigative approach to the trial given that they are the ultimate finder of fact and should not remain aloof. Civil judges were encouraged to adopt an active case management approach following the Woolf reforms. A judge should also intervene when there are real concerns about the administration of justice.
There was no suggestion throughout the trial that the Judge had either prejudged matters, or that the Respondent was unable to put his case and challenge the Appellants’ witnesses. The Respondent could not be categorised as an ill- prepared ingenue, wholly unfamiliar with the court process. He had previously instructed highly competent solicitors and counsel (apparently generating approximately £100,000 in costs) and continued receiving the benefit of assistance from the solicitors throughout the trial, as was apparent from the Properties of the Word file comprising the Respondent’s closing submissions. Although Polish, his English was fluent and clear and he was complimented by the Judge on his closing speech.
The Judge was rightly concerned about two important areas, namely: (1) that the Respondent may have committed perjury; and (2) adequate disclosure.
Whilst the Judge was at times robust, that would be insufficient to render a trial unfair. This has the inevitable and highly undesirable consequence of considerable extra court time and costs, and causes severe prejudice to litigants such as the Appellants who are thereby forced to re-commence litigation through no fault on their part.
Further, characterisation of the Judge having an animus towards the Respondent is too vague a concept to render the trial unfair.
The criteria for permission to appeal
Freedom of expression is a fundamental human right. The balance to be struck between it and the protection of reputation is of general public importance. The stated purpose of Parliament in the 2013 Act was to rebalance defamation law in favour of freedom of expression in order to reduce the number of claims and the “chilling” effect of the threat of a claim.8 The Parliamentary debate emphasised the need to protect public interest journalism.9 The main way in which such journalism is protected is by a user-friendly public interest defence.
Lachaux v Independent Print Ltd  UKSC 27 is the only case on the interpretation of the Act to have reached the Supreme Court. It concerned the serious harm requirement in s.1. Lord Sumption noted at  that: “Broadly speaking, it [the Act] seeks to modify some of the common law rules which were seen unduly to favour the protection of reputation at the expense of freedom of expression”. Lord Sumption identified one of the concerns that s.1 sought to address.
The principal concerns that s. 4 seeks to address are evident from (a) its wording in the context of preceding developments in the common law and (b) (insofar as is necessary to consider it) the Parliamentary debate. These concerns are that the application of the common law Reynolds defence was inflexible, the ten factors were, in practice, treated as a rigid checklist and the assessment of the defendant’s journalism was vulnerable to hindsight bias.10
17. S.4 is more a favourable defence than Reynolds. This is evident from its wording (considered further in  below). To use an expression favoured by the US Supreme Court, it is intended to provide the “breathing space” that enables freedom of expression to flourish.11 The corollary is that there will be claimants who are subject to false imputations and the law of defamation will no longer
8 See for example, the Ministerial Foreword to the draft Bill .
9 See, for example, the Report of the Joint Committee on the Draft Defamation Bill at .
10 See, for example, the Joint Committee at  and Lord McNally, Minister of State for Justice and the promoter in the House of Lords, (Hansard, House of Lords, Grand Committee, 19th December 2012, column GC534).
11 “That erroneous statement is inevitable in free debate, and that it must be protected if the freedoms of expression are to have the "breathing space" that they "need . . . to survive.”” New York Times Co.
v. Sullivan (1976) 376 U.S. 254 per Brennan J at 272.
provide a remedy. The balance between these competing interests is a question of public policy for Parliament.
Common law developments relevant to s.4 start with Derbyshire CC v Times Newspapers Ltd  AC 534. This was the first explicit recognition of the “chilling effect” of the common law requirement to prove truth. Having cited New York Times Co. v Sullivan (1976) 376 U.S. 254, Lord Keith observed at 548E: “What has been described as "the chilling effect" induced by the threat of civil actions for libel is very important. Quite often the facts which would justify a defamatory publication are known to be true, but admissible evidence capable of proving those facts is not available. This may prevent the publication of matters which it is very desirable to make public”.
The chilling effect is exacerbated by artificial features of the common law that determine the “truth” that must be proved, such as the single meaning rule and the repetition rule.
In 1999 the House of Lords formulated the Reynolds defence in order to address “the 'chilling' effect of this rigorous, reputation protective principle”.12 It was conceived as a form of qualified privilege arising from the duty of the media to impart information on matter of public interest and the interest of the public in receiving it. The standard was “responsible journalism”. Lord Nicholls set out 10 non-exclusive factors that might be taken into account in determining whether the standard had been met. The House rejected a Sullivan type defence which would require the claimant to prove malice, on the basis that it would not provide sufficient protection to reputation.13
Next, in Jameel v Wall Street Journal Europe Sprl  AC 359 the House of Lords made clear that the ten factors were not a series of hurdles. Criticism was made of the restrictive way it which the defence had been applied by first instance judges. Baroness Hale observed at : “It should by now be entirely clear that the Reynolds defence is a “different jurisprudential creature” from the law of privilege…. It is not helpful to analyse the particular case in terms of a specific duty and a specific right to know. That can, as experience since Reynolds has shown, very easily lead to a narrow and rigid approach which defeats its object”. The House upheld the defence after it had failed at first instance due to the failure to wait for the claimants to comment.
12 See Lord Nicholls at 192H.
13 See Lord Nicholls at 201F-H.
It should be noted that the failure to give the claimant a reasonable opportunity to comment was never a requirement of the Reynolds defence, let alone a tenacious one, as suggested by the Court of Appeal at . The Court did not refer to GKR Karate (UK) Limited v Porch  EMLR 410, an investigative article with similarities to the present, where Sir Oliver Popplewell regarded it as a factor to be “put in the balance on the claimant's behalf”.14 He upheld the defence having found the journalist to be “an honest, sensible and responsible person on whose evidence I could rely” and that the “criticisms” of her conduct “neither individually or cumulatively, outweigh the interest in the free flow of information in this case”.15
Finally, in Flood v Times Newspapers Ltd  2 AC 273, the only Supreme Court Reynolds case prior to the Act, emphasis was placed on the need to give weight to editorial discretion. However, the standard remained “responsible journalism”.
The Draft Defamation Bill published on 15 March 2011 provided for a defence of “Responsible publication on matter of public interest” in s.2. The wording of s.2(1) is essentially the same as s.4(1), as enacted, with the substitution of “reasonably believed that publishing the statement complained of was in the public interest” for “acted responsibly in publishing the statement complained of”. S.2(2) essentially replicated Lord Nicholls’ factors in relation to whether the defendant acted responsibly. In contrast, section 4(2) merely refers to “all the circumstances of the case” in relation to whether the belief was reasonable.
Over the course of the Parliamentary debate, a variation of Sullivan was proposed that would require the defendant simply to believe that publication was in the public interest.16 S.4 is plainly a compromise involving subjective and objective elements.
There are some obvious but important points that can be made about the wording of s.4. First, there is no reference to the defendant or the journalism being responsible. Its relevance can only arise through the application of the words “ín the public interest”. Second (and relatedly), there is no reference to Lord Nicholls’ factors. Third, the defendant does not have to prove that publication was in the public interest, let alone that the journalism was responsible. The section is directed to what the defendant believed and whether the belief was reasonable.
While the Appellants accept that the interpretation of the words “reasonable belief” and “in the public interest”, which appear in other statutes, may vary depending on
14 At page 428.
15 At pages 429-430.
16 See Lord McNally, Minister of State for Justice (Hansard, House of Lords, Grand Committee, 19th December 2012, column GC533 to 535).
the context, some consequences are inherent in their use. In particular, (a) the judge may regard the belief to be wrong, but nevertheless reasonably held and (b) where the judge believes that publication was in the public interest (as the Judge found in the present case) it must follow that it was reasonable for the defendant to believe that it was. This is so even if the judge’s belief is based on different grounds to the defendant. All that matters is that the defendant’s (subjective) belief was (objectively) reasonable.17
It is apparent that the s.4 defence is a materially more flexible and generous defence than Reynolds, even in the most favourable of the five judgments in its final iteration in Flood. In contrast, the Court of Appeal’s interpretation is significantly less favourable to a defendant than Flood.
There has been one previous Court of Appeal decision on s.4, Economou v De Freitas  EMLR 7, cited in the Judgment at  and . It dismissed an appeal from the decision of Warby J to uphold the defence.
While it is not necessary for the Appellants to take issue with Sharp LJ’s judgment in Economou in order to succeed in the proposed appeal, it is submitted that it too closely aligns the s.4 defence to Reynolds. It does not give proper recognition to the points made in  and  above that flow from the differences in wording between s.4 and Reynolds.
In Lachaux, the Supreme Court overturned the Court of Appeal’s narrow interpretation of s.1, which, effectively, preserved the status quo. Lord Sumption noted at  that if the Court of Appeal was correct: “it is difficult to see that any substantial change to the law of defamation has been achieved by what was evidently intended as a significant amendment”.
There is, at least, an arguable case to justify permission to appeal that the approach of the Court of Appeal in the present case deprives journalists and others who find themselves sued for defamation of the protection that Parliament intended by passing s.4. The grounds of appeal raise issues of law of general public importance that ought to be resolved definitively by the Supreme Court now.
The proposed appeal also gives rise to a discrete point of law of general public importance relating to the standard of appellate review. The issue was left open in Economou (at ). It was also left open by Lord Phillips in Flood who described it as “a matter of general importance” (at  to ). He appeared to favour
17 See, for example, Chesterton Global Ltd v Nurmohamed  IRLR 837
the more limited form of intervention, which accords with the approach of Lord Kerr in Stocker. The argument in favour of this approach has grown stronger since Flood for a number of reasons. First, it is consistent with subsequent Supreme Court decisions involving multi-factorial findings, such as McGraddie v McGraddie  UKSC 58, cited in Stocker. Second, “in the public interest” is a vaguer standard than “responsible journalism” and involves a greater number of factors.18 Third, s.4 has an enhanced subjective element, affording greater advantage to the trial judge from seeing the defendant give evidence.
The correct test of whether a trial has been rendered unfair by the conduct of the judge, encompassing the extent to which a judge in a civil trial may adopt a more investigative approach, is a point of law of general public importance.
There is a tension between the approach of the Court of Appeal in the present case and Sarabjeet Singh v Secretary of State for the Home Department  4 WLR 183 where Davis LJ observed at : “In general terms, there need be no bar on robust expression by a judge, so long as it is not indicative of a closed mind. In fact, sometimes robust expression may be positively necessary in order to displace a presumption or misapprehension, whether wilful or otherwise, on the part of an advocate or litigant on a point which has the potential to be highly material to the case.”
The result of the approach taken in the present case is to constrain how High Court judges are free to manage and try civil trials, inevitably leading to lengthier hearings. The consequences of the very public opprobrium of a highly respected High Court judge will have a chilling effect on other members of the wider judiciary, both full-time and Deputy High Court Judges contemplating full-time application.
Where a judge’s conduct is improper, for example by being overly rude, but does not impact on the fairness of the trial, a complaint to the Judicial Conduct Investigations Office is sufficient recourse.