Reform of the law of defamation – the defence of fair and reasonable publication

Hurdles via flickrSection 26 of the Defamation Act 2009 (also here) introduced a new defence of fair and reasonable publication into Irish defamation law. In Meegan v Times Newspapers Ltd [2016] IECA 327 (09 November 2016) Hogan J for the Court of Appeal (Finlay Geoghegan and Peart JJ concurring) explained that the

section is clearly designed to provide a defence for publishers who show that they acted bona fide and that the publication was fair and reasonable having regard, in particular, to the matters set out in section 26(2) of the 2009 Act. Section 26 may be regarded as an endeavour by the Oireachtas to move away in some respects from the strict liability nature of the common law tort of libel and to introduce – in, admittedly, some specific and limited respects – a negligence based standard in actions for defamation under the 2009 Act. ([2016] IECA 327 (09 November 2016) [10])

For all that section 26 is a centre-piece of the reforms worked by the 2009 Act, it is, in my view, a legislative dead letter. It is over-complex, placing far too many hurdles in the way of a successful invocation. Meegan illustrates the point. The plaintiff garda claimed that she had been defamed by the defendant’s allegations that she had supplied sensitive information to a paramilitary group. The defendant pleaded the defence of fair and reasonable publication; and the High Court ordered discovery of the journalist’s notes and other background material relevant to article ([2015] IEHC 696 (06 November 2015)). The Court of Appeal reversed, on the grounds that it was

premature to assess whether the discovery sought is genuinely necessary for the proper conduct of this litigation, at least until the scope and extent of the section 26 defence is clarified and particulars of the facts proposed to be relied upon by the defendant in support of that defence are duly ascertained, whether by further pleading or by particulars. ([2016] IECA 327 (09 November 2016) [18] (Hogan J) (emphasis added)).

Although from the context, it is clear that Hogan J meant that the scope and extent of the section 26 defence that needed to clarified were the scope and extent of the defence as pleaded on the facts, such that clarification could be forthcoming following delivery of particulars, in my view the lack of clarity as to the scope and extent of the defence must have been contributed to in no small part by the complexity and opacity of the terms of section 26 itself. A simpler, clear, statutory statement of the defence would have made for less ambiguity in what the defendant was relying upon and what the plaintiff could seek by way of discovery. There are so many hurdles in section 26 that it was not clear what the defendant was relying on. This raises very serious questions about the workability of the defence.

By way of contrast, in England and Wales, section 4 of the Defamation Act 2013 introduced a similar defence of publication on matter of public interest; it is in much simpler terms; and it was successfully invoked for the first time in Economou v de Freitas [2016] EWHC 1853 (QB) (27 July 2016). In this post, I want to discuss that case, and to present that defence as a template for the reform of the defence of fair and reasonable publication in the 2009 Act.

Economou v de Freitas is the first case in which the defence of publication on matter of public interest in section 4 of the Defamation Act 2013 has been successfully invoked. Eleanor de Freitas and Alexander Economou had been in a relationship. In January 2013, de Freitas accused Economou of rape. He was arrested, but never charged. In August 2013, he started a private prosecution against her for perverting the course of justice; and the Crown Prosecution Service (CPS) continued it. Four days before trial, in April 2014, she killed herself. In various interviews later that year, her father, David de Freitas, said that he wanted the inquest into her death expanded to include the role of the CPS. Economou alleged that, in those interviews, de Freitas had libellously accused him both of having falsely prosecuted Ms de Freitas for perverting the course of justice and of having raped her. Warby J held that some of the publications referred to Economou, had a defamatory meaning, and caused “serious harm” to his reputation. de Freitas therefore sought to rely on the public interest defence in section 4 of the Defamation Act 2013, which provides:

(1)     It is a defence to an action for defamation for the defendant to show that—

(a)     the statement complained of was, or formed part of, a statement on a matter of public interest; and
(b)     the defendant reasonably believed that publishing the statement complained of was in the public interest.

(2)     Subject to subsections (3) and (4), in determining whether the defendant has shown the matters mentioned in subsection (1), the court must have regard to all the circumstances of the case.

(6)     The common law defence known as the Reynolds defence is abolished.

Warby J held that there was a strong public interest in the question of whether the CPS, as a public authority, had gone wrong in deciding to prosecute a mentally unwell rape complainant who had killed herself almost on the eve of her trial. Furthermore, the role of the inquest and the desirability of permitting private prosecutions were additional matters of public interest. Against that background, he therefore held that that de Freitas could rely on the defence:

250. Mr de Freitas could and did properly consider the publication to be in the public interest; and that a judgment in favour of Mr Economou would represent an interference with Mr de Freitas’ free speech rights out of any reasonable proportion to the need to protect and vindicate Mr Economou’s reputation.

Section 4 had been discussed in Barron v Vines [2015] EWHC 1161 (QB) (29 April 2015), and Yeo v Times Newspapers Ltd [2015] EWHC 3375 (QB) (25 November 2015), but Economou v de Freitas is the first time it has been successfully invoked. Counsel for the claimant submitted that, in determining whether a defendant’s belief that publication was in the public interest was reasonable for the purposes of section 4, the court should be guided by the checklist proposed by Lord Nicholls in Reynolds v Times Newspapers [1999] UKHL 45, [2001] 2 AC 127 (28 October 1999). That case concerned the limits of the common law defence of qualified privilege. The House of Lords emphasised that the elasticity of the defence enables a court to give appropriate weight to the importance of freedom of expression by the media on matters of public concern; and Lord Nicholls offered ten illustrative matters which could be taken into account in deciding whether the public interest privilege was established. In Jameel v Wall Street Journal Europe Sprl [2007] AC 359, [2006] UKHL 44 (11 October 2006) the House of Lords reformulated Reynolds as a test of responsible journalism in the public interest. And, in the application of Jameel, the UK Supreme Court in Flood v Times Newspapers Ltd [2012] 2 AC 273, [2012] UKSC 11 (21 March 2012) provided significant latitude to editorial judgment. However, section 4 replaced these developments, and expressly abolished the common law defence. The Scottish Law Commission Discussion Paper on Defamation (Discussion Paper no 161, 2016, pdf) noted that section 4 is not a direct statutory codification of the common law. Nevertheless, paragraph 29 of the Explanatory Notes to that section explain:

This section creates a new defence to an action for defamation of publication on a matter of public interest. It is based on the existing common law defence established in Reynolds v Times Newspapers and is intended to reflect the principles established in that case and in subsequent case law.

There is an ambiguity between “creat[ing] a new defence” (emphasis added) and merely “reflect[ing] the principles established in” Reynolds. In Economou v de Freitas, Warby J neither accepted nor rejected counsel’s submission that the court should be guided by the Reynolds checklist. Instead, he acknowledged that some features of the Reynolds defence had indeed carried through into section 4:

240. Among these are flexibility and, by statutory definition, adaptability to the circumstances of the individual case. A third is a recognition that, as the … [Grand Chamber of the European Court of Human Rights] put it in Hrico v Slovakia 49418/99 (2005) 41 EHRR 18, [2004] ECHR 365 (20 July 2004) [40](g) “There is little scope under Article 10(2) of the Convention for restrictions on … questions of public interest …”. Fourthly, there is the allowance for editorial judgment which the court is required to make by section 4(4). I would not go all the way with Mr Barca’s submission [for the defendant] that this “must simply refer to the subjective judgment of the defendant”. It seems to me that this is linked with the requirement of flexibility and the need to demonstrate convincingly that a restriction on Article 10(1) rights is necessary and proportionate in pursuit of an identified legitimate aim.

241. I would consider a belief to be reasonable for the purposes of section 4 only if it is one arrived at after conducting such enquiries and checks as it is reasonable to expect of the particular defendant in all the circumstances of the case. Among the circumstances relevant to the question of what enquiries and checks are needed, the subject-matter needs consideration, as do the particular words used, the range of meanings the defendant ought reasonably to have considered they might convey, and the particular role of the defendant in question.

Warby J should have decisively rejected the submission based on Reynolds. At common law, in reformulating Reynolds, the House of Lords and Supreme Court moved away from the checklist in Jameel and Flood; there is no justification to re-introduce it into the defence of defence of publication on matter of public interest in section 4; and it is is not necessary to reach the conclusion in para 241. All that it can do is add unnecessary uncertainty.

Moreover, section 4 did not just replace the checklist in Reynolds, it replaced the reformulation in Jameel as well. That reformulation focussed on the responsibility of the defendant’s journalism, whereas section 4

operates to all intents and purposes not on the basis of the responsibility of the journalism but rather on the reasonableness, or otherwise, of the belief held by the defendant in any given case that publication of the statement in question was in the public interest. (SLC Discussion Paper [6.8] 41).

This change from Reynolds and Jameel to section 4 was to emphasise the importance of editorial discretion (ibid [6.10] 42), and this was the basic issue in Economou v de Freitas. For all of these reasons, the reference in that case to Reynolds was, at best, unnecessary. The abolition of the Reynolds defence in section 4(6) should be given full effect, and the defence in section 4(1) should be given effect in its own terms.

It is a much simpler defence that the Irish equivalent in section 26 of the 2009 Act. Subsection (1) sets out the terms of the defence. Subsection 2 goes on to provide a list of 10 factors which the court can take into account in determining whether the defence in subsection (1) is made out. The section was introduced on foot of recommendations in the Report of the Legal Advisory Group on Defamation (2003) (pdf here and here):

Looked at in the round, therefore, the jurisprudence would suggest that there is a strong case in favour of the idea of affording something like a broad qualified privilege on matters of public interest which has due regard to good faith principles. ([7] p6) …

It is clear that the law is out of step with developments in other jurisdictions which share our legal tradition [such as Australia, New Zealand and the United Kingdom]. It is arguable, indeed, that it is out of step both with the values which are enshrined in the Constitution and … the European Convention of Human Rights. ([11] p7)

I was a member of that Group. And while I stand by the argument of principle that Irish law should provide a public interest defence, I now think that we got the implementation of that principle in section 26 very wrong. That section was referred to, but not discussed in any great detail, in Lowry v Smyth [2012] 1 IR 400, [2012] IEHC 22 (10 February 2012), Griffin v Sunday Newspapers Ltd [2012] IEHC 132 (30 March 2012), Walsh v News Group Newspapers [2012] IEHC 353 (10 August 2012), McAdam v ABP Food Group [2014] IEHC 536 (14 November 2014), and Ryanair v Quigley [2015] IEHC 776 (09 December 2015). The first time it had an impact on the outcome of a case is Meegan v Times; that impact was a baleful one; and the contrast between that case and Economou v de Freitas helps to illustrate why I now think that section 26 is a legislative dead letter. In Meegan, the complexity of section 26 meant that it was not sufficiently clear what the defendant’s defence actually was. On the other hand, in Economou the simplicity of section 4 meant that the defence could be successfully relied upon.

There are two dense layers to the complexity of section 26. The first relates to the intricacies of subsection (1). There are eleven hurdles to be jumped before the defence is established, as follows:

(1)   It shall be a defence (to be known, and in this section referred to, as the “defence of fair and reasonable publication”) to a defamation action for the defendant to prove that-

(a)   the statement in respect of which the action was brought was published-

(i)    I in good faith, and
(ii)   in the IIa course of, or for the IIb purpose of, the discussion of a subject of III public interest, the discussion of which was for the IV public benefit,

(b)   in all of the circumstances of the case, the V manner and VI extent of publication of the statement did not exceed that which was VII reasonably  VIII sufficient, and

(c)   in IX all of the circumstances of the case, it was X fair and XI reasonable to publish the statement.

Many of these terms seem synonymous; but, because of the basic rule of statutory interpretation that all of the words of a statute should be given effect if possible, even the (near-)synonyms must be given separate effect. And since each of these terms must be given separate effect, this means that there are many hurdles to be jumped. In my view, there are so many hurdles that it is almost impossible to see how they can all be jumped, and the terms of the defence fully made out. In my view, there are far too many such hurdles. In much the same way that Warby J in Economou v de Freitas relied on ECHR cases such as Hrico v Slovakia (above) to guide his interpretation of section 4, it may be that a similar reliance might ameliorate some of the complexity of section 28. But this is only a faint prospect. And, anyway, I doubt that it would be sufficient to make section 28 workable. We on the Legal Advisory Group on Defamation should have recommended a much simpler subsection (1), and I hope that this is considered in the review of the 2009 Act currently being conducted by the Department of Justice.

The second problematic layer of complexity in section 26 relates to subsection (2), which provides 10 factors which the court can take into account in determining whether the defence in subsection (1) has been established. These factors add many more hurdles to the 11 already identified in subsection (1). And if subsection (1) is unworkable in its own right, this unworkability is compounded by the addition of the factors in subsection (2).

Worse, the interpretation of the factors is likely to be confused by reference to the checklist in Reynolds, to which the factors in subsection (2) are similar but not identical. Section 26 and Reynolds have already both been discussed together, almost interchangeably, by Peart J in the High Court in Ryanair Ltd v Somner [2014] IEHC 634 (12 December 2014), and by Dunne J in the Supreme Court in Leech v Independent Newspapers (Ireland) Ltd [2014] IESC 78 (19 December 2014). If questions can arise under the 2013 Act in England and Wales about the inter-relationship between Reynolds and section 4, even where section 4(6) has expressly abolished the Reynolds defence, it is not surprising that the text of section 36 of the 2009 Act invites similar questions.

Reynolds was (probably) followed as a matter of Irish law by Ó Caoimh J in Hunter v Duckworth [2003] IEHC 81 (31 July 2003) (noted here and here). And Jameel was (probably) followed as a matter of Irish law by Charleton J in Leech v Independent Newspapers [2007] IEHC 223 (27 June 2007) (noted here). There do not seem to be any Irish case on Flood v Times Newspapers (above), or the more recent Privy Council decision in Pinard-Byrne v Linton [2015] UKPC 41 (12 October 2015). Jameel and Flood went well beyond the Reynolds factors; and they were ultimately displaced in the law of England and Wales, if not by those decisions, then by section 4(b) of the 2013 Act. Section 28(2) froze the development of the law at something similar to Reynolds, where Jameel and Flood – to say nothing of section 4 of the 2013 Act – became far more sophisticated.

Crossed fingersIf, in section 28, the complexity of subsection (1) was an error, then that error was compounded by the addition of subsection (2). At this point, the best remedy would be to repeal section 28 in its entirety, and replace it with something akin to section 4 of the 2013 Act in England and Wales. As Oliver Lock notes of Economou v de Freitas (see Inforrm | Farrer), that latter provision successfully places emphasis on the specific circumstances of each case and, in particular, on the role of the defendant in deciding to publish the words complained of, and it can do this without recourse to an over-elaborate formulation of the defence (such as is to be found in section 28(1) of the 2009 Act), or to a fossilized list of possibly relevant factors (such as is to be found in section 28(2)). The Recommendations to the Department of Finance on the Reform of Defamation Law in Northern Ireland (June 2016; pdf), prepared by Prof Andrew Scott of the LSE, suggested that Northern Irish law should adopt an equivalent of section 4 of the 2013 Act, and Scotland (Discussion Paper (2016) (pdf)) is considering a similar development. There is every good reason to suggest that Irish law should do the same, and to hope that this will be one of the outcomes of the review of the 2009 Act currently being conducted by the Department of Justice. Fingers crossed.