Taming the ‘Chilling Effect’ of Defamation Law: English Experience and Implications for Australia

The United Kingdom: Thornton and Reaffirming Existing Elements

Prior to the Defamation Act 2013, a threshold requirement of seriousness (‘substantial’ harm) had already been recognised in Thornton v Telegraph Media Group Ltd (‘Thornton’) as part of the common law definition of defamation. ¹⁵ This was not a new element: it was a re-affirmation of existing principles of defamatory capacity. So much was clear from the precedents relied upon in Thornton, which did not support a standalone substantive requirement, beyond the tort’s established elements. Justice Tugendhat had relied upon Lord Atkin’s contrast, in Sim v Stretch (‘Sim’), ¹⁶ between attacks on character, and mere bad manners and discourtesy. ¹⁷ That contrast was one between substantive imputations capable of being defamatory and those which are not; this confirmed that mere bad manners and discourtesy could never be defamatory, not that they were insufficiently substantial but nevertheless capable of defaming. Lord Atkin was extrapolating what was meant by ‘tend[ing] to lower the Plaintiff in the estimation of right-thinking members of society generally’, not adding a further element ¹⁸ ; his Honour was illustrating (though not defining) a seriousness threshold. ¹⁹ In defining this inherent seriousness threshold, Justice Tugendhat had also relied upon Jameel v Dow Jones & Co (‘Jameel’), ²⁰ where the Court confirmed trivial claims could be struck out for abuse of process. ²¹ That was no more than an exercise of the Court’s inherent jurisdiction and general discretion to strike out a claim where continuing it would, as a procedural matter, be a disproportionate use of judicial resources, given that a real and substantial tort had not been committed; it was not a recognition of an additional, substantive element of substantial harm in the tort. ²² Indeed, it has been confirmed that the new serious harm threshold in section 1 is separate from, and has not displaced, that inherent jurisdiction. ²³

Thornton therefore added nothing more to the existing common law than that the ‘sting’ of defamation must be substantial as opposed to trivial (in the sense of Lord Atkin’s bad manners or discourtesy). ²⁴ That requirement did not entail assessing the broader circumstances of the publication and its effects, as well. ²⁵ Thornton certainly did not introduce into defamation law the need to prove special damage; the tort remained actionable per se. Indeed, a major impetus behind codifying a standalone seriousness element in the Defamation Act 2013 was the fact that, despite Thornton, there was deemed to be an insufficiently high threshold for a defamation claim to be brought, ²⁶ contributing to the perceived ‘chilling effect’.

Australia: Thornton, Triviality and a High Onus on Defendants

Australian courts adopted Thornton as requiring proof that the impugned words tended to do substantial harm to the plaintiff’s reputation ²⁷ : this was merely an ‘elucidation’ of the principle that the imputation must tend to lower the estimation of the plaintiff by the community. ²⁸ The courts interpreted this threshold matter as being consistent with (and no more than) Lord Atkin’s articulation of defamatory capacity in Sim, ²⁹ and rejected that Thornton (or Jameel) introduced a new element to the tort, or a requirement to examine factors external to the pleaded imputation. ³⁰ Therefore, Australian common law required no proof of special damage and no additional, standalone element of seriousness.

Section 33 of the MDPs codified a defence of triviality under which the defendant bore the onus of proof to show the circumstances of publication meant the plaintiff was unlikely to suffer harm. The MDAPs have repealed that defence and effectively replaced it with the new serious harm threshold. ³¹ Unlike the Thornton seriousness threshold, this defence did require an assessment of the circumstances of the publication, rather than merely the meaning and its inherent tendency to harm reputation. ³² The defendant had to show that, regardless of whether any harm actually ensued, and notwithstanding that the publication could lead ordinary reasonable people to think less of the plaintiff, the circumstances were such that the plaintiff was unlikely to sustain harm as a result. ³³ These circumstances included the content, extent and recipients of the publication. ³⁴

The combination of the Thornton requirement as adopted in Australia and the statutory triviality defence could be seen to have generated a ‘chilling effect’. While the plaintiff was required only to adduce evidence of defamatory capacity to pass the seriousness threshold, the defendant was required to adduce evidence of external circumstances of the publication to demonstrate triviality. Furthermore, given that Thornton was effectively captured by the classical Sim requirement of defamatory capacity, and given that the triviality defence placed a ‘high’ onus on the defendant to show the plaintiff was unlikely, in a prospective sense, to suffer ‘any’ harm, ³⁵ neither the seriousness threshold nor the triviality defence were of use to defendants wishing to argue the impugned statements were too anodyne or weak, in the context as a whole, to defame or cause harm. ³⁶ In the recent Armstrong v McIntosh (No 4) (‘Armstrong’) case, which involved four text messages sent by the defendant to his friend, alleging his ex-wife’s brother (the plaintiff) had acted in an unchristian, mendacious and immoral way to engineer the defendant’s ruination, the plaintiff satisfied the Thornton requirement, and the triviality defence failed; damages were recovered in the sum of $6,500, including both compensation and aggravation. ³⁷

With the new serious harm threshold, the MDAPs have effectively transferred the higher onus as to triviality (or, inversely, seriousness), in the circumstances of the publication from the defendant to the plaintiff. ³⁸ This effect is consistent with the judicial interpretation of the new serious harm threshold in the Defamation Act 2013.

Initial Interpretations: Presumption of Damage Abolished

It is critical to understand how judicial interpretations of the serious harm threshold diverged before the Supreme Court judgement in Lachaux SC, to understand the precise way in which this new element has narrowed the tort to address the ‘chilling effect’, without eroding its core protective capacity altogether. In one of the first judgements handed down in the Queen’s Bench after section 1 came into effect, Justice Bean held that the new provision required proof (by evidence or inference) of specific reputational harm. ⁴¹ This represented a significant change from pre-existing doctrine. The tort could effectively no longer be understood as actionable per se, apart from in irresistibly clear cases involving widespread publication of egregious allegations. ⁴² This initial interpretation may have given some credence to arguments that the defamation reforms narrowed the tort to such an extent that claimants might try to bring their defamation claims under alternative, more capacious, causes of action, including data protection and misuse of private information. ⁴³

It was also held that Parliament intended to raise the Thornton and Jameel requirements of ‘substantial’ harm to a higher bar of seriousness: this was supported by the legislative history of section 1, which saw the word ‘serious’ introduced during parliamentary proceedings to replace ‘substantial’. ⁴⁴ Justice Bean also reasoned that section 1 demanded an assessment of the circumstances of publication to determine serious harm, rather than merely the pre-existing common law test of whether the words had a tendency to defame, and this was connected with the requirement to prove specific harm, separate from defamatory capacity. ⁴⁵ Finally, it was confirmed that the reference in section 1 to whether serious harm is ‘likely to be caused’ permitted the plaintiff, as at date of trial, to look prospectively to potential harms that might flow from publication, in order to show the publication meets the threshold. ⁴⁶

This temporal aspect of section 1 has given rise to concerns about whether plaintiffs ‘might be able to sue successfully, hence recover compensatory damages, for a loss that has not yet occurred − and of course might never occur at all’. ⁴⁷ Yet this consequence of section 1, and particularly the construction in Cooke v MGN Ltd (‘Cooke’) of ‘likely to cause harm’ as justifying a prospective examination of probable harm, was accepted without concern (albeit obiter) in the High Court, where the Judge recognised ‘the status of a publication may change from non-defamatory to defamatory. A cause of action may lie inchoate until serious harm is caused or its future occurrence becomes probable’. ⁴⁸ In fact, this prospective element of the serious harm threshold has led to a finding that a potentially defamatory publication became non-defamatory due to actions taken following the publication, which demonstrated that publication was not likely to cause serious harm: evidence of the plaintiff having effectively restored his reputation following the impugned publication was accepted as showing that publication was not likely to cause the plaintiff serious harm in the future. ⁴⁹

The evidentiary burden imposed on plaintiffs in Cooke was tempered to some extent in Ames v Spamhaus Project Ltd (‘Ames’), where Justice Warby conceded that the circumstances in which serious harm could be inferred (without special evidence) might be more common than the exceptional, egregious instances suggested in Cooke. ⁵⁰ However, Justice Warby did not negate Cooke in respect of section 1 requiring proof of specific harm, breaking with pre-existing common law. ⁵¹ That was consistent with the opinion of some commentators, following Cooke, who foresaw that plaintiffs would at least ‘be forced to offer up evidence at an early stage of how the harm they have suffered is serious’, and that that may ‘increase the likelihood that unmeritorious cases will not be brought’. ⁵² At this stage of its doctrinal evolution, the new threshold had definitively narrowed the tort relative to what was available to plaintiffs before the reforms and had in this way responded to the ‘chilling effect’.

The Cooke principle of specific harm was then explicitly confirmed by Justice Warby in the first-instance judgement in Lachaux HC: section 1 had indeed abolished the presumption of damage in the tort of defamation. ⁵³ Lachaux HC involved newspaper publications falsely alleging the plaintiff had, amongst other things, been controlling and violently abusive towards his ex-wife, that he had abducted their son from her, and that he himself had falsely accused her of abducting their son. Justice Warby found on these facts that serious harm, as a requirement to prove special harm over and above a tendency to lead ordinary reasonable people to think less of the plaintiff, had been established in respect of most of the publications. ⁵⁴

Following that judgement, subsequent first-instance judgements continued to apply section 1 as requiring proof of damage. ⁵⁵ It was confirmed that ‘serious harm’ required something more than injury to feelings, however grave; that, although the plaintiff and defendant were both permitted to call evidence to prove or challenge serious harm, a court was entitled to draw inferences from adduced evidence, so that specific evidence of serious harm may sometimes be unnecessary; and that the prospective temporal aspect of section 1 meant that the ‘grapevine effect’, or on-publication beyond the original publisher, could be relevant to serious harm. ⁵⁶

Justice Warby’s confirmation in Lachaux HC that section 1 had abolished the presumption of damage did not, however, abolish the tort itself, as his Honour made clear in that judgement, ⁵⁷ and then again demonstrated in Monroe: a series of tweets had met the serious harm threshold (and ultimately amounted to defamation warranting £24,000 in damages), even though they had not caused ’grave’ harm. ⁵⁸ The circumstances and consequences of the tweets’ publication proved that serious harm had been caused: Twitter analytics were accepted as demonstrating the impugned tweets had appeared on screen, been re-published, were replied to, and were ‘liked’, a significant number of times, and the Judge accepted that the abuse the plaintiff had received was a result of those tweets, their reach, and the engagement with them (re-tweets, replies and likes). That was evidence of serious harm under section 1. ⁵⁹

Reversion to Thornton in the Court of Appeal

This growing body of principles arising from High Court decisions, in particular that section 1 abolished the presumption of damage, was then put into doubt by the Court of Appeal. In its judgement in Lachaux CA, the Court held that section 1 had not abolished the presumption of damage, but had only raised the common law threshold set in Thornton from substantial harm to serious harm. ⁶⁰ The Court reasoned that Parliament had in fact not intended to impose a new burden on plaintiffs to show on a balance of probabilities that serious harm had been or would be caused ⁶¹ ; rather, the correct reading of section 1 was that the words had a tendency to cause serious harm, drawn from the meaning and the context. On the Court’s view, section 1 simply gave ‘statutory status’ to Thornton, though it raised the substantive bar. ⁶² A court could draw an inference of seriousness from the gravity of the statement and the context of the publication, and that inference could be rebutted by evidence to the contrary. ⁶³

The Court’s interpretation was more conservative than, and indeed inconsistent with, the normative concerns and impetus underlying the reforms, which highlighted the need to narrow the tort and address the ‘chilling effect’. ⁶⁴ Furthermore, the Court’s rejection of section 1 abolishing the presumption of damage and actionability per se appeared to be at odds with its contention that harm could be inferred from meaning and context, but then rebutted with evidence to the contrary. Some High Court judgements that followed the Court of Appeal’s decision saw Judges emphasise the appellate Court’s statement as to inferences from meaning and context, and the possibility of rebuttal of such inferences with evidence to the contrary, as a way of preserving the Cooke principles, and downplay the Court of Appeal’s contention that section 1 merely codified Thornton and added nothing more. ⁶⁵

In the High Court cases that followed the Court of Appeal’s judgement, but preceded the Supreme Court’s decision on appeal, the outcome on the serious harm threshold issue in each case ultimately turned on the facts (including the context of publication), and it cannot be said that there was a discernible swing back in favour of plaintiffs, due to the more conservative reading of section 1. ⁶⁶ However, the Court of Appeal’s interpretation remained (at least apparently) at odds with the core purpose of the defamation reforms, and the exact requirements of section 1 remained unclear until the Supreme Court delivered its judgement in July 2019.

Supreme Court Confirmation: A Standalone Element of ‘External’ Harm to Tame the ‘Chilling Effect’

Parliamentary intention and drafting were central to the Supreme Court’s unanimous confirmation that section 1 had indeed gone further than simply raising the common law threshold. Lord Sumption confirmed that the intention behind the legislation was ‘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’, ⁶⁷ and thereby generate a ‘chilling effect’. His Lordship reasoned that, with the Act’s overall purpose in mind, but primarily in view of the language used in section 1, the serious harm threshold ‘not only raises the threshold of seriousness above that envisaged in… Thornton, but requires its application to be determined by reference to the actual facts about its impact and not just to the meaning of the words’. ⁶⁸ That is a clear acknowledgement of both parliamentary intent and the language used in section 1: ‘has caused or is likely to cause’ harm denotes a requirement to demonstrate harm external to the ‘sting’ itself, and not a requirement merely to show liability to cause harm given the tendency of the words. ⁶⁹ Indeed, ‘if serious harm can be demonstrated only by reference to the inherent tendency of the words, 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’. ⁷⁰

Having confirmed Parliament’s will to reform defamation law in this way — to tame the ‘chilling effect’ — Lord Sumption importantly specified the exact extent to which Parliament had changed the common law: ‘the defamatory character of the statement no longer depends only on the meaning of the words and their inherent tendency to damage the claimant’s reputation’. ⁷¹ That did not, in his Lordship’s view, and echoing Justice Warby at first instance, revolutionise the law of defamation. Rather, it meant that a statement involving a particularly serious allegation, or particularly intense ‘sting’, might not meet the threshold because the circumstances of publication of that statement may be of such a nature as not to have caused or be likely to cause serious harm. Lord Sumption confirmed that the externalities — the harm as an external value — were now material to establishing the tort of defamation:

Suppose that the words amount to a grave allegation against the claimant, but they are published to a small number of people, or to people none of whom believe it, or possibly to people among whom the claimant had no reputation to be harmed. The law’s traditional answer is that these matters may mitigate damages but do not affect the defamatory character of the words. Yet it is plain that section 1 was intended to make them part of the test of the defamatory character of the statement. ⁷²

Therefore, the question of whether serious harm has been suffered must be determined by reference to the facts as to the impact of the statement, and not solely the meaning of the words. However, inferences of fact can be drawn from the publication’s circumstances, including the ‘inherent probabilities’ of the publication in context. ⁷³ Relevant factors included (i) the scale of the publication; (ii) the fact that the statements had come to the attention of people who knew the plaintiff; (iii) the fact that the statements were likely to have come to the attention of others who either knew the plaintiff or would come to know him in future and (iv) the gravity of the statements themselves, according to the legal meaning attributed to them. ⁷⁴

Even though Lord Sumption referred to ‘inherent probabilities’, it would be disingenuous to interpret that as reverting to the Court of Appeal’s reasoning and focus on meaning and tendency, irrespective of the context and impact of publication. His Lordship referred to the inherent probabilities of a publication in light of its nature and circumstances, and not the inherent tendency of the meaning of words to defame the plaintiff. ⁷⁵ Indeed, his Lordship emphasised that ‘inherent tendency’ alone was insufficient, and the impact of publication had to be accounted for under section 1. ⁷⁶

In this way, the Supreme Court highlighted and clarified the conceptual difference between the harmful consequences caused by or likely to be caused by a publication (which can be called ‘external harm’), and the inherent reputational harm that may be implicit in the meaning of the impugned words (which can be called ‘internal harm’). The Court unanimously confirmed that Parliament had intended to narrow the tort of defamation and shift the boundary between reputational protection and freedom of expression, by importing into the tort a new requirement to prove external harm, either by calling specific evidence or by drawing inferences from established facts. Though this is not a radical transformation of the tort, it is a clear and intentional doctrinal shift from the less burdensome requirement to show ‘internal harm’ only — harm that is presumed in imputations that could lead ordinary reasonable people to think less of the plaintiff.

That authoritative statement of law from the highest court in the United Kingdom not only settles the judicial disagreement between the first-instance judges and the Court of Appeal, the resulting lack of clarity about whether harm is presumed or must be proven, and whether any inferences of harm can be rebutted ⁷⁷ ; it also puts out of any doubt that the tort of defamation, following the most recent reforms in the United Kingdom, has undergone a decisive, though not revolutionary, change in scope and character: from a tort that is straightforwardly actionable per se, to a tort with an additional, new, standalone element requiring proof of damage.

Remaining Uncertainties

Although the purpose and requirements of the serious harm threshold are now settled, the procedural stage at which the issue should be decided remains to be determined. The Supreme Court was not required to determine that matter. At first instance, Justice Warby treated it as a preliminary issue, ⁷⁸ while the Court of Appeal would have left the issue for full trial. ⁷⁹ Even where the issue is dealt with as a pre-trial or preliminary matter, the strength of legislative purpose behind the provision and the clarity of the Supreme Court decision in Lauchaux will mean that this procedural direction alone should not discharge the substantive requirement that the plaintiff actually prove serious harm from the impact of the publication, as a matter that is external to how intense the words’ sting might be.

Another matter that is unsettled, and difficult to predict, is the actual outcome of particular disputes about serious harm, in various factual situations. The Supreme Court’s ruling confirms that much depends upon how the particular court adjudicates particular sets of facts, and this potential for variability can be observed in the High Court judgements handed down after the Court of Appeal’s decision in Lachaux CA but before the Supreme Court’s judgement on appeal ⁸⁰ : it is difficult, if not impossible, to discern a guiding pattern across different sets of facts, which may indicate how any future case may be determined, even as between those cases that were subject to the interpretation of the Court of Appeal, and those cases preceding that ruling, or following the Supreme Court’s ruling.

Similarly, it is unlikely that, following the Supreme Court’s ruling, certain claims on particular fact situations will be routinely struck out for not meeting the threshold. ⁸¹ As Lord Sumption reasoned, the narrowing of the tort, though it is a way of addressing the ‘chill’ on freedom of expression, is not a revolutionary change, and outcomes will continue to turn on the facts of each case: a judge’s ‘task [is] to evaluate the material before him, and arrive at a conclusion on an issue on which precision will rarely be possible’. ⁸² In one case following the Supreme Court’s decision, the Judge found there was a real prospect of establishing serious harm on the basis of the inherent probabilities of the publication. ⁸³ In another case, the threshold was held to have been met (and general damages awarded in the sum of £25,000), where a former disgruntled client published an online review of a solicitor’s firm making false allegations of dishonesty. ⁸⁴ In a more recent High Court case, the Supreme Court’s interpretation of section 1 was arguably stretched to its limit, when the Judge held he was satisfied that, ‘based on the meaning alone, and without any necessary additional evidence’, the threshold was met, in circumstances where, although the scope of the publication was limited to one person (the defendant’s sister), ‘the nature of the language allows one to infer serious harm’. ⁸⁵ The Judge reasoned that the assessment of damage to reputation under section 1 was not ‘a numbers game’, and the statement in this case — a false allegation of rape — was particularly egregious. ⁸⁶

The United Kingdom: Reynolds and the Problem of Responsible Journalism

It is important to recall the English courts acknowledged the need to protect freedom of expression, and press freedom, particularly on matters of public interest, long before the Defamation Act 2013 brought into effect a new public interest defence. ¹⁰⁴ Following this basic principle, the House of Lords recognised in 1999 that the common law defence of qualified privilege to the tort of defamation extended to media reports on matters of public interest, subject to a test of ‘responsible journalism’. ¹⁰⁵ In those circumstances of responsible media reportage of material ‘which it is proper for the public to know’, ¹⁰⁶ the defence could be available and was not limited by the classic requirement of reciprocal duty and interest as between publisher and publishee: publications at large could benefit from this extended qualified privilege defence.

Acknowledging the inherent uncertainty in determining whether material was in the public interest and whether the responsible journalism requirements were met, and the risk that such uncertainty might ‘chill the publication of true statements of fact’, ¹⁰⁷ Lord Nicholls set out ten factors that could be taken into account ¹⁰⁸ : seriousness of allegation; the extent to which the subject-matter is of public concern; source of information; steps taken to verify the information; status of the information; urgency of the matter; whether plaintiff had been approached for comment; whether the material contained the gist of the plaintiff’s side of the story; tone and circumstances of publication including timing.

Reynolds was authority for the propositions that freedom of expression was a crucial principle that must inform the scope of the tort of defamation, that the ‘press discharges vital functions as a bloodhound as well as a watchdog’, and that courts should be ‘slow to conclude’ against public interest and the public’s right to know, particularly vis-à-vis political discussion. ¹⁰⁹ However, the defence of qualified privilege was not extended to all political communications, as it was considered that would encroach too far on reputational rights, which were integral to protecting individual dignity. Likewise, although ‘doubts [in any particular case] should be resolved in favour of publication’, ¹¹⁰ the defence itself hinged on whether the responsible journalism standards had been observed. That was an important limit on this extended qualified privilege defence, and it preserved the critical difference between English and American jurisprudence on defamation in the context of political or public interest reportage, and the deeper issue of the boundary between reputational protection and freedom of expression. ¹¹¹

The limit of responsible journalism ultimately caused insurmountable problems for defendants who wished to rely upon Reynolds. In one judgement following Reynolds, the defence was held to be unavailable for a sensationalised story which unambiguously alleged serious wrongdoing, including criminal guilt: the tone of the publication indicated responsible journalism standards had not been met. ¹¹² Yet, it is inaccurate to claim the courts applied the responsible journalism requirement and Lord Nicholls’s ten factors in a prohibitively rigid or unduly inflexible manner; rather, they recognised the important function and varying practices of journalists in reporting on matters of public interest. ¹¹³ The Supreme Court emphasised, echoing Lord Nicholls’s own assertions, that not all of the ten factors were prerequisites for responsible journalism in every case, especially as ‘public interest privilege involves a spectrum’ of circumstances ranging from pure reportage to reportage that is not neutral. ¹¹⁴

Nevertheless, concerns remained that the courts were applying the responsible journalism requirement too rigidly, ¹¹⁵ and that the framing of the Reynolds defence, including the ten factors, rendered it too difficult for media defendants to access and litigate successfully. ¹¹⁶ Media and publishers found the defence too complicated and expensive to run, which was due in part also to a lack of clarity or certainty on how that defence operated in practice, there having been limited development of the common law following Reynolds. ¹¹⁷ There was also uncertainty about how Reynolds should be applied outside mainstream journalism, and that its limited scope might generate a ‘chilling effect’ on such publishers. ¹¹⁸

Australia: A Statutory Reynolds-Like Defence with Similar Problems

Australian courts did not welcome the developments in Reynolds, ¹¹⁹ and instead preserved the classical qualified privilege defence, which depends upon a reciprocal duty and interest, is defeasible by malice and is not available for publications to the world at large. ¹²⁰ It is noteworthy that the High Court, even before Reynolds, recognised that common law qualified privilege in Australia specifically protects publications on political or governmental matters, given that, by virtue of the Constitution, all Australians have an interest in receiving these communications. ¹²¹ That specific defence, however, being based on the implied constitutional freedom of political communication, is limited to political communication only and has not been the basis of further development of qualified privilege in Australia. ¹²²

Aside from the common law, the MDPs contain a separate statutory defence of qualified privilege, oriented towards but not limited to media defendants, which is not restricted by the reciprocal duty-interest requirement. ¹²³ The section 30 defence requires the recipient to have an interest or apparent interest in receiving the information on that subject, ¹²⁴ and the defendant’s conduct in publishing the matter to be reasonable in the circumstances. The provision sets out a non-exhaustive list of ten factors pertaining to reasonableness, which are based upon but not identical to Lord Nicholls’s Reynolds factors, and include: the public interest, any relevance to public functions, whether the matter distinguishes between suspicions and proven facts, the nature of the defendant’s business operations, source of the information, inclusion of the plaintiff’s side of the story and steps taken to verify the allegations. This defence is defeasible by proof of malice, which can be established if it is shown the matter was published for a purpose, or with a motive, that is foreign to the circumstances and context giving rise to the interest on which the defence hinges. ¹²⁵

As in the United Kingdom with Reynolds, in Australia it was recognised that the section 30 adjusted qualified privilege defence was not readily pleaded by defendants, and when it was, it was not often successful. ¹²⁶ In a recent case involving serious allegations of the plaintiff’s involvement in an international bribery scandal, where the section 30 defence failed, the publication was found to be of ‘considerable public interest’, by virtue of the plaintiff being a public figure (albeit a reluctant one). ¹²⁷ However, even though malice had not been made out, the defendant media company’s conduct was unreasonable in all the circumstances, including: that the journalist who authored the impugned article had failed to consider and analyse crucial details of the alleged conspiracy to bribe the United Nations General Assembly President, including that relevant individuals had not even been alleged to have participated, let alone been charged; that asserting the plaintiff might remain in China to avoid extradition was for sensationalist purposes, without any reliable source to verify that assertion, and was unwarranted and damaging; that the assertions that certain offending web pages supposedly revealing the plaintiff’s connection with the conspiracy were ‘scrubbed from cyberspace’ were inaccurate, unjustified and damaging; that there was an inadequate and unfair presentation of the plaintiff’s side of the story; that the article as a whole contained ‘sensational and hyperbolic language and… [a] general sneering and deprecating tone’; and that there was ‘a lack of any demonstrable care and caution’ in reporting serious allegations against the plaintiff including the failure to take adequate steps to ensure defamatory imputations were not conveyed. Damages were awarded in the sum of $280,000. ¹²⁸

The Lange defence is also not often raised (successfully) in defamation cases. Concerns about limited and uncertain scope and application in practice, and overly rigid application, arose in Australia in the same way as they did in the United Kingdom. ¹²⁹ Given that it was still encased in a qualified privilege defence (albeit an adjusted and liberated version of that defence), the element of public interest was not at the forefront of the section 30 defence, and certainly it was not the central or definitive feature of that defence. ¹³⁰

The lack of clarity in how these extended qualified privilege defences operated where publications were considered to be of public interest, and the lack of success of defendants pleading these defences, motivated the introduction of a new, standalone, defence of public interest in both the United Kingdom and Australia. ¹³¹ It is intended to concretise, clarify and broaden the place of public interest in the defences to a defamation action. To that extent, it is intended to narrow the tort relative to pre-existing doctrine, by alleviating the overall complexities and burdens faced by defendants wishing to argue public interest. ¹³²

The Purpose and Drafting of Section 4: Abolishing Reynolds but Reflecting Established Principles

Although section 4 introduced a new, standalone, public interest defence, it was intended to be ‘based on the existing common law defence established in Reynolds…and…to reflect the principles established in that case and in subsequent case law’. ¹³⁵ The abolition of the Reynolds defence is therefore intended to consolidate existing principles into a newly framed defence, not erase those principles outright. ¹³⁶ The essential substantive change brought about by section 4 is that the Reynolds requirement to show the defendant acted responsibly in publishing the matter is replaced with a requirement to show they reasonably believed that publication was in public interest. ¹³⁷ That core element, of public interest coupled with reasonableness, has been adopted in the new public interest defence in the MDAPs.

Even though concerns about uncertainty as to the scope and proper operation of the Reynolds defence motivated the codification of a standalone public interest defence in the United Kingdom, it was considered that there was no need, in that new defence, to define what is meant by the ‘public interest’: that was ‘a concept…well-established in the English common law and that in view of the very wide range of matters which are of public interest and the sensitivity of this to factual circumstances attempting to define it in statute would be fraught with problems’. ¹³⁸ Instead, it was intended that ‘a holistic view may be taken of the statement in the wider context of the document, article etc in which it is contained’. ¹³⁹ Indeed, there is an explicit instruction that the court must have regard to all of the circumstances, ¹⁴⁰ and the abolition of the ‘Reynolds defence’ ¹⁴¹ makes it even clearer that the ten Reynolds factors are not constitutive or determinative of public interest or reasonableness. This was considered as being ‘helpful to send a signal to the courts and practitioners to make clear the wish of Parliament that the new defence should be applied in as flexible a way as possible in light of the circumstances’. ¹⁴²

Yet, doubts were raised following the enactment of section 4 about how far the courts would move away from the Reynolds factors. Some argued they would continue to assess those same considerations under the new defence, especially as the new statutory provision deliberately avoided providing more detail about when a matter was in the public interest and when a defendant reasonably believed publication was in the public interest. ¹⁴³

Supreme Court Confirmation: The Correct Place of Reynolds under Section 4

In fact, the courts have confirmed section 4 rules out the use of the ten Reynolds factors as a checklist ‘to which reference ought to be made, in particular order to check whether a preliminary conclusion should be confirmed’. ¹⁴⁴ As the Supreme Court ruled in Serafin, ‘[e]ven if, at the time of the decision in the Reynolds case, it was appropriate to describe the factors identified by Lord Nicholls as a check list, it is clearly inappropriate so to regard them in the context of the statutory defence’. ¹⁴⁵ Although in that case, the Supreme Court confirmed that the origin of the new defence lies in the common law, and that the relevant common law principles underpinning section 4 are contained in Reynolds, Jameel v WSJ and Flood, ¹⁴⁶ Parliament had clearly not intended in section 4 that the courts reach for the Reynolds ten factors, ‘by way of checklist’, and apply them to every case in which the public interest defence is raised. ¹⁴⁷ The Court acknowledged the parliamentary intention that the new defence be applied flexibly, as inferred from both the drafting and the legislative history, which included the deliberate removal of the Reynolds factors from the draft provision, and pre-legislative consultation confirming the Bill should not be interpreted as a checklist. ¹⁴⁸ Such was the error of the Court of Appeal in Serafin when it resorted to the Reynolds factors as a ‘checklist’, that, given there had to be a new trial, the Supreme Court ordered the new trial judge to determine the public interest defence without any reference to the Court of Appeal’s interpretation of it.

The Supreme Court in Serafin approved and applied the Court of Appeal’s reasoning in an earlier judgement, Economou, which had in turn upheld Justice Warby’s interpretation of the defence at first instance. The Court of Appeal had reasoned that the statute could have made reference to the ten Reynolds factors, but, given it had not, and had instead explicitly directed courts to account for all of the circumstances, those ten factors did not apply as a rule in every case. ¹⁴⁹ Indeed, the factors taken into account, and the weight given to them, ‘would vary from case to case’. ¹⁵⁰ That meant that, depending on the facts of the case, the non-exhaustive list of Reynolds factors may be relevant to the outcome of the new public interest defence. ¹⁵¹ After all, the new defence’s replacement of responsibility with reasonableness did not erase the principles that underpinned the pre-existing common law: that statutory change did not mean ‘the rationale for the Reynolds defence and for the public interest defence are materially different, or that the principles that underpinned the Reynolds defence, which sought to hold a fair balance between freedom of expression on matters of public interest and the reputation of individuals, are not also relevant when interpreting the public interest defence’. ¹⁵² That is consistent with the legislative intention to consolidate existing principles into a newly framed and codified defence, as confirmed by the Supreme Court, and as had already been proposed by Justice Warby in the High Court: ‘there are some features of the Reynolds defence which … must on any view carry through into the new law’. ¹⁵³

Justice Warby’s reference to ‘features’ of Reynolds carrying through, which was upheld by the Court of Appeal and approved by the Supreme Court, ¹⁵⁴ was a reference to basic common law principles that give meaning to the new public interest defence, rather than to the ten factors listed by Lord Nicholls as a rubric to be transferred onto the new defence’s ‘reasonableness’ inquiry. Indeed, one of Justice Warby’s ‘features’, or principles, was the need to maintain flexibility in the judicial approach to determining public interest and reasonableness. ¹⁵⁵

Existing Principles to Determine Reasonable Belief

According to Justice Warby, another principle derived from existing common law was that ‘a belief [can] 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’. ¹⁵⁶ This is a principled interpretation of what ‘reasonable belief’ entails, as it appears in the new statutory defence, rather than a robotic reverting back to the ten factors in Reynolds. Justice Warby went on to confirm that the ‘enquiries and checks that can reasonably be expected must be bespoke, depending on the precise role that the individual plays,’ ¹⁵⁷ so that, where the defendant has actual knowledge of the falsity of an allegation, but nevertheless publishes that information, that defendant cannot be said to have reasonably believed the publication to be in the public interest: as a matter of law, no reasonable person could believe that it is in the public interest to publish information known or believed to be false. That proposition has subsequently been applied in High Court cases, ¹⁵⁸ including one case which involved imputations that the plaintiff delighted in using Nazi symbols. In that case, the public interest defence failed because the defendants did not believe, and there were no grounds to believe, that the plaintiff was a Nazi or partook in Nazi symbolism. ¹⁵⁹

This interpretation of section 4 is not an application of a checklist from Reynolds; instead, the courts are carrying through common law principles that best give effect to the core elements of the new defence. The fact that the principles relevant to responsible journalism that flowed from Reynolds through Jameel v WSJ are relevant to the interpretation of reasonableness in the new defence ¹⁶⁰ does not mean the new defence is merely a restatement of Reynolds. Justice Warby’s approach to public interest and reasonableness in this regard was upheld by the Court of Appeal, where it was emphasised that, in articulating the principles of reasonableness, Justice Warby was — correctly — not accepting that the Reynolds factors were key to determining reasonableness in that case, or any other case. ¹⁶¹ Rather, Justice Warby was acknowledging that certain underlying principles of common law had to flow through in the interpretation of the new statutory defence.

This approach, of recognising and applying the underlying common law principles that remain relevant to the new defence despite the abolition of the Reynolds defence, was followed in the High Court in a case where the public interest defence was upheld. ¹⁶² In Onwude v Dyer (‘Onwude’), a defamation claim was brought by a gynaecologist against a journalist, the British Medical Journal, and its editor, vis-à-vis an article reporting a decision of the General Medical Council disciplinary tribunal, which had found the gynaecologist guilty of misconduct on several charges and had erased him from the medical register. The tribunal decision was overturned on appeal, and that fact was reported by the defendants only after the plaintiff had informed them of the fact. The public interest defence succeeded, and the Court highlighted the change brought about by section 4: ‘the question is not (as it was under the common law regime) whether the defendants’ journalism was responsible, but whether the defendants reasonably believed that publishing the first article (and abstract) was in the public interest’. ¹⁶³

Although basic common law principles persist as to the importance of proper journalistic conduct and editorial judgement, ¹⁶⁴ whereas the former defence, hinging on responsible journalism, might have failed in Onwude, the new defence, which hinges on reasonable belief in public interest, was established. That is one decision on the new section 4 defence which, though it recognised and applied pre-existing common law principles (including from Reynolds), ¹⁶⁵ nevertheless clearly demonstrates that the new public interest defence has narrowed the tort of defamation relative to how it had been expected to operate in light of the previously framed — now abolished — Reynolds defence.

Existing Principles to Protect Individual Reputation

The Court of Appeal in Economou confirmed another common law principle which remained relevant to the interpretation of the new defence: the ‘importance of reputation to the individual and to society more generally’. ¹⁶⁶ Crucially, the Court reasoned that the impact on individuals of the publication of false imputations was more serious today, in the context of online communications and an ‘era of distrust and “fake news”’, than it was when Reynolds was decided. ¹⁶⁷ It was relevant to the interpretation of what counts as public interest, or a reasonable belief in public interest, that ‘it is more important than ever that the public should, so far as possible, be put in the picture as to where the truth, or some approximation of it, lies’. ¹⁶⁸

Since that reasoning invokes the very normative essence of the tort of defamation — the imperative to protect individual reputation from the harm in proliferating falsehoods — it confirms that, although the new public interest defence has narrowed the tort by alleviating some complexities and burdens previously faced by defendants, the new defence has not diminished the importance of the principle at the heart of the tort: the law’s protection and vindication of individual reputation in the face of false imputations.

Existing Principles to Limit Journalistic Excesses

A similar principle can be discerned from the approach taken to the new public interest defence in the most recent High Court decision in Lachaux. ¹⁶⁹ This principle is that, although the new public interest defence is intended to narrow the tort of defamation by lessening the risks associated with the previous defence and borne by media defendants in particular, and, although this new defence includes a specific requirement that courts take into account editorial judgement where they deem it appropriate, ¹⁷⁰ the new defence does not, as a matter of law, give a new, special latitude, autonomy or deference to journalists or editors in their reportage of matters that may be of public interest. ¹⁷¹ The replacement of the Reynolds responsible journalism requirement with the requirement to show reasonable belief in public interest does not mean the new public interest defence carves out of the tort of defamation a novel protected space for journalists or editors. That was not the purpose of the defence, and it would go beyond narrowing the tort to reduce excessive ‘chilling effects’ on public interest communications generally.

Therefore, just as it was contemplated in pre-existing common law, it can be the very actions of journalists in their reportage of public interest stories that demonstrates they, as defendants, did not reasonably believe the publication was in the public interest. In Lachaux, Justice Nicklin was highly critical of the publishers for failing to ensure basic journalistic standards were upheld, and having found the public interest defence failed, his Honour awarded £120,000 in damages. In the circumstances of this case, under the new public interest defence, it was incumbent upon the defendants to verify the allegations against the plaintiff in light of their gravity. The failure to do so, and, in particular, the failure of any journalists or editors to approach the plaintiff for comment and fairly represent his response in the articles, meant, in the circumstances of such grave and widely published allegations, the defendants could not reasonably have believed it was in the public interest to proceed with such publications. Referring to the Court of Appeal’s and Supreme Court’s endorsement of the principle that whether appropriate pre-publication checks were undertaken is relevant to whether the defendant reasonably believes publication is in the public interest, Justice Nicklin reasoned that principle could also be inferred from section 4(3): that provision ‘is an implicit recognition that efforts to verify will usually be regarded as an important factor in the assessment of the reasonableness of a defendant’s belief that publication was in the public interest’. ¹⁷²

Where the circumstances involve media or journalist defendants, ¹⁷³ the public interest and reasonableness issues at the heart of the public interest defence may therefore hinge on whether the defendants acted in accordance with the relevant journalistic standards. ¹⁷⁴ This should not be seen as a re-juridification of the Reynolds responsible journalism standards, contrary to the intention and proper construction of section 4. It is a straightforward interpretation of the reasonableness element of the public interest defence, in the context of media reportage. To ignore or downplay journalistic standards when considering the public interest defence, in cases where journalists or media editors are defendants, would be to elevate journalists or editors above the law of defamation and effectively see the courts abdicate their adjudicative authority to editorial judgement in cases of media reportage.

The English courts have, in their engagement with the new public interest defence so far, recognised that defence has effected substantive change to the law of defamation, and has narrowed the tort of defamation to make it less difficult to argue a defence based upon publication in the public interest, than it had been previously. The defence is clearly targeted at reducing ‘chilling effects’. However, the courts have emphasised that the new defence has not transformed the tort so fundamentally as to give special protection to editors and render journalistic standards wholly irrelevant, to weaken the core normative importance of protecting reputation against falsehoods, or to eliminate pre-existing common law principles that are germane to the interpretation of public interest and reasonable beliefs.