• Specific Year
    Any

Gould, Kim --- "Locating a 'Threshold of Seriousness' in the Australian Tests of Defamation" [2017] SydLawRw 15; (2017) 39(3) Sydney Law Review 333


Locating a “Threshold of Seriousness” in the Australian Tests of Defamation

Kim Gould[∗]

Abstract

This article addresses the question whether the Australian tests of defamation carry a seriousness threshold, akin to the Thornton threshold found in the English common law tests, notwithstanding differences between the two legal and constitutional environments. The question is important given that the statutory triviality defence has disappointed as the primary mechanism for weeding out ‘unworthy’ defamation claims. If left unchecked, such claims can threaten the proper balance between freedom of speech and the protection of reputation, not to mention unreasonably burden the administration of justice. It will be argued that the threshold is inherent in the Australian tests, but if additional support is required, then recourse may be made to the emerging Bleyer abuse of process jurisdiction, albeit not without some qualification. This article advances the ongoing defamation reform debate in Australia by highlighting the potential availability of a largely neglected filter and, in so doing, uncovers the proper basis for assessing the need for additional or alternate measures for enhancing the law’s response to unworthy defamation claims.

I Introduction

A key element in an action for defamation is that the publication in question is defamatory. What is ‘defamatory’ is not defined in the Australian uniform national defamation legislation,[1] and its elucidation is left, instead, to the common law.[2]

In Radio 2UE Sydney Pty Ltd v Chesterton,[3] the High Court of Australia declared that ‘(t)he general test [for defamation] ... [i]s whether the published matter is likely to lead an ordinary reasonable person to think the less of the plaintiff’,[4] although this does not necessarily shut out other common law tests.[5] Clearly the tests frame and constrain a plaintiff’s defamation claim and changing their scope will directly affect the size of the potential catchment area for successful claims.

In 2011 in Thornton v Telegraph Media Group Ltd,[6] Tugendhat J identified a ‘threshold of seriousness’ in the English common law tests of defamation. The threshold quickly became an established feature of United Kingdom (‘UK’) defamation law. Recently, New Zealand (‘NZ’) High Court judges in CPA Australia Ltd v New Zealand Institute of Chartered Accountants[7] and Opai v Culpan[8] accepted such a limitation for the NZ common law tests. It is only natural to ask whether the Australian tests also carry such a threshold given their common pedigree.

To date there has been no such declaration by an Australian court. The first, and (to the author’s knowledge) only, time that an Australian superior court has addressed the question of Thornton and ‘a minimum threshold of seriousness’ was the 2017 decision of the Full Court of the Supreme Court of South Australia in Lesses v Maras.[9] However, in that case, there does not appear to have been full argument on the question, and the Court’s consideration only extended to a few paragraphs.[10] The Court stated that

The passage from the judgment of Tugendhat J [in Thornton] ... should be understood as merely an elucidation of the requirement that, to be defamatory, an imputation must tend to lower the estimation of the plaintiff by the community and an emphasis that an adverse opinion may be expressed about a person without its having such a tendency. The seriousness of the adverse opinion is obviously a factor to be taken into account in determining whether its expression does tend to lower the estimation of the plaintiff by the community. The passage should not be understood as creating an additional element of the cause of action for defamation.[11]

Denying the threshold operates as ‘an additional element of the cause of action for defamation’,[12] does not deny the possibility of the threshold being inherent in the defamation tests.[13] Further, a determination of the proper meaning and effect of Thornton regarding a ‘seriousness threshold’ does not necessarily determine the status of such a threshold for the Australian tests of defamation. Accordingly, the question is still alive and awaits further judicial consideration in Australia.

The question is important given its capacity to exclude ‘unworthy’ defamation claims. Such capacity is valuable for the purpose of maintaining the balance the law purports to strike between free speech and protection of reputation.[14] Allowing unworthy claims to proliferate without restraint is apt to tip that balance unduly in favour of reputation at the expense of free speech. It is also liable to burden the administration of justice unreasonably in terms of resources, costs and delay. The importance of this question will be heightened if existing Australian filters for unworthy claims are found wanting, especially under pressure from new communications technologies providing platforms for unmoderated communication by ordinary people with unprecedented audience reach. It has also become more pressing in the wake of recent calls for Australia to follow the UK’s lead and adopt its new ‘serious harm’ test found in Defamation Act 2013 (UK) s 1(1).[15] However, it would be useful to examine the current position regarding seriousness under Australian defamation law before proceeding in that direction. The analysis in this article will contribute to the necessary audit.

It is trite to say that English decisions are no longer binding on Australian courts; but, they, like other foreign decisions, may be persuasive.[16] A key factor that may detract from their persuasiveness, however, relates to differences between the legal and constitutional contexts of Australia and the other jurisdiction. Two potentially relevant differences between Australia and the UK are the absence of an equivalent to the Human Rights Act 1998 (UK) (‘HRA’) in all of the Australian jurisdictions and the absence of a dedicated statutory triviality defence to a defamation action in the UK.[17] The same may also be said of the relevant differences between Australia and NZ, save, of course, for replacing the HRA with the New Zealand Bill of Rights Act 1990 (NZ). However, it is too superficial simply to dismiss the possibility of a seriousness threshold for Australia on the basis of these differences alone. A more nuanced and critical analysis is required, especially in the wake of recent Australian cases regarding the abuse of process jurisdiction, as well as the statutory triviality defence.

The aim of this article is to conduct such an analysis. For this purpose, the Australian tests of defamation will be taken to include the ‘hatred, contempt, or ridicule’ test[18] and the ‘shun and avoid’ test,[19] in addition to the Radio 2UE Sydney ‘general test’ stated above.[20] It will be argued that a case can be made for finding an inbuilt, but as yet largely undeclared, ‘threshold of seriousness’ in these tests, notwithstanding the absence of a Human Rights Act as well as the presence of the statutory triviality defence throughout Australia. Should additional assistance be required, then resort may be made to the emerging Bleyer[21] principle of proportionality as a head of abuse of process, although not without some qualification. Further support may be garnered from certain policy considerations of the kind that resonated in CPA Australia.[22] Nevertheless, the minimum level at which the revealed threshold for seriousness is pitched is unclear but likely to be lower than Thornton and possibly too low to address adequately the 21st century need to exclude unworthy defamation claims.

This article is structured as follows. Part II analyses the Thornton decision with a view to teasing out the reasoning underpinning the seriousness threshold identified by Tugendhat J in the English defamatory meaning tests, as well as identifying questions relevant to its discovery in the Australian tests. Part III then interrogates the Thornton primary and secondary reasons and their relevance in the Australian context. Part IV completes the English story by introducing the s 1(1) ‘serious harm’[23] test through the lens of early case law, but with the particular aim of emphasising its differences to the Thornton test. Although they both speak to minimum thresholds, they are very different thresholds. Part V follows with an analysis of the reasoning in the two NZ decisions, aiming to discern their relevance for Australia, as well as to uncover the true nature of the threshold recognised. Part VI goes on to examine key aspects of the Australian legal and constitutional context that may bear upon the recognition of a seriousness threshold. Prominent among these, apart from the absence of an HRA equivalent in all of the Australian jurisdictions, are the emergence of the Bleyer principle of proportionality and the narrow scope of the statutory triviality defence. Part VII then draws on the foregoing analysis to scrutinise the Australian defamatory meaning tests for a seriousness threshold and consider whether — and, if so, to what extent — external drivers may be needed and are available to support such a threshold. Finally, Part VIII provides an opportunity reflect on the ramifications of recognising a seriousness threshold in the Australian tests, including whether this step would go far enough to address the concerns prompting calls to import the English s 1(1) test.

II The Thornton “Threshold of Seriousness”

The proceedings in Thornton[24] concerned an application for summary dismissal of an action for defamation arising out of an allegation of ‘copy approval’,[25] made in a review of the plaintiff’s book, published in the print and online versions of the defendant’s newspaper. In granting the defendant’s application, Tugendhat J held that the ‘copy approval’ allegation was not capable of conveying a defamatory meaning either personally or professionally but, if it was, then it was not capable of conveying a ‘sufficiently’ defamatory meaning; that is, one that passed the relevant ‘threshold of seriousness’.[26] In his Honour’s view, ‘whatever definition of “defamatory” is adopted, it must include a qualification or threshold of seriousness, so as to exclude trivial claims’.[27] This ‘threshold of seriousness’ was conceived of as ‘some tendency or likelihood of adverse consequences for the claimant’.[28] And it was stated to apply to both personal and business/professional defamation.[29] His Honour gave two main reasons for this conclusion which are worth quoting in full:

(i) It is in accordance with the true interpretation of Lord Atkin’s speech in Sim v Stretch [1936] 2 All ER 1237. It is also in accordance with the decision of Sharp J in Ecclestone v Telegraph Media Group [2009] EWHC 2779 with which I respectfully agree. (ii) It is required by the development of the law recognised in Jameel (Yousef) v Dow Jones & Co Inc [2005] EWCA Civ 75; [2005] QB 946 as arising from the passing of the Human Rights Act 1998: regard for article 10 and the principle of proportionality both require it.[30]

Justice Tugendhat examined seven definitions of defamatory meaning,[31] including the classic common law trio familiar to Australia — namely, the ‘hatred, contempt, or ridicule’ test,[32] the ‘shun and avoid’ test[33] and Lord Atkin’s ‘lowering in the estimation of others’ test.[34] With one exception,[35] Tugendhat J found that a ‘threshold of seriousness’ was ‘explicitly or implicitly’ carried in the terms of each test,[36] although in some tests — and notably the ‘hatred’, ‘contempt’ and ‘shun and avoid’ tests — it may be set ‘too high’.[37] The exception was what his Honour hailed as having ‘most often been used’[38] in recent times and that was Lord Atkin’s test. Justice Tugendhat differentiated this definition on the basis ‘that it directs attention to the “estimation” of right-thinking persons, and make [sic] no express mention of any adverse consequences that might result. Attention is directed only to what is in the mind of the publishee’.[39] Interestingly, while the ‘hatred’ and ‘contempt’ tests also appear to be confined to ‘what is in the mind of the publishee’, Tugendhat J was of the view that ‘[i]f the feelings of the publishee are sufficiently strong to be signified by these words, then adverse consequences must be implicit’.[40] It would seem then that ‘lowering in the estimation’ was not considered to be ‘sufficiently strong’ of itself to give rise implicitly to ‘adverse consequences’.[41]

Nevertheless, Tugendhat J was able to find the ‘threshold of seriousness’ for Lord Atkin’s definition in a later passage in his Lordship’s speech in Sim,[42] which read:

[t]hat juries should be free to award damages for injuries to reputation is one of the safeguards of liberty. But the protection is undermined when exhibitions of bad manners or discourtesy are placed on the same level as attacks on character; and are treated as actionable wrongs.[43]

So, while the threshold of seriousness was not discoverable from the oftquoted terms of Lord Atkin’s test alone, it emerged upon a proper reading of his Lordship’s speech, which entailed reading those terms in conjunction with another passage in that speech.[44] Justice Tugendhat conceded that this connection between the two passages had largely escaped attention though not completely.[45] His Honour also acknowledged that the threshold arose in the rediscovered passage ‘by way of example or illustration and not by formulating a test for the degree of seriousness required’.[46]

Justice Tugendhat’s second reason drew upon two external factors or drivers. One is Jameel (Yousef) v Dow Jones & Co Inc[47] and the other is the HRA, which may also be seen to operate indirectly through Jameel.

The HRA is generally understood as ‘incorporating’ the European Convention on Human Rights[48] (‘ECHR’) into the domestic law of the UK,[49] although the proposition may need some qualification.[50] Its effect for present purposes is to require ‘defamation law ... [to] be understood in a rights-compliant manner’.[51] The key ECHR articles in this regard are art 10 (‘[f]reedom of expression’) and art 8 (‘[r]ight to respect for private and family life’), noting that art 8 has been interpreted as extending protection to reputation.[52] Further, it has been held that ‘neither article has as such precedence over the other’.[53]

Jameel concerned an article in the defendant’s online subscription journal, together with a hyperlinked donor list allegedly imputing financial support of an infamous terrorist organisation. The proceedings for defamation were permanently stayed as an abuse of process. Although the imputation was serious, the extent of publication in England was very small. Only five subscribers in England had accessed the material and three of these had some connection with the plaintiff.[54] Applying the same test for setting aside permission to serve tort proceedings outside of the jurisdiction, the English Court of Appeal found that ‘the five publications ... did not ... amount to a real and substantial tort’ and concluded, therefore, that they constituted an abuse of process.[55] Significantly, Lord Phillips MR explained that

[a]n abuse of process is of concern not merely to the parties but to the court. It is no longer the role of the court simply to provide a level playing-field and to referee whatever game the parties choose to play upon it. The court is concerned to ensure that judicial and court resources are appropriately and proportionately used in accordance with the requirements of justice.[56]

Then, in what has become an oft-cited passage, his Lordship declared,

If the claimant succeeds in this action and is awarded a small amount of damages, it can perhaps be said that he will have achieved vindication for the damage done to his reputation in this country, but both the damage and the vindication will be minimal. The cost of the exercise will have been out of all proportion to what has been achieved. The game will not merely not have been worth the candle, it will not have been worth the wick.[57]

In the particular case his Lordship concluded that ‘[i]t would be an abuse of process to continue to commit the resources of the English court, including substantial judge and possibly jury time, to an action where so little is now seen to be at stake’.[58]

Thus emerged the Jameel proportionality principle, as a head of abuse of process, from a cradle comprised of the Civil Procedure Rules 1998 (UK) (‘CPR’)[59] (especially the r 1.1 ‘overriding objective’s requirement for proportionality’) as well as the HRA.[60] This principle became a regular feature of the UK defamation landscape prior to Thornton,[61] although its exercise was regarded as ‘exceptional’ having regard to its serious consequences for claimants.[62] Further, while it is not confined to defamation actions,[63] it may nevertheless have more work to do in that context in the absence of ‘a proportionate small claims procedure’.[64]

In addition to these two primary reasons, Tugendhat J also drew support from a claimed mutual symbiotic relationship between the ‘threshold of seriousness’ and the presumption of damage:

If the likelihood of adverse consequences for a claimant is part of the definition of what is defamatory, then the presumption of damage is the logical corollary of what is already included in the definition. And conversely, the fact that in law damage is presumed is itself an argument why an imputation should not be held to be defamatory unless it has a tendency to have adverse effects upon the claimant.[65]

Further, his Honour relied upon the precedential status of Sim as House of Lords authority, and the absence of any authority to the contrary.[66] Subsequently, in Dell’Olio v Associated Newspapers Ltd,[67] Tugendhat J expressly proclaimed another pillar for the Thornton ‘threshold of seriousness’ in ECHR art 8, in view of case law requiring ‘that intrusions must reach a certain level of seriousness to engage the operation of the Convention’.[68]

Justice Tugendhat then went on to proffer his preferred definition of ‘what is defamatory?’ as follows: ‘the publication of which he complains may be defamatory of him because it [substantially] affects in an adverse manner the attitude of other people towards him, or has a tendency so to do’.[69] This is a revised version of one of Neill LJ’s definitions,[70] which imports the sufficiency of ‘a tendency or likelihood’ regarding the publication’s requisite impact from other tests and authority and makes clear that it carries a ‘threshold of seriousness’ pitched at the level of ‘substantial’.[71] In his Lordship’s view, ‘substantial’ was ‘the lowest threshold that might be envisaged’ from the case law.[72]

The Thornton seriousness threshold was assumed (without deciding the issue) by the English Court of Appeal in Cammish v Hughes[73] and again in Elliott v Rufus.[74] In Cammish, Arden LJ declared:

The law does not provide remedies for inconsequential statements, that is, of trivial content or import. It is necessary that there should be some threshold test of seriousness to avoid normal social banter or discourtesy ending up in litigation and to avoid interfering with the right to freedom of expression conferred by article 10 of the European Convention on Human Rights.[75]

The imputation in that case of ‘a seriously incompetent business person’ was found to have crossed the Thornton seriousness threshold.[76] However, ‘[m]inor criticisms’ of a junior administrative assistant’s work performance,[77] ‘[‘insulting’] references to [a ‘very well known’ woman’s] lifestyle, money and wealth’,[78] and the suggestion that ‘[an employee] sent pompous messages’[79] failed to do so.

Several questions arise concerning the Thornton seriousness threshold. One is, to what is the threshold directed? Is it the gravity of the imputation or this plus other factors including the extent of publication? Support for both contenders may be found in Cammish. While the Court declared that ‘seriousness is a multi-factorial question’, it did not specify a set of relevant factors and its Thornton enquiry seemed confined to the ‘inherent gravity’ of the allegation in question, in accordance with the opening sentence of the passage from the judgment quoted above.[80] Gatley on Libel and Slander specifies a range of relevant factors that extend beyond ‘the nature and inherent gravity of the allegation’ to include, for example, the ‘number of publishees’.[81] But these appear to speak to the more wide-ranging assessments undertaken on a Jameel proportionality inquiry and also under the statutory ‘serious harm’ test. Yet, the Thornton threshold has been described as ‘a specific version of the broader Jameel abuse of process jurisdiction’,[82] which speaks to it having a narrower compass consistent with being confined to an imputation’s gravity, although not necessarily exclusively.

Another question is, what does ‘substantial’ mean in terms of the level at which the threshold is pitched? Some guidance may be obtained from the denotation of claims sought to be excluded by the threshold as ‘trivial’.[83] But, is everything that is not ‘trivial’ necessarily ‘substantial’ or is there a level (or two or more) in between? While the trio of failed meanings reported above demonstrates that trivial imputations will not cross the threshold,[84] they do not of themselves negate the possibility of a marginally-more- than- trivial imputation not failing to do so. A more comprehensive survey of the decided cases may yield a better picture of where the cut-off lies in practice, but it may not consist of a single point. Some convergence may be achieved by articulating more precisely the requisite level, but this is no easy task. Where, for example, do ‘significant’, ‘substantial’ and ‘serious’ sit on a seriousness scale? And how low is ‘trivial’ on this scale and where does it sit in relation to ‘minor’, ‘mild’ and ‘slightly’?

There is also the question of the treatment of ‘ridicule’ in Thornton. Unlike the ‘hatred’ and ‘contempt’ tests, Tugendhat J did not specifically locate a seriousness threshold for the ‘ridicule’ test for reasons that are not clear. Granted ridicule was excluded as an issue,[85] but so was the ‘shun and avoid’ test and yet his Honour still went ahead to locate its seriousness threshold.[86] The capacity of ridicule to register low levels of seriousness would seem to enliven the issue of a minimum ‘ridicule’ threshold, especially considering it had been located elsewhere, including Australia.[87] If a minimum threshold was already inherent in the English ‘ridicule’ test, then it would be reasonable to expect that it would be articulated and its level assessed as part of Tugendhat J’s detailed examination of the English defamation tests, noting in particular that it extended to revealing the ‘implicit’ thresholds carried by ridicule’s two compatriots in the old composite test.[88] Although his Honour’s conclusion regarding the presence of a seriousness threshold was expressed in the all-encompassing terms of ‘whatever definition of “defamatory” is adopted’,[89] the position regarding the ‘ridicule’ minimum threshold is not necessarily without doubt.

III Interrogating the Thornton Reasons

In preparation for the examination of the Australian defamation tests in Part VII, it is necessary to interrogate the Thornton reasons further with respect to their reliance upon certain UK drivers and, specifically, the HRA and its progeny, such as the Jameel proportionality principle. Any such reliance will be inversely proportional to the transportability of the Thornton threshold to jurisdictions lacking in these drivers. The smaller that reliance, the greater the threshold’s transportability, and vice versa. Close inspection of Thornton reasoning reveals that it may not be as dependent on HRA considerations as may appear to be the case at first sight.

In the first place, it is to Lord Atkin’s ‘lowering in the estimation of others’ test that Tugendhat J’s two main reasons[90] appear to be largely directed, noting that of the tests reviewed it was the only test not found to carry a seriousness threshold inherent in its usual terms.[91] It is arguable, then, that if nothing further was required to locate a seriousness threshold in the other tests, HRA considerations could be treated as ‘a check or test of the correctness of’[92] a threshold that already exists as opposed to being integral to the supply of such a threshold for those tests.

Second, of the two main reasons, only one is expressly tied to the HRA and that is the second. It is also tied to that Act implicitly through its reference to Jameel. However, to proclaim, as that reason does, that the HRA and Jameel both require [the seriousness threshold]’[93] is different to asserting that both are necessary to sustain this threshold or that it could not be sustained by one or other alone or, perhaps, even by means other than these authorities. So, it is arguable that the discovery of a seriousness threshold in the absence of one or other of these authorities, or possibly even both, is not inconsistent with this Thornton reason.

Third, while the first limb of the first Thornton reason seems to be impervious to HRA considerations given that Sim was decided well before the introduction of the HRA, old authorities may be reinterpreted in the light of changed conditions and, indeed, Tugendhat J foreshadowed this possibility earlier in his judgment.[94] However, this does not necessarily preclude ‘the true interpretation of Lord Atkin’s speech’,[95] as revealed by Tugendhat J, emerging otherwise than through a HRA lens. Further, the potential of the second limb of this reason to open a portal to HRA infusion by virtue of its reference to Ecclestone v Telegraph Media Group Ltd,[96]

a postHRA decision, is limited by the difficulty of constituting this limb as an essential step in Thornton reasoning. It presents, rather, as additional support for the threshold already established by the first limb and can expect only modest assistance from Ecclestone’s precedential status as a High Court decision.

Fourth, the linking of the seriousness threshold to the presumption of damage in mutually supportive ways[97] also appears, on its face, to be HRA agnostic. It could be argued, however, that closer inspection reveals a small area of HRA sensitivity by way of employing what appears to be a revised version of the presumption. The revision takes account of the concession made in Jameel to the potential for incompatibility between the presumption and ECHR art 10 to arise ‘in the rare case where a claimant[’s] ... reputation has suffered no or minimal actual damage’,[98] albeit not such as to warrant the abolition of the presumption because in that ‘rare case’ the defendant would not be left without any course of action.[99] Nevertheless, the significance of any HRA sensitivity thereby arising may be largely discounted by the subsidiary nature of this Thornton reason.

Consequently, the relevance of HRA considerations to the location of the Thornton seriousness threshold appears to be largely confined to Lord Atkin’s test. This, in turn, appears to turn largely on the correctness of Tugendhat J’s analysis of Lord Atkin’s speech in Sim. If that speech can stand up to scrutiny, then there seems to be no need to resort to an external driver to supply a threshold. The revision of Lord Atkin’s test effectively puts it in the same position as the other tests in that a seriousness threshold may now be found inherent in its terms (albeit now enlarged) and the HRA is only relevant for ancillary purposes. At the highest, it operates as an additional, as distinct from a necessary, reason for the threshold. However, it may be necessary to resort to the HRA and Jameel for the purpose of setting the threshold at ‘substantial’ given that the examples contained in Lord Atkin’s rediscovered passage appear to speak to a lower level of seriousness. If, however, Tugendhat J’s analysis of Lord Atkin’s speech cannot stand up to scrutiny, then something extra would be required to both supply the threshold and set it at ‘substantial’. In this event, HRA considerations and/or the Jameel proportionality principle may operate as fallback reasons, the possibility of which was anticipated by Tugendhat J.[100]

From this analysis, it becomes apparent that the Thornton seriousness threshold may be freed from HRA and Jameel considerations, as regards its existence if not the level at which it is pitched as well. The transportability thereby effectively accorded the threshold regarding jurisdictions in which such considerations do not have a foothold will be of critical significance when it comes to assessing the threshold’s status in Australia.

IV Two Different English Thresholds

The UK has moved on from the ‘twin-track approach [of Thornton and Jameel] to the elimination of trivial defamation claims’,[101] to add a statutory ‘serious harm’ filter. This forms part of the defamation law reform package legislated in Defamation Act 2013 (UK) that commenced operation in 2014.[102] It is important to pause here to emphasise and explain that this is a very different threshold to the Thornton threshold and that the difference is shaping up to be much more than simply a difference in pitch.

Defamation Act 2013 (UK) s 1(1) provides: ‘A statement is not defamatory unless its publication has caused or is likely to cause serious harm to the reputation of the claimant.’ According to the Explanatory Notes, it ‘builds on’ Thornton and Jameel, but ‘raises the bar for bringing a claim’.[103] A stricter test for filtering out trivial defamation claims was sought in the ‘public interest’ of preventing ‘wealthy individuals and organisations ... [from stifling] comment and debate that has no significant impact on their reputation’.[104] It is still early days for this test and its features are still being worked out by the courts. At the time of writing, it has yet to receive appellate scrutiny, although judgment has been reserved by the English Court of Appeal in the appeals from Lachaux v Independent Print Ltd.[105] Nevertheless, several significant differences between the statutory test and its Thornton common law progenitor can be discerned in the developing first instance jurisprudence.

Although the s 1(1) test is acknowledged to be stricter than the Thornton test,[106] any enhanced stringency is not the result of simply changing the dial on the Thornton filter from ‘substantial’ to ‘serious’,[107] but by changing the actual filter. This is not unexpected given its Jameel bloodline. Whereas the Thornton test concentrates on tendency to harm reputation, the focus of the s 1(1) test is reputational harm having been caused or likely to be caused, which requires proof on the balance of probabilities.[108] In other words, Thornton ‘tendency’ is not the same as the combined effect of s 1(1) ‘likely to cause’ plus ‘has caused’.

Consequently, the enquiry into s 1(1) ‘serious harm’ is wider than that concerning the Thornton ‘seriousness’ threshold, extending beyond ‘the defamatory meaning of the words and the harmful tendency of that meaning’ to ‘all the relevant circumstances, including evidence of what has actually happened after publication’.[109] In Lachaux, an article that was found to convey a defamatory meaning of ‘wife-beat[ing]’, did not satisfy the s 1(1) test because the impugned words were not ‘the primary focus’ of the article, constituting a small portion at its end, but also had a small online audience reach.[110] The publication in Cooke v MGN Ltd[111] failed the s 1(1) test because of ‘a prompt and prominent apology’.[112]

Further, whereas Thornton tendency can be assessed from the impugned words,[113] s 1(1) requires the claimant ‘to prove [serious harm] as a fact on the balance of probabilities’.[114] This could be a very onerous burden for the claimant.[115] However, it also appears to be accepted that proof for this purpose does not always necessitate adducing direct evidence of ‘tangible adverse consequences’, but can instead be satisfied by the drawing of inferences.[116] The ‘obvious’ example, given in Cooke, is where ‘a national newspaper with a large circulation wrongly accuses someone of being a terrorist or a paedophile’.[117] However, it is unclear whether proof by inference is also available in ‘less obvious cases’.[118]

There is another, related sense in which the two tests may be said to differ in kind that has been highlighted in the legal commentary — that is, whereas the Thornton test is a test of what is defamatory, the s 1(1) test is a test of whether defamation is actionable.[119] Nevertheless, Descheemaeker has opined that even the ‘actionability’ characterisation ‘appears misguided’.[120]

Flowing from their differences in kind, the two tests differ markedly in their ramifications for settled principles of defamation law. The s 1(1) test has been construed as impliedly abolishing the presumption of damage in defamation law.[121] If it does, then this constitutes a major departure from a long-standing principle of defamation law. It also differs from the approach taken in Thornton where the presumption, albeit in a modified form, was relied upon to support the seriousness threshold. Although recognising the Jameel concession to ‘no or minimal damage’ claims, Tugendhat J was clearly not of the view that it was such as to swallow up the whole presumption.[122]

It also appears that under the s 1(1) test, the defamatory ‘status’ of a publication (in the sense of its actionability) can change over time from defamatory to non-defamatory and also from non-defamatory to defamatory.[123] However, the precise time for determining ‘serious harm’ and, in particular, whether it should be ‘the date on which the claim is issued’[124] or ‘the time at which the issue is determined’,[125] remains unsettled.[126] This ‘shifting’ quality constitutes another departure from ordinary defamation law principles,[127] let alone the change from the time of publication as the critical time for determining actions for defamation.

The two tests also have different costs ramifications. To the extent that the Thornton threshold carries a cost burden, this is arguably much less than that potentially entailed by the s 1(1) requirement to prove serious harm, even accounting for any relief by way of proof by inference. The s 1(1) costs issue is attracting a lot of attention in the case law, with concern expressed regarding the prospect of ‘wasteful duplication’ should a claimant pass the s 1(1) hurdle at a preliminary issues hearing and then proceed to trial.[128] It remains to be seen how far ‘appropriate case management’[129] can remedy the situation.

Accordingly, as the authorities currently stand, moving to the s 1(1) test poses a very different proposition to recognising the Thornton threshold. In addition to the matters mentioned above, and in consideration of those matters, it would almost certainly be regarded as a reform outside the proper power of Australian courts to initiate and implement and require instead parliamentary assistance as it did in the UK. Whether s 1(1) offers a better mechanism for weeding out unworthy claims than the Thornton threshold, let alone the best of all possible mechanisms for this purpose, warrants an investigation beyond the scope of this article. Nevertheless, the foregoing analysis has highlighted two major challenges regarding its adoption in Australia. These are its potential to disrupt well-settled principles of defamation law as well as to increase costs early in defamation proceedings.

V A “Seriousness Threshold” for New Zealand

The proceedings in CPA Australia[130] arose out of ‘criticisms’ made by the acting Chief Executive Officer of one professional accounting body about another professional accounting body in two conference addresses. In refusing a declaration of liability for defamation, the NZ High Court held that the plaintiff corporate body had failed to show relevant pecuniary loss pursuant to the Defamation Act 1992 (NZ).[131] However, in obiter dicta remarks, Dobson J went on to ‘endorse’ a seriousness threshold for the NZ common law tests of defamatory meaning.[132]

His Honour drew support from the NZ equivalent of ECHR art 10 in the New Zealand Bill of Rights Act 1990 (NZ) s 14, the provision of local statutory alternatives to litigation for defamation, and policy considerations around promoting better decision-making on the question of litigating defamation as well as limiting ‘unjustified [free speech] infringements’.[133]

The proceedings in Opai,[134] consisted of several interlocutory applications relating to an action for defamation concerning four workplace documents authored by the claimant’s supervisor. In determining the defamatory quality of the impugned statements, the NZ High Court recognised the Thornton seriousness threshold, and applied the Thornton test of defamatory meaning to find that all the meanings found capable of being conveyed by the publications in question were capable of being defamatory.[135]

However, it is not clear whether the threshold applied in CPA Australia was the Thornton threshold or some other threshold possibly more akin to the English s 1(1) test.[136] In Opai, the Thornton threshold was kept separate from the s 1(1) test, but its adoption was not supported by any explanation or authority. It was simply declared, although similarities between the NZ and English legal and constitutional environments were highlighted in the judgment, including ‘the general trend in New Zealand ... to recognise the [Jameel abuse jurisdiction]’.[137]

The apparent conflation of the two English seriousness thresholds and the absence of supporting analysis coupled with NZ’s different legal context detracts from the persuasiveness of these decisions before Australian courts, even putting aside their precedential status as decisions of single judges of the NZ High Court. Nevertheless, policy considerations appealed to in CPA Australia may resonate in Australia.

VI The Australian Legal and Constitutional Context

Before scrutinising the Australian tests of defamation, it is necessary to briefly explore key aspects of the Australian legal and constitutional context that may bear upon the location of a seriousness threshold.

A Common Tests But Not Without Difference

It is apparent from the earlier articulation of the Australian tests of defamation,[138] that they share a pedigree with the English tests, noting in particular that the Radio 2UE Sydney ‘general test’ is derived from Lord Atkin’s ‘lowering in the estimation of others’ test.[139] However, there are two important local differences.

First, there is authority recognising a seriousness threshold for the ‘ridicule’ test in Australia since at least Ettingshausen v Australian Consolidated Press Ltd,[140] which pre-dates Thornton by some 20 years. In Ettingshausen, a photograph of an Australian rugby league footballer in the shower was published in a magazine without consent. Justice Hunt cast his assessment of the defamatory capacity of one of the pleaded imputations in terms that it was ‘capable of subjecting the entirely blameless plaintiff to a more than a trivial degree of ridicule’.[141] Interestingly, this threshold was not sourced in English common law, but in a United States case.[142] The prior recognition of a seriousness threshold for this test is significant in that it provides a precedent for the recognition of such thresholds in Australia in the absence of HRA considerations and in the presence of a statutory triviality defence.[143] It also it fills an apparent lacuna left by the Thornton analysis in relation to the ‘ridicule’ test. However, while there are first instance authorities accepting the ridicule threshold,[144] they do not number many and the threshold has yet to be tested at the appellate level.[145] This inactivity could be a function of a small number of ridicule cases coming before the courts, but, in any event, the existence of the ridicule threshold does not present a live issue. It may be a different matter, however, regarding the level at which it is pitched. The free speech sensitivities of the 21st century may call for a higher setting. Further, while the recognition of the ridicule threshold may promote receptiveness to the idea of a seriousness threshold in Australia, it does not of itself transport this requirement to the other Australian tests of defamation.

The second local difference of note concerns the Radio 2UE Sydney ‘general test’ and its retirement of Lord Atkin’s ‘right-thinking members of society generally’ in favour of the ‘ordinary reasonable person’ as the relevant hypothetical referee.[146] This was done to disoblige, yet not disallow, this person from having regard to ‘moral or social standards’ when assessing defamatory meaning.[147] This revision does not undermine the possibility of locating a seriousness threshold in the Radio 2UE Sydney ‘general test’. On the contrary, there is post-Thornton English authority to the effect that the ‘ordinary reasonable sensible person’ model ‘implicit[ly]’ carries a seriousness threshold in that such a person would not ‘think the less of [someone]’ because of ‘[m]inor criticisms’.[148] However, the level of seriousness may not be the same as that effected by Lord Atkin’s hypothetical referee.

B Defamation Law and the Australian Federation

In the Australian federal system, defamation law is a matter regulated by the states and territories and not the Commonwealth. However, for more than a decade, Australia has enjoyed (substantially) uniform defamation legislation by virtue of each state and territory enacting legislation in (substantially) identical terms.[149] Nevertheless, this legislation is not a code, and defamation law in Australia consists instead of a mix of legislation and common law.[150]

The challenge of maintaining the uniformity of this law cannot be underestimated and manifests in two important ways in the present context.

First, while an Intergovernmental Agreement between the states and territories facilitates the maintenance of the uniformity of the national defamation legislation,[151] the availability and operation of measures to maintain the uniformity of the common law component of the national scheme is less certain. The well-known precedent rule supporting adherence to intermediate appellate court decisions by equivalent and lower courts concerning the interpretation of national uniform legislation ‘unless convinced that that interpretation is plainly wrong’,[152] does not directly speak to decisions regarding the common law component of such schemes that are not the direct product of interpretation of its legislative component. Further, it is not clear whether and, if so, to what extent the principle of ‘one common law in Australia’[153] may be called upon to assist. Although the principle has been proclaimed by the highest court in Australia with some regularity,[154] it is not without its critics,[155] and its operation may also be challenged in areas, such as defamation law, that do not give rise to frequent appeals to the High Court and so provide frequent opportunities for that court’s guidance and correction.[156]

Second, uniformity of the national defamation scheme may be undermined by lack of uniformity regarding the regulation of aspects falling outside its purview, but which nevertheless affect a defamation claim. The national defamation scheme does not, for example, extend to the case management of such claims before the courts, and legislation and rules regarding this aspect are not otherwise made uniform across all Australian jurisdictions. The condition of varying state/territory legislation has also been highlighted as a challenge to the ‘one common law’ concept.[157]

Consequently, any change by the courts to the common law component of Australian defamation law, such as recognising a seriousness threshold, may struggle to be effected uniformly throughout all the states and territories and especially where it relies upon ‘allied’ laws for support.

C The Absence of a Human Rights Act

As yet, Australia does not have a Bill of Rights and there is no express guarantee of free speech in the Australian Constitution. While the courts have recognised an implied constitutional guarantee of political communication,[158] this is generally regarded as ‘a frail shield’.[159] But even if the implied guarantee does not require a seriousness threshold for the Australian defamatory meaning tests, that of itself is not an impediment to one being found.[160]

At the state/territory level, only two out of the eight jurisdictions participating in the national defamation scheme have enacted statutory bills of rights, both of which contain express guarantees of ‘freedom of expression’ and ‘privacy and reputation’.[161] So even if they could be relied upon to support the recognition of a seriousness threshold in the defamatory meaning tests, their effect is likely to be limited at this stage.

Nevertheless, the absence of this potential catalyst does not necessarily foreclose the discovery of a seriousness threshold in the Australian defamatory meaning tests based on Thornton reasoning. However, the force of this proposition does depends on the correctness of the Thornton revision of Lord Atkin’s test and, to the extent necessary, the availability of the Jameel proportionality principle, or something akin to this principle, in Australia.

D The Emergence of the Bleyer Proportionality Principle

In Bleyer, the New South Wales (‘NSW’) Supreme Court granted a permanent stay of defamation proceedings pursuant to Civil Procedure Act 2005 (NSW) s 67 (‘CPA’) ‘on the grounds that the resources of the court and the parties that will have to be expended to determine the claim are out of all proportion to the interest at stake’.[162] Justice McCallum characterised ‘such disproportionality ... as a species of abuse of process’.[163] The proceedings arose out of search results (including hyperlinked material) generated by the defendant search engine, which allegedly conveyed imputations of criminal conduct, but had only been accessed by three people in Australia, only one of whom had done so after the defendant was notified of the alleged defamation. As the defendant corporation was based in the United States, it was also accepted that damages would be unenforceable.[164]

Bleyer has been hailed as ‘the first application of the Jameel principle in Australia’[165] and, as such, ‘a landmark judgment’.[166] However, it is arguable that the basis of its recognition of proportionality lies in local legal instruments and accepted principle either as an alternative to, if not to the exclusion of, importing Jameel into Australian common law. In other words, rather than simply adopting Jameel, it is argued that Bleyer recognised a home-grown proportionality principle emerging from the cradle comprising the CPA ss 5662, 67 and the Uniform Civil Procedure Rules 2005 (NSW) (‘UCPR’), supplemented by the well accepted ‘proposition ... that the just allocation of the finite resources of the court is a relevant consideration of the court’s authority, at least in civil matters’.[167] This methodology is apparent from her Honour taking as ‘[t]he obvious starting point in determining’ what constitutes an abuse of process, the relevant NSW legislative instruments[168] and expressly eschewing the need ‘to resort to ... [English] law’ for that purpose.[169] By looking to local instruments for guidance, however, McCallum J may be seen to be following Jameel methodology, if not directly importing its substantive results.

The distinction is important because a local version of the proportionality principle will not be susceptible to attack based upon divergences from the UK legal context that saw the emergence of the Jameel principle, such as the absence of an HRA equivalent and different court rules. If the absence of an express reference to proportionality in the UCPR s 56 ‘overriding purpose statement’ is relevant to the emergence of a local proportionality principle, it will be for reasons other than the inclusion of such a reference in the ‘overriding objective’ of the English rules.[170] Recognition of a home-grown proportionality principle will go a long way then to addressing concerns raised in pre-Bleyer case law regarding the application of the Jameel proportionality principle in Australia.[171] But even if Bleyer is interpreted as importing the Jameel proportionality principle, then it is important to note that McCallum J downplayed the significance of HRA considerations in Jameel reasoning by characterising them ‘as a check or test of the correctness of the conclusion it would otherwise reach’,[172] and also highlighting a point made earlier by Basten JA in Bristow v Adams[173] that such considerations themselves may pull in different directions.[174]

Justice Basten also flagged another potential obstacle to the acceptance of a Jameel-type proportionality principle in Australia and that is the presence of the s 33 triviality defence in the uniform defamation legislation.[175] For McCallum J, Basten JA was questioning whether the proportionality principle ‘can comfortably sit alongside the [triviality] defence’,[176] to which her Honour responded in the affirmative, on the basis that:

The source of the power to stay proceedings as an abuse of process is the institutional authority of the court. Defences protect defendants. The existence of a defence to the action is of little avail to the court in protecting the integrity of its own processes (assuming, as I think I should, that includes the fair and just allocation of finite resources).[177]

In other words, the two mechanisms are directed to protecting different interests and stakeholders in a defamation action and, in that sense, operate in different spaces. Judge Gibson, speaking extra-judicially, however, distilled an alternate question from Basten JA’s reference going to the ‘sufficien[cy]’ of the statutory triviality defence.[178] The question might be framed in terms of whether a proportionality principle is necessary given the provision of the statutory triviality defence. The answer to this question turns on the scope of the defence, a matter upon which Judge Gibson expressed some misgivings, especially in the online space.[179] The limited scope of the s 33 defence is explored in more detail in the next section.

The emergence in Australia of a proportionality principle as a case management tool — whether it be a Jameel import or a local product akin to Jameel — is significant for the question of a seriousness threshold for the Australian defamatory meaning tests, because it could act as a driver or catalyst for the location of this threshold if and to the extent additional support is required. It could even also act as a proxy for absent HRA considerations to the extent necessary.

However, the force of these arguments is limited somewhat at this stage by the limited penetration to date of the Bleyer proportionality principle, both upwards in the NSW court hierarchy and outwards across other Australian jurisdictions. The principle has been applied in NSW as a ground for staying or dismissing defamation proceedings as an abuse of process by McCallum J in the Supreme Court,[180] and Gibson DCJ in the District Court.[181] Further, its application in both courts is now supported by Practice Notes.[182] However, it has yet to secure appellate approval. Nevertheless, the NSW Court of Appeal in Ghosh Appeal[183] arguably left the door open in the observation that, ‘the dismissal of proceedings simply upon the basis of a lack of proportionality, without the presence of further factors favouring that result, is likely to be justified only rarely’.[184] To say the proportionality principle operates ‘only rarely’ concedes that it at least operates, albeit to a limited extent. And, more recently the principle did not attract adverse comment from that Court in Toben v Nationwide News Pty Ltd.[185] It might be said then that the principle is currently ‘in park’ at the appellate level in NSW, waiting for an appropriate case that squarely raises the issue.

In the meantime, however, the Bleyer proportionality principle is also struggling to gain a foothold outside NSW. The challenge appears to be one of gaining traction in statutory contexts varying from that prevailing in NSW, especially as regards court legislation and rules, and notably in the absence of an equivalent to CPA ss 56 and/or 60.[186] An interesting question arises, however, as to whether the principle may be disconnected from that context and located more directly in the ‘inherent jurisdiction [of ‘every court’] to stay proceedings which are an abuse of its process’.[187] If it can,[188] then the variation in the state/territory statutory contexts may not pose a problem for the emergence of a proportionality principle in each state and territory, although it will not necessarily guarantee the operation of a uniform proportionality principle in all Australian jurisdictions. Different statutory contexts may, nevertheless, exert different influences on and shape differently the contours of that principle so that different versions of the principle may operate in different states and territories. This disparateness could compromise the principle’s ability to lend support to the seriousness threshold, if such support is required.

The next step in the Smith v Lucht litigation[189] is visited in the following section. However, it may pay now to foreshadow that the plaintiff’s defamation action ultimately failed at trial because the statutory triviality defence succeeded.[190] What conclusion may be drawn from this for present purposes, however, is difficult to say especially given that McGill DCJ in the interlocutory application cast doubt on whether there was relevant disproportionality in the circumstances amounting to an abuse of process.[191] In the case, the defendant called the plaintiff ‘Dennis Denuto’,[192] on three occasions (an email and two oral utterances), to two close relations of the plaintiff, while a party to a family law suit involving one of those relations for whom the plaintiff acted as solicitor. District Court Judge McGill declined to rule out the possibility of damages exceeding a nominal amount having regard to recent Queensland appellate authority.[193] So, to what do the Smith cases speak? Different enquiries, depending on whether Bleyer proportionality or the triviality defence is in issue? Or, essentially same enquiry, about whose outcome reasonable minds may differ?

E The Presence of the Statutory Triviality Defence

A relevant feature of Australian defamation law, which is absent from the UK’s is the statutory triviality defence.[194] Its ‘apparent purpose ... [is] to ... discourage actions for trivial defamation’,[195] and is cast in terms that are similar to those of its immediate precursor in the Defamation Act 1974 (NSW) s 13, as follows: ‘It is a defence to the publication of defamatory matter if the defendant proves that the circumstances of publication were such that the plaintiff was unlikely to sustain any harm.’[196]

The central element of the defence is ‘the circumstances of publication’.[197] These ‘include (inter alia) the content of the publication, the extent of the publication, the nature of the recipients and their relationship with the applicant ... [but] not ... the previous bad reputation of a plaintiff’.[198] Nor, it seems, do the relevant circumstances ‘include individual characteristics of the plaintiff, at least insofar as they are unknown to the defendant at the time of publication’.[199] The typical combination of circumstances said to engage the defence is ‘where a slightly defamatory statement is made in jocular circumstances to a few people in a private home’,[200] although it has been emphasised that it is not confined to those circumstances.[201] To succeed, the defendant must show that at ‘the time of publication’,[202] these circumstances displayed a ‘proneness to cause harm’,[203] as opposed to ‘actual’ harm.[204] The requisite likelihood is ‘“the absence of a real chance” or the “absence of a real possibility of harm”’ and not the lesser “probably did not suffer harm”.[205] And the requisite quantum of s 33 ‘harm’ is ‘“harm” at all’ as opposed to ‘great or substantial harm’.[206] The stringency of the resulting burden of proof on the defendant is well recognised,[207] and so it is not surprising that the defence has not enjoyed a great track record,[208] although it has managed to secure two wins of late, both of which have survived appellate scrutiny.[209] The facts and decision in Smith Appeal have already been outlined. In Barrow v Bolt,[210] the defence succeeded in relation to an email calling the plaintiff ‘a vexatious litigant’,[211] sent by a journalist to the managing editor of his employer in response to a complaint by the plaintiff about the journalist’s blog to the Australian Press Council (‘APC’), which was then forwarded to the executive director of the APC.

To examine the potential interplay between a seriousness threshold and the s 33 defence, the two questions employed earlier for the Bleyer proportionality principle will again frame the discussion. The question of ‘comfortable fit’ with the s 33 defence is likely to be more acute for the seriousness threshold than for the Bleyer/Jameel proportionality principle given that they would sit in the same plane, being elements, broadly speaking, of the same action, thereby facilitating potential for incursions into each other’s space. Complicating the task is uncertainty attending key aspects of both mechanisms. Questions concerning the range of considerations relevant to the threshold enquiry, as well as its precise setting, have already been flagged. For the s 33 defence, there has been uncertainty regarding the meaning of s 33 ‘any harm’ and, in particular, whether it extends beyond reputational injury to include hurt feelings.[212] It is arguable that the recent majority decision of the Queensland Court of Appeal in Smith Appeal[213] in favour of the narrow view does not settle the issue given there appear to be reasonable grounds for arguing that the ruling is obiter dicta.[214] Further, it is arguable that the broader view is not without foundation, especially in view of the insertion of ‘any’ before ‘harm’ in s 33.[215] It may be that the proper construction of s 33 ‘harm’-type will not be put beyond doubt unless and until there is an appeal to the High Court (absent legislative clarification in the meantime).

If s 33 ‘harm’-type is interpreted to extend to hurt feelings, then the defence risks becoming ‘virtually unworkable’,[216] in which case compatibility with the threshold would largely become a non-issue. If s 33 ‘harm’-type does not extend to hurt feelings, compatibility will depend in the first place on whether or not threshold considerations are confined to the gravity of the imputation. If they are, then there could be an overlap with the s 33 defence given that gravity is also one of the recognised ‘circumstances of publication’. It is conceivable, for example, that the defence may not be engaged in its classic setting, as ‘a slightly defamatory statement’ would arguably be knocked out by the threshold before the question of defences arises. In this scenario, the two mechanisms are essentially doing the same work and so it might be asked, why have the threshold? One answer is that it may offer a costs-saving benefit given that it operates early in the proceedings, as opposed to waiting to the end to filter out the trivial claim. However, a threshold confined to gravity will not deprive the defence of all work. Consider, for example, a serious imputation published to a handful of people. It will definitely cross the threshold (meaning the claim remains on foot) as its small audience will be ignored and may also pass the s 33 test (meaning the defence will succeed and so the claim will ultimately fail). It is accepted that the defence can succeed for ‘serious’ as well as ‘slightly defamatory’ imputations, depending on the other ‘circumstances of publication’,[217] the key one of which in the present example is the very limited audience reach.

If the threshold considerations extend beyond the gravity of the imputation, then the potential for overlap with the s 33 defence is much greater — although the extent to which threshold considerations would coincide with s 33 ‘circumstances’ is not clear. Nevertheless, the two mechanisms will be distinguishable at least in terms of their burden of proof. Whereas it would be for the plaintiff to establish the requisite threshold, it is for the defendant to establish the defence. Moreover, the two mechanisms could yield different results. Take, for example, a more-than-trivial imputation conveyed to a small audience. The limited publication may be enough to prevent the imputation crossing the threshold (meaning that the claim would fail). However, it is conceivable that this publication will also fail the s 33 test (meaning that the defence will fail and the claim may succeed). This is because of the demanding burden of proof required by the defence compared with that of the threshold. Whereas the threshold is concerned with ‘tendency to substantially harm’, the defence requires ‘unlikelihood of any harm at all’. In other words, an imputation that has the tendency to cause some harm, but less than substantial harm, will not cross the threshold (meaning that the claim will fail), but it will also not satisfy the s 33 test (meaning that the claim will succeed).

So, what conclusions can be drawn from this state of affairs? It could be argued that it shows that the threshold does not sit well with the s 33 defence, which would tend against recognising it. Alternatively, it could be seen as grounds for confining the threshold to the gravity of the imputation. A third option, relevant for the second question, is that any problem lies with the defence and, in particular, that its bar is set too high.

It is worth emphasising that even if s 33 ‘harm’-type is confined to reputational harm, the defence is still pitched at a ‘very high’ level.[218] There are other factors contributing to this high setting, apart from its harm quantum. Even ‘the circumstances of publication’ element can have an unduly limiting effect depending on the weight placed by the court on the various circumstances. The defence has failed, for example, in the face of extremely limited publication, including to one person only,[219] although that is not to say that such publication can never be serious.[220] Further, the traditional approach to some of these ‘circumstances’, and notably communication in permanent form as well as extremely wide-audience reach and the potential for a healthy ‘risk of repetition or the “grape-vine effect”’,[221] may pose significant challenges for the defence on social network sites.[222] The recent s 33 wins in the appeals of Barrow and Smith regarding emails may not necessarily have had to meet all of these challenges.[223] If the s 33 defence has a narrow scope of operation, then the potential for incompatibility with a seriousness threshold is considerably reduced and, with that, grounds for resisting the recognition of the threshold based on the s 33 defence.

Before leaving the first question of compatibility, it is worth making explicit by way of consolidation of the above discussion that the threshold/defence compatibility will also be affected by the level at which the seriousness threshold is pitched. Generally speaking, the higher the threshold, the smaller the area of operation left for the s 33 defence.

On the second question of necessity, a narrow scope of operation for the s 33 defence not only leaves room but may also invite, if not call, for the introduction of other mechanisms to help weed out unworthy defamation claims. So, there are grounds for answering this question in the affirmative. Ultimately however, it will depend on how unworthy is conceived in the 21st century and, in particular, whether the rise of new communication technologies, such as social network sites, and the way people use them, requires revisiting traditional notions of ‘triviality’.

VII Scrutinising the Australian Defamatory Meaning Tests for a Seriousness Threshold

The stage is now set to scrutinise the Australian tests of defamation for a seriousness threshold and to determine whether external assistance is needed and available — and, if so, to what extent.[224]

The finding in Thornton that the ‘shun and avoid’ test as well as the ‘hatred’ and ‘contempt’ tests each carry a seriousness threshold inherent in their terms will probably not give rise to much controversy in Australia, notwithstanding the absence of a HRA-equivalent in all of the Australian jurisdictions. Nor would the finding in Thornton that these thresholds are probably pitched ‘too high’. Even if these tests were viewed through an HRA lens in Thornton, anything thereby added is arguably negligible to what is yielded on a bare reading of these tests given the nature and strength of their terms.

As noted above in Part VI(A), a seriousness threshold has already been located in the Australian ‘ridicule’ test by Australian case law and the more pressing question is whether it is pitched at an appropriate level. It is not clear, for example, what ‘more than a trivial degree of ridicule’ means and whether, for example, it equates with ‘substantial’ or sits somewhere below this level and, if so, how far below. In addition, concern has been expressed that the current setting may not be sensitive enough to exclude satire,[225] although that is not to say that there is necessarily consensus as to what satire is and whether all satire should be protected and protected equally.

While still featuring regularly in defamation pleadings, these old definitions have relatively little work to do outside of the Radio 2UE Sydney ‘general test’. That test is the test ordinarily applied in Australia and so is the key test to interrogate for a seriousness threshold. And it is in relation to this test that controversy around the existence of a seriousness threshold is likely to arise. Like its English progenitor (Sim),[226] a seriousness threshold is not immediately apparent on its terms, at least not in the Thornton sense of ‘adverse consequences’.[227] While deployment of the ‘ordinary reasonable person’ as the hypothetical referee may incorporate a seriousness threshold, as suggested in Daniels,[228] it may not coincide with the Thornton seriousness threshold without an ‘adverse consequences’[229] dimension. Controversy is also likely to attend the impact of differences between the Australian and UK legal and constitutional contexts, especially around the existence or otherwise of drivers for the recognition of a seriousness threshold.

As explained in Part VI(C) above, a Thornton-style seriousness threshold can exist in the absence of both the HRA and the Jameel proportionality principle depending on the strength of Tugendhat J’s analysis of Lord Atkin’s speech in Sim. While Lord Atkin’s rediscovered passage has not escaped judicial attention in Australia,[230] it is arguable that its capacity to ground the threshold is undermined by ambiguity. Lord Atkin appears to be setting a threshold in this passage, but it is not clear to what this threshold is directed. ‘[E]xhibitions of bad manners or discourtesy’[231] could sound in defamation, but they may not. So, is the threshold directed to delineating between defamation and no defamation (without regard to seriousness) or between different grades of defamation? The latter assumes that there is defamation in the first place. The approach taken in the Australian case referring to this passage is consistent with the former interpretation.[232]

To the extent that Tugendhat J’s analysis of Sim may be challenged, however, the significance of the absence of the HRA may be weakened in Australian eyes by local judicial acknowledgement that its relevant considerations, and notably freedom of expression (ECHR art 10) and the right to a fair trial (ECHR art 6), may pull in different directions.[233] Moreover, the absence of this UK driver may be compensated to a significant extent by the emergence in Australia of a Jameel-like proportionality principle, in the form of the Bleyer proportionality principle, as part of a court’s case management toolbox. To reiterate, saying that both HRA and Jameel require a seriousness threshold in the UK does not necessarily mean that both (or their equivalents) are required to sustain a seriousness threshold in another jurisdiction. One may suffice (and on both counts). It may be argued, then, that the emergence of the Bleyer proportionality principle in Australia requires the recognition of a seriousness threshold for the Radio 2UE Sydney ‘general test’. However, even if the Bleyer proportionality principle is capable of acting as such a catalyst, its strength in that regard is weakened at this stage by the absence of appellate approval, as well as its limited jurisdictional reach — the extension of which beyond NSW is not without difficulty.

An argument based on the presumption of damage per se has the potential to resonate in Australia given that that presumption is a recognised ‘basic principle of defamation law’.[234] However, as framed in Thornton, based on an apparently HRA/Jameel moderated version of the presumption, it may lose traction before Australian courts, at least at this early stage of Jameel-equivalent recognition. Further, even if the s 1(1) ‘serious harm’ requirement does abolish the presumption of damage in the UK, that abolition does not, of itself, automatically extend to Australian law. Moreover, it does not necessarily mean that adopting the Thornton seriousness threshold in Australia will have the same effect because, as demonstrated in Part IV above, they are different tests. Although removing one of the planks of Thornton reasoning could be destabilising to its threshold, any such effect is likely to be limited given that it was only a subsidiary plank and confined to English law.

The steps taken in NZ towards locating a ‘seriousness threshold’ in its common law tests of defamatory meaning will pique interest in Australia — although there are limitations on the persuasiveness of their guiding authorities for Australia at this stage. Nevertheless, the broad tenor of the policy considerations advanced in CPA Australia,[235] especially regarding the savings (and not just financial) to be made from weeding out unworthy defamation claims, are of relevance in Australia and sound in the objects of the Australian uniform defamation legislation ‘[to avoid] plac[ing] unreasonable limits on freedom of expression’, ‘to provide effective and fair remedies [for reputational harm]’, and ‘to promote speedy [dispute resolution]’.[236] While these considerations also speak to the s 33 defence, there is arguably an important difference in that the savings to be made from a seriousness threshold are potentially greater given that it would operate as a front-end mechanism and save a trip to the defences in an unworthy claim. A possible counterargument might be made, based on the potential for duplication arising from the use of multiple mechanisms in certain circumstances and the associated ‘costs’ in terms of additional complexity and expense. However, to the extent that the s 33 defence is limited in its scope, the undesirable consequences of any duplication will be reduced.

Assuming that the Radio 2UE Sydney ‘general test’ is found to carry a threshold of seriousness, the next question is at what level is it pitched? In particular, is it the same as in Thornton (of ‘substantial’), or lower? To the extent that Lord Atkin’s rediscovered passage is capable of supplying the threshold for this test, its level is arguably at least the same as that conveyed by the examples provided in that passage. But the highest this might be put is ‘more than trivial’. That would make it consistent with the threshold already recognised for the ‘ridicule’ test in Australia. To the extent that Bleyer may be relied upon to support a seriousness threshold, the question becomes whether this may also raise the bar in the way that the Jameel proportionality principle may have done for the Thornton seriousness threshold. An affirmative answer to this question seems to be less forthcoming if, as contended by this article, Bleyer gives rise to a home-grown proportionality principle, as opposed to importing the Jameel proportionality principle into Australian common law having regard to its ‘real and substantial tort’ dimension.

Finally, although the position is not certain, it is arguable that locating a seriousness threshold in the Australian defamatory meaning tests is compatible with the s 33 defence, given its limited availability. However, compatibility may entail some accommodation being made to the threshold in the form of confining its consideration to the gravity of the imputation and possibly also moderating the level at which it is pitched. The lower the threshold level, the less likely it is to be incompatible with the s 33 defence (in the sense of depriving it of work). But, by the same token, the less likely it is to be able to supplement the s 33 defence in the task of filtering out unworthy defamation claims and so stake a claim to its recognition on the basis of necessity.

VIII Final Reflection

A tenable argument may be made that the Australian common law tests of defamation already carry a Thornton-type seriousness threshold inherent in their terms notwithstanding the absence of an HRA equivalent and also notwithstanding the existence of the s 33 triviality defence. Should additional support be required to locate the threshold for the Radio 2UE Sydney ‘general test’, then the Bleyer proportionality principle may be called upon — albeit arguably not to its full potential at this time. Certain policy considerations may also assist, especially around the advantages of front-end mechanisms compared with back-end mechanisms for limiting unworthy defamation claims.

Assuming a seriousness threshold can be recognised, what difference will it make? If the threshold is inbuilt, and has essentially always been there (albeit undeclared), then it may be argued that to recognise it as such will have little, if any, practical impact in terms of the outcome of cases. Nevertheless, it may be countered that its express recognition is important in several ways. First, in making transparent the true nature and operation of the defamatory meaning tests — and, in particular, that some imputations, though otherwise defamatory, may be eliminated for failure to meet the threshold. Second, in enhancing the serviceability of the threshold by enabling greater threshold sensitivity on the part of both the parties and the courts. Third, in opening the threshold to scrutiny, evaluation and the possibility of review and reform if considered necessary. And in that regard, it is easier to argue for a change to a threshold that exists than to argue for its introduction in the first place. Fourth, and importantly, the recognition of a seriousness threshold may invite a remodelling of the Radio 2UE Sydney ‘general test’, in the way it did for Neill LJ’s Berkoff test in Thornton, to make both the existence of the threshold and the level at which it is pitched more apparent.

The question, however, is whether the discovered threshold goes far enough to contribute effectively to addressing the policy imperatives around the limitation of unworthy defamation claims. Although the setting for the Radio 2UE Sydney ‘general test’ is not completely certain, it is unlikely to be higher than ‘more than trivial’. A higher level is not immediately apparent from either Lord Atkin’s rediscovered passage in Sim or the Bleyer proportionality principle. Whether it should be higher invites consideration of a range of factors, including: the high costs of defamation litigation; over-taxed court resources; and the apparent trends of ordinary people suing ordinary people in defamation[237] and for litigants to proceed self-represented.[238] But, at the heart of the question is what constitutes ‘unworthy’ in this context? Traditionally, the metric used has been expressed in terms of ‘triviality’, but that is a multi-factorial concept that does not speak clearly at a universally accepted pitch. It may also need reconceptualising in order to remain relevant in the new communication paradigms provided by online technologies. In addition, an ‘unworthy’ claim may not necessarily relate to ‘unworthy’ speech. A seriousness threshold may not only weed out ‘trivial’ claims, but also claims relating to valuable speech. The challenge, however, is to calibrate the threshold so that it can be sensitive to this difference.

If there is an Australian seriousness threshold, but it is pitched too low to be able to address effectively 21st century needs in relation to unworthy defamation claims, then simply recognising it may not be enough to stave off the need for (further) reform. It may even provoke reform in the interests of reducing the unnecessary duplication, complexity and associated costs potentially attending what would be a unique ‘[triple]-track approach to the elimination of trivial defamation claims’ for Australia,[239] consisting of a Thornton-type seriousness threshold, the Bleyer proportionality principle and the s 33 triviality defence. But would an appropriately pitched threshold fare any better? In particular, would it stave off calls to adopt the English statutory ‘serious harm’ test? It is difficult to come to a concluded view about this while critical features of the new statutory test are still being worked out by the English courts. But as the authorities currently stand, adopting it poses a very different proposition to recognising a Thornton-style threshold, from the method of its implementation to its ramifications for defamation principles, not to mention its impact on litigation costs. The degree of increased strictness over the Thornton threshold also remains to be seen. It may turn out that a statutory ‘serious harm’ test is not ‘worth the candle’[240] for Australia, at least not in its present form.

In the meantime, however, this article contributes to the local conversation around improving the law’s capacity to exclude unworthy defamation claims by shining a light on what appears to be the largely overlooked option of recognising a seriousness threshold in the Australian defamation tests,[241] operating either with or without the support of the emerging Bleyer proportionality principle. This revelation is important, given the shortcomings of the primary mechanism for filtering out unworthy defamation claims in the form of the s 33 triviality defence. Something extra is required in the interests of maintaining a proper balance between freedom of speech and protection of reputation, as well as avoiding placing an unreasonable burden on the administration of justice in terms of resources, costs and delay. The analysis in this article will also assist with evaluating the need for adopting additional or alternative filters, such as the English statutory ‘serious harm’ requirement, by revealing that the proper basis from which to make such assessments is one that includes a seriousness threshold, as opposed to no threshold at all.


[∗] Senior Lecturer, Faculty of Law, University of Technology Sydney, Australia.

I would like to thank the anonymous reviewers for their valuable comments, points and suggestions for improving this article and acknowledge their contributions generally. I also thank and acknowledge the Sydney Law Review editorial team for their valuable suggestions and assistance. However, any errors or other deficiencies remaining are my responsibility.

[1] Civil Law (Wrongs) Act 2002 (ACT) ch 9; Defamation Act 2005 (NSW); Defamation Act 2006 (NT); Defamation Act 2005 (Qld); Defamation Act 2005 (SA); Defamation Act 2005 (Tas); Defamation Act 2005 (Vic); Defamation Act 2005 (WA). Future references to this legislation will be to provisions in the Defamation Act 2005 (NSW), unless otherwise indicated.

[2] Defamation Act 2005 (NSW) s 6; Born Brands Pty Ltd v Nine Network Australia Pty Ltd (No 6) [2013] NSWSC 1651 (12 November 2013) [63].

[3] [2009] HCA 16; (2009) 238 CLR 460 (‘Radio 2UE Sydney’).

[4] Ibid 467 (French CJ, Gummow, Kiefel and Bell JJ).

[5] Kim Gould, ‘The Common Law Tests of Defamatory Meaning in the Wake of Radio 2UE Sydney v Chesterton(2015) 41(1) Australian Bar Review 44; contra Des Butler and Sharon Rodrick, Australian Media Law (Lawbook, 5th ed, 2015) 42; Alastair Mullis and Richard Parkes (eds), Gatley on Libel and Slander (Thomson Reuters, 12th ed, 2013) 36–7 n 32.

[6] [2010] EWHC 1414; [2011] 1 WLR 1985 (‘Thornton’).

[7] [2015] NZHC 1854 (6 August 2015) (Dobson J) (‘CPA Australia’).

[8] [2016] NZHC 3004 (13 December 2016) (Bell AsJ) (‘Opai’).

[9] [2017] SASCFC 48 (15 May 2017) (‘Lesses’).

[10] Ibid [122]–[125].

[11] Ibid [125].

[12] Ibid.

[13] See also David Rolph, ‘Triviality, Proportionality and A Threshold of Serious Harm’, Gazette of Law and Journalism, 25 August 2017 <http://www.glj.com.au/2911-article> .

[14] Lange v Australian Broadcasting Corporation [1997] HCA 25; (1997) 189 CLR 520, 568 (‘Lange’).

[15] c 26. See, eg, Judge Judith Gibson, ‘The Trouble with Australian Defamation Law – 1’, Gazette of Law and Journalism, 28 August 2015 <http://www.glj.com.au/2511-article> Richard Coleman, ‘The Trouble with Australian Defamation Law – 4’, Gazette of Law and Journalism, 2 September 2015 <http://www.glj.com.au/2517-article> Phoebe J Galbally, ‘A “Serious” Response to Trivial Defamation Claims: An Examination of S 1(1) of the Defamation Act 2013 (UK) from an Australian Perspective’ (2015) 20(3) Media and Arts Law Review 213, especially 250–1; ‘Practice Notes – Matthew Lewis’, Gazette of Law and Journalism, 10 May 2016 <http://www.glj.com.au/2636-article> .

[16] Cook v Cook [1986] HCA 73; (1986) 162 CLR 376, 390.

[17] See Defamation Act 2005 (NSW) s 33.

[18] Parmiter v Coupland [1840] EngR 168; (1840) 6 M & W 105, 342 (Parke B) (‘Parmiter’). Its three limbs can operate as independent tests, and these will be referred to as the ‘hatred’ test, the ‘contempt’ test and the ‘ridicule’ test: see, eg, Ettingshausen v Australian Consolidated Press Ltd (1991) 23 NSWLR 443 (‘Ettingshausen Trial’); Berkoff v Burchill [1996] EWCA Civ 564; [1996] 4 All ER 1008 (‘Berkoff’).

[19] Youssoupoff v Metro-Goldwyn-Mayer Pictures Ltd (1934) 50 TLR 581, 587 (‘Youssoupoff’).

[20] It is arguable that of the two older tests, only the ‘shun and avoid’ test survives Radio 2UE Sydney [2009] HCA 16; (2009) 238 CLR 460: Gould, ‘The Common Law Tests of Defamatory Meaning’, above n 5, especially 66–77.

[21] Bleyer v Google Inc [2014] NSWSC 897; (2014) 88 NSWLR 670 (‘Bleyer’).

[22] [2015] NZHC 1854 (6 August 2015).

[23] Note, Defamation Act 2013 (UK) s 1(2) deals with a particular application of the s 1(1) test to ‘a body that trades for profit’, the relevance of which in Australia is significantly limited by the restriction on ‘[c]ertain corporations’ suing for defamation pursuant to Defamation Act 2005 (NSW) s 9 and so will not be examined in this article.

[24] [2010] EWHC 1414; [2011] 1 WLR 1985.

[25] This is the ‘practice’ of according ‘interviewees the right to read what [the author] proposed to say about them and alter it’: ibid 1989 referring to Statement of Claim (6.2).

[26] Ibid 2010–1.

[27] Ibid 2008.

[28] Ibid 2002.

[29] Ibid 2009.

[30] Ibid 2008–9.

[31] These definitions were drawn from the listing given by Neill LJ in Berkoff [1996] EWCA Civ 564; [1996] 4 All ER 1008: ibid 1994–6.

[32] Is the publication ‘calculated to injure the reputation of another, by exposing him to hatred, contempt, or ridicule’?: Parmiter [1840] EngR 168; (1840) 6 M&W 105, 342.

[33] ‘[Whether the matter] tends to make the plaintiff be shunned and avoided’?: Youssoupoff (1934) 50 TLR 581, 587 (Slesser LJ).

[34] ‘[W]ould the words tend to lower the plaintiff in the estimation of right-thinking members of society generally?’: Sim v Stretch [1936] 2 All ER 1237, 1240 (‘Sim’).

[35] That is, putting aside the ‘ridicule’ limb, which is discussed below.

[36] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2002.

[37] Ibid 2002, 2006, 2008.

[38] Ibid 2002.

[39] Ibid.

[40] Ibid.

[41] Ibid.

[42] Ibid 2002, 2005.

[43] [1936] 2 All ER 1237, 1242.

[44] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2004–5, 2007, 2008.

[45] Ibid 2004–5.

[46] Ibid 2005, 2008.

[47] [2005] EWCA Civ 75; [2005] QB 946 (‘Jameel’).

[48] Convention for the Protection of Human Rights and Fundamental Freedoms European Convention on Human Rights, opened for signature on 4 November 1950, 213 UNTS 221 (entered into force 3 September 1953) (‘ECHR’).

[49] See HRA c 42, s 1.

[50] Andrew Nicol, Gavin Millar and Andrew Sharland, Media Law and Human Rights (Oxford University Press, 2nd ed, 2009) 42–3; N A Moreham and Mark Warby (eds), Tugendhat and Christie: The Law of Privacy and the Media (Oxford University Press, 3rd ed, 2016) 17 n 73.

[51] Mullis and Parkes, above n 5, 13 referring specifically to HRA ss 3(1), 6(1).

[52] See, eg, Flood v Times Newspapers Ltd [2012] UKSC 11; [2012] 2 AC 273, 290.

[53] In Re S (A Child) (Identification: Restrictions on Publication) [2005] 1 AC 593, 603 (emphasis in original). See also, In Re Attorney-General’s Reference (No 3 of 1999) [2009] EWCA Civ 1365; [2010] 1 AC 145, 164.

[54] Jameel [2005] EWCA Civ 75; [2005] QB 946, 957 (‘members of the claimant’s camp’).

[55] Ibid 970.

[56] Ibid 965.

[57] Ibid 969–70.

[58] Ibid 970.

[59] SI 1998/3132.

[60] Jameel [2005] EWCA Civ 75; [2005] QB 946, 966.

[61] See Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2003–4 and the examples cited at [62]–[63].

[62] Ames v The Spamhaus Project Ltd [2015] EWHC 127; [2015] 1 WLR 3409, 3419 (‘Ames’).

[63] Ibid, citing Sullivan v Bristol Film Studios Ltd [2012] EMLR 27 (‘Sullivan’).

[64] Jameel [2005] EWCA Civ 75; [2005] QB 946, 970. And see Sullivan [2012] EMLR 27, 663.

[65] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009.

[66] Ibid 2008.

[67] [2011] EWHC 3472 (QB) (20 December 2011) (‘Dell’Olio’).

[68] Ibid [13]–[15], quoting R (Gillan) v Commissioner of Police for the Metropolis [2006] UKHL 12; [2006] 2 AC 307, 344 and see other authorities cited in [15].

[69] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009–10 (emphasis and square brackets in original).

[70] See Berkoff [1996] EWCA Civ 564; [1996] 4 All ER 1008, 1018.

[71] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2003, 2009.

[72] Ibid 2009.

[73] [2012] EWCA Civ 1655; [2013] EMLR 13, 302 (‘Cammish’).

[74] [2015] EWCA Civ 121; [2015] EMLR 17, 401, 404.

[75] [2012] EWCA Civ 1655; [2013] EMLR 13, 302.

[76] Ibid 302–3.

[77] Daniels v British Broadcasting Corporation [2010] EWHC 3057 (QB) (24 November 2010)

[48]–[50] (‘Daniels’).

[78] Dell’Olio [2011] EWHC 3472 (QB) (20 December 2011) [28], [32].

[79] Briggs v Jordan [2013] EWHC 3205 (QB) (24 October 2013) [21].

[80] Cammish [2012] EWCA Civ 1655; [2013] EMLR 13, 302–3.

[81] Mullis and Parkes, above n 5, 39, but cf 41.

[82] Alastair Mullis and Andrew Scott, ‘Tilting at Windmills: The Defamation Act 2013’ (2014) 77(1) Modern Law Review 87, 105.

[83] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2008.

[84] See above nn 77–80 and accompanying text.

[85] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 1997.

[86] Ibid 1997, 2002.

[87] See below nn 140–42 and discussion in accompanying text.

[88] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2002.

[89] Ibid 2008 (emphasis added).

[90] Set out in the passage quoted at above n 30 and accompanying text.

[91] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2002.

[92] To borrow the characterisation by McCallum J in Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 678 of the role played by HRA considerations in the emergence of the Jameel proportionality principle, see below n 172 and discussion in accompanying text.

[93] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009 (emphasis added).

[94] Ibid 1996.

[95] Ibid 2008.

[96] [2009] EWHC 2779 (6 November 2009) (‘Ecclestone’).

[97] See passage quoted in text accompanying n 65 above.

[98] Jameel [2005] EWCA Civ 75; [2005] QB 946, 962, referred to in Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009.

[99] Jameel [2005] EWCA Civ 75; [2005] QB 946, 962.

[100] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009.

[101] Lachaux v Independent Print Ltd [2015] EWHC 2242; [2016] QB 402, 420 (Warby J) (‘Lachaux’).

[102] Defamation Act 2013 (UK) s 1(1).

[103] Explanatory Notes, Defamation Act 2013 (UK) [11].

[104] Joint Committee on the Draft Defamation Bill, Draft Defamation Bill, House of Lords Paper No 203, House of Commons Paper No 930-1, Session 2010-12 (2011) 25.

[105] [2015] EWHC 2242; [2016] QB 402.

[106] Ibid 416; Theedom v Nourish Training [2015] EWHC 3769; [2016] EMLR 10, 192 (‘Theedom’).

[107] ‘Serious’ is not defined in the Act, and is taken to be ‘an ordinary word in common usage’: see, eg, Cooke v MGN Ltd [2014] EWHC 2831; [2015] 1 WLR 895, 905 (‘Cooke’).

[108] Lachaux [2015] EWHC 2242; [2016] QB 402, 419–20, 424.

[109] Ibid 424.

[110] Ibid 409, 433, 445–6.

[111] [2014] EWHC 2831; [2015] 1 WLR 895.

[112] Ibid 901–2, 907.

[113] Lachaux [2015] EWHC 2242; [2016] QB 402, 412, 422.

[114] Ibid 419; Theedom [2015] EWHC 3769; [2016] EMLR 10, 186.

[115] Ames [2015] EWHC 127; [2015] 1 WLR 3409, 3424.

[116] See, eg, Cooke [2014] EWHC 2831; [2015] 1 WLR 895, 907 (quoting from the defendants’ submission); Ames [2015] EWHC 127; [2015] 1 WLR 3409, 3424; Theedom [2015] EWHC 3769; [2016] EMLR 10, 186, 193.

[117] Cooke [2014] EWHC 2831; [2015] 1 WLR 895, 907.

[118] For the view that it is, see Lachaux [2015] EWHC 2242; [2016] QB 402, 422.

[119] Mullis and Parkes, above n 5, 40–1; Erik Descheemaeker, ‘Three Errors in the Defamation Act 2013(2015) 6(1) Journal of European Tort Law 24, 27. Also, the fact that this seems to have been appreciated during the course of parliamentary debate on the Defamation Bill has not gone unnoticed: Descheemaeker.

[120] Descheemaeker, above n 119.

[121] Lachaux [2015] EWHC 2242; [2016] QB 402, 423.

[122] Thornton [2010] EWHC 1414; [2011] 1 WLR 1985, 2009, referring to Jameel [2005] EWCA Civ 75; [2005] QB 946, 962.

[123] Cooke [2014] EWHC 2831; [2015] 1 WLR 895, 903, 907; Lachaux [2015] EWHC 2242; [2016] QB 402, 424.

[124] Cooke [2014] EWHC 2831; [2015] 1 WLR 895, 903.

[125] Lachaux [2015] EWHC 2242; [2016] QB 402, 424.

[126] Sobrinho v Impresa Publishing SA [2016] EWHC 66; [2016] EMLR 12, 251.

[127] Lachaux [2015] EWHC 2242; [2016] QB 402, 424.

[128] See, eg, Theedom [2015] EWHC 3769; [2016] EMLR 10, 194.

[129] Lachaux [2015] EWHC 2242; [2016] QB 402, 424.

[130] [2015] NZHC 1854 (6 August 2015).

[131] Ibid [101]–[103], [222].

[132] Ibid [120].

[133] Ibid [107], [114], [115], [119].

[134] [2016] NZHC 3004 (13 December 2016).

[135] Ibid [32]–[33], [42], [53].

[136] See CPA Australia [2015] NZHC 1854 (6 August 2015) [106], [120]–[121], [222]. The latter view expressed above accords with the interpretation of this case in Opai [2016] NZHC 3004 (13 December 2016) [77].

[137] Opai [2016] NZHC 3004 (13 December 2016) [77]–[78].

[138] See n 4 above and accompanying text and nn 32–3 above.

[139] Radio 2UE Sydney [2009] HCA 16; (2009) 238 CLR 460, 466, 477.

[140] (1991) 23 NSWLR 443.

[141] Ibid 449 (emphasis added). This finding was not disturbed on appeal: Australian Consolidated Press Ltd v Ettingshausen [1993] NSWCA 10 (13 October 1993) (‘Ettingshausen Appeal’).

[142] Burton v Crowell Pub Co 82 F 2d 154, 156 (Learned Hand J) (2nd Cir, 1936). See Ettingshausen Trial (1991) 23 NSWLR 443, 448.

[143] See Defamation Act 1974 (NSW) s 13, repealed by the Defamation Act 2005 (NSW).

[144] See Obermann v ACP Publishing Pty Limited [2001] NSWSC 1022 (16 November 2001) [37]–[40]; Hanson-Young v Bauer Media Ltd [2013] NSWSC 1306 (11 September 2013) [37]; Hanson-Young v Bauer Media Ltd (No 2) [2013] NSWSC 2029 (9 December 2013) [20].

[145] The issue was not addressed by the NSW Court of Appeal in Ettingshausen Appeal [1993] NSWCA 10 (13 October 1993) and ‘the correctness of Ettingshausen [Trial]’ was not challenged in McDonald v The North Queensland Newspaper Co Ltd [1997] 1 Qd R 62, 65–6.

[146] Radio 2UE Sydney [2009] HCA 16; (2009) 238 CLR 460, 467, 478, 484.

[147] Ibid 477–80, 484; Corby v Allen & Unwin Pty Ltd (2014) Aust Torts Reports 82–184, 67,723–67,724.

[148] Daniels [2010] EWHC 3057 (24 November 2010) [50], and adapting to the Radio 2UE Sydney ‘general test’.

[149] See above n 1.

[150] Defamation Act 2005 (NSW) s 6 confirms a role for ‘the general law’ (defined in s 4 as ‘the common law and equity’).

[151] See Standing Committee of Attorneys-General (SCAG), Annual Report 2005–06 <http://webarchive.nla.gov.au> . See also New South Wales, Parliamentary Debates, Legislative Assembly, 13 September 2005, 17641 (Bob Debus, Attorney-General, Minister for the Environment, and Minister for the Arts) (Second Reading Speech, Defamation Bill 2005 (NSW)).

[152] Australian Securities Commission v Marlborough Gold Mines Ltd [1993] HCA 15; (1993) 177 CLR 485, 492. For a more recent statement, see Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 230 CLR 89,

151–2 (‘Farah’).

[153] Lange [1997] HCA 25; (1997) 189 CLR 520, 563.

[154] See, eg, ibid; Lipohar v The Queen (1999) 200 CLR 485, 500 (Gleeson CJ), 505–10 (Gaudron, Gummow and Hayne JJ); 551–2 (Kirby J) (‘Lipohar’); Farah (2007) 230 CLR 89, 152.

[155] See, eg, Lipohar (1999) 200 CLR 485, 574–84 (Callinan J); L J Priestley, ‘A Federal Common Law in Australia?’ (1995) 6 Public Law Review 221, 230–3; Sir Anthony Mason, ‘The Interaction of Statute Law and Common Law’ (2016) 90(5) Australian Law Journal 324, 326–7.

[156] Priestley, above n 155, 232, referring to ‘[t]he brooding presence of the High Court’.

[157] See, eg, ibid 232–3, adopted by Callinan J in Lipohar (1999) 200 CLR 485, 580–2; Mason, above n 155, 327.

[158] See, eg, Lange [1997] HCA 25; (1997) 189 CLR 520; Coleman v Power [2004] HCA 39; (2004) 220 CLR 1; McCloy v New South Wales [2015] HCA 34; (2015) 257 CLR 178.

[159] George Williams, ‘Anti-Terrorism Laws and Human Rights’ (2015) 19(2) Review of Constitutional Studies 127, 140.

[160] Lange [1997] HCA 25; (1997) 189 CLR 520, 571.

[161] See, Charter of Human Rights and Responsibilities Act 2006 (Vic) ss 15, 13; Human Rights Act 2004 (ACT) ss 16, 12, respectively.

[162] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 681.

[163] Ibid.

[164] Ibid 686–7.

[165] David Rolph, ‘A Landmark Judgment’ Gazette of Law and Journalism, 18 August 2014 <http://www.glj.com.au/2281-article> . See also Freeburn v The Cake Decorators Association of NSW Inc (No 2) (2014) 19 DCLR(NSW) 232, 242 (Gibson DCJ) (‘Freeburn’).

[166] Judge Judith Gibson, ‘Case Management Issues in Costs Appeals (Costs Assessors Annual Seminar, 20 June 2015) 6 <http://www.districtcourt.justice.nsw.gov.au/Pages/Speeches-by-Judicial-Officers.aspx> Rolph, ‘A Landmark Judgment’, above n 165, also highlighting the decision’s significance for the issue of internet search engine liability.

[167] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 681 and see generally 679–82. See also Galbally, above n 15, 243–5, although the suggestion that McCallum J ‘reject[s] Jameel’ (at 243) may need some clarification. It is submitted that McCallum J does not actually decide whether Jameel is applicable in Australia, but, rather, proceeded on the basis that the abuse of process question can be decided without regard to this decision: see below n 169 and accompanying text. Further, her Honour arguably renders Jameel more amenable to importation into Australian common law by diminishing its dependence upon HRA considerations: see nn 172–4 below and accompanying text.

[168] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 679.

[169] Ibid 680.

[170] See CPR r 1.1.

[171] See, eg, Bristow v Adams [2012] NSWCA 166 (22 May 2012) [41] (Basten JA) (‘Bristow’); Manefield v Child Care NSW [2010] NSWSC 1420 (15 December 2010) [185]–[187]; Barach v University of New South Wales [2011] NSWSC 431 (24 May 2011) [128]–[129].

[172] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 678 (emphasis added).

[173] [2012] NSWCA 166 (22 May 2012).

[174] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 678, referring to ibid [41] and ECHR arts 6, 10.

[175] Bristow [2012] NSWCA 166 (22 May 2012) [41].

[176] Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 681.

[177] Ibid.

[178] Judge Judith Gibson, ‘It Came from CyberSpace: Defamation Law and the Internet (NSW State Legal Conference, 28 August 2014) 23 <http://www.districtcourt.justice.nsw.gov.au/Pages/Speeches-by-Judicial-Officers.aspx> .

[179] Ibid 23. See also Judge Judith Gibson, ‘From McLibel to E-Libel: Recent Issues and Recurrent Problems in Defamation Law (Speech delivered to the State Legal Convention, 30 March 2015) 14, 16–7 <http://www.districtcourt.justice.nsw.gov.au/Pages/Speeches-by-Judicial-Officers.aspx> Gibson, ‘The Trouble with Australian Defamation Law – 1’, above n 15.

[180] See YZ v Amazon (No 7) [2016] NSWSC 637 (20 May 2016) (‘YZ’).

[181] See, eg, Ghosh v TCN Channel Nine Pty Ltd (No 4) (2014) 19 DCLR (NSW) 38 (although note that on appeal, the NSW Court of Appeal was of the view that the primary judge had not found disproportionality beyond UCPR r 12.7 ‘want of due despatch’: Ghosh v NineMSN Pty Ltd [2015] NSWCA 334; (2015) 90 NSWLR 595, 602–3) (‘Ghosh Appeal’); Freeburn (2014) 19 DCLR(NSW) 232; Mohareb v Palmer (No 2) [2015] NSWDC 141 (30 July 2015); Elzahed v Commonwealth of Australia [2015] NSWDC 271 (18 November 2015); Park v Lee (2016) 22 DCLR(NSW) 250.

[182] Supreme Court of New South Wales, Practice Note SC CL 4Supreme Court Common Law Division – Defamation List, 5 September 2014, [23]; District Court of New South Wales (Civil Jurisdiction), Practice Note 6District Court Defamation List, 3 June 2015, [20].

[183] [2015] NSWCA 334; (2015) 90 NSWLR 595.

[184] Ibid 603 (Macfarlan JA with Adamson J agreeing) (emphasis added).

[185] [2016] NSWCA 296; (2016) 338 ALR 329. Special leave to appeal to the High Court was refused: [2017] HCASL 73

(30 March 2017).

[186] Smith v Lucht [2014] QDC 302 (17 November 2014) [6], [24]–[25], [27] (McGill DCJ) (‘Smith Application’); Lazarus v Azize [2015] ACTSC 344 (5 November 2015) [21], [23], [39] (Mossop AsJ).

[187] Clyne v NSW Bar Association [1960] HCA 40; (1960) 104 CLR 186, 201; Jago v District Court of NSW (1989)

[1989] HCA 46; 168 CLR 23, 25 (Mason CJ), 74 (Gaudron J).

[188] Support for such disconnection may be gleaned from Jameel [2005] EWCA Civ 75; [2005] QB 946, 963–7.

[189] Comprising Smith Application [2014] QDC 302 (17 November 2014), Smith v Lucht [2015] QDC 289 (20 November 2015) (‘Smith Trial’), Smith v Lucht [2016] QCA 267 (20 October 2016) (‘Smith Appeal’).

[190] Smith Trial [2015] QDC 289 (20 November 2015), affd [2016] QCA 267 (20 October 2016).

This did not escape attention in YZ [2016] NSWSC 637 (20 May 2016) [57].

[191] Smith Application [2014] QDC 302 (17 November 2014) [23], [27].

[192] This is the name of the home owner’s solicitor in the iconic Australian movie, The Castle, and was found to convey an imputation of incompetence and unprofessionalism: Smith Trial [2015] QDC 289 (20 November 2015) [17], [31] (Moynihan DCJ).

[193] Smith Application [2014] QDC 302 (17 November 2014) [23], [27], referring to Roberts v Prendergast [2013] QCA 47; [2014] 1 Qd R 357 and Cerutti v Crestside Pty Ltd [2016] 1 Qd R 89.

[194] See, eg, Defamation Act 2005 (NSW) s 33.

[195] Chappell v Mirror Newspapers Ltd (1984) Aust Torts Reports 80-691, 68,947 (‘Chappell’).

[196] Defamation Act 2005 (NSW) s 33.

[197] Chappell (1984) Aust Torts Reports 80-691, 68,946-8; Barrow v Bolt [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,698 (‘Barrow Appeal’); Smith Appeal [2016] QCA 267 (20 October 2016) [11], [102].

[198] Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,698 (Kaye JA with Ashley and McLeish JJA agreeing) (endnotes omitted). See also, Morosi v Mirror Newspapers Ltd [1977] 2 NSWLR 749, 800 (‘Morosi’); Chappell (1984) Aust Torts Reports 80-691, 68,948; King and Mergen Holdings Pty Ltd v McKenzie (1991) 24 NSWLR 305, 309–11 (‘King’); Perkins v New South Wales Aboriginal Land Council (Unreported, New South Wales Supreme Court, Badgery-Parker J, 15 August 1997) 27 (‘Perkins’); Jones v Sutton [2004] NSWCA 439; (2004) 61 NSWLR 614, 618, 621 (‘Jones’); Smith Appeal [2016] QCA 267 (20 October 2016) [37].

[199] Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,702 (Kaye JA with Ashley and McLeish JJA agreeing).

[200] Morosi [1977] 2 NSWLR 749, 800; Papaconstuntinos v Holmes à Court [2009] NSWSC 903 (4 September 2009) [104] (‘Papaconstuntinos’).

[201] Smith Appeal [2016] QCA 267 (20 October 2016) [50].

[202] Morosi [1977] 2 NSWLR 749, 799; Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,698; Smith Appeal [2016] QCA 267 (20 October 2016) [33].

[203] Chappell (1984) Aust Torts Reports 80-691, 68,947; Barrow Appeal (2015) Aust Torts Reports 82248, 69,698; Smith Appeal [2016] QCA 267 (20 October 2016) [102].

[204] Morosi [1977] 2 NSWLR 749, 799; Enders v Erbas & Associates Pty Ltd [2014] NSWCA 70; (2014) Aust Torts Reports 82-161, 67,111 (‘Enders Appeal’); Smith Appeal [2016] QCA 267 (20 October 2016) [35].

[205] Jones [2004] NSWCA 439; (2004) 61 NSWLR 614, 624, 625 (citations omitted); Enders Appeal [2014] NSWCA 70; (2014) Aust Torts Reports 82-161, 67,111; Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,698; Smith Appeal [2016] QCA 267 (20 October 2016) [36].

[206] King (1991) 24 NSWLR 305, 309 (Mahoney JA); Papaconstuntinos [2009] NSWSC 903 (4 September 2009) [105]; Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,698–9.

[207] See, eg, King (1991) 24 NSWLR 305, 309 (Mahoney JA) (‘significant’); Barrow Appeal (2015) Aust Torts Reports 82248, 69,698–9 (‘high’).

[208] Gibson, ‘From McLibel to E-Libel’, above n 179, 14.

[209] Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248; Smith Appeal [2016] QCA 267 (20 October 2016).

[210] [2014] VSC 599 (2 December 2014) (‘Barrow Trial’), affd Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248.

[211] See Barrow Appeal [2015] VSCA 107; (2015) Aust Torts Reports 82-248, 69,694.

[212] See, eg, ibid 69,699–700 (Kaye JA with Ashley and McLeod JJA agreeing).

[213] [2016] QCA 267 (20 October 2016) (Flanagan J with Philippides JA agreeing, Margaret McMurdo P not deciding the issue).

[214] Ibid [12] (Margaret McMurdo P).

[215] Ibid [8]–[11] (Margaret McMurdo P). ‘Harm’ in the s 13 precursor was not qualified by ‘any’.

[216] Szanto v Melville [2011] VSC 574 (15 November 2011) [162]. See also Smith Trial [2015] QDC 289 (20 November 2015) [37]; Smith Appeal [2016] QCA 267 (20 October 2016) [97], [100].

[217] Perkins (Unreported, New South Wales Supreme Court, Badgery-Parker J, 15 August 1997) 27; Jones [2004] NSWCA 439; (2004) 61 NSWLR 614, 618; Smith Appeal [2016] QCA 267 (20 October 2016) [50]. See, eg, Lang v Willis [1934] HCA 51; (1934) 52 CLR 637 concerning Defamation Act 1912 (NSW) s 5, repealed by the Defamation Act 1958 (NSW).

[218] Smith Appeal [2016] QCA 267 (20 October 2016) [16] (Philippides JA).

[219] See, eg, Jones [2004] NSWCA 439; (2004) 61 NSWLR 614.

[220] Haji-Ioannou v Dixon [2009] EWHC 178 (QB) (6 February 2009) [31], quoted with apparent approval in Kingsfield Holdings Pty Ltd v Rutherford [2016] WASC 117 (11 April 2016) [385].

[221] Jones [2004] NSWCA 439; (2004) 61 NSWLR 614, 625.

[222] Kim Gould, ‘The Statutory Triviality Defence and the Challenge of Discouraging Trivial Defamation Claims on Facebook’ (2014) 19(2) Media and Arts Law Review 113. See also Rothe v Scott (No 4) [2016] NSWDC 160 (5 August 2016) [127].

[223] See Gibson, ‘From McLibel to E-Libel’, above n 179, 17, with particular reference to Barrow Trial [2014] VSC 599 (2 December 2014).

[224] To avoid unnecessary repetition, references to authorities will generally not be given in this Part where they have been provided in earlier Parts.

[225] Roger S Magnusson, ‘Freedom of Speech in Australian Defamation Law: Ridicule, Satire and Other Challenges’ (2001) 9(3) Torts Law Journal 269, 281.

[226] [1936] 2 All ER 1237, 1240.

[227] [2010] EWHC 1414; [2011] 1 WLR 1985, 2002.

[228] [2010] EWHC 3057 (24 November 2010) [50].

[229] [2010] EWHC 1414; [2011] 1 WLR 1985, 2002.

[230] Cameron v Consolidated Press Ltd [1940] SAStRp 57; [1940] SASR 372, 377.

[231] [1936] 2 All ER 1237, 1242.

[232] Cameron v Consolidated Press Ltd [1940] SAStRp 57; [1940] SASR 372, 379.

[233] Bristow [2012] NSWCA 166 (22 May 2012) [41]; Bleyer [2014] NSWSC 897; (2014) 88 NSWLR 670, 678.

[234] Bristow [2012] NSWCA 166 (22 May 2012) [25].

[235] [2015] NZHC 1854 (6 August 2015).

[236] Defamation Act 2005 (NSW) ss 3(b)–(d) respectively.

[237] Judge Judith Gibson, ‘“Ridiculousness”: Ridicule and Defamatory Meaning in the Age of the Internet’ (Speech delivered at the College of Law, Sydney, 27 August 2014) 6 <http://www.districtcourt.justice.nsw.gov.au/Pages/Speeches-by-Judicial-Officers.aspx> .

[238] Gibson, ‘From McLibel to E-Libel’, above n 179, 2; ‘Practice Notes — Matthew Lewis’, above n 15.

[239] To borrow and adapt the words of Warby J in Lachaux [2015] EWHC 2242; [2016] QB 402, 420.

[240] Jameel [2005] EWCA Civ 75; [2005] QB 946, 970.

[241] But see David Rolph, ‘The Trouble with Australian Defamation Law – 2’, Gazette of Law and Journalism, 31 August 2015 <http://www.glj.com.au/2513-article> .