Abou-Lokmeh v Harbour Radio Pty Ltd

McCOLL JA: The applicant, Bassem Abou-Lokmeh, seeks leave to appeal and to appeal from a judgment of Gibson DCJ granting the first, second and third respondents, respectively Harbour Radio Pty Ltd (Harbour Radio), Ray Hadley and Mahassen Issa, leave to file a proposed amended defence to defamation proceedings commenced by Mr Abou-Lokmeh, and rejecting his application to strike out parts of that defence. [1]

The leave application was originally listed for hearing before Basten and Simpson JJA on 23 March 2016, on which occasion their Honours referred the application to a Full Court to be heard as a concurrent matter with the appeal if leave be granted. For the reasons that follow, I am of the view that while there should be a limited grant of leave to appeal, the appeal should be dismissed with costs.

Mr Hadley is the presenter of the Ray Hadley Morning Show on Radio 2GB. On 3 November 2014, in the course of a broadcast on that show, Mr Hadley had a conversation with Ms Issa, who was married to, but separated from, Mr Abou-Lokmeh. A transcript of the broadcast is set out in the Schedule to these reasons.

It suffices to describe the subject of the broadcast to say that it was an interview between Mr Hadley and Ms Issa concerning events to which the latter alleged she was subjected by Mr Abou-Lokmeh when she visited Lebanon with a new partner. On an earlier occasion, Harbour Radio had broadcast an interview between Mr Hadley and Ms Issa’s brother, Mr Issa. Mr Issa apparently said during that interview that, while in Lebanon, Ms Issa married her new partner while still married to Mr Abou-Lokmeh. It appears that Ms Issa was invited to participate in the interview with Mr Hadley to respond to Mr Issa’s allegations. She denied them and, in response to Mr Hadley’s request, gave her version of the events. [2] Mr Abou-Lokmeh alleges that Ms Issa defamed him while doing so.

On 3 March 2015 Mr Abou-Lokmeh commenced proceedings in defamation in respect of the November 2014 broadcast. He pleaded that the following imputations arose from the parts of the matter complained of indicated in the brackets and that each imputation was defamatory of him:

The respondents filed their defence on 1 June 2015. They pleaded, inter alia, a defence of contextual truth pursuant to s 26 of the Defamation Act 2005 (NSW) (2005 Act) and common law qualified privilege on the basis of reply to an attack (reply to attack defence), both of which defences were the subject of the strike-out application. The attack particularised was one said to have been made by Mr Issa on the Ray Hadley Morning Show during the earlier broadcast to which I have referred. The earlier broadcast was not in evidence before the primary judge or in this Court, however portions of it were extracted as particulars to the reply to attack defence. In addition, the respondents pleaded that imputations (b) and (c) were substantially true and a defence of justification and qualified privilege pursuant to s 25 and s 30 of the 2005 Act respectively.

Section 26 of the 2005 Act provides a defence to the publication of defamatory matter if the defendant proves that the matter carried, in addition to the defamatory imputations of which the plaintiff complains, one or more other imputations (contextual imputations) that are substantially true, and the defamatory imputations do not further harm the reputation of the plaintiff because of the substantial truth of the contextual imputations.

The contextual truth defence pleaded the following contextual imputations:

In July 2015 the respondents sent Mr Abou-Lokmeh a proposed form of amended defence and indicated they would seek leave to file it. Leave was granted by Order 1 of the primary judgment. Although Mr Abou-Lokmeh now seeks leave to appeal from the primary judge’s order granting the respondents leave to file the amended defence, the portions of the amended defence relevant to Mr Abou-Lokmeh’s application for leave to appeal were in the same form in the original defence. Nevertheless, in the course of the amendment application Mr Abou-Lokmeh sought to strike out the contextual truth and reply to attack defences. [4]

The primary judge summarised Mr Abou-Lokmeh’s submission that the contextual truth defence should be struck out as follows:

Mr Abou-Lokmeh submitted that the reply to attack defence should be struck out because it was Mr Issa, not he, who mounted any attack.

The strike-out application failed.

Mr CJ Dibb, who appeared for Mr Abou-Lokmeh on appeal and below, submitted that leave to appeal should be granted because the appeal raised important issues relating to the defences to defamation proceedings and had good prospects of success. He contended the appeal should be dealt with at the interlocutory stage to prevent the proceedings from becoming an inappropriately wide-ranging enquiry into the whole course of Mr Abou-Lokmeh and Ms Issa’s marriage, and to give effect to what he described as the “overriding objective and proportionality principle” contained in Civil Procedure Act 2005 (CPA), ss 56 – 60.

Mr ATS Dawson, who appeared for the respondents on appeal with Ms Barnett, but not below, submitted that the primary judge’s decision was an exercise of discretion in relation to a matter of practice and procedure and was not of the kind that warranted appellate intervention. He contended that the application for leave to appeal involved no question of principle, and there would be no injustice should the matter proceed to trial on the basis of the amended defence.

As Mr Dawson submitted, the primary judge's decision concerned matters of practice and procedure. Thus, the application for leave to appeal attracts the proposition articulated by Jordan CJ in Re Will of Gilbert [6] that if a tight rein is not kept upon allowing appeals to an appellate court on such matters, the result would be disastrous for the proper administration of justice. Accordingly, appellate courts should exercise particular caution in reviewing decisions pertaining to practice and procedure. [7]

An applicant for leave must demonstrate something more than that the trial judge was arguably wrong in the conclusion arrived at. [8] Generally speaking, it is only appropriate to grant leave to appeal in matters that involve issues of principle, questions of public importance or in circumstances where it is reasonably clear that an injustice has occurred by reason of error in the judgment, going beyond what is merely arguable. [9]

In my view, for the reasons which follow, the application to strike out contextual imputation (i) and the reply to attack defence raises issues of principle warranting a grant of leave to appeal, but the application relating to contextual imputations (ii) – (iv) does not.

Mr Dibb submitted before the primary judge that the matter complained of did not convey any imputation of killing, other than that identified in the passages pleaded in respect of imputation (a) in the statement of claim. He contended that contextual imputation (i) was essentially the same as imputation (a) and was not therefore reasonably capable of being found by a jury to be “in addition to” his imputation (a). [10]

Ms Barnett, who appeared for the respondents below, submitted that imputation (a) related to a specific act, namely Mr Abou-Lokmeh allegedly giving his lawyer a large amount of money with orders either to have Ms Issa put in jail or killed, whereas contextual imputation (i) pleaded a condition, and related more generally to Mr Abou-Lokmeh’s state of mind and, in particular, his animus against Ms Issa. She further contended that none of Mr Abou-Lokmeh’s imputations dealt with his state of mind, although the thrust of the matter complained of was his jealousy-fuelled quest for vengeance against Ms Issa, which led him to engage in a series of acts designed to harm or imprison her, only one of which was to pay his lawyer to have her killed. [11]

The primary judge made the following finding in respect of contextual imputation (i):

The grounds of appeal contend that the primary judge erred:

In this Court Mr Dibb essentially repeated the submissions made below. He accepted that contextual imputation (i) was carried by the matter complained of because the allegation that Mr Abou-Lokmeh paid money to a lawyer to have Ms Issa murdered necessarily conveyed that he wanted her killed. However, he submitted that that imputation did not arise separately and “in addition to” imputation (a). He also argued that the primary judge erred in taking into account the fact that the jury might find imputation (a) to be proved true, given that the respondents did not plead a justification defence in respect of that imputation.

As a general proposition, Mr Dawson contended in his written submissions that in order for Mr Abou-Lokmeh’s strike out application to succeed in respect of the contextual imputations, he was required to establish that each meaning of the contextual imputations pleaded was not one any reasonable jury, properly instructed, could find to be carried “in addition to” his imputations. Mr Dawson submitted that this was a very high bar. If it was a question upon which reasonable minds might differ, he argued the contextual imputations must be left to the jury.

Turning to contextual imputation (i), Mr Dawson repeated the submission made below that that imputation was “in addition to” imputation (a) because it concerned Mr Abou-Lokmeh’s state of mind, rather than the specific act attributed to him in imputation (a). He submitted it was open to a jury to find the requisite difference existed.

Secondly, Mr Dawson submitted that the primary judge’s reference to the possibility of imputation (a) being found to be true was not infelicitous. Rather, he contended her Honour was applying the test of whether contextual imputation (i) differed in substance from imputation (a) by looking at the evidence the respondents would rely upon to seek to justify contextual imputation (i). He submitted that the particulars of the contextual truth defence demonstrated that the respondents would seek to justify contextual imputation (i) by proving a number of occasions on which Mr Abou-Lokmeh told Ms Issa either that he would kill her or hoped that she would die, or was suspected to have tampered with Ms Issa’s car with a view to causing it to explode or start a fire. He submitted that those particulars demonstrated the substantial difference between contextual imputation (i) and imputation (a) and, accordingly, that it was carried “in addition to” imputation (a).

Mr Dawson also submitted that contextual imputation (i) was of a general nature, and therefore of a Maisel type [12] compared to the specific nature of imputation (a) and that this was another reason the “in addition to” test was satisfied.

Finally, in this context Mr Dawson submitted that the Court should not assume a jury would conclude imputation (a) was conveyed. If a jury were to form the view that imputation (a) was not conveyed, he argued it would be open to the respondents to rely on contextual imputation (i) in relation to such other of Mr Abou-Lokmeh’s imputations as were found to be conveyed and from which it differed in substance.

The Court’s strike out jurisdiction should not be exercised in Mr Abou-Lokmeh’s favour unless the contextual imputations are obviously untenable, [13] or “unarguably bad”. [14] The Court should “take the defendant’s case at its highest as to which of the plaintiff’s imputations were conveyed (that is, to compare the particulars of contextual truth with the least serious of the plaintiff’s imputations).” [15]

A defence of contextual truth must defeat the whole defamatory matter of which the plaintiff complains, that is to say, all the plaintiff’s stings or imputations. [16] The tribunal of fact must be able to conclude that, because of the substantial truth of the contextual imputations, the defamatory imputations which constitute the plaintiff’s cause of action do not further harm the plaintiff’s reputation. The focus is on comparing the contextual imputations with the plaintiff’s cause of action. [17] As McCallum J explained in McMahon, [18] “the defence does not compare imputation with imputation. [Its] essence … is to permit the defendants to put the plaintiff’s imputations in their factual context according to the content of the whole of the article.”

In order to be carried “in addition to” the plaintiff’s defamatory imputations, the contextual imputations must differ in substance from the plaintiff’s imputations. [19] As I observed in Zeccola (at [48]), in Ange v Fairfax Media Publications Pty Ltd, [20] Nicholas J explained that the words “in addition to” in s 26(a) removed the confusion under s 16 of the Defamation Act 1974 (NSW) (1974 Act) about what was required to establish a contextual truth imputation as a defence. In Ange, Nicholas J said of the “differ in substance” test:

In addition to the matters to which Nicholas J referred, the question whether a contextual imputation differs in substance from the plaintiff’s imputation may be determined either by identifying what the defendant must prove in order to justify the contextual imputation, [21] or by asking “what may be proved by way of justification to each such imputation” (emphasis in original). [22]

The “differ in substance” test can also be satisfied by a defendant pleading a contextual imputation in general terms in the sense permitted by Maisel. [23] As I have said, this is one way the respondents put their argument in support of contextual imputation (i).

Hunt J explained the Maisel approach in the contextual truth context in Allen as follows:

The “policy” of which Hunt J was speaking was that introduced by the contextual truth defence. Under that defence as available both pursuant to s 16 of the 1974 Act and s 26 of the 2005 Act, a plaintiff cannot by selective pleading of a less serious imputation, when a more serious one is carried by the matter complained of, prevent a defendant from seeking to justify that more serious imputation. That had been the position at common law. [25]

The requirement that the contextual imputation(s) be conveyed at the same time as the plaintiff’s imputation(s) is to enable the tribunal of fact to “weigh or … measure the relative worth or value of the imputation or imputations for which each party contends”. [26]

The impressionistic nature of the differ in substance test to which Nicholas J referred in Ange can be illustrated as follows.

In Hyndes v Nationwide News Pty Limited, [27] McCallum J accepted that it was possible for words to convey an act and, separately, a condition. However, her Honour held that in the circumstances of that case, the condition which the defendant sought to rely upon as a contextual imputation (willingness to lend money to persons the plaintiff believed to be drug dealers) to the plaintiff’s imputations (which in various ways alleged that he had lent money to persons engaged in drug dealing) was necessarily implicit in the act identified in the plaintiff's imputations. Accordingly, in her Honour’s view, the contextual imputations could not be said to arise at the same time, and in addition to, the plaintiff’s imputations.

McCallum J distinguished State of New South Wales v Deren, [28] in which Priestley JA (with whom Powell and Stein JJA agreed on this issue) held that the State’s contextual imputation that the plaintiff was a child molester, differed in substance from the plaintiff’s imputation which directly accused the plaintiff of particular sexual assaults on particular young children coming from a particular location during a particular time. His Honour accepted that it may be that the contextual imputation could be “derived from” the plaintiff’s imputation, but said that even if that were right, and even if the ideas conveyed by the parties’ respective imputations had “much in common”, it would not follow that the meaning of the two imputations was the same. [29] His Honour’s reasoning reflects that of Hunt J in Allen to which I have referred concerning the same subject matter. [30]

As Hyndes and Deren demonstrate, the question whether a particular charge of wrongdoing carries a general charge “may depend on the context in which the words are used” and the “the gravity of the misconduct imputed in the particular charge”. [31]

Thus, the mere fact that contextual imputation (i) may be “derived from” (which expression, in my view, would comprehend the notion of being “implicit” in another imputation) in imputation (a), does not answer the question, as Mr Dibb’s submissions tended to assume, whether the former is “in addition to” the latter.

There are, accordingly, a number of reasons why the primary judge did not err in refusing to strike out contextual imputation (i).

First, while imputation (a) and contextual imputation (i) are derived from that part of the matter complained of in which Ms Issa alleged Mr Abou-Lokmeh paid to have her killed, the particulars said to make contextual imputation (i) substantially true range over a greater factual matrix than the incident referred to in the matter complained of, unrelated to the specific instance identified in the matter complained of. [32] That goes to satisfying the differ in substance test, at least at a strike-out stage. [33]

Secondly, contextual imputation (i) falls within the principles Hunt J identified in Allen. Here, I would accept the respondents’ submission, that imputation (a) pleads an imputation of a specific nature, that Mr Abou-Lokmeh paid to have Ms Issa murdered, whereas contextual imputation (i), although related to the same subject matter, contends that he “wanted her killed” and is of a general nature. Contextual imputation (i) arguably conveys “a different, and potentially broader, sting” than imputation (a), [34] thus satisfying the differ in substance test.

Thirdly, in most cases a general allegation will differ in substance from the particular. [35] An imputation that Mr Abou-Lokmeh wanted Ms Issa killed (and demonstrating his state of mind) is capable at a strike out stage of being substantially different from an imputation that he paid to have her murdered (being an imputation relating to an act). [36]

Fourthly, in this case in my view the gravity of imputation (a) is such that the general charge contained in imputation (i) is capable of being conveyed in addition to it.

Finally, I do not accept that the primary judge’s reference [37] to the possible fate of imputation (a) distracted her Honour from applying the correct test.

I am conscious that the question whether an imputation amounts to a general charge against the character of a plaintiff is one to be approached with care, to avoid unduly opening up avenues for defences of justification and contextual truth which would otherwise be barred to a defendant and thus prolong the hearing of defamation cases. [38]

However, in my view Mr Abou-Lokmeh has not established that contextual imputation (i) is obviously untenable or unarguably bad. It is at least arguable that the jury could find that Mr Abou-Lokmeh’s imputation (a) and contextual imputation (i) differ in substance. [39] That is sufficient at the strike-out stage to make it inappropriate to strike it out. The primary judge did not err in refusing to do so.

Mr Dibb submitted before the primary judge that neither of contextual imputations (ii) and (iii) was capable of being conveyed or conveyed by any part of the matter complained of other than the same passages which conveyed imputation (a). Accordingly, he contended neither contextual imputation could be conveyed “in addition to” his imputations. [40]

Ms Barnett submitted contextual imputations (ii) and (iii) were conveyed by passages of the matter complained of other than those which conveyed imputation (a). She relied upon what she contended was a discernible sound of fear in Ms Issa’s voice during the interview which would be apparent to the ordinary reasonable listener. She also relied upon those passages in which Ms Issa said that she could have been one of many women that are still trapped overseas: “I would have been one of them Ray”, as well as “I would have been another number” as supporting the proposition Ms Issa was in fear by reason of Mr Abou-Lokmeh’s conduct. [41]

The primary judge declined to strike out contextual imputations (ii) and (iii), saying:

Mr Abou-Lokmeh’s notice of appeal contends the primary judge erred in holding that the matter complained of was capable of conveying contextual imputations (ii) and/or (iii) at all and/or in addition to his imputations.

Again, Mr Dibb essentially repeated the arguments advanced below. He contended that contextual imputations (ii) and (iii) were not capable of being carried by the matter complained of at all, or in addition to imputation (a). He submitted there was nothing in the matter complained of to suggest Ms Issa was “at any time in fear of her life” or that she was aware at the time of the broadcast that Mr Abou-Lokmeh had allegedly paid to have her killed. He contended that listening to the broadcast “reveals not the slightest fear or concern by [the] tone of [her] voice”. He argued that if it were thought that Ms Issa’s voice indicated, by itself or in combination with other factors, that she was afraid, there was nothing in the matter complained of to suggest that any conduct on Mr Abou-Lokmeh’s part caused such feelings. Mr Dibb did not address any submissions to the “in addition to” point.

In his oral submissions, Mr Dibb contended that the primary judge’s reasoning in relation to contextual imputations (ii) and (iii) was so erroneous that there was latent error in the sense explained in House v R [42] such as would warrant appellate intervention.

Mr Dawson supported the primary judge’s reasoning. He submitted that the question of the tone of the broadcast was quintessentially a matter for the jury. He contended that the primary judge was correct in finding that the combined effect of the language, subject matter and tone used by Ms Issa were sufficient, at least on a capacity basis, for her Honour to conclude that both contextual imputations (ii) and (iii) were capable of being carried by the broadcast. He also argued that each of these contextual imputations conveyed a different and distinct sting that was not captured by Mr Abou-Lokmeh’s imputations. He relied on the fact that contextual imputations (ii) and (iii) related to Ms Issa’s state of mind rather than the acts referred to in Mr Abou-Lokmeh’s imputations. He also drew attention to a number of other passages in the matter complained of similar to those to which the primary judge referred (at [22]) as supporting the capacity of the matter complained of to carry these contextual imputations.

I would refuse leave to appeal in relation to these contextual imputations. In my view Mr Abou-Lokmeh has not identified any issue of principle, question of public importance or demonstrated that an injustice has occurred by reason of the primary judgment. The passages to which the respondents refer are capable of demonstrating that Ms Issa was left in fear by a number of matters, not least that Mr Abou-Lokmeh had paid to have her killed. It is a question for the jury whether they hear the tone of fear the respondents contend was apparent in her voice. Generally, as the primary judge held, the subject matter of the matter complained of, which was to a large extent about Mr Abou-Lokmeh’s conduct directed to keeping Ms Issa trapped in a foreign country against her will, was capable of conveying contextual imputations (ii) and (iii). Further, contextual imputations (ii) and (iii) are carried by other passages in the matter complained of. Finally, they differ in substance from Mr Abou-Lokmeh’s imputations in the sense I have explained. The respondents seek to justify these contextual imputations by proof of matters unrelated to the two incidents to which Mr Abou-Lokmeh’s imputations relate.

At the strike-out application, Mr Dibb challenged the “seriousness” of contextual imputation (iv), [43] rather than the capacity of the matter complained of to carry it. He submitted that an imputation that Mr Abou-Lokmeh dishonestly registered his marriage in Lebanon when he knew Ms Issa had filed for divorce was “self-evidently” much less serious than any of his imputations. He further submitted that, pursuant to what he described as “the principle of proportionality” and the court’s overriding obligation pursuant to CPA, s 56 to ensure that proceedings be conducted with a view to determining the real issues in a “just, cheap and quick” fashion, the primary judge should refuse to permit contextual imputation (iv) to stand. [44]

The primary judge identified the sting in contextual imputation (iv) as being that Mr Abou-Lokmeh dishonestly abused the legal process, in that he had his marriage to Ms Issa registered in Lebanon even though he knew his marriage was over and that she had filed for divorce. [45] Her Honour continued:

Accordingly, her Honour declined to strike out contextual imputation (iv).

Mr Abou-Lokmeh contended in his amended summons seeking leave to appeal that the primary judge erred in holding that contextual imputation (iv) was capable of so affecting Mr Abou-Lokmeh’s reputation as to found a defence under s 26 of the 2005 Act to any of his imputations. Tellingly, there was no ground of appeal in the draft notice of appeal relating to this contextual imputation.

Mr Dibb did not advance any written submissions in respect of contextual imputation (iv). In oral argument, he conceded that Mr Abou-Lokmeh was “not particularly troubled” by this contextual imputation.

I would refuse leave to appeal in relation to contextual imputation (iv). In my view Mr Abou-Lokmeh has not identified any issue of principle, question of public importance or demonstrated that an injustice has occurred by reason of any error in the primary judgment. As the primary judge said, the only issue Mr Abou-Lokmeh sought to canvass was whether by reason of the substantial truth of contextual imputation (iv) his imputations did not further harm his reputation. [46] That is an issue a jury should determine by comparing contextual imputation (iv) with such of Mr Abou-Lokmeh’s imputations as are found to have been carried by the matter complained of.

The primary judge dealt with Mr Abou-Lokmeh’s attempt to strike out the reply to attack defence as follows:

Accordingly, her Honour refused Mr Abou-Lokmeh’s application to strike out the defence.

Mr Dibb submitted the primary judge erred in holding that the reply to attack defence was arguable. He contended that the attacks particularised in the amended defence as giving rise to the occasion for response were two broadcasts in which Mr Issa made remarks about Ms Issa. However, the defence did not plead that Mr Abou-Lokmeh was in any way responsible for, or complicit in, Mr Issa’s attacks. Accordingly, he argued the reply to attack defence was not open. He also contended that what Ms Issa said about Mr Abou-Lokmeh did not constitute a response to the particularised attack, giving as an illustration Ms Issa’s assertion that Mr Abou-Lokmeh had paid to have her killed.

In his oral submissions, Mr Dibb contended her Honour erred in referring to Dixon J’s statement in Loveday v Sun Newspapers Ltd [47] as his Honour was writing in dissent, his statement had been “read too widely”, and, in any event, his statements were not essential to his reasoning to his conclusion. He relied on Kelly J’s reasons in McNicol v ACT Health Authority, [48] which he contended were persuasive, as standing in the way of accepting Dixon J’s statement and as more accurately reflecting what had been said in Coward v Wellington. [49] He also submitted that, in contrast to Loveday, the respondents did not allege Mr Abou-Lokmeh was complicit in Mr Issa’s attack.

Mr Dawson characterised Mr Abou-Lokmeh’s submissions as raising two complaints in respect of the reply to attack defence. First, that the defence did not allege he was complicit in the attack. Secondly, that Ms Issa’s reply was unconnected with the particularised attack. Mr Dawson submitted that the principal difficulty with both complaints was that they raised factual issues for trial.

In respect of Mr Dibb’s “complicity” submission, Mr Dawson contended that given the attack was an exposition of the dispute between Mr Abou-Lokmeh and Ms Issa which was more than favourable to Mr Abou-Lokmeh, and was made by Mr Issa against her, it was at least “on the cards” that he was complicit in it. He argued that pre-trial processes and cross-examination would reveal the extent of that complicity. Further, Mr Dawson submitted it was clearly arguable that the reply to attack defence was available, whether or not Mr Abou-Lokmeh was complicit in the attack.

In respect of the second complaint, Mr Dawson submitted that, given the subject matter of the attack, it was difficult to see how Ms Issa could have replied without defaming Mr Abou-Lokmeh.

Mr Dawson submitted that Harbour Radio Pty Ltd v Trad [50] established that the question of whether the reply was sufficiently connected to the attack, and therefore sufficiently relevant to the occasion, was a matter for the jury at trial.

Finally, Mr Dawson submitted the reply to attack defence could not be said to be so manifestly hopeless or untenable as to satisfy the General Steel summary dismissal test.

Pursuant to s 22(2) of the 2005 Act, a jury is to determine whether the defendant has published defamatory matter about the plaintiff and, if so, whether any defence raised by the defendant has been established. Nothing in s 22, however, requires or permits a jury to determine any issue that, at general law, is an issue to be determined by the judicial officer. [51]

At general law it was for the judicial officer to determine whether any publication took place on an occasion of qualified privilege and, too, to be satisfied that there was sufficient evidence to go to the jury on matters within its province. [52] It was for the jury to determine whether the communication was privileged as a matter of fact, that is to say, whether the defendant used the occasion for the reason for which it was recognised or, in the case of reply to attack, whether the matter complained of exceeded legitimate bounds in the vindication either of character or proprietary interest. [53] It was also for the jury to determine any disputed questions of fact upon which the matter for judicial determination depended. [54]

The width of the principles governing qualified privilege for defamation makes it necessary, in deciding whether they apply, to make a close scrutiny of the circumstances of the case, of the situation of the parties, of the relations of all concerned and of the events leading up to and surrounding the publication, involving a consideration of all the circumstances leading up to publication. [55]

In determining whether the defendant’s statements purportedly in reply to an attack are “commensurate with the occasion”, [56] “no nice scales should be used.” [57] In Turner v Metro-Goldwyn-Mayer Pictures Ltd, [58] Lord Oaksey drew “an analogy between the criminal law of self defence and a man’s right to defend himself against written or verbal attacks” in both of which cases “he is entitled, if he can, to defend himself effectively, and he only loses the protection of the law if he goes beyond defence and proceeds to offence”, adding “[t]he law does not concern itself with niceties in such matters. If you are attacked by a prize fighter you are not bound to adhere to the Queensberry rules in your defence.” However, the privilege will be lost if what is published is not “reasonably appropriate to [the] duty or right”, or is “unconnected with and irrelevant to the main statement”. [59]

In Trad, the plurality (Gummow, Hayne and Bell JJ) approved Dixon J’s statement of principle at first instance in Penton concerning the foundation of the common law qualified privilege defence of reply to an attack. [60] Dixon J said:

The plurality also referred, [62] with apparent approval, to Dixon J’s statement in Mowlds v Fergusson, [63] relevantly that:

Where an attack is published to the public at large, defence to attack qualified privilege extends to publication of the answer in the newspapers or in any other manner that would reach the public generally. The reason for the privilege of the newspaper publisher is that it is right and for the common convenience and welfare of society that the publisher should lend the aid of the newspaper to the party who is entitled publicly to repel the attack or answer the criticism. [64] In this context, however, “the consideration of what is relevant to the attack requires particular care [as] [t]he response must be commensurate with an occasion which is in an exceptional category.” [65]

The most obvious form of a defence to attack qualified privilege is where the defendant has responded to the plaintiff’s attack in order to vindicate his or her reputation and, in the course of doing so, has defamed the plaintiff. Subject to the response satisfying the principles referred to above, and the plaintiff not being able to establish malice operative in the publication, the defence is established.

However, the privilege is not so confined.

First, the attack need not expressly be made by the plaintiff. Rather, as is apparent from Loveday, it may be sufficient if it can be found that an attack ostensibly in another’s name was authorised by the plaintiff or, to use the language of counsel in this case, if it can be found that the plaintiff was complicit in the attack. In such circumstances the plaintiff can also be seen to have been responsible for the attack. [66] I deal with Loveday in more detail below. However, in this respect, there is force in Mr Dibb’s complaint that the respondents have not pleaded that Mr Abou-Lokmeh was complicit in the attack on Ms Issa. Rather, the particulars of the reply to attack defence set out the substance of the interview in which the attack was made. They conclude by asserting the matter complained of was a “reply to the attack by Mr Issa.”

Accordingly, at this stage, the submission that the respondents may establish Mr Abou-Lokmeh was complicit in the attack appears to be a matter of speculation. It cannot be dismissed however as groundless. Mr Abou-Lokmeh and Ms Issa were married for many years before the events which led to publication of the matter complained of. It cannot be said that there is “no possibility” of circumstances emerging demonstrating Mr Abou-Lokmeh’s involvement in Mr Issa’s attack. [67]

While this may be a sufficient basis on which to dispose of the challenge to the reply to attack defence, in case minds may differ, the alternative basis on which the respondents put their case should also be considered.

Secondly, the response to the attack need not necessarily be made by the person attacked. Rather, the respondent may be a body such as the Army Council in Adam v Ward, which had a duty to investigate the plaintiff’s attack and, if appropriate, vindicate the victim to the world. [68] In that case, the plaintiff was formerly an officer in a cavalry regiment who was subsequently elected a member of Parliament. He gave a speech in the House of Commons, falsely charging the General commanding the brigade of which his late regiment formed part with sending confidential reports to Army Headquarters about officers under his command which contained wilful and deliberate misstatements. The General referred the matter to the Army Council as he was required to do, whereupon the defendant, as secretary to the Council and at their direction, wrote a letter to the General, vindicating him against the charge made by the plaintiff and containing defamatory statements about the plaintiff, and sent it to the Press for publication. The letter was widely published in the British and Colonial Press. The House of Lords held that the letter was published on a privileged occasion because it was the duty of the Army Council “to inquire into the truth of this charge and to make the result of that inquiry known as widely as possible”, [69] to vindicate the General’s conduct. [70]

Another illustration is Bowen-Rowlands v Argus Press Ltd, [71] in which a relative successfully relied on a reply to attack defence when sued for defamation in respect of defamatory statements she made about the plaintiff in the course of vindicating her father’s reputation.

Thirdly, it seems that the reply to attack defence may succeed even if the reply does not defame the attacker, but, as ostensibly in this case (subject to the complicity issue), a third party. Coward v Wellington illustrates that proposition. In that case, a customer accused a tradesman of charging for goods which were never delivered. The tradesman responded in writing suggesting that an employee of the customer was guilty of fraud. The employee was dismissed. The employee’s husband brought proceedings for special damage he sustained by reason of his wife’s dismissal. [72] The defence was that the letter was a privileged communication, written by the defendant “in his own defence”. His letter was found to be protected by qualified privilege. Littledale J held that “[i]f a man bonâ fide writes a letter in his own defence, and for the defence and protection of his interests and rights, and is not actuated by any malice, that letter is privileged, although it may impute dishonesty to another; but in such cases, malice may either be proved by the letter itself, or by other evidence.” [73]

Coward v Wellington has been referred to in the High Court and elsewhere with apparent approval. Barton ACJ referred to it in Norton, [74] to illustrate, it would appear, the ambit of a defence of reply to attack.

In Mowlds, [75] Williams J referred to Coward v Wellington as illustrating the proposition that “[i]f the reputation of a person holding a public position is publicly attacked he has an interest to reply to such attack and publish it to the members of the public who are likely to have become aware of it, and such members of the public would have an interest to receive it”. His Honour also referred with approval to Dixon J’s reasons in Loveday. [76]

The defendant in Mowlds was a senior police officer, who was severely criticised by a Royal Commissioner in respect of a report he had written in 1934 in a matter involving the conduct of the plaintiff. In 1937, as a result of a request originating from the Premier, the defendant wrote a further report on the same matter in which, in the course of seeking to justify his 1934 report, he defamed the plaintiff. The defendant showed his 1937 report to C, who had been the Commissioner of Police from whom he had obtained advice about his 1934 report, but who had since resigned. The High Court unanimously held that the publication to C was on an occasion of qualified privilege. Mowlds demonstrates how the notions of corresponding duty or interest must be “very widely interpreted” when defamatory matter is published in self-defence, in defence or protection of an interest, or by way of vindication of reputation following an attack. [77]

As Mr Dawson submitted, Mowlds is an example of a publisher being able to rely upon defence to attack qualified privilege in resisting defamation proceedings by a person who was not that attacker, but was defamed in the course of the reply.

In Loveday, the defendant newspaper published an article containing extracts from a letter written to the newspaper by the secretary of a District Unemployed Relief Council (DURC) attacking Canterbury Municipal Council (Council) with respect to its refusal of relief work to the plaintiff. The same article contained a statement in reply, prepared by the Town Clerk of the Council, saying the plaintiff had been refused relief work and giving reasons which were defamatory of him. The plaintiff brought defamation proceedings against the newspaper and the Town Clerk, but was nonsuited on the basis that the publication was made on an occasion of qualified privilege which protected both the newspaper and the Town Clerk and there was no evidence of malice fit to be left to the jury.

The plaintiff’s application to set aside the nonsuit failed before the Full Court and unanimously in the High Court. The plaintiff’s contention in the High Court was that before considering the nonsuit application, the trial judge should have asked the jury to determine whether he authorised the sending of the letter to the newspaper, [78] a point which had not been raised at trial. [79]

The Court (Latham CJ, Starke and Dixon JJ) held that if the plaintiff had authorised the DURC to send the letter to the newspaper, both defendants were entitled to rely upon a reply to attack qualified privilege. [80] Latham CJ and Starke J were both of the view that there was evidence at trial to support a finding that the plaintiff had authorised the DURC to send the letter to the newspaper and, accordingly, that the nonsuit was right. [81]

Dixon J disagreed on this point. His Honour cited Coward v Wellington in support of the proposition that “ ... if A attacks B and in order to defend himself B reasonably publishes defamatory matter of C, who has been no party to the attack, B's privilege will protect him as against C”. [82] His Honour did so apparently to demonstrate that the qualified privilege defence would run even if the plaintiff (C) had not authorised the DURC (A) to send the letter to the newspaper attacking the municipal council (B) as long as the DURC letter had been published prior to the defamatory publication. [83] However, in his Honour’s view, if it could not be established that the plaintiff consented, invited or incited the newspaper to publish the DURC letter and the Town Clerk’s response, the newspaper could not claim the benefit of a qualified privilege defence where it published the Town Clerk’s response “not in answer, but as a fresh item on which a general interest, as distinguished from a particular interest already aroused”. [84] In his Honour’s view, “upon the evidence as it stood the jury were not bound to find that the plaintiff did authorize the letter to the newspaper” and, accordingly, the newspaper was not entitled to a nonsuit. [85] His Honour was of a different view in relation to the Town Clerk who had “procured the publication of the defamatory matter for the purpose of meeting an attack to be published at the same time.” [86]

In Palmer v Belan, [87] a defamatory leaflet written by the defendant, Mr Belan, in response to an “attack” by the plaintiff, Mr Palmer, “repeatedly named” a Mr Boner. Kirby J referred with approval to Dixon J’s reasons in Loveday, [88] in support of the proposition that publication of the leaflet was protected by reply to attack qualified privilege. In his Honour’s view, “[a]lthough Mr Boner had nothing to do with the attack, the victim may, in his response, deal with the complicity of third parties, provided the allegations against them are not wholly unconnected with the defence”. [89]

In McNicol, in the passage upon which Mr Dibb relied, after setting out the facts of Coward v Wellington and Littledale J’s statement to which I have referred, Kelly J said:

In my view neither Coward v Wellington nor the passage in Williams J’s reasons in Mowlds to which I have referred above is authority for the limited proposition Kelly J enunciated. The question is whether the statements the defendant makes to vindicate his or her reputation in response to an attack are “commensurate with the occasion” or, as Kiefel J emphasised in Trad, [90] are relevant to the matter which gives rise to the occasion. [91] No narrow limit should be placed upon those propositions.

Dixon J’s reasons in Loveday are those of one of the most eminent judges in the Australian judiciary. Speaking in 1987, Sir Anthony Mason said:

I accept that it would not be proper to seek to extract a binding authority from an opinion expressed in a dissenting judgment. [93] However, the fact that a judgment of the High Court is a dissenting one does not necessarily diminish its persuasive power. It is, of course, not unknown for a dissenting judgment to become the orthodox wisdom. One well-known example is Lord Atkin’s dissent in Liversidge v Anderson. [94] As Gummow and Hayne JJ accepted in Jones v Bartlett, [95] “perhaps we should not make a shibboleth of any requirement that there may be in this context that dissenting judgments should be disregarded; they may at least contain weighty dicta” or “deserve respectful consideration”. [96]

In any event, Dixon J’s statement is supported by other authority. Indeed, as Mr Dawson submitted, understood as an example rather than a statement of principle, Kelly J’s statement in McNicol describes the reply to attack defence the respondents seek to advance. Ms Issa was arguably defending herself against Mr Issa’s attack by showing that Mr Abou-Lokmeh was really the person to whom the finger of blame should be pointed.

In my view the explanation for Coward v Wellington lies in the rationale of the defence to attack qualified privilege which permits “the party attacked free scope to place his case before the body whose judgment the attacking party has sought to affect. In this instance, it is assumed to be the entire public” [97] and the recognition that in vindicating his or her reputation, the respondent may impugn the reputation of even third parties, as long as the response is commensurate with, or relevant to, the occasion. [98]

That is not to say that Coward v Wellington is without critics. Spencer Bower said it “seems difficult to explain or support”. [99] However, in my view, Dixon J’s uncritical recitation of Coward v Wellington and the application of that case in other authorities are weighty considerations in finding that the respondent’s defence to attack qualified privilege is not so unarguable in the General Steel sense as to be struck out prior to trial. It is for the Court on that occasion to undertake the careful scrutiny of all the circumstances of the publication to determine whether the defence is made good.

The primary judge did not err in refusing to strike out the reply to attack defence.

Finally, Mr Abou-Lokmeh complained that the primary judge erred in declining to take account of the proportionality principle in considering the impugned portions of the proposed amended defence and in failing to exercise her discretion to strike them out as a preliminary matter.

It was unclear whether Mr Dibb finally pressed this argument. Without considering how any principle of proportionality might apply in other circumstances, there is no room for its application in my view in this case.

I propose the following orders:

GLEESON JA: I have had the advantage of reading in draft the reasons of McColl JA and Payne JA.

I agree with McColl JA for the reasons her Honour gives that leave to appeal should be refused in relation to the challenge to the primary judge’s refusal to strike out contextual imputations (ii), (iii) and (iv).

I agree with Payne JA for the reasons that his Honour gives that leave to appeal should be refused in relation to the primary judge’s refusal to strike out contextual imputation (i) and the common law defence of qualified privilege on the basis of reply to an attack.

In relation to contextual imputation (i) I would add that the reasons of McColl JA demonstrate that the applicant has not shown an injustice which in the circumstances should be addressed, there being no suggestion that the proposed appeal on this ground raised any issue of principle or question of general importance.

As to the “reply to an attack” defence, since the first ground given by McColl JA is a sufficient basis to refuse leave to appeal, the question of principle sought to be raised by the applicant does not require determination at this stage and, depending upon what occurs at the trial, it may never arise. In those circumstances, and without intending to detract in any way from what her Honour has written on that topic, I would prefer to leave the issue of principle for another day when the facts are either fully found or otherwise agreed for the purposes of determination of a separate question.

PAYNE JA: I have had the advantage of reading in draft the reasons of McColl JA. I agree with her Honour in relation to contextual imputations (ii), (iii) and (iv) that leave to appeal should be refused for the reasons her Honour gives.

While I agree generally with her Honour’s reasons in addressing contextual imputation (i) and the defence of common law qualified privilege based on a reply to attack, I would refuse leave to appeal on those grounds.

Generally speaking, leave will only be granted concerning matters involving issues of principle, questions of general public importance or involving an injustice which is reasonably clear, in the sense of being more than merely arguable: Carolan v AMF Bowling Pty Ltd (Court of Appeal (NSW), 16 November 1995, unrep) per Kirby P; Be Financial Pty Ltd as trustee for Be Financial Operations Trust v Das [2012] NSWCA 164 at [32]-[38].

The only issue in the present case is whether leave should be granted to determine whether the primary judge was correct to decline to strike out parts of a defence. That is not a promising basis as an initial consideration for the grant of leave.

The test to be applied in applications to strike out pleadings was succinctly stated in State of New South Wales v Williams [2014] NSWCA 177 at [71] by Emmett JA (Macfarlan and Simpson JJA agreeing):

There is no rule in defamation proceedings that makes these general principles inapplicable.

The applicant contends that contextual imputation (i) does not fall within the statutory defence of contextual truth contained in s 26 of the 2005 Act. That section provides:

The sole issue is whether contextual imputation (i) is carried “in addition to” the plaintiff’s imputation (a) within the meaning of s 26 of the Act. There is no issue raised by the applicant about the correct test to be applied in determining this question.

As McColl JA demonstrates, for a contextual imputation to be carried “in addition to” another imputation it must differ in substance from that other imputation (Fairfax Media Publications v Zeccola [2015] NSWCA 329 at [42]-[46] and [48] per McColl JA, Macfarlan JA agreeing; [114] per Sackville AJA).

I agree with McColl JA that the primary judge was correct to conclude that contextual imputation (i) is carried in addition to the plaintiff’s imputation (a). The particulars for contextual imputation (i) raise different matters to the defamatory publication and the incident it describes.

The respondent has particularised as part of contextual imputation (i) and will seek to prove at the trial that:

Those matters tend strongly to suggest that contextual imputation (i) is carried in addition to plaintiff’s imputation (a).

Further, imputation (a) pleads a specific act (making a payment to have the respondent murdered) whereas the contextual imputation (i) pleads a general state of mind or condition (the plaintiff wanted the third defendant to be killed). In my view, the two imputations differ in substance.

There is no issue of principle or question of general public importance raised by this ground. It is common ground that the test to be applied in determining this question is found the judgment of McColl JA in Zeccola. It is only the application of that test to particular facts which is in issue.

The question is not one involving an injustice which is reasonably clear. To the contrary, the primary judge did not err in refusing to strike out contextual imputation (i).

I would refuse leave to appeal in relation to contextual imputation (i).

I agree with McColl JA that the respondents’ “reply to attack” defence is not so clearly untenable that it cannot succeed.

In particular, I agree with her Honour that at this stage of the proceedings the respondents may establish at trial that Mr Abou-Lokmeh was complicit in the attack upon Ms Issa. If that is so, the question of whether the defence is maintainable in circumstances where the reply defames a third party will not arise.

That is a sufficient basis to dispose of this matter. The primary judge was correct not to strike out the defence on that basis.

The alternative basis upon which McColl JA considers the matter, involving a close consideration of Coward v Wellington (1836) 7 Car & P 531; 173 ER 234 and the authorities in the High Court and this Court which have discussed that case, is thus unnecessary to determine the present case.

As McColl JA demonstrates, any complete consideration of the question of the defence of common law qualified privilege based on a reply to attack would involve an analysis of Dixon J’s statement in Loveday v Sun Newspapers Ltd [1938] HCA 28; (1938) 59 CLR 503. Whist cognizant that his Honour’s remarks were contained in a dissenting judgement, those remarks were plainly closely considered dicta and should not be departed from by this Court without a powerful reason. I agree with McColl JA that the only reason to do so suggested by the applicant – a first instance decision of the ACT Supreme Court in McNicol v ACT Health Authority [1988] ACTSC 55 – does not provide any basis to depart from Dixon J’s judgement.

A complete consideration of the question would also involve an analysis of the decision of the High Court in Harbour Radio Pty Ltd v Trad [2012] HCA 44; (2012) 247 CLR 31. In that case the plurality referred with approval (at [25]) to the following statement of Dixon J in Mowlds v Fergusson [1940] HCA 38; (1940) 64 CLR 206 at 214-215:

The plurality continued at [35]:

In my view it would be open to a jury to conclude that the matter complained of was sufficiently connected with the content of the attack.

In light of the above I am of the view that the primary judge did not err in refusing to strike out the respondents’ defence of common law qualified privilege based on a “reply to attack”.

The issue raised is not one of principle or a question of general public importance. The questions sought to be agitated by the applicant about the defence of common law qualified privilege based on a reply to attack do not yet arise and may never arise in this case. The only question is whether the primary judge was correct not to strike out an arguable defence. It is inappropriate for this Court to embark upon a summary determination of a legally arguable defence on the contingent basis advanced by the applicant.

The issue does not involve an injustice which is reasonably clear. In my view the primary judge was clearly correct not to strike out the defence of common law qualified privilege based on a reply to attack. I would refuse leave to appeal.

Accordingly I propose the following orders: