Fairfax Media Publications Pty Ltd v Pedavoli

1. [1]

   McCOLL JA: I have had the advantage of reading Sackville AJA’s judgment in draft. I agree with the orders his Honour proposes and, save for what follows on the issue of aggravated damages, with his Honour’s reasons. I have also read Simpson JA’s judgment in draft. I agree with her Honour’s observations on Grounds 11–16. I would also add the following observations about the quantum of the award of damages. I repeat the facts only to the extent necessary for these reasons.

2. [2]

   The primary judge rightly described the defamatory publication as “grossly” defaming the respondent, Melinda Pedavoli. Her Honour identified the conduct of the appellants, Fairfax Media Publications Pty Ltd and Anne Davies, in publishing the matter complained of as “unjustifiable” in the sense discussed in Triggell v Pheeney. [1] That conclusion was based, at least in part, on her Honour’s conclusion that Fairfax’s and Ms Davies’ whole approach to the publication of the matters complained of “smacked of reckless indifference.” [2] Gross irresponsibility would be another description.

3. [3]

   The appellants did not plead any defence either at common law or pursuant to Part 4, Division 2 of the Defamation Act 2005 (NSW). Their only defence was to plead, pursuant to s 18 of the Defamation Act, that the respondent had failed unreasonably to accept their offer of amends. The primary judge rejected that defence. [3] A challenge to that finding was abandoned at the commencement of the hearing of the appeal.

4. [4]

   The facts, in short compass, were that, “by the careless inclusion of incorrect information” in an article (the “matter complained of”) written by Ms Davies, the appellants identified the respondent completely falsely as the teacher who was the perpetrator of unlawful sexual misconduct with a number of boys at St Aloysius College (the “College”). [4] The respondent was falsely identified because she was the only teacher at the College who fitted the description of the teacher referred to in the matter complained of.

5. [5]

   The matter complained of was extensively published at 9 pm on Thursday 30 January 2014 on the internet, in the print edition of the Sydney Morning Herald on Friday 31 January 2014 and in the tablet app edition of the Sydney Morning Herald the same day. It was also made available to Twitter followers of the Sydney Morning Herald’s Twitter account for 30 and 31 January 2014 and to the publisher and operator of The Age website.

6. [6]

   The impact, as the primary judge said, of the publication of the defamation was to:

7. [7]

   The circumstances in which the matter complained of came to be published were detailed by the primary judge. [6] Some aspects deserve repetition.

8. [8]

   There is no doubt the matter complained of concerned a serious topic, but it reported sensational allegations. [7] Those allegations had, in fact, been made against a different teacher at the College (the “teacher concerned”) whose name was suppressed at trial and remains suppressed. That teacher had resigned from the College prior to 29 January 2014, as Ms Davies discovered that day when she spoke to the Rector of the College, Father Hosking. [8] Ms Davies had been told by Ms Kate McClymont (the original source of the information concerning the allegations) that she had heard the teacher concerned was a 29 year old English teacher. [9] By the time of Ms Davies’ conversation with Father Hosking, Ms Davies had identified a number of persons, including the respondent, she thought could fit the description of the teacher the subject of the allegations, which was “an English and/or drama teacher of approximately 28 or 29 years of age”. [10]

9. [9]

   Following her conversation with Father Hosking, Ms Davies called the College switchboard and asked to speak to the respondent. When the person who answered the phone said she would be put through to the respondent, Ms Davies realised that, because she was still at the College, the plaintiff could not have been the teacher concerned. Accordingly, she hung up the phone before speaking to her. [11]

10. [10]

    Because Ms Davies decided to hang up the phone without speaking to the respondent, she deprived herself of:

11. [11]

    Ms Davies then ascertained through another enquiry a name very similar to the correct name of the teacher concerned. She sought to make further enquiries about that person including by accessing the College’s website, Facebook and the telephone directory. However, because she did not have the precise spelling of the teacher concerned’s name, she was unable to obtain “any concrete information” about her. [13]

12. [12]

    As the primary judge found, by 29 January 2014 as a result of phoning the College, Ms Davies must have realised that she was wrong in concluding that the teacher concerned in the “allegations of the most serious sexual misconduct was Ms Pedavoli.” However:

13. [13]

    Ms Davies did not give evidence at the trial. However, the circumstances in which she had engaged in such irresponsible journalism emerged from answers to interrogatories tendered at trial.

14. [14]

    The primary judge found that the matters complained of conveyed all the defamatory imputations the respondent pleaded, they being:

15. [15]

    The appellant faintly sought to challenge the primary judge’s finding that imputation (a) was conveyed. As Sackville AJA has found, that challenge must fail. In my view, it was devoid of merit.

16. [16]

    The effect the publication of such imputations would have on the respondent, a teacher still relatively new to the teaching profession, embarking upon what she wished be a successful career in a profession she had always wanted to pursue and to whom “teaching [was] everything” [16] was, as the primary judge found, devastating. [17]

17. [17]

    In considering the appellant’s ground of appeal that the damages award was unreasonable (a ground asserted in more traditional terms in the appellant’s written submissions as “manifestly excessive”), it is necessary to bear in mind the purposes served by damages awarded for defamation. As the primary judge directed herself, they are:

18. [18]

    These purposes “overlap considerably in reality and ensure that the amount of a verdict is the product of a mixture of inextricable considerations”. [19]

19. [19]

    The nature of the defamation and the circumstances of it may be such as to make vindication of particular importance. [20] It is clear the primary judge was of the view this was such a case. [21] The imputations conveyed by the matter complained of were such as to strike at the heart of the respondent’s reputation relevant to her career.

20. [20]

    Further, the tribunal of fact may award the plaintiff “a substantial sum by way of damages for the plaintiff’s injury apart from the claim for vindication of reputation.” In such circumstances “it will be unnecessary to add a further sum for vindication [as] the award of that substantial sum will in itself serve to vindicate the plaintiff’s reputation.” [22] Such damages “for distress and anguish are the result of a social judgment, made by the [the tribunal of fact] and monitored by the appellate courts, of what, in the given community at the given time, is an appropriate award or … solatium for what has been done.” [23]

21. [21]

    A challenge to an award of damages as manifestly excessive means the appellants do not identify “a specific error of principle or fact”; rather they contend that the sum awarded “is evidently wrong”, such that the primary judge must have failed “properly to exercise [her] discretion in fixing the amount to be awarded”. [24] The question for this Court is whether the primary judge’s award “bespeaks error”. [25]

22. [22]

    If an appellate court is convinced, not that in its own view the amount awarded is too high or too low, but that the amount awarded is so high or so low that it is outside the range of what could reasonably be regarded as appropriate to the circumstances of the case, it should intervene to prevent a miscarriage of justice. [26]

23. [23]

    At the end of the day, Mr Blackburn of Senior Counsel, who appeared for Fairfax and Ms Davies at trial and on appeal with Mr R A Jedrzejczyk, conceded that if the Court was against the appellants on the extent of publication grounds of appeal, it would be hard to press the manifestly excessive ground.

24. [24]

    That was a proper concession. The primary judge’s award was properly placed at a level which approached the statutory cap by reason of any of the inextricable combination of the need for vindication of the respondent’s reputation, the extreme injury to the respondent by way of personal distress and hurt and the aggravation of damages caused by the circumstances of the publication.

25. [25]

    The matters I have set out concerning the imputations conveyed by the matters complained of, the circumstances of the publication and its effect on the respondent and her reputation, serve as background to the primary judge’s consideration of the aggravated damages issue of which the appellants complain. It is in this context that I, with respect, disagree with Sackville AJA.

26. [26]

    The respondent made an extensive claim for aggravated damages, the most significant of which claims concerned the circumstances of the publication of the matters complained of.

27. [27]

    The respondent also complained of aspects of the conduct of the trial she alleged had aggravated the hurt to her feelings, based on the uncontroversial proposition that, in assessing damages for defamation, the tribunal of fact is “entitled to look at the whole conduct of the defendant from the time the libel was published down to the time [of] verdict.” [27]

28. [28]

    The aspects of the conduct of the hearing which the primary judge found to be unjustifiable related to an exchange during Mr Blackburn’s cross-examination of the respondent her Honour set out verbatim. [28] That cross-examination concerned a statement the respondent had made in the affidavit she swore for the purposes of the proceedings expressing her concerns about the potential effect of the publication of the matters complained of on her future employment.

29. [29]

    Mr Blackburn SC challenged the genuineness of the respondent’s concern “that any resume I submitted to another school would immediately raise concerns that a potential paedophile was seeking employment.” [29] He put to the respondent that there was “no basis at all for the concern” she had expressed in that paragraph of her statement, a proposition her Honour accepted the respondent denied. Her Honour also noted that in the course of that exchange the respondent “began crying again.” [30]

30. [30]

    The primary judge concluded that that line of cross-examination amounted to “an accusation of dishonesty or, at best, disingenuity” for which there was no warrant. Rather, in her Honour’s view, the proposition “reflected the weary scepticism of a frequent defendant which … was wholly unjustifiable in the present case.” Her Honour noted that the respondent was “visibly upset by the exchange” and accepted “that it had aggravated the hurt occasioned to her by the defamation”. [31]

31. [31]

    Fairfax and Ms Davies contend that Mr Blackburn’s cross-examination constituted no more than the “mere pursuit of a bona fide defence” [32] which could not be characterised as unjustifiable such as to attract an award of aggravated damages.

32. [32]

    It is uncontroversial that a “bona fide defence raised properly or justifiably in the circumstances known to the defendant” cannot be used to support an award of aggravated damages. However the conduct of the defence may be taken into consideration as improperly aggravating the injury done to the plaintiff, if there is a lack of bona fides in the defendant’s conduct or it is improper or unjustifiable. [33]

33. [33]

    In actions involving damages for loss of reputation there is a “highly subjective element” in respect of which the plaintiff, “in case the libel, driven underground, emerges from its lurking place at some future date … must be able to point to a sum awarded by a jury sufficient to convince a bystander of the baselessness of the charge.” [34] The “lurking place” to which Lord Hailsham referred, is that part of “the grapevine” along which the defamation may be said to have spread and, further, where it may rest and strike the plaintiff again at an unexpected time. It is in that respect that the plaintiff must be able to demonstrate the “baselessness of the charge”. [35]

34. [34]

    The fears the respondent expressed in her statement upon which she was cross-examined reflected a concern about the grapevine effect leading to the defamatory imputations lurking in such places, possibly unbeknownst to her, as could adversely influence her future employment prospects. There could be no doubt, in my view, that objectively that was a genuine and legitimate concern on her part, or would have been on the part of anyone else in her position. The primary judge was best placed to evaluate that issue insofar as the respondent herself was subjectively concerned.

35. [35]

    Mr Blackburn’s cross-examination was a grave attack in circumstances where, as Mr B R McClintock of Senior Counsel, who appeared for the respondent with Mr M F Richardson at trial and on appeal, submitted, the respondent had been wrongly labelled a paedophile. Once it was accepted that the matters complained of identified the respondent “particularly in the circles in which [the respondent’s] reputation counts most”, Mr Blackburn’s cross-examination could not, in my view, be characterised as no more than vigorous persistence in a bona fide defence. [36]

36. [36]

    The cross-examination amounted to an unjustifiable (and inexplicable) denial of a fundamental principle of the law of defamation concerning the likely spread, in this case, of baseless and grave charges. It accorded no weight to the entirely legitimate context in which the respondent expressed her concern: the reckless publication by the appellants of false and grossly defamatory imputations likely to identify her particularly to those in her professional sphere. Rather, as the primary judge found, by his cross-examination Mr Blackburn accused her of “dishonesty or, at best, disingenuity.” [37] That cross-examination was clearly capable of increasing the hurt to the respondent’s feelings by accusing her of a different sort of misconduct to that the appellants had already falsely accused her of in the matter complained of.

37. [37]

    The primary judge had an overwhelming advantage over this Court in characterising both Mr Blackburn’s cross-examination of the respondent and, in that light, the real effect of the defamation and his conduct on the respondent. [38] This Court should be slow to interfere with a finding of this nature.

38. [38]

    In my view, the primary judge was entitled to find that the cross-examination was unjustifiable in the sense referred to in Triggell v Pheeney.

39. [39]

    SIMPSON JA: The relevant facts and circumstances are set out in the judgment of Sackville AJA, which I have read in draft. With respect to Ground 10, I agree with his Honour, for the reasons he gives, that the inference the primary judge drew from the Google Analytics document was one well open to her.

40. [40]

    I wish to make some observations about Grounds 11-16 (concerning the identification of the respondent by readers of the articles after having access to the College’s website). And, having regard to the difference of opinion between Sackville AJA and McColl JA, it is necessary that I state my views on one aspect of the issue of aggravated damages.

41. [41]

    Grounds 11-16 arise out of a single proposition of law advanced on behalf of the appellants. That proposition, stated starkly, is that, in order for a publication that does not name the plaintiff to be defamatory of him or her, it is necessary that the identifying facts or circumstances be known to the recipient at the time of publication. It is not sufficient that subsequent enquiries reveal information that permits the identification to be made. (For the meaning of “publication” in this context, see below [44]-[45].)

42. [42]

    There was no issue that the respondent was identifiable from the information contained in the article. Nor was there any issue that she was in fact identified by some readers. The question that gives rise to these grounds concerns the number of readers who were shown to have done so, and, therefore, the extent of the publication of the defamatory imputations; and, specifically, whether it is legitimate, in determining the extent of publication, to take into account readers who identified the plaintiff as a result of information obtained subsequent to publication.

43. [43]

    The appellants’ proposition, as stated above, is important to their argument on the quantification of damages. That is because the primary judge admitted evidence that, on the day of publication, there had been a surge in access to the College website. The website revealed that the respondent was the only teacher of both English and drama at the College, unequivocally (but wrongly) identifying the respondent as the teacher the subject of the articles. The primary judge drew an inference that the reason for the sudden interest in the College website was an intention on the part of those visitors to ascertain the identity of the teacher concerned. That enlarged, significantly, the extent of the defamatory publication.

44. [44]

    A preliminary diversion is in order. To understand what follows, it is necessary to understand the meaning of “publication” in the law of defamation. For that purpose, it does not bear its common meaning - it might reasonably be thought that, in ordinary language, publication (for example, of a newspaper or a radio program) takes place when the newspaper is printed or distributed, or when the radio program is broadcast. For the law of defamation, publication takes place when the newspaper is read and comprehended, or the radio broadcast is heard and comprehended. In other words, publication depends upon the content being received. That is made clear (although in the context of the location of the commission of the tort of defamation) in Dow Jones and Co Inc v Gutnick. [39] And, in a judgment that ought not to be allowed to lie fallow, Palmer J discussed what he plainly saw as the peculiarities or idiosyncrasies of the meaning of “publication” for the purposes of the law of defamation. [40]

45. [45]

    What can be drawn from the cited passages in Gutnick, is that a defamatory imputation is not published in the sense used in the law of defamation until its defamatory meaning is received by the recipient. Thus, where the plaintiff is not named in the matter complained of, no defamatory imputation is published until a recipient of the publication makes the necessary identifying connection. These basic principles need to be understood in order to make sense of what follows.

46. [46]

    A further preliminary observation that may here be made is that whether a plaintiff has been identified by any recipient of the matter complained of is a question of fact to be determined on the evidence adduced.

47. [47]

    I return now to the foundational proposition advanced on behalf of the appellants. It is important to be clear how senior counsel for the appellants stated it. It was that, for the defamatory imputations to have been conveyed, it was necessary that knowledge of the identifying fact - that is, in this case, that the respondent was the only teacher of both English and drama at the College - was in the possession of the recipient at the instant he or she read the article, and not later. Thus, on the facts of this case, a parent who read the article after the children had left for school, but who questioned them after school as to the name of the English and drama teacher, would not be a person to whom the defamatory imputations were published. A parent who telephoned another parent or, indeed the College, or any other source of information, to make such an enquiry would, similarly, not be a person to whom the defamatory imputations were published. Only those readers who knew, at the moment of reading the article, that the respondent was the only teacher of English and drama at the College could be persons to whom the defamatory imputations were published. Those who made the connection after visiting the College’s website were not such persons.

48. [48]

    Ultimately, under considerable pressure from the Court, senior counsel was driven to accept that, for example, a parent who read the article over the breakfast table and immediately asked a child the name of the teacher of English and drama, and was given the respondent’s name, would be a person to whom the defamatory imputations were published. His explanation for this modest concession lay in the relative contemporaneity of the receipt of the publication and the receipt of the identifying information. The two are not, however, truly contemporaneous, and, in my opinion, the concession exposes the flaw in, and even destroys, the argument. Once it is conceded that a short gap in the time between receipt of a publication and receipt of the facts that identify the person the subject of the publication does not preclude a conclusion that a defamatory imputation is published, the question becomes one of fact and degree. What is sufficient to establish that a reader identified the respondent as the subject of the publication?

49. [49]

    It might be useful, before attempting any answer to that question, to explore the foundation for the appellants’ proposition. That foundation lies in the decision of the English Court of Appeal in Grappelli v Derek Block (Holdings) Ltd. [41] The relevant passage is in the judgment of Lord Denning MR and is as follows:

50. [50]

    It is important to be aware of the relevant facts in that case. A famous jazz musician (and a not so famous one) had been booked (without their knowledge) by their agent (the defendant) to perform on several days at different locations in England. On learning of the bookings, the plaintiffs required the agent to cancel them. The agent telephoned the relevant people, and told them that the concert had been cancelled because Mr Grappelli was very seriously ill in Paris and was unlikely ever to tour again. The agent did this in September 1976. The statement that Mr Grappelli was ill was entirely false (but not defamatory). Subsequently (in November 1976) notices appeared in various newspapers to the effect that the plaintiffs would be performing in various concert halls, not those at which the agent had booked them, but on or close to the dates of those bookings.

51. [51]

    The plaintiffs sued, alleging that, by reason of the notices in the newspapers, those who had received the initial communication from the agent that the concerts had been cancelled because of Mr Grappelli’s illness would have inferred that the plaintiffs had given a false reason for the cancellations. It was in this context that Lord Denning (together with Templeman and Dunn LJJ) ordered that the statement of claim, so far as it pleaded a cause of action in defamation, should be struck out. There was a gap of two months between the publication of the (false, but not defamatory) statements concerning Mr Grappelli’s health, and the events (publication of the newspaper notices that the plaintiffs would be performing) that were said to transform the original statements into defamatory ones.

52. [52]

    Three features distinguishing Grappelli from the present case may immediately be noted. The first is that Grappelli was a case involving, not identification, but true (or legal) innuendo (as Lord Denning stated). The second is that the facts that were said to render the publication defamatory were not known publicly prior to or at the time of the publication the subject of the proceedings (and, indeed, did not exist at that time). The third is that (as is often the case where true innuendo is alleged), absent knowledge of the facts said to render the publication defamatory, the statements (that Mr Grappelli was seriously ill and might never tour again) were entirely innocuous and, of themselves, incapable of conveying any defamatory imputation.

53. [53]

    The last of these is what marks the difference between defamation by true innuendo and defamation of a person not named in the relevant publication, but identifiable to those in possession of relevant identifying information. In the case of the latter, the defamatory imputation is contained in the publication; the extrinsic fact that makes it defamatory of a particular person is the identification of that person. In the former, as in Grappelli, there may be nothing defamatory in the publication, except to those in possession of extrinsic facts that cast the publication in a different light. In Grappelli, both Lord Denning MR and Dunn LJ quoted from a New Zealand decision, [43] Lord Denning as follows:

54. [54]

    Dunn LJ quoted the following passage:

55. [55]

    That Grappelli concerned true innuendo and not identification is significant; because (as was the case in Grappelli) circumstances that come in to existence after the publication may put a defamatory complexion on an otherwise innocent publication. That was what gave rise to the strong statements of principle in Grappelli. In the case of identification, the identifying facts are in existence at the time of publication, although they may not be known to some recipients.

56. [56]

    Lord Denning referred to the dictum in Grappelli only three months later, in Hayward v Thompson. [46] This was a case that did involve a publication that conveyed defamatory imputations but did not name the plaintiff, and was, therefore, like the present, a case in which the issue was identification. The newspaper reports the subject of the proceedings concerned a notorious allegation of solicitation to murder. Articles were published in two consecutive editions of a Sunday newspaper. In the first it was asserted that two more people had become involved in the investigation, one of whom was “a wealthy benefactor of the Liberal party” (a description that fitted the plaintiff) but who was not named. The second article, published in the newspaper one week later, did name the plaintiff. The plaintiff sued on both articles. He claimed that readers who did not identify him on reading the first article were able to do so retrospectively on reading the second.

57. [57]

    In its defence, the newspaper publisher relied upon the principle stated in Grappelli. Lord Denning MR quoted the passage from Grappelli, and then said:

58. [58]

    His Lordship went on to discuss publication in the context of the intention of the publisher to refer to the plaintiff. He concluded:

59. [59]

    The question has also arisen in this State, in Baltinos v Foreign Language Publications Pty Ltd. [48] A Greek language newspaper, published in Sydney, referred in defamatory terms to a class, or group of people (“slickers”) who financially exploited “foreign visitors” who wished to seek permanent residence in Australia. No individual “slicker” was named. However, the article then referred to a television program to be aired that night, and finished with:

60. [60]

    Mr Baltinos sued on both the newspaper publication and the television program. It is only the newspaper publication that is of present relevance. Mr Baltinos claimed that the recommendation to readers to view the television program was an invitation to ascertain the identity of the persons the subject of the article by reference to the program. Hunt J (as he then was) said:

61. [61]

    Hunt J spent a good deal of his judgment discussing what he saw as recent, and contrary to principle, incursions into the established understanding that identification of a person the subject of a defamatory publication is to be ascertained objectively, and not subjectively (by reference to the intention of the publisher). [50] Having done that, Hunt J turned his mind to what he called an extension of the Grappelli principle in Hayward v Thompson. He considered (and ultimately accepted) a submission that the plaintiff was entitled to rely upon:

62. [62]

    He said:

63. [63]

    I note that Hunt J considered that the qualification to the Grappelli principle (perhaps “dictum” is a better word, since it has been taken to extend well beyond the circumstances of the case) made in Hayward v Thompson was restricted to subsequent publications by the same defendant. I do not read Hayward so restrictively. The predominant and determinative question in Hayward was whether the defendant had, in the first article, intended to refer to the plaintiff. The second article was accepted as relevant to cast light on that question. I do not read Lord Denning as pronouncing an exception to the principle stated in Grappelli: rather, he “readily accepted” that that principle did not apply “where the words were defamatory on the face of them, and the only question was one of identification”. [53] The “principle” was therefore limited to cases of true innuendo.

64. [64]

    Hunt J further extended what he saw as the restricted principle in Hayward so as to apply to the subsequent publication by another publisher where the first publisher has expressly directed the attention of the recipient to that subsequent publication.

65. [65]

    But, it might be asked, why should the extension stop there? In fact, it has not. In Strasberg v Westfield Ltd, [54] the question arose again. The facts are somewhat sparsely stated in this interlocutory and apparently extempore judgment, one of a series delivered in the Defamation List of the Common Law Division of the Supreme Court. It seems that the publication the subject of the judgment was an internal letter to retailers in a shopping centre written by the shopping centre’s management. It contained statements which the plaintiff alleged were defamatory of her. However, it named “Marianne Schwartz”. The plaintiff sued by the name of Marianne Strasberg. The letter referred to a television current affairs program, to be aired that night which (it seems) named the subject as Marianne Strasberg. It concluded by inviting retailers in the shopping centre to refer customer enquiries to a named person in centre management. Levine J said:

66. [66]

    This decision represents a further encroachment into the absolute nature of the “principle” as stated in Grappelli. It was now not necessary that the subsequently identifying information be published by the same publisher (Hayward); it was not necessary that the allegedly defamatory publication expressly invite attention to identifying material published by another publisher (Baltinos); now, it was sufficient that the allegedly defamatory publication nominate a publication reference to which would identify the plaintiff.

67. [67]

    The proposition advanced on behalf of the appellants gains some small support from a 1952 (that is, predating Grappelli and its progeny) decision of the Full Court of the Supreme Court of NSW in Cross v Denley. [56] The facts in this case were as follows. The plaintiff was a printer carrying on business in Auburn under the name “X-Press Printery”. The defendant was also a printer, carrying on business under a very different name. The defendant was asked to print some pamphlets of a politically charged nature, critical of a currently showing film, and likely to engender hostility in some quarters. Instead of putting his own business name on the pamphlets, the defendant printed “Express Printery, Lidcombe”. Some of the pamphlets were showered on the audience in the cinema where the film was being shown. The following day, the incident was reported in a newspaper, which added that the pamphlets bore the imprint of “Express Print, Auburn”. Some of the plaintiff’s customers who read the newspaper report identified the plaintiff. There appears to have been no evidence that any of the cinema attendees had done so.

68. [68]

    The plaintiff sued the defendant on the publication of the pamphlets in the cinema, on the basis that the subsequent newspaper report allowed recipients of the pamphlets in the cinema to identify him as the publisher of the politically controversial material.

69. [69]

    At first instance, the plaintiff succeeded. In order to do so, he had to prove that some recipients of the pamphlets were able to identify him as the proprietor of the “X-Press Printery”. He sought to do this by evidence that those of his customers who (the following day) read the newspaper and knew that he was the proprietor of the “X-Press Printery” linked him with the distribution of pamphlets in the cinema. Owen J (with whom Street CJ and Herron J agreed) held that he could not do that by reference to the later publication of the name of his business in the newspaper; he said: [57]

70. [70]

    A seemingly contrary view was taken in Ware v Associated Newspapers Ltd. [58] The plaintiff (the well-known swimmer, Dawn Fraser, suing in her married name) sued upon three separate articles published in the same newspaper, the first on 1 March 1965, the second and third on the following day. She was not named in the first of the articles, but she was named in the second and third. Her action on the first and third publications failed. It was inferred that the reason for her failure on the first count was the absence of identification of her as one of the persons referred to. Walsh JA said that the judge’s directions to the jury were:

71. [71]

    The present appeal was argued on the basis that the statements of Lord Denning in Grappelli established a principle, applicable in this jurisdiction, that publication concerning an unnamed subject does not convey a defamatory imputation unless the recipient, at the instant of receipt of the publication, identifies that person. That principle, as it was put to this Court, allowed for no interlude between receipt of the publication, and identification. The further proposition was that some limited exceptions have been grafted onto the original principle, in Hayward, and Baltinos, but that the principle remains.

72. [72]

    In my opinion, the original proposition cannot be sustained.

73. [73]

    There are two components to the Grappelli principle, as stated by Lord Denning. The first is uncontroversial - a cause of action in defamation is complete on publication of the defamatory imputation. What that means is that, at that point, the person the subject of the defamatory imputation is entitled to sue, and (subject to the facts) to succeed. Nothing more is needed. The second component - that any extrinsic facts that render an otherwise non-defamatory publication defamatory must be known to the recipient at the time the publication is received - although stated in the most general terms, must be seen in the context of the facts there under consideration, and in the light of subsequent refinements to the general principle. The most important of the facts under consideration in Grappelli was that the circumstances relied upon to render the publication defamatory did not come into existence until two months later. The subsequent refinements to the general principle allow for post-publication identification in a number of circumstances.

74. [74]

    That the second component of the Grappelli “principle” is open to doubt was made clear in Elaine Chase v News Group Newspapers Ltd. [61] There, Eady J drew a distinction, exemplified in Hayward, between

75. [75]

    Similar doubt had been expressed a decade earlier in Ballantyne v Television New Zealand Ltd, [62] by Williamson J who said:

76. [76]

    Notwithstanding the apparent acceptance by Hunt J of the commanding (and binding) force of the statement in Grappelli, I am of the view that that statement has been read as standing for a more absolute rule than it can now justify, at least in cases where the issue is identification of the plaintiff. One of the beauties of rejecting that view of Grappelli is that it accords with common sense. It is a natural human response, when confronted with serious allegations against an unnamed person, to enquire as to the identity of that person. That may be done in the ways alluded to above - by enquiries of those who might be supposed to have the relevant information - or, in the 21st century, by access to electronic media, or by a variety of other ways. Hayward and Baltinos ought not to be seen as representing exceptions to a “rule” as laid down in Grappelli. Rather, they should be seen as recognition that the rule was too broadly stated in the first place (as, indeed, Lord Denning acknowledged in Hayward, when he “readily accept[ed]” that the statement did not apply where the only question was one of identification).

77. [77]

    Indeed, the editors of Gatley, [63] in another footnote [64] have doubted whether Grappelli and Hayward are “truly reconcilable”.

78. [78]

    It is not, I venture to suggest, necessary, in order for subsequently acquired information to permit identification of a plaintiff, that a publication contain within it an express or implicit invitation to the recipient to have resort to some particular source of external information, although, where that has happened, the case is clear, as in Baltinos and Strasberg; nor is it necessary that subsequent publication be that of the original publisher. In virtually every case where identification is in issue, it may be supposed (depending, perhaps, at least in part on the level of salaciousness, or gravity of the allegations) that recipients will seek (with a greater or lesser degree of vigour) to identify the subject.

79. [79]

    It was, therefore, unnecessary, but not necessarily wrong, for the primary judge to locate, in the articles, an express or implied invitation to enquire to the identity of the teacher concerned. As I have indicated, such an invitation is, in any event, implicit in almost any defamatory publication that does not name its subject.

80. [80]

    The invitation perceived by the primary judge was not limited to enquiries by way of electronic media, calling for new responses to deal with new technology. Electronic media, in this case, was, on the evidence, a readily accessible (and accessed) source of the identifying information, but it was not the only potential source, and the primary judge did not so limit her observations. She said that the articles naturally prompted the question: “Do you know who she is?”. That question could be asked, and answered, in a variety of ways. The focus was on electronic media enquiries because the evidence provided a clear inference that readers had in fact resorted to that source.

81. [81]

    It is not, in my opinion, the law that a publication which does not name the plaintiff conveys defamatory imputations only to those recipients who, at the moment of first receipt of the publication, know the identity of the person referred to. Whether or not the plaintiff has been identified is a matter of fact, and will depend upon the circumstances of each case. It was to deal with the particular circumstances of their own cases that the judgments in Hayward, Baltinos and Strasberg developed (rather than provided exceptions to) the “principle” as stated. It may well be that questions of remoteness arise, but that is not this case.

82. [82]

    I agree that Grounds 11-16 ought to be rejected.

83. [83]

    The cross-examination that gives rise to this issue was related to the respondent’s evidence (given in affidavit form) that she feared that the damage occasioned by the defamatory imputations would come back to haunt her in the event that she later applied for employment in some other school. Only one question directly challenged the assertion that she had that fear: it was a question that, in the primary judge’s view “amounted … to an accusation of dishonesty or, at best, disingenuity”. I agree with McColl JA that the question was unjustifiable; in fact there was no apparent factual basis for it to be put. There is no suggestion anywhere in the proceedings that the respondent had given evidence in such a way as to produce concerns as to her veracity.

84. [84]

    The remainder of the cross-examination went to the basis for her asserted fear. It could be legitimate to seek to establish that a feared consequence of the publication of defamatory imputations is unlikely, as a matter of reality, to occur, but that is so if the plaintiff puts her fear as not only genuine, but also realistic. In other words, if she put the feared consequence as a probable result of the defamation, therefore sounding in damages. There is nothing in the proceedings before this Court to suggest that the respondent’s fear was put on that basis. The question challenging the veracity of the respondent’s evidence of her fear, was no more than, in the vernacular, a cheap shot. It should not have been asked. It caused distress. The subsequent questioning, concerning the legitimacy of her belief, seems to me to be largely irrelevant to the issues in the trial, but prolonged the respondent’s distress.

85. [85]

    I agree with McColl JA that the primary judge was in the best position to assess both the legitimacy of the questioning, and its effect. There is thus no need to attempt to discern the impact the findings may have had on the quantification of damages.

86. [86]

    I agree that the appeal ought to be dismissed.

87. [87]

    SACKVILLE AJA: This is an appeal against a decision of a Judge of the Common Law Division (McCallum J) awarding the respondent $350,000 as damages in respect of publications which, in her Honour’s words, “grossly” defamed the respondent. [65]

88. [88]

    The primary Judge summarised the nature of the case and the outcome as follows: [66]

Background

1. [89]

   In early 2014, the respondent was a teacher at St Aloysius College (College), a Catholic high school for boys. At that time, the respondent was in her late 20s and was the only teacher at the College who taught both English and drama at the College.

2. [90]

   In December 2013, the College became aware of allegations that a female teacher had engaged in sexual relations with a number of Year 12 boys. The College undertook an internal investigation and, as a result, asked the teacher against whom the allegations were made to resign. The teacher did so. I refer to that teacher, as did the primary Judge, as “the teacher concerned”. [67]

3. [91]

   The respondent is not the teacher concerned. The teacher concerned was older than the respondent and taught only English.

4. [92]

   At various times during January 2014, the second appellant (Ms Davies) investigated rumours about sexual misconduct at the College. Ms Davies wrote an article concerning the allegations. The article did not name the respondent but said that the teacher concerned was in her late 20s and taught English and drama at the College.

5. [93]

   At 8.57 pm on Thursday, 30 January 2014, the article was made available for downloading on Fairfax’s website, www.smh.com.au. The link to the article was headed “Sydney teacher quits amid Year 12 sex claims”. This publication was described in the pleadings as “the second matter complained of”.

6. [94]

   The article was duly published on pages 2 and 3 of the print edition of the Sydney Morning Herald on Friday, 31 January 2014 under the headline “Female teacher quits top Catholic school after claims of sex with boys”. The article was also published that day on the Sydney Morning Herald Tablet App. This version of the article was “the first matter complained of”.

7. [95]

   Apart from the difference in the headline the content of the articles was the same. The article was as follows:

8. [96]

   At 9.11 pm on 30 January 2014, the respondent received a text message from a colleague at the College with a link to the second matter complained of. Shortly thereafter, the respondent received a telephone call from a friend from her school days who alerted her to a group message on Facebook concerning the article. The Facebook message involved people who knew the respondent.

9. [97]

   From 7:47 am on 31 January 2014, the respondent began receiving messages from friends asking about the article. She had a discussion by way of messages on Facebook with one friend in which she explained that she was not the teacher concerned. At 8.36 am on that day, the friend sent a text message to an acquaintance at the Sydney Morning Herald, informing him that the story had wrongly implicated the respondent. At approximately 8.42 am, Fairfax removed the references in the online version of the article to the teacher’s age and the subjects she taught.

10. [98]

    Fairfax estimated that on 30 January 2014, there were 18,693 unique page views of the second matter complained of in New South Wales and a further 2,831 unique page views in other States and Territories. The print edition of the Sydney Morning Herald on 31 January 2014 had a distribution of 142,233 copies in New South Wales and an estimated readership of 769,000 for that day. On 31 January 2014, there were 9,129 unique page views of the first matter complained of, utilising the Sydney Morning Herald Tablet App.

11. [99]

    In addition to the publications already referred to, the second matter complained of was made available on 30 and 31 January 2014 to followers of the Sydney Morning Herald’s Twitter account. An estimated 1,155 persons read the article via Twitter.

12. [100]

    On 3 February 2014, solicitors for the respondent wrote to the Sydney Morning Herald asserting that the articles clearly and falsely identified the respondent as the offending teacher. The letter advised that defamation proceedings would be commenced. The letter requested that, in order to mitigate damages, Fairfax should publish an apology in the form set out in the letter.

13. [101]

    The solicitor for Fairfax responded on 5 February 2014 with an offer to make amends under Pt 3 of the Defamation Act 2005 (NSW) (Defamation Act). The offer included compensation in the sum of $50,000 and an apology in similar terms to that proposed by the respondent’s solicitors, but with some words omitted.

14. [102]

    On 13 February 2014, an apology was published on page 2 of the Sydney Morning Herald in the terms proposed in Fairfax’s offer to make amends. An apology was also posted on the Sydney Morning Herald’s website via a hyperlink to the word “Apology”. On the same day, the amended online article was removed in its entirety.

15. [103]

    The respondent’s solicitors filed the statement of claim on 19 February 2014. Unusually for defamation cases, the statement of claim was not amended. The trial took place over three hearing days (27-29 October 2014) and the Primary Judgment was delivered on 27 November 2014. Thus the proceedings were completed at first instance approximately nine months after they were commenced.

Primary Judgment

1. [104]

   The primary Judge found that both matters complained of identified the respondent as the teacher who had resigned from the College because of the identifying details included in the articles. [68] Her Honour also found that the articles conveyed three defamatory imputations, as follows: [69]

2. [105]

   The primary Judge first addressed the appellants’ defence based on the offer to make amends. The appellants claimed that the offer complied with s 18(1) of the Defamation Act and that the failure of the respondent to accept the offer therefore constitutes a defence to her claim. [70] The primary Judge rejected this defence on several grounds. [71] The appellants’ amended notice of appeal challenged the rejection, but the challenge was abandoned shortly before the appeal was heard.

3. [106]

   Since the appellants have abandoned this challenge, there is no need to recount in detail the primary Judge’s careful reasons for concluding that the offer to make amends did not comply with the statutory criteria. It is convenient to note, however, that one reason for her Honour’s finding that the offer was not “in all the circumstances … reasonable”, [72] was that the offer to pay compensation of $50,000 was inadequate.

4. [107]

   The primary Judge considered that the seriousness of the defamatory imputations conveyed by the articles required an offer of monetary compensation if the offer was to be regarded as reasonable. [73] An offer of compensation would have been required even if the appellants’ apology had been of equal prominence to the matter complained of and carefully targeted to reach the same audience (which her Honour found was not the case). Her Honour also considered that at the time the offer was made the appellants should have appreciated that the defamation was such as to cause extreme hurt and distress to the respondent. [74] Her Honour continued as follows: [75]

5. [108]

   The primary Judge proceeded to assess damages. Her Honour set out in uncontentious fashion the basic principles governing an award of damages for defamation. [76] She noted that the statutory cap for damages for non-economic loss, adjusted in accordance with s 35(1) of the Defamation Act, was $366,000. [77]

6. [109]

   In her Honour’s view, the assessment of damages had to begin with the proposition that but for the fact that the respondent had not been named in the article, the seriousness of the defamation and the degree of hurt suffered by the respondent placed the case within the worst class of defamation. It called for an award of damages at or near the statutory cap, subject to relevant factors of mitigation and aggravation. [78]

7. [110]

   The appellants had submitted to the primary Judge that the case was towards the lower end of the scale in terms of harm to the respondent’s reputation because the number of people who would have identified her as the teacher concerned was “quite limited”. [79] Her Honour said that this submission raised a threshold question of principle. [80]

8. [111]

   There was no issue that to anyone who knew the identifying facts, the description in the article pointed unequivocally to the respondent. However, the appellants had contended that the respondent’s cause of action was limited to the persons who knew of the identifying facts at the time of publication. This was said to follow inexorably from the general principle that a cause of action for defamation is complete upon publication of the matter complained of. Accordingly, so the appellants had argued, the publication is limited to those persons who had the knowledge required to understand the matter in its defamatory sense at the time of publication. Thus the respondent could not recover damages in respect of a publication to persons who only acquired the requisite knowledge at a later time, such as by undertaking searches on the internet.

9. [112]

   The primary Judge pointed out that the issue was significant because there had been a spike of interest in the College’s website when the article was published online, suggesting that people accessed the website to ascertain the identity of the teacher described in the article. At the time the articles were published, the respondent was the only teacher identified on the website as teaching drama and English (the description in the article). The issue therefore was whether the respondent could rely on a reader’s knowledge of extrinsic identifying facts, where knowledge of those facts was acquired after the matter complained of had been published. [81]

10. [113]

    After referring to the judgment of Hunt J in Baltinos v Foreign Language Publications Pty Ltd [82] her Honour quoted two remarks made by Eady J in the English decision of Chase v News Group Newspapers Ltd. [83] Eady J’s observations were as follows: [84]

11. [114]

    The primary Judge continued as follows: [85]

12. [115]

    A document referred to as the Google Analytics document, produced on subpoena by the College, demonstrated that there had been a dramatic spike of interest in the staff page of the College’s website following publication of the article. [86] Her Honour was satisfied that a large number of people consulted the website for the purpose of ascertaining the identity of the teacher described in the article. [87] It followed that the extent of publication was not as limited as the appellants had suggested.

13. [116]

    The appellants had also submitted that even among people who appreciated that the article referred to the respondent, only a limited number would have believed that she was the teacher concerned and of those, only a smaller subset would have laboured under the misapprehension for any length of time. Her Honour rejected the submission:

14. [117]

    The primary Judge found that the respondent had suffered extreme hurt to her feelings and had experienced physical distress after seeing some particularly hurtful comments on Youtube clips. [88] Her Honour also found that the hurt to feelings suffered by the respondent was continuing. [89]

15. [118]

    The primary Judge accepted that the prompt removal of the incorrect detail from the online article on the morning of 31 January 2014 and the apology “stemmed the harm to [the respondent’s] reputation”. [90] But, contrary to the appellants’ submission, the respondent’s recent promotion to Acting Assistant Head of English at the College said little about the likely reach of the defamation. [91]

16. [119]

    The primary Judge considered that the conduct of the appellants warranted an award of aggravated damages: [92]

17. [120]

    Her Honour also considered that, in one respect, the cross-examination of the respondent was unjustifiable. Senior counsel for the appellants had put to the respondent that she did not have a genuine fear (as she claimed) that if she sought employment at another school concerns would be raised that a potential paedophile was seeking a position. The primary Judge thought that there was no warrant for putting to the respondent, in effect, that she was being dishonest or at best disingenuous. The cross-examination had aggravated the hurt occasioned to her by the defamation. [94]

18. [121]

    The primary Judge concluded as follows: [95]

19. [122]

    After allowing for interest of $8,685, her Honour gave judgment for the respondent in the sum of $358,685.

Reasoning

1. [123]

   As I have noted, the appellants did not press all grounds of appeal. It is convenient to address the remaining grounds in a different order to that identified in the amended notice of appeal.

2. [124]

   Mr Blackburn SC, who appeared with Mr Jedrzejczyk for the appellants, submitted, albeit rather faintly, that the primary Judge should not have found that the material complained of conveyed imputation (a), namely that the respondent was “a sexual predator who used boys at the school at which she taught for sexual gratification”. Mr Blackburn argued that the articles were not “sensational in the tabloid sense” and would not have been understood by ordinary, reasonable readers [96] as conveying that the respondent was a sexual predator. He submitted, as he did to the primary Judge, that imputation (a) was a “pleader’s flourish” and failed to recognise that boys in years 12 would most likely be 17 or 18 years old and thus above the age of consent.

3. [125]

   The primary Judge, in rejecting this submission, pointed out that the story was not presented as an account of the lawful sexual encounters of young men. It related to the conduct of the teacher concerned with “boys” she was teaching and with whom she had had inappropriate relationships. The articles stated as facts that the teacher concerned had resigned after an internal investigation and that the investigation had “revealed” that the teacher had engaged in “inappropriate relationships” with a number of boys and had sex with at least two. The articles also asserted that the findings of the investigation had been reported to the police and to the Ombudsman, who had oversight of child protection laws. In her Honour’s view, the ordinary reasonable reader would readily perceive that the purpose of child protection laws is to protect children from sexual predators. Such a reader would not understand the report to relate to “an unexceptional cluster of lawful sexual encounters with different young men”, but to conduct that was “predatory, sexual and self-interested”.

4. [126]

   I agree with the primary Judge’s analysis. It is correct, as Mr Blackburn submitted, that the ordinary reasonable reader is to be taken as a person of fair average intelligence. Such a reader is not perverse, morbid or suspicious of mind or avid for scandal. [97] But the articles reported as fact the conduct of a teacher in a position of responsibility and trust towards her students. The fact that the articles were not overtly sensational in character does not detract from the impression that would be taken away by a reasonable reader, namely that the teacher concerned had exploited and seriously abused her position of trust and responsibility to gratify her sexual desires. That behaviour would be understood by the reader to be predatory, even taking into account the likely ages of the boys.

5. [127]

   While the appellants did not pursue their challenge to the primary Judge’s decision to admit into evidence the Google Analytics document prepared by the College, Mr Blackburn did not abandon Ground 10 in the amended notice of appeal. Ground 10 challenges the inference drawn by the primary Judge from the Google Analytics document, namely that there had been a dramatic spike of interest in the College’s staff page on its website in the immediate period after the articles were published. [98] He did not elaborate in oral argument on the appellants’ very brief written submissions on this issue.

6. [128]

   The only argument advanced in the written submissions, other than the contention that the Google Analytics document was inadmissible, was that the “dramatic spike in attention” to the staff page occurred at about the time the school year recommenced. This was said to be a “natural time” for consulting the staff list and thus the apparent spike in visitors to the website did not justify an inference that the spike was related to the matter complained of.

7. [129]

   The Google Analytics document recorded the number of sessions, users and page views for the College’s website. While there was no evidence interpreting the document, it clearly indicates that on 31 January 2014, there were 2,921 visitors to the College’s website and, of these, 1,289 visited the Senior School Staff List on that day. These figures were vastly greater than the average for the three month period covered by the Google Analytics document and very much greater than for the days preceding and following 31 January 2014.

8. [130]

   The 2014 school year commenced on 28 January 2014. The spike in visits to the College’s website did not occur until the day the newspaper article was published. The appellants’ explanation for the spike is not plausible. The primary Judge was justified in drawing the inference she did.

9. [131]

   Mr Blackburn presented the argument on Grounds 11-16 in the amended notice of appeal together. The grounds are as follows:

10. [132]

    Mr Blackburn submitted that the primary Judge erred in adopting the approach taken by Eady J in Chase v News Group Newspapers since that approach involved an unwarranted extension of the principle stated in Baltinos. In particular, holding a publisher liable for a publication merely because it was reasonably foreseeable that the plaintiff could be identified through inquiries undertaken by readers of the publication, was an unwarranted extension of orthodox principle. According to Mr Blackburn it is inappropriate to import into the law of defamation concepts applied in other areas of the law, such as the law of negligence. Further, importing the concept of reasonable foreseeability into this area of the law is inconsistent with the fundamental proposition that a cause of action in defamation arises at the time of publication.

11. [133]

    Mr Blackburn argued that the primary Judge’s approach represents (variously) a “massive”, “radical” and “dramatic” expansion of the scope of a publisher’s liability, particularly when so much information is shared and readily accessible on the internet. If the test is reasonable foreseeability, there would be few limits on a publisher’s potential liability for publications which do not identify anyone, but which provide enough information to allow a diligent or prurient reader to undertake all the inquiries needed to ascertain the subject’s identity. This would involve a major shift in the balance between freedom of speech and protection of reputation.

12. [134]

    Mr Blackburn also submitted that if the correct test is whether the matter complained of invites readers to ascertain the identity of the teacher concerned, it was not open to the primary judge to find that the articles contained any such invitation. The articles did not expressly invite readers to search the College’s website. They were not written in a sensational fashion and the primary Judge used inappropriately emotional language to describe the information provided about the teacher concerned as “tantalising details”. Nor was it a case where, in the language used by Eady J in Chase v Newsgroup Newspapers, the appellants had “lit a fuse which [was], objectively judged, going inevitably to lead to the claimant’s identification”.

13. [135]

    As Mr Blackburn principally relied on Baltinos, it is convenient to start with that decision. One issue in Baltinos was whether the plaintiff had been defamed by an article in a Greek language newspaper which did not name him. The article referred to a number of “rapacious con-men” preying on foreign visitors. The article stated that SBS would show a television program later that day on “this very interesting topic” and recommended that readers watch the program. The television program identified the plaintiff as a person alleged to have taken money from foreign visitors. The defendant argued that the plaintiff could not rely on the television program to establish that the newspaper article was defamatory, because the program was broadcast after the defendant’s newspaper had been published and the defendant itself bore no responsibility for the program. [99]

14. [136]

    Hunt J referred to the “basic principle” of defamation law that a plaintiff’s action in defamation arises at the time of the defendant’s act of publication and not later. He accepted that there is a qualification to that principle in that a plaintiff may rely on a subsequent publication by the defendant where the matter complained of is defamatory on its face and the subsequent publication is used only to identify the plaintiff as the person referred to. [100] As his Honour pointed out, the qualification was stated and applied by the English Court of Appeal in Hayward v Thompson. [101]

15. [137]

    Clearly the qualification to the general principle could not assist the plaintiff in Baltinos because the defendant was not responsible for broadcasting the television program identifying the plaintiff. Nonetheless, Hunt J devoted most of his judgment to a critique of Hayward v Thompson. His Honour identified two possible justifications for the qualification to the general principle accepted in that case. The first was that the subsequent publication showed that the defendant actually intended to refer to the plaintiff in the defamatory publication. The second justification was that the subsequent publication led the reader to infer from what had already been published that the defendant had intended to refer to the plaintiff (regardless of the defendant’s actual intention). [102]

16. [138]

    Hunt J considered that the first justification was contrary to the fundamental principle that liability for defamation does not depend on the intention of the defamer, but upon the fact of defamation. In his Honour’s view, the only intention of the defendant relevant to the issue of identification is “that which is perceived by the reader to have been his intention, gained only from what has been published by the defendant”. [103] Nonetheless, Hunt J regarded himself as bound by the decision of the Privy Council in Lloyd v David Syme & Co Ltd [104] that the defendant’s intention is relevant to the issue of identification. He took this approach notwithstanding what he said was “the serious and fundamental objection in principle to what was held by the Privy Council”. His Honour therefore accepted, albeit with great reluctance, the first justification for the principle applied in Hayward v Thompson.

17. [139]

    So far as the second justification for the qualification was concerned, Hunt J said that there could be no objection in principle to taking into account material previously published by the defendant. This could be done in order to demonstrate that the defendant’s apparent intention (that is, apparent from the publication) was to refer to the plaintiff. He accepted that the extension in Hayward v Thompson, permitting reference to a subsequent publication for the same purpose, was supported by authority in New South Wales. [105] Since the extension focussed on the defendant’s apparent intention it was sound in principle.

18. [140]

    The next question – critical to the outcome of Baltinos – was whether, as Hunt J framed it, the proposition stated in Hayward v Thompson should be extended to cover: [106]

19. [141]

    Hunt J said that the policy justification for restricting the plaintiff to subsequent publications by the defendant itself is that if the defamatory matter does not identify the plaintiff, the defendant should not be held responsible for the conduct of a third party in pointing the finger at the plaintiff. While the third party might be held separately liable for adopting the defendant’s earlier publication: [107]

20. [142]

    Hunt J considered, however, that the case before him was different: [108]

21. [143]

    His Honour therefore concluded that the plaintiff’s entitlement to rely on the television program was a “wholly justified extension” of the principle in Hayward v Thompson. Since there was to be a jury trial in Baltinos, Hunt J added this observation: [109]

22. [144]

    Several propositions emerge from Baltinos and the cases discussed by Hunt J. First, where a publication is not defamatory, a subsequent publication cannot be used as an extrinsic fact in order for the publication to be given a meaning it does not bear on its face. [110] The reason is that the cause of action in defamation is the publication of the defamatory material of and concerning the plaintiff. The cause of action arises when the words are published.

23. [145]

    Secondly, where a publication is defamatory on its face, but does not identify the plaintiff by name, it is permissible to look at a later publication by the defendant to ascertain to whom the first publication referred. This represents a qualification to the general principle stated in Grappelli. [111] The qualification can be justified as a matter of principle on the ground that the later publication shows, not the actual intention of the defendant, but that the first publication itself apparently intended to refer to the plaintiff. [112]

24. [146]

    Thirdly, if a defamatory publication does not name the plaintiff, but incorporates an apparent invitation to the reader to ascertain the identity of the person defamed by reading or viewing a later publication, the plaintiff may rely on the subsequent publication on the issue of identification. [113]

25. [147]

    The decision in Baltinos can be interpreted in one of two ways. The broad interpretation is that if a reasonable reader would understand that the matter complained of invited him or her to read or view a subsequent publication in order to ascertain the identity of the person to whom the defendant intended to refer, the plaintiff may have regard to the content of the subsequent publication. The plaintiff may do so both on the issue of identification and in assessing the number of readers who would have understood the original defamatory publication to be referring to the plaintiff.

26. [148]

    A narrower interpretation of Baltinos is available if the decision is to be confined to its particular facts. The newspaper article in that case expressly invited readers to view the television program which was to be broadcast later in the day and which identified the plaintiff. The decision in Baltinos, narrowly construed, might therefore stand for the proposition that:

27. [149]

    That Baltinos should not be confined to its own facts is suggested by the decision of Levine J in Strasberg v Westfield Ltd. [115] His Honour applied the “Baltinos principle” to an internal letter which did not identify the plaintiff. [116] The letter did not expressly invite readers to watch a television program to be broadcast later that day which named the plaintiff, but did refer to the program. Levine J inferred that recipients of the letter would view the program which identified the plaintiff. Notwithstanding the absence of an express invitation to recipients to watch the program, there was a “sufficient nexus” between the letter and the program to attract the Baltinos principle. [117]

28. [150]

    In interpreting the scope of the “Baltinos principle” it is necessary to bear in mind the objects of the law of defamation. As the joint judgment in Dow Jones & Co Inc v Gutnick observed, [118] the law attempts to balance, on the one hand, society’s interest in freedom of speech and in the free exchange of information against an individual’s interest in preserving his or her reputation from unwarranted slurs or damage. It is because of the law’s concern with reputation, that damage is the gist of an action in defamation. [119]

29. [151]

    In Baltinos, [120] Hunt J remarked that it would be unfair to make a defendant liable for the actions of a third party in pointing the finger at a person who could not otherwise be identified from what had been published at the time of publication. Unfairness is perhaps not the touchstone in this context. A more cogent way of making the same point is that to impose liability on a defendant not responsible for identifying the plaintiff would involve too great a constraint on freedom of speech. It is for that reason that care must be taken not to extend the reach of defamation law too far.

30. [152]

    Even so, there are good reasons not to give an unduly narrow interpretation to the “Baltinos principle”. A defamatory publication which does not name the plaintiff may cause grave reputational damage, even among readers or viewers who cannot identify the plaintiff at the moment they read or view the publication. For example, a defamatory publication may be expressed in a manner clearly designed to excite the interest of readers in ascertaining the identity of the unnamed person who has been defamed. The publication might also make it abundantly clear that the unnamed person can be identified by going to an easily accessible source. Depending on the particular circumstances, it might be very difficult to conclude that if a reader or viewer chooses to identify the person by going to the source suggested, the publisher of the defamatory material is being held responsible, not for his or her own acts, but for the acts of a third party.

31. [153]

    In Baltinos, the source of information accessed by readers happened to be a television program that the publication both named and recommended to readers. In Strasberg v Westfield Ltd, the program was named but not recommended. Why, as a matter of principle, should it be different if the identifying source is implied rather than expressed? This was the point of the comment by Eady J in Chase v News Group Newspapers, [121] cited by the primary Judge, that invitations can be extended by implication or by “nudges and winks”.

32. [154]

    In the present case, the readers of the articles who identified the plaintiff as the teacher concerned by accessing the College’s website utilised a resource (the College’s website) that was actually in existence at the time of publication of the articles. It was a resource known to the appellants, since the journalist whom Ms Davies had asked to make inquiries in Ms Davies’ absence on holidays had consulted the College’s website. Furthermore, the appellants published the defamatory material on its own website and to subscribers to the Sydney Morning Herald’s Tablet App. They clearly appreciated that readers using computers, smartphones and tablets could access other websites virtually instantaneously if they wished to ascertain the identity of the teacher concerned.

33. [155]

    To hold the appellants liable in these circumstances does not imply that the law is being judicially reformed to cope with the “quantum leap of technological capacity” [122] that characterises communications in the early twenty first century. Nor does it mean that the law has undergone a “massive”, “radical”, or “dramatic” transformation. It simply means that, in accordance with the tradition of the common law, established principles have to be applied to changed circumstances. In this case, the changed circumstances include novel means of communicating and accessing information.

34. [156]

    The primary Judge recognised that the articles were not “inappropriately sensational”. Nonetheless, she found that they were bound to excite interest as to the identity of the teacher concerned. Contrary to Mr Blackburn’s submission, it would not have only been those with a prurient interest in the published allegations who would have wished to identify the teacher concerned. For example, readers associated then or in the past with the school community would have had an interest in doing so for reasons other than mere prurience.

35. [157]

    The primary Judge attributed particular significance to the “tantalising details” provided about the teacher’s identity. In my view, the significance of the details provided is not so much whether they can be described as “tantalising”, but whether they imparted sufficient information to interested readers to allow them easily to ascertain the identity of the teacher concerned. The details provided in the articles were quite sufficient to cause an ordinary reasonable reader to realise that a simple search of the College’s website would almost certainly reveal the identity of the teacher concerned. That, the primary Judge found, explained the spike in the visits to the College’s website on 31 January 2014.

36. [158]

    The factors that in my view justify the finding that the articles implicitly invited reasonable readers to visit the College’s website to ascertain the identity of the teacher concerned were the following:

37. [159]

    For these reasons, I think that her Honour was correct to find that the articles implicitly invited readers to ascertain the identity of the teacher concerned by accessing the College’s website. There was therefore no error in her Honour’s finding that the appellants were liable to all persons who read the matter complained of and who either knew the identifying detail at the time or ascertained it from the College’s website. [123]

38. [160]

    I add this observation. Much of Mr Blackburn’s argument was directed to the primary Judge’s acceptance of the proposition that a publisher should be held liable for a publication where it is reasonably foreseeable that the plaintiff will be identified as a result of the defendant’s actions. Her Honour put this forward as an alternative basis for her decision. While I think there is force in Mr Blackburn’s criticism of reasonable foreseeability as a criterion of liability in this area of the law, it is not necessary to attempt a definitive ruling. The first ground advanced by her Honour, understood as I have explained, supports her finding that the articles invited readers to ascertain the identity of the teacher concerned from the College’s website.

39. [161]

    The appellants accepted that there must have been people who read the articles and mistakenly formed the view that the respondent was the teacher concerned. Mr Blackburn submitted, however, that the number of people who laboured under that misapprehension must have been quite limited. He relied on evidence which suggested that students who attended the senior school in 2013 and 2014 and staff at the College must have known that the teacher who resigned was not the respondent. He pointed out that the Rector of the College had advised staff on 24 January 2014 that the teacher who had been accused of inappropriate conduct had been asked to resign and had done so. In addition, the Rector and the Chair of the College’s Council, in a letter dated 23 January 2014, had advised parents of students in the senior school that the teacher concerned had resigned. Some parents contacted the College following publication of the articles. They were advised that the articles were incorrect and that the teacher concerned was not in her late 20s and was not a teacher of both drama and English.

40. [162]

    These submissions seem to be directed at findings that the primary Judge did not make, rather than those she did. Her Honour expressly rejected what she perceived to be the proposition advanced by the appellants, namely that the readers of the articles who understood them to refer to the respondent coincided almost exactly with those readers who knew she was not the teacher concerned. Her Honour listed the categories of readers who would be likely to have understood the articles to refer to the respondent. [124] The categories identified by her Honour do not include staff of the College, students at the senior school or the parents of those students. Her Honour no doubt accepted that most within the College community would have appreciated that the respondent was not the teacher concerned (bearing in mind that a defamatory statement made to someone who does not believe it nonetheless constitutes the tort of defamation). [125] Her finding was that many readers who knew the respondent or knew of her would have laboured under the misapprehension that she was the teacher concerned. The arguments advanced by Mr Blackburn in relation to this ground of appeal cast no doubt on her Honour’s finding.

41. [163]

    Mr Blackburn submitted that the primary Judge failed to take into account post-publication evidence indicating that the respondent’s reputation had not been damaged to the extent she claimed. Mr Blackburn relied particularly on the respondent’s promotion in 2014 to the position of Assistant Head of English at the College. This, he said, constituted a signal to the wider school community that the respondent continued to enjoy a “stellar reputation” as a teacher of English and drama.

42. [164]

    As Mr Blackburn submitted, evidence of post-publication material going to a person’s reputation is admissible to ensure that the damages awarded for defamation accurately reflect the plaintiff’s reputation at the time the damages are awarded. [126] The evidence may go to mitigate or aggravate the damage suffered by the plaintiff. [127] However, the primary Judge did not depart from this principle.

43. [165]

    The primary Judge admitted the evidence as to the respondent’s promotion and did not overlook the evidence. Indeed her Honour accepted that the promotion demonstrated that the respondent’s ability as a teacher was recognised by the decision-makers at the College, none of whom could have been under any misapprehension as to whether she was in fact the teacher concerned. [128] The point her Honour made was that the promotion said little as to the likely reach of the defamation among the much larger group of persons that read the articles and would have understood the publications to refer to the respondent.

44. [166]

    Mr Blackburn did not identify evidence which contradicted the primary Judge’s assessment that the promotion did little to curb the damage to the respondent’s reputation among the large group of readers beyond the College community. No error has been shown in her Honour’s finding.

45. [167]

    Mr Blackburn challenged the primary Judge’s finding that the cross-examination of the respondent aggravated the hurt occasioned to her by the defamatory publications. He submitted that as the appellants’ senior counsel he was perfectly entitled to cross-examine the respondent about her claim as he had and that there was no basis for the finding that the cross-examination was “wholly unjustifiable”. [129]

46. [168]

    Mr Blackburn did not dispute that the primary Judge correctly identified the relevant principle as that stated in the joint judgment in Triggell v Pheeney: [130]

47. [169]

    The single passage of cross-examination on which her Honour relied for the finding that the appellants’ conduct aggravated the hurt occasioned to the respondent tested the respondent’s claim in her affidavit that she feared a prospective employer, upon seeing her resume, would immediately raise concerns that she was a paedophile.

48. [170]

    The cross-examiner put to the respondent that she did not have the fear she claimed in her affidavit. It no doubt would have been less abrasive and less likely to produce an emotional reaction from the respondent had the questions been confined to testing whether the fear, although genuinely held, had a sound factual basis. The answer to that question would have to take into account the respondent’s high standing at the College and the excellent references she would be likely to obtain if she ever left her position there. The respondent rebuffed the suggestion that she did not really fear that she might be seen as a paedophile and the issue was not pursued, except by questioning directed to showing that there was no reasonable basis for the fear expressed by her.

49. [171]

    It is true, as Mr McClintock SC submitted on behalf of the respondent, that the primary Judge was well placed to assess the impact of the challenge to her veracity on this issue. But in my view there was a reasonable basis for regarding the particular fear expressed by the respondent to be somewhat exaggerated. In these circumstances I do not think that brief questioning suggesting that the fear was not genuinely held should be characterised as unjustifiable and lacking bona fides, even though it clearly caused some distress to the respondent. The questioning addressed only one relatively minor aspect of the respondent’s case and did not travel beyond a vigorous presentation of a legitimate defence on that issue. [132]

50. [172]

    In my opinion, however, this conclusion does not warrant any reduction in the aggravated damages awarded to the respondent or in the overall quantum of damages. The primary Judge did not separately quantify aggravated damages, but she found that the most significant aggravating factor was the failure of the publisher to take steps to prevent readers falling into the same trap as Ms Davies. [133] That was a reference to the fact that Ms Davies telephoned the College’s switchboard on 29 January 2014 and asked to speak to the respondent. Before the telephone call, Ms Davies had been given a name which was very similar to the correct name of the teacher who had left the College, with only one letter that was wrong. [134] The person answering the telephone at the College said that the call would be put through, whereupon Ms Davies (as she admitted in answer to interrogatories) realised that the respondent could not be the teacher concerned. Ms Davies chose to hang up without speaking to the respondent and thus lost the opportunity to have it put beyond all possible doubt that the teacher concerned was not in her late 20s and had not taught both English and drama. Although Ms Davies made a few further perfunctory inquiries, a minor spelling error apparently constituted an insuperable obstacle to successfully completing what would seem to have been the rather straightforward task of ascertaining the true identity of the teacher concerned. Had that been done, or at least had the articles not described the teacher concerned in a way that could only have referred to a teacher still on staff (and who could not have been the teacher concerned), the reputational damage and distress to the respondent would have been avoided.

51. [173]

    The primary Judge remarked that it “beggars belief” that Fairfax proceeded to publish the articles knowing that the respondent was not the teacher concerned and that the details included in the article pointed to the respondent as being the teacher concerned. Compared with this abject failure of investigative journalism, the impact of a brief passage in the cross-examination on the respondent fades into insignificance. The primary Judge’s award of aggravated damages should stand.

52. [174]

    Section 34 of the Defamation Act requires the court, in determining the amount of damages to be awarded in defamation proceedings, to ensure that there is an appropriate and rational relationship between the harm sustained by the plaintiff and the amount of damages awarded. Section 35(1) provides that unless the court otherwise orders under subsection (2), the maximum amount of damages that can be awarded for non-economic loss in defamation proceedings is that determined in accordance with the section ($366,000 at the date of the trial). Section 35(2) provides that a court may award damages exceeding the maximum:

53. [175]

    Within this statutory framework, as the primary Judge observed, [135] an award of damages for defamation serves the three purposes identified by the High Court in Carson v John Fairfax & Sons Pty Ltd: [136]

54. [176]

    For these reasons, an award of damages for defamation is regarded as a discretionary judgment. Thus an appellate court can interfere with a damages award only if it appears that the trial judge proceeded upon a wrong principle or if the court can infer that in some way the discretion of the trial judge must have miscarried. [138] The appellate court does not intervene merely because it would have assessed damages in some other amount.

55. [177]

    Mr Blackburn accepted in his oral submissions that if the appellants’ other grounds of appeal were unsuccessful, the appellants would find it difficult to press the contention that the award of $350,000 was unreasonable. He also accepted that rejection of the appellants’ other submissions constitute an obstacle to the grounds of appeal asserting that the primary Judge’s exercise of a discretionary judgment was vitiated by failing to take relevant matters into account or by taking irrelevant matters into account.

56. [178]

    Nonetheless, Mr Blackburn submitted that the defamation in the present case was insufficiently serious to warrant an award of damages close to the statutory cap, even if it incorporated a component for aggravated damages. He drew attention to authorities on the statutory cap in New South Wales, where trial courts have followed the approach of Bell J in Attrill v Christie: [139]

57. [179]

    Mr Blackburn did not submit that the primary Judge was not cognisant of the need to consider whether the damages sustained by the respondent in consequence of the defamatory publications was sufficiently serious to justify an award close to the statutory cap. His complaint was that her Honour overstated the seriousness of the damage to the respondent’s reputation and of the consequential distress she experienced.

58. [180]

    The primary Judge found that by reason of the defamatory publications the respondent suffered hurt to her feelings, felt powerless and experienced a deep sense of embarrassment and shame despite not having had any involvement whatsoever in the conduct reported in the articles. [140] Her Honour accepted, moreover, that the respondent felt “scared, worried, anxious, distressed, alone, depressed, angry and insecure”. Mr Blackburn did not challenge these findings. They are hardly surprising given the extensive reach of the articles.

59. [181]

    There was a good deal of evidence that parents of students and the respondent’s friends and acquaintances within the broader College community were aware of the articles. Some believed, on the basis of the identifying features, that the respondent was the teacher concerned. Others were uncertain and confused as to whether the respondent might be the teacher concerned. Publication of the articles in a widely distributed newspaper with hundreds of thousands of readers, as the primary Judge found, drew them to the attention of people well beyond the broader College community, many of whom knew or knew of the respondent.

60. [182]

    The “spike” in traffic on the College’s website reflected the efforts of readers to match the identifying features of the teacher concerned, as stated in the articles, with the attributes of the College’s teachers recorded on the staff page on the website. There was only one match.

61. [183]

    The “grapevine” effect of matching the articles with the respondent’s identifying features is evident from the references to the events at the College in social media, such as Facebook and Twitter. Some communications identified the respondent, if not by name then by her achievements, including the fact that she was a University medallist (a fact not referred to in the articles). Commentary on social media was frequently aimed at the teacher concerned, but assumed that the respondent was that teacher. Some of the comments were vile and inevitably caused great distress to the respondent.

62. [184]

    In my opinion, the appellants have not established any basis for interfering with the primary Judge’s award of damages for this very serious defamation.

Orders

1. [185]

   The appeal must be dismissed. The appellants must pay the respondent’s costs of the appeal.