AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2004 >> [2004] MelbULawRw 13

Database Search | Name Search | Recent Articles | Noteup | LawCite | Help

Kenyon, Andrew T --- "Lange and Reynolds Qualified Privilege: Australian and English Defamation Law and Practice" [2004] MelbULawRw 13; (2004) 28(2) Melbourne University Law Review 406


LANGE AND REYNOLDS QUALIFIED PRIVILEGE: AUSTRALIAN AND ENGLISH DEFAMATION LAW AND PRACTICE

ANDREW T KENYON[∗]

[Australian and English case law has developed qualified privilege defences that are available to the media and appear to protect more political or public interest speech than traditional defamation law. This article draws on judicial decisions and qualitative research into defamation litigation to examine the defences’ scope, strength and practicality in litigation. England’s Reynolds privilege emerges as a well-supported, relatively strong, flexible and innovative defence, especially compared with Australia’s narrower and weaker privileges under Lange and New South Wales legislation. The research strongly supports the further development of Australian privilege defences, as well as more careful consideration of judge and jury roles in each country. A closer understanding of Reynolds offers important benefits for protecting the publication of public interest news and commentary, and it is particularly useful in light of recent, and proposed, Australian law reforms.]

CONTENTS

INTRODUCTION

Defamation law is often said to deter speech. The existence of such a ‘chilling effect’[1] has some research support. For example, the leading United Kingdom study, Libel and the Media: The Chilling Effect,[2] extensively investigated the practices of journalists, editors and their legal advisers. It argues that both direct and structural chilling effects exist under traditional defamation law. Media speech is chilled directly when lawyers recommend editing the content of publications, and is chilled structurally when journalists internalise the law’s restrictive principles.[3] A separate Australian study, ‘Defamation Law’s Chilling Effect’, involved a comparative content analysis of more than 1400 Australian and United States newspaper articles.[4] The study suggests that in the US — where defamation plaintiffs face much heavier burdens than under the Anglo-Australian law[5] — defamatory allegations against political and corporate actors are published more frequently than in Australia.[6] In the study’s sample, the US articles contained defamatory allegations at nearly three times the rate of the Australian articles. In particular, the Australian media appeared to be less comfortable making allegations about corporate affairs than its US counterpart. Admittedly, the concept of a ‘chilling effect’ has obvious rhetorical appeal for participants in defamation practice or academic commentary. The concept, however, is also supported by existing empirical research which suggests that traditional Anglo-Australian defamation law may well chill media content.[7]

One important aspect of the apparent chilling effect is that defendants must generally prove the truth of the factual allegations they publish.[8] This follows in part from defamation law’s unusual burden of proof for civil actions. Defamation plaintiffs need not prove that the publisher was at fault, nor that the publication was false. In addition, once published material about the plaintiff is shown to be defamatory, general damages are presumed.[9] Truth is central for defendant publishers because the primary defences of justification and fair comment require a publication’s factual basis to be proven true. While truth need not be shown for the other major defences of absolute and qualified privilege, these defences have traditionally applied to few media publications, beyond fair reports of court proceedings or parliaments. Thus, defamation law has imposed liability on many media defendants who cannot prove their publications to be true. In this way, the law appears to focus more on protecting reputation than promoting wide debate about public interest issues.

The frequent need to prove truth is something that Australian and English developments in qualified privilege have sought to address. Since the mid-1990s, extended forms of qualified privilege have developed through Lange v Australian Broadcasting Corporation[10] and Reynolds v Times Newspapers Ltd.[11] The defences appear to protect more political or public interest speech than traditional law. Reynolds can be seen as ‘conceptually, a different species of qualified privilege’ from the general duty-interest defence,[12] or at least as a substantial expansion of the circumstances in which the defence can be satisfied.[13] Reynolds privilege seeks to protect defamatory material of public importance where defendants have published responsibly, irrespective of the material’s truth or falsity.[14] Lange has also been described as making ‘fundamental changes’ to privilege’s ‘conceptual foundations’.[15] Under Lange, privilege has become a ‘relatively egalitarian’ defence that promotes ‘“free” discussion ... in which all citizens, so long as they act “reasonably”, may participate on equal terms, rather than with some enjoying the status of “privileged publishers”’.[16]

This article considers whether the developments, in practice, do offer accessible defences for media publications. It combines case law analysis with qualitative investigation of defamation litigation in each country.[17] The research suggests that Reynolds is far more likely to reduce any chilling effect than Lange, but difficult issues of litigation practice remain for both defences. Part II briefly outlines the traditionally limited protection for media publications, before examining recent case law. It explains how the two new defences differ doctrinally, both in scope and in strength. Reynolds privilege is broad and flexible, and the judges who apply it appear sensitive to the free speech concerns underlying its development. It seems to offer the media meaningful benefits, at least for non-tabloid investigative reporting, and it may also develop to support wider commentary and public debate. Lange privilege is comparatively narrow, with its focus on political communication. More significantly, however, Lange privilege is weak. As with the longstanding statutory privilege under s 22 of the Defamation Act 1974 (NSW), publishers appear likely to have trouble demonstrating to courts’ satisfaction that they acted reasonably. Doctrinal analysis also suggests that judge and jury roles are problematic under the English and Australian defences.[18]

Part III examines defamation lawyers’ perspectives on the defences, obtained in interviews with 50 leading practitioners in London, Melbourne and Sydney. The interviews suggest that Reynolds privilege is quite supported in England, despite some uncertainties about its future development and litigation practicality. Reynolds also appears to be affecting the media’s pre-publication conduct and encouraging what may be seen as more balanced reporting. In Australia, Lange privilege appears to be a barely useable defence, primarily due to the reasonableness requirement. Practitioners suggest revising the defence, with preferable models offered by Reynolds or an expanded duty-interest privilege in the style of the earlier Australian approach under Theophanous v Herald & Weekly Times Ltd.[19] The fieldwork also suggests that privilege defences need to allocate judge and jury roles carefully. Otherwise, the greater protection of at least some types of speech, which appears to have been sought by developments in qualified privilege, will be substantially undercut.

As a final introductory matter, it is worth noting that this article’s interview-based research takes a traditional sociolegal approach that is focused on legal practice. A longstanding interest of legal fieldwork has been legal actors, whether lawyers in courts,[20] or regulators outside courts.[21] Empirical research did display early concern with non-legal actors,[22] which has resurfaced since the late 1980s in qualitative investigations of the experiences and attitudes of laypeople who interact with the law.[23] The present research, however, may differ slightly in focus from some lawyer-centred sociolegal work. It starts from the expectation that doctrinal analysis can benefit from empirical research into litigation practice, at least for analysing laws that are commonly litigated. As a result, the social in this research is quite confined: it is defamation law and practice.[24]

II DEVELOPING QUALIFIED PRIVILEGE

A The Traditional Defence

Defamation law has long recognised various categories of qualified privilege, said to exist for the ‘common convenience and welfare of society’.[25] Qualified privilege was available where publishers were under a legal, social or moral duty, or acted to protect an interest, and recipients had a corresponding duty or interest.[26] Shared duties or interests could exist for material published to small audiences,[27] but widespread publications were likely to see the defence fail.[28] Thus, while qualified privilege is longstanding for fair reports of parliamentary and court proceedings,[29] the defence has not generally protected media publications of political or public interest.[30] Established occasions of privilege can be criticised in many ways, not least for their historical class bases.[31] Qualified privilege did not protect media publications about suspected corruption. Instead, alleged malfeasance had to be reported to what courts regarded as proper authorities. Nor did privilege protect publications reasonably believed to be true. While the defence was narrow, it was strong because only malice would defeat publications made on, and relevant to, a privileged occasion. In both England and Australia, however, qualified privilege developed during the 1990s.

B England — Reynolds v Times Newspapers Ltd

Reynolds arose after The Sunday Times published a story in its London edition criticising Albert Reynolds, who had recently resigned as Ireland’s Taoiseach (or Prime Minister).[32] Reynolds had played a major role in the Northern Ireland peace process, and his resignation was significant throughout the United Kingdom. In its 1999 decision, the House of Lords held that the duty-interest requirement for privilege could be satisfied by media publications which the public had a right to know in all the circumstances.[33] This was a major development, making Reynolds the ‘most important’ recent privilege decision,[34] ‘an important and potentially far-reaching reform’[35] and ‘a marked liberalisation’.[36]

The House of Lords rejected a generic privilege for political material in order to maintain what it saw as adequate protection for reputation.[37] A qualified privilege for all political material was thought to protect too strongly the wrong type of speech. Protection would be too strong because, influenced by the European Convention for the Protection of Human Rights and Fundamental Freedoms,[38] journalists’ confidential sources are very strongly protected. This makes it ‘virtually impossible’ to prove malice and defeat privilege.[39] In addition, the wrong speech would be protected in that a political defence would be too narrow. Instead, the defence should focus on the public interest qualities of a publication. While political speech is an important example of protected material,[40] Reynolds privilege is defined by a test of public interest.[41]

The House of Lords in Reynolds required multiple factors to be considered when deciding whether defendants have established privilege, with Lord Nicholls listing 10 illustrative factors.[42] A court will need to consider matters about the publication — including its public importance, urgency and overall tone, and whether it included the claimant’s position[43] — as well as matters about the information’s source, such as steps taken by the publisher or another relevant body in verification.[44]

Commentators quickly suggested that courts would treat the Reynolds factors as a ‘checklist’[45] or ‘the standard template’.[46] Courts have done this, focusing on the idea of ‘responsible’ journalism. Subsequent cases suggest that the defence may fail where publications are sensational and repeated,[47] sources are unreliable, claimants are not contacted prior to publication (at least where publication has no urgency),[48] or suspicion is presented as fact.[49] But the defence’s scope extends to matters of public importance such as bribery in professional soccer,[50] the use of child labour[51] and the detection of terrorism.[52] It is not limited to political communication. Reynolds privilege has flexibility, with the Privy Council emphasising the need to apply the responsible journalism standard practically.[53] Privilege need not be judged according to the meanings found to be conveyed by a jury. The standard looks more to conduct than to a publication’s meaning: ‘a journalist should not be penalised for making a wrong decision on a question of meaning on which different people might reasonably take different views.’[54]

Cases also show the defence’s potentially far-reaching effects. Reynolds privilege can protect media reports of allegations being made by others, at times without publishers even attempting to verify the allegations. For example, Al-Fagih v HH Saudi Research & Marketing (UK) Ltd[55] concerned a political dispute within the Saudi Arabian community. In a series of articles, allegations by each disputing party against the other were published without any attempt at verification. The Court of Appeal held that the fact that allegations existed was important and could properly be reported. The repetition rule, which applies to the defence of justification,[56] does not limit qualified privilege. The rule does not ‘require that an unadopted allegation ... be treated in the same way as an allegation asserted to be true’ for the purposes of qualified privilege.[57] Al-Fagih’s support for neutral reportage shows the significance of a publication’s tone under Reynolds — for example, are allegations presented as suspicions or facts, and are they merely reported or adopted by the publisher? Subsequently, the Court of Appeal has noted that reportage may have a wider application, with Simon Brown LJ stating:

I am certainly prepared to recognise that the approach adopted in Al-Fagih may need to be taken further still — rather than perhaps confined merely to the reporting of statements (attributed and unadopted) by both sides to a political dispute ...[58]

Litigating claims of Reynolds privilege, however, may be difficult, particularly in relation to judge and jury roles.[59] Defamation law traditionally seeks a simple division under which judges determine whether occasions of publication are privileged, while juries decide if publications were affected by malice. Although there can be complications in judges deciding if a publication’s content is relevant to privileged occasions,[60] these complications do not arise as a matter of course, nor do they depend on disputed facts. However, the Reynolds factors encompass matters traditionally dealt with under malice.[61] Many disputed facts could be relevant to the Reynolds factors, with juries needing to determine each of these facts before judges decide whether occasions are privileged. This could lead to very complex jury questionnaires, as in Loutchansky v Times Newspapers Ltd [No 4] where the jury sat an ‘examination paper’.[62] Eady J has commented that in cases in which Reynolds is argued:

It is now almost inevitable that ... the jury will be asked to answer a series of questions in order for the court to arrive at the necessary factual substratum upon which to base the ruling. The questions will be directed no doubt largely to establishing the raw data for answering those of the ten non-exhaustive tests identified by Lord Nicholls in Reynolds ... as may be relevant to the case in hand.[63]

Under this approach, judges retain significant powers to interpret and apply those jury answers, with ‘[a]ny value judgments, and to a large extent also any inferences to be drawn from the raw data’ reserved for judges.[64]

Overall, Reynolds privilege is broad, not being limited to political publications. With strong protection for journalists’ sources, a flexible approach to meaning and the recognition of some neutral reportage, Reynolds privilege appears to offer journalists meaningful benefits, while also requiring professional journalistic conduct. English protection for matters of political and public interest appears to have strengthened, at least for the classic investigative reporting that courts envisage.

C Australia — Lange v Australian Broadcasting Corporation

Australian privilege developed in two stages, each influenced by Australia’s constitutional protection for political communication. The Constitution does not expressly protect speech, but creates what is called an implied freedom of political communication.[65] This limits legislative power to restrict speech about government or political matters. Any restriction must be appropriate to a legitimate end of the legislation. In addition, the common law — including its traditional approach to qualified privilege — is shaped by the implied protection.

The first stage in development was the 1994 Theophanous decision.[66] It created a constitutional defence for publications concerning political and government matters. The defence could apply where defendants were unaware that publications were false, had not published recklessly without caring about truth or falsity, and publication was reasonable in the circumstances.[67] Theophanous also expanded common law qualified privilege to cover media publications about political or government matters.[68] For such publications, defendants need not have met the constitutional defence’s reasonableness requirement. The expanded common law privilege was defeated only by malice, making it the most significant development for publishers in Theophanous.

The majority judgments in Theophanous also suggested that political communication was wide, and close to being communication about ‘public affairs’.[69] Political communication included speech about governments, politicians, candidates, public officers, political and public bodies, and people whose activities were matters of political debate, such as trade union leaders and political commentators.[70] The majority cited Eric Barendt to describe ‘political speech’ as ‘all speech relevant to the development of public opinion on the whole range of issues which an intelligent citizen should think about.’[71] With a privilege defence that could only be defeated by malice, Theophanous suggested that many public interest matters would be protected far more than previously: it appeared to be a wide and strong defence.

But the defence did not survive. The High Court’s divisions in Theophanous and related cases led it to reconsider matters. The 1997 Lange decision forms the second and current stage of Australian developments. It concerned the leading current affairs program, Four Corners, which criticised the conduct of New Zealand’s then Prime Minister, David Lange. The plaintiff pleaded that the program meant he was unfit to hold public office.[72] The High Court heard a pre-trial challenge to the availability of the Theophanous defences. In a unanimous judgment, the High Court confirmed the constitutional protection for political communication, but refashioned its effect on defamation law. The judgment combined elements of the constitutional and common law Theophanous defences into a new form of qualified privilege.

The High Court held that political communication could give rise to a privileged occasion, as ‘each member of the Australian community has an interest in disseminating and receiving information, opinions and arguments concerning government and political matters that affect the people of Australia.’[73]

This means that Lange privilege could protect widespread publications, as did common law privilege under Theophanous. Protecting wide publications, in turn, requires defendants to establish that publication was reasonable, as under the Theophanous constitutional defence. To establish reasonableness, defendants must generally establish that they had reasonable grounds to believe publications were true, that they did not believe publications were false, and that they had made proper inquiries to verify them. In addition, defendants must have sought and published responses from potential plaintiffs, except where this was not practical or necessary.[74] Although malice defeats Lange privilege, it will generally have little room to operate.[75] The focus will be on whether defendants can establish reasonableness.

Lange privilege is weaker than Theophanous privilege because all political publications face the reasonableness test. In contrast, the Theophanous common law privilege was only defeated by malice. In addition, Lange is narrower because political communication appears to have been confined since Theophanous. Lange privilege appears to encompass only matters about politics and government ‘in an electoral and parliamentary sense.’[76] Political communication may be limited to information for decisions about voting. For example, most criticism of judicial officers may fall outside the protection.[77] Criticism calling for a judge’s dismissal may come within the protection,[78] but perhaps not criticism related merely to an individual case.[79] In terms of its formal doctrine, Lange privilege may protect far fewer media publications than Theophanous.

The defence’s narrow scope helps explain the relative lack of subsequent decisions on Lange privilege. The cases, however, illustrate three related points about the defence’s scope. First, the defence clearly does not extend to all matters of public interest.[80] Material about the corporate sector, whether profit-seeking or non-profit, can be expected to be excluded. For example, in Rowan v Cornwall [No 5], criticisms from a public review panel about the administration of a women’s refuge that was run by a non-government organisation were outside the defence.[81]

Second, discussion of public administration may come within the defence. For example, Conservation Council of SA Inc v Chapman[82] concerned a bitterly fought bridge development which had become an issue of national public concern. Lange privilege did not apply to allegations that property developers threatened litigation to silence residents opposing development, even for publications couched in terms of free speech and misuse of the legal process.[83] Potentially within the defence’s scope, however, were publications suggesting developers had engaged in ‘token’ consultation with indigenous groups, which was less than the statutory planning process required.[84]

Third, discussion of commercially significant matters can exceed the scope of Lange privilege, even when occurring within apparently political contexts. In West v Nationwide News Pty Ltd,[85] discussion by councillors at a council meeting and its subsequent media reporting were outside the defence. The discussion concerned development approval that had been obtained by a plaintiff, the degree to which development conditions had been complied with and the effects on local residents.[86] Although courts have repeatedly held that no narrow approach should be taken to the defence’s scope,[87] success in Lange privilege may exist primarily for discussion about political candidates.[88]

Under Lange privilege, publishers’ conduct must be reasonable. The history of statutory privilege under s 22 of the Defamation Act 1974 (NSW) suggests that reasonableness will be hard to meet. The High Court considered Lange on the basis of New South Wales law, and held that it would unreasonably burden political discussion if not for s 22.[89] This meant that the common law required the expanded privilege set out by the High Court in order to meet the constitutional protection for political communication. Lange, however, left unresolved a doctrinal issue of possible importance. Under s 22, most defendants must establish their honest belief in the publication.[90] Defendants must also establish that the publication’s manner and extent ‘did not exceed what was reasonably required’, that they exercised reasonable care in terms of making inquiries, and that their conclusions of fact or opinion ‘followed logically, fairly and reasonably’ from the information revealed by those inquiries.[91] In practice, these appear to be onerous requirements in the way they are applied. The approach ‘effectively negates the availability’ of the defence.[92] The requirements far exceed those for common law fair comment, for example, where opinion can be exaggerated, prejudiced or obstinate.[93] It is very difficult to establish the s 22 defence for publications based on facts not proven true, or where defendants did not make ‘proper’ inquiries.[94] The defence appears to be commonly pleaded, but has succeeded in only a handful of cases since 1974.[95] The issue Lange left unresolved is whether defendants need to establish their belief in a publication’s truth, or merely reasonable grounds to believe in its truth. Academic commentary favours the latter approach, which differs from s 22 and is said to make Lange privilege more widely available.[96]

Case law has not yet dealt explicitly with this possible difference between the Lange and s 22 defences. But recent cases have noted that reasonableness under Lange, as well as under s 22, should not be inflexible.[97] The Lange factors[98] are neither ‘principles of law’ nor ‘essential elements’ in the defence.[99] For example, it can be reasonable to publish without seeking a response,[100] although failing to meet a Lange factor would ‘in most cases prove fatal.’[101] Overall, however, courts’ interpretation of reasonableness closely tracks s 22, notwithstanding the peculiar New South Wales action where each pleaded imputation is a separate cause of action.[102]

A final issue about Lange privilege is the division of roles in a jury trial. It appears that the jury, where there is one, rules on questions of primary fact relevant to the defence — for example, it may need to answer multiple questions about whether certain inquiries were made by the defendant before publication. In light of the jury answers, the judge then rules whether the defence is made out. Although trial practices have varied, the Victorian Court of Appeal has unanimously endorsed this approach.[103] Thus, Lange privilege seems likely to follow the traditional approach in Australia, which is also reflected in the English developments discussed above. Equivalent questions arise about the complexity of this approach. Another possibility would be for the jury to determine whether the publication was reasonable, rather than any disputed facts relevant to reasonableness. Evidence from practice may suggest which approach is likely to be preferable in litigation.

D Doctrinally Comparing the English and Australian Defences

The new English and Australian privilege defences have been analysed above in terms of their scope, strength and trial practicality. These points are summarised here before examining litigation practice. English cases suggest the value of a broad defence, not limited to narrow, institutional conceptions of politics.[104] In addition, English cases apply a reasonably wide variety of factors in assessing publishers’ conduct, and appear to apply them quite flexibly.[105] The English approach to publications’ tone is a notable example,[106] though it is one that favours non-tabloid publications.[107] In comparison, the Australian approach under Lange privilege is narrow — it is not a public interest defence and rarely encompasses speech about commercial issues. It also appears to be weak because of the requirement for publication to be proven reasonable. Reasonableness under Lange, however, may be a more achievable standard than under s 22, depending on whether defendants will be required to establish their belief in the truth of the publication. In any event, neither Lange nor s 22 offer the strength of protection offered by Theophanous common law privilege. In terms of litigation practicality, Theophanous common law privilege and traditional qualified privilege offer a reasonably clear division between judge and jury roles. The jury addresses malice without needing to answer an ‘examination paper’ of questions about disputed events. It appears that such examination papers may arise under both Reynolds and Lange privileges, which may undercut the benefits aimed for in developing the defences.

Next, qualitative material about litigation practice is examined in relation to these issues about the defences’ scope, strength and practicality in litigation.

III ENGLISH AND AUSTRALIAN PRACTICE

A Fieldwork

To investigate privilege in action, interviews were conducted with 50 defamation practitioners in England, New South Wales and Victoria.[108] The 23 London interviewees comprised specialist defamation judges, barristers from two leading defamation chambers,[109] solicitors with substantial defamation practices and in-house lawyers for media companies.[110] In Australia, 27 defamation practitioners were interviewed from the two largest defamation jurisdictions, New South Wales and Victoria.[111] They included a specialist judge and leading defamation barristers, solicitors and in-house lawyers.[112] Defamation formed the major part of most interviewees’ practices, often being a sizeable majority. Given the concentrated media law practices in each country, the interviewees represented a good cross-section of lawyers with significant defamation involvement.[113]

Each interview lasted approximately one hour, with all except two being audio recorded and transcribed.[114] Transcriptions were analysed according to the legal issues discussed, with major areas being the scope, strength and litigation effects of qualified privilege. Quotations are direct transcriptions from interviews, with only minimal changes made to interviewees’ words for clarity and readability.[115] To protect interviewees’ confidentiality, no quotation is identified with a named interviewee. Instead, interviewees are referred to by number, along with their professional role: solicitor (‘S’), barrister (‘B’), Queen’s Counsel (‘QC’) or Senior Counsel (‘SC’).[116] All those who assisted in relation to the interviews are sincerely thanked.[117]

The 50 interviews generated a very large amount of material,[118] but offered the most viable source of empirical information. Alternatives to interviews such as observing litigation were not pursued. Litigation activities take place over many months in multiple locations and observation only reaches the small portion of the litigation process visible in court. Interviews can encompass activities across a wide range of times and locations.[119] Their flexible structure compared with other methods, such as questionnaires, was also more likely to reveal differences in the use of terminology in each jurisdiction, which is an important issue for comparative research.[120]

In what way can extrapolations be drawn from the material?[121] Does the research have what is often called reliability and validity?[122] Using recognised legal categories goes some way towards addressing reliability concerns. One could expect broadly similar material would be found by another researcher or presented by these interviewees at another time. At the same time, the transcription and coding method aimed to maintain good access to the material — to what people said and the way they described the legal categories. As well as concerns about reliability, researchers often consider their material’s validity. In this context, validity can be understood as referring to realistic representation of social phenomena. In the limited context of this legal research, that description can be accepted. That is not to suggest that validity is an unproblematic concept,[123] but the representations drawn from this material should be meaningful in terms of how legal practitioners operate with certain categories in defamation law. In addition to remembering the limited legal scope of this project, aiming at a comprehensive treatment of the interview material and including atypical cases improves validity. It addresses a weakness in some qualitative research — namely, its anecdotalism. For example, a researcher can quote a few comments from interviews, without it being apparent how representative the responses are and without contrary examples being considered.[124] Here, the extensive footnoting allows readers to assess both issues, at least to some degree — that is, a simple counting of responses within the material is made relatively transparent to the reader.[125]

B English Interviews

Three broad, related issues are drawn from interviews with London practitioners and discussed here. First, despite uncertainties about its future development, Reynolds privilege is strongly supported and is affecting pre-publication legal advice and media conduct. Second, particular concerns exist about each party’s position under Reynolds privilege and the focus on media conduct. Third, the division between judge and jury roles is also problematic at trial, but may be difficult for trial judges to improve.

First, Reynolds is overwhelmingly seen as significant. It offers a comparatively strong defence for media publications. There is uncertainty about how it will develop and how its multiple factors will apply in any litigated case. Two factors, however, are seen as central: attempting to contact the subjects of potential publications, and publications’ tone or balance. These factors mean Reynolds privilege appears to alter pre-publication conduct. Not surprisingly, London interviewees described Reynolds as important.[126] It offers a ‘whole new defence’, arguable in almost all media cases,[127] which is ‘revolutionary’[128] and a ‘massive change’.[129] It is a ‘significant improvement’ for the media,[130] as it provides a useable, or even strong, defence.[131] For some, initial doubts or disappointments have given way: ‘I remember the first time I read Reynolds thinking that this is all rather a disappointment ... [B]ut gradually re-reading it, one began to see its potential and the actual results have been really very considerable.’[132] More cautious responses still saw the potential in Reynolds for a ‘very useful defence’.[133]

While some interviewees suggested that Reynolds’ flexibility had value,[134] a major area of criticism was its uncertainty and unpredictability.[135] An experienced barrister commented: ‘I think the problem with Reynolds in practice is that it is a “sniff test”. Does it have a bad smell about it or does the judge feel warm about it?’[136] How will courts balance publications’ public interest qualities, the potential harm to subjects’ reputations, and defendants’ conduct before publication? It is difficult to predict how any particular situation will be interpreted in light of relevant factors:

Things like how urgent was it? Was it really necessary to publish this? How far do you have to check it in the heat of the moment? ... [It is] very difficult to say [and] newspapers rather take a punt.[137]

It is difficult to predict whether some publications will be found to have ‘the right calibre of public interest’,[138] or how the degree of public interest relates to the required level of investigation.[139] Yet the checklist should not be ‘rigid’[140] — ‘fluidity’ is wanted, rather than having the 10 factors ‘set in stone’.[141] While subsequent decisions have clarified somewhat Reynolds’ scope and strength,[142] the defence’s uncertainty — its ‘inherently case-by-case’ analysis — seems likely to continue.[143]

Of the 10 Reynolds factors, however, two were more influential for interviewees: contacting a publication’s subject,[144] and carefully considering its tone or balance.[145] Notwithstanding comments about the defence’s uncertainty in litigation, these two factors were seen as particularly important and were also reported as being central to changing pre-publication conduct. Speaking to the subject and publishing a response is seen as ‘the litmus test’,[146] although it is ‘not clear how much time you should give them to respond’.[147] At the least, the defence is very likely to fail if there has been no attempt to contact the subject.[148] In theory, perhaps Reynolds should not affect what is published, ‘because a newspaper ... ought not ever have published unless it had taken reasonable care and honestly believed that what it was saying was true’.[149] However, the way in which tone is ‘critically important’ changes this position.[150] Interviewees suggested that Reynolds assists the media to raise more allegations or suspicions,[151] and to publish more public interest or whistle-blowing stories.[152] At the same time, it encourages what may be seen as more balanced reporting — publications are more likely to include subjects’ points of view.[153] Before Reynolds, ‘if you had a story which was clearly defamatory, but you had a pretty good source there would be little point in publishing it in a balanced way. So you’d simply go for it’.[154] But now publishers are likely to consider Reynolds and moderate the tone of publications. Television journalism in particular ‘will be well served by Reynolds’ because its longer lead times mean that everyone involved can work to ‘the Reynolds programme’.[155] These comments suggest Reynolds is affecting publishing practices more widely and publishers are exercising greater care when they research and write stories.[156]

Second, concerns were expressed about each party’s position under Reynolds. Claimants face particular difficulties because they lack knowledge of many Reynolds factors.[157] Most often, claimants will know only whether they were approached for comment before publication. They will not know what research journalists have undertaken. This differs from other defences that involve truth or a clear occasion of privilege, such as reporting parliamentary debates. Correspondence between the parties, often referring explicitly to Reynolds factors, may help to clarify the position for claimants.[158] But Reynolds, which ‘spotlights the newspaper’, can make claimants feel that their interests have been overlooked.[159] One interviewee stated that ‘[c]lients do not really care whether the journalists were ethical or not; they care that a falsehood has been stated about them’,[160] and several interviewees suggested claimants would like to seek declarations of falsity under Reynolds.[161] For claimants, the uncertain ‘shadow of Reynolds[162] deters claims and encourages settlements.[163] Reynolds is ‘very expensive to defeat’[164] with ‘almost no claimant [being] confident of victory, and therefore reasonable settlements for reasonable awards, or indeed for apologies and costs, are becoming [more] common’.[165]

The position of defendants also drew concerns. As for claimants, the costs of trying the defence were raised: ‘even in the case of a newspaper article, and a relatively short one at that, you can have a four week trial involving a lot of detailed factual issues and a number of legal issues and vast expense and an enormous amount of time.’[166] Other respondents, however, noted that publications could involve comparatively small amounts of background material. The information sources and documentary evidence held prior to publication were ‘finite’,[167] though the amount varied between ‘very little’ and ‘massive’.[168] Justification, by way of contrast, can draw on a far wider range of material, and was described as potentially a more unwieldy defence.[169] Concerns were also raised about focusing on publishers’ conduct:[170] ‘for defendants, the Reynolds defence is a bit of a Trojan horse because it changes the dynamics of the trial. If you run a Reynolds defence alone the trial will turn into a trial of your journalists.’[171]

Journalists’ motives and actions will be more closely scrutinised than has been routine, notwithstanding the issues that historically have arisen on malice.[172] There is a ‘much less benign atmosphere for media defendants than was expected’,[173] and the evaluation of journalists’ conduct by courts ‘can be very painful’.[174] Interviewees were unsure whether courts could fully understand the pressures associated with daily news production:[175] ‘[i]f you’re a newspaper on a 24 hour turnaround ... you’re not going to be able to do all the things that they want you to do’.[176] Judges and journalists appear to hold markedly different views about standards: ‘the highest level of journalistic endeavour’ achieved in the media is seen as being ‘miles short of where the court sets the lowest standard for journalistic endeavour’.[177]

Third, interviewees strongly criticised the division between judge and jury,[178] emphasising the complexity of the defence in practice.[179] Its ‘odd dichotomy’[180] of roles creates a ‘bad system’.[181] An experienced barrister stated: ‘I can’t imagine a more clumsy and unsatisfactory way of determining an issue’.[182] Practitioners thought courts had been ‘blind to the practical consequences of simply saying the jury can find the primary facts’.[183] The key difficulty appears to be the number of factual determinations juries may need to make under Reynolds.[184] A trial can end with ‘twenty questions for the jury, all on tiny bits of fact about who said what, what was the journalist told, and did the journalist phone twice to check or only once?’[185] Asking the jury to address many detailed questions is ‘completely different from how juries have always been regarded’ in English defamation and subverts their traditional role of delivering general verdicts.[186] At a trial’s outset, it is unclear what questions the jury will need to answer. Specific questions emerge from evidence — ‘if anyone ... tried to forecast in advance what questions they would have to have decided, it would [be] folly’.[187] These complexities mean that ‘if you asked the jury afterwards to explain what was going on [with] qualified privilege, [jury members] wouldn’t have a bloody clue’.[188] To the ‘bemused’ jury,[189] the questions must appear ‘slightly irrelevant and thoroughly trivial’.[190] Through such specific questions, juries will seek to deal with whether publishers deserve to succeed. As one interviewee suggested: ‘I’m not sure that’s very satisfactory at all, because embedded in the long list of factual questions which the jury [is] going to answer is always the ultimate question: was this behaviour responsible?’[191] In its allocation of judge and jury roles, Reynolds is ‘totally unsatisfactory’.[192] It makes the jury’s role far more difficult, if not ‘almost impossible’.[193]

These views led to suggestions either to increase the jury’s role on qualified privilege or sideline it. Some respondents suggested that the jury ‘should be asked general questions. Do you find for the claimant or the defendant? ... I think this intermediate idea of asking them a huge taxonomy of different questions is just unsatisfactory and counterproductive’.[194] Other interviewees suggested that early judicial rulings on privilege would be valuable.[195] But such rulings require that no factual issues be in dispute. This is difficult to achieve given the range of factors that can be relevant to Reynolds privilege. Defendants may decide ‘to concede facts’ to seek a summary ruling.[196] Some interviewees suggested removing juries entirely, or removing them in relation to qualified privilege: ‘I think the simple answer to all this is that we should get rid of the juries’.[197] Other respondents made a related point that Reynolds privilege makes more sense in judge-only trials[198] and that this change may be happening, to a degree, through judicial rulings that cases can be tried conveniently without juries.[199] Interviewees also suggested that the jury is being sidelined in that judges have room to decide whether Reynolds privilege succeeds by interpreting jury answers,[200] sometimes because of the inconsistency of the jury’s answers.[201]

Overall, although the Reynolds defence is seen as having difficulties at trial, London practitioners describe the developments as ‘sensible’[202] and ‘evidently justifiable’.[203]

C Australian Interviews

Four issues from Australian interviews are considered here. First, Lange privilege is not seen as a useable defence. Its reasonableness requirement, which almost all practitioners equate with the defence under s 22 of the Defamation Act 1974 (NSW), is the primary reason for this failing. Second, notwithstanding its legal weaknesses, Lange appears to have reduced suits by politicians. Third, the roles of judge and jury under Lange are seen as problematic, echoing concerns from England. Fourth, most interviewees suggested that Lange needs review, with Reynolds being seen as one preferable approach and the Theophanous form of common law privilege also receiving substantial support.

First, in marked contrast to the English situation under Reynolds, Australian interviewees suggested that Lange has not improved the position of publishers.[204] While Lange privilege allows defendants to raise more issues, it ‘was quite clearly constructed to be ... in results, pro-plaintiff’.[205] Judges have suggested that political discussion requires greater protection, but Lange has not achieved this ‘at all’.[206] The media has attempted to ‘stretch the boundaries’[207] of the defence. Where stories have a political ‘slant’[208] or a ‘hint’[209] of politics, Lange will be pleaded.[210] Judges, however, retain ‘the discretion to knock things out on what seem to be fairly technical grounds’.[211] For example, the interpretation of political and government matters in Popovic v Herald & Weekly Times Ltd[212] to exclude much criticism of judges was not supported,[213] with a wider scope being preferred.[214]

While Lange privilege’s scope raised some concerns, interviewees more strongly criticised the reasonableness requirement. Its restrictive interpretation was seen as a key failing in the defence.[215] Under Lange ‘you walk straight into reasonableness’.[216] Reasonableness ‘is the stumbling block and will always be the stumbling block’.[217] It is the ‘killer’[218] that one ‘always come[s] back to’:[219]

Why should a defendant have the added burden of reasonableness when the High Court [judges] are trying to espouse a principle that in political speech there should be more freedom. In fact, there’s less freedom because defendant publishers can never or hardly ever get up to that standard of reasonableness.[220]

A related point was raised about s 22 of the Defamation Act 1974 (NSW). Lange privilege is seen to have done little in New South Wales because it is merely a subset of the s 22 defence.[221] Lange is believed to operate, within the area of political communication, in effectively the same restrictive manner as s 22. As noted in Part II, academic commentary and law reform documents suggest differences between the Lange and s 22 defences.[222] Under s 22, publishers must virtually always believe publications to be true, while under Lange, it appears that they only need reasonable grounds to believe publications to be true. Notably, however, many interviewees did not see this distinction as having any practical importance.[223] It is ‘too fine’ to change arguments at trial,[224] except in very rare cases. A small minority of respondents suggested or hoped a different approach would emerge for Lange reasonableness,[225] or even that reasonableness under s 22 would be interpreted less strictly to meet constitutional requirements for political communication.[226] But the overwhelming impression was that reasonableness, under Lange or s 22, remains a very difficult hurdle.

Although comparatively narrow in scope and weak in strength, Lange privilege appears to have one effect like Reynolds: it encourages the media to seek comment.[227] Lange privilege ‘brings into sharp focus what was done before the publication, and ... the need to provide the opportunity for comment’.[228] The attention paid to investigation is good,[229] and not seeking comment would most probably cause the defence to fail.[230] To a degree this has clarified journalists’ responsibilities,[231] but interviewees were concerned about uncertainty in the concept of reasonableness.[232] Journalists need some clear benchmarks,[233] applied with flexibility in the context of each publication.[234] The law, however, sets out ‘a suggested way that things are to happen, in relation to publications ... generally ... [W]hat is reasonable in the circumstances is never looked at ... There is no differentiation for the circumstances.’[235]

Second, notwithstanding their limited impact on doctrine, Lange and the preceding High Court decisions have reduced suits by politicians and related figures,[236] and lowered the costs of settlement.[237] The ‘spectre’ of Lange privilege dissuades lawsuits.[238] Lange has given politicians ‘pause to stop’;[239] politicians now appear to have a ‘generic belief’ that they cannot sue.[240] This reported shift in litigation patterns may explain why ‘almost every barrister and judge in the country’ believes Lange has not changed anything in substance, but ‘almost every journalist in the country thinks it has’.[241]

Third, Australian interviewees echoed concerns from England about judge and jury roles. Doctrinal orthodoxy might suggest that the roles under Lange should resemble those under Reynolds — namely, where there is a jury, the jury would resolve disputed facts while judges would determine whether occasions of publication are privileged. When interviews were conducted, Australian case law had not yet examined the division of roles in any detail.[242] Some juries had been asked the general question: Was this publication reasonable?[243] Interviewees supported this approach, generally strongly.[244] It was seen as a jury question in principle,[245] and one of great importance:[246] ‘whether something is reasonable or not is quintessentially a jury question. Juries have grappled with that question always’.[247] If the jury deals with reasonableness, it need not address a potentially complex series of questions on specific factual matters.[248] As in England, practitioners supported juries having a general role in addressing reasonableness.

Fourth, Lange privilege had little support overall, apart from dissuading some political plaintiffs: ‘when you actually get to trial ... Lange’s hopeless, completely hopeless, about as hopeless as you can possibly get and still be a defence’.[249] It is ‘a bit of a disaster’.[250] Some interviewees did suggest that Lange may have more effect in time,[251] with reasonableness being interpreted less restrictively.[252] Most interviewees, however, suggested that Lange should be reviewed.[253] ‘We have all been waiting for the High Court to really take Lange to the next stage in saying “well clearly it’s been set too high, the benchmark” and we should have known that in the first place’.[254] Australian interviewees gave far more support to Reynolds.[255] The ‘very different’,[256] ‘strong’[257] and ‘sensible’[258] defence ‘has provided a better basis for evolution of doctrine than Lange’:[259]

We really need to look at this in the context of Reynolds and whether there should be a broad based public interest privilege defence ... Is there a role for it in Australia in terms of lowering the benchmark for reasonableness [and] widening the ambit outside politics?[260]

In any event, the application of Lange privilege could be developed. Although the current High Court may not be ‘minded to expand’ the defence,[261] reasonableness need not be narrow: Lange privilege could be applied far more like Reynolds while remaining unchanged in terms of law.[262]

An alternative development that interviewees raised would be a duty-interest style of privilege for media publications.[263] This existed for political publications under Theophanous. These lawyers saw the reasonableness requirement of Lange privilege as ‘wrong in principle’,[264] compared with the traditional defence that would only be defeated by malice. Such a defence would give the jury a clearer role and place defendant publishers in a much stronger position. Without such changes, truth will remain the vital defence,[265] and the aim of Lange privilege to reduce the chill of defamation law appears unlikely to be achieved.

Overall, Lange privilege was seen to offer only the smallest improvement on the traditional common law defence, and none on New South Wales’ restrictive s 22. Lange was not described with anything like the support that English practitioners offered for Reynolds, even allowing for the qualifications they had about Reynolds’ predictability and trial practicality. There was much support for reforming Australian Lange and s 22 privileges, or at least substantially developing their application.

IV CONCLUSION

Roy Baker has suggested, in a detailed comparison of Reynolds and Lange, that both Australian and English courts had given the media what it ‘had sought for decades’ by ‘significantly’ extending privilege’s application ‘to stories of public importance appearing in the mass media’.[266] The empirical research underlying this article suggests that this is true for the English developments under Reynolds, but unfortunately Baker may have been too optimistic about the Australian changes: reasonableness under Lange is ‘more onerous’ for defendants than responsibility under Reynolds.[267] Lange appears not to have improved the position of publishers to any significant degree. It has not reduced the apparent chill of defamation law, apart perhaps from reducing suits by politicians. Lange is ‘completely hopeless’[268] as a defence and, if anything, is ‘constructed to be ... in results, pro-plaintiff’.[269] There is no evidence to support the suggestion made by Callinan J in Australian Broadcasting Corporation v Lenah Games Meats Pty Ltd that Lange ‘had the impact of the detonation of a hydrogen bomb’ on defamation practitioners.[270] But Lange does appear to have encouraged more diverse reporting, in the limited sense of encouraging publications to include subjects’ points of view.

Reynolds privilege has done more to affect media conduct through the wide range of factors that are clearly relevant to the defence and the more achievable standard that appears to be required of publishers. Importantly, English practitioners suggested that the media can publish more allegations and stories of public interest under Reynolds.[271] This follows from the approach of English courts to a publication’s tone. Nevertheless, tone remains a problematic factor for tabloid newspapers,[272] which do not produce ‘the sort of journalism ... judges want to read’.[273] Grobbelaar v News Group Newspapers Ltd,[274] for example, may suggest that a publisher will be ‘penalised by the court for being a tabloid newspaper.’[275] Overall, however, the English courts’ approach to tone, particularly in relation to reportage and the degree to which publications distinguish suspicions and allegations of fact,[276] is innovative. It may grow to support a far more diverse range of public debate. These matters about tone and reportage do not arise explicitly under Lange and have been little considered in Australia.[277] The research suggests that Australian courts should make clear that a wide range of factors can be relevant to Lange reasonableness, and those factors should be applied flexibly.

The empirical research also underlined difficulties in litigation practice that may follow from judge and jury roles. The problems in seeking jury responses to multiple, detailed questions may thwart many of the benefits sought through the doctrinal changes. In England, the division in roles was very strongly criticised,[278] while in Australia there was strong support for asking juries a general question about whether publication was reasonable.[279] Subsequent Australian case law is against that approach,[280] but the research reported here suggests that both reasonableness under Lange and responsibility under Reynolds should be determined by the jury. To do otherwise will make trials difficult to run and could severely undermine protection for discussion of matters of political or public importance. While another response to the interview material could be to advocate judge-only trials for defamation,[281] existing research suggests the benefits in simplifying defamation litigation that are offered by jury trial.[282] Most Australian courts, in jurisdictions where defamation juries are available, have some flexibility to use the jury in a general role of assessing reasonableness.[283] This article’s research strongly suggests that Australian courts should take that opportunity. English trial courts may have more difficulty, after statements about judge and jury roles in Reynolds,[284] but the preferable approach is equally clear. If English law cannot move directly to having the jury determine the responsibility or otherwise of publication, English courts will need to contain the complexities posed by a large number of jury questions,[285] without unnecessarily limiting the defence. Appellate courts should not be ‘blind to the practical consequences’ of judge and jury roles in moving to simplify those roles.[286]

Lange and Reynolds are worth examining comparatively not just because they are significant legal developments with potentially important influences on speech. Reynolds also appears to have been under-appreciated in some Australian material. A 2002 New South Wales report certainly did not overlook the Reynolds decision.[287] It examined the factors set out in Reynolds, noting that no factor required publishers to establish that they believe in their publications’ truth. It recommended amending s 22 of the Defamation Act 1974 (NSW) to similar effect:

the ‘reasonableness requirement’ in s 22 should have attached to it a statutory set of factors to be considered by a court in determining whether the publication is protected by qualified privilege. Such a list ought to make clear to decision makers that it is not necessary for a publisher who wishes to invoke qualified privilege to prove that they had objective grounds for believing in the truth of the matter published.[288]

New South Wales legislation now includes a wider set of factors to consider in relation to reasonableness.[289] Interviewees, however, suggested these changes added little or nothing to the law.[290] The list of factors is ‘insulting’[291] and ‘an absolute waste of time’.[292] It merely re-expresses what courts have always been able to do.[293] However, the new statutory factors reflect the Reynolds approach. Given widespread support for Reynolds by Australian interviewees, s 22(2A) of the Defamation Act 1974 (NSW) may offer a very useful basis for innovative arguments to revise reasonableness under the New South Wales statutory defence. For example, the legislation now explicitly refers to ‘the extent to which the matter published distinguishes between suspicions, allegations and proven facts’.[294] This could support arguments drawing on the English reportage cases.[295] Since s 22 is wide — it is not limited to Lange privilege’s political communication — the arguments should be open in many cases. This article’s examination of Reynolds in action suggests that such a reinterpretation of s 22 is warranted and that decisions on Reynolds privilege offer much to New South Wales courts.

While the recent New South Wales report did not overlook Reynolds, other Australian proposals have sidestepped its possibilities. A 2003 report from Western Australia, for example, notes that the State has power to broaden or strengthen the constitutionally-based Lange defence. But the report simply suggests: ‘The [proposed new Western Australian defamation] Act should do no more than refer to the existence of this defence [related to discussion on political or government matters] — it should not attempt to define it or prescribe its limits.’[296] Similarly, a brief discussion paper issued by the Commonwealth Attorney-General in March 2004 does not refer to Reynolds at all.[297] It suggests that a statutory defence for reasonable publication should replace all qualified privilege. The defence would be modelled on s 22 of the Defamation Act 1974 (NSW), notwithstanding its very restrictive history.[298] The research reported in this article suggests that the Attorney-General’s initial proposal for qualified privilege would do nothing to encourage diverse speech on matters of public concern. Rather, such a form of statutory qualified privilege would restrict speech and maintain the chill that appears to exist under Australian defamation law.

This article’s doctrinal and empirical research points towards a broader and stronger Australian defence — one that is neither limited to a narrow version of political communication, nor requires an onerous standard of reasonableness. First, Lange should be broader. The diverse range of publications that can seek protection on Reynolds privilege,[299] the criticism of Lange’s narrow scope in interviews[300] and existing academic literature[301] all support a broader Australian defence. As Lord Cooke noted in Reynolds itself:

It is doubtful whether the ... new defence could sensibly be confined to political discussion. There are other public figures who exercise great practical power over the lives of people or great influence in the formation of public opinion or as role models. [Their] power or influence may indeed exceed that of most politicians. The rights and interests of citizens in democracies are not restricted to the casting of votes. Matters other than those pertaining to government and politics may be just as important in the community; and they may have as strong a claim to be free of restraints on freedom of speech.[302]

The High Court has noted that it is possible for the common law defence under Lange to be broader than the constitutional protection for political communication.[303] Thus the development appears to be open in Australian law, and it should be made. Existing hesitancy by appellate courts should not dispose of the issue,[304] and Reynolds should be drawn on to develop Lange privilege. Second, Lange should be stronger. The difficult task of meeting the reasonableness requirement was one of the clearest problems seen through the interview research.[305] Again, as already discussed for s 22 in the New South Wales legislation, Reynolds offers very useful material.

In terms of how developments in Lange or s 22 may be promoted, recent comments by two members of the High Court are noteworthy. In Rogers v Nationwide News Pty Ltd,[306] Gleeson CJ and Gummow J discussed the need for greater consideration of the circumstances of news production:

In the respondent’s written submissions, reference was made, without elaboration, to ‘the circumstances in which daily newspapers are published’. It may be enlightening if ... courts were given more evidence as to those circumstances. ... Where, as here, serious errors are made, and attributed to ‘the circumstances in which daily newspapers are published’, a court would be in a better position to judge the reasonableness of the publisher’s conduct if it were told exactly what those circumstances were, why they prevailed, and how they contributed to the error.[307]

Their Honours’ invitation for more detailed material on news production accords with ongoing academic research into defamation law and news production.[308] It suggests that if the media can provide more substantial material about what is reasonable, then the courts’ strict approach could be revised as they adopted more appropriate approaches to imposing ‘“quality control” standards within journalism’.[309] The English experiences outlined in this article underline how poorly Australian law has responded to defamation law’s traditional limits.

English and Australian law is attempting to overcome the traditional limitations on speech that follow from requiring many defendants to prove truth. As Barendt and his co-researchers noted of the situation before Reynolds:

in order to have a defence against every possible defamation ... [newspapers] would require certainty that in every case there were witnesses willing and able to appear in court ... or that conclusive and legally admissible documentary evidence be in the editor’s hands. If such certainty were required for everything controversial, there would be no newspapers worth reading.[310]

Reynolds offers much to address such concerns, and it may inform Australian developments more easily than moving to the stronger protection offered by a Theophanous-style privilege that applied to matters of public interest.[311] That, however, may be where the law must move if it is to meet the aims of Lange and Reynolds to reduce defamation law’s burden of proving truth.[312] Whichever developments are pursued, if privilege defences are to assist diverse speech and reduce any chill from defamation law, courts would do well to remain ‘slow to conclude that a publication [is] not in the public interest’ and resolve doubts ‘in favour of publication’.[313]


[∗] LLB (Hons) (Melb), LLM (Dist) (London), PhD (Melb); Senior Lecturer and Director, Centre for Media and Communications Law (‘CMCL’), Law School, The University of Melbourne; Editor, Media & Arts Law Review. This paper draws on research supported by the University of Melbourne and its Faculty of Law (ECR Grant 2003) and research supported by the Australian Research Council (Andrew T Kenyon and Tim Marjoribanks, Defamation Law in Context: Australian and US News Production Practices and Public Debate (Discovery Project DP0343258, 2003–05)). Many thanks to Chris Dent (Research Fellow, CMCL) and Jason Bosland for their research assistance, and to all those who participated in the fieldwork: see below n 117.

[1] For a review of the term’s United States history, see Frederick Schauer, ‘Fear, Risk and the First Amendment: Unravelling the “Chilling Effect”’ (1978) 58 Boston University Law Review 685.

[2] Eric Barendt et al, Libel and the Media: The Chilling Effect (1997).

[3] Ibid 191–4.

[4] Chris Dent and Andrew T Kenyon, ‘Defamation Law’s Chilling Effect: A Comparative Content Analysis of Australian and US Newspapers’ (2004) 9 Media & Arts Law Review 89.

[5] For detail about the US position, see Practising Law Institute, Sack on Defamation: Libel, Slander, and Related Problems (3rd ed, at April 2002); Libel Defense Resource Center, LDRC 50-State Survey 2002–03: Media Libel Law (2002).

[6] Dent and Kenyon, above n 4, 106.

[7] Further content analysis and news production research is being undertaken as part of ongoing research at the CMCL: see author’s note. Among other issues, it should allow more detailed understanding of the wide variety of factors influencing media content and the particular contexts in which different defamation laws operate.

[8] Barendt et al, above n 2, 77.

[9] The presumption of general damages, which is another key aspect in the idea of a chilling effect, has received wide criticism: see, eg, Eric Barendt, ‘What Is the Point of Libel Law?’ (1999) 52 Current Legal Problems 110; John Fleming, ‘Retraction and Reply: Alternative Remedies for Defamation’ (1978) 12 University of British Columbia Law Review 15.

[10] [1997] HCA 25; (1997) 189 CLR 520 (‘Lange’).

[11] [2001] 2 AC 127 (‘Reynolds’). Similar changes have occurred in New Zealand, India and South Africa: see Adrienne Stone and George Williams, ‘Freedom of Speech and Defamation: Developments in the Common Law World’ [2000] MonashULawRw 15; (2000) 26 Monash University Law Review 362.

[12] David Price and Korieh Duodu, Defamation: Law, Procedure and Practice (3rd ed, 2004) 107.

[13] Jameel v Wall Street Journal Europe SPRL [No 2] [2004] EMLR 196, 202–7 (Eady J) (‘Jameel’). The terms ‘Reynolds privilege’ and ‘Lange privilege’ are used in this article for clarity and brevity, rather than to suggest the defences have any necessary or complete separation from traditional qualified privilege.

[14] See Loutchansky v Times Newspapers Ltd [Nos 2–5] [2001] EWCA Civ 1805; [2002] QB 783. Reynolds privilege attaches to publications to the world at large, rather than to occasions in the traditional sense: Kearns v General Council of the Bar [2003] EWCA Civ 1167; [2003] 1 WLR 1357, 1358 (Simon Brown LJ).

[15] Michael Chesterman, Freedom of Speech in Australian Law: A Delicate Plant (2000) 106.

[16] Ibid 109.

[17] Left aside for separate analysis is the defence of qualified protection under the Queensland and Tasmanian defamation Codes, which appears wider and stronger than Lange in protecting publications made in good faith ‘for the public good’: see Des Butler and Sharon Rodrick, Australian Media Law (2nd ed, 2004) 86–92. Similarly not examined is the Australian Capital Territory defence of ‘non-negligent publication’ under s 66 of the Civil Law (Wrongs) Act 2002 (ACT), which has not yet received judicial interpretation: see Matt Collins, ‘New Defamation Law for the ACT’ (2001) 6 Media & Arts Law Review 335. Also worth noting is the parallel New Zealand development of qualified privilege in Lange v Atkinson [2000] NZCA 95; [2000] 3 NZLR 385: see Stone and Williams, above n 11, 374–7; Michael Gillooly, The Third Man: Reform of the Australasian Defamation Defences (2004) 129–39; Rosemary Tobin, ‘Political Discussion in New Zealand: Cause for Concern?’ [2003] New Zealand Law Review 215.

[18] Some material in Part II is examined in greater detail in Andrew T Kenyon, ‘Developments in Qualified Privilege: England, Australia and Ireland?’ in Eoin O’Dell (ed), Freedom of Expression (forthcoming).

[19] [1994] HCA 46; (1994) 182 CLR 104 (‘Theophanous’). See below nn 6671 and accompanying text.

[20] See, eg, Doreen McBarnet, ‘Magistrates’ Courts and the Ideology of Justice’ (1981) 8 British Journal of Law & Society 181; Richard Ingleby, Solicitors and Divorce (1992).

[21] See, eg, Richard Johnstone, The Court and the Factory: The Legal Construction of Occupational Health and Safety Offences in Victoria (PhD Thesis, The University of Melbourne, 1994).

[22] Stewart Macaulay, ‘Non-Contractual Relations in Business: A Preliminary Study’ (1963) 28 American Sociological Review 55.

[23] Significant northern hemisphere examples include: Belinda Fehlberg, Sexually Transmitted Debt: Surety Experience and English Law (1997); Patricia Ewick and Susan Silbey, The Common Place of Law: Stories from Everyday Life (1998); Austin Sarat and Thomas Kearns (eds), Law in Everyday Life (1993); William O’Barr and John Conley, ‘Lay Expectations of the Civil Justice System’ (1988) 22 Law & Society Review 137. Some Australian use of similar methods exists, for example, in sex discrimination: Rosemary Hunter and Alice Leonard, The Outcomes of Conciliation in Sex Discrimination Cases (1995).

[24] Issues of social power and access to law remain largely distinct from this doctrinal focus, although they appear acute for defamation: see, eg, Michael Newcity, ‘The Sociology of Defamation in Australia and the United States’ (1991) 26 Texas International Law Journal 1; Chesterman, above n 15, 173–4.

[25] Toogood v Spyring [1834] EngR 363; (1834) 1 Cr M & R 181, 193; [1834] EngR 363; 149 ER 1044, 1050 (Parke B).

[26] Adam v Ward [1917] AC 309, 334 (Lord Atkinson).

[27] Chapman v Ellesmere [1932] 2 KB 431.

[28] Braddock v Bevins [1948] 1 KB 580. The media might be able to rely on an ancillary or derivative protection, for example, if it published one person’s reply to an attack by another.

[29] Some jurisdictions have significant statutory fair report defences: see, eg, Defamation Act 1996 (UK) c 31, s 15, sch 1; Defamation Act 1974 (NSW) ss 24, 25, sch 2.

[30] See, eg, Wason v Walter (1868) LR 4 QB 73; Curry v Walter [1796] EngR 913; (1796) 1 Bos & P 525; 126 ER 1046. For a re-examination of the history of qualified privilege and widespread publications, see Chris Dent, ‘“The Privileged Few” and Public Discourse: Practices of Classification in a History of Comment and Qualified Privilege’ (Paper presented at the Australian and New Zealand Law and History Society, 23rd Annual Conference, Perth, 3 July 2004).

[31] See, eg, Geoffrey Robertson and Andrew Nicol, Media Law (4th ed, 2002) 73–5; Chesterman, above n 15, 81–91.

[32] Alan Ruddock and John Burns, ‘Goodbye Gombeen Man’, The Sunday Times (London), 20 November 1994, 19.

[33] Reynolds [2001] 2 AC 127, 195 (Lord Nicholls).

[34] Richard Shillito and Eric Barendt, ‘Libel Law’ in Eric Barendt and Alison Firth (eds), The Yearbook of Copyright and Media Law 2000 (2000) 391, 407.

[35] Ian Loveland, ‘Freedom of Political Expression: Who Needs the Human Rights Act?’ [2001] Public Law 233, 233.

[36] Kevin Williams, ‘Defaming Politicians: The Not So Common Law’ (2000) 63 Modern Law Review 748, 754. However, it has also been suggested that English judges were merely ‘tiptoeing towards free speech’: Geoffrey Robertson, ‘Tiptoeing towards Free Speech’, The Guardian (London), 29 October 1999, 22, cited in Ian Loveland, ‘Political Libels — Whose Right Is It Anyway?’ (2000) 53 Current Legal Problems 333, 356–7.

[37] Reynolds [2001] 2 AC 127, 200, 204 (Lord Nicholls), 217 (Lord Cooke agreeing), 237 (Lord Hobhouse agreeing), 210–11 (Lord Steyn), 234–5 (Lord Hope). This followed the rejection by earlier law reform reports of a defence for public interest statements believed to be true and published with reasonable care, notably the Faulks Committee report: United Kingdom, Report of the Committee on Defamation, Cmnd 5909 (1975) 53–5.

[38] Opened for signature 4 November 1950, 213 UNTS 221 (entered into force 3 September 1953).

[39] Shillito and Barendt, above n 34, 408. See also Loveland, ‘Political Libels’, above n 36, 351; Paul Mitchell, ‘Political Qualified Privilege’ (2000) 11 King’s College Law Journal 114, 115; Gaddafi v Telegraph Group Ltd [2000] EMLR 431.

[40] See Loveland, ‘Freedom of Political Expression’, above n 35, 233: ‘While the Reynolds principle does not extend to political information as a generic category, the judgment will undoubtedly provide a substantial degree of legal protection to diligent press coverage of political stories.’

[41] See, eg, Jameel [2004] EMLR 196, 205 (Eady J).

[42] [2001] 2 AC 127, 205:

1. The seriousness of the allegation. The more serious the charge, the more the public is misinformed and the individual harmed, if the allegation is not true. 2. The nature of the information, and the extent to which the subject matter is a matter of public concern. 3. The source of the information. Some informants have no direct knowledge of the events. Some have their own axes to grind, or are being paid ... 4. The steps taken to verify the information. 5. The status of the information. The allegation may have already been the subject of an investigation which commands respect. 6. The urgency of the matter. News is often a perishable commodity. 7. Whether comment was sought from the plaintiff ... An approach to the plaintiff will not always be necessary. 8. Whether the article contained the gist of the plaintiff’s side of the story. 9. The tone of the article. A newspaper can raise queries or call for an investigation. It need not adopt allegations as statements of fact. 10. The circumstances of the publication, including the timing.

[43] The Civil Procedure Rules 1998 (UK), which commenced on 26 April 1999, have substantially changed civil litigation in England, including altering some terminology: plaintiffs have become claimants, writs are called claim forms, and pleadings have been renamed statements of case. In relation to the Rules’ influence on defamation, see Price and Duodu, above n 12, 271–5.

[44] Although the House of Lords confirmed that a wider qualified privilege defence existed in law, a majority held that the defence was unavailable to The Sunday Times at retrial because the publisher had failed to put Reynolds’ side of the story when making such serious allegations of political misconduct: Reynolds [2001] 2 AC 127, 206 (Lord Nicholls), 217 (Lord Cooke agreeing), 237 (Lord Hobhouse agreeing); cf 216–17 (Lord Steyn), 237 (Lord Hope).

[45] Shillito and Barendt, above n 34, 410.

[46] Williams, above n 36, 753. See also Roy Baker, ‘Extending Common Law Qualified Privilege to the Media: A Comparison of the English and Australian Approaches’ (2002) 7 Media & Arts Law Review 87; and the detailed consideration of each factor in Robertson and Nicol, above n 31, 130–3.

[47] Grobbelaar v News Group Newspapers Ltd [2001] EWCA 33; [2001] 2 All ER 437 (Court of Appeal). Note that this aspect of the decision was not challenged on appeal: Grobbelaar v News Group Newspapers Ltd [2002] UKHL 40; [2002] 1 WLR 3024 (House of Lords).

[48] Jameel [2004] EMLR 196, 209–10 (Eady J). There is a danger, however, that urgency will be evaluated with hindsight, rather than from the perspective of publishers who must decide, based on their own skill and experience, whether delaying a story is likely to produce more information about a claimant’s version of events: Ian Cram, ‘Reducing Uncertainty in Libel Law after Reynolds v Times Newspapers? Jameel and the Unfolding Defence of Qualified Privilege’ (2004) 15 Entertainment Law Review 147, 149.

[49] James Gilbert Ltd v MGN Ltd [2000] EMLR 680; see also Khan v Euro Bangla Newspaper (Unreported, Queen’s Bench Division, Eady J, 13 January 2004).

[50] Grobbelaar v News Group Newspapers Ltd [2001] EWCA 33; [2001] 2 All ER 437 (Court of Appeal).

[51] James Gilbert Ltd v MGN Ltd [2000] EMLR 680, 700 (Eady J).

[52] Jameel [2004] EMLR 196.

[53] Bonnick v Morris [2002] UKPC 31; [2003] 1 AC 300, 309 (Lord Nicholls).

[54] Ibid.

[55] [2001] EWCA Civ 1634; [2002] EMLR 215 (‘Al-Fagih’).

[56] See Shah v Standard Chartered Bank [1998] EWCA Civ 612; [1999] QB 241.

[57] Al-Fagih [2001] EWCA Civ 1634; [2002] EMLR 215, 231 (Simon Brown LJ).

[58] Mark v Associated Newspapers Ltd [2002] EMLR 839, 856 (Simon Brown LJ), 859 (Mummery and Dyson LJJ agreeing). A differently constituted Court of Appeal has referred to Al-Fagih without disapproval: Loutchansky v Times Newspapers Ltd [Nos 2–5] [2001] EWCA Civ 1805; [2002] QB 783, 806–8 (Lord Phillips MR, Simon Brown and Tuckey LJJ). The decision has also been applied at first instance: Al Misnad v Azzaman Ltd [2003] EWHC 1783 (Unreported, Queen’s Bench Division, Gray J, 21 July 2003).

[59] Shillito and Barendt, above n 34, 411.

[60] See, eg, Bashford v Information Australia (Newsletters) Pty Ltd (2004) 204 ALR 193, 201 (Gleeson CJ, Hayne and Heydon JJ), 247 (Kirby J), cf 203 (McHugh J).

[61] See Loveland, ‘Political Libels’, above n 36, 357.

[62] [2001] EMLR 898, 912 (Gray J). The questions should be framed to seek ‘yes’ or ‘no’ answers: Jameel [2004] EMLR 196, 200 (Eady J).

[63] Jameel [2004] EMLR 196, 199–200 (Eady J).

[64] Ibid.

[65] See especially Australian Capital Television Pty Ltd v Commonwealth (1992) 177 CLR 106 (‘ACTV’); Nationwide News Pty Ltd v Wills [1992] HCA 46; (1992) 177 CLR 1; Theophanous [1994] HCA 46; (1994) 182 CLR 104; Stephens v West Australian Newspapers Ltd [1994] HCA 45; (1994) 182 CLR 211 (‘Stephens’); Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1; Lange [1997] HCA 25; (1997) 189 CLR 520; Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579; Roberts v Bass [2002] HCA 57; (2002) 212 CLR 1.

[66] Theophanous [1994] HCA 46; (1994) 182 CLR 104. See also Stephens [1994] HCA 45; (1994) 182 CLR 211. For an examination of the Theophanous defence and subsequent developments, see Butler and Rodrick, above n 17, 77–82; Sally Walker, Media Law: Commentary and Materials (2000) 207–28; Michael Gillooly, The Law of Defamation in Australia and New Zealand (1998) 188–96.

[67] Theophanous [1994] HCA 46; (1994) 182 CLR 104, 140 (Mason CJ, Toohey and Gaudron JJ).

[68] Ibid.

[69] The Theophanous joint judgment referred approvingly to the use by Mason CJ of the term ‘public affairs’ in ACTV (1992) 177 CLR 106, 138–40: Theophanous [1994] HCA 46; (1994) 182 CLR 104, 122–3 (Mason CJ, Toohey and Gaudron JJ).

[70] Theophanous [1994] HCA 46; (1994) 182 CLR 104, 124 (Mason CJ, Toohey and Gaudron JJ).

[71] Ibid, citing Eric Barendt, Freedom of Speech (1985) 152.

[72] Lange [1997] HCA 25; (1997) 189 CLR 520, 521.

[73] Ibid 571.

[74] Ibid 574.

[75] Desiring to harm political opponents does not constitute malice: ibid. See also Brander v Ryan [2000] SASC 446; (2000) 78 SASR 234, 250 (Lander J), 235 (Prior J agreeing), 250 (Bleby J agreeing); Roberts v Bass [2002] HCA 57; (2002) 212 CLR 1, 14 (Gleeson CJ), 31–3 (Gaudron, McHugh and Gummow JJ), 70 (Kirby J).

[76] Herald & Weekly Times Ltd v Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [504] (Warren AJA) (‘Popovic’). Some examples, however, do exist of wider approaches being taken to political communication, which may support arguments for a wider defence; for example, in a context outside defamation, see Australian Broadcasting Corporation v Lenah Games Meats Pty Ltd (2001) 208 CLR 199, 281–2 (Kirby J) (‘Lenah’).

[77] Ibid [6], [9]–[10] (Winneke ACJ), [498]–[500], [504] (Warren AJA), cf [247]–[251] (Gillard AJA). See also O’Shane v John Fairfax Publications Pty Ltd [2004] NSWSC 140 (Unreported, Smart AJ, 16 March 2004) [187]–[196].

[78] Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [6], [9]–[10] (Winneke ACJ), [247]–[251] (Gillard AJA), cf [498]–[500], [504] (Warren AJA).

[79] Ibid [6], [9]–[10] (Winneke ACJ).

[80] See, eg, Amalgamated Television Services Pty Ltd v Marsden [2002] NSWCA 419 (Unreported, Beazley, Giles and Santow JJA, 24 December 2002) [1160] (‘Marsden’).

[81] [2002] SASC 160; (2002) 82 SASR 152 (Debelle J). Similarly, in NRMA v John Fairfax Publications Pty Ltd [2002] NSWSC 563 (Unreported, Master Macready, 26 June 2002), material about aspiring candidates for a corporate board was beyond Lange’s scope.

[82] [2003] SASC 398; (2003) 87 SASR 62 (‘Chapman’). For general background to the development, see Margaret Simons, The Meeting of the Waters: The Hindmarsh Island Affair (2003), or the brief background provided by Besanko J in the Full Court: Chapman [2003] SASC 398; (2003) 87 SASR 62, 102–12.

[83] Chapman [2003] SASC 398; (2003) 87 SASR 62, 71 (Doyle CJ), 127–8 (Besanko J), cf 98 (Gray J). This publication also exceeded the defence’s scope at trial, even when the concept of ‘Strategic Lawsuits Against Public Participation’ (‘SLAPPs’) was raised in the publications: Chapman v Conservation Council of South Australia [2002] SASC 4; (2002) 82 SASR 449, 522–4 (Williams J). For a brief background to the development of SLAPPs, see Andrew T Kenyon, ‘Defamation and Critique: Political Speech and New York Times v Sullivan in England and Australia’ [2001] MelbULawRw 18; (2001) 25 Melbourne University Law Review 522, 524, 530–2.

[84] Chapman [2003] SASC 398; (2003) 87 SASR 62, 72 (Doyle CJ), 97 (Gray J), 142 (Besanko J). This publication also came within the defence’s scope at trial: [2002] SASC 4; (2002) 82 SASR 449, 528 (Williams J). Interlocutory decisions supporting the idea that Lange applies to discussion about public administration include: Bristile Ltd v Buddhist Society of Western Australia Inc [1999] WASC 259; [2000] Aust Torts Reports 81-548; Cock v Hughes [2002] WASC 108 (Unreported, Hasluck J, 14 May 2002); Archer v Channel Seven Perth Pty Ltd [2002] WASC 160 (Unreported, Hasluck J, 21 June 2002); Marsden [2002] NSWCA 419 (Unreported, Beazley, Giles and Santow JJA, 24 December 2002) [1149]; Shave v West Australian Newspapers Ltd [2003] WASC 83 (Unreported, Hasluck J, 5 May 2003). In Marsden, however, the New South Wales Court of Appeal expressed in obiter doubts that Lange did have this breadth: [2002] NSWCA 419 (Unreported, Beazley, Giles and Santow JJA, 24 December 2002) [1162].

[85] [2003] NSWSC 505 (Unreported, Simpson J, 15 August 2003).

[86] Ibid [135]. Simpson J held that

the councillors were expressing their concerns, not about matters of policy or government, but about what they perceived to be the flouting by the plaintiff of the terms of the development consent, and their opposition to the development consent ... granted by the Land and Environment Court.

[87] See, eg, Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [8] (Winneke ACJ), [244] (Gillard AJA).

[88] See Featherston v Tully [No 2] [2002] SASC 338; (2002) 83 SASR 347 (Court of Disputed Returns); Brander v Ryan [2000] SASC 446; (2000) 78 SASR 234.

[89] [1997] HCA 25; (1997) 189 CLR 520, 573–5.

[90] See Morgan v John Fairfax & Sons Ltd [No 2] (1991) 23 NSWLR 374 (Court of Appeal).

[91] Ibid 388 (Hunt AJA).

[92] Steven Rares, ‘Can I Say That?’ (2004) 25 Australian Bar Review 45, 47.

[93] See, eg, Merivale v Carson [1887] UKLawRpKQB 202; (1887) 20 QBD 275, 281 (Lord Esher MR), 283–4 (Bowen LJ). See also Branson v Bower [2001] EWHC 460; [2002] QB 737, 741 (Eady J).

[94] Sally Walker, ‘Lange v ABC: The High Court Rethinks the “Constitutionalisation” of Defamation Law’ (1998) 6 Torts Law Journal 9, 21.

[95] Pinniger v John Fairfax & Sons Ltd [1920] ArgusLawRp 15; (1979) 26 ALR 55; Barbaro v Amalgamated Television Services Pty Ltd (1989) 20 NSWLR 493; Bowin Designs Pty Ltd v Australian Consumers Association [1996] 1070 FCA 1 (Unreported, Lindgren J, 6 December 1996); Lear v Malter (Unreported, Supreme Court of New South Wales, Donovan AJ, 14 March 1997). In addition, the defence was set aside on appeal in Evatt v Nationwide News Pty Ltd [1999] NSWCA 99 (Unreported, Sheller, Powell and Giles JJA, 19 April 1999).

[96] See, eg, Walker, ‘Lange v ABC’, above n 94, 28.

[97] See, eg, Rogers v Nationwide News Pty Ltd [2003] HCA 52; (2003) 201 ALR 184, 191 (Gleeson CJ and Gummow J).

[98] See above n 74 and accompanying text.

[99] Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [92], [197] (Gillard AJA).

[100] In Chapman, no response needed to be sought where the subjects of a publication were ‘making ample use of the media’ to convey their points of view: [2003] SASC 398; (2003) 87 SASR 62, 143 (Besanko J), 73 (Doyle CJ agreeing), 98 (Gray J). See also Brander v Ryan [2000] SASC 446; (2000) 78 SASR 234, 249, 250 (Lander J), 235 (Prior J agreeing), 250 (Bleby J agreeing); Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [213]–[218] (Gillard AJA), [453] (Warren AJA agreeing); O’Shane v John Fairfax Publications Pty Ltd [2004] NSWSC 140 (Unreported, Smart AJ, 16 March 2004) [204]; Cock v Hughes [2002] WASC 263 (Unreported, McLure J, 11 November 2002) [42]; Mickelberg v 6PR Southern Cross Radio Pty Ltd [2003] WASC 209 (Unreported, Master Newnes, 31 October 2003) [52]–[62].

[101] Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [92] (Gillard AJA).

[102] Courts do at times recognise that this means decisions under s 22 may have limited influence for the common law defence: see, eg, ibid [201] (Gillard AJA). On the strong, apparent benefits of the common law cause of action, see Andrew T Kenyon, ‘Imputation of Publication: The Cause of Action in Defamation Law’ [2004] UNSWLawJl 5; (2004) 27 University of New South Wales Law Journal 100.

[103] Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003) [13] (Winneke ACJ), [104], [116]–[117] (Gillard AJA), [509] (Warren AJA).

[104] See New South Wales Attorney-General’s Task Force on Defamation Law Reform, Defamation Law: Proposals for Reform in NSW (2002) 20, citing Kenyon, ‘Defamation and Critique’, above n 83, 544.

[105] In particular, Bonnick v Morris [2002] UKPC 31; [2003] 1 AC 300 illustrates the value of a non-technical approach to a publication’s meaning for the privilege defence. Cf the history of s 22 in New South Wales: see above nn 8996 and accompanying text. As to the very influential role of meaning in defamation law and litigation more generally, see Andrew T Kenyon, Word Games: Meaning in Defamation Law and Practice in England, New South Wales and Victoria (PhD Thesis, The University of Melbourne, 2002).

[106] Al-Fagih [2001] EWCA Civ 1634; [2002] EMLR 215.

[107] Grobbelaar v News Group Newspapers Ltd [2001] EWCA 33; [2001] 2 All ER 437 (Court of Appeal).

[108] The interviews were associated with the two projects mentioned in the author’s note, which examined many aspects of defamation law and practice.

[109] The two chambers at 1 Brick Court and 5 Raymond Buildings account for a substantial majority of all London defamation work.

[110] All English interviews were conducted in May 2003. They are identified here as 1 to 23. Five interviewees were women, and all interviews were conducted separately except for two pairs: 16 and 17, and 18 and 19.

[111] There were 16 interviewees from New South Wales and 11 from Victoria.

[112] Two Australian interviews were conducted in December 2002 (40 and 41), with all the others occurring in June and July 2003. The interviewees are identified here as 24 to 39 from New South Wales, and 40 to 50 from Victoria. Five interviewees were women, and all interviews were conducted separately, except for 45 and 46.

[113] Obtaining a sample of experienced defamation lawyers means a sizeable minority of interviewees acted predominantly for the media — for example, in solicitors’ firms retained by media companies to give pre-publication and litigation-related legal advice. However, a diverse group of experienced lawyers was sought, and the most common background for interviewees was to do significant amounts of work for both defamation plaintiffs (and claimants) and for defamation defendants (whether they were media or non-media defendants). A small minority of interviewees did more work for defamation plaintiffs (and claimants). In any event, no direct relationship was apparent between interviewees’ professional backgrounds and their views about defamation law and litigation. For example, one strong comment within the Australian interviews was that Lange had not improved the position of publishers (see below n 204), while in England many interviewees suggested Reynolds had improved the media’s position (see below nn 130–3 and accompanying text). In Australia, the 13 responses came from five lawyers who predominantly did media defence work, one lawyer who conducted more plaintiff work, and seven lawyers with roughly even practices. In England, the 16 responses came from six lawyers with predominantly media practices, three lawyers with mainly claimant practices, and seven lawyers with roughly equal practices.

[114] Handwritten notes were made at two interviews: once due to background noise preventing audio-recording, and once to accord with the interviewee’s wishes.

[115] Such editing of interview transcripts is common; for a discussion of the general approach, see, eg, Ewick and Silbey, above n 23, 259.

[116] QC is the title used in England, while both QC and SC are used in Australia. In this article, QC is used for senior English barristers and SC for Australian ones. Judges are referred to as QC or SC to mirror their major prior experience.

[117] These individuals assisted with the interviews used here (each is identified with any title then held): Adrian Anderson, Ian Angus, Peter Bartlett, Alastair Brett, Desmond Browne QC, Godwin Busuttil, Bruce Burke, Siobhain Butterworth, David Caspersonn, Iain Christie, Richard Coleman, Dr Matthew Collins, Stephen Collins, Jacob Dean, the Hon Justice David Eady, Anne Flahvin, the Hon Justice Charles Gray, Tim Hale SC, Alister Henskens, Will Houghton QC, Jan Johannes, the Hon Justice David Levine RFD, Jennifer McDermott, Michael Martin, Tom Molomby SC, Patrick Moloney QC, Justine Munsie, Leanne Norman, Peter O’Donahoo, Laurence Maher, Matthew Nicklin, Stephen O’Meara, Adrian Page QC, Marcus Partington, Susan Poffley, Richard Potter, David Price, Nicholas Pullen, Richard Rampton QC, Steven Rares SC, Rhory Robertson, Georgina Schoff, Richard Shillito, Michael Skrein, Michael Smyth, Adam Speker, Belinda Thompson, Robert Todd, Mark Warby QC and Simon Wilson QC. Many other practitioners assisted with the two larger research projects: see author’s note. Their contributions will be reported elsewhere.

[118] The interview transcriptions drawn on here, which dealt with many aspects of defamation law and practice, totalled 400 000 words.

[119] See Clive Seale, The Quality of Qualitative Research (1999) 59.

[120] For a brief discussion of questions about meaning-equivalence in comparative research, see Tim May, Social Research: Issues, Methods and Process (1993) 161–2. Clive Seale provides an overview of similar concerns about survey research: ibid 133–5.

[121] This may be a more useful concept than ‘generalizability’: Pertti Alasuutari, Researching Culture: Qualitative Method and Cultural Studies (1995) 156–7, cited in David Silverman, Doing Qualitative Research: A Practical Handbook (2000) 110–11.

[122] See Silverman, Doing Qualitative Research, above n 121, 175. For an overview of current qualitative approaches to reliability, see Seale, above n 119, ch 10.

[123] See Silverman, Doing Qualitative Research, above n 121, 177.

[124] Ibid 10–11.

[125] The approach follows that of David Silverman, Interpreting Qualitative Data: Methods for Analysing Talk, Text and Interaction (1993) 163. See generally Seale, above n 119, ch 9. Note, however, that not every interviewee commented on each issue. Interviews were semi-structured, allowing respondents a degree of latitude to focus on what they believed to be the most significant matters about defamation law and litigation. The footnoting of interviewee numbers is not intended to suggest percentage responses to particular issues. Rather — and similarly to the way in which it would be misleading to treat interviewees primarily as advocates for, or against, the media (see above n 113) — the material reported here should be assessed with the understanding that: where an issue receives many comments, practitioners see the issue as important; where many interviewees hold a particular view, that view is likely to be common among experienced practitioners; and divergences of views are noted in the reporting.

[126] Almost all interviews emphasised the decision’s importance: 1S, 2S, 3S, 4QC, 5QC, 8B, 9B, 11S, 12B, 14S, 16S, 17S, 18S, 19S, 21QC, 22S. Along with some of these respondents, other interviewees suggested that Reynolds will develop significance over time (7QC, 10B) while 20QC suggested that it was too early to tell whether it would achieve the desirable strengthening of qualified privilege. 6B emphasised the defence’s unpredictability, with which 15QC agreed (while noting it has led to innovative subsequent decisions). 13QC supported a qualified privilege defence for media publications but believed that the traditional duty-interest style of qualified privilege, expanded to include at least some media publications, would be far preferable.

[127] 11S; similar 2S, 6B, 10B, 19S, 21QC. However, many respondents noted the differing positions of broadsheet and tabloid newspapers: eg, 16S, 17S, 22S. Others also noted the importance of deciding whether to plead Reynolds in any particular case because of the way it alters a trial’s focus: 12B, 16S, 20QC, 23S.

[128] 1S.

[129] 8B.

[130] 3S, 5QC; similar 2S, 8B, 14S, 16S, 19S, 22S.

[131] 1S, 4QC, 17S, 21QC, 22S.

[132] 4QC; similar 1S.

[133] 9B; similar 7QC, 10B, 20QC.

[134] 1S, 3S, 9B.

[135] 2S, 4QC, 5QC, 6B, 7QC, 10B, 11S, 13QC, 15QC, 16S, 20QC, 21QC, 22S.

[136] 5QC. Others commented on the power the approach gives judges: 8B, 14S, 18S, 23S.

[137] 11S.

[138] 20QC.

[139] 7QC.

[140] 9B; similar 17S.

[141] 11S.

[142] 2S, 11S.

[143] 21QC; similar 11S.

[144] 1S, 3S, 4QC, 5QC, 7QC, 8B, 12B, 16S, 17S, 19S, 22S.

[145] 2S, 7QC, 8B, 12B, 14S, 16S, 17S, 19S, 22S.

[146] 5QC.

[147] 16S. The courts’ attitude to urgency was also raised as a matter needing clarification: 10B, 12B, 14S, 17S.

[148] Attempting contact is necessary if not sufficient: 7QC. Not attempting to make contact ‘is such a moronic failure, you’re almost holed below the water line’: 12B.

[149] 21QC.

[150] 22S; similar 7QC, 16S, 19S.

[151] 2S (due to the tone factor), 8B, 16S. 7QC also noted the reportage line of cases. ‘Reportage’ refers to Al-Fagih and related decisions: see above nn 55–8 and accompanying text. Other interviewees saw the decision in Al-Fagih as ‘altogether odd’ (1S; similar 9B), or at least a distinct development beyond what would have been expected with Reynolds (7QC, 8B); cf 21QC, who strongly endorsed the development.

[152] 3S, 16S, 17S, 19S, 22S.

[153] 1S, 3S, 4QC, 5QC, 7QC, 8B, 12B, 16S, 17S, 19S, 22S.

[154] 8B.

[155] 12B; similar 16S.

[156] 1S, 2S, 3S, 4QC, 8B, 11S, 14S, 22S. While solicitors and in-house media lawyers reported that pre-publication practices were changing, and that uncertainty was not a problem at that stage, an experienced barrister thought difficulties in predicting how a judge might apply the Reynolds factors would also exist at the pre-publication stage.

[157] 4QC, 6B, 7QC, 21QC.

[158] 19S. Similarly, 2S noted claimants are able to judge the content of publications, if not any underlying investigation, which is important in assessing their position.

[159] 11S; similar 1S, 4QC, 10B, 20QC, 21QC.

[160] 21QC.

[161] 1S, 2S, 4QC.

[162] 21QC.

[163] 2S (there is a ‘chilling effect’ on claimants), 7QC, 12B, 15QC, 21QC.

[164] 6B.

[165] 21QC; similar 7QC.

[166] 7QC; similar 6B, 8B, 10B.

[167] 11S; similar 2S, 19S, 22S.

[168] 6B; similar 11S.

[169] 10B, 11S.

[170] 3S, 5QC, 6B, 7QC, 12B, 17S, 20QC, 21QC, 23S.

[171] 12B; similar 17S, 23S.

[172] 3S, 5QC, 6B.

[173] 20QC; similar 5QC, 7QC, 21QC.

[174] 20QC.

[175] 14S, 17S.

[176] 16S.

[177] 12B.

[178] 4QC, 6B, 8B, 9B, 11S, 12B, 13QC, 16S, 20QC, 21QC, 22S.

[179] 4QC, 6B, 8B, 9B, 11S, 13QC, 22S.

[180] 11S.

[181] 22S.

[182] 13QC.

[183] 4QC; similar 6B 12B, 15QC, 21QC.

[184] 4QC, 6B, 8B, 9B, 21QC.

[185] 8B; similar 12B who had planned to ask a particular jury 40 questions.

[186] 8B; similar 13QC.

[187] 4QC.

[188] 14S.

[189] 22S.

[190] 13QC; similar 11S, 14S, 21QC.

[191] 20QC.

[192] 13QC.

[193] 12B; similar 9B, 10B.

[194] 9B; similar 4QC, 20QC.

[195] 5QC, 6B, 9B, 21QC.

[196] 5QC.

[197] 9B; similar 12B, 16S, 22S.

[198] 20QC, 21QC.

[199] 4QC, 20QC, 21QC. See Supreme Court Act 1981 (UK) c 54, s 69 as to jury trial being usual in defamation.

[200] 1S, 8B, 18S, 22S.

[201] 13QC.

[202] 2S, 3S, 22S; similar 1S, 5QC, 9B, 14S.

[203] 21QC; similar 17S, 20QC.

[204] 24SC, 25SC, 31S, 33SC, 34S, 35S, 36S, 38S, 39S, 42B, 44S, 47SC, 50B. There were only two more supportive views: ‘I see the Lange defence as a very important defence for publishers, a very practical defence in terms of what it allows to be published’: 40S; and Lange ‘has been very positive for the media’: 41S.

[205] 33SC; similar 47SC, 50B.

[206] 50B.

[207] 34S; similar 28SC, 50B.

[208] 50B.

[209] 47SC.

[210] 37S, 47SC. It is pleaded ‘more often than not’ by the media: 43B. One barrister went further to suggest that ‘in almost all media cases involving public figures ... Lange generally has a run’: 33SC.

[211] 43B; similar 29S.

[212] [2002] VSC 174 (Unreported, Bongiorno J, 21 May 2002). See also Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003). See above nn 74–7 and accompanying text.

[213] 40S, 41S, 44S, 47SC (Popovic is ‘bizarre’), 48B, 50B.

[214] 34S, 40S, 41S, 44S, 47SC, 48B, 50B.

[215] 25SC, 28SC, 29S, 30S, 31S, 33SC, 35S, 36S, 37S, 38S, 39S, 43B, 44S, 45S, 46S, 47SC, 50B.

[216] 38S.

[217] 28SC; similar 30S: ‘People regard the reasonableness test as too high to be jumped over and frankly judges pretty much make it too high to be jumped over’.

[218] 47SC.

[219] 37S; similar 47SC.

[220] 47SC.

[221] 24SC, 25SC, 26B, 27B, 28SC, 33SC, 34S, 35S, 36S, 37S, 38S, 39S, 48B.

[222] See, eg, Walker, ‘Lange v ABC’, above n 94; New South Wales Attorney-General’s Task Force on Defamation Law Reform, above n 104, 219.

[223] 25SC, 26B, 34S, 35S, 36S, 38S, 43B.

[224] 43B; similar 34S.

[225] 29S, 32S. Others expressed far less confident hopes for such a change: 31S, 39S.

[226] 45S.

[227] 32S, 35S, 40S, 44S.

[228] 46S; similar 31S, 32S, 41S, 45S, 47SC.

[229] 28SC, 34S, 40S, 45S.

[230] 42B, 49SC.

[231] 32S.

[232] 35S, 43B, 44S, 49SC, 50B.

[233] 35S.

[234] 42B.

[235] 38S.

[236] 31S, 35S, 37S, 41S, 42B, 43B, 44S, 47SC, 49SC.

[237] 41S, 42B, 45S.

[238] 37S.

[239] 49SC.

[240] 35S; similar 42B, 43B, 44S.

[241] 30S. Further, ‘the media loves Lange [and is] obsessed with it’: 42B.

[242] See now Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003).

[243] See, eg, Popovic v Herald & Weekly Times Ltd [2002] VSC 174 (Unreported, Bongiorno J, 21 May 2002).

[244] 40S, 43B, 44S, 47SC, 49SC, 50B. It was contrasted by some interviewees with New South Wales’ s 22 defence where reasonableness is an issue for the judge: eg, 47SC.

[245] 44S, 47SC, 48B, 50B.

[246] 49SC, 50B.

[247] 50B.

[248] 49SC.

[249] 42B.

[250] 39S.

[251] 30S.

[252] 29S, 32S, 42B, 45S (who could see the arguments extending to s 22).

[253] Eg, 25SC, 28SC, 29S, 30S, 31S, 34S, 35S, 36S, 39S, 40S, 41S, 42B, 43B, 44S, 45S, 47SC, 48B, 50B. Only four interviewees offered any notable support for the current law on privilege: 24SC, 26B and 27B approved of s 22, while 49SC suggested that Lange struck a happier balance between parties than the common law qualified privilege defence under Theophanous.

[254] 35S.

[255] 28SC, 30S, 35S, 36S, 41S, 42B, 50B. One solicitor suggested that ‘the outcome is pretty much the same’ under both defences: 45S.

[256] 42B.

[257] 36S.

[258] 50B.

[259] 28SC; similar 30S.

[260] 35S.

[261] 35S; similar 45S. Another participant suggested that ‘the High Court [is] champing at the bit to have another go at it’: 36S.

[262] 42B.

[263] 29S, 35S, 36S, 39S, 44S, 45S, 47SC, 50B. Some interviewees, however, did not support such a ‘massive change’: 24SC; similar 49SC.

[264] 47SC.

[265] 25SC, 31S, 42B, 43B, 49SC.

[266] Baker, ‘Extending Common Law Qualified Privilege to the Media’, above n 46, 88.

[267] See Chesterman, above n 15, 104.

[268] See above n 249 and accompanying text.

[269] See above n 205 and accompanying text.

[270] (2001) 208 CLR 199, 333 (Callinan J), cf 287 (Kirby J). Callinan J made this comment in relation to both Theophanous and Lange. While research does not support this characterisation in relation to Lange, it offers some support for the possibility that the Theophanous duty-interest defence had a meaningful impact on defamation law. That impact, however, would appear to be more a matter of ‘measured reform’ than the detonation of any bomb.

[271] See above nn 149–52 and accompanying text.

[272] 3S, 8B, 14S, 16S, 17S, 22S, 23S. 5QC commented that Reynolds itself suggested a ‘transformed’ judicial attitude to tabloids, although later cases have seen a ‘slipping back’ against tabloid styles of presentation.

[273] 14S.

[274] [2001] EWCA 33; [2001] 2 All ER 437 (Court of Appeal). See above n 47 and accompanying text.

[275] Sarah Thomas, ‘Red Devils’, The Guardian (London), 4 November 2002, Media Guardian 10. Others, however, supported the decision: see, eg, Paul Robertshaw, ‘The Review Roles of the Court of Appeal: Grobbelaar v News International(2001) 64 Modern Law Review 923, 927: ‘This detailed cogitation by the Court of Appeal of the publication and its context ... demonstrate[s] not only judicious balance, but a sound grasp of the way in which contemporary tabloid journalism is produced and published.’

[276] See above nn 55–8, 150–6 and accompanying text.

[277] Tone, however, was expressly considered in Rowan v Cornwall [No 5] [2002] SASC 160; (2002) 82 SASR 152. In addition, the way in which Lange qualified privilege may protect comment is an underdeveloped area in Australia: see Roy Baker, ‘Defamatory Comment on the Judiciary: Lange Qualified Privilege in Popovic v Herald & Weekly Times(2002) 7 Media & Arts Law Review 213.

[278] See above nn 178–93 and accompanying text.

[279] See above nn 244–8 and accompanying text.

[280] Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003).

[281] Attorney-General’s Department, Commonwealth of Australia, Outline of Possible National Defamation Law (2004) 5.

[282] See Kenyon, Word Games, above n 105.

[283] Victorian trial courts remain bound by Popovic [2003] VSCA 161 (Unreported, Winneke ACJ, Gillard and Warren AJJA, 21 November 2003).

[284] See above nn 5962 and accompanying text.

[285] See, eg, Jameel [2004] EMLR 196, 199. See also above nn 63–4 and accompanying text.

[286] See above n 190 and accompanying text.

[287] New South Wales Attorney-General’s Task Force on Defamation Law Reform, above n 104, 28.

[288] Ibid 29.

[289] Defamation Act 1974 (NSW) s 22(2A) provides that:

In determining ... whether the conduct of the publisher ... is reasonable in the circumstances, a court may take into account the following matters and such other matters as the court considers relevant:

[290] 25SC, 28SC, 33SC, 35S, 37S, 39S (‘unless judges decide to interpret [reasonableness differently]’).

[291] 35S.

[292] 33SC.

[293] 37S.

[294] Defamation Act 1974 (NSW) s 22(2A)(d).

[295] See above nn 55–8 and accompanying text.

[296] Defamation Law Reform Committee, Western Australian Defamation Law: Committee Report on Reform to the Law of Defamation in Western Australia (2003) 18.

[297] Attorney-General’s Department, above n 281.

[298] The Attorney-General’s discussion paper takes this approach primarily from the Australian Law Reform Commission, Unfair Publication: Defamation and Privacy, Report No 11 (1979), which predated the widespread recognition of failings in New South Wales’ s 22 defence: ibid 74–5.

[299] See above nn 50–8 and accompanying text.

[300] See, eg, above nn 213–14 and accompanying text.

[301] See, eg, Chesterman, above n 15, 99; Baker, ‘Extending Common Law Qualified Privilege to the Media’, above n 46; Kenyon, ‘Defamation and Critique’, above n 83.

[302] [2001] 2 AC 127, 220. This quotation has some currency in New Zealand at least: Geoff McLay, ‘Lange v Atkinson: Not a Case for Dancing in the Streets’ [2000] New Zealand Law Review 427, 435–6; John Burrows, ‘Lange v Atkinson 2000: Analysis’ [2000] New Zealand Law Review 389, 392.

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

[304] See, eg, Marsden [2002] NSWCA 419 (Unreported, Beazley, Giles and Santow JJA, 24 December 2002) [1168]–[1170].

[305] See above n 215 and accompanying text.

[306] [2003] HCA 52; (2003) 201 ALR 184.

[307] Ibid 191–2.

[308] See, eg, Timothy Marjoribanks and Andrew T Kenyon, ‘Negotiating News: Journalistic Practice and Defamation Law in Australia and the US’ (2003) 25 Australian Journalism Review 31.

[309] Roger Magnusson, ‘Freedom of Speech in Australian Defamation Law: Ridicule, Satire and Other Challenges’ (2001) 9 Torts Law Journal 269, 297.

[310] Barendt et al, above n 2, 77.

[311] The New South Wales Attorney-General’s Task Force on Defamation Law Reform, above n 104, 31–2 offers one illustration that such changes removing the reasonableness requirement may be difficult to achieve. Two of the Task Force’s four members recommended introducing a statutory privilege, subject only to malice, for broadly ‘political’ material, but the recommendation did not appear in the subsequent Bill.

[312] See, eg, Sallie Spilsbury, ‘Bloodhounds and Watchdogs — Qualified Privilege, Malice and the Publication of Material in the Public Interest’ (2000) 11 Entertainment Law Review 43, 46–7, who suggests that judges should determine privileged occasions according to whether publications have public interest content, and that juries should deal with the circumstances of publication in terms of malice.

[313] Reynolds [2001] 2 AC 127, 205 (Lord Nicholls).


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2004/13.html