United Kingdom House of Lords Decisions

You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Reynolds v. Times Newspapers Ltd and Others [1999] UKHL 45; [1999] 4 All ER 609; [1999] 3 WLR 1010 (28th October, 1999)
URL: http://www.bailii.org/uk/cases/UKHL/1999/45.html
Cite as: [2000] HRLR 134, [1999] UKHL 45, 7 BHRC 289, [1999] 3 WLR 1010, [2001] 2 AC 127, [2000] EMLR 1, [1999] 4 All ER 609

[New search] [Buy ICLR report: [1999] 3 WLR 1010] [Buy ICLR report: [2001] 2 AC 127] [Help]

JISCBAILII_CASES_TORT

Reynolds v. Times Newspapers Ltd and Others [1999] UKHL 45; [1999] 4 All ER 609; [1999] 3 WLR 1010 (28th October, 1999)

HOUSE OF LORDS

Lord Nicholls of Birkenhead Lord Steyn Lord Cooke of Thorndon
Lord Hope of Craighead Lord Hobhouse of Wood-borough

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT IN THE CAUSE

REYNOLDS
(RESPONDENT)

TIMES NEWSPAPERS LIMITED AND OTHERS
(APPELLANTS)

ON 28 OCTOBER 1999

LORD NICHOLLS OF BIRKENHEAD

My Lords,

This appeal concerns the interaction between two fundamental rights: freedom of expression and protection of reputation. The context is newspaper discussion of a matter of political importance. Stated in its simplest form, the newspaper's contention is that a libellous statement of fact made in the course of political discussion is free from liability if published in good faith. Liability arises only if the writer knew the statement was not true or if he made the statement recklessly, not caring whether it was true or false, or if he was actuated by personal spite or some other improper motive. Mr. Reynolds' contention, on the other hand, is that liability may also arise if, having regard to the source of the information and all the circumstances, it was not in the public interest for the newspaper to have published the information as it did. Under the newspaper's contention the safeguard for those who are defamed is exclusively subjective: the state of mind of the journalist. Under Mr. Reynolds' formulation, there is also an objective element of protection.

The events giving rise to these proceedings took place during a political crisis in Dublin in November 1994. The crisis culminated in the resignation of Mr. Reynolds as Taoiseach (prime minister) of Ireland and leader of the Fianna Fáil party. The reasons for Mr. Reynolds' resignation were of public significance and interest in the United Kingdom because of his personal identification with the Northern Ireland peace process. Mr. Reynolds was one of the chief architects of that process. He announced his resignation in the Dáil (the House of Representatives) of the Irish Parliament on Thursday, 17 November 1994. On the following Sunday, 20 November, the 'Sunday Times' published in its British mainland edition an article entitled 'Goodbye gombeen man.' The article was the lead item in its world news section and occupied most of one page. The article was sub-headed 'Why a fib too far proved fatal for the political career of Ireland's peacemaker and Mr. Fixit'. On the same day the Irish edition of the 'Sunday Times' contained a three page article headed 'House of Cards' concerning the fall of the Government. This article differed in a number of respects from the British mainland edition.

Mr. Reynolds took strong exception to the article in the British mainland edition. In the libel proceedings which followed, Mr. Reynolds pleaded that the sting of the article was that he had deliberately and dishonestly misled the Dáil on Tuesday, 15 November 1994 by suppressing vital information. Further, that he had deliberately and dishonestly misled his coalition cabinet colleagues, especially Mr. Spring, the Tanaiste (deputy prime minister) and minister for foreign affairs, by withholding this information and had lied to them about when the information had come into his possession. The author of the article was Mr. Ruddock, the newspaper's Irish editor. Times Newspapers Ltd. was the publisher of the newspaper, and Mr. Witherow was the editor. They were defendants in the proceedings. The background facts are further elaborated in the judgment of the Court of Appeal, reported at [1998] 3 W.L.R. 862, 869-873. It was common ground before your Lordships that by instituting and prosecuting his libel action Mr. Reynolds had waived his immunity under the Irish constitution in respect of proceedings in the Dáil. His ability to do so was not questioned in your Lordships' House.

The action was tried by French J. and a jury between 14 October and 19 November 1996. The issues at the trial were: the meaning of the article, qualified privilege at common law, justification, malice and damages. During the trial the defendants abandoned pleaded defences that the words were fair comment on a matter of public interest and that they were a fair and accurate report of proceedings in public of the Irish legislature.

The jury verdict took the form of answers to questions. The jury decided that the defamatory allegation of which Mr. Reynolds complained was not true. So the defence of justification failed. The jury decided that Mr. Ruddock was not acting maliciously in writing and publishing the words complained of, nor was Mr. Witherow. So, if the occasion was privileged, and that was a question for the judge, the defence of qualified privilege would succeed. Despite their rejection of the defence of justification, the jury awarded Mr. Reynolds no damages. The judge substituted an award of one penny. In the light of this nil award, costs were the only remaining issue. On this the defence of qualified privilege was still a live question. If this defence was available to the defendants, they had a complete defence to the action, and the judge would have ordered Mr. Reynolds to pay the defendants' costs of the action. The judge then heard submissions on the question of qualified privilege. The defendants unsuccessfully contended for a wide qualified privilege at common law for 'political speech'. The judge ruled that publication of the article was not privileged.

Mr. Reynolds appealed, contending that the judge had misdirected the jury in certain respects. The defendants cross-appealed against the judge's decision on the qualified privilege point. The Court of Appeal, comprising Lord Bingham of Cornhill C.J., Hirst L.J. and Robert Walker L.J., allowed Mr. Reynolds' appeal. They concluded, with regret because of the consequences for the parties, that the misdirections identified by the court were, cumulatively, such as to deny Mr. Reynolds a fair trial of his claim. They set aside the verdict, finding and judgment of the court below and ordered a new trial. The Court of Appeal also considered whether the defendants would be able to rely on qualified privilege at the retrial. The court held they would not. Your Lordships' House gave leave to the defendants to appeal against this ruling, since it raised an issue of public importance. That is the issue now before your Lordships.

Defamation and truth

The defence of qualified privilege must be seen in its overall setting in the law of defamation. Historically the common law has set much store by protection of reputation. Publication of a statement adversely affecting a person's reputation is actionable. The plaintiff is not required to prove that the words are false. Nor, in the case of publication in a written or permanent form, is he required to prove he has been damaged. But, as Littledale J. said in McPherson v. Daniels (1829) 10 B. & C. 263, 272, 'the law will not permit a man to recover damages in respect of an injury to a character which he does not or ought not to possess'. Truth is a complete defence. If the defendant proves the substantial truth of the words complained of, he thereby establishes the defence of justification. With the minor exception of proceedings to which the Rehabilitation of Offenders Act 1974 applies, this defence is of universal application in civil proceedings. It avails a defendant even if he was acting spitefully.

The common law has long recognised the 'chilling' effect of this rigorous, reputation protective principle. There must be exceptions. At times people must be able to speak and write freely, uninhibited by the prospect of being sued for damages should they be mistaken or misinformed. In the wider public interest, protection of reputation must then give way to a higher priority.

Honest comment on a matter of public interest

One established exception is the defence of comment on a matter of public interest. This defence is available to everyone, and is of particular importance to the media. The freedom of expression protected by this defence has long been regarded by the common law as a basic right, long before the emergence of human rights conventions. In 1863 Crompton J. observed in Campbell v. Spottiswoode (1863) 3 B. & S. 769, 779, that 'it is the right of all the Queen's subjects to discuss public matters'. The defence is wide in its scope. Public interest has never been defined, but in London Artists Ltd. v. Littler [1969] 2 QB 375, 391, Lord Denning M.R. rightly said that it is not to be confined within narrow limits. He continued:

'Whenever a matter is such as to affect people at large, so that they may be legitimately interested in, or concerned at, what is going on; or what may happen to them or others; then it is a matter of public interest on which everyone is entitled to make fair comment.'

Traditionally one of the ingredients of this defence is that the comment must be fair, fairness being judged by the objective standard of whether any fair-minded person could honestly express the opinion in question. Judges have emphasised the latitude to be applied in interpreting this standard. So much so, that the time has come to recognise that in this context the epithet 'fair' is now meaningless and misleading. Comment must be relevant to the facts to which it is addressed. It cannot be used as a cloak for mere invective. But the basis of our public life is that the crank, the enthusiast, may say what he honestly thinks as much as the reasonable person who sits on a jury. The true test is whether the opinion, however exaggerated, obstinate or prejudiced, was honestly held by the person expressing it: see Diplock J. in Silkin v. Beaverbrook Newspapers Ltd. [1958] 1 W.L.R. 743, 747.

It is important to keep in mind that this defence is concerned with the protection of comment, not imputations of fact. If the imputation is one of fact, a ground of defence must be sought elsewhere. Further, to be within this defence the comment must be recognisable as comment, as distinct from an imputation of fact. The comment must explicitly or implicitly indicate, at least in general terms, what are the facts on which the comment is being made: see the discussion in Duncan and Neill on Defamation, 2nd ed. (1983), pp. 58-62.

One constraint does exist upon this defence. The comment must represent the honest belief of its author. If the plaintiff proves he was actuated by malice, this ground of defence will fail.

Privilege: factual inaccuracies

The defence of honest comment on a matter of public interest, then, does not cover defamatory statements of fact. But there are circumstances, in the famous words of Parke B. in Toogood v. Spyring (1834) 1 C.M. & R. 181, 193, when the 'common convenience and welfare of society' call for frank communication on questions of fact. In Davies v. Snead (1870) L.R. 5 Q.B. 608, 611, Blackburn J. spoke of circumstances where a person is so situated that it 'becomes right in the interests of society' that he should tell certain facts to another. There are occasions when the person to whom a statement is made has a special interest in learning the honestly held views of another person, even if those views are defamatory of someone else and cannot be proved to be true. When the interest is of sufficient importance to outweigh the need to protect reputation, the occasion is regarded as privileged.

Sometimes the need for uninhibited expression is of such a high order that the occasion attracts absolute privilege, as with statements made by judges or advocates or witnesses in the course of judicial proceedings. More usually, the privilege is qualified in that it can be defeated if the plaintiff proves the defendant was actuated by malice.

The classic exposition of malice in this context is that of Lord Diplock in Horrocks v. Lowe [1975] A.C. 135, 149. If the defendant used the occasion for some reason other than the reason for which the occasion was privileged he loses the privilege. Thus, the motive with which the statement was made is crucial. If desire to injure was the dominant motive the privilege is lost. Similarly, if the maker of the statement did not believe the statement to be true, or if he made the statement recklessly, without considering or caring whether it was true or not. Lord Diplock. at p. 150, emphasised that indifference to truth is not to be equated with carelessness, impulsiveness or irrationality in arriving at a positive belief that it is true:

'In ordinary life it is rare indeed for people to form their beliefs by a process of logical deduction from facts ascertained by a rigorous search for all available evidence and a judicious assessment of its probative value. In greater or in less degree according to their temperaments, their training, their intelligence, they are swayed by prejudice, rely on intuition instead of reasoning, leap to conclusions on inadequate evidence and fail to recognise the cogency of material which might cast doubt on the validity of the conclusions they reach. But despite the imperfection of the mental process by which the belief is arrived at it may still be 'honest', that is, a positive belief that the conclusions they have reached are true. The law demands no more.'

Over the years the courts have held that many common form situations are privileged. Classic instances are employment references, and complaints made or information given to the police or appropriate authorities regarding suspected crimes. The courts have always emphasised that the categories established by the authorities are not exhaustive. The list is not closed. The established categories are no more than applications, in particular circumstances, of the underlying principle of public policy. The underlying principle is conventionally stated in words to the effect that there must exist between the maker of the statement and the recipient some duty or interest in the making of the communication. Lord Atkinson's dictum, in Adam v. Ward [1917] A.C. 309, 334, is much quoted:

'. . . a privileged occasion is . . . an occasion where the person who makes a communication has an interest or a duty, legal, social, or moral, to make it to the person to whom it is made, and the person to whom it is so made has a corresponding interest or duty to receive it. This reciprocity is essential'.

The requirement that both the maker of the statement and the recipient must have an interest or duty draws attention to the need to have regard to the position of both parties when deciding whether an occasion is privileged. But this should not be allowed to obscure the rationale of the underlying public interest on which privilege is founded. The essence of this defence lies in the law's recognition of the need, in the public interest, for a particular recipient to receive frank and uninhibited communication of particular information from a particular source. That is the end the law is concerned to attain. The protection afforded to the maker of the statement is the means by which the law seeks to achieve that end. Thus the court has to assess whether, in the public interest, the publication should be protected in the absence of malice.

In determining whether an occasion is regarded as privileged the court has regard to all the circumstances: see, for example, the explicit statement of Lord Buckmaster L.C.in London Association for Protection of Trade v. Greenlands Ltd. [1916] 2 A.C. 15, 23 ('every circumstance associated with the origin and publication of the defamatory matter'). And circumstances must be viewed with today's eyes. The circumstances in which the public interest requires a communication to be protected in the absence of malice depend upon current social conditions. The requirements at the close of the twentieth century may not be the same as those of earlier centuries or earlier decades of this century.

Privilege and publication to the world at large

Frequently a privileged occasion encompasses publication to one person only or to a limited group of people. Publication more widely, to persons who lack the requisite interest in receiving the information, is not privileged. But the common law has recognised there are occasions when the public interest requires that publication to the world at large should be privileged. In Cox v. Feeney (1863) 4 F. & F. 13, 19, Cockburn C.J. approved an earlier statement by Lord Tenterden C.J. that 'a man has a right to publish, for the purpose of giving the public information, that which it is proper for the public to know'. Whether the public interest so requires depends upon an evaluation of the particular information in the circumstances of its publication. Through the cases runs the strain that, when determining whether the public at large had a right to know the particular information, the court has regard to all the circumstances. The court is concerned to assess whether the information was of sufficient value to the public that, in the public interest, it should be protected by privilege in the absence of malice.

This issue has arisen several times in the context of newspapers discharging their important function of reporting matters of public importance. Two instances will suffice, together with one instance of the publication in book form of information originating with the publisher. Purcell v. Sowler (1877) 2 C.P.D. 215 concerned a newspaper report of a public meeting of poor-law guardians. During the meeting the medical officer of the workhouse at Knutsford was said to have neglected to attend pauper patients when sent for. In deciding that publication of this allegation was not privileged, the Court of Appeal looked beyond the subject-matter. The court held that the administration of the poor-law was a matter of national concern, but there was no duty to report charges made in the absence of the medical officer and without his having had any opportunity to meet them. The meeting was a privileged occasion so far as the speaker was concerned, but publication in the press was not. In Allbutt v. General Council of Medical Education and Registration (1889) 23 QBD 400, 410, the defendants published a book containing minutes of a meeting of the council recording that the plaintiff's name had been removed from the medical register for infamous professional conduct. This was after an inquiry at which the plaintiff had been represented by counsel. The Court of Appeal held that the publication was privileged. Giving the judgment of the court, Lopes L.J. expressly had regard to the nature of the tribunal, the character of the report, the interests of the public in the proceedings of the council and the duty of the council towards the public. Perera v. Peiris [1949] AC 1, 21, was an appeal to the Privy Council from the Supreme Court of Ceylon. The 'Ceylon Daily News' had published extracts from a report of the Bribery Commission which was critical of Dr. Perera's lack of frankness in his evidence. The Judicial Committee upheld a claim to qualified privilege. In the light of the origin and contents of the report and its relevance to the affairs of Ceylon, the due administration of the affairs of Ceylon required that the report should receive the widest publicity.

The courts have recognised that the status and activities of certain bodies are such that members of the public are entitled to know of their proceedings. Then privilege derives from the subject-matter alone. Fair and accurate reports of the proceedings of these organisations are privileged. A leading instance is Wason v. Walter (1868) L.R. 4 Q.B. 73, 89, concerning newspaper reports of debates in Parliament. The Court of Queen's Bench held, by analogy with reports of judicial proceedings, that fair and accurate reports of parliamentary proceedings were privileged. Cockburn C.J. observed that it was of paramount public and national importance that the proceedings of either House of Parliament should be communicated to the public 'who have the deepest interest in knowing what passes within their walls, seeing that on what is there said and done, the welfare of the community depends'.

In Webb v. Times Publishing Co. Ltd. [1960] 2 Q.B. 535 the defendants attempted to obtain similar blanket (or 'generic') protection for another category of subject-matter: reports of foreign judicial proceedings. There 'The Times' newspaper had published a report of the criminal trial in Switzerland of a British subject. Pearson J. rejected this approach, but he upheld the claim to privilege by applying the general principle enunciated in the line of authorities exemplified by Cox v. Feeney 4 F. & F. 13, Allbutt v. General Council of Medical Education and Registration 23 Q.B.D. 400 and Perera v. Peiris [1949] AC 1.

Similarly, in Blackshaw v. Lord [1984] 1 Q.B. 1, 6 the Court of Appeal rejected a claim to generic protection for a widely stated category: 'fair information on a matter of public interest'. A claim to privilege must be more precisely focused. In order to be privileged publication must be in the public interest. Whether a publication is in the public interest or, in the conventional phraseology, whether there is a duty to publish to the intended recipients, there the readers of the 'Daily Telegraph', depends upon the circumstances, including the nature of the matter published and its source or status.

The decision of the Court of Appeal in Braddock v. Bevins [1948] 1 K.B. 580, on which the appellant newspaper placed some reliance, is consistent with this approach. The court held that Mr. Bevins' election address at a local election was the subject of qualified privilege. In reaching its conclusion the court applied the classic requirements necessary to confer qualified privilege: see the judgment of Lord Greene M.R., at pp. 589-590. This decision was reversed by section 10 of the Defamation Act 1952:

'A defamatory statement published by or on behalf of a candidate in any election to a local government authority or to Parliament shall not be deemed to be published on a privileged occasion on the ground that it is material to a question in issue in the election, whether or not the person by whom it is published is qualified to vote at the election.'

Parliament seems to have taken the view that the defence of comment on a matter of public interest provided sufficient protection for election addresses. Whether this statutory provision can withstand scrutiny under the Human Rights Act 1998 is not a matter to be pursued on this appeal. Suffice to say, Braddock v. Bevins did not place election communications into a special category.

In Derbyshire County Council v. Times Newspapers Ltd. [1993] AC 534 this House held that it was contrary to the public interest for organs of central or local government to have any right at common law to maintain an action for defamation. This is an instance, in the field of political discussion, of the court's concern to remove all unnecessary fetters on freedom of speech. Beyond that, this decision does not assist in the present appeal.

In its valuable and forward-looking analysis of the common law the Court of Appeal in the present case highlighted that in deciding whether an occasion is privileged the court considers, among other matters, the nature, status and source of the material published and the circumstances of the publication. In stressing the importance of these particular factors, the court treated them as matters going to a question ('the circumstantial test') separate from, and additional to, the conventional duty-interest questions: see [1998] 3 W.L.R. 862, 899. With all respect to the Court of Appeal, this formulation of three questions gives rise to conceptual and practical difficulties and is better avoided. There is no separate or additional question. These factors are to be taken into account in determining whether the duty-interest test is satisfied or, as I would prefer to say in a simpler and more direct way, whether the public was entitled to know the particular information. The duty-interest test, or the right to know test, cannot be carried out in isolation from these factors and without regard to them. A claim to privilege stands or falls according to whether the claim passes or fails this test. There is no further requirement.

Statutory privilege

Many, if not all, of the common law categories of case where reports of proceedings attract privilege are now the subject of statutory privilege. Successive statutes have extended the categories. The Law of Libel Amendment Act 1888 granted qualified privilege to fair and accurate reports published in newspapers of a limited range of public meetings. In 1948 the Report of the Committee on the Law of Defamation (Cmd. 7536), chaired by Lord Porter, recommended that the classes of reports subject to qualified privilege should be extended, and that they should be re-classified into two categories: those where statements were privileged without explanation or contradiction, and those where privilege was conditional on publication on request of a letter or statement by way of explanation or contradiction. The Defamation Act 1952 gave effect to these recommendations. Among the publications having qualified privilege without explanation or contradiction was a fair and accurate report of proceedings in public of the Irish legislature. Until abandoned, this was one of the defendants' pleaded defences in the present proceedings.

In 1975 the committee on defamation chaired by Faulks J. considered a proposal that a statutory qualified privilege should be created to protect statements made, whether in a newspaper or elsewhere, if the matter was of public interest and the publisher believed the statement of facts was true and he had taken reasonable care in relation to such facts. In its report (Cmnd. 5909) the committee did not accept this proposal. The committee considered this would seriously alter the balance of the law of defamation against a defamed plaintiff. The committee noted that the common law defence of qualified privilege was available to the media as much as anyone else, and referred to the Cox v. Feeney line of cases.

In 1991 the Supreme Court Procedure Committee, chaired by Neill L.J., in its Report on Practice and Procedure in Defamation considered that fair and accurate coverage by the British media of statements and proceedings abroad ought to be protected by qualified privilege in circumstances which would attract privilege if comparable statements or proceedings occurred in this country. The committee recommended this result should be achieved by statute. The committee regarded the 'duty' test as too stringent in modern conditions and productive of too much uncertainty. The committee was opposed to the introduction of a defence similar to the 'public figure' defence enunciated by the United States Supreme Court in New York Times Co. v. Sullivan (1964) 376 U.S. 254.

The Defamation Act 1996 broadly gave effect to the Neill committee recommendations. The Act contained an extended list of categories of statutory qualified privilege. In the Act of 1996 and the Act of 1952 statutory privilege was additional to any common law privilege, but did not protect publication of any matter which was not of public concern and the publication of which was not for the public benefit: see section 15 of the Act of 1996 and section 7 of the Act of 1952.

In other countries

Before turning to the issues raised by this appeal mention must be made, necessarily briefly, of the solutions adopted in certain other countries. As is to be expected, the solutions are not uniform. As also to be expected, the chosen solutions have not lacked critics in their own countries.

In the United States the leading authority is the well-known case of New York Times Co. v. Sullivan 376 U.S. 254. Founding itself on the first and fourteenth amendments to the United States Constitution, the Supreme Court held that a public official cannot recover damages for a defamatory falsehood relating to his official conduct unless he proves, with convincing clarity, that the statement was made with knowledge of its falsity or with reckless disregard of whether it was false or not. This principle has since been applied to public figures generally.

In Canada the Supreme Court, in Hill v. Church of Scientology of Toronto (1995) 126 D.L.R. (4th) 129, rejected a Sullivan style defence, although that case did not concern political discussion. The Supreme Court has not had occasion to consider this issue in relation to political discussion.

In India the Supreme Court, in Rajagopal v. State of Tamil Nadu (1994) 6 S.C.C. 632, 650, held that a public official has no remedy in damages for defamation in matters relating to his official duties unless he proves the publication was made with reckless disregard of the truth or out of personal animosity. Where malice is alleged it is sufficient for the defendant to prove he acted after a reasonable verification of the facts.

In Australia the leading case is Lange v. Australian Broadcasting Corporation (1997) 189 CLR 520. The High Court held unanimously that qualified privilege exists for the dissemination of information, opinions and arguments concerning government and political matters affecting the people of Australia, subject to the publisher proving reasonableness of conduct. The High Court regarded its decision as an extension of the categories of qualified privilege, and considered that the reasonableness requirement was appropriate having regard to the greater damage done by mass dissemination compared with the limited publication normally involved on occasions of common law qualified privilege. As a general rule a defendant's conduct in publishing material giving rise to a defamatory imputation would not be reasonable unless the defendant had reasonable grounds for believing the imputation was true, took proper steps, so far as they were reasonably open, to verify the accuracy of the material and did not believe the imputation to be untrue. Further, the defendant's conduct would not be reasonable unless the defendant sought a response from the person defamed and published the response, except where this was not practicable or was unnecessary.

In South Africa the issue has not been considered by the Constitutional Court. In National Media Ltd. v. Bogoshi 1998 (4) S.A. 1196, 1212 the Supreme Court of Appeal broadly followed the approach of the Court of Appeal in the present case and the Australian High Court in the Lange case. Press publication of defamatory statements of fact will not be regarded as unlawful if, upon consideration of all the circumstances, it is found to have been reasonable to publish the particular facts in the particular way and at the particular time. In considering the reasonableness of the publication account must be taken of the nature, extent and tone of the allegations. Greater latitude is usually to be allowed in respect of political discussion.

In New Zealand the leading case is the Court of Appeal decision in Lange v. Atkinson [1998] 3 N.Z.L.R. 424. The Court of Appeal held that members of the public have a proper interest in respect of statements made about the actions and qualities of those currently or formerly elected to Parliament and those seeking election. General publication of such statements may therefore attract a defence of qualified privilege. The exercise of reasonable care by the defendant is not a requirement of this defence. This decision is currently under appeal to the Privy Council. The Judicial Committee heard this appeal shortly before the Appellate Committee of your Lordships' House, similarly constituted, heard the parties' submissions on the present appeal.

A new category of privileged subject-matter?

I turn to the appellants' submissions. The newspaper seeks the incremental development of the common law by the creation of a new category of occasion when privilege derives from the subject-matter alone: political information. Political information can be broadly defined, borrowing the language used by the High Court of Australia in the Lange case, as information, opinion and arguments concerning government and political matters that affect the people of the United Kingdom. Malice apart, publication of political information should be privileged regardless of the status and source of the material and the circumstances of the publication. The newspaper submitted that the contrary view requires the court to assess the public interest value of a publication, taking these matters into account. Such an approach would involve an unpredictable outcome. Moreover, it would put the judge in a position which in a free society ought to be occupied by the editor. Such paternalism would effectively give the court an undesirable and invidious role as a censor or licensing body.

These are powerful arguments, but I do not accept the conclusion for which the newspaper contended. My reasons appear from what is set out below.

My starting point is freedom of expression. The high importance of freedom to impart and receive information and ideas has been stated so often and so eloquently that this point calls for no elaboration in this case. At a pragmatic level, freedom to disseminate and receive information on political matters is essential to the proper functioning of the system of parliamentary democracy cherished in this country. This freedom enables those who elect representatives to Parliament to make an informed choice, regarding individuals as well as policies, and those elected to make informed decisions. Freedom of expression will shortly be buttressed by statutory requirements. Under section 12 of the Human Rights Act 1998, expected to come into force in October 2000, the court is required, in relevant cases, to have particular regard to the importance of the right to freedom of expression. The common law is to be developed and applied in a manner consistent with article 10 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (Cmd. 8969), and the court must take into account relevant decisions of the European Court of Human Rights (sections 6 and 2). To be justified, any curtailment of freedom of expression must be convincingly established by a compelling countervailing consideration, and the means employed must be proportionate to the end sought to be achieved.

Likewise, there is no need to elaborate on the importance of the role discharged by the media in the expression and communication of information and comment on political matters. It is through the mass media that most people today obtain their information on political matters. Without freedom of expression by the media, freedom of expression would be a hollow concept. The interest of a democratic society in ensuring a free press weighs heavily in the balance in deciding whether any curtailment of this freedom bears a reasonable relationship to the purpose of the curtailment. In this regard it should be kept in mind that one of the contemporary functions of the media is investigative journalism. This activity, as much as the traditional activities of reporting and commenting, is part of the vital role of the press and the media generally.

Reputation is an integral and important part of the dignity of the individual. It also forms the basis of many decisions in a democratic society which are fundamental to its well-being: whom to employ or work for, whom to promote, whom to do business with or to vote for. Once besmirched by an unfounded allegation in a national newspaper, a reputation can be damaged for ever, especially if there is no opportunity to vindicate one's reputation. When this happens, society as well as the individual is the loser. For it should not be supposed that protection of reputation is a matter of importance only to the affected individual and his family. Protection of reputation is conducive to the public good. It is in the public interest that the reputation of public figures should not be debased falsely. In the political field, in order to make an informed choice, the electorate needs to be able to identify the good as well as the bad. Consistently with these considerations, human rights conventions recognise that freedom of expression is not an absolute right. Its exercise may be subject to such restrictions as are prescribed by law and are necessary in a democratic society for the protection of the reputations of others.

The crux of this appeal, therefore, lies in identifying the restrictions which are fairly and reasonably necessary for the protection of reputation. Leaving aside the exceptional cases which attract absolute privilege, the common law denies protection to defamatory statements, whether of comment or fact, proved to be actuated by malice, in the Horrocks v. Lowe [1975] A.C. 135 sense. This common law limitation on freedom of speech passes the 'necessary' test with flying colours. This is an acceptable limitation. Freedom of speech does not embrace freedom to make defamatory statements out of personal spite or without having a positive belief in their truth.

In the case of statements of opinion on matters of public interest, that is the limit of what is necessary for protection of reputation. Readers and viewers and listeners can make up their own minds on whether they agree or disagree with defamatory statements which are recognisable as comment and which, expressly or implicitly, indicate in general terms the facts on which they are based.

With defamatory imputations of fact the position is different and more difficult. Those who read or hear such allegations are unlikely to have any means of knowing whether they are true or not. In respect of such imputations, a plaintiff's ability to obtain a remedy if he can prove malice is not normally a sufficient safeguard. Malice is notoriously difficult to prove. If a newspaper is understandably unwilling to disclose its sources, a plaintiff can be deprived of the material necessary to prove, or even allege, that the newspaper acted recklessly in publishing as it did without further verification. Thus, in the absence of any additional safeguard for reputation, a newspaper, anxious to be first with a 'scoop', would in practice be free to publish seriously defamatory misstatements of fact based on the slenderest of materials. Unless the paper chose later to withdraw the allegations, the politician thus defamed would have no means of clearing his name, and the public would have no means of knowing where the truth lay. Some further protection for reputation is needed if this can be achieved without a disproportionate incursion into freedom of expression.

This is a difficult problem. No answer is perfect. Every solution has its own advantages and disadvantages. Depending on local conditions, such as legal procedures and the traditions and power of the press, the solution preferred in one country may not be best suited to another country. The appellant newspaper commends reliance upon the ethics of professional journalism. The decision should be left to the editor of the newspaper. Unfortunately, in the United Kingdom this would not generally be thought to provide a sufficient safeguard. In saying this I am not referring to mistaken decisions. From time to time mistakes are bound to occur, even in the best regulated circles.. Making every allowance for this, the sad reality is that the overall handling of these matters by the national press, with its own commercial interests to serve, does not always command general confidence.

As high-lighted by the Court of Appeal judgment in the present case, the common law solution is for the court to have regard to all the circumstances when deciding whether the publication of particular material was privileged because of its value to the public. Its value to the public depends upon its quality as well as its subject-matter. This solution has the merit of elasticity. As observed by the Court of Appeal, this principle can be applied appropriately to the particular circumstances of individual cases in their infinite variety. It can be applied appropriately to all information published by a newspaper, whatever its source or origin.

Hand in hand with this advantage goes the disadvantage of an element of unpredictability and uncertainty. The outcome of a court decision, it was suggested, cannot always be predicted with certainty when the newspaper is deciding whether to publish a story. To an extent this is a valid criticism. A degree of uncertainty in borderline cases is inevitable. This uncertainty, coupled with the expense of court proceedings, may 'chill' the publication of true statements of fact as well as those which are untrue. The chill factor is perhaps felt more keenly by the regional press, book publishers and broadcasters than the national press. However, the extent of this uncertainty should not be exaggerated. With the enunciation of some guidelines by the court, any practical problems should be manageable. The common law does not seek to set a higher standard than that of responsible journalism, a standard the media themselves espouse. An incursion into press freedom which goes no further than this would not seem to be excessive or disproportionate. The investigative journalist has adequate protection. The contrary approach, which would involve no objective check on the media, drew a pertinent comment from Tipping J. in Lange v. Atkinson [1998] 3 N.Z.L.R. 424, 477:

'It could be seen as rather ironical that whereas almost all sectors of society, and all other occupations and professions have duties to take reasonable care, and are accountable in one form or another if they are careless, the news media whose power and capacity to cause harm and distress are considerable if that power is not responsibly used, are not liable in negligence, and what is more, can claim qualified privilege even if they are negligent. It may be asked whether the public interest in freedom of expression is so great that the accountability which society requires of others, should not also to this extent be required of the news media.'

The common law approach does mean that it is an outside body, that is, some one other than the newspaper itself, which decides whether an occasion is privileged. This is bound to be so, if the decision of the press itself is not to be determinative of the propriety of publishing the particular material. The court has the advantage of being impartial, independent of government, and accustomed to deciding disputed issues of fact and whether an occasion is privileged. No one has suggested that some other institution would be better suited for this task.

For the newspaper, Lord Lester's fall-back position was that qualified privilege should be available for political discussion unless the plaintiff proved the newspaper failed to exercise reasonable care. One difficulty with this suggestion is that it would seem to leave a newspaper open to publish a serious allegation which it had been wholly unable to verify. Depending on the circumstances, that might be most unsatisfactory. This difficulty would be removed if, as also canvassed by Lord Lester, the suggested limitation was stated more broadly, and qualified privilege was excluded if the plaintiff proved that the newspaper's conduct in making the publication was unreasonable. Whether this test would differ substantially from the common law test is a moot point. There seems to be no significant practical difference between looking at all the circumstances to decide if a publication attracts privilege, and looking at all the circumstances to see if an acknowledged privilege is defeated.

I have been more troubled by Lord Lester's suggested shift in the burden of proof. Placing the burden of proof on the plaintiff would be a reminder that the starting point today is freedom of expression and limitations on this freedom are exceptions. That has attraction. But if this shift of the onus were applied generally, it would turn the law of qualified privilege upside down. The repercussions of such a far-reaching change were not canvassed before your Lordships. If this change were applied only to political information, the distinction would lack a coherent rationale. There are other subjects of serious public concern. On balance I favour leaving the onus in its traditional place, on him who asserts the privilege, for two practical reasons. A newspaper will know much more of the facts leading up to publication. The burden of proof will seldom, if ever, be decisive on this issue.

For Mr. Reynolds, Mr. Caldecott submitted that in the context of political speech a report which 'failed to report the other side' should always fail the common law test and, further, that there should be a burden on the newspaper to establish a cogent reason why it should be excused from proving the truth of the assertion. I cannot accept either of these suggested requirements. Failure to report the plaintiff's explanation is a factor to be taken into account. Depending upon the circumstances, it may be a weighty factor. But it should not be elevated into a rigid rule of law. As to the second requirement, it is not clear to what extent, and in what respects, this suggestion covers ground different from the ground already covered by the common law principle.

Human rights jurisprudence

The common law approach accords with the present state of the human rights jurisprudence. The immensely influential judgment in Lingens v. Austria (1986) 8 EHRR 407 concerned expressions of opinion, not statements of fact. Mr. Lingens was fined for publishing in his magazine in Vienna comments about the behaviour of the Federal Chancellor, Mr. Kreisky: 'basest opportunism', 'immoral' and 'undignified'. Under the Austrian criminal code the only defence was proof of the truth of these statements. Mr. Lingens could not prove the truth of these value judgments, because Mr. Kreisky's behaviour was capable of more than one interpretation. In a passage, often overlooked, at pp. 420-1, in para. 46 of its judgment, the European Court of Human Rights stated that a careful distinction needs to be made between facts and value judgments. The existence of facts can be demonstrated, whereas the truth of value judgments is not susceptible of proof. The facts on which Mr. Lingens founded his value judgments were undisputed, as was his good faith. Since it was impossible to prove the truth of value judgments, the requirement of the relevant provisions of the Austrian criminal code was impossible of fulfilment and infringed article 10 of the Convention. The court has subsequently reiterated the distinction between facts and value judgments in De Haes and Gijsels v. Belgium (1997) 25 E.H.R.R. 1, 54 at para. 42.

In Fressoz and Roire v. France (unreported), 21 January 1999, Case No. 29183/95, paragraph 54, the court adverted to the need for accuracy on matters of fact. Article 10 protects the right of journalists to divulge information on issues of general interest provided they are acting in good faith and on 'an accurate factual basis' and supply reliable and precise information in accordance with the ethics of journalism. But a journalist is not required to guarantee the accuracy of his facts. Bladet Tromso and Stensaas v. Norway (unreported), 20 May 1999, Case No. 21980/93 involved newspaper allegations of fact: cruelty by seal hunters. The Court of Human Rights considered whether the newspaper had a reasonable basis for its factual allegations. Similarly, in Thorgeirson v. Iceland (1992) 14 EHRR 843 two newspaper articles reported widespread rumours of brutality by the Reykjavik police. These rumours had some substantiation in fact: a policeman had been convicted recently. The purpose of the articles was to promote an investigation by an independent body. The court held that although the articles were framed in particularly strong terms, they bore on a matter of serious public concern. It was unreasonable to require the writer to prove that unspecified members of the Reykjavik police force had committed acts of serious assault resulting in disablement.

None of these three latter cases involved political discussion, but for this purpose no distinction is to be drawn between political discussion and discussion of other matters of public concern: see the Thorgeirson case, at pp. 863-4, 865 para. 61, 64.

Conclusion

My conclusion is that the established common law approach to misstatements of fact remains essentially sound. The common law should not develop 'political information' as a new 'subject-matter' category of qualified privilege, whereby the publication of all such information would attract qualified privilege, whatever the circumstances. That would not provide adequate protection for reputation. Moreover, it would be unsound in principle to distinguish political discussion from discussion of other matters of serious public concern. The elasticity of the common law principle enables interference with freedom of speech to be confined to what is necessary in the circumstances of the case. This elasticity enables the court to give appropriate weight, in today's conditions, to the importance of freedom of expression by the media on all matters of public concern.

Depending on the circumstances, the matters to be taken into account include the following. The comments are illustrative only.

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 for their stories.

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. He may have information others do not possess or have not disclosed. 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.

This list is not exhaustive. The weight to be given to these and any other relevant factors will vary from case to case. Any disputes of primary fact will be a matter for the jury, if there is one. The decision on whether, having regard to the admitted or proved facts, the publication was subject to qualified privilege is a matter for the judge. This is the established practice and seems sound. A balancing operation is better carried out by a judge in a reasoned judgment than by a jury. Over time, a valuable corpus of case law will be built up.

In general, a newspaper's unwillingness to disclose the identity of its sources should not weigh against it. Further, it should always be remembered that journalists act without the benefit of the clear light of hindsight. Matters which are obvious in retrospect may have been far from clear in the heat of the moment. Above all, the court should have particular regard to the importance of freedom of expression. The press discharges vital functions as a bloodhound as well as a watchdog. The court should be slow to conclude that a publication was not in the public interest and, therefore, the public had no right to know, especially when the information is in the field of political discussion. Any lingering doubts should be resolved in favour of publication.

Privilege and the facts of this case

The appellant newspaper's primary submission was that they never had the opportunity of pleading and proving a case that the 'circumstantial test' was satisfied, because this test had not been formulated until the Court of Appeal gave judgment. I am not persuaded by this line of argument. Mr. Reynolds' case before the judge was that all the circumstances had to be taken into account. He specifically relied on the gravity of the charge, the presentation of lying as an allegation of fact and not as an opinion or value judgment, the omission of Mr. Reynolds' defence as given by him in the Dail debate on Wednesday, 16 November 1994, and the difference between the versions in the mainland and Irish editions. In the exercise of its discretion the Court of Appeal decided to rule on the issue of qualified privilege, rather than leave this matter to be dealt with by the trial judge at the re-trial.

I can see no sufficient ground for interfering with that decision. Further, despite the defendants' criticisms of some of the grounds set out by the Court of Appeal at [1998] 3 W.L.R. 862, 911-912, the facts relied upon by Mr. Reynolds before the judge were clear and undisputed. A most telling criticism of the article is the failure to mention Mr. Reynolds' own explanation to the Dáil. Mr. Ruddock omitted this from the article because he rejected Mr. Reynolds' version of the events and concluded that Mr. Reynolds had been deliberately misleading. It goes without saying that a journalist is entitled and bound to reach his own conclusions and to express them honestly and fearlessly. He is entitled to disbelieve and refute explanations given. But this cannot be a good reason for omitting, from a hard-hitting article making serious allegations against a named individual, all mention of that person's own explanation. Particularly so, when the press offices had told Mr. Ruddock that Mr. Reynolds was not giving interviews but would be saying all he had to say in the Dáil. His statement in the Dáil was his answer to the allegations. An article omitting all reference to this statement could not be a fair and accurate report of proceedings in the Dáil. Such an article would be misleading as a report. This article is not defended as a report, but it was misleading nonetheless. By omitting Mr. Reynolds' explanation English readers were left to suppose that, so far, Mr. Reynolds had offered no explanation. Further, it is elementary fairness that, in the normal course, a serious charge should be accompanied by the gist of any explanation already given. An article which fails to do so faces an uphill task in claiming privilege if the allegation proves to be false and the unreported explanation proves to be true.

Was the information in the 'Sunday Times' article information the public was entitled to know? The subject matter was undoubtedly of public concern in this country. However, these serious allegations by the newspaper, presented as statements of fact but shorn of all mention of Mr. Reynolds' considered explanation, were not information the public had a right to know. I agree with the Court of Appeal this was not a publication which should in the public interest be protected by privilege in the absence of proof of malice. The further facts the defendants wish to assert and prove at the retrial would make no difference, either on this point or overall. I would dismiss this appeal.

LORD STEYN

My Lords,

I gratefully adopt the account of the background given by Lord Bingham of Cornhill, C.J., in sections I, II, and III of the judgment of the Court of Appeal (reported at [1998] 3 W.L.R. 862, 868H-876F), as well as the summary given by my noble and learned friend Lord Nicholls of Birkenhead. I therefore turn directly to the central issues.

The New Landscape

Important issues regarding the reconciliation of the colliding right of free speech and the right to reputation need to be considered in the light of the new legal landscape. In what was at the time regarded as a classic direction on fair comment to the jury Diplock J. in Silkin v. Beaverbrook Newspapers Ltd. [1958] 1. W.L.R. 743, 746 observed:

"In the first place, every man, whether he is in public life or not, is entitled not to have lies told about him; and by that is meant that one is not entitled to make statements of fact about a person which are untrue and which redound to his discredit, that is to say, tend to lower him in the estimation of right-thinking men."

The present case involves a defamatory and factually false statement which the newspaper honestly believed to be true. If the observation of Diplock J. is taken not only as the starting point but as reflecting an absolute rule, there would be no room for any qualified privilege in respect of political speech. But the law has not stood still. In Attorney-General v. Guardian Newspapers Ltd. (No. 2) [1990] 1 AC 109, 283-4, Lord Goff of Chieveley observed that there was in principle no difference between article 10 of the European Convention of Human Rights and the English law of confidence. Article 10 is in the following terms:

"1. Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers . . . 2. The exercise of these freedoms, since it carriers with it duties and responsibilities, may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of national security, territorial integrity or public safety, for the prevention of disorder or crime, for the protection of health or morals, for the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence, or for maintaining the authority and impartiality of the judiciary."

In Derbyshire County Council v. Times Newspaper Ltd. [1993] AC 534, 551G, Lord Keith of Kinkel, speaking for a unanimous House, endorsed in a carefully considered passage Lord Goff's observations in the context of article 10 of the Convention and the law of defamation.

It is worth considering why Lord Goff and Lord Keith could so confidently assert that the law of England and article 10 of the Convention is the same. In my judgment the reasons are twofold. First, there is the principle of liberty. Whatever is not specifically forbidden by law individuals and their enterprises are free to do: see Lord Goff, at p. 283G, where he stated that in England "everybody is free to do anything, subject only to the provisions of the law." By contrast the executive and judicial branches of government may only do what the law specifically permits. Secondly, there is a constitutional right to freedom of expression in England: see Broome v. Cassell & Co. Ltd. [1972] AC 1027, 1133 A-B per Lord Kilbrandon. By categorising this basic and fundamental right as a constitutional right its higher normative force is emphasised. These are perhaps some of the considerations which enabled Lord Goff in 1988 and Lord Keith in 1993 to hold that article 10 of the Convention and the English law on the point are in material respects the same. Now the Human Rights Act 1998, which will corporate the Convention into our legal order, is on the statute book. And the government has announced that it will come into force on 2 October 2000. The constitutional dimension of freedom of expression is reinforced. This is the backcloth against which the present appeal must be considered. It is common ground that in considering the issues before the House, and the development of English law, the House can and should act on the reality that the Human Rights Act 1998 will soon be in force.

The new landscape is of great importance inasmuch as it provides the taxonomy against which the question before the House must be considered. The starting point is now the right of freedom of expression, a right based on a constitutional or higher legal order foundation. Exceptions to freedom of expression must be justified as being necessary in a democracy. In other words, freedom of expression is the rule and regulation of speech is the exception requiring justification. The existence and width of any exception can only be justified if it is underpinned by a pressing social need. These are fundamental principles governing the balance to be struck between freedom of expression and defamation.

The Issues

The issues to decide are as follows: (1) Is there a generic qualified privilege extending to publication by a newspaper to the public at large of information including assertions of fact concerning government and political matters which affect the people of the United Kingdom? If there is such a generic qualified privilege, the appeal must succeed. If the answer is in the negative, further issues arise. (2) After stating the traditional issues of duty and interest applicable to qualified privilege, the Court of Appeal enunciated what it described as "a circumstantial test." The second issue is whether that test is correct in law. (3) If neither the generic test nor the circumstantial test is correct, what is the applicable law regarding qualified privilege in respect of political speech containing a defamatory and factually false statement which was honestly believed to be true? Under this heading the requirements and conditions applicable to such a qualified privilege (if any) are in dispute. (4) Whatever test is laid down, what are the respective functions of judge and jury? (5) Depending on the way the issues of law are resolved, should the decision of the Court of Appeal be affirmed or should it be quashed? (6) What order should be made?

Issue (1): Generic qualified privilege and political speech.

Counsel for the newspaper did not invite your Lordships to develop English law in line with the landmark case of New York Times Co. v. Sullivan (1964) 376 U.S. 254. The United States Supreme Court unanimously held that a public official could not succeed in an action for libel without proving that the defendant was actuated by actual malice, that is, at least with a reckless disregard of the truth. The question was whether a particular advertisement forfeited constitutional protection by reason of the falsity of some of the factual statements and the alleged defamation of a public official. The Supreme Court declared the relevant state law unconstitutional. In the present case counsel for the newspaper cited passages from the classic judgment of Brennan J. in the Sullivan case about the chilling effect on freedom of speech of too broad a defamation law. Perhaps for present purposes the most important passage is the following (at pp. 278-279):

"The state rule of law is not saved by its allowance of the defence of truth. . . . A rule compelling the critic of official conduct to guarantee the truth of all his factual assertions--and to do so on pain of libel judgments virtually unlimited in amount--leads to a comparable 'self-censorship.' Allowance of the defence of truth, with the burden of proving it on the defendant, does not mean that only false speech will be deterred. . . . Under such a rule, would-be critics of official conduct may be deterred from voicing their criticism, even though it is believed to be true and even though it is in fact true, because of doubt whether it can be proved in court or fear of the expense of having to do so. They tend to make only statements which 'steer far wider of the unlawful zone.' . . . The rule thus dampens the vigour and limits the variety of public debate. It is inconsistent with the First and Fourteenth Amendments."

Given the limited way in which counsel used the Sullivan case I need not explore the subsequent development of the doctrine in Gertz v. Robert Welch, Inc. (1974) 418 U.S. 323 and other cases.

Counsel submitted that the House should recognise a qualified privilege extending to the publication by a newspaper to the public at large of factual information, opinions and arguments concerning government and political matters that affect the people of the United Kingdom. For convenience, I will call this a generic qualified privilege of political speech. A distinctive feature of political speech published by a newspaper is that it is communicated to a large audience. And this characteristic must be kept in mind in weighing the arguments in the present case. It is further essential not to lose sight of the factual framework in which the question arises, namely a defamatory and factually incorrect statement which the newspaper believed to be true.

It is now necessary to explain what is meant by a generic qualified privilege. It is to be contrasted with each case being considered in the light of its own particular circumstances, that is, in an ad hoc manner, in the light of the concrete facts of the case, and balancing in each case the gravity of the damage to the plaintiff's reputation against the value of publication on the particular occasion. A generic privilege, on the other hand, uses the technique of applying the privilege to a category or categories of cases. An example is the rule in the Sullivan case, which requires proof of malice in all defamation actions by public officials and public figures. In the present case counsel for the newspaper argues for a generic test not applicable to a category of victim (such as public figures) but dependent on the subject matter (political speech).

This is a branch of law in which common law courts have arrived at sharply divergent solutions. In the Sullivan case the United States Supreme Court upheld a public figure defence. In Lange v. Australian Broadcasting Corporation (1997) 189 CLR 520 the Australian High Court allowed a qualified privilege of political speech subject to a requirement of due care. In impressive and valuable judgments Elias J. (now Chief Justice) and the Court of Appeal of New Zealand allowed a generic defence of free speech, the rationale of the decisions being policy considerations applicable to New Zealand: Lange v. Atkinson [1997] 2 N.Z.L.R. 22 and [1998] 3 N.Z.R. 424. And in Reynolds v. Times Newspaper Limited [1998] 3 W.L.R. 862 the Court of Appeal enunciated a circumstantial test depending substantially on the source of the information. There are at stake powerful competing arguments of policy. They pull in different directions. It is a hard case in which it is unrealistic to say that there is only one right answer. And in considering the decisions in other jurisdictions it is right to take into account that cultural differences have played an important role.

Counsel for Mr. Reynolds submitted that a generic qualified privilege of political speech, defeasible only by proof of malice or reckless disregard of the truth, would make the prospect of suing a newspaper which published defamatory and false allegations about a politician without checking the facts unduly difficult.óKóóK On the other hand, counsel for the newspaper argued that in the case of an unchecked publication alleging grave misconduct the newspaper would be at significant risk of an adverse jury verdict on the ground of recklessness. He submitted that in the absence of a generic qualified privilege investigative journalism into political matters is inadequately protected. He argued that the generic test will result in more predictable decisions. And he emphasised that it would be consistent with the spirit of the new legal landscape to develop the law in this way.

Weir, A Casebook on Tort, 8th ed., (1996) describes defamation as "the oddest" of the torts. He explains (at p. 525):

"he (the plaintiff) can get damages (swingeing damages!) for a statement made to others without showing that the statement was untrue, without showing that the statement did him the slightest harm, and without showing that the defendant was in any way wrong to make it (much less that the defendant owed him any duty of any kind)"

Weir, at p. 530, observes that "the courts could arguably have done more to prevent the law becoming as absurd, complex and unfair as it is, without resigning themselves to saying, as Diplock L.J. did, that the law of defamation "has passed beyond redemption by the courts" (Slim v. Daily Telegraph Ltd. [1968] 2 Q.B. 157, 179). Weir states that "the law of England is certainly stricter than that of any free country . . ." at p. 528. The argument for addressing the chilling effect of our defamation law on political speech and for striking a better balance between freedom of speech and defamation is strong: see Eric Barendt and others, Libel and the Media: The Chilling Effect, (1997), Clarendon Press, Oxford, pp. 191-192. But the burden is on counsel for the newspaper to demonstrate that the development he advocated would in practice be fair and workable, and could sensibly be accommodated in our legal system.

On balance two particular factors have persuaded me to reject the generic test. First, the rule and practice in England is not to compel a newspaper to reveal its sources: see section 10 of the Contempt of Court Act 1981; R.S.C., Ord. 82, r.6; and Goodwin v. United Kingdom (1996) 22 EHRR 123, 143 at para. 39. By contrast a plaintiff in the United States is entitled to a pre-trial enquiry into the sources of the story and editorial decision-making: Herbert v. Lando (1979) 441 U.S. 153. Without such information a plaintiff suing for defamation in England will be substantially handicapped. Counsel for the newspaper observed that the House could recommend a reform of the procedural rule. This is an unsatisfactory basis to embark on a radical development of the law. Given the procedural restrictions in England I regard the recognition of a generic qualified privilege of political speech as likely to make it unacceptably difficult for a victim of defamatory and false allegations of fact to prove reckless disregard of the truth. Secondly, a test expressed in terms of a category of cases, such as political speech, is at variance with the jurisprudence of the European Court of Human Rights which in cases of competing rights and interests requires a balancing exercise in the light of the concrete facts of each case. While there is as yet no decision directly in point, it seems to me that Professor John Fleming is right in saying that the basic approach of the European Court of Human Rights has been close to the German approach by insisting on individual evaluation of each case rather than categories: "Libel and Constitutional Free Speech," in Essays for Patrick Atiyah, ed. Cane and Stapleton (1991), p. 333 at pp .337 and 345. Our inclination ought to be towards the approach that prevails in the jurisprudence on the Convention. In combination these two factors make me sceptical of the value of introducing a rule dependent on general categorisation, with the attendant sacrifice of individual justice in particular cases.

I would answer question (1) by saying that there is no generic qualified privilege of political speech in England.

Issue (2): Soundness of the circumstantial test

My Lords, it is important to appreciate that the judgment of the Court of Appeal marked a development of English law in favour of freedom of expression. In the context of political speech the judgment recognised a qualified privilege, dependent on the particular circumstance of the case, provided that three requirements are fulfilled. The first and second are the familiar requirements of duty and interest. The Court of Appeal then stated a third and separate requirement. The passage in the judgment reads as follows (at pp. 899G-900B):

"Were the nature, status and source of the material, and the circumstances of the publication, such that the publication should in the public interest be protected in the absence of proof of express malice? (We call this the circumstantial test.)

Later in the judgment the Court of Appeal observed (pp. 909H-910C):

it is one thing to publish a statement taken from a government press release, or the report of a public company chairman, or the speech of a university vice-chancellor, and quite another to publish the statement of a political opponent, or a business competitor or a disgruntled ex-employee;

The circumstantial test was not put forward in the Court of Appeal by either side or raised in argument. But the development was well within the power of the Court of Appeal. On balance, however, I am satisfied that the support for it in the authorities is not great. Except for obiter dicta in Blackshaw v. Lord [1984] Q.B. 1, 42 the other decisions relied on by the Court of Appeal (see [1998] 3 W.L.R. 862, 894H-899D) are cases of institutional reporting which are materially different fromóKóóK reports resulting from investigative journalism. And Blackshaw v. Lord predates the Derbyshire case [1993] AC 534.

The Court of Appeal observed "We question whether in practice this [the circumstantial test] is a test very different from the test of reasonableness upheld in Australia." This is a reference to the decision of the High Court of Australia in Lange v. Australian Broadcasting Corporation (1997) 189 CLR 520. The Lange decision was substantially influenced by a New South Wales statute which imposed a standard of reasonable care on publishers: see Michael Tilbury, "Uniformity, The Constitution and Australian Defamation Law at the Turn of the Century," in Torts Tomorrow: A tribute to John Fleming, ed. Mullany and Linden, (1998), p. 244 for a perceptive analysis of the distinctive Australian context. In reply counsel for the newspaper put forward the Lange solution, with the legal burden on the plaintiff, as an alternative solution. In my view such a development would involve a radical re-writing of our law of defamation. Contrary to the submissions of counsel I also do not think it is a satisfactory way of redressing the imbalance between freedom of speech and defamation in England. I would reject this argument.

For the newspaper counsel argued that the particular requirements of the circumstantial test stated by the Court of Appeal are unduly restrictive. There is force in this argument. I will return to it. Counsel for Mr. Reynolds pointed out in his case:

in general,

He submitted that there is a structural flaw in the circumstantial test. He invited your Lordships not to adopt it. I would not accept the circumstantial test is soundly based. Having reached this point I would not wish to be taken to reject entirely the reasoning of the Court of Appeal. It will be recalled that the Court of Appeal had observed (at 910 B-C):

unless the circumstances of the publication are such as to make

it proper, in the public interest, to afford the publisher immunity from liability in the absence of malice.

After all, this is the core of the reasoning of the Court of Appeal.

I would however rule that the circumstantial test should not be adopted.

Issue (3): The alternative tests of duty and interest

If both the generic test and the circumstantial test are rejected, as I have done, the only sensible course is to go back to the traditional twofold test of duty and interest. These tests are flexible enough to embrace, depending on the occasion and the particular circumstances, a qualified privilege in respect of political speech published at large.

The critical question is then to decide what requirements should be imposed in respect of qualified privilege in the context of political speech. In my view the passages in the Court of Appeal judgment which I have cited should not be elevated to legal requirements. Those passages, with a distinction drawn between official and "unofficial sources," and between "a government press release" and "the statement by a political opponent," could create the impression that if information is not obtained from a prima facie authoritative source, a privileged occasion does not arise. A rule, principle or approach that in considering a plea of qualified privilege of political speech greater weight should be given to what is said on behalf of the government than what is said on behalf of the opposition, other political parties or pressure groups is unacceptable in our democracy. And I am confident that the Court of Appeal did not intend to make such a ruling.

Counsel for Mr. Reynolds did not invite your Lordship to endorse the observations of the Court of Appeal. Instead he submitted that in the context of political speech qualified privilege must always fulfil as part of the duty test three legal requirements: (1) that the occasion must be one in respect of which it can fairly be said that it is in the public interest that the information should be published; and (2) that a report which "failed to report the other side" would always fail the test; (3) that there is a burden on a publisher of a report to prove that there is a cogent reason why it should be excused in the particular circumstances from justifying the truth of the assertion.

My Lords, the first proposition involves nothing radical or extravagant. It builds on the web of existing law. I am content to accept that it should be the governing principle. The second proposition put forward by counsel as an independent legal requirement is implausible. A failure to report the other side will often be evidence tending to show that the occasion ought not to be protected by qualified privilege. But it would not necessarily always be so, e.g. when the victim's explanation is unintelligible or plain nonsense. This was recognised in the Australian Lange case: 189 CLR 520, 574. The suggested strict requirement runs counter both to the pragmatic approach of the common law and a test dependent on particular circumstances. The third proposition overlaps with the first requirement. But as expressed it would emasculate the qualified privilege of political speech. I would reject it.

Returning now to the requirement that the occasion must be one in respect of which it can fairly be said to be in the public interest that the information about political matters should be published, I would accept that it may be objected that this requirement is imprecise. But this is a corner of the law which could do with the minimum of legal rules. And what is in the public interest is a well-known and serviceable concept. It will, of course, have to be given practical content. Inevitably the question will arise in concrete cases whether the newspaper was entitled to rely on the information it had obtained before publishing. This issue can be accommodated within the test of an occasion in the public interest warranting publication. In my view such an approach complies with the requirement of legal certainty. And in practice the issue will have to be determined on the whole of the evidence. If a newspaper stands on the rule protecting its sources, it may run the risk of what the judge and jury will make of the gap in the evidence.

The context in which the qualified privilege of free speech should be applied is all important. It was said by counsel for the newspaper that the English courts have not yet recognised that the press has a general duty to inform the public of political matters and that the public has a right to be so informed. If there is any doubt on the point this is the occasion for the House to settle the matter. It is an open space in the law which can be filled by the courts. It is true that in our system the media have no specially privileged position not shared by individual citizens. On the other hand, it is necessary to recognise the "vital public watchdog role of the press" as a practical matter: see Goodwin v. The United Kingdom (1996) 22 EHRR 123, 143, para. 39. The role of the press, and its duty, was well described by the European Court of Human Rights in Castells v. Spain (1992) 14 EHRR 445, 476, para. 43 in the following terms:

inter alia

"Freedom of the press affords the public one of the best means of discovering and forming an opinion of the ideas and attitudes of their political leaders. In particular, it gives politicians the opportunity to reflect and comment on the preoccupations of public opinion; it thus enables everyone to participate in the free political debate which is at the very core of the concept of a democratic society."

In De Haes Gijsels v. Belgium (1997) 25 E.H.R.R. 1 the European Court of Human Rights again emphasised that the press plays an essential role in a democratic society. The court trenchantly observed (at p. 53; para. 39):

"It is incumbent on the press to impart information and ideas of public interest. Not only does the press have the task of imparting such information and ideas: the public also has a right to receive them."

This principle must be the foundation of our law on qualified privilege of political speech.

The correct approach to the line between permissible and impermissible political speech was indicated by the European Court of Human Rights in Lingens v. Austria (1986) 8 E.H.R. 407, as follows (at 419, para. 42):

"The limits of acceptable criticism are accordingly wider as regards a politician as such than as regards a private individual. Unlike the latter, the former inevitably and knowingly lays himself open to close scrutiny of his every word and deed by both journalists and the public at large, and he must consequently display a greater degree of tolerance. No doubt article 10(2) enables the reputation of others--that is to say, of all individuals--to be protected, and this protection extends to politicians too, even when they are not acting in their private capacity; but in such cases the requirements of such protection have to be weighed in relation to the interests of open discussion of political issues."

Implicit in that dictum is the distinction that speech about political matters has a higher value than speech about private lives of politicians. The dictum in the Lingens case was reinforced by the European Court of Human Rights in Oberschlick v. Austria (1991) 19 E.H.R.R. 389, 422, para. 59. Moreover, it will always be necessary to take into account the dynamics of the role of the press and that "news is a perishable commodity and to delay its publication, even for a short period, may well deprive it of all its value and interest": The Sunday Times v. United Kingdom No. 2 (1991) 14 E.H.R.R. 229, 242 (para. 51). If the matter is approached in this liberal way the balance in our law between freedom of information and the right to reputation should fulfil the Convention requirement of being necessary in a democracy.

In the result I would uphold qualified privilege of political speech, based on a weighing of the particular circumstances of the case.

Issue (4): The function of the judge and jury

My Lords, the American Law Institute, Restatement of the Law, Torts, 2d, (1977) summarises in paragraph 619 the function of judge and jury in the following terms in regard to privilege:

"(1) The court determines whether the occasion upon which the defendant published the defamatory matter gives rise to a privilege. "(2) Subject to the control of the court whenever the issue arises, the jury determines whether the defendant abused a conditional privilege."

The commentary on subsection (1) reads as follows:

"Whether a privilege exists at all is a question for the court. This requires the court to determine whether the circumstances under which the publication was made were such as, . . . to make the publication privileged. This is true whether the issue involves the existence of an absolute privilege or of a conditional privilege. If the facts are in dispute, the jury is called upon to consider the evidence and pass upon the issues thus raised. It is for the court, however, to decide whether the facts found by the jury made the publication privileged or to instruct the jury as to what facts they must find in order to hold the publication privileged."

For the sake of completeness the commentary on subsection (2) is as follows:

"The question whether the defendant acted for an improper purpose or in an improper manner is material if the publication is conditionally privileged . . . Under these circumstances, the qualified protection thus created is lost if the defendant has utilized the privilege for a purpose other than that for which the privilege was created, or if he otherwise abused the privilege. . . . These questions are for the jury to determine unless the facts are such that only one conclusion can reasonably be drawn.

For my part these principles admirably and accurately state the English law and practice on the topic of qualified privilege: see Hebditch v. MacIlwaine [1894] 2 QB 54, 58; Adam v. Ward [1917] A.C. 309, 318; Minter v. Priest 1930 A.C. 558, 571-572; Kingshott v. Associated Kent Newspapers [1991] 1 Q.B. 88, 101A-C. I would apply it to the qualified privilege of political speech.

The particular qualified privilege which I have held to exist may or may not involve issues of primary or secondary fact which are for the jury. But the judge may withdraw the issue from the jury if only one conclusion can be drawn and, in any event, in the light of the jury's findings of fact it is for the judge to decide whether the occasion was privileged.

Issue 5: The decision of the Court of Appeal

The question arises how the appeal should be resolved.

The Court of Appeal enunciated a test of qualified privilege, which marked a new development of the law. As a result of the speeches in the House today a different approach has been adopted. In this very difficult case nobody could at the time of trail realistically have foreseen this outcome. Given that a retrial, involving a different judge and jury has been ordered, I regard it as fair that the issue of qualified privilege should be before the judge and jury to be considered in accordance with the speeches delivered today. In any event, on the basis of a transcript of evidence not placed before the Court of Appeal, it is now clear that the Court of Appeal's assumption "that there was no evidence before the jury that Mr. Spring authorised Mr. Finlay to accuse Mr. Reynolds of lying" was wrong: at p. 911F. Furthermore, the finding that "Mr Spring did not in terms accuse Mr. Reynolds of lying to the Dáil" is arguably contrary to the findings of the jury and, in any event, debatable. Indeed counsel for Mr. Reynolds described it as a complex issue. Moreover, the issue of justification will have to be reconsidered at the retrial and the evidence on it may overlap with the evidence on qualified privilege. It is fair that both the issues of justification and qualified privilege should be considered by the judge and jury.

The only escape from this outcome is to say that a failure to publish the explanation given by Mr. Reynolds in the Dáil precludes the newspaper as a matter of law from relying on the qualified privilege of political speech. My Lords, I have already explained why I would not put the law in such a rigid straight jacket. And my understanding is that there is no support for such a rule in the speeches delivered today.

For these reasons I would hold that the Court of Appeal's ruling (at p. 912A) that "this was not a publication which should in the public interest be protected by privilege in the absence of malice" should not be upheld. Issue 6: The disposal of the appeal.

I would allow the appeal and remit the issue of qualified privilege to be considered at the retrial.

LORD COOKE OF THORNDON

My Lords,

I am in full agreement with the speech of my noble and learned friend Lord Nicholls of Birkenhead.

The article sued on is a mixture of allegations of fact, comment and reporting. The chief defence at the trial was justification: that is to say, truth. The sting of the article was that Mr. Reynolds had lied to and deceived by non-disclosure the Dáil and his colleague in government, Mr. Spring. An impugning of what was said in the Dáil was thus at the heart of the case, but it became common ground in the argument of this appeal that the plaintiff was entitled to waive and had waived parliamentary privilege. It would seem that implied waiver may likewise explain Adam v. Ward [1917] A.C. 309; contrast Prebble v. Television New Zealand Ltd. [1995] 1 AC 321.

The defence of justification in the present case was disposed of by the jury's finding that the allegation complained of was not true in substance--a finding reached notwithstanding certain misdirections which the Court of Appeal held to have had the effect of denying the plaintiff a fair trial. The part of the Court of Appeal's judgment concerning misdirections has not been challenged on the appeal to your Lordships. At first sight it seems odd that the jury awarded no damages. Lord Lester of Herne Hill Q.C. for the appellants suggested during the argument before the Appellate Committee that the jury's reason for no award was that in evidence before them there were some Irish newspapers containing similar material, on which the plaintiff had not sued in Ireland. That may explain the apparent inconsistency in the verdict, but has little if any bearing on the issue of qualified privilege which your Lordships have to determine.

Other defences pleaded had been fair comment on a matter of public interest, and a fair and accurate report of proceedings in public of the Irish legislature (Defamation Act 1952, section 7 and Schedule, paragraphs 1 and 14; cf. Defamation Act 1996, section 15 and First Schedule, paragraph 1). But both these defences were abandoned at the outset of the trial. Fair (that is to say, honest) comment would have failed because, as the jury in effect found, the basic facts were not truly stated. To the extent that the article was a fair and accurate report of proceedings in the Dáil, it would have been protected by statutory qualified privilege; but it was not a fair and accurate report, as it omitted the explanation given to the Dáil by Mr. Reynolds. In any event the reporting of the proceedings in the Dáil was mixed up with other allegations, including lying, which the newspaper appeared to adopt as its own or to accept; and these would have been outside the statutory reporting privilege.

In that situation the defence could not succeed unless the case could be brought within the protection of the subsisting principles of common law regarding qualified privilege (which are not limited or abridged by the statutory privileges: see section 7(4) of the Act of 1952 and cf. section 15(4) of the Act of 1996); or unless the court could be persuaded to introduce into English law a new generic head of qualified privilege for political discussion, on lines similar, for instance, to that proposed for New Zealand by the New Zealand courts in Lange v. Atkinson [1997] 2 N.Z.L.R. 22; [1998] 3 N.Z.L.R. 424, contemporaneously under appeal to the Privy Council. The less-sweeping new generic head established by the High Court of Australia in Lange v. Australian Broadcasting Corporation (1997) 189 CLR 520 might be an alternative approach.

In Reynolds the defence preferred to take the high ground. As in the courts below, counsel for the appellants to your Lordships concentrated on arguing for a new generic privilege for political discussion, limited merely by the possibility of the plaintiff's proving malice. Only in the dying stages of Lord Lester's reply was a less radical new generic privilege put forward as an alternative. This would have some similarity to the solution evolved for Australia in Lange v. Australian Broadcasting Corporation, with a major difference as to onus. The fallback position of the present appellants would involve placing on the plaintiff the burden of proving unreasonable conduct or lack of reasonable care on the part of the defendant, whereas the Australian solution requires the defendant to establish reasonableness.

Arguments invoking freedom of speech in a democracy have ready moral, intellectual and emotional appeal, and in this instance their presentation by Lord Lester and Mr. James Price Q.C. lacked nothing in potency. Some famous observations were cited. Your Lordships' Committee were reminded that it was eloquently said by Judge Learned Hand in United States v. Associated Press 52 F. Supp. 362, 372 (1943) that the First Amendment " . . . presupposes that right conclusions are more likely to be gathered out of a multitude of tongues, than through any kind of authoritative selection. To many this is, and always will be, folly; but we have staked upon it our all." In like vein was the pronouncement of Holmes J., dissenting but with the concurrence of Brandeis J., in Abrams v. United States 250 U.S. 616, 630 (1919) " . . . the best test of truth is the power of the thought to get itself accepted in the competition of the market . . . "

Such observations are most naturally apposite, however, to freedom to express ideas and convey news. Neither of the cases in which they were made was a defamation case. It would be dangerous to stretch them out of context. As to defamatory allegations of fact, even in the United States the opinions of jurists differ on the extent to which the collectively cherished right of free speech is to be preferred to the individually cherished right to personal reputation; and it is certain that neither in the United Kingdom nor anywhere else in the Commonwealth could it be maintained that the people have knowingly staked their all on unfettered freedom to publish falsehoods of fact about political matters, provided only that the writer or speaker is not actuated by malice. It would be a mistake to assume that commitment to the cause of human rights must lead to a major abandonment of established common law limitations on political allegations of fact. See, for instance, the Francis Mann lecture by Sir Sydney Kentridge Q.C. (as he now is) Freedom of Speech: Is it the Primary Right?, published in (1996) 45 I.C.L.Q. 253, wherein Sir Sydney argues against introducing a New York Times Co. v. Sullivan-type defence for political discussion (see 376 U.S. 254). "It should not be beyond a court's ability," he says at page 268, "to distinguish in any particular case between hard-hitting political criticism and truly libellous allegations of fact." I would follow that approach.

As I understand it, none of your Lordships who sat in this case and in the New Zealand Lange case favours any new form of generic privilege for political discussion; and I am of the same mind for the following main reasons:

(i) Although investigative reporting can be of public benefit, the commercial motivation of the press and other sections of the media can create a temptation, not always resisted, to exaggerate, distort or otherwise unfairly represent alleged facts in order to excite the interest of readers, viewers or listeners. This very case may conceivably be an illustration. On the same date, 20 November 1994, as that of the British mainland edition of the 'Sunday Times' containing the article sued on, the Irish edition of the same paper carried a much longer article on the same subject. It presented Mr. Reynolds as a victim of circumstances, which it traced in much detail, and its tone is markedly less dramatic and more objective. As the Court of Appeal records, Mr. Reynolds accepts the article in the Irish edition as being very largely accurate and on the whole unobjectionable. It is common ground that those responsible for the British mainland edition knew what was to be published in the Irish edition. A possible inference, albeit supported by no direct evidence, is that it may have been felt that as full, factually detailed and balanced an account would not have the same appeal for British mainland readers. Be that as it may, there is in my opinion no good reason why politicians should be subjected to a greater risk than other leading citizens, or for that matter any other persons, of false allegations of fact in the media.

(ii) In the United Kingdom a succession of well-qualified committees on the reform of defamation law have rejected anything approaching the generic privilege for which the appellants primarily contend. They have specifically rejected for the United Kingdom a Sullivan approach. Counsel for the present appellants disclaimed seeking to go as far as that case. Still, they adopted as part of their argument certain letters from a leading New York libel attorney, which they tendered during the hearing of this appeal, including an assessment that in the United States public officials and public figures not only have a genuine opportunity to meet the Sullivan test but often do so by succeeding in actual litigation.

What is being proposed for the appellants is, or is at least close to, Sullivan in a limited sphere (politicians but not at this stage other public figures) but without any assurance that, on the issue of malice, the plaintiff will have access to the defendant's sources. As I understand it, plaintiffs do commonly have such access in the United States. In the United Kingdom the common law and practice regarding protection of media sources has been strengthened by section 10 of the Contempt of Court Act 1981, prohibiting any court from requiring disclosure of a journalistic source unless satisfied that disclosure is necessary in (inter alia) the interests of justice. A contemporary textbook, Carter-Ruck on Libel and Slander 5th ed. (1997), pp. 105-107, refers to the difficulty of predicting when disclosure will be ordered, citing X Ltd. v. Morgan-Grampian (Publishers) Ltd. [1991] 1 A.C. 1 and Goodwin v. United Kingdom 22 EHRR 123. See also Maxwell v. Pressdram Ltd. [1987] 1 W.L.R. 298 for a vivid illustration of this uncertainty in the defamation field.

(iii) There are further reasons why the exception of malice is a dubious safeguard. Few persons contemplating bringing a defamation suit would derive much confidence from advice that, if the case were skilfully handled, their lawyers might succeed in proving malice. The defendant is entitled to a direction that, while recklessness as to whether the facts are true or not amounts to malice, mere carelessness, impulsiveness, vehemence of language, and even gross and unreasoning prejudice, do not: see Horrocks v. Lowe [1975] A.C. 135, 145 to 146 and 150, per Viscount Dilhorne and Lord Diplock respectively. So too, although much was made for the present appellants of the ability of a jury to find malice if a defendant newspaper elects not to reveal its sources, the defendant will normally be entitled to a direction that in itself unwillingness to reveal confidential sources is not evidence of malice. The burden of proving malice is a heavy one and it may be extremely difficult to establish: Spring v. Guardian Assurance Plc. [1995] 2 AC 296, 329 per Lord Slynn of Hadley, 346 per Lord Woolf.

(iv) It is doubtful whether the suggested 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. Such 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.

(v) The existing balance between the right to personal reputation and freedom of speech has been carefully and gradually developed over the years by common law and statutes. It is true that the restrictions on freedom of speech that have been thought necessary to give reasonable protection to personal reputation may have a tendency to chill the publication, not only of untruths, but also of that which may be true but cannot be proved to be true. But there is nothing new in this. Nor, as far as I am aware, is there any way of assessing which tendency is the greater--although experience of libel litigation is apt to generate a suspicion that it is the former. A new generic qualified privilege of the width primarily urged for the appellants would do violence to the present pattern of the law without any compelling evidence of necessity. As regards discussion of government and political matters, the defences of justification, fair comment and privilege for fair and accurate reports of certain proceedings would all, at one stroke, be rendered virtually obsolete. No longer would the defendant have to prove the truth of any defamatory allegations or substratum of fact. No longer would any report have to be fair and accurate. The sole safeguard would be the possibility of the plaintiff's proving malice, as to the adequacy of which I have already expressed misgivings. Of course a trial judge may point out that the truth of an allegation has not been pleaded or proved, but such niceties can be buried beneath the general impression conveyed to the public of who has won or lost the case.

(vi) The foregoing considerations do not exert the same force against the solution evolved in the Australian Lange case 189 CLR 520. Reconciling the differences of opinion in Theophanous v. Herald & Weekly Times Ltd. (1994) 182 C.L.R. 104 and Stephens v. West Australian Newspapers Ltd. (1994) 182 C.L.R. 211 and in some respects modifying the view of the majority in those cases, the High Court in Lange settled on a new common law privilege for communications to the general public on government or political matters, conditioned by a defence onus of proving reasonableness of publication. New South Wales statute law was held to be consistent with this solution. The federal Constitution, providing for representative and responsible government, was now seen, not as a direct source of the privilege, but as a restriction on legislative power and a background or context helping to demonstrate a need to develop the common law of Australia. As I see it, however, the United Kingdom is no less a representative democracy with responsible government than Australia. The same can be said of other comparable jurisdictions, including New Zealand. For the purposes of defamation law, the background or context does not seem materially different. The constitutional structures vary, but the pervading ideals are the same. Freedom of speech on the one hand and personal reputation on the other have the same importance in all democracies.

But the Australian solution is not supported by either side in the present litigation (the fallback position of the appellants differing as to onus), and I share the view that your Lordships should not impose it without at least some difference of emphasis. The whole purpose of defamation law is to enable a plaintiff to clear his or her name. The privilege required for reasonable freedom of speech does run counter to that purpose in some cases. A major expansion of the privilege, such as may have been achieved in Australia, shifts the focus of political defamation to the conduct of the defendant. In practice it may leave a politician plaintiff without redress. His or her private life may be immune from the extended privilege, but otherwise the opportunity of a public clearing of name may be virtually gone. If the Australian solution has disadvantages, they may lie in this change of focus and in the singling out of politicians as acceptable targets of falsehood.

Further, it is hard not to see something a little incongruous or awkward about the proposition stated in the Australian Lange case:

(189 CLR 520

This appears to set an Australian judge the task of determining what would have been the common law of England on the case at hand, at some unspecified date, before the judge can decide whether to disapply that law. If workable in Australia, it could hardly be appropriate in England.

(vii) In the judgment now under appeal the Court of Appeal described the New Zealand Lange case as the sheet anchor of Lord Lester's arguments [1998] 3 W.L.R. 862, 906D. In that case the New Zealand courts struck out on a new line in deciding not to apply the decision of the New Zealand Court of Appeal in Templeton v. Jones [1984] 1 N.Z.L.R. 448. That decision was not in proceedings against a newspaper or a broadcasting station but, even so, the approach in the New Zealand Lange case is different.

The parties in Templeton v. Jones were rival candidates for a parliamentary seat at a general election. The defendant had made to the annual general meeting of the electorate branch of his political party a speech in which he said (inter alia), apparently as a statement of fact, that the plaintiff despised Jews. Copies of his speech were distributed by the defendant to the parliamentary press gallery. As a result the allegation in the speech was broadcast in a national television news programme. The action was brought on that publication, for which it was not denied that the defendant was responsible. On an appeal from a pre-trial ruling the Court of Appeal held that there was neither any general privilege protecting publication of political matter to the public at large, nor in the particular circumstances any privilege protecting the publication of this allegation by the defendant.

On the question of general privilege, the authorities cited specifically in the judgment were from several jurisdictions. From England Duncombe v. Daniell (1837) 8 C. & P. 222; Braddock v. Bevins [1948] 1 K.B. 580; Plummer v. Charman [1962] 1 W.L.R. 1469, 1474, per Diplock L.J.; and Blackshaw v. Lord [1984] Q.B. 1. From Australia Lang v. Willis (1934) 52 C.L.R. 637, 667, per Dixon J. From Canada Douglas v. Tucker [1952] 1 D.L.R. 657; Globe and Mail Ltd. v. Boland (1960) 22 D.L.R. (2d) 277; Jones v. Bennett (1968) 2 D.L.R. (3d) 291; and Lawson v. Chabot (1974) 48 D.L.R. (3d) 556. From New Zealand Bradney v. Virtue (1909) 28 N.Z.L.R. 828, 839, per Edwards J.; Truth (N.Z.) Ltd. v. Holloway [1960] N.Z.L.R. 69; Dunford Publicity Studios Ltd. v. News Media Ownership Ltd. [1971] N.Z.L.R. 961; and Brooks v. Muldoon [1973] 1 N.Z.L.R. 1. In the light of these authorities, and notwithstanding New York Times Co. v. Sullivan 376 U.S. 254, the New Zealand Court of Appeal in Templeton v. Jones declined to introduce in New Zealand a new generic privilege.

My Lords, with the benefit of the arguments in the present appeal and in the appeal to the Privy Council in the New Zealand Lange case, I have returned to the authorities on which Templeton v. Jones was founded. As the authorities stood in 1984, I continue to regard the decision in Templeton v. Jones as inevitable. It is as well to add that in Horrocks v. Lowe Lord Diplock remarked (see [1975] A.C. 135, 152) that qualified privilege does cover what local councillors say at meetings of the council or its committees; but that appears to be an exception to and not to undermine his broader proposition in Plummer v. Charman [1962] 1 W.L.R. 1469, 1474:

"I need hardly say that there is no privilege known to the law which entitles persons engaged in politics to misstate a fact about their opponent provided that they say it honestly even though untruthfully. They can comment upon the conduct of persons in public life, provided they do so honestly and without malice."

It is also true that, (unlike earlier authorities and the New Zealand Lange case and the present case) Templeton v. Jones related to a television programme. The power of the media and the facility of communicating with the general public have certainly been much increased by television. It seems to me, however, that this is far from a ground for extending the heads of privilege. On the contrary, if anything it adds to the importance of principles aimed at ensuring journalistic responsibility.

But the common law nowhere stands still. In this field of much international debate, I think that it was open to the New Zealand Court of Appeal in the Lange case to reconsider Templeton v. Jones. Indeed I would put it more strongly. In the light of the intervening line of cases across the Tasman--namely, the Theophanous, Stephens and Australian Lange cases--I would accept that reconsideration of Templeton v. Jones was either incumbent on the New Zealand courts or at least highly appropriate. It is the result, so far, of their reconsideration with which I respectfully disagree. In the Reynolds case Lord Bingham of Cornhill L.C.J. in giving the judgment of the English Court of Appeal has said that in the New Zealand Lange case " . . . no or at least insufficient weight is given to the proper balance . . . " (see [1998] 3 W.L.R. 862, 907H). For the reasons set out in my present speech, I agree with the Lord Chief Justice and his colleagues. At the same time, as a party to the Privy Council judgment in the New Zealand Lange case, I am equally clear that there is a high element of judicial policy in the resolution of the issue, and that the best course is to refer the New Zealand Lange case back to the New Zealand Court of Appeal to enable account to be taken of the Reynolds case. In other words, the possibility of a difference between English and New Zealand common law on the issue has to be accepted, albeit not advocated.

(viii) International human rights law, whenever relevant, should have an important part to play in developments of the common law. For United Kingdom courts, particular importance must attach to the European Convention for the Protection of Human Rights and Fundamental Freedoms, bearing in mind that by section 6(1) of the Human Rights Act 1998 it is unlawful for a public authority to act in a way which is incompatible with a Convention right. By section 6(3)(a) "public authority" here includes a court or tribunal. By section 2(1)(a) decisions of the European Court of Human Rights must be taken into account. The Convention rights here relevant are to be found in article 10.1 (which includes rights to freedom of expression, and to receive and impart information and ideas) and are subject to article 10.2 (which speaks of accompanying duties and responsibilities and authorises such restrictions as are prescribed by law and are necessary in a democratic society for the protection of the reputation or rights of others). The Act is not yet in force, but naturally the appeal was argued on the footing that regard should be had to it.

The jurisprudence of the European Court of Human Rights has been reviewed by Lord Nicholls. I need say only that it reveals first an emphatic distinction between fact and opinion, secondly a careful examination of all the circumstances of a particular case before a decision is reached as to whether freedom of expression is to be treated as the dominant right. As the European case law stands at present, no trace is to be found of endorsement of a generic privilege in the political context. This is not surprising in view of the balance aimed at by article 10. I am afraid that the arguments for the appellants would tend, in effect, to divert your Lordships from the European path.

(ix) The Human Rights Act also has a special provision, original to the United Kingdom, pointing to the answer to the present problem. It is section 12:

"12.--(1) This section applies if a court is considering whether to grant any relief which, if granted, might affect the exercise of the Convention right to freedom of expression. . . .

(4) The court must have particular regard to the importance of the Convention right to freedom of expression and, where the proceedings relate to material which the respondent claims, or which appears to the court, to be journalistic, literary or artistic material (or to conduct connected with such material), to -

(i) the material has, or is about to, become available to the public; or

(ii) it is, or would be, in the public interest for the material to be published;

(5) In this section -

court includes a tribunal; and

relief includes any remedy or order (other than in criminal proceedings)."

The omitted subsections contain restraints on ex parte and pre-trial relief, and the whole section is inspired by the purpose of ensuring a due measure of media freedom. What are significant in the present context are the references to journalistic material and especially to the extent to which it is, or would be, in the public interest for the material to be published. The focus appears to be on the particular material rather than the general subject matter. Of course the general subject matter is a factor to be considered, but I do not think that a court would discharge its responsibility under the section by going no further than satisfying itself that the material related to government or political matters. A more specific examination appears to be contemplated by Parliament. The common law of qualified privilege should evolve in harmony with that legislative approach.

Cumulatively the reasons under the foregoing heads lead me to the view that the Court of Appeal in the present case and an earlier Court of Appeal in Blackshaw v. Lord [1984] Q.B. 1 adopted substantially the right approach. The categories of qualified privilege are not closed. When a case cannot be brought within an established generic category, it may nevertheless call for a finding of privilege if in all the circumstances the classical tests of reciprocal interest and duty or common interest are satisfied. Although sometimes newspaper privilege has been put on the ground of common interest (see Perera v. Peiris [1949] AC 1, 22), the weight of authority favours the former test and it has the advantage of underlining media responsibility. There are occasions when the media may rightly claim to have a social or moral duty to publish defamatory material to the world at large; but there is no room for any suggestion that the motive of increasing readership or audience is a sufficient interest, nor does it seem altogether realistic to treat the media as no more than citizens communicating with other citizens on matters of common interest. It was indeed the duty of the media on which in his sustained argument Lord Lester placed constant stress.

In the judgment now under appeal the circumstantial test was separated to some extent from the duty and interest tests. Not entirely, I think, for it was said that, while the duty and interest tests were "in general" satisfied, the circumstantial test was not. This may indicate that the Court of Appeal thought that the general subject matter of the article brought it potentially within qualified privilege but that the particular context and surrounding circumstances ruled the privilege out. The threefold analysis is largely a matter of arrangement. I agree that the twofold classical test is enough, once it is accepted that all the circumstances of the publication are to be taken into account. It is undeniable that a privilege depending on particular circumstances may produce more uncertainty and require more editorial discretion than a rule-of-thumb one. But in other professions and callings the law is content with the standard of reasonable care and skill in all the circumstances. The fourth estate should be as capable of operating within general standards.

A more formidable argument against the approach of the Court of Appeal is that it introduces at the stage when the existence of privilege is determined issues which are said to be relevant only to malice or abuse of the occasion. In the leading case of London Association for Protection of Trade v. Greenlands Ltd. [1916] 2 A.C. 15, 23, Lord Buckmaster L.C. said:

"Again, it is, I think, essential to consider every circumstance associated with the origin and publication of the defamatory matter, in order to ascertain whether the necessary conditions are satisfied by which alone protection can be obtained, but in this investigation it is important to keep distinct matter which would be solely evidence of malice, and matter which would show that the occasion itself was outside the area of protection."

Lord Lester pointed out that at the end of his speech (at p. 27) Lord Buckmaster L.C. indicated that whether the material published had been checked went only to malice. The argument is that failure to include the plaintiff's account, or to give him an opportunity of contradicting the article to be published, are examples of matters which do not bear on the existence of privilege, but only on the loss of privilege.

The answer to that argument, in my opinion, is to be found in the nature of the publication. The Greenlands case was one of publication in confidence to a single potential customer. Many qualified privilege cases are concerned with very limited publications. Then, the occasion and the subject matter being identified, there is normally no reason to go further. When a publication to the world at large is in issue, however, the policy of the law is different. For reports, fairness and accuracy are essential at common law (Wason v. Walter (1868) L.R. 4 Q.B. 73), just as they invariably are for statutory reporting privileges (see now the Defamation Act 1996, section 15 and Schedule 1). Some of the latter also require compliance with requests to publish reasonable statements by way of explanation or contradiction.

Hitherto the only publications to the world at large to which English courts have been willing to extend qualified privilege at common law have been fair and accurate reports of certain proceedings or findings of legitimate interest to the general public. In Blackshaw v. Lord [1984] Q.B. 1, Templeton v. Jones [1984] 1 N.Z.L.R. 448, and now the present case, the law is being developed to meet the reasonable demands of freedom of speech in a modern democracy, by recognising that there may be a wider privilege dependent on the particular circumstances. For this purpose I think it reasonable that all the circumstances of the case at hand, including the precautions taken by the defendant to ensure accuracy of fact, should be open to scrutiny. Lord Nicholls has listed, non-exhaustively, matters to be taken into account. As the Court of Appeal suggested, this brings English law into a position probably not very different from that produced by the Australian reasonableness test, but perhaps rather more consonant with common law tradition. Onus becomes unimportant, except in the sense that evidence of the circumstances surrounding the publication is necessary. The contents of the publication in those circumstances become all-important.

The established common law rule, for which Adam v. Ward [1917] A.C. 309 is the leading authority, is that disputed questions of fact relevant to an issue of qualified privilege are for the jury, but otherwise it is for the judge to determine whether the privilege applies: see Gatley on Libel and Slander, 9th ed. (1998), para. 34.15. The editors of that work evidently regard this common law rule as unaffected by Kingshott v. Associated Kent Newspapers Ltd. [1991] 1 Q.B. 88, which they deal with elsewhere therein, particularly in para. 15.5, note 43. In this I think they are right. The Kingshott case held that Adam v. Ward had not overruled earlier decisions on what is now section 15(3) of the Defamation Act 1996, a provision excluding statutory reporting privilege if matter published to the public, or a section of the public, "is not of public concern and the publication . . . is not for the public benefit." Distinguishing Adam v. Ward as not concerned with the statutory privilege, the Court of Appeal in Kingshott held that under the statute public concern and public benefit were matters for the jury. The principal judgment was given by Bingham L.J., as he then was, and it may be significant that in his Reynolds judgment on common law privilege he makes no mention of the Kingshott case.

At common law any value judgment required in determining whether a publication is privileged has been widely understood, in England and I believe elsewhere in the Commonwealth, as falling to the judge. I would be loath to entrench upon that understanding. Defamation cases are already difficult enough for juries, and the drastic judicial surgery that has had to be undertaken to curb extravagant awards of damages (see John v. MGN Ltd. [1997] QB 586) suggests that it may now be over-romantic to conceive of juries as champions of freedom of speech as in the days of Penn and Mead's case (1670) 6 St.Tr. 951 and Bushell's case (1670) 6 St.Tr. 999.

As for the application of the principles to the circumstances of the present case, I cannot do better than reproduce the Court of Appeal's words reported in [1998] 3 W.L.R. 862, 911-912-

"As already noted, in the present case there was only one issue of fact which was pertinent to qualified privilege left to the jury, namely whether the words complained of correctly reported Mr. Spring's stated reasons for withdrawing from the government. This question was answered in the defendants' favour, and is not the subject matter of the plaintiff's appeal. We can therefore proceed on the footing that this answer was correct, and that otherwise the relevant facts are not in issue.

The circumstances in which Mr. Reynolds's government fell from power were matters of undoubted public interest to the people of Great Britain. We think it clear that the defendants had a duty to inform the public of these matters and the public had a corresponding interest to receive that information. So the duty and interest tests were, in general, satisfied. We cannot, however, regard the circumstantial test as satisfied.

(1) The allegation that Mr. Reynolds had lied was attributed in the article to an unidentified colleague of Mr. Spring. This source was later identified, as a result of the exchange of witness statements, as a Mr. Finlay, who was not a deputy but was described in the Dáil as 'Mr. Spring's programme manager.' There was no evidence before the jury that Mr. Spring authorised Mr. Finlay to accuse Mr. Reynolds of lying, and Mr. Finlay (although present in court for part of the trial) was never called as a witness. In the bitter aftermath of these events, a member of the staff of one of Mr. Reynolds's leading political opponents could scarcely be judged an authoritative source for so serious a factual allegation. (2) Mr. Spring did not in terms accuse Mr. Reynolds of lying to the Dáil. He did, in his speech on Wednesday, 16 November, strongly criticise Mr. Reynolds for failing to disclose what he had known on Tuesday, 15 November about the Duggan case; but his criticism was consistent with an honest but mistaken omission on Mr. Reynolds's part. (3) The defendants wholly failed to record Mr. Reynolds's own account of his conduct, as described by him when addressing the Dáil in the Wednesday debate. (4) The defendants did not, between the debate on Wednesday and publication on Sunday, alert Mr. Reynolds to their highly damaging conclusion that he had lied to his coalition colleagues and knowingly misled the Dáil so as to obtain his observations on it. (5) The defendants failed to resolve whether Mr. Reynolds was a victim of circumstance, as conveyed to Irish readers in the 'House of Cards' article, or a devious liar, as conveyed to readers on the mainland of Britain. It should have been obvious that he could not be both.

Given the nature, status and source of the defendants' information, and all the circumstances of the publication, this was not in our judgment a publication which should in the public interest be protected by privilege in the absence of proof of actual malice."

Subject to the refinement that the circumstantial test should not be treated as something apart from the duty-interest test, I would be content to adopt all of that. Various explanations were offered for the appellants, but they do not shake the essential accuracy of what the Court of Appeal said. It does seem to me to be correct that there was no evidence before the jury that Mr. Spring authorised Mr. Finlay to accuse Mr. Reynolds of lying. There was some evidence to the effect that one might naturally assume from Mr. Finlay's association with Mr. Spring that he spoke with the latter's authority; but that is a different point and does not in my view affect the balance of the case as to qualified privilege. If their primary argument for generic privilege fails, the appellants seek to have the issue of qualified privilege determined at a new trial and on possibly different evidence. My Lords, I cannot think that this would be just. They had every opportunity at the trial of calling such evidence as they saw fit. In the light especially of Blackshaw v. Lord [1984] Q.B. 1, it was readily foreseeable that any privilege might be held to depend on the particular circumstances; and the pleadings and arguments for the defendants were wide enough to cover this possibility, although it was not the outcome for which they primarily contended. The new trial has not been ordered because of any defect in the trial having anything to do with the ruling against qualified privilege. On that issue the defendants have had their day in court--indeed many days in three courts--and, if a new trial does take place, it should be, as I see the justice of the case, on the basis that the article is not eligible for privilege. I would dismiss the appeal.

LORD HOPE OF CRAIGHEAD

My Lords,

Among the issues which are raised by this case are two important questions which relate to the structure of the law of defamation in its application to qualified privilege.

The first question is whether discussion of matters relating to the public conduct of those elected to positions of responsibility in government (referred to as "political discussion" by Lord Lester of Herne Hill Q.C. in the course of his argument) should be accorded the benefit of a generic common law qualified privilege. If that were so, all defamatory statements of fact made in the course of such discussion would be protected by the privilege. And the benefit of the defence would extend not only to the newspapers but to all sections of the media. The result would be that all statements of fact falling within the scope of this category would be presumed to have been made without malice. The burden of proving malice would rest in all such cases on the person who claimed that the statement was defamatory.

The second question assumes that the availability of the defence will continue to depend upon the facts of each case. It relates to the tests which must be applied in order to decide whether, in the particular circumstances, the occasion on which the statement was made was one which entitled the maker of it to the protection of the qualified privilege. Giving the judgment of the Court of Appeal, Lord Bingham of Cornhill C.J. said that three tests required to be satisfied: the duty test, the interest test and the circumstantial test: [1998] 3 W.L.R. 862, 899D-G. At the end of the judgment, at p. 911E, he said that the duty and interest tests were, in general, satisfied in this case but that the court could not regard the circumstantial test as satisfied. In the last paragraph of the judgment, at p. 912A, he said:

"Given the nature, status and source of the defendants' information, and all the circumstances of the publication, this was not in our judgment a publication which should in the public interest be protected by privilege in the absence of proof of actual malice."

The question is whether, in its formulation of the circumstantial test, the court went further than it ought to have done in defining the circumstances of the occasion by introducing into that test matters of fact which might be thought to be relevant to the issue of malice - indicating abuse of the occasion - rather than to the question whether the occasion itself was privileged.

The generic privilege

The occasion for which the appellants seek to be accorded the benefit of a generic qualified privilege was the publication in an edition of the "Sunday Times" newspaper circulating in the United Kingdom of an article relating to the resignation of Mr. Albert Reynolds, who had just resigned as Taoiseach in the Irish government, and the collapse of his coalition government. Mr. Reynolds claims that passages in that article meant and were understood to mean that he had deliberately and dishonestly lied to the Dáil by suppressing information which he possessed about the suitability for promotion of the Irish Attorney-General whose appointment to the Presidency of the High Court of Ireland he was promoting, and that he deliberately and dishonestly misled his coalition cabinet colleagues by withholding that information from them and lying about when the information came into his possession.

The generic privilege for which the appellants contend was formulated in various ways by Lord Lester. But in essence his submission was that it should extend to any discussion of a governmental or political matter affecting the people of the United Kingdom. He made it clear that the privilege for which he contended was intended to apply only in respect of criticism of political conduct and not to private conduct. The theme which he stressed throughout was that the justification for the generic privilege was that it was necessary in a modern democratic society, in view of the strong public interest in free speech in general and in freedom of expression on political issues on the press and other sections of the media in particular.

An examination of this issue must start from familiar first principles. The foundation of an action of defamation is malice. If words are used which are defamatory and untrue the law implies malice. That presumption is rebutted if the occasion when the words were used is privileged. The privilege destroys the presumption. But it remains open to the claimant to prove that there was malice in fact. At the heart of the matter is the question whether "the occasion" is privileged. This occurs where the person who makes the communication has an interest or duty to make it to the person to whom it is made, and the person to whom it is made has a corresponding interest or duty to receive it: Adam v. Ward [1917] A.C. 309, 334, per Lord Atkinson. The interest or duty may be a legal or moral duty or it may arise from social circumstances. But it is essential that there should be a reciprocity of duty and interest as to the matter which is being communicated. As Lord Atkinson pointed out, the communication is often loosely described as privileged. But strictly speaking it is the occasion itself which is privileged.

It is important not to lose sight of this point as we examine the issues raised by this case. It is essential to a proper understanding of the structure of this branch of the law. The privilege is given not to communications nor to the people who communicate them, but to the occasion. No individual or organisation, such as a newspaper or any other section of the media, can assert that it is entitled to the benefit of qualified privilege simply because of who or what that individual or organisation is or what it does. It is the occasion of the communication which must be examined, to see whether there was an interest or duty to make it and a corresponding interest or duty to receive it, having regard to its particular subject matter.

The application of these principles to particular facts and circumstances may show that there are some occasions of qualified privilege which can be regarded as falling into a recognisable group or category. Cases of that kind may be regarded as attracting what has been described as a generic common law qualified privilege. All occasions falling within that group or category will be treated as occasions of qualified privilege, and proof of actual malice will always be required before the words used can be held to be defamatory. This assists free speech and full and frank disclosure of the facts. It removes the inhibiting, or "chilling," effect which the law of defamation imposes on the discussion of matters of public interest. As a general rule it is beneficial and in the public interest that communication between parties with the necessary reciprocal duty and interest in the matter should not be inhibited.

But there is another general rule. The circumstances in which the common law defence of qualified privilege will be applied have always been defined broadly. In Perera v. Peiris [1948] AC 1, 20 Lord Uthwatt, giving the judgment of the Privy Council, said that their Lordships preferred to relate their conclusions to the wider general principle which underlies the defence of privilege in all its aspects rather than to debate the question whether the case fell within some specific category. This approach recognises the fact that the question is ultimately one as to striking the right balance between competing interests. In order to achieve this the primary concern of the law must be to maintain its flexibility. The advantages of certainty, which is the product of recognising that cases falling within a certain class or category will always attract qualified privilege, must be measured against the disadvantages which tend to flow from rigidity.

I think that three factors are relevant to the issue as to whether a generic qualified privilege can be recognised. The first relates to the precision with which the category can be described. The second relates to the persons to whom the material is to be communicated. The third relates to the issue of malice. As the only protection left against the damaging effect of communicating defamatory false statements is proof of malice which will remove the privilege, care should be taken not to give the benefit of the privilege too readily to persons or organisations whose sources of information are themselves protected to an extent which renders the issue of malice inscrutable.

If the category cannot be described precisely, it will be at risk of enlargement or erosion case by case and thus of losing touch with the underlying justification for the creation of the category. Where imprecision is unavoidable, the better course would seem to be to take each case on its own facts and circumstances. If the category is of a kind where the communication is made to a particular person or group of persons, and not to the public generally, it may be thought that the advantages of precision outweigh those which come with flexibility. The consequences to the person who is the subject of the communication are likely to be less serious than they would be if the defamatory statement of fact is published generally. But where the category involves communication to the public, the question must be whether the public interest in the receipt of the information will always outweigh the general public interest in protecting the reputation of the individual. This is a question which is particularly sensitive to changing circumstances, whether they be social or political, and to changes in the way in which information is presented or disseminated. As for the issue of malice, the less open the communicator is to scrutiny, the more important it is likely to be to retain the benefits of flexibility. Qualified privilege, in other words, should not be given to a category where the occasion of the communication is such that the privilege is at risk of becoming, in practice, absolute.

Against that background I regard the election cases as providing the most useful starting point for an examination of the authorities. This is a clearly recognisable group of cases, as to which the limits of the application of a generic qualified privilege can be readily identified. Lord Lester's argument is that the temporal and geographical limits which have been laid down in these cases are out of date. He said that the law should now recognise that there is a point of principle in them which should be applied more generally. I think that there is much force in that argument. But is first necessary to examine the cases to see what that principle is, and whether the limits which have been set by these cases for the application of the qualified privilege are indeed too narrow and should now be modified.

In Duncombe v. Daniell (1837) 8 C. & P. 222, the defendant was a voter in a parliamentary election. He wrote two letters which were published in a newspaper, the "Morning Post," which reflected upon the character of one of the candidates in his constituency. The plaintiff was awarded damages, whereupon the defendant applied for a new trial. One of the grounds for the application was that it was justifiable for an elector bona fide to communicate to the constituency any matter respecting a candidate which he believed to be true and believed to be material to the election. The application was refused. In the course of the argument Coleridge J., at p. 229, said that the defendant had to go further than that and show that the elector was entitled to publish it to all the world, as the publication was in a newspaper. Counsel for the defendant submitted that if no more was done than was necessary to make the matters known to the electors the publication was privileged, and that whether or not anything more was done was a question for the jury. Giving judgment Lord Denman C.J. said at p. 229:

"However large the privilege of electors may be, it is extravagant to suppose that it can justify the publication to all the world of facts injurious to a person who happens to stand in the situation of a candidate."

The same point was raised in several Scottish cases towards the end of the nineteenth century. Two of these cases concerned statements made by individuals about the fitness for office of candidates who were seeking election at the time when the statements were published: Anderson v. Hunter (1891) 18 R. 467 and Bruce v. Leisk (1892) 19 R. 482. Two of them concerned anonymous letters, one about a candidate for election, the other about a public official, which were published in local newspapers: Brims v. Reid & Sons (1885) 12 R. 1016 and McKerchar v. Cameron (1892) 19 R. 383. It is worth mentioning these cases, as they contain some observations which are relevant to the issue which we have to decide.

In Anderson v. Hunter the pursuer was seeking election as a county councillor for a division of the county of where a parish had been divided into two electoral divisions for county council purposes. The defender lived in the same parish but he was an elector in the other division. He had made various statements to people in the parish that the pursuer was not fit to be elected as he would soon be bankrupt. His argument was that the statements were made in circumstances that were privileged, as he was a ratepayer in the parish and the pursuer was a candidate for the public post of county councillor on one of the divisions of that parish. It was rejected, simply on the ground that he was not a voter in the election with reference to which he was said to have made the statements complained of.

In Bruce v. Leisk the defender was an elector in the same ward of the burgh in which the pursuer was seeking election as a councillor, so the geographical problem which was the basis of the decision in Anderson v. Hunter did not arise. He was also a member of the ward committee appointed by the ratepayers to recommend suitable candidates for election. The Lord Ordinary, Lord Stormonth Darling, said, at p. 484 that even if the defender were only an elector, it seemed to him that the case was one of privilege and that the pursuer must prove malice. In his view it was contrary to public policy to deny electors latitude in discussing the qualifications of those who were standing for election. His decision was reclaimed to the Inner House, which upheld his decision that the action should be dismissed as the pursuer was not willing to aver malice. Lord President Robertson said, at p. 485; that it was clear case for connecting the language used with the fulfilment of a public duty, and that when electors are considering who shall be elected they are quite entitled to state to other people, similarly concerned, what they know, or believe they know, on the question whether or not the person should be elected. Lord Adam said at pp. 486-487:

Anderson v. Hunter

Lord Kinnear was also of the opinion that the occasion was privileged, as it was clear from the pursuer's own statement that the words of which he complained were uttered when the defender was engaged in the exercise of a public right, with a view to the performance of a public duty.

Non-disclosure of its sources by a newspaper was the issue in Brims v. Reid & Sons. In that a newspaper had published an anonymous letter concerning the fitness for office of the pursuer who was seeking re-election as a member of a town council and to the public office of Dean of Guild. The publisher refused to disclose the name of the writer of the letter which he had published in his newspaper. It was held that he could not plead privilege in action to the pursuer's action of damages. Lord President Inglis gave his reasons, at p. 1020; in a passage from which is worth quoting at some length, as it covers a number of the issues raised in the present case in the course of the argument:

"It appears to me that, whatever might be the case if these statements had been made in an editorial article, about which I give no opinion, the fact that they were made in an anonymous letter is quite sufficient for the decision of this case. It is difficult to define the exact extent of the privilege of comment which the editor of a newspaper undoubtedly has to some extent upon the doings of public men; it is difficult to define what the class of public men is with reference to whose doings he enjoys that privilege, or what the kind of accusations that may be brought against the conduct of public men is; and yet again it is difficult to distinguish between the doings of a public man, as a public man, and as a private individual.

"But we are relieved of all these difficulties in the present case by the fact that the statements complained of are contained in an anonymous letter to the editor. The editor has declined to disclose the author. The effect of this in point of law is not to entitle this letter to be dealt with as if it had appeared in a leading article or in some part of the paper in which the editor speaks for himself. The law is that the editor accepts the position of the anonymous writer with every liability which could have been laid upon that writer if he had been disclosed. The question, then, is whether malice would require to be put in issue against the writer if he had been disclosed.

umbra

"The newspaper editor can be in no better position then than the anonymous writer himself. Now, if the letter was written with malice, it is conceded that the pursuer is entitled to damages. But how can anyone prove malice on the part of a person of whom he knows nothing at all? What can he tell of his state of mind, or his relation to the matter on which he comments? Or how, on the other hand, can malice in such a case be disproved?"

Lord Shand made it clear, at p. 1021; that, if the question had arisen with reference to editorial comments in a leading article about the conduct of a public man seeking re-election to office on the eve of the election, he would have been in favour of the view that the occasion was privileged. But the writer of an anonymous letter could not be given the benefit of qualified privilege, and the editor of a newspaper could not, by adopting the letter, invest the writer with the privilege which might have attached to his own articles.

McKerchar v. Cameron was another case involving an anonymous letter published in a local newspaper. The letter contained statements indicating that the pursuer, who was a salaried official, was unfit for his post as a teacher in a public school. It was argued that the ratepayers and inhabitants of the neighbourhood had an interest and a right to know the contents of what was published, but the decision in Brims v. Reid & Sons was followed. It was held that there was no room for the defence of privilege, so there was no need for the pursuer to plead malice. The court did not need to decide whether a member of the public, in attacking any person holding any office under any public body, was entitled to the defence of privilege. But Lord McLaren observed; at p. 386; that it was difficult to see what duty or right there was on the part of a member of the public, as such, to criticise the conduct of a public servant who was in the public employment.

These cases indicate that the extent of the qualified privilege in relation to discussion of the public conduct of public officials, and especially of those who were seeking election to a public office or re-election, was already the subject of a vigorous debate one hundred years ago. Various strands of thought can be detected. It seems unlikely that the Scottish courts, by the end of the nineteenth century, would have taken the same line as was taken in Duncombe v. Daniell 8 C. & P. 222 where the privilege was held not to be available to an elector who published his statements under his own name in a newspaper. It seems to be implicit in the two Scottish cases which I have mentioned about the publication of anonymous letters by newspapers that the writers of those letters would have been able to argue that they were entitled to the defence of privilege if their names had been disclosed and they had been sued. The newspapers were small circulation, local newspapers. But at least it can be said that these cases were not decided on the narrow ground that publication in a newspaper was in itself enough to rule out the question of privilege.

As for the classes of persons by whom and about whose conduct comment might be made with the benefit of privilege, the criterion which was being applied was whether they were electors and candidates in the same electoral ward, district or constituency. But there are signs, particularly in the opinion of Lord President Inglis in Brims v. Reid & Sons, of a recognition that people in public positions generally, in regard to their conduct as such, were in a different position from private individuals. The Lord President referred to some of the difficulties in defining the class which have been raised in the present case. But he did not say that that was a fruitless exercise because comment of that kind could never attract the privilege.

I think that the geographical and temporal limitations which are apparent from the election cases provide a good illustration of the kind of situation which will attract a generic qualified privilege: cases falling within clearly defined limits, within which the elements of duty and interest in the publication of relevant matter will always be found. But that is not to say that there will not be other cases - of which the public conduct of public persons, especially those holding or aspiring to an elected political office, is the clearest example - where the privilege will be available. The difficulty as to these cases is one of definition, not one of principle.

Developments in regard to recognition of the fundamental right of free speech and to the nature of the electoral process since the end of the nineteenth century have reinforced the arguments in favour of the wider availability of the qualified privilege to those who publish material to the general public on matters of general public interest. There are powerful dicta to the effect that there is no inconsistency between article 10 of the European Convention for the Protection of Human Rights and Fundamental Freedoms and the English common law on freedom of speech: see Attorney-General v. Guardian Newspapers Ltd. (No. 2) [1990] 1 AC 109, 283 per Lord Goff of Chieveley. But there can be no doubt that the incorporation of the Convention into English law by the Human Rights Act 1998 has strengthened the arguments in favour of the principles which are set out in that article.

In Braddock v. Bevins [1948] 1 K.B. 580 it was recognised that it was necessary for the welfare of society that there should be a frank exchange of information and opinions on matters relating to the exercise of the franchise by the electorate. Since that decision the width and subject matter of that exchange have been enlarged still further by a greater concentration upon the parties rather than on individuals in the electoral process. The growth of public opinion polls, both during election campaigns and between elections, has tended to shift attention towards the performance of the parties relative to each other throughout the entire calendar. The public conduct of leading politicians is now seen as the embodiment of a party's performance and credibility. Recent developments in the method of electing candidates through party lists in the elections for the Welsh Assembly, the Scottish Parliament and the European Parliament have added to that development. These developments show that the case law which confined the privilege to comment on individual candidates at election time and to the electoral process within their own constituencies has become outdated. They support the argument, in a wider public interest, for the wider availability of the defence of qualified privilege.

But the question remains: should we now recognise a common law generic qualified privilege for political discussion? On balance I am of the opinion that this would not be satisfactory, bearing in mind the nature of the occasion and the use which would be likely to be made of it. It may be difficult to achieve a satisfactory definition of the category which will eliminate the risk of its being applied more widely to discussion about people in public life generally. A category which went that far was not asked for in this case, and I would regard it as unacceptable. But the greater risk is of defamatory statements of fact being communicated to a wide audience, based upon information communicated to the media by sources which those who publish the information must protect and consequently will not be revealed to the individual. The balance is a delicate one, as there are powerful arguments in favour of the constitutional right of free speech and, where politicians are involved, the interest and duty tests are likely to be satisfied in most cases without too much difficulty. But the importance which must be attached to the principle which justifies the protection of their sources by the media - which has an essential part to play in the role of the media in a free and democratic society - carries with it certain penalties.

One of these, I believe, is the discipline of having to justify each claim to the benefit of qualified privilege should the statements of fact which are made by the media turn out to be defamatory. The description of this discipline as having a "chilling" effect on free speech, as if this in itself shows that something is wrong with it, is too simple. Of course, it does "chill" or inhibit the freedom of the communicator. But there are situations in which this is a necessary protection for the individual. The first line of protection is removed, if the occasion justifies it, by the defence of qualified privilege. Proof of malice, which is the second line of protection, is likely to be very difficult, if not impossible, if the sources of the information cannot be identified. Taken on a case by case basis, the risk that this will be so is one which can be accepted as being in the public interest and therefore justified. But I would be unwilling to extend that risk to political comment generally. I would decline to recognise in this area of our public life a generic qualified privilege.

On this aspect of the case therefore I too am in full agreement with the speech of my noble and learned friend Lord Nicholls of Birkenhead.

The circumstantial test

As I see it, the application of this third test to the facts of this case raises an issue about the taxomony, or structure, of the common law relating to qualified privilege. There is no doubt that the Court of Appeal broke new ground when it identified this as an additional test which had to be satisfied in relation to any individual occasion when applying the law of qualified privilege. I do not see this, in itself, as a basis for criticising what was, on any view, an admirable, forward-looking and imaginative judgment. Initiatives of this kind are part of the life-blood of the common law. We all benefit from the constant process of adjustment and refinement as one case follows upon another and new problems reveal how the law can be explained better or further clarified.

The difficulty is, perhaps, more one of detail rather than of principle. In the past it has always been necessary to consider the circumstances in order to decide, as a matter of law, whether the interest and duty tests were satisfied. These are not abstract concepts. The occasion has to be identified, because it is the occasion which attracts the qualified privilege. To identify the occasion one must examine the nature of the material, the persons by whom and to whom it was published and in what circumstances. It may be necessary to resolve some questions of fact before the issue of law can be addressed as to whether the occasion was privileged. But the point is that if the issue of law is resolved in favour of the publisher and the argument is then taken against him that because he has misused the occasion he has lost the benefit of the privilege, further questions of fact will be raised. They too will involve a consideration of the circumstances. But it does not follow that the circumstances which will be relevant at this stage of the inquiry will be the same as those which were relevant to the question whether the occasion was privileged. On the contrary, they are likely to be different, as the question which must be answered at this stage is a different question.

As Lord Diplock explained in Horrocks v. Lowe [1975] A.C. 135, 149:

"With some exceptions which are irrelevant to the instant appeal, the privilege is not absolute but qualified. It is lost if the occasion which gives rise to it is misused. For in all cases of qualified privilege there is some special reason of public policy why the law accords immunity from suit - the existence of some public or private duty, whether legal or moral, on the part of the maker of the defamatory statement which justifies his communicating it or of some interest of his own which he is entitled to protect by doing so. If he uses the occasion for some other reason he loses the protection of the privilege."

Lord Diplock then reviewed the various situations in which it may be proved against the publisher that there was some other dominant and improper motive on his part. The commonest case is where the dominant motive is not to perform the relevant duty or to protect the relevant interest but to give vent to a personal spite or ill-will towards the person defamed. This may be proved by direct evidence or by inference from the circumstances.

The test of malice; if I may paraphrase Brett L.J., as he then was, in Clark v. Molyneux (1877) 3 QBD 237 at 247 is: has it been proved that the defendant did not honestly believe that what he said was true, that is, was he either aware that it was not true or indifferent to its truth or falsity. It was contended in Horrocks v. Lowe that the inference of malice should be drawn from the contents of the speech, the circumstances in which it was made, the defendant's failure two days later to apologise and the evidence which he gave in the witness box. But in the end the judge, who was sitting without a jury, declined to draw the inference that the defendant was indifferent to the truth or falsity of what he said.

In my opinion that case shows that the question of malice also involves an examination of the circumstances. But there is this difference. The question whether the occasion was privileged is a question of law for the judge. The question whether the occasion was abused because of malice is a question of fact which, if the trial is by jury, the jury must decide. This separation of function is clearly identified in the Scottish cases which I mentioned earlier. They were all decided on the preliminary issue of law as to whether the occasion was privileged. In each case the pursuer was seeking the approval of issues which were to be put before the jury at a jury trial. They were unwilling or unable to allege malice, so the whole question turned on the issue of qualified privilege. If the defence was upheld and malice was not averred there was no issue which could be put to the jury. In the United States of America the same separation of function is to be found in paragraph 619 of the American Law Institute, Restatement of the Law (1977), Torts 2d, Ch 26:

"(1) The court determines whether the occasion upon which the defendant published the defamatory matter gives rise to a privilege.

"(2)Subject to the control of the court whenever the issue arises, the jury determines whether the defendant abused a conditional privilege."

I think that the circumstantial test tends to obscure this difference of function and, perhaps even more importantly, to obscure the difference between questions which go to the question of malice and the question whether the occasion was privileged. It is too widely formulated. It includes "the nature, status and source of the material, and the circumstances of the publication" without any qualification as to the purpose of examining this evidence [1998] 3 W.L.R. 862, 899G. It has had the effect in this case of introducing, at the stage of examining the question of law whether the occasion was privileged, assumptions which I think are relevant only to the question of fact as to the motive of the publisher: as where it is said that it is one thing for him to publish a statement taken from a government press release or the report of a public company chairman or the speech of a university vice-chancellor, and quite another to publish the statement of a political opponent or a business competitor or a disgruntled ex-employee: p. 909H-910A. In its application to the facts of this case, it has introduced questions as to the use of sources, as to a failure to publish Mr. Reynold's own account of his conduct, as to the appellants' failure to alert him prior to the publication of their conclusion that he had lied to his coalition colleagues and knowingly misled the Dáil so as to obtain his observations on it: p. 911F-H. In my opinion these considerations go to the question whether the appellants abused the occasion. This is a question of fact for Mr. Reynolds to establish upon a review of all the evidence. They do not go to the question whether the occasion itself was privileged.

In my opinion the circumstantial test is confusing and it should not be adopted.

Conclusion

I consider that the Court of Appeal were wrong to hold at p. 912A, as a matter of law, that in the light of the issues which they considered in their application of the circumstantial test the publication was not protected by qualified privilege. Although there is plainly a question as to whether the occasion was abused, I would hold that the prior question as to whether the occasion itself was privileged has not been properly addressed. It seems to me still to be an open one.

I would allow the appeal. In my opinion the question of law as to whether the occasion was privileged should be reconsidered by the judge at the new trial.

LORD HOBHOUSE OF WOODBOROUGH

My Lords,

I agree that this appeal should be dismissed for the reasons given by my noble and learned friend Lord Nicholls of Birkenhead. Like my noble and learned friend Lord Cooke of Thorndon, I am in full agreement with the speech of Lord Nicholls. The few words which I will add should not be read as in any way detracting from the clarity of that agreement.

This case is concerned with the problems which arise from the publication of factual statements which are not correct--i.e. do not conform to the truth. This case is not concerned with freedom of expression and opinion. The citizen is at liberty to comment and take part in free discussion. It is of fundamental importance to a free society that this liberty be recognised and protected by the law.

The liberty to communicate (and receive) information has a similar place in a free society but it is important always to remember that it is the communication of information not misinformation which is the subject of this liberty. There is no human right to disseminate information that is not true. No public interest is served by publishing or communicating misinformation. The working of a democratic society depends on the members of that society being informed not misinformed. Misleading people and the purveying as facts statements which are not true is destructive of the democratic society and should form no part of such a society. There is no duty to publish what is not true: there is no interest in being misinformed. These are general propositions going far beyond the mere protection of reputations.

The law of civil defamation is directly concerned with the private law right not to be unjustly deprived of one's reputation and recognises the defence of privilege. The justification for this defence is at least in part based upon the needs of society. It can sensibly be asked why society or the law of defamation should tolerate any level of factual inaccuracy. The answer to this question is that any other approach would simply be impractical. Complete factual accuracy may not always be practically achievable nor may it always be possible definitely to establish what is true and what is not. Truth is not in practice an absolute criterion. Nor are the distinctions between what is fact and innuendo and comment always capable of a delineation which leaves no room for disagreement or honest mistake. The free discussion of opinions and the freedom to comment are inevitably liable to overlap with factual assumptions and implications. Some degree of tolerance for factual inaccuracy has to be accepted; hence the need for a law of privilege.

There is another aspect of the law which needs to be identified. Save as provided in section 1 of the Defamation Act 1996, any publication of a defamatory statement exposes the publisher of that statement to tortious liability. This is so whether or not he is the originator of the statement or is simply republishing what someone else has said. This rule is relevant to the defence of privilege and the media. Journalists very often have no personal knowledge of the truth or falsity of the facts which they report and publish. Typically they are reporters of material derived from others. The character of the source is relevant to the kind and the extent of the privilege which should be afforded to the publisher. For example, privilege attaches to the reporting of legal proceedings and of the evidence given to and the findings of Inquiries. It is in the public interest that the public should be informed about such matters and this is so even if some of what has been said during such proceedings may not have been true. But the same cannot be said of casual gossip overheard by a journalist; there is no public interest in its repetition unless it be factually true. Between these two extremes there is a spectrum of possible circumstances.

To attract privilege the report must have a qualitative content sufficient to justify the defence should the report turn out to have included some misstatement of fact. It is implicit in the law's insistence on taking account of the circumstances in which the publication, for which privilege is being claimed, was made that the circumstances include the character of that publication. Privilege does not attach, without more, to the repetition of overheard gossip whether attributed or not nor to speculation however intelligent.

The decided cases confirm both the recognition of the element of public interest in the law of privilege (e.g. Perera v. Peiris [1949] AC 1) and the limits within which it must be kept (Blackshow v. Lord [1984] Q.B. 1; see also Truth (N.Z.) Ltd. v. Holloway [1960] N.Z.L.R. 69). The publisher must show that the publication was in the public interest and he does not do this by merely showing that the subject matter was of public interest. The decided cases also show that, anyway in English law, the doctrine of express malice does not provide an adequate safeguard. It is a very narrow doctrine as explained by Lord Diplock in Horrocks v. Lowe [1975] A.C. 135. The plaintiff has to prove that the publisher did not have an honest belief in the truth of what he was publishing: "the law demands no more" (p.150E). The subjective character of this criterion makes the plaintiff's burden of proof one which it is difficult to discharge in all but the most blatant cases. It is also inadequate to meet the objective requirements of a satisfactory law of privilege. Both in England and in other countries there have been statutory interventions which affect the structure of this part of the law. In New South Wales this is manifestly so and in New Zealand the statutory definition of malice in section 19 of the Defamation Act 1992 clearly has to be taken into account. In England the provisions of the Defamation Act 1996 take the form of providing the media with additional special defences and therefore do not provide a justification for introducing the further modifications of the existing law for which the appellants have contended.

As your Lordships agree, there is no generic privilege. There are reasons of principle and practical reasons for this. No genus is satisfactory, nor is any genus more satisfactory than the criterion of what it is in the public interest that the public should know and what the publisher could properly consider that he was under a public duty to tell the public. It is clearly established in English law that the duty/interest test is not confined to private duties and interests. The public dimension recognised by the law encompasses in a satisfactory and adaptable manner those types of publication to which privilege should attach. Any generic category will tend to be both too wide and too narrow. It will fail to take account of the differing character and circumstances of the publications which may fall within it. It will fail to afford privilege to publications which fall outside its definition but are equally deserving of privilege.

Your Lordships were urged by the appellants to endorse an approach which would leave it to publishers to decide whether or not to publish and to uphold their privilege to do so save where the plaintiff can prove actual bad faith on the part of the publisher. Such an approach would of course be attractive to the media but it would be handing to what are essentially commercial entities a power which would deprive the subjects of such publications of the protection against damaging misinformation. Such persons and the public are entitled to the disinterested and objective involvement of the law. It is for the publisher to establish to the satisfaction of the law that the publication was privileged. It is only once the publisher has done this that a burden of proof passes to the plaintiff. As previously stated, the burden of proof which the plaintiff then has to discharge is not a light one.

There are advantages for the media in the present state of the law as the experience of the United States of America subsequent to the Sullivan case (376 U.S. 254) has shown. The present law is consistent with the publisher being able, if he so chooses, to preserve the confidentiality of his sources. The burden of proving circumstances justifying privilege is upon the publisher. Whether or not he chooses to disclose his sources in order to assist him to do so is (in general) a matter for him. If on the other hand there is some generic privilege which without more confers privilege, the aggrieved party must in justice be able to obtain discovery of all the relevant facts and documents to enable him to displace that privilege. This is what has happened in the United States. The trade-off for the more extensive defence has been the requirement of full disclosure by way of extensive and onerous pretrial discovery.

I agree with Lord Nicholls that the circumstances of publication have to be taken into account in determining whether any particular publication was privileged. This, as the authorities he cites show, is an established part of English law. The criticism to be made of the Court of Appeal judgment is that it sought to treat the circumstances as a separate and distinct element. This was unnecessary and mistaken. But the substance of the judgment must be upheld. The Court of Appeal also reached the right conclusion upon the application of the law to the essentially undisputed facts of the present case. There is no justification for allowing the defendants to reopen that aspect of the case on the retrial.