B e f o r e :
LORD JUSTICE WARD LORD JUSTICE SEDLEY and LORD JUSTICE LONGMORE ____________________
____________________
(Transcript of the Handed Down Judgment of Smith Bernal Reporting Limited, 190 Fleet Street London EC4A 2AG Tel No: 020 7421 4040, Fax No: 020 7831 8838 Official Shorthand Writers to the Court)
Mr. M. Tugendhat QC and Mr. R. Parkes (instructed by Farrer & Co.) for the Appellant Mr. C. Hollander QC (instructed by Simmons and Simmons) for the Respondents Mr D. Mayhew (Solicitor Advocate) sitting for Interested Party – Financial Services Authority Mrs Maryann Mahon & Mr Kelwin Nicholls (instructed by Clifford Chance) for Reuters ____________________
HTML VERSION OF JUDGMENT AS APPROVED BY THE COURT ____________________
Crown Copyright ©
Lord Justice Sedley:
This appeal
On 19 December 2001 Lightman J, on an interlocutory application, granted a Norwich Pharmacal order against the five defendant news media requiring them to preserve and within 24 hours to deliver up to the claimant their original copies of a leaked and (as the judge found) partially forged document about a contemplated takeover by the claimant, a major brewer based in Belgium, of South African Breweries.
The first limb of the order is uncontroversial: Mr Michael Tugendhat QC for the defendants points out that the issue of the proceedings will anyway have made it a contempt not to preserve the documents. The dispute is about the requirement of immediate delivery up, which is stayed pending this appeal by the five defendants. The claimant does not cross-appeal against the judge’s dismissal of its claim in breach of confidence. That element therefore leaves the stage, although its voice will be heard occasionally from the wings.
It is the defendant appellants’ contention that the outcome of the appeal is likely to be directly affected by the House of Lords’ decision in Ashworth Hospital v MGN Ltd [2001] 1 WLR 515, CA. The appeal, for which their Lordships have given leave, is to be heard later this term. Although the claimant submits that it is not capable of affecting the issues before us, it initially seemed to us impossible to be certain that this was so, and undesirable to come to a conclusion which, if it went against the defendants, might be falsified when it was too late to reverse the outcome. The opposite risk of the trail going cold seemed overstated, although the difficulty for the claimant of preparing a fresh bid with the risk of a further leak was real. But in the event our decision to stand the present appeal out had to be reconsidered when it became known that their Lordships might not be able to give their decision in the Ashworth appeal until the summer. In these circumstances the balance of convenience moved in the claimants’ favour, and we proceeded to hear out the appeal. The basis on which we have done so has therefore been that we are bound by the Ashworth decision. For reasons which follow, however, there is nothing particularly dramatic in this.
The facts
A full account of the relevant events is set out in Lightman J’s judgment, from which I gratefully adopt the following paragraphs:
The law
The Contempt of Court Act 1981, s.10
Section 10 of the Contempt of Court Act 1981 provides:
It will be observed that this provision creates no power or right of disclosure: what it does is assume the existence of such a power or right and place a strong inhibition on its exercise. It governs material received with a view to publication, whether published or not: see X Ltd v Morgan-Grampian Ltd [1991] 1 AC 1, 40, per Lord Bridge.
This court in Ashworth followed the now dominant line of authority which attributes a more catholic meaning to the phrase “the interests of justice” in s.10 of the Contempt of Court Act 1981 than was initially given to it in Secretary of State for Defence v Guardian Newspapers [1985] AC 339, 350, where Lord Diplock limited it to the technical interests of the administration of justice. By common consent our approach is that of Lord Bridge in X v Morgan Grampian [1991] 1 AC 1, 43: the phrase is large enough to include the exercise of legal rights and self-protection from legal wrongs, whether or not by court action.
The genesis of s.10 is of some interest and relevance. It was introduced into the Contempt of Court Bill in committee by Lord Scarman, following the decision of the House in November 1980 in British Steel Corporation v Granada Television [1981] AC 1096, in which the majority of their Lordships acknowledged, but only as a contingent discretionary factor, the public interest in the protection of journalists’ sources. Lord Scarman, quoting Lord Salmon’s dissenting speech in favour of much stronger protection, explained that his two proposed exemptions from a general privilege for journalists’ sources - national security and the prevention of disorder or crime - were derived from article 10(2) of the European Convention on Human Rights. He withdrew his amendment upon the promise of Lord Hailsham LC to consider the issue, and Lord Morris thereafter introduced what is now s.10, spelling out the interests of justice as a further ground of exception. It should not be forgotten that in this country, then as now, the principal source of unattributable leaks to the media – in the form of off-the-record briefings - and therefore the principal beneficiary of a rule protecting the secrecy of sources, was government itself (see Peter Hennessy, Whitehall (1989) p.363-4).
The European Convention on Human Rights
Article 10 provides:
In Camelot Group plc v Centaur Communications Ltd [1999] QB 124 , 128, Schiemann LJ remarked that s.10 “was passed in order that our domestic law might reflect article 10 …”. While this is certainly true of the exceptions in Lord Scarman’s original amendment, and while the interests of justice may correspond with – though they clearly go beyond – the further purpose of maintaining the authority and impartiality of the judiciary, the protection of journalistic sources was not at that date a recognised purpose of art. 10. It was only with the decision of the European Court of Human Rights in Goodwin v United Kingdom (1996) 22 EHRR 123 (the sequel to X Ltd v Morgan-Grampian Ltd ), building on Jersild v Denmark (1994) 19 EHRR 1, that the jurisprudence of the Court drew out of art. 10 the principle proleptically incorporated in s.10 of the 1981 Act. It did so, it is worth noting, not by extracting from paragraph (1) of the article the words “freedom …to receive … information … without interference by public authority” and conferring this on the press rather on its readers, but by drawing out of the article’s overriding purpose a protection without which the purpose itself would be subverted.
Lord Scarman, the first judicial advocate of patriation of the Convention, remarked in his dissenting speech in the Guardian Newspapers case in 1984:
He went on to point out the structural similarity between s.10 and the qualified Convention rights. That there is, however, a difference of content or of emphasis between them was exemplified in the decision of the European Court of Human Rights in Goodwin (above). The House of Lords had concluded that the need to prevent an unknown source from disseminating commercially damaging information called for the lifting of the bar on disclosure under s.10, notwithstanding that an injunction had been obtained in time to prevent publication; but the Court considered that the injunction removed much of the justification for lifting the bar, and that the remaining risks were not sufficient to make disclosure a proportionate measure under art. 10(2).
The Ashworth case
In the Ashworth case (above), which concerned the unauthorised disclosure of a patient’s medical records to the press, Lord Phillips MR considered the narrower meaning of “the interests of justice” adopted by Lord Diplock in Secretary of State for Defence v Guardian Newspapers Ltd (above, at 350), which confined the phrase to its technical sense of court proceedings, as against the wider meaning of the exercise of rights and self-protection against wrongs preferred by Lord Bridge in Morgan-Grampian (above, at 43). His judgment, which draws extensively on the judgment of the European Court of Human Rights in Goodwin (above), includes these passages:
Laws LJ, concurring in the result, added this:
The right to delivery up
There are varying rights to production of documents held by others, and differing purposes for which the right exists. The right which Interbrew asserts to production of the documents held by the respective defendants is derived from the decision of the House of Lords in Norwich Pharmacal Co v Customs and Excise Commissioners [1974] AC 133 : a person who, albeit innocently, facilitates the tortious act of another must co-operate in righting the wrong by disclosing the wrongdoer’s identity to the wronged party, and can be made by the court to do so if no other expedient is available. The basis of the newly asserted jurisdiction being the old equitable bill of discovery, the power does not run against a mere witness; but it runs in respect of equitable as well as common law wrongs.
Criminal sanctions
At the time of this leak the Financial Services Act 1986 provided by s. 47:
This legislation is now superseded by the Financial Services and Markets Act 2000 which by Part XI confers on the Financial Services Authority extensive investigative powers directed to the detection and prosecution of crimes, including market manipulation.
The issues
It is said by Mr Charles Hollander QC on behalf of Interbrew that the source, or somebody supplying the source, has plainly been guilty not only of a breach of confidence, probably within GS or Lazards, but of forgery and of market manipulation. By publishing the products of the first two wrongs and thereby promoting the third, the defendants have all laid themselves open to the remedy of delivery up and brought themselves within the statutory exceptions from non-disclosure of sources. The defendants, jointly represented by Mr Michael Tugendhat QC, contest each step in this reasoning.
In the light of the arguments addressed to us, we have to consider these questions:
i) Is the detection of crime a proper object of the Norwich Pharmacal power?
ii) Has an entitlement to delivery up been established?
iii) If so, is it blocked by s.10? In particular –
iv) What are the interests of justice which disclosure will serve here?
v) Is the prevention of crime engaged?
vi) Should the court treat the allegations of forgery and market manipulation as proved because they are uncontradicted?
vii) Is the necessity of disclosure affected by the source’s motive?
viii) Is it affected by an express promise of confidentiality?
ix) In what sense is the decision under s.10 a matter of discretion?
Lightman J initially described his approach to the last of these issues in this succinct way:
But the defendants criticise him for departing from this correct stance later on in his judgment by describing the exercise in terms of a rebuttable presumption, and by then saying (para. 32(vi)):
To this they now add the judge’s reasons for refusing permission to appeal. These begin: “The decision is or involves an exercise of discretion.” Aldous LJ, granting permission, commented: “I do not accept that the reasons for refusing permission to appeal given by the judge are correct.”
Not all the other questions were canvassed before Lightman J as they have been before us; but criticism is made of his reliance, in reaching his judgment, on the fact that the source had obtained no express promise of confidentiality from any of the defendants and that none could be implied from the circumstances, and of his refusal to – as he put it – second-guess the advisers who had concluded that the source could not be detected without access to the originals received by the defendants. For my part, and whatever our conclusions, I would wish to pay tribute to the care and thoroughness of the judge’s fact-finding and reasoning in a judgment produced under enormous pressure of time.
Discussion
Is the detection of crime a proper object of the Norwich Pharmacal power?
There are sound reasons, both historical and contemporary, for holding that the answer is no.
Mr Hollander points to the strong evidence that the source’s purpose was criminal under s.47 of the Financial Services Act 1986 and submits that if facilitation of tort exposes an innocent party to an order for delivery up, facilitation of crime must do so too. I do not accept that this follows, and nothing in either the reasoning of the House of Lords in Norwich Pharmacal or the earlier decisions on which it was based suggests that it does. Lord Phillips MR in Ashworth , paragraphs 55-66, considered the nature of the wrongdoing needed to attract the Norwich Pharmacal power and concluded that nothing in authority or in principle confined it to torts; but the argument there was whether it reached breach of confidence.
The point of the bill of discovery was that it permitted a party with a cause of action against a person whose identity he did not know to extract from an unwitting participant – that is, somebody who was involved but was not a wrongdoer - the identity of the wrongdoer. The purpose, manifestly, was to complete the cause of action (though the exemption of “mere” witnesses from the obligation to assist seems to me inexplicable as a matter of principle and may be why Lightman J spoke of the Norwich Pharmacal procedure as “a rule of practice, not a rule of law”). Even in the period before the mid-19 th century, when the majority of prosecutions were privately brought, the identification of criminals does not feature in the reported cases as a proper purpose of the bill of discovery. Today, when prosecution, subject to rare exceptions, is the task of the state, I can see no justification for introducing it. The prosecution of offences, notably in the field of financial services, is the business of specialised agencies equipped with statutory powers of search and seizure. Each of these powers, none of them unlimited, represents a carefully struck balance between the needs of the community and the rights of the individual. To undercut them with a civil right to compel production of documents or data at the instance of a person claiming to be the victim of a crime would be to court catastrophe. For what purpose would the documents be exigible? If prosecution, that will rarely be the applicant’s intent. Here, for instance, Interbrew say that they are entirely content to leave the question of criminal proceedings to the FSA. If the purpose is civil process, what is the relevance of the commission of a crime? If it has to be crime of which the applicant is a victim, there is almost bound to be a civil cause of action. If it were able to be a crime of which the applicant is not a victim, the limited Norwich Pharmacal purpose would be replaced by a practically untrammelled right to disclosure. It may appear paradoxical that once s.10 of the Contempt of Court Act 1981 is reached, the prevention of crime becomes an admissible reason for ordering disclosure of sources. I consider the effect of this below; but it cannot require the prevention of crime to be read back into the equitable right to disclosure which is the predicate of s.10. Parliament can override or modify the common law and equity, but it cannot determine what they are.
Mr Tugendhat has been able to show us authority and commentary which corroborates, if indeed it does not bindingly establish, this view of equity and the law. Bray on Discovery (1885, p.3) advances as an introductory proposition:
“The right of discovery exists only in aid of civil proceedings”.
In support of his corollary “Not in aid of an action founded on a felony” Bray cites first the decision of Lord Eldon LC in Cartwright v Green (1803) 8 Ves. Jun. 406 allowing a demurrer to a bill of discovery. It is clear from the report, however, that Lord Eldon acted on the ground that the defendants might otherwise be made to incriminate themselves, which is not quite the same thing. But in the earlier case of Montague v Dudman (1751) 2 Ves. Sen. 396, also cited by Bray, Lord Hardwicke had said:
Lord Hardwicke was adjudicating on a demurrer to a prayer for an injunction to stay a claim to a common law mandamus, and for discovery in aid of it; but the passage cited above is central to his reasons for allowing the demurrer as to discovery. Mr Anthony Walton QC, replying at the bar of the House in Norwich Pharmacal (at 166), therefore had good grounds for his submission that equity refused to aid the obtaining of evidence in relation to criminal proceedings or evidence which would result in self-incrimination. Nothing said in their Lordships’ speeches indicates a different view.
Private prosecutions are still permissible and from time to time prove their value as a longstop behind an inert public authority. I would want to leave open the situation where an applicant can show a genuine need or wish to bring a private prosecution but requires the respondent’s help in identifying the wrongdoer. It is not this case. In a case such as this, which is the ordinary case, I would hold that it is immaterial to the Norwich Pharmacal jurisdiction that the wrongdoer may have been guilty of a crime. What matters is that the applicant means to bring a civil action, or otherwise to assert its legal rights (see British Steel , above, per Lord Wilberforce at 1174), as soon as it knows who the correct defendant is.
I do not accept Mr Hollander’s submission, at least as a general proposition, that something less will do. I can accept that in the extreme circumstances confronting Sir Richard Scott V-C in P v T Ltd [1997] 1 WLR 1309 it was legitimate to exercise the power in a case where not only the name of the tortfeasor but the full nature of the suspected tort was unknown; but here Interbrew know enough to be able to show the court that they have been wronged. The decision of Neuberger J in Murphy v Murphy [1999] 1 WLR 282 intelligibly extends the principle from the identification of persons answerable for wrongdoing in which the applicant has an interest to the identification of trustees answerable for the management of funds in which the applicant has an interest. But that is all. Neither opens the door to a roving commission of inquiry into the authorship of crimes.
Has an entitlement to delivery up been established?
Interbrew must therefore establish that each defendant has facilitated a civil wrong committed by an unknown person against whom they aim to seek redress. This, in agreement with Lightman J, I consider they have succeeded in doing, though not without some effective opposition.
Mr Hollander submits that on the evidence before the court the source has committed against Interbrew at least the actionable wrong of breach of confidence. He is also liable to them (it is argued) for malicious falsehood and unlawful interference with Interbrew’s economic interests. If he worked with others he is liable civilly for conspiracy to injure. If he worked for Interbrew he is liable for breach of contract.
Working backwards through this litany of wrongs, there is no evidence that the source was employed by Interbrew and much to suggest that he was not. There is no evidence that more than one person was involved, and nothing therefore to indicate a conspiracy. Malicious falsehood and interference with economic interests require proof of damage in order to establish a tort, and I accept Mr Tugendhat’s submission that any such damage, on the evidence and findings outlined above, is entirely speculative. Interbrew cannot on the one hand assert that they were engaged in no more than routine monitoring and on the other simultaneously assert that their takeover bid has been thwarted; and even if they could, proof of consequential loss would still be a long way off. This is apart from the question, to which I will come, of establishing that a material part of the document is false.
That leaves breach of confidence. The source has passed on information which he must have known was confidential. But, Mr Tugendhat submits, this is only the beginning. First of all, Interbrew’s own case is that critical parts of the document were false, and there can be no confidentiality in false information. This Mr Hollander accepts. As to the rest, Mr Tugendhat submits that any confidentiality evaporated when Mr Powell, on the record and without qualification, spoke to the Financial Times – the first disseminator – about the document and admitted that research had been done, albeit routinely, on South African Breweries. This I do not accept. In particular I do not accept the contention that it was up to Mr Powell to assert confidentiality. Financial journalists of the calibre of those employed by these defendants were not born yesterday. They had in their hands an apparently authentic document about an impending bid by one of the world’s largest brewing conglomerates for a coveted company, delivered anonymously and disclosing things that Interbrew and its advisers plainly needed to keep secret. Mr Powell’s options when he learnt from Goldman Sachs that the FT had the document were therefore pretty limited. If he said and did nothing, the story would run in full. If he spoke off the record there would still be nothing to contradict it. By going on the record to accept routine market-watching but to deny any concrete plans, he was trying intelligibly to limit the harm; but all he achieved, and all he could probably in retrospect have hoped to achieve, was publication of his denial alongside the full story. I see no possible case of waiver or acquiescence here, unless it is to be the law that the victim of a leak is damned if he speaks to the media and damned if he does not.
To this relatively modest extent, therefore, I agree with Lightman J that a breach of confidence is made out. What is not in my view made out is the bigger and better cause of action in respect of the “lethal cocktail” of fact and falsehood, since the element of falsehood can neither form part of the protected confidence nor stand on its own as a discrete tort. Nor, it seems to me, is it material to the Norwich Pharmacal jurisdiction that Interbrew will be able to dismiss the source, if they find him or her within their organisation, for misconduct in falsifying as well as leaking documents. That is well within the ambit of the interests of justice on the broad interpretation of s.10 if and when one reaches it; but for the same reasons as are set out above in relation to the detection of crime, it can have no place in the jurisdiction to require disclosure in support of a cause of action.
From this point on I agree with Mr Hollander that each defendant, by disseminating the leaked information, has innocently lent itself to the wrongful purpose. None is a mere bystander or witness. All, not solely (as Mr Tugendhat submits) the Financial Times, disseminated the disclosure. It is unnecessary in these circumstances to consider Mr Hollander’s unpromising submission that Interbrew would be entitled to an order for delivery up even if the defendants had published nothing. If he were right, there would be a built-in incentive to publish regardless, for which enterprises like Interbrew would not, I suspect, be grateful.
Is entitlement to production blocked by s.10?
The purpose of s.10 of the Contempt of Court Act 1981 is to limit to the necessary minimum any requirement upon journalists to reveal their sources. It has now to be read and applied by our courts, so far as possible, compatibly with the Convention rights: Human Rights Act 1998, s.3(1). For reasons touched on earlier in this judgment, there should be no difficulty about this; but that is not to say that the Convention can simply be treated as background, for it and its jurisprudence may both amplify and modify the hitherto accepted meaning and effect of s.10. For present purposes the Convention right which is in play is the qualified right spelt out in art. 10. Mr Tugendhat’s late invocation of art. 8 seems to me to be a right too far: the right to respect for their correspondence (which is what I take him to have had in mind) is qualified by a power of interference so long as this is regulated by law and proportionately deployed for the protection of the rights of others. That is exactly what s.10, mediated through art.10, sets out to gauge, and it does not seem to me to help the argument if one adds to the major right of free expression, which is clearly in play, the ancillary right (as it is in the present context) to respect for one’s correspondence.
Taking the interests of justice to have the meaning adopted in Morgan-Grampian , and allowing Mr Tugendhat to reserve, if not the argument, at least the hope that the narrower view taken in Guardian Newspapers is to be preferred, we reach the paradox that while the prevention of crime cannot ordinarily found a right to disclosure, once the right is separately established the prevention of crime becomes – it appears – a ground on which the protection of journalists’ sources can be defeated: becomes, in other words, a ground of disclosure. In point of law Parliament’s words are clear and Interbrew is entitled to the benefit of them. Whether the benefit is real in point of fact, however, will have to be considered below.
The issue for us, therefore, since the question is one of law, is whether it is necessary in the interests of justice or for the prevention of crime to breach the confidentiality of sources which the law otherwise protects. It is, I think, uncontroversial that the effect of ss. 2 and 3 of the Human Rights Act 1998 is to require our judgment to conform so far as possible to the requirements of art.10, and to take into account the Court’s jurisprudence in determining what those requirements are. This must mean that, to be necessary within what is now the meaning of s.10, disclosure must meet a pressing social need, must be the only practical way of doing so, must be accompanied by safeguards against abuse and must not be such as to destroy the essence of the primary right: see Clayton and Tomlinson, The Law of Human Rights (2000) paras. 6.43-47 and the cases there cited. To the extent that these criteria refine the s.10 test of what is necessary and highlight the need for a methodical approach to it, they may go beyond the approach taken by the House of Lords in In re an Inquiry [1988] AC 660, 704, where Lord Griffiths confined his construction of the word to the lexicographical range of meanings, leaving the rest to the judge of fact. As Laws LJ indicated in Ashworth (cited above), it is an unvarying premise that it is prima facie contrary to the public interest for press sources to be disclosed; and the debate which follows must today be whether in a given case there is “an overriding public interest, amounting to a pressing social need” to which the need for confidentiality should give way. This vocabulary reflects that of the European Court of Human Rights in Goodwin (above; see especially para. 39).
For all this, I would not necessarily endorse Mr Tugendhat’s critique of Lightman J for having described the s.10 shield in terms of a rebuttable presumption. What matters in this as in all cases is not how the mechanism is described but how it is operated.
What Lightman J said was this:
It seems to me, with great respect, that of the judge’s three particular reasons for withdrawing the shield, the second and third are not tenable and the first is, at lowest, controversial. I will explain why.
The interests of justice
The interests of justice are the more obvious s.10 ground for removing the shield. They include both the bringing of proceedings by Interbrew against the source and their exercise of any legal rights they may possess against him. They do not include the prevention of crime, since the prevention of crime is a separate ground in s.10. But the detection of crime, which is not separately spelt out, is paradigmatically in the interests of justice. Does it not follow that once a claimant has passed through the Norwich Pharmacal door by establishing a civil cause of action, it may enlarge its claim on the identity of the source by contending that disclosure will assist in the detection of crime?
I do not think it does follow. Section 10, as I have said, does not create either a freestanding privilege from disclosure or a freestanding withdrawal of the privilege in specified conditions. It assumes the existence of a prima facie right to disclosure and qualifies it in strictly limited terms, requiring a value judgment to be formed in every case. Absent an initial right, s.10 has no bearing. If this is a correct analysis of it, it seems to me to follow that an initial right to disclosure for one purpose does not necessarily allow the statutory privilege which protects journalistic sources to be abrogated for a different purpose. The point can perhaps be illustrated by considering the exception in s.10 in favour of national security. A newspaper publishes a story from a confidential source, believed by it to be completely reliable, which alleges A to be British spy. A, considering himself defamed and wanting to prevent the source from disseminating further libels on him, brings Norwich Pharmacal proceedings. The newspaper invokes s.10. Can A argue that it is in the interests of national security to trace the source, since the source is likely to be within the security services? Or must that be for the Attorney General if, and only if, he decides to bring his own proceedings to trace and discipline the source as a threat to national security? The latter seems right. On the other hand, it is harder to say that a company trying to trace an insider who is feeding information to anarchists planning to disrupt its operations cannot rely on the s.10 exception for the prevention of disorder. For such reasons as these, as well as because of the plain language of s.10, there cannot be a simple rule confining the available grounds for abrogating the privilege to those which found the initial right to disclosure. But there is an undoubted problem in the apparent availability of an armoury of fresh weapons to any claimant able to find the key to the Norwich Pharmacal door.
The solution, in my view, is in the pivotal word “necessary”. Read, as it must now be read, in the light of the Convention and its jurisprudence, it requires close regard to be had to the relationship between the mischief and the measure. If the mischief is a civil wrong, the measure which needs to be justified as relevant and proportionate is one which will right the wrong. In the present case, that will mean a measure in the interests of civil justice, not – except peripherally – for the detection of crime. The latter is a matter for the FSA with its quite different statutory powers. I have accepted that the language of s.10 does not make it possible – and its purpose does not always make it desirable – to shut out other listed grounds of abrogation of the privilege; but the effect of the Human Rights Act and the Convention is to place the focus on the statutory analogue of the ground established for disclosure.
The prevention of crime
This reasoning, if it is right, deals equally with the place of the s.10 exception for the prevention of crime. If Interbrew can show that the source has committed or been party to a crime of market manipulation and is likely to do so again unless caught, it will not be irrelevant to the present claim; but neither will it be a freestanding ground for abrogating the s.10 privilege. In re an Inquiry [1988] AC 660, 704-5 (an example of the class of case where the prevention of crime exception is pivotal because it mirrors the ground of the application) makes it clear that the prevention of crime under s.10 can go much wider than simply preventing repetition, but Interbrew do not need it to do so.
Forgery and market manipulation
Everything proceeded below on the assumption that crimes of forgery and market manipulation were proved. But both, or at least the former, depend upon the falsity of the pages in the document showing bid price and timetable. False they were if one goes on such evidence as was before the court. But by definition their falsity is alleged against an absent and silent accused. We have no way of knowing, any more than – as Mr Tugendhat stressed – the five defendants do, whether the source, if cornered, would demonstrate that he had simply assembled authentic documents from different places within Interbrew, GS and Lazards.
I have to say that I find this aspect of the Norwich Pharmacal procedure troubling. A commercial enterprise which may very well have its own reasons for denying the authenticity of a document gets a clear run against a media defendant which can only, save in rare cases, take a neutral stand on the question. The court of first instance needs to be extremely circumspect before accepting evidence, especially when, as here, it is second- or third-hand, that goes to the heart of the case and cannot be controverted. I would not be prepared to accept Mr Tugendhat’s submission that even the source of a lie, and one knowingly published, would rank for protection under s.10 – at least not in practice; but he is entitled to draw attention to the one-sidedness of the evidence about whether what has been published is a lie.
The source’s motive and purpose.
The source may have had many possible motives for leaking information. He or she may also have had more than one purpose in mind. Mr Tugendhat’s grounds of appeal argued that the latter is irrelevant; but, as he accepted in argument, it is motive rather than purpose which needs to be excluded. Not only is motive ordinarily pure guesswork in the absence of the source; it is immaterial to the legal issues. Indeed, if Laws LJ is right in Ashworth (cited above), there is in the chilling effect of disclosure orders an affirmative policy reason for ignoring motive in individual cases. The purpose of the leak, by contrast, is likely to be highly material. If it is to bring wrongdoing to public notice it will deserve a high degree of protection, and it will not matter – assuming that it could anyway be ascertained - whether the motive is conscience or spite. If the purpose is to wreck legitimate commercial activity, again it will not matter whether it is done for political motives or personal gain: it will be the less deserving of protection. For these reasons the court of first instance, while not speculating about motive, needs to form the best view it can of the source’s purpose. To this extent I would not, with respect, agree with the view of Laws LJ (ante) that it is of no consequence that the information may be of no legitimate objective public interest, or that it is contrary to the court’s proper purposes to create a chilling effect in relation to such material. If – as ex hypothesi will be the case – a primary entitlement to disclosure and a s.10 ground for overriding the privilege have both been established, the presence or absence of an overriding public interest in disclosure of the source cannot surely be the same whether the information is of real public importance or, say, merely prurient. Where a case for overriding the privilege against disclosure is made out, the consequent chilling effect, as Lord Phillips MR said in Ashworth , may be no bad thing.
The promise of confidentiality
What in my judgment can never matter is whether the promise of confidentiality is explicit or implicit. To prioritise an express over an implied promise might well be to give the knowing and artful source better protection than the frightened and elusive source, for to give an express promise the journalist has ordinarily to know either who the source is or how to find him or her. Mr Hollander’s submission to the contrary seems to me a recipe for injustice. On the other hand – although it does not arise here – I do not think Mr Tugendhat can have the penny and the bun by arguing, as he goes on to do, that an express promise of confidentiality may be relevant. If the circumstances are not such as to bring the public interest in anonymity of sources into play, I do not see how a promise can bridge the gap; and once the interest is in play, I do not see how it can be enhanced by a promise, nor (for reasons I have given) why it should be able to be. I say that the issue does not arise here because I do not accept that the assertion of the Independent’s managing editor that the document was received by the paper’s journalists “subject to a clear understanding that they would not reveal the source of the material held by them” is more than a description of the implied terms on which all such material is received.
Is there a discretion under s.10?
Perhaps the hardest of all the questions is whether, and if so in what measure, the court has a discretion to exercise under s.10.
I have no difficulty in concluding that the central exercise is not in any true sense one of discretion. Deciding whether disclosure is necessary for one of the listed purposes is a matter of hard-edged judgment, albeit one of both fact and law, and none the less so for having to respect the principles of proportionality. To this extent I respectfully accept that Lightman J erred when, in refusing permission to appeal, he wrote: “The decision is, or involves, an exercise of discretion”.
Mr Hollander has devoted the second half of his skeleton argument to the proposition that the final balancing exercise is an exercise of discretion. For authority he relies on what Lord Phillips MR said in paragraphs 91-2 in Ashworth. But those paragraphs, cited earlier in this judgment, have to be read in the context of the one which precedes them. As I understand his reasoning, the Master of the Rolls is saying no more than that, accepting for the sake of argument that this is how the reasoning process can be characterised, it does not differ in practice from the approach of the European Court of Human Rights.
For my part, I can see every reason for not holding that the rigorous approach to necessity which the Convention requires in this context is equivalent to, say, a decision as to where the balance of convenience lies on an application for an interlocutory injunction. Discretion is exercised where the court has to make a choice between two or more legitimate courses. On both Strasbourg and domestic authority, by contrast, the question which arises under s.10 is what the legitimate course is: is there a lawful aim? is disclosure necessary to achieve it? will disclosure destroy the essence of the protected right? and if not, does its importance outweigh the public interest in protecting journalists’ sources? To each of these questions, once the facts are found, there can in law be only one answer. As Lord Griffiths said in In re an Inquiry (above) at 704, “whether a particular measure is necessary, although described as a question of fact for the purpose of s.10, involves the exercise of a judgment upon the established facts”. His next remark, that “[i]n the exercise of that judgment different people may come to different conclusions on the same facts”, does not reduce the exercise to one of discretion. As Lord Bridge was later to explain in X v Morgan-Grampian (above, at 44):
This is why the judge was wrong, in my respectful view, to say that his decision was an exercise of discretion. But it must not be forgotten that to establish the factors required by s.10 to override confidentiality is to do no more than restore the jurisdiction initially invoked in order to obtain an order for disclosure. If within this jurisdiction there are distinct grounds, such as unclean hands or delay, for refusing an order, the judge in his discretion can refuse it. The s.10 bar will have been lifted, but other, discretionary, bars may still operate. To this extent Lightman J was therefore right to say that the court’s judgment involved an exercise of discretion; but not materially to the present case.
Conclusions
In the light of the foregoing I have come to the conclusion, though not without misgiving, that the order for disclosure was rightly made against all the defendants. Nothing which has been put before us suggests that the court will be significantly better placed at an eventual trial than it is now to decide the key issues; and there is at least some force in Interbrew’s complaint that its hands are tied vis-à-vis South African Breweries, and possibly other targets too, unless and until it can prevent a recurrence of this spoiling operation.
For the reasons given earlier in this judgment, Interbrew’s prima facie entitlement to delivery up of the documents is established because – and solely because – it may enable them to ascertain the identity of the proper defendant to a breach of confidence action relating to the relatively anodyne, though not the explosive, parts of the document. From the sweep of Interbrew’s original case and the “lethal cocktail” on which Lightman J founded his conclusions, the basis of the application now shrinks to this little measure. But though little, it is far from insignificant for Interbrew.
With it, one turns to s.10 of the 1981 Act. The section begins by barring any order for disclosure in circumstances such as these, since its very object is to discover the source of information which the defendants have published. Interbrew, however, can invoke one of the listed purposes for lifting the bar: that disclosure is necessary in the interests of justice. That it may also go to the prevention of crime cannot be ruled out as irrelevant, but it is peripheral because it is not a purpose for which Interbrew themselves are entitled to disclosure. Then is the public interest in the doing of justice sufficient in the particular circumstances of this case to make disclosure necessary? Reading that question through the lens of the Convention and its jurisprudence, as we are now required to do, the following elements separate themselves out.
First, what is the nature and weight of the public interest in the confidentiality of sources? The right of free expression enshrined in art. 10 is undifferentiated, but as the European Court of Human Rights said in Goodwin , “freedom of expression constitutes one of the essential foundations of a democratic society”, and “[p]rotection of journalistic sources is one of the basic conditions for press freedom” (paras. 39, 40). Much judicial authority in this country says the same. The news media, in consequence, enjoy in s.10 of the 1981 Act a high initial level of protection, not in their own but in the public interest.
The entitlement to reverse the balance is prescribed by law, as art. 10(2) requires. A required ground for doing so, founded on the no less important public interest in an effective system of justice, is present. Is it then necessary to reverse the balance? The following elements seem to me to matter here:
• What is “necessary” within s.10 must also be what is “necessary in a democratic society” within art. 10(2). A democratic society is one in which press sources are properly protected. To this extent the balancing exercise of which Lord Bridge and others speak begins with a juridical finger in the scales.
• Any invasion of the protection must meet a pressing social need – not merely an individual one. This meets the need to counterpose one public interest to another. Here the need, in terms of s.10, is to enable Interbrew to restrain by court action any further breach of confidence by the source and possibly to recover damages for losses already sustained. In terms of art. 10(2) it is to protect the rights of Interbrew.
• There must be no less invasive alternative. While I would not adopt the judge’s approach of simply refusing to second-guess the view of the aggrieved parties’ solicitors, I would infer from the evidence that as much has been done as can at present be done by the use of reputable private detectives to trace the source. I can see that to demand particulars of the admittedly general assertions about this could jeopardise the exercise.
• There is a linear and logical relationship between the production of the document and the interests of justice.
• The basis on which production is sought does not negate the right of free expression: it seeks to limit it on prescribed grounds in a particular fact situation.
• Whether production is then a proportionate response is a value judgment which further analysis cannot assist. It requires, rather, a synthesis of what has been established as legally relevant in fact and in law.
It seems to me that once the legitimacy of Interbrew’s intended resort to law is accepted, the relatively modest leak of which they are entitled to complain does not diminish the prospective seriousness for them of its repetition. For the media, on the other hand, the public interest in their freedom to publish is constant, and with it the public interest in the confidentiality of their sources. While I do not think that the character of the material is irrelevant as a matter of law, I agree with Laws LJ in Ashworth at least to this extent, that it cannot be for the court to decide how interesting or important it thinks the material is. That is for journalists and their editors. But it may follow that the more the press decide to make of a story, the greater will be the affected party’s legitimate interest in finding and suppressing its source.
What in my judgment matters critically, at least in the present situation, is the source’s evident purpose. It was on any view a maleficent one, calculated to do harm whether for profit or for spite, and whether to the investing public or Interbrew or both. It is legitimate in reaching this view to have regard not only to what Interbrew assert is the genuine document but also to the interpolated pages; for whether they are forged or authentic, integral or added, they were calculated to maximise the mischief. To this factual extent the “lethal cocktail” is material, despite its legal irrelevance to the earlier stages of the inquiry. The public interest in protecting the source of such a leak is in my judgment not sufficient to withstand the countervailing public interest in letting Interbrew seek justice in the courts against the source.
This approach deliberately avoids setting a legal ambush for the news media: it depends upon nothing that will not have been known to them at the time of publication. In particular it depends upon what appeared on the face of the leaked document and not upon the authenticity of its components. I would be unhappy about any legal or factual approach which undid the protection of press sources on grounds which the media themselves were not in a position to appraise when they decided to publish; but while that may have to be considered again in another case, it does not arise here. What the media cannot do, and (to be fair) have only faintly sought to do in argument before us, is to plead a public duty to circulate any story of sufficient interest which comes into their hands, subject only to an ethical obligation to leaven it with any on-the-record denial which is offered. As they know better than anybody, denial can be jam on the bread and butter of a good story.
Because the foregoing reasoning does not depend upon any part of the document being forged, the relevance of truth and falsehood to the question of necessity does not directly arise. But if this were a case in which the falsehood of the nub of the story had been established (as Mr Hollander says it has been) it would in my judgment be an additional reason for overriding the protection accorded to the source. Art. 10 of the Convention explicitly links the right of free expression with the freedom of the public “to receive … information … without interference by public authority”. If, as the Court stressed in Goodwin (above, para. 39), the central purpose of the shielding of journalists’ sources is to enable the press to provide accurate and reliable information, to the extent that that purpose is departed from the rationale of the protection recedes, and with it, arguably, the weight which the courts are called on to accord to it. There is no public interest in the dissemination of falsehood. The fact that Interbrew routinely kept an eye on the brewing market was not news. The news, if news it was, was that they were preparing a bid for a named brewer at a named price on a named timetable. The defendants faithfully reported the denial that any of this was true, but they reported parts of the allegedly false story too. While newspapers cannot be asked to guarantee the veracity of everything they report, they in turn have to accept that the public interest in protecting the identity of the source of what they have been told is disinformation may not be great. Equally, however – and this is another reason for preferring to steer off the whole question – the media may well not know who is telling them the truth, and it may be a legitimate decision to publish both sides rather than neither side of the story. Again, everything is likely to depend on fact and context.
I would therefore dismiss this appeal. The Independent, against which at one stage it seemed to me that the judge had made a speculative order incapable of enforcement, does not contest the evidence suggesting that it too received a copy of the document. The order needs to be adjusted accordingly so as to take the same form in all five cases.
Lord Justice Longmore:
I agree.
Lord Justice Ward:
I also agree.
Order : appeal dismissed; counsel to lodge an agreed minute of order.
(Order does not form part of the approved judgment)