Various Claimants v MGN Ltd [2019] EWCA Civ 350

[2019] EWCA Civ 350
IN THE COURT OF APPEAL (CIVIL DIVISION)
[2018] EWHC 708 (Ch)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
BUSINESS LIST (ChD)
MR JUSTICE MANN
Venue Royal Courts of Justice Strand, London, WC2A 2LLDate 07/03/2019
Between
VARIOUS CLAIMANTSRespondents/ ClaimantsMGN LIMITEDAppellant/Defendant
Before
LORD JUSTICE McCOMBETHE SENIOR PRESIDENT OF TRIBUNALS (SIR ERNEST RYDER)LORD JUSTICE FLOYD- - - - - - - - - - - - - - - - - - - - -Between:
Richard Spearman QC (instructed by Reynolds Porter Chamberlain LLP) for theAppellantDavid Sherborne, Sara Mansoori and Julian Santos (instructed by Atkins Thomson) for RespondentsHearing Hearing date: 13 February 2019
Approved Judgment
[1]This appeal raises a question concerned with the protection from disclosure of confidential journalistic sources enjoyed by publishers under section 10 of the Contempt of Court Act 1981. It arises in the phone hacking litigation brought against the appellant, MGN Limited (“MGN”), by the respondents, who bring claims for misuse of private information as a result of the unlawful interception of voicemail messages left on their mobile phones. MGN claims that an order made in the proceedings in 2015 for early disclosure by MGN gives rise to a real risk, in the case of one claimant, that MGN will have to disclose material which would reveal the identity of a confidential journalistic source. MGN accordingly applied to the judge for a variation of the order made in all the claims so as to prevent any disclosure occurring without the consent of the alleged source. The judge, Mann J, who is the Managing Judge in this and other phone hacking litigation, refused the application to vary. MGN appeals to this court against his judgment, dated 22 March 2018, and his consequent order, with the permission of Newey LJ granted on 11 April 2018.

Background to the application

[2]On 21 May 2015 Mann J handed down his judgment in Gulati v MGN Limited [2015] EWHC 1482 (Ch). At paragraphs 6 to 10 he gave a general introduction to the techniques used for phone hacking, which is all that is necessary for an understanding of the issues on this appeal:
“6. Although the colloquial expression “phone hacking” has for some time been used to describe the activities which lie at the heart of this case, (and I shall continue to use it) a more accurate description would be “mobile voice-message interception”
. It does not involve listening in to actual two-way phone conversations. It works in the following way. 7. A mobile telephone account comes with a voicemail box in which the account holder can receive, and listen to, voice messages left by callers when he or she does not answer the phone. The messages can be retrieved from the phone itself, or by ringing in from an outside line. The mailbox can be protected from an unauthorised person ringing and listening to voicemail messages by a PIN code. At the time of the events covered by this action some of the mobile phone companies had default PIN codes, and many users did not bother to change them (if indeed they knew that they existed). Other users changed the PIN codes but used codes that were predictable - perhaps consecutive numbers, or years of their birth or of others close to them. An unauthorised person who knew or could guess those PIN codes could use them to access another’s voicemail account. 8.There were two ways in which an unauthorised person could access another’s voicemail box. The trick is to get through to the voicemail system, and not the owner of the phone in person. If the owner answered the phone the hacker would not be in the voicemail system. So, unless one knows that the owner will not pick up (so that the call goes through to voicemail after a number of unanswered rings) one has to ring when one knows the phone line is in use so that the call goes through to voicemail. The first hacking technique produces this effect. It involves the use of two phones simultaneously. The technique was described by Mr Evans, one of the witnesses. The first phone was used to ring the telephone number of the victim. He would count to three and then use the second phone to dial the same number. Because a call was already being made the second phone was likely to go to the caller’s voicemail box. Phone 1 was immediately disconnected, usually without actually ringing the receiving phone so as to alert the owner of the phone, so the owner would often not know that an attempt had been made to ring his or her phone. At that point the unauthorised caller could concentrate on the second phone, and the voicemail box. Once the voicemail system had responded the hacker could behave like the owner, and press keys which allowed the entry of a PIN, which in turn allowed access to the messages that had been left there. This, of course, required knowledge of the PIN, but because so many people never changed their PIN number from the default number provided by the network, or used a predictable number which could be accurately guessed by the hacker, their phone accounts were vulnerable to this activity. This technique was given the slang term “double tapping”. 9. The other access route applied only to the Orange network. This network provided a separate number which would be dialled into from any phone in order to get voicemail messages. The dialler would have to identify the phone number whose messages were being recalled, and then enter the PIN. This route in was available to anyone who knew the number of the phone in question and could ascertain or guess the PIN. The voicemail messages could then be accessed. 10. This activity proved to be a very useful source of information for the newspapers involved in this case. Once into a user’s voice mailbox, a journalist could listen to messages left there and often find out the numbers of the people who left them. The content of the message might itself be of significance, as might the identity of the person who left it. Furthermore, the number of the person who left it could also be ascertained, and if it was a mobile number an attempt could be made to hack into that user’s voicemail as well. In that way a pattern of sources of information could be built up. Mr Evans, the principal witness about the actual hacking activities in the case, described this extended activity as “farming”.”[3]In his judgment in Gulati, Mann J concluded that unlawful phone hacking was rife throughout the Daily Mirror, The Sunday Mirror and The People, which are all newspapers published by MGN. He proceeded to assess and award damages to eight representative claimants for the intrusion and harm caused to them by the hacking of their mobile phones. An appeal to this court against his judgment and order was dismissed on 17 December 2015: Gulati v MGN Ltd [2015] EWCA Civ 1291. At paragraph 106 Arden LJ (with the agreement of Rafferty and Kitchin LJJ) said:
“Indeed, so far as I can see, there were no mitigating circumstances at all. The employees of MGN instead repeatedly engaged in disgraceful actions and ransacked the respondents' voicemail to produce in many cases demeaning articles about wholly innocent members of the public in order to create stories for MGN's newspapers. They appear to have been totally uncaring about the real distress and damage to relationships caused by their callous actions. There are numerous examples in the articles of the disclosure of private medical information, attendance at rehabilitation clinics, domestic violence, emotional calls to partners, details of plans for meeting friends and partners, finances and details of confidential employment negotiations, which the judge found could not have been made if the information had not been obtained by hacking or some other wrongful means. The disclosures were strikingly distressing to the respondents involved.”
[4]The reference to damage to relationships perhaps requires some further explanation. Members of the public who found details of their personal and private lives appearing in the newspapers (unknown to them via the hacking of their phones) naturally came to suspect that friends, relatives or other close associates were leaking this information to the press. This suspicion was supported by the fact that newspaper articles based on the private information would often describe the source as a friend of the claimant. It is not difficult to see how these suspicions could lead to the poisoning of relationships and the breakdown of trust between individuals.[5]Because the unlawful phone hacking activities had been undertaken over a number of years, and because much of the phone hacking had been carried out using pay-as-yougo mobile phones which had been disposed of or destroyed, there was little in the way of a documentary record of the extent to which any individual’s phone had been hacked. One of the important threads of the evidence was constituted by the existence of some call data from MGN’s landlines to the claimants’ mobile telephones. This was in itself of some value to the claimants, but the process known as “farming” (described by Mann J in the passage cited above) meant that close associates of the claimants would be subjected to phone hacking as well, and private information of the claimants would often be obtained by hacking into these associates’ phones as well and obtaining access to private voicemail messages left by the claimant. In a judgment dated 8 July 2015 (Various Claimants v MGN Limited [2015] EWHC 2381 (Ch)) Mann J noted:
“The hacking of associates is as significant in many cases, or as likely to be as significant in many cases, as the hacking of the claimant. Merely to produce the fruits of a search in relation to calls made to the claimant’s number may produce an entirely false impression.”
[6]It was on the basis of these considerations that Mann J, shortly after handing down his judgment in Gulati, provided for a regime of early disclosure of call data in the further claims, grouped together as Wave 2. At the 9th Case Management Conference on 8/9 July 2015, he made an order the effect of which was to ensure that, after issue of the claim form, each claimant will receive from MGN:(a) MGN landline call data (where it exists) in relation to the claimant and four of the claimant’s nominated associates;(b) private investigator invoices relating to the claimant and four of the claimant’s nominated associates; and(c) all of the articles published in MGN’s newspapers which refer to the claimant. The order was predicated on the assumption that all call data would be relevant to the claimants’ claims, as calls from MGN to the claimants would be indicative of phone hacking. The order contained no exception or limitation so as to enable MGN to exclude call data which related to irrelevant communications between MGN and the claimants.[7]Thus, each claimant was able to nominate four associates (family, friends or others) all of whose call data would be disclosed. This early disclosure was designed in part to enable claimants to enter settlement negotiations with MGN. The judge was told that this regime had been successful in causing cases to settle. It prevented claimants having to “settle blind”. It also saved the expenditure of much time and money and court resources. Given the scale of the litigation, this was obviously an enormous benefit.[8]It is fair to say that MGN has never been comfortable with the early disclosure regime in the form in which it exists. On 24 July 2015, only two weeks after the CMC, and before the order was sealed, MGN argued that the early disclosure regime should require the claimants to seek and provide proof of consent from their associates before any call data was provided by MGN. This proposed requirement was rejected by the Managing Judge. Two months later, on 11 September 2015, MGN applied to the court to stay the early disclosure regime pending its appeal to the Court of Appeal, and again to seek a condition requiring claimants to obtain and provide proof of consent from associates prior to receiving such disclosure. The basis for seeking to impose this condition was suggested by MGN to be the risk that it would reveal a confidential journalistic source. The application was refused by Mann J (see Various Claimants v MGN [2015] EWHC 2861 (Ch)) largely because there was no concrete evidence, as opposed to a speculative risk, that any particular source would be identified. The judge refused permission to appeal from his order and no subsequent appeal was pursued to this Court.

The application

[9]The present application was launched by application notice dated 12 March 2018. It sought an order in the following terms:
“Paragraph 30.d of the 9th CMC Order dated 9 July 2015 is amended to read as follows: “The mobile telephone number(s) of not more than 4 person(s) associated with the Claimant during the Relevant Period who have provided their written consent to theDefendant providing the Claimant with the documents setout a paragraph 31.a and b. relating to them (“Associates”) together with the copies of documents signed by eachAssociate confirming such consent.”” “The mobile telephone number(s) of not more than 4 person(s) associated with the Claimant during the Relevant Period who have provided their written consent to theDefendant providing the Claimant with the documents setout a paragraph 31.a and b. relating to them (“Associates”) together with the copies of documents signed by eachAssociate confirming such consent.””
[10]In the course of the hearing of the appeal MGN made it clear that it does not seek to vary the order in respect of paragraph 31.b, that is to say in respect of Private Investigator Invoices relating to the claimant and the four associates. It is difficult to see why that variation was ever applied for, or what connection it could possibly have to the identification of a possible source.[11]MGN’s application was supported by the first witness statement of Alex Wilson, a solicitor at MGN’s solicitors. He explained that an issue had recently arisen “in a particular case”. This was that the disclosure to a claimant of call data (both incoming and outgoing) relating to one of his or her associates “will reveal that Associate as a source of information to the Defendant”. (“Incoming” and “outgoing” call data are defined by reference to whether they are incoming to or outgoing from MGN). Mr Wilson went on to explain that “for obvious reasons we cannot provide further details so as to avoid the risk of revealing the identity of that source”. He said: “The Defendant is also unable to justify resisting disclosure relating to the Associates on this ground in the individual case(s) in question as this would also run the risk of identifying a source (by identifying the very small pool of potential sources i.e. the named Associates)”.[12]MGN’s position was therefore that the consent procedure was a necessary prior step to obtaining the call data, and that any less burdensome approach which confined the impact of the protection of the source to the case in which it actually arose was not viable. Mr Wilson went on to explain that MGN had spoken to the relevant journalist who had confirmed that the Associate in this instance was a confidential source of information. He had also seen confirmation of this fact from the source. Finally, Mr Wilson explained that, having reviewed the call data, he was satisfied that it would clearly reveal the Associate as a confidential source “as it demonstrates the extensive contact and communication between them and the Defendant’s journalists”.[13]Mr Wilson’s witness statement was answered in the fortieth witness statement of James Heath, the lead solicitor for the claimants. Mr Heath pointed out that it had previously been made clear in the litigation that the claimants did not require disclosure of incoming (to MGN) call data, and it was MGN who had asserted that, if outgoing call data was to be disclosed, it was necessary for MGN to provide incoming call data to avoid providing a misleading or incomplete picture. He accordingly suggested that the “simple cure” was for MGN to refrain from voluntarily providing incoming call data in relation to all four associates in the relevant case. He said that he would find it impossible to identify the fact that someone was acting as a source simply by looking at outgoing call data. Moreover, even armed with bi-directional call data, he would find it impossible to say what the information passing to MGN was. The associate might be engaged in wholly unrelated publicity activity in relation to their work, providing information in relation to a charitable project they were engaged in, or co-operating with a story relating to themselves. All the call data would show is that the associate had a lot of contact with the newspaper and nothing more.[14]Mr Heath also went on to explain why he considered that the variation of the order to require consent would be unworkable and onerous. There is no need for our purposes to go into these details.[15]In his second witness statement, in reply to Mr Heath, Mr Wilson, whilst adhering to his view that both incoming and outgoing call data would reveal the Associate to be a source, explained that “in this instance, the relevant data that would reveal the Associate to be a source of information is almost entirely outgoing calls … which record the extensive and frequent communications between the Associate and the Defendant.”

The legal framework

[16]Section 10 of the Contempt of Court Act 1981 provides:
“No court may require a person to disclose, nor is any person guilty of contempt of court for refusing to disclose, the source of information contained in a publication for which he is responsible, unless it be established to the satisfaction of the court that disclosure is necessary in the interests of justice or national security or for the prevention of disorder or crime.”
[17]Article 10 of the European Convention on Human Rights and Fundamental Freedoms (“ECHR”) provides, so far as material:
“(1) Everyone has the right to freedom of expression. This right shall include freedom to … receive and impart information and ideas without interference by public authority … (2) The exercise of these freedoms, since it carries with it duties and responsibilities, may be subject to such … restrictions … as are prescribed by law and are necessary in a democratic society … for the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence ...”
[18]The protection of journalistic sources has long been recognised to be a principle of high importance. Encroachments on this protection are capable of having a chilling effect on the free flow of information to journalists, and therefore amount to an inhibition on the freedom of the press protected by Article 10. Without such protection, sources may be deterred from assisting the press in informing the public on matters of public interest.[19]The protection afforded against disclosure of journalistic sources is not, however, absolute. Measures requiring the disclosure of such sources can be justified by “an overriding requirement in the public interest”: see paragraph 39 of the judgment of the ECtHR in Goodwin v United Kingdom [1996] 22 EHRR 123 at page 143. This reflects the test of “necessary in a democratic society” in Article 10(2) ECHR, which requires the court to weigh whether the restriction is proportionate to the legitimate aim pursued (Goodwin at [40]). The ECtHR went on to explain in the same case that “necessity” must, in any case be “convincingly established”. At paragraph 45 the court said:
“… it will not be sufficient, per se, for a party seeking disclosure of a source to show merely that he or she will be unable without disclosure to exercise the legal right or avert the threatened legal wrong on which he or she bases his or her claim in order to establish the necessity of disclosure.”
[20]Both parties drew our attention to the helpful summary of principles by Warby J in Arcadia Group Limited and others v Telegraph Media Group Limited [2019] EWHC 96 (QB) at [15], which I accept as correct: “The following principles are now clearly established, and not controversial:-(1) The onus lies on the applicant to show that disclosure should be ordered.(2) It must be shown that disclosure is necessary for one of the four legitimate purposes identified in s 10. It is not enough, for this purpose, to show that the information is relevant to the claim or defence: Maxwell v. Pressdram [1987] 1 WLR 298 at 310 G-H (Parker LJ). It is not even enough to show that the claim or defence cannot be maintained without disclosure: Goodwin v UK [1996] 22 EHRR 123 [39], [45]. The need for the information in order to bring or defend a particular claim is not to be equated with necessity "in the interests of justice".(3) In In re An Inquiry under the Company Securities (Insider Dealing) Act 1985 [1988] AC 660, 704, Lord Griffiths gave this guidance as to the meaning of the term "necessary" in this context: "I doubt if it is possible to go further than to say that 'necessary' has a meaning that lies somewhere between 'indispensable' on the one hand, and 'useful' or 'expedient' on the other, and to leave it to the judge to decide towards which end of the scale of meaning he will place it on the facts of any particular case. The nearest paraphrase I can suggest is 'really needed.'"(4) This requires proof that the interests of justice in the context of the particular case are "so pressing as to require the absolute ban on disclosure to be overridden": X Ltd vMorgan-Grampian (Publishers) Ltd [1991] 1 AC 1 at53C (Lord Oliver). In the language of Strasbourg, the disclosure order must correspond to a pressing social need, and must be proportionate. It must be "justified by an overriding requirement in the public interest": Goodwin [39].(5) Hence, it is necessary for the applicant to satisfy the Court, on the basis of cogent evidence, that the claim or defence to which the disclosure is relevant is sufficiently important to outweigh the private and public interests of source protection, and that disclosure is proportionate.(6) When making this assessment, the Court must bear in mind that incursions into journalistic confidentiality may have detrimental impacts on persons other than the individual source(s). Disclosure may have a "detrimental impact … on the newspaper against which the order is directed, whose reputation may be negatively affected in the eyes of future potential sources by the disclosure, and on the members of the public, who have an interest in receiving information imparted through anonymous sources and who are also potential sources themselves": Goodwin [69].(7) The court must be satisfied that there is, "no reasonable, less invasive, alternative means" of achieving whatever aim is pursued by a source disclosure application: Goodwin ibid.” "I doubt if it is possible to go further than to say that 'necessary' has a meaning that lies somewhere between 'indispensable' on the one hand, and 'useful' or 'expedient' on the other, and to leave it to the judge to decide towards which end of the scale of meaning he will place it on the facts of any particular case. The nearest paraphrase I can suggest is 'really needed.'"[21]In his speech in X Ltd v Morgan-Grampian Lord Bridge emphasised the following: a) “… where a judge asks himself the question: “Can I be satisfied that disclosure of the source of this information is necessary to serve this interest?” he has to engage in a balancing exercise.” (see 41E); b) The starting assumptions in that exercise are(i) the protection of sources is itself a matter of high public importance;(ii) nothing less than necessity will serve to override it, and(iii) that necessity can only arise out of another matter of high public importance, being one of the four matters listed in the section, see (41E-F); c) Whether necessity of disclosure is established is a question of fact, not of discretion, but, like such questions as whether someone has acted reasonably, it is one which requires “the exercise of a discriminating and sometimes difficult value judgment” (see 44C); d) The balance is between the weight to be attached to the importance of disclosure in the interests of justice on the one hand and that of protection from disclosure in pursuance of the policy which underlies section 10 on the other hand, (see 44 C-D).[22]There was some debate before us as to the extent to which the court might vary the weight to be given to the protection of the source dependent on the nature of the information which is sought to be protected. Lord Bridge in Morgan-Grampian said at 44 E-F:
“One important factor will be the nature of the information obtained from the source. The greater the legitimate public interest in the information which the source has given to the publisher … the greater will be the importance of protecting the source”
[23]One must be careful how far one takes that proposition. It is certainly not the case that one ceases to afford protection to the source because the source is providing information which is low down on the public interest spectrum. Read as a whole, I understand Lord Bridge’s speech to be saying that one starts with the assumption that the protection of the source is always a matter of high importance, and it becomes yet more difficult to override that public interest in cases where there is a real public interest in the information provided by the source.

The judgment of Mann J

[24]Having set out the relevant principles in terms which are not the subject of any criticism, the judge set out the parties’ respective arguments at paragraphs 23 and 24. He recorded the claimants’ case as being that, if there was a risk to the source, then this was a case where the interests of justice overrode the protection of that particular source. At paragraph 25 the judge expressed his conclusion that the case of the claimants was correct. He went on to expand on that conclusion at paragraphs 26 to 30 of his judgment. At paragraph 26 he recorded some important findings:(i) The information which the claimants acquire under the early disclosure regime in relation to associates’ call data was “of great significancein the litigation”.(ii) Techniques for dealing with phone hacking cases generally had been refined over the years. The early disclosure regime was “very important to the just and proportionate disposal of cases.”(iii) He was prepared to infer that “hundreds of cases” had been settled, because the early disclosure gave claimants “what they need in order to assess the strength of their respective cases at an early stage, and achieves some sort of balance between information which the claimants will inevitably not have as to how much unlawful activity may have gone on and information which the defendant does have in its hands.” There was “a significant imbalance of information as between the claimants on one hand and the defendant on the other”, because of the covert nature of the operations meaning that evidence of them lies, almost exclusively, in the hands of the defendant.(iv) The fair disposal of the cases involved considering techniques to achieve early settlement, or a more focussed pursuit of the cases if they did not settle.(v) Cutting down early disclosure in the manner proposed would risk putting claimants in a position in which they could not fairly assess the strength of their respective cases and would put them in a position in which, in large numbers, they would either have to settle partially blind or pursue a claim which they might otherwise have settled.(vi) The numbers involved were not small. In the current wave there were, on a rough and ready estimate, at least 40 or 50 cases which were likely to be affected, and in future waves there would be even more. There would therefore be a very significant number of claimants who would potentially be affected.[25]At paragraph 27 the judge recorded that the effect of the proposed variation of the early disclosure regime would be significant. There was a high risk of a significant number of “non-returns”, that is to say associates who, upon being asked for consent, did not respond. In the course of argument before him, MGN had offered a concession that the call data could be disclosed without consent if it was shown that reasonable efforts to obtain it had proved unfruitful. If the question of whether enough had been done to obtain their consent remained unresolved, the claimant would be deprived of important information. To resolve that issue would require a further significant costs for a claimant and loss of time. There would be associates whom the claimant would be reluctant to contact at all, such as ex-partners or spouses. Others would refuse consent for reasons other than being the supposed source. The judge continued:
“The end result of this assessment is that considerable burdens will be placed on claimants adding time and costs, and there is a real risk that information which a claimant genuinely needs will not be forthcoming even if all queries about nonresponsive sources, as between claimants and the defendant, are resolved in favour of disclosure. In short, [MGN’s proposed] scheme does not produce a workable scheme which operates in the way in which the overall scheme was intended to work and claimants will suffer as a result.”
[26]The judge also observed, at paragraph 28, that it was “of some weight in the debate … that the defendant has not really specified how it is that the source will be identified.”[27]What was particularly significant (paragraph 29) was the “impact on the many in order to protect the identity of one associate in one case.” All the outgoing call data was disclosable on standard disclosure in any event, in all of the cases. Yet MGN sought to qualify the disclosure in all cases in order to protect the identity of a source in only one of them. He added:
“The protection of sources is and remains important, but the techniques to be adopted to protect them have their limits, and prejudicing dozens of cases in which the point does not arise is not likely to be in the interests of justice.”
[28]Also of real significance (paragraph 30) was the nature of the information, and the nature of the source, which the defendant was seeking to protect. It was likely that the source was someone close to one of the claimants who had been passing private information gleaned from being in a position of some confidence with the claimant in question. It had not been suggested that that sort of information had any public interest element to it.

The judge concluded:

“Taking all these matters into consideration, I am quite satisfied that the balance lies in favour of leaving the early disclosure regime where it is. To invoke the qualifications which [MGN] seeks to invoke would be to risk materially prejudicing a large number of claimants who have absolutely no association to the source protection issues and there would be a serious and disproportionate imbalance against them were they now to be deprived of the sort of early call information which has proved so important in the proper conduct of this litigation. Balancing their rights against the proper interests which are invoked in considering journalistic source protection, and the other matters referred to above, I consider that the interests of justice clearly lie in retaining the current necessary, and properly operated, early disclosure regime.”

The appeal

[30]Mr Richard Spearman QC for MGN submitted that the judge had failed to apply section 10 of the Contempt of Court Act. He had made no finding that early disclosure of call data was necessary in the interests of justice. In fact, none of the claimants contended that they had any legitimate interest in the disclosure of the source as such. Disclosure of the source would be a by-product of the early disclosure regime. The judge had adopted a different and incorrect approach, by balancing the risk of materially prejudicing the claimants against the interests of protecting journalistic sources. His conclusion, quoted above, that “the interests of justice clearly lie in the retention of the current necessary and properly operated, early disclosure regime” was not a proper application of the test under section 10.[31]Mr Spearman also attacked the judge’s conclusions on the basis that he had not properly investigated other, less intrusive alternatives. MGN’s proposed variation had not even been tried out in practice. Although the claimants had given examples of cases where the obtaining of consent could give rise to difficulties, these were not typical of what would happen in the ordinary case where the associate would be likely to co-operate. His primary position was that the claimants should not get the call data at all where they had not obtained the relevant consent, except where the associate had died. This was a retreat from the concession made before the judge, where MGN had been prepared to accept that the data should be disclosed when reasonable efforts had been made without success to obtain consent. If MGN’s concession were withdrawn, the problems which the judge saw in resolving issues about whether adequate steps had been taken to obtain consent would not arise.[32]Mr Spearman also submitted that the judge had failed to have regard to the consideration that call data relating to contact between MGN and the source was not disclosable on standard disclosure because it did not relate to the claimants’ claims of unlawful information gathering.[33]Mr Spearman also criticised the judge for relying on the degree of risk that the source would be identified. He submitted that this was irrelevant. Once the threshold of a sufficient risk of identification was crossed section 10 was engaged. One did not revisit that conclusion when balancing the interests involved. Likewise, Mr Spearman attacked the judge’s reference to the nature of the information provided by the source.[34]In the course of argument we were shown an example of the format in which the call data was disclosed. The call data is a printout from a database kept by MGN. It consists of lines of data with columns allocated to date, call time, call duration, calling party (i.e. the MGN internal extension), and the number dialled. I enquired why it was not possible, in the case of the particular associate identified as a source, to exclude from the printout the lines of data which identified calls to the associate in his or her capacity as a (willing) confidential source of information. As the call data printout is created for the purposes of the litigation by MGN, there should be no difficulty, as a practical matter, in not printing out those lines without providing any clues as to other activities of the source. Neither Mr Spearman, nor when it came to it, Mr Sherborne for the claimants, had any knock-out reason why such an approach would not work, although both acknowledged that the practicalities had not been closely examined. To allow MGN to withhold such data would require a carve-out from the present early disclosure regime, which has no in-built criterion of relevance. It could, however, be expressly limited so as to allow MGN, in the case of the single case identified in Mr Wilson’s evidence, to withhold call data relating to the associate acting as the source.[35]If the associate in question only acted as a source, then such a carve out would result in a nil return for that associate. There would be nothing to identify the source to the claimants if that turned out to be the case. If some of the calls were phone hacking calls, but others were to the associate acting as a source, elimination of the latter would again not give the claimants any basis for believing that the associate in question was the source, and the phone hacking call data would still be disclosed.[36]One possible difficulty with this course is how MGN would differentiate (if that were necessary) between calls to associates which are phone hacking and calls which are not. At the moment MGN claims to be able to do so because the outgoing calls are so extensive, but the sample printout we were shown appears to show phone hacking calls lasting up to 20 minutes, sometimes with 5 or more phone calls on the same day. Given that these events are now a long time in the past, accurate differentiation may be difficult. As this proposal has not been addressed at all in the evidence or skeleton arguments, it would be wrong to take it into account in dealing with this appeal. I do propose, however, to return to it once I have dealt with the arguments which are before us.[37]Mr Sherborne supported the conclusion arrived at by the judge, and the reasons he gave for it. He stressed the unique nature of the litigation which involved a large number of cases, and the importance of the call data within it. There was a strong public interest, over and above the interests of the individual litigants, in securing proper redress for the victims of phone hacking. The application to vary, he submitted, was “overkill” in that it sacrificed the interests of numerous claimants in order to protect the identity of a single a source in a single case.

Discussion

[38]I address, first of all, MGN’s submission that the judge did not properly investigate the less intrusive alternative of seeking consent from the associates. I am firmly of the view that we have no basis for interfering with the judge’s conclusion that this alternative was simply unworkable. Mann J has been managing this extremely complex, multi-claimant litigation with great skill for more than 5 years, and has, in the process, acquired an intimate knowledge of the way it works and gained invaluable experience of how it is to be properly managed. His assessment of the problems to which the proposed variation would lead was based on the whole of that experience, and not simply examples of difficulties with the variation discussed in the evidence before him. He was not obliged to conduct a road test of a proposal which he could already see would lead to a heavy burden on the claimants and risk of injustice to them. This court is in no position to second-guess his rejection of the variation as unworkable.[39]As I have indicated, MGN attempts to undermine the judge’s conclusion by retreating from the concession which they made before him which, in effect, deems consent to have been given when reasonable efforts have been made to obtain it. Whilst the withdrawal of the concession makes the cumbersome procedure somewhat less so, it has the effect that there will be some claimants on whom an injustice will fall because their associates have been negligent or reluctant in responding to the request for consent. MGN was right to consider that the concession was a necessary one to make.[40]I turn, next, to the heart of MGN’s appeal, namely the argument that the judge did not ask himself whether the disclosure of the source was necessary in the interests of justice, but instead asked himself what was in the interests of justice in the particular litigation concerned. The argument is, in essence, that the judge fell into the error identified in the second of Warby J’s uncontested principles in Arcadia of equating what was in the interests of justice with what was needed by the claimants to bring the claim.[41]It is, in my judgment, quite clear that the judge had well in mind that it was not sufficient for the claimants merely to show that they needed the call data to maintain their claim. That is why he quoted extensively from the speech of Lord Bridge in X v Morgan-Grampian which speaks of the balancing exercise and of “estimating the weight to be attached to the importance of disclosure in the interests of justice on the one hand and that of protection from disclosure in pursuance of the policy which underlies section 10 on the other.” He was also well aware that a journalistic source was entitled to a high level of protection in the interests of a free press, given his citation from Goodwin v UK at paragraph 17 of his judgment, as well as the other cases cited at [16] – [21]. It is also why he drew attention to the unique aspects of the litigation, such as the large numbers of litigants and the potential for large savings in costs and court time that the call data presented.[42]The argument, to my mind, depends critically on being able to separate out and deal separately with disclosure of the identity of the source and disclosure of the call data which the claimants need for the proper assessment and prosecution of their claims. If the claimants were asking for disclosure of all phone hacking call data and, in addition, separable and separate call data identifying the source, then of course the court would be obliged to address those two issues separately. In those hypothetical circumstances the claimants would not be able to demonstrate any legitimate interest in disclosure of the source. Indeed, it is common ground that the disclosure of the identity of the source is not relevant. The claimants’ cases are concerned with the unlawful interception of their voicemail messages, not with activities of their associates as confidential journalistic sources. The claimants would have nothing to place in the balance against the high public importance attached to the protection of confidential journalistic sources.[43]The judge was not, however, dealing with a scenario in which these two types of information could be neatly separated into airtight compartments. On the contrary, as the case was presented to him, the risk of identifying the source was an unavoidable by-product of the requirement that the call data should be disclosed. In those circumstances, the judge was entitled to weigh the public interest in the ability of the multiple claimants to assess and prosecute their claims with the benefit of the call data against the public interest in protection of the identity of the source. I am entirely satisfied that this is what he in fact did.[44]Once it is appreciated that the call data and the data which identifies the source are mixed, it is not difficult to see the answer to MGN’s criticism that the judge failed to give weight to the fact that the identity of the source was not relevant. The mixing of the two elements of the call data means that it is simply not possible to consider the necessity of disclosure of the source in isolation. The judge was well aware that the claimants did not seek disclosure of the source if it could be avoided: but on the evidence before him it could not, except by varying the regime in a manner which he held was unworkable. For the same reason, on the basis of the case presented to him, the judge was correct to hold that the call data which identified the source was material which would have to be disclosed on standard disclosure in any event.[45]In the assessment of whether the interests of justice overrode the interest in protection of the source, Mr Sherborne was correct to stress the unique nature of this litigation. Whilst it is correct to say that that there are cases where the need to protect a journalistic source can be sufficient to defeat altogether a right which is asserted against the publisher, the judge was entitled to take the view that this litigation was not such a case. The volume of claims was one basis which distinguished it. If there are 100 cases in the pipeline (and we were told that figure may be roughly correct), there are 99 in which the call data is highly relevant and important disclosure and in which there is no possible objection to its production (because the call data will not identify the source). It is only in case 100 where there exists a risk of identification. Although Mr Spearman sought to downplay the importance of numbers, they are in my judgment highly relevant to the exercise which the judge was properly conducting. The numbers of cases affect the question of whether a restriction on the protection of the source is a proportionate response to the legitimate aim of affording justice to this class of litigants. The judge was also entitled to have in mind the effect that the early disclosure regime had on the settlement of cases and the saving of time and expense, multiplied across the whole of the litigation.[46]I do not accept MGN’s contention that, once it is decided that the risk of the source being identified engages section 10, any further consideration of the level of risk of identification of the source is precluded. Whilst in many cases it may not be necessary to come back to it, it seems to me that the level of risk is a factor which can at least come back in to resolve an evenly balanced case. In any event, I do not read the judge as saying any more than it was a minor factor: he said it was “of some weight” in contrast to other matters which were “of real significance” and “particularly significant”.[47]Finally, I think the judge was entitled to draw attention to the fact that this was not a case in which any public interest was claimed in the information provided by the source. He was plainly alive to the fact that such a public interest could only serve to enhance the protection afforded to the source. He had set out the part of Lord Bridge’s speech in X v Morgan Grampian which emphasises the high importance which the law affords to the protection of the source. In weighing the interests at stake, the judge was entitled to have in mind that this was not a case where the source was entitled to enhanced protection of the kind which Lord Bridge was referring to.[48]In the end, despite the very able and clear submissions of Mr Spearman, I am not able to detect any error by the judge in his approach to the balancing exercise required of him. This court would not therefore be justified in substituting its own “value judgment” (Lord Bridge’s words) for that of the judge.[49]Nevertheless, at the end of the argument, I remained concerned that there might exist a possible regime, based on the discussion which I have set out at paragraphs 34 to 36 above, which would confine the impact of the journalistic source protection to the actual claim in which it arose, and also avoid any risk of disclosure, thereby serving both the important interests at stake. This is not a criticism of the judge. The application to him was based on MGN’s proposed variation alone, and the judge was right to dismiss it. Indeed, Mr Wilson’s evidence might be said to have made it clear that all schemes based on a challenge by MGN to the relevance of the disclosure would be likely themselves to involve disclosure of the identity of the source. For the reasons discussed in those paragraphs, however, I am not convinced that is so. There remains a possibility, how substantial I do not know, that the source could be given secure protection without any impact on the countervailing public interest in the proper prosecution of these claims.[50]I have considered whether, for this reason, we should decline to deal finally with the appeal, and refer the issue of this alternative variation to the judge under CPR 52.10(2)(b). Although I was at one point attracted to this course, I have in the end rejected it. It remains open to MGN to apply to the judge for an alternative variation of the early disclosure regime if it should choose to do so, although it will have to explain why this simpler expedient was not raised earlier. Such an application would have to address the question of how it is proposed that phone hacking call data can be distinguished from call data relating to the source acting as such to the satisfaction of the court and the claimants. It is better to leave such an entirely fresh set of case management issues to the judge, who has unrivalled experience of managing this litigation. The judge will either accept or reject the application, if it is made. No purpose would be served by the matter being retained for final disposal in this court.[51]For the reasons I have given, I would dismiss the appeal.

Senior President of Tribunals:

[52]I agree.[53]I also agree.

Cited in 6 later judgments