SUPREME COURT OF THE UNITED KINGDOM
LORD PHILLIPS, President
LORD HOPE, Dep President
13 JULY 2011
The issue that arises on this appeal is whether the court has the power to order a "closed material procedure" as described in the preliminary issue that was tried by Silber J for the whole or part of the trial of a civil claim for damages and, if so, in what circumstances it is appropriate to exercise the power. The preliminary issue was in these terms:
Could it be lawful and proper for a court to order that a 'closed material procedure' (as defined below) be adopted in a civil claim for damages?
Definition of 'closed material procedure'
A 'closed material procedure' means a procedure in which
For the purposes of this definition, disclosure is contrary to the public interest if it is made contrary to the interests of national security, the international relations of the United Kingdom, the detection and prevention of crime, or in any other circumstances where disclosure is likely to harm the public interest.
Silber J answered the question raised by the preliminary issue in the affirmative:  EWHC 2959 (QB). The Court of Appeal (Lord Neuberger of Abbotsbury MR, Maurice Kay and Sullivan LJJ)  EWCA Civ 482,  3 WLR 1069 allowed the claimants' appeal and held that the court has no such power in an ordinary civil claim for damages. The defendants appeal with the permission of the Supreme Court.
The preliminary issue was raised in proceedings in which the claimants alleged that the Security Service and other organs of the state (the appellants) had been complicit in the detention and ill-treatment of them by foreign authorities at various locations including Guantanamo Bay. The pleaded causes of action included false imprisonment, trespass to the person, conspiracy to injure, torture and breach of the Human Rights Act 1998. A more detailed exposition of the factual background is set out in paras 5-6 of the judgment of Lord Neuberger.
The appellants filed an open defence in which they admitted that the claimants had been transferred and detained, but they put in issue the alleged mistreatment and denied any liability for the claimants' detention or alleged mistreatment. At a case management hearing, the appellants said that they were in possession of material which they wished the court to consider, but which they would be obliged in the public interest to withhold from disclosure. This material was contained in a closed defence. The course contended for by the appellants would require parallel open and closed proceedings and parallel open and closed judgments. Special advocates would represent the interests of the claimants in the closed hearings.
The claimants objected to this course. They argued that a conventional public interest immunity ("PII") exercise should be conducted ex parte by a judge in relation to the "closed" material. Lord Clarke describes the PII procedure in detail at para 145 below. In response, the appellants emphasised the difficulties that would be caused by the vast amount of sensitive material in their possession and the enormous scale of any PII exercise. The evidence filed on behalf of the appellants suggested that there might be as many as 250,000 potentially relevant documents, and that PII might have to be considered in respect of as many as 140,000 of them. It might take three years to complete the exercise of deciding in respect of which documents PII could properly be claimed.
Against this background, directions were sought from the court for the determination of four preliminary issues. On 24 September 2009, Burnett J ordered that the first of these issues should be tried first. This was the issue which, in its final form, is the subject of the present appeal.
After the decision of the Court of Appeal, but before the appeal came on for hearing before the Supreme Court, the claims were settled on confidential terms. A question therefore arose as to whether the court should permit the appeal to continue. It raises an important point of principle which was the subject of a full and carefully reasoned decision of the Court of Appeal. In my view, it was right to entertain the appeal. Having had the benefit of full legal argument over a period of two days, I am in no doubt that the Supreme Court should decide the issue raised by the preliminary issue so far as it is able to do so.
The positions of the parties in outline
The appellants submit that the right to a fair trial is absolute, but the means of satisfying that right vary according to the circumstances of the case. The procedures of the court are the means of achieving real justice between the parties. As a general rule, real justice and a fair trial can only be achieved by open hearings, open disclosure, each side confronting the other's witnesses and open judgments. But in certain circumstances, a closed procedure may be necessary in order to achieve real justice and a fair trial. Such procedures are adopted in certain classes of case (for example, cases involving children and confidential information). There is no reason in principle why in the exercise of its inherent jurisdiction the court should not be able to order such a procedure in other classes of case, such as cases where a defendant cannot deploy its defence fully (or sometimes not at all) if it is required to follow an open procedure. The appellants' primary case is that a court has the power to substitute, at least in exceptional cases, a closed material procedure for a conventional PII exercise.
The respondent says that open procedures are fundamental to our system of justice. His case is that a closed material procedure would be such a fundamental change to the way in which ordinary civil litigation (including judicial review) is conducted that it should not be introduced by the courts. Any such change can only be made by Parliament.
The essential features of a common law trial
There are certain features of a common law trial which are fundamental to our system of justice (both criminal and civil). First, subject to certain established and limited exceptions, trials should be conducted and judgments given in public. The importance of the open justice principle has been emphasised many times: see, for example,
R v Sussex Justices, Ex p McCarthy  1 KB 256, at p 259, per Lord Hewart CJ,
Attorney General v Leveller Magazine Ltd  AC 440, at pp 449H-450B, per Lord Diplock, and recently
R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs (No 2) (Guardian News and Media Ltd intervening)  QB 218, paras 38-39, per Lord Judge CJ.
The open justice principle is not a mere procedural rule. It is a fundamental common law principle. In Scott v Scott  AC 417, Lord Shaw of Dunfermline (p 476) criticised the decision of the lower court to hold a hearing in camera as "constituting a violation of that publicity in the administration of justice which is one of the surest guarantees of our liberties, and an attack upon the very foundations of public and private security." Lord Haldane LC (p 438) said that any judge faced with a demand to depart from the general rule must treat the question "as one of principle, and as turning, not on convenience, but on necessity".
Secondly, trials are conducted on the basis of the principle of natural justice. There are a number of strands to this. A party has a right to know the case against him and the evidence on which it is based. He is entitled to have the opportunity to respond to any such evidence and to any submissions made by the other side. The other side may not advance contentions or adduce evidence of which he is kept in ignorance. The Privy Council said in the civil case of Kanda v Government of Malaya  AC 322, 337:
If the right to be heard is to be a real right which is worth anything, it must carry with it a right in the accused man to know the case which is made against him. He must know what evidence has been given and what statements have been made affecting him: and then he must be given a fair opportunity to correct or contradict them.
Another aspect of the principle of natural justice is that the parties should be given an opportunity to call their own witnesses and to cross-examine the opposing witnesses. As was said by the High Court of Australia in Lee v The Queen (1998) 195 CLR 594, at para 32:
|Confrontation and the opportunity for cross-examination is of central significance to the common law adversarial system of trial.|
I do not believe that any of this is controversial, but it needs to be emphasised because, unlike the law relating to PII, a closed material procedure involves a departure from both the open justice and the natural justice principles. In recent years, both the courts and Parliament have been exercised by the problem of how to balance
the interest that we all have in maintaining a fair system of justice which, so far as possible, respects the essential elements of these principles and
the interest that we also all have in the protection of national security, the international relations of the United Kingdom and the prevention, detection and prosecution of crime.
Thus, Parliament has reacted to the threat of terrorism to our national security interests by introducing a form of closed material procedure (with the use of special advocates) for use in certain categories of case, for example, by enacting the Prevention of Terrorism Act 2005 and the Counter-Terrorism Act 2008.
A striking example of a case where the court had to balance these two competing interests is Carnduff v Rock  EWCA Civ 680,  1 WLR 1786. A registered police informer brought an action against the police to recover payment for information and assistance provided to the police. The defendants denied any contractual liability to make such payments or that any information or assistance provided by the plaintiff had led to the arrests or the prosecutions alleged. The claim was struck out by the Court of Appeal on the grounds that a fair trial of the issues raised by the pleadings would require the police to disclose, and the court to investigate and adjudicate upon, sensitive information which should in the public interest remain confidential to the police. Laws LJ (with whom Jonathan Parker LJ agreed) said at para 36:
.... it is to my mind inevitable that the court's duty would be to hold that the public interest in withholding the evidence about it outweighed the countervailing public interest in having the claim litigated on the available relevant evidence. In reality such a position could only be avoided if the police made comprehensive admissions which absolved the court from the duty to enter into any of these issues. But a case which can only be justly tried if one side holds up its hands cannot, in truth, be justly tried at all.
This is the only case that was cited to us in which the court has decided that the public interest in maintaining confidential information trumps the public interest in the administration of justice to the extent that on that ground a trial has been denied altogether.
This is the background against which the important issues raised by this appeal fall to be considered.
The inherent power of the court to regulate its own procedure
A distinction should be made at the outset between the court
exercising its inherent power to control its own procedure and
exercising its general power to develop the substantive common law incrementally.
We are not here concerned with (ii), a paradigm example of which would be the incremental development by the courts of the law of negligence. We are concerned with (i). In his seminal article "The Inherent Jurisdiction of the Court", Current Legal Problems 1970, Sir Jack Jacob said: "the source of the inherent jurisdiction of the court is derived from its nature as a court of law, so that the limits of such jurisdiction are not easy to define, and indeed appear to elude definition." But there is no doubt that the court's inherent power to regulate its own procedures is not unlimited. For example, the power may not be exercised in contravention of legislation or rules of court. In the words of Sir Jack Jacob, loc cit at p 24: "the court may exercise its inherent jurisdiction even in respect of matters which are regulated by statute or by rule of court, so long as it can do so without contravening any statutory provision." In such a case, its power has been removed by statute and cannot be exercised.
In proceedings which are not regulated by statute or statutory rules, it might be thought that there are no limits to the inherent power of the court to regulate its own procedure and that it has an untrammelled power to manage litigation in whatever way it considers necessary or expedient in the interests of justice.
There are many examples of the court in the exercise of its inherent power introducing procedural innovations in the interests of justice. Thus it invented the power to grant Mareva injunctions and make Anton Piller orders. These orders were devised to prevent misuse of the court's procedure and to ensure that its procedure is effective. The PII procedure was also a creature of the common law devised by the court in the exercise of its inherent power to regulate its own procedures. The remedy of discovery (now known as disclosure) was developed by the courts of equity in order to aid the administration of justice. Upon the amalgamation of the Court of Chancery and the common law courts into the High Court by the Judicature Acts, that remedy came to be governed by the Rules of Court. It is now contained in CPR Part 31. The rules governing disclosure recognised that conflict may arise between the public interest in the administration of justice and other public interests which preclude the disclosure of all relevant materials. The law of PII was developed to deal with such situations. The court was exercising its inherent power in controlling its own procedures by deciding the scope of disclosure in cases involving confidential material. The scope of disclosure has long been seen as a matter on which the court has jurisdiction to decide.
But even in an area which is not the subject of statute or statutory procedural rules, there are limits to the court's inherent jurisdiction to regulate how civil and criminal proceedings should be conducted. In my view, there is considerable force in what Professor Martin Dockray said in "The Inherent Jurisdiction to Regulate Civil Proceedings" (1997) 113 LQR 120, 131:
.... a matter which is procedural from the position of an applicant may be constitutional in the eyes of the respondent. The fact that procedural law can be described as subordinate or adjectival because it aims to give effect to substantive rules should not conceal the truth that procedures can and do interfere with important human rights, while the means by which a decision is reached may be just as important as the decision which is made in the end. Where procedure is as important as substance, procedural change requires the same degree of political accountability and economic and social foresight as reform of an equivalent rule of substantive law. Major innovations in procedural law should therefore be recognised as an institutional responsibility, not a matter on which individual judges should respond to the pleas of particular litigants. Procedural revolutions should appear first in statutes or in the Rules of Court, not in the law reports.
For example, it is surely not in doubt that a court cannot conduct a trial inquisitorially rather than by means of an adversarial process (at any rate, not without the consent of the parties) or hold a hearing from which one of the parties is excluded. These (admittedly extreme) examples show that the court's power to regulate its own procedures is subject to certain limitations. The basic rule is that (subject to certain established and limited exceptions) the court cannot exercise its power to regulate its own procedures in such a way as will deny parties their fundamental common law right to participate in the proceedings in accordance with the common law principles of natural justice and open justice. To put the same point in a different way, the court must exercise the power to regulate its procedure in a way which respects these two important principles which are integral to the common law right to a fair trial.
Is there a common law power to require a closed material procedure?
Mr Crow QC submits that the PII system suffers from five serious defects. These are that
the balancing exercise is inherently difficult to perform, because the two public interests are fundamentally different (it involves, as he put it, a comparison of apples and pears);
if the balance is struck against disclosure, relevant evidence is excluded from the trial (thereby reducing the chances that the court will reach the right result);
the party holding the sensitive undisclosed material knows its contents which may inform its preparation and/or conduct of the trial, thereby putting it at an unfair advantage over the other party;
if the balance is struck in favour of disclosure, the party holding the material that is ordered to be disclosed is faced with the invidious choice of disclosing the material, despite the harm that this would or might cause to the public interest, or refusing disclosure and facing the possibility of having to concede the whole of the other side's case. As Carnduff demonstrates, the court may be faced with the stark choice between depriving a defendant of the means to defend himself and striking out the claim on the grounds that it is untriable; and
the scale of the PII exercise is sometimes so vast that it can take years to complete.
As regards (v), as already pointed out, it is said that in the present case as many as 140,000 documents may have to be the subject of the PII process: a massive and expensive task which will inevitably mean that it will not be possible for the trial to take place for a very long time.
Mr Crow submits that the closed material procedure does not suffer from these defects. If such a procedure is adopted, the court is able to see and take into account at trial relevant material (written and oral) whose disclosure would, applying the PII principles, be excluded from the trial process altogether, regardless of whether the material is of assistance to the claimant or the defendant. If the balancing exercise favours non-disclosure, the defendant may have no adequate basis to defend itself in reliance on the open material when, if the truth were known and the sensitive material were referred to, it had a complete answer to the claim. The court would be able to review relevant material in the overall interests of justice and to do so with the assistance of special advocates who would be able to make submissions on behalf of the party from whom open disclosure was being withheld. In short, in an appropriate case, a closed procedure is more likely than PII to achieve justice through a fair trial.
Since the shortcomings in the PII process to which Mr Crow draws attention are inherent in the process and are of general application, the logic of his arguments ought to lead to the conclusion that the court should exercise its power in most if not all cases to adopt a closed material procedure rather than PII. But his submission is more modest. He seeks to secure from the court no more than an acknowledgement that, in principle, the court has the power to adopt the closed material procedure and that it should exercise that power in exceptional cases where this is necessary in the interests of justice. He derives the necessity test from Scott v Scott per Lord Haldane LC at p 436.
Scott v Scott was addressing the very important principle that justice should be administered in public and recognised that there may be a departure from that principle where that is necessary in the interests of justice.
It is one thing to say that the open justice principle may be abrogated if justice cannot otherwise be achieved. As Lord Bingham of Cornhill said in R v Davis  UKHL 36,  1 AC 1128 at para 28, the rights of a litigating party are the same whether a trial is conducted in camera or in open court and whether or not the course of the proceedings may be reported in the media. It is quite a different matter to say that the court may sanction a departure from the natural justice principle (including the right to be present at and participate in the whole or part of a trial). Scott v Scott is no authority for such a proposition. How can such a step ever satisfy the requirements of justice? And if the court does have the power to deny a litigant this fundamental common law right, in what circumstances is it appropriate to exercise it? These are the questions that lie at the heart of this appeal.
Before attempting to answer these questions, I think that it is instructive to consider Davis in more detail. It concerned the question whether the judge at a criminal trial could permit witnesses to give evidence for the prosecution under conditions of anonymity. The perceived need for anonymity arose because, owing to threats of intimidation, the witnesses would not be willing to give their evidence without it. In the Court of Appeal  1 WLR 3130, para 13, Sir Igor Judge P said that the court possesses an inherent jurisdiction at common law to control its own proceedings, if necessary by adapting and developing its existing processes "to defeat any attempted thwarting of its process": see per Lord Morris of Borth-y-Gest in Connelly v Director of Public Prosecutions  AC 1254, 1301. The court concluded that the anonymity ruling did not prevent proper investigation with the witnesses in open court of the essential elements of the defence.
The House of Lords disagreed. Lord Bingham referred at para 5 to the "long-established principle of the English common law that, subject to certain exceptions and statutory qualifications, the defendant in a criminal trial should be confronted by his accusers in order that he may cross-examine them and challenge their evidence". The authorities to which Lord Bingham referred at para 5 in support of this proposition were in the field of both criminal and civil law. Thus, for example, in 1720 in a civil case the court declared in Duke of Dorset v Girdler (1720) Prec Ch 531, 532 that "the other side ought not to be deprived of the opportunity of confronting the witnesses, and examining them publicly, which has always been found the most effectual method for discovering of the truth."
In rejecting the conclusion of the Court of Appeal, Lord Bingham said at para 34 that at no point in its judgment did the Court of Appeal "acknowledge that the right to be confronted by one's accusers is a right recognised by the common law for centuries, and it is not enough if counsel sees the accusers if they are unknown to and unseen by the defendant."
It is worthy of note that the House of Lords reached its decision in the face of arguments advanced on behalf of the Crown which included
the problem of witness intimidation is real and prevalent and, unless witnesses are allowed to give evidence under conditions of anonymity, dangerous criminals will walk free and society and the administration of justice will suffer;
as Lord Haldane LC said in Scott v Scott, the paramount object must always be to do justice and if, in order to do justice, some adaptation of ordinary procedure is called for, it should be made, so long as the overall fairness of the trial is not compromised; and
the Strasbourg jurisprudence does not condemn the use of measures to protect witnesses.
These arguments did not prevail. The common law right to be confronted by one's accusers (an essential element of a fair trial) could not be abrogated by the courts. Any such abrogation was a matter for Parliament. Lord Bingham said at para 28 that it was pertinent to recall the observations of Lord Shaw in Scott at pp 477-478: "There is no greater danger of usurpation than that which proceeds little by little, under cover of rules of procedure, and at the instance of judges themselves." Lord Rodger of Earlsferry said at para 45 that Parliament was the proper body to decide whether a change in the law on the way that witnesses gave their evidence where intimidation is a problem was required, and, if so, to devise an appropriate system which still ensures a fair trial. Lord Brown of Eaton-under-Heywood said at para 66 that it was for the government to decide whether to legislate in the field. Meanwhile, the emasculation of the common law principle must not only be halted but reversed. And Lord Mance made similar observations at para 98.
Lord Brown at para 78 of his judgment in the present case suggests that there is a real difference between this case and Davis in that legislation to meet the problem was envisaged in Davis, whereas it is highly doubtful whether Parliament will legislate speedily to introduce a closed procedure. But I do not think that the House of Lords' principled refusal to abrogate the common law right to be confronted by one's accusers can be explained by reference to a belief that Parliament would speedily intervene to resolve the problem.
The House did not state whether the protective measures imposed by the trial judge were ones which the court had no power to impose, or whether, although the power existed, it was inappropriate to exercise it. That is no doubt because it makes no practical difference. Lord Bingham was content to say at para 35 that the protective measures "hampered the conduct of the defence in a manner and to an extent which was unlawful and rendered the trial unfair".
It is not difficult to see the similarities between the arguments that were rejected by the House of Lords in Davis and those advanced by Mr Crow. The problem here is not that dangerous criminals will walk free. It is that in sensitive material cases, parties will not be able to develop their true case on the basis of all the relevant material with the result that parties will sometimes lose cases that they should win. A defendant who is ordered to disclose sensitive material on a PII hearing has the invidious choice of disclosing material that will damage some important public interest or making admissions and, in an extreme case, conceding the claim.
I return to the questions that lie at the heart of this appeal. In my view, the analogy with Davis is compelling. As I have said, the fact that Davis was a criminal case is not material. The issues considered were of application to trials generally. It decided that, subject to certain exceptions and statutory qualifications, the right to be confronted by one's accusers is such a fundamental element of the common law right to a fair trial that the court cannot abrogate it in the exercise of its inherent power. Only Parliament can do that. The closed material procedure excludes a party from the closed part of the trial. He cannot see the witnesses who speak in that part of the trial; nor can he see closed documents; he cannot hear or read the closed evidence or the submissions made in the closed hearing; and finally he cannot see the judge delivering the closed judgment nor can he read it.
Can all of these flaws be cured by a special advocate system? No doubt, special advocates can mitigate these weaknesses to some extent and in some cases the litigant may be able to add little or nothing to what the special advocate can do. For example, this will be the case where the litigant has no knowledge of the matters to which the closed material relates and can give no instructions which will enable the special advocate to perform his function more effectively. But in many cases, the special advocate will be hampered by not being able to take instructions from his client on the closed material. A further problem is that it may not always be possible for the judge (even with the benefit of assistance from the special advocate) to decide whether the special advocate will be hampered in this way.
The limitations of the special advocate system, even in the context of the statutory contexts for which they were devised, were highlighted by the Joint Committee on Human Rights in their report on Counter-Terrorism Policy and Human Rights (Sixteenth report): Annual Renewal of Control Orders Legislation 2010 (HL Paper 64/HC 395) (dated 26 February 2010) in the context of the Prevention of Terrorism Act 2005 and cases heard by the Special Immigration Appeals Commission. This report was based on the first-hand experience of those who have acted as special advocates. As the Court of Appeal noted at para 57, it is the Committee's view after five years of operation that the closed material procedure (with special advocates) operated under the statutory regimes is not capable of ensuring the substantial measure of procedural justice that is required. At para 210 of its earlier report, HL Paper 157, HC 394, (published on 30 July 2007), the Committee had concluded:
After listening to the evidence of the Special Advocates, we found it hard not to reach for well worn descriptions of it as 'Kafkaesque' or like the Star Chamber. The Special Advocates agreed when it was put to them that, in the light of the concerns they had raised, 'the public should be left in absolutely no doubt that what is happening .... has absolutely nothing to do with the traditions of adversarial justice as we have come to understand them in the British legal system.' Indeed, we were left with the very strong feeling that this is a process which is not just offensive to the basic principles of adversarial justice in which lawyers are steeped, but it is very much against the basic notions of fair play as the lay public would understand them.
These views may not sufficiently take account of specific statutory protections, (such as those set out in rule 54 of the Employment Tribunals Rules of Procedure contained in Schedule 1 to the Employment Tribunals (Constitution and Rules of Procedure) Regulations 2004 (SI 2004/1861)) to which Lord Mance refers at para 10 of his judgment in Tariq v Home Office  UKSC 35 but they do throw light on the limitations of the special advocate system.
In my view, Mr Crow provided no satisfactory answer to the questions (i) why it is necessary to introduce a procedure which would deprive a litigant of his fundamental common law rights or (ii) in what circumstances the power to order a closed material procedure should be exercised. Since he disavows the wholesale replacement of PII with the closed procedure, it is difficult to see the relevance of the defects alleged by him in the one and the advantages he claims for the other. As regards the problem of the time-consuming exercise that has to be performed in a PII process, I agree with what Lord Clarke says at para 152 of his judgment: a closed procedure would not achieve any saving.
Mr Crow's answer to the two questions that I have posed is that the court has the power to order a closed material procedure in exceptional cases where this is necessary in the interests of justice. But this simply prompts the further question: in what circumstances can it ever be in the interests of justice to deny a litigant in ordinary civil claims (including claims for judicial review) the rights which are entrenched in our common law system as being fundamental requirements of justice itself? I deal with special cases at paras 63 to 65 below. Mr Crow did not give a concrete example of a case where the court would or might exercise this exceptional power. This amply demonstrates that the test that he proposes is vague. It is likely to cause uncertainty in the minds of litigants and their advisers and to lead to unwelcome satellite litigation.
Lord Clarke agrees that the court has the power to order a closed procedure where this is necessary in the interests of justice, and at paras 159 to 164 and 178 to 181 he suggests circumstances in which it may be appropriate to exercise the power. He says that, in the exercise of its inherent jurisdiction, the court has the power after the PII process has been completed to order some form of closed procedure involving special advocates. Thus the claimants might seek a closed procedure if they thought that there were advantages in such a procedure, especially if their case was thought to depend to any significant extent on documents in the possession of the defendants. The defendants might also wish to have such a procedure depending on the circumstances. Lord Clarke says that the court has adopted novel procedures in not dissimilar situations by way of development of the common law and pursuant to its obligation under CPR Part 1 to deal with cases justly. There is no reason why the common law should not be able to develop along these lines. It would be a "development of the common law of PII".
I respectfully disagree with this approach. First, no form of closed material procedure can properly be described as a development of the common law of PII, although there is no objection to the use of special advocates to enhance the PII process (see para 49 below). In many ways, a closed procedure is the very antithesis of a PII procedure. They are fundamentally different from each other. The PII procedure respects the common law principles to which I have referred. If documents are disclosed as a result of the process, they are available to both parties and to the court. If they are not disclosed, they are available neither to the other parties nor to the court. Both parties are entitled to full participation in all aspects of the litigation. There is no unfairness or inequality of arms. The effect of a closed material procedure is that closed documents are only available to the party which possesses them, the other side's special advocate and the court. I have already referred to the limits of the special advocate system.
Secondly, it is obviously true that party A who is in possession of the closed material will know whether there is material on which it may wish to rely and will therefore be in a position to decide whether to ask the court to order a closed procedure in relation to that material. But it is difficult to see how opposing party B will know whether his case will be assisted by, or even depend to a significant extent on, the closed material held by A without knowing what the material is and what it contains. If a special advocate is appointed, he might be able to assess the importance of some of the documents, but the scope for doing so without being able to take instructions from B is bound to be limited. It follows that, if the power to order a closed material procedure turns on such considerations, it is likely to operate in favour of A and to the disadvantage of B. In my view, this is an approach which is inherently unfair. It is certainly not necessary in the interests of justice.
Thirdly, it is difficult to see how on the suggested approach the court would be able to judge whether to order a closed procedure in any particular case. Would the court be called upon to decide whether the closed documents are likely to advance the case of one or other of the parties and, if so, how would it do this and what test would it apply? Would the material have to be crucial to the case of the party seeking the closed procedure or would it be sufficient that it provides some support for it? This is not an exercise that the court should be required to perform. It would be likely to give rise to argument about whether the closed material is or is not likely to help the cause of the party seeking to invoke the procedure. It would be a recipe for satellite litigation. It would merely add to the complexity and expense of the whole process.
Fourthly, to allow a closed procedure in circumstances which are not clearly defined could easily be the thin end of the wedge. This is the point that was made by Lord Shaw in his celebrated speech in Scott v Scott (see para 31 above). Mr Crow's undefined exceptional circumstances in the interest of justice could develop into something more defined and exorbitant. So too could Lord Clarke's suggested approach. This would be a big step for the law to take in view of the fundamental principles at stake. In my view, this is a matter for Parliament and not the courts.
Fifthly, like Lord Clarke (subject to the one qualification that he mentions as to which I express my views at paras 60 and 61 below), I accept the fundamental principles stated by the Court of Appeal at para 70 of its judgment which he has quoted at para 167 below. But in my view these principles do not sit happily with what he proposes. A closed procedure in the circumstances that he suggests would cut across the fundamental principles of the right to a fair trial and the right to know the reasons for the outcome. It would complicate a well-established procedure for dealing with the problem, namely the PII procedure. And for the reasons that I have given would be likely to add to the uncertainty, cost, complexity and delay of all stages of the litigation.
As I understand it, Lord Mance (with whom Lady Hale agrees) adopts the view of Lord Clarke that the closed material procedure should not be an alternative to PII, but that it may be ordered in addition to PII in certain circumstances. They differ, however, over the circumstances in which a closed procedure may be ordered. Lord Mance limits himself to cases where closed material is in the defendant's possession and the claimant, in order to avoid his or her claim being struck out, consents to a closed material procedure. The differences between them show that different views may be held as to how the test of "necessity" should be applied in this context. These differences lend further support to the view that, if a closed material procedure is to be available in ordinary civil claims, the decision as to when it might be "necessary" for such a procedure to be used should be left to Parliament. I should make it clear that, like the Court of Appeal (para 71), I leave open the question whether a closed material procedure can properly be adopted where the parties agree. We heard no argument on this point.
Closed material procedures and the use of special advocates continue to be controversial. In my view, it is not for the courts to extend such a controversial procedure beyond the boundaries which Parliament has chosen to draw for its use thus far. It is controversial precisely because it involves an invasion of the fundamental common law principles to which I have referred. I would echo what Lord Phillips said in Secretary of State for the Home Department v AF (No 3)  UKHL 28,  2 AC 269. It is true that this was a case concerning the requirements of a fair trial under article 6 of the European Convention on Human Rights, but in my view it is equally applicable in relation to the common law requirements of a fair trial. At para 64, he said that the best way of producing a fair trial is to ensure that a party has the fullest information of the allegations against him and the evidence (both oral and documentary) that is relied on in support of those allegations. Both our criminal and civil procedures set out to achieve those aims. In some circumstances, however, they run into conflict with other aspects of the public interest. He then said:
How that conflict is to be resolved is a matter for Parliament and for government, subject to the law laid down by Parliament.
The common law principles to which I have referred are extremely important and should not be eroded unless there is a compelling case for doing so. If this is to be done at all, it is better done by Parliament after full consultation and proper consideration of the sensitive issues involved. It is not surprising that Parliament has seen fit to make provision for a closed material procedure in certain carefully defined situations and has required the making of detailed procedural rules to give effect to the legislation.
There is no compelling reason for taking the course that is urged by Mr Crow or that which is suggested by Lord Clarke. The PII process is not perfect. Perfection cannot be achieved in any system. It has been improved over time as the history of its development shows. One particular development to note is the use of special advocates to enhance the PII process. There can be no objection to the use of special advocates for that purpose, since the PII process fully respects the principles of open justice and natural justice. There is nothing objectionable about excluding a party from the PII process. There can, therefore, be no objection to improving the position of that party in the process by the use of a special advocate.
It is true that, if a closed material procedure were introduced, it might not be necessary to strike out a claim such as Carnduff. Looked at in isolation, that would be a good thing. But the problem cannot be looked at so narrowly and in any event it seems that cases such as Carnduff are a rarity. They do not pose a problem on a scale which provides any justification (let alone any compelling justification) for making a fundamental change to the way in which litigation is conducted in our jurisdiction with all the attendant uncertainties and difficulties that I have mentioned.
As in the Court of Appeal, so here Mr Crow relies on a number of previous authorities where the use of special advocates in a closed material procedure was approved. These were discussed by the Court of Appeal at paras 58 to 66 of their judgment. Lord Clarke suggests at para 166 below that the Court of Appeal distinguished these cases on the basis that they were not ordinary "civil" claims. But that was not the only reason given by the Court of Appeal for refusing to follow those decisions. In none of the cases was proper consideration given to the question whether a closed material procedure was a permissible development of the common law.
Thus, in Secretary of State for the Home Department v Rehman  UKHL 47,  1 AC 153, para 31 Lord Woolf MR stated (obiter) that, albeit only "in the most extreme circumstances", the Court of Appeal could hear submissions in the absence of a party and his counsel under the inherent jurisdiction of the court on the basis that the party's interests could be protected by a special advocate. But there is no suggestion that the contrary was argued.
R v Shayler  UKHL 11,  1 AC 247 was concerned with a preparatory hearing in relation to a defendant who was charged with unauthorised disclosure of material under the Official Secrets Act 1989. A question arose as to whether the defendant could disclose the material to his lawyers. At para 34, Lord Bingham said that, following what Lord Woolf said in Rehman, in the unlikely event of the court having to consider the material which could not be disclosed to the defendant's lawyers, a special advocate could be appointed. As the Court of Appeal said at para 61, there would have been no question of the defendant himself being in ignorance of the material or of his being excluded from the hearing where it was considered. It was, therefore, not a closed material procedure at all. In any event, what Lord Bingham said was based on para 31 of Rehman and does not appear to have been the subject of contrary argument.
In R v H  UKHL 3,  2 AC 134 at para 22, Lord Bingham made some observations about the use of a special advocate in a PII procedure. But as discussed at para 49 above, that is not the kind of closed material procedure with which we are concerned.
In R (Roberts) v Parole Board  UKHL 45,  2 AC 738, the Parole Board decided that evidence on which the Home Secretary intended to rely should be heard in the absence of, and not disclosed to, the claimant or his legal representatives, since to do so would put an informant at risk. Instead they directed that the evidence should be disclosed only to a special advocate. The Board had express power under the relevant rules to withhold material which otherwise had to be served on a prisoner. In other words, there was express statutory power to adopt a closed material procedure. One of the issues was whether there was power to require a special advocate. The Board had power under paragraph 1(2)(b) of Schedule 5 to the Criminal Justice Act 1991 to do anything incidental or conducive to the discharge of its functions. It was held that appointing a special advocate to mitigate the adverse effects of the closed procedure would be an exercise of that power. Roberts, therefore, provides no basis for concluding that a court may determine a civil claim on its merits using a closed material procedure in the absence of a specific enactment authorising it to do so.
In R (Malik) v Manchester Crown Court  EWHC 1362 (Admin),  4 All ER 403, when giving the judgment of the Divisional Court, I reviewed a number of the authorities about the use of special advocates. But as the Court of Appeal said at para 64, no argument was advanced that the Crown Court had no power to order a closed material procedure in the absence of an enactment enabling it to do so. The sole question was whether the court should have appointed a special advocate of its own motion.
In A v HM Treasury  EWCA Civ 1187,  3 WLR 25, when giving the majority judgment of the Court of Appeal, Sir Anthony Clarke MR said at paras 58 and 60 that in an appropriate case the court would have power to authorise or request the use of a special advocate even where this was not sanctioned by Parliament. But here too there does not appear to have been any argument as to whether the court had power to order a closed material procedure in the first place.
Finally, in R (AHK) v Secretary of State for the Home Department (Practice Note)  EWCA Civ 287,  1 WLR 2049 at paras 37-38, Sir Anthony Clarke MR giving the judgment of the Court of Appeal gave guidance as to the circumstances in which a special advocate could be appointed. Here too, there was no debate as to whether the court could order a closed material procedure. It seems to have been assumed that it could.
None of these authorities is binding on this court. In none of them was there any consideration of the issues which have been considered in detail over two days of argument by this court. They do not amount to a secure and established line of authority to support the proposition that the court has power to order a closed material procedure in the absence of statutory authority.
The Civil Procedure Rules
At para 43, the Court of Appeal said that they considered that the defendants faced "very serious difficulties in their contention that the closed material procedure is compatible with the CPR". They explained why this was the case at paras 41 to 48. I see the force of the points made by the Court of Appeal and they are supported by Ms Rose QC and Mr Howell QC. In short, there are detailed rules for the filing and service of a defence (CPR r 15.2 and CPR r 16.5); disclosure (CPR r 31.5); giving of evidence orally (CPR r 32.2) and the provision of witness statements (CPR r 32.4 and 32.5). There is no provision for the filing and service of a closed defence or for closed disclosure or the giving of evidence in a closed hearing or the provision of closed witness statements. By contrast, CPR Parts 76 and 79 explicitly modify or disapply those Parts of the CPR for the purpose of the particular proceedings in which Parliament has decided that a closed material procedure may be used.
These points based on the CPR provide some further support for the conclusion which I have reached for other reasons. The rules make no provision for a closed procedure except in circumstances where it is authorised by statute. On the face of it, the general rules are inconsistent with a closed material procedure. But I do not consider that, if the argument based on the CPR stood alone, it would have been sufficient to carry the day for the respondent. It is not sufficiently clear that a closed material procedure would contravene the CPR to say that on that account the court has no power to order such a procedure.
"Ordinary civil claims"
I agree with Lord Clarke, for the reasons that he gives, that there can be no principled basis for distinguishing between ordinary civil claims and claims for judicial review. I would accept the submission of Mr Howell that the mere fact that there may be a public interest involved in the determination of a case does not mean that the court may disregard the duty imposed on it by the law relating to PII or may override the fundamental rights of a party to civil litigation recognised at common law.
But I agree that there are certain classes of case where a departure from the normal rule may be justified for special reasons in the interests of justice. Thus as Baroness Hale of Richmond said in Secretary of State for the Home Department v MB  UKHL 46,  1 AC 440 at para 58:
If .... the whole object of the proceedings is to protect and promote the best interests of a child, there may be exceptional circumstances in which disclosure of some of the evidence would be so detrimental to the child's welfare as to defeat the object of the exercise.
Wardship proceedings are an obvious example of such a case: see In re K (Infants)  AC 201, per Lord Devlin at p 241A. Cases involving children raise different considerations from those which arise in ordinary civil litigation. That is because the interests of children are paramount. It follows that where the interests of the child are served, so too are the interests of justice.
Similarly, where the whole object of the proceedings is to protect a commercial interest, full disclosure may not be possible if it would render the proceedings futile. This problem occurs in intellectual property proceedings. It is commonplace to deal with the issue of disclosure by establishing "confidentiality rings" of persons who may see certain confidential material which is withheld from one or more of the parties to the litigation at least in its initial stages. Such claims by their very nature raise special problems which require exceptional solutions. I am not aware of a case in which a court has approved a trial of such a case proceeding in circumstances where one party was denied access to evidence which was being relied on at the trial by the other party.
In my view, the children and confidentiality cases cannot be relied on to justify creating a rule of general application in ordinary civil litigation. These are two narrowly defined categories of case where a departure from the usual rules of procedure has been held to be justified. So far as I am aware, the procedures adopted in these cases are not regarded as controversial and work satisfactorily. By contrast, the closed material procedure is controversial and, in some quarters at least, is regarded as unsatisfactory.
Reference was also made by Mr Crow to proceedings before the Competition Appeal Tribunal ("CAT") and, for example, the case of Carphone Warehouse Group v Office of Communications  CAT 37 where the tribunal refused disclosure for the purpose of the substantive hearing. But the CAT emphasised the distinction between the inquisitorial procedure that is adopted in statutory appeals before it and the adversarial procedure adopted in civil trials.
As Lord Clarke has emphasised, the common law is flexible. It develops over time in response to changing circumstances. Sometimes, it takes giant steps forward. More often, it evolves gradually and cautiously. But any change must be justified, otherwise the law becomes unstable. This is particularly important where a change involves an inroad into a fundamental common law right. The introduction of a closed material procedure in ordinary civil claims (including claims for judicial review) would do just that. Mr Crow suggests that the court should have the power to replace the PII process with a closed material procedure in exceptional circumstances where this is in the interests of justice. Lord Clarke suggests that the court should be able to supplement the PII process with such a procedure in exceptional circumstances. For the reasons that I have given, there is no compelling reason for change. The PII process is not perfect, but it works well enough. In some cases, it is cumbersome and costly to operate, but a closed material procedure would be no less so.
It is true that, by a majority, this court has decided in Tariq v Home Office  UKSC 35 that the use of a statutory closed material procedure before the Employment Tribunal is lawful under article 6 of the European Convention on Human Rights ("the Convention") and EU law. But the lawfulness of a closed material procedure under article 6 and under the common law are distinct questions. As Lord Bingham said in R (Ullah) v Special Adjudicator  UKHL 26,  2 AC 323 at para 20 "[i]t is of course open to member states to provide for rights more generous than those guaranteed by the Convention". It is, therefore, open to our courts to provide greater protection through the common law than that which is guaranteed by the Convention.
As the Court of Appeal said at para 69 of their judgment, "never say never" is often an appropriate catchphrase to use in the context of the common law. Nobody can predict how the law will develop in the future. We are concerned with the position as it is now. But for the reasons that I have given, I agree with the Court of Appeal that the issues of principle raised by the closed material procedure are so fundamental that a closed material procedure should only be introduced in ordinary civil litigation (including judicial review) if Parliament sees fit to do so. No doubt, if Parliament did decide on such a course, it would do so in a carefully defined way and would require detailed procedural rules to be made (such as CPR Parts 76 and 79) to regulate the procedure.
The issues in this case have been set out so clearly and so comprehensively by Lord Clarke and Lord Dyson that I can go directly to the heart of the question that is before us.
The word "never" is such a blunt and uncompromising expression that one has a natural reluctance to lean against it, especially in a case such as this where, as Lord Clarke points out in para 125, we are being asked to decide the issue without reference to the facts of any particular case. But the issue is essentially one of principle. I can see the attractions of the approach that Lord Phillips takes, which is to deal only with the preliminary issue which was before the Court of Appeal and not to engage in a consideration of the wider question as to whether it would ever be right for the closed material procedure to be introduced for ordinary civil litigation without the authority of Parliament: see paras 192 and 197. I would like him give a negative answer, for the reasons that Lord Clarke has explained, to the question whether the established PII procedure should be replaced by some sort of closed procedure at common law. I agree that such a fundamental change to that long-established procedure cannot properly be seen as a development of the common law and that it could only be brought about by Parliament. But the argument both in the Court of Appeal and before this court was addressed to the wider issue too. I do not think that it would be right for us to leave the wider issue in a state of uncertainty.
I have always believed that a court of unlimited jurisdiction is the master of its own procedure. But that does not mean that the court can do what it likes. Everything that it does must have regard to the fundamental principles of open justice and of fairness. The principle of legality demands nothing less than that. There is, of course, a very wide area of procedure where these issues of principle are not engaged at all. There comes a point, however, where the line must be drawn between procedural choices which are regulatory only and procedural choices that affect the very substance of the notion of a fair trial. Choices as to how the conduct of the court's business may be simplified, made less expensive or made easier to understand are one thing. Choices that cut across absolutely fundamental principles such as the right to a fair trial, the right to be confronted by one's accusers and the right to know the reasons for the outcome are entirely different. The court has for centuries held the line as the guardian of these fundamental principles.
I can see the force of the argument that there are circumstances where justice cannot be done unless a closed procedure is adopted. It is advanced, as Lord Brown puts it in para 81, as the least bad solution to a difficult problem. But I think that the court must resist the temptation to go down that road. It would, at best, be an uncertain journey, beset by problems of the kind that Lord Dyson refers to in para 43. It would also run the risk of opening the door to something else. As the Court of Appeal said, it is a melancholy truth that a procedure or approach which is sanctioned by the court expressly on the basis that it is applicable only in exceptional circumstances none the less often becomes common practice:  3 WLR 1069, para 69. Lord Shaw of Dunfermline's warning in Scott v Scott  AC 417, 477-478, against the usurpation of fundamental rights that proceeds little by little under the cover of rules of procedure remains just as true today as it was then. This is not the time to weaken the law's defences. On the contrary, any weakening in the face of advances in the methods and use of secret intelligence in a case such as this would be bound to lead to attempts to widen the scope for an exception to be made to the principle of open justice. That would create a state of uncertainty in an area of our law which would be inimical to the concept of a fundamental right.
The proposition that a closed material procedure should only be introduced in ordinary civil litigation if Parliament sees fit to do so should not be seen as surrendering to Parliament something which lies within the area of the court's responsibility. Instead it is a recognition that the basic question raises such fundamental issues as to where the balance lies between the principles of open justice and of fairness and the demands of national security that it is best left for determination through the democratic process conducted by Parliament, following a process of consultation and the gathering of evidence. The detailed working out of any change to the procedure that Parliament may sanction would no doubt be left to the court in the exercise of its rule-making powers. The court will, of course, be conscious of its responsibility to see that, so far as it is possible to do so, anything that Parliament enacts is read and given effect in a way that is compatible with the Convention rights.
The question whether it would be open to the court to adopt a closed material procedure if the parties agreed to this raises difficult issues on which we did not hear any argument, not the least of which is the pressure that might be brought to bear on a claimant to agree. Like Lord Kerr, I very much doubt whether this would ever be appropriate. But, as we do not need to decide it in this case, the proper course is to leave the question open for the time being.
I would dismiss the appeal for the reasons given by Lord Dyson.
I have had the advantage (a very real advantage in this particular case) of reading in draft the judgments of Lord Clarke and Lord Dyson. Lord Clarke envisages circumstances in which the courts could properly order a closed material procedure (with a special advocate) in ordinary civil litigation and so would allow the appeal. Lord Dyson agrees with the Court of Appeal that this would be impermissible without express parliamentary authorisation. I find it difficult to suppose that either viewpoint could be more persuasively expressed. Yet neither conclusion do I find entirely satisfactory. Let me try to explain.
If and in so far as the real issue before us is whether Parliament alone could provide for so fundamental an inroad into the principle of open justice as is proposed here whether, in other words, such a step is beyond the permissible development of the common law I am in Lord Dyson's camp. There seems to me at least as good a reason for saying that Parliament alone could sanction this development as for saying, as the House of Lords did in R v Davis  UKHL 36;  1 AC 1128, that Parliament alone could sanction the use of anonymous evidence in a criminal trial an invasion of the common law principle that the defendant has the right to be confronted by (or, as I preferred to put it, the right to know the identity of) his accusers. But there is to my mind a real difference between this case and Davis. This is not, as Lord Clarke suggests (at para 187), because Davis was a criminal case. Rather it is because, whereas the majority in Davis envisaged early legislation to meet the problem legislation did indeed follow within days although whether it meets the problem remains to be decided by the long-awaited judgment of the Grand Chamber in Strasbourg in Al-Khawaja v United Kingdom it seems to me highly doubtful whether Parliament will legislate speedily here and even more doubtful whether the introduction of a closed procedure, certainly along the lines both Lord Clarke and Lord Dyson appear to envisage, would indeed meet the very real problems that arise in cases like this.
I speak of "cases like this" because it seems to me quite impossible to discuss the issue raised here on an entirely abstract basis and really only useful to do so in a specific context the very context, I would suggest, in which the question here was raised: a claim against the Intelligence Services and their sponsoring departments of the nature summarised by Lord Clarke (at para 132 of his judgment), essentially alleging complicity in the claimants' extraordinary rendition, false imprisonment, torture and other ill-treatment, involving (as Lord Clarke notes at para 135) up to 250,000 potentially relevant documents of which up to 140,000 may involve considerations of national security - so that a conventional public interest immunity process, notwithstanding the employment of 60 lawyers specifically for the purpose, would be likely to take upwards of three years.
Both Lord Clarke and Lord Dyson envisage that, before any question could arise of introducing a closed procedure into the proceedings (save in so far as necessary to improve the PII process itself, as recognised by the House of Lords in R v H  UKHL 3,  2 AC 134 see Lord Clarke at para 150 and Lord Dyson at paras 49 and 54) the conventional PII process would have to be completed with first the Minister (see R v Chief Constable of West Midlands Police, Ex p Wiley  1 AC 274) and then the court striking the relevant balances. Lord Clarke makes this plain at paras 152 and 153, Lord Dyson at para 38. Both, in short, appear to envisage that the only real object of the closed procedure proposed here would be to enable a party, wishing to rely on documents held to be undisclosable, and thus inadmissible, in the public interest, to rely on them in closed proceedings. In all likelihood, of course, such a procedure would be invoked by the Crown in just such a case as this where many of the relevant documents would almost inevitably be undisclosable in the interests of national security. (Although theoretically it would be open to claimants to seek this procedure in order, first, through their special advocate, to examine the undisclosed documents and then, if thought favourable to their case, to deploy them in closed proceedings, frankly this appears a somewhat unrealistic prospect.)
The advantage of this approach, suggests Lord Clarke (at para 159), is that it avoids both the unpalatable alternative outcomes: either the trial proceeding without the undisclosable documents (generally, it must be supposed, to the Crown's considerable disadvantage) or the action being struck out on the basis that without the disclosed documents a fair trial is not possible (as in Carnduff v Rock  EWCA Civ 680,  1 WLR 1786). In short, contend its proponents, it is the least bad solution to a difficult problem.
For my part, however, I am unpersuaded of this. In the first place, it offers no solution at all to the very real problems of having to conduct a conventional PII process in a case like this. To my mind there need to be compelling reasons to justify the enormous expense, effort and delay involved in such a process here. Secondly, the problem is surely not confined to the disclosure, and thus admissibility, of documents. What about oral evidence? Presumably in a case like this the Crown would wish, indeed need, to call witnesses from the Security Service and the Secret Intelligence Service. Documentation aside, how is it suggested that such evidence could satisfactorily be adduced at a public hearing (indeed, any hearing attended by the claimants themselves)? Thirdly, any closed material procedure with special advocates raises problems all its own, considerations of open justice apart. As the Court of Appeal observed at para 70(e) of its judgment an observation expressly concurred in by both Lord Clarke (para 168) and Lord Dyson (para 45) the envisaged closed material procedure "is likely to add to the uncertainty, cost, complication and delay in the initial and interlocutory stages of proceedings, the trial, the judgment, and any appeal". Further, as Lord Dyson observes at para 43, it would be a recipe for satellite litigation. And on top of all this, and by no means least in importance, is the whole question of open justice.
One need not take so extreme a view as that expressed by the Joint Committee on Human Rights last year (see Lord Dyson's judgment at para 37) to recognise the grave inroads into our fundamental principles of open justice and fair trials that are made by closed procedures. Without "A-type disclosure" (see my judgment in Tariq v Home Office  UKSC 35, at para 86), the claimants may not learn sufficient of the case against them to enable them to give effective instructions to the special advocate to meet it. With such disclosure, however, national security may still be put at risk. Whatever the extent of disclosure, moreover, it will be difficult for the claimants' lawyers to advise on the merits (and difficult for the same reason to secure public funding), difficult similarly to advise on settlement offers, and difficult too to advise on any appeal. But beyond all these considerations would be the damage done by a closed procedure to the integrity of the judicial process and the reputation of English justice. As Lord Atkin, invoking Milton's Areopagitica, famously said in Ambard v Attorney General for Trinidad and Tobago  AC 322, 335: "Justice is not a cloistered virtue: she must be allowed to suffer the scrutiny and respectful, even though outspoken, comments of ordinary men." A closed procedure in the present context would mean that claims concerning allegations of complicity, torture and the like by UK Intelligence Services abroad would be heard in proceedings from which the claimants were excluded, with secret defences they could not see, secret evidence they could not challenge, and secret judgments withheld from them and from the public for all time. As the Court of Appeal observed below (at para 56):
If the court was to conclude after a hearing, much of which had been in closed session attended by the defendants but not the claimants or the public, that for reasons, some of which were to be found in a closed judgment that was available to the defendants but not the claimants or the public, that the claim should be dismissed, there is a substantial risk that the defendants would not be vindicated and that justice would not be seen to have been done. The outcome would be likely to be a pyrrhic victory for the defendants whose reputation would be damaged by such a process, but the damage to the reputation of the court would in all probability be even greater.
Lord Clarke (para 161) understands it to be common ground that there could be no objection to a closed procedure were the parties to agree to it (as claimants might, were the only alternative to be the striking out of their claims). For my part I respectfully disagree. The rule of law and the administration of justice concern more, much more, than just the interests of the parties to litigation. The public too has a vital interest in the conduct of proceedings. Open justice is a constitutional principle of the highest importance. It cannot be sacrificed merely on the say so of the parties. In Scott v Scott  AC 417 itself, after all, there had been no objection at first instance to the proceedings being conducted in camera. That did not prevent the process thereafter being roundly and resoundingly condemned by the House of Lords.
What, then, of the several exceptions to the open justice principle that have been recognised by the courts? In so far as these exceptions have a statutory basis, of course, no problem arises. Tariq v Home Office  UKSC 35 is, obviously, a case in point. So, too, the control order cases. And R (Roberts) v Parole Board  UKHL 45,  2 AC 738 seems to me similarly explicable on essentially the same basis: given that the Parole Board was expressly empowered to withhold evidence from the prisoner and exclude him from hearings devoted to its consideration, the House of Lords understandably (although still only by the narrowest of majorities) sanctioned the use of special advocates specifically so as to provide additional protection for the prisoner. No problem arises either in certain well-established classes of case of which the most obvious (perhaps the only) examples are wardship/children proceedings (see Lord Dyson's judgment at para 63) and, yet more narrowly circumscribed, certain intellectual property proceedings to protect commercial interests (see Lord Dyson's judgment at para 64). In so far, however, as the appellants seek to rely on other recent decisions or dicta of the Divisional Court or the Court of Appeal most notably in R (Malik) v Manchester Crown Court  EWHC 1362 (Admin),  4 All ER 403, A v HM Treasury  EWCA Civ 1187,  3 WLR 25 and R (AHK) v Secretary of State for the Home Department (Practice Note)  EWCA Civ 287,  1 WLR 2049 (albeit in none of these cases does it appear that the propriety of a closed procedure was ever actually argued) I would simply repeat what I said in Davis (at para 66): "If .... the government now think it right to legislate in this field, so be it. Meantime, however, the creeping emasculation of the common law principle must be not only halted but reversed."
I come then, finally, to the way ahead. If, as I believe, it would be quite wrong for the common law to be developed to provide for a closed procedure in cases like this, and if, as I have suggested, not even the introduction of such a procedure by Parliament would really solve the problems arising, what should instead be done? For my part I have reached the reluctant conclusion that, by their very nature, claims of the sort advanced here, targeted as they are principally against the Intelligence Services, are quite simply untriable by any remotely conventional open court process. The problems they raise, of oral no less than documentary evidence, are just too deep-seated to be capable of solution within such a process. Far too little would be gained, and far too much lost, by the appellants' proposed development of the common law. In short, some altogether more radical solution is, I believe, required. Realistically there seem to be only two possible solutions. Either cases of this kind, necessarily involving highly sensitive security issues, should go for determination by some body akin to the Investigatory Powers Tribunal which does not pretend to be deciding such claims on a remotely conventional basis (see my judgment in Tariq v Home Office). Or they must simply be regarded as untriable and struck out on the basis that, as Laws LJ put it in Carnduff at para 36: "[They] cannot, in truth, be justly tried at all."
Obviously, I need hardly add, claims of the sort made here of the complicity of the Intelligence Services in torture ought not simply to be swept under the carpet. That, of course, explains why, these particular claims having been settled without admission of liability, they are to be the subject of an inquiry under the chairmanship of Sir Peter Gibson. It is to be hoped that, in the light of that inquiry's findings, together with the responses to the Government's proposed Green Paper, an acceptable way ahead may be found for the resolution of this type of case. Meanwhile, whatever else is to be done, it is certainly not the development of the common law in the way proposed. This is one of those cases where the court should indeed say "never". The appeal should be dismissed.
For the reasons given by Lord Dyson, with which I fully agree, I too would dismiss this appeal.
As I have observed in the associated case of Tariq v Home Office  UKSC 35, the right to know and effectively challenge the opposing case has long been recognised by the common law as a fundamental feature of the judicial process. I referred in my judgment in that case to various celebrated expressions of that principle and I need not repeat them here. The right to be informed of the case made against you is not merely a feature of the adversarial system of trial, it is an elementary and essential prerequisite of fairness. Without it, as Upjohn LJ put it in In re K (Infants)  Ch 381, a trial between opposing parties cannot lay claim to the marque of judicial proceedings.
And so the key nature of this right and its utter indispensability to the fairness of proceedings must occupy centre stage in the debate as to whether it may be compromised to serve the interests which the appellants claim require to be served and which are said to justify a departure from it.
The appellants have advanced two principal arguments in support of the claim that a closed procedure is required in this case. The first of these is pragmatic; the second purportedly a matter of principle. It is first asserted that the exercise involved in conducting a conventional public interest immunity exercise would be so daunting that some means should be found to avoid it. The second argument is that the adoption of a closed procedure will actually conduce to a fairer trial than would otherwise be possible.
The first of these arguments can easily be and, in the judgments of Lord Clarke and Lord Dyson, has been disposed of and I need say little more about it. As has been observed, unless there is to be complete abandonment of public interest immunity procedure as a means of catering for the tension between disclosure of relevant material and protection of the public interest, the exercise cannot be avoided. For the reasons given by Lord Clarke, to desert that procedure, so deeply embedded in our system of law, for reasons of expediency simply cannot be contemplated. The seemingly innocuous scheme proposed by the appellants would bring to an end any balancing of, on the one hand, the litigant's right to be apprised of evidence relevant to his case against, on the other, the claimed public interest. This would not be a development of the common law, as the appellants would have it. It would be, at a stroke, the deliberate forfeiture of a fundamental right which, as the Court of Appeal has said in para 70 of its judgment  3 WLR 1069, has been established for more than three centuries.
The appellants' second argument proceeds on the premise that placing before a judge all relevant material is, in every instance, preferable to having to withhold potentially pivotal evidence. This proposition is deceptively attractive - for what, the appellants imply, could be fairer than an independent arbiter having access to all the evidence germane to the dispute between the parties? The central fallacy of the argument, however, lies in the unspoken assumption that, because the judge sees everything, he is bound to be in a better position to reach a fair result. That assumption is misplaced. To be truly valuable, evidence must be capable of withstanding challenge. I go further. Evidence which has been insulated from challenge may positively mislead. It is precisely because of this that the right to know the case that one's opponent makes and to have the opportunity to challenge it occupies such a central place in the concept of a fair trial. However astute and assiduous the judge, the proposed procedure hands over to one party considerable control over the production of relevant material and the manner in which it is to be presented. The peril that such a procedure presents to the fair trial of contentious litigation is both obvious and undeniable.
In the solution offered by the appellants a state party can supply evidence to the judge with only (at best) the inquiring confrontation of the special advocate. Quite apart from the reasons so clearly stated by Lord Dyson about the necessary, inevitable but ultimately inherent frailties of the special advocate system, the challenge that the special advocate can present is, in the final analysis, of a theoretical, abstract nature only. It is, self evidently and admittedly, a distinctly second best attempt to secure a just outcome to proceedings. It should always be a measure of last resort; one to which recourse is had only when no possible alternative is available. It should never be regarded as an acceptable substitute for the compromise of a fundamental right such as is at stake in this case.
At a somewhat more prosaic level, the arguments against the case made by the appellants appear to me to be overwhelming. If one starts, as I believe we must, with the position that there is a constitutional, common law right to be informed of the case made against you in civil litigation, it becomes clear that what we are here being asked to do is to create an entirely new, hitherto unrecognised exception which has no statutory underpinning. The proposal for the radical change in the law which the appellants seek is openly policy driven. On that account alone, one should be very wary of it. Moreover, the claims that the present system creates impossible logistical burdens or that it produces unfair results on a massive scale are not backed up by any evidence. To accept that claim we would simply be acting on the ipse dixit of counsel.
At the moment with PII, the state faces what might be described as a healthy dilemma. It will want to produce as much material as it can in order to defend the claim and therefore will not be too quick to have resort to PII. Under the closed material procedure, all the material goes before the judge and a claim that all of it involves national security or some other vital public interest will be very tempting to make.
In this connection, one should not lose sight of the public interest in maintaining confidence in the administration of justice referred to so pertinently by Lord Neuberger MR in para 56 of his judgment in the Court of Appeal in the present case. For the reasons that he has given, I consider that this is an extremely important consideration and one which ought not to be overlooked.
On the question whether closed material might be provided to a judge where the claimant consents, I confess to grave misgivings as to how this might operate in practice. Consent to the submission of closed material under threat of a Carnduff application would not be the most propitious basis on which to found a jurisdiction which would not otherwise exist.
I also entertain considerable doubt as to whether it is possible as a matter of principle to invest the court with jurisdiction in this way and tend to agree with what Lord Brown has had to say on the subject. Since, however, this matter was not argued on the appeal, I consider that it is unnecessary to express any final view on it.
(with whom Lady Hale agrees)
A conventional PII exercise at common law involves comparing the public interest in the administration of justice secured by availability of the material with the public interest identified by the certificate and court favouring its suppression; and in striking a balance between these two competing interests: see Lord Reid's speech in Conway v Rimmer  AC 910, pp 940 and 951-952, quoted by Lord Clarke in paras 142-143.
The balance is struck somewhat differently in criminal and civil law contexts (see eg Balfour v Foreign and Commonwealth Office  1 WLR 681, at pp 688H-689A). In a criminal context, the general rule is that, if material is necessary to prove the defendant's innocence or avoid a miscarriage of justice, then "the balance comes down resoundingly in favour of disclosing it": R v Keane  1 WLR 746, 751-752, per Lord Taylor of Gosforth CJ. If the Crown still does not wish to disclose the material, it can and must forgo further prosecution.
In a civil law context, the liberty of the subject is not at stake. Where a prima facie case of public interest immunity is made out, a party who wishes to invite the court to inspect material before determining whether it should after all be deployed must show that it is likely to give substantial support to his or her case: Air Canada v Secretary of State for Trade  2 AC 394. When the court is balancing the competing interests, Cross & Tapper on Evidence, 12th ed (2010), p 484 suggests, with reference to case law from various common law jurisdictions, that
the court will take into account factors such as the seriousness of the claim for which disclosure is sought, whether or not the government is itself a party or alleged to have acted unconscionably, the relevance of the particular evidence to the dispute, taking into account other possible sources of evidence, and on the other side, the nature of the state's interest, and the length of time that has elapsed since the relevant discussion took place.
Thus, in both a criminal and a civil context, a judge dealing with an issue of PII has necessarily to form a view as to the relevance of the material for which PII is claimed. This is a fortiori the position if a judge, having concluded that, in the public interest, material cannot be disclosed, goes on to consider whether, as a result, the case has become untriable: see Carnduff v Rock  EWCA Civ 680,  1 WLR 1786, below.
Mr Jonathan Crow QC representing the Crown identifies various problems about a conventional PII exercise. Lord Dyson has set them out in para 23. They include the obvious difficulty in some cases of comparing two such different interests as the administration of justice between parties and the general public interest in withholding from outside eyes sensitive state material, the fact that a successful claim to PII can leave a party with relevant knowledge that neither the other party nor the decision-maker will have and may even make a case untriable and the dilemma (whether or not to pursue its case) in which an unsuccessful claim to PII can leave the state. They also include essentially pragmatic concerns such as the cost and time-burden imposed by a conventional PII exercise.
In their light, the Crown's primary case is that a court can decide to replace a conventional PII exercise by a closed material procedure, even without statutory authorisation such as arises for consideration in the linked appeal in Tariq v Home Office  UKSC 35. The court should be able, at least in exceptional cases, to order a closed procedure under which a special advocate would first ascertain the case being advanced by the litigant whose interests he or she was to serve; the material which the Crown seeks to withhold would then be made available to the judge and a special advocate, for the latter to make such forensic use of it as was possible. A balancing of the competing interests in disclosure would become unnecessary, since the judge and the special advocate would have access to everything. Whether or not it might, after a conventional PII exercise, have been excluded or made available would become irrelevant. As a secondary and alternative case, the Crown suggests the possibility that a closed material procedure might be ordered after or at the end of a conventional PII exercise.
Ms Dinah Rose QC's submission on behalf of the respondent is that there are inalienable features of a civil trial which no English court can or should abandon or qualify, at least without Parliamentary authority, even though this would involve no infringement of the Convention rights domesticated by the Human Rights Act 1998 (cf Tariq v Home Office  UKSC 35). Lord Dyson has discussed them under the heads of open and natural justice in paras 10 to 13 and Lord Clarke has identified the principal features in para 126. Ms Rose also noted a number of practical consequences which could flow from their abandonment, including difficulty on the part of the party without access to the material withheld to assess, take proper legal advice on the merits of or fund a case.
Taking the Crown's primary submission, I see no basis for the complete substitution of a conventional PII exercise by a closed material procedure, even if this were a possibility limited to "exceptional circumstances". The line between cases where a traditional PII exercise was undertaken and others would be unprincipled and uncertain. It is inherent in a conventional PII exercise that there may be difficulty in comparing the interests of the administration of justice and the general public interest, that a successful claim to PII can leave one party with relevant knowledge, or may even make a case untriable and may place the Crown in a dilemma whether or not to pursue its case after an unsuccessful claim to PII. It would be entirely unclear when these features might justify a judge in abandoning any attempt at a conventional PII exercise. Further, the special advocate would often be likely to become engaged in a modified form of PII exercise, involving arguments as to whether material should after all be disclosed to the litigant, bearing in mind the potential disadvantages for the litigant of a closed material procedure; nothing would then be gained, except to shift the whole burden of conducting that form of PII exercise onto the special advocate.
The Crown's alternative submission, that a closed material procedure might be recognised after or as supplementary to a conventional PII exercise, merits close scrutiny. As I understand it, no member of the Supreme Court doubts the approach in Carnduff v Rock  1 WLR 1786 as a possibility: see Lord Dyson at para 15, Lord Brown at para 86 and Lord Clarke at para 157. In other words, a successful claim for PII can make an issue untriable, so that the court will simply refuse to adjudicate upon the case. In some circumstances, therefore, the court is faced not with a binary choice, between trial with or without the material for which PII has been claimed, but with a trinary choice: the third possibility is no trial at all - whoever happens to be the claimant then has no access to the court at all.
Logically, this third possibility may be capable of feeding back into the decision whether a claim for PII should be allowed. If the effect of a successful claim to PII is that the case will not be tried at all, that introduces a different dimension, which may affect the striking of the balance of competing interests (para 103 above).
Lord Brown assumes (para 81) that a trial without the judge having any access to the PII material would be likely to be "to the Crown's considerable disadvantage". But it is not right to assume that the executive never errs or that material for which it claims PII is necessarily in its favour. In any event, issues regarding PII can arise between non-state parties, as for example in Asiatic Petroleum Co Ltd v Anglo-Persian Oil Co Ltd  1 KB 822. And it must certainly be regarded as being to a claimant's considerable disadvantage, if as a result of the withholding of PII material, the court concludes, as in Carnduff, that the case is not triable at all.
Like a number of other members of the court, I believe that the issue upon which the Court of Appeal and now this Court has embarked is fraught with danger and in principle undesirable; and that any answer that we give should at least be confined to situations such as the present where the defendant is the state and has the material withheld in its possession.
If the court never has jurisdiction (in the strict sense) to order a closed material procedure, that means that, even where a court concluded that a claimant must be denied access to material and the case must otherwise be struck out as untriable, it would be impossible for the court to order, with the consent of the claimant, a closed material procedure. There would be no way in which the material could be put before a judge, with the claimant's interests being represented to the best extent possible by a special advocate. I would be surprised if the court's inherent jurisdiction (in the strict sense) were inhibited to this extent.
I note that the judgment of the Court of Appeal, whose decision the respondent has invited the Supreme Court to uphold "as correct for the reasons given by it", expressly leaves open the question of "whether a closed material procedure can properly be adopted, .... in an ordinary civil case such as the present, where all parties agree, or in a civil claim involving a substantial public interest dimension", and adds that, although this is an issue to be considered as and when it arises, "principle and the authorities relied upon [in the courts] below .... suggest that a different conclusion may well be justified in such cases, albeit only in exceptional circumstances" (para 71). Ms Rose did not challenge this qualification in her submissions. On the contrary, she went even further than the Court of Appeal. Her case (para 133) states:
There is similarly no need for this court to reach a decision on whether a closed material procedure would be permissible if the parties consented to it, or in different sorts of proceedings, where the task of the judge was not simply to adjudicate on a private law claim for damages. However, insofar as necessary, the respondent would submit:
Cases of consent are also outside the "basic rule" which Lord Dyson identifies in para 22, that the court cannot exercise its power to "deny" parties their fundamental common law right to participate in proceedings in accordance with the common law principles of natural justice and open justice. An inability to allow a voluntarily accepted closed material procedure, as an alternative to striking a claim out as untriable, would be to deny something even more basic, that is any access to justice at all. Lord Dyson in the first sentence of para 22 uses the phrase "at any rate, not without the consent of the parties" and may therefore also accept this.
Further, once it is accepted, as Lord Dyson does (para 63), that "there are certain classes of case where a departure from the normal rule may be justified for special reasons in the interests of justice", for example wardship and other cases where the interests of children are paramount, that to my mind also makes it difficult to suggest that the court lacks jurisdiction in a strict sense to vary the basic principles of open and natural justice mentioned in para 107 above.
There is however a real distinction between having jurisdiction and exercising it. Principles as important as open and natural justice ought to be regarded as sacrosanct, as long as they themselves do not lead to a denial of justice. Absent statutory authorisation, any significant deviation from the ordinary process and consequences of a conventional PII exercise can and should only be under the compulsion of necessity, in order to avoid such a denial. Mr Crow acknowledged this as a possible approach. He cited Viscount Haldane LC's well-known statements in Scott v Scott  AC 417, 437-438, that exceptions to the principle that justice be administered in public may arise from considerations of necessity (not convenience) as "the outcome of a yet more fundamental principle that the chief object of Courts of justice must be to secure that justice is done". In this connection, Viscount Haldane referred to two cases: (a) "wards of court and of lunatics", where the court "is really sitting primarily to guard [their] interests" and "the broad principle which ordinarily governs it therefore yields to the paramount duty, which is the care of the ward or the lunatic" the court's role here, now regulated statutorily under the Mental Health Act 1983 and Mental Capacity Act 2005, has a parallel in that which it has in relation in children - and (b) "litigation as to a secret process, where the effect of publicity would be to destroy the subject-matter" and "justice could not be done at all if it had to be done in public", in which case "As the paramount object must always be to do justice, the general rule as to publicity, after all only the means to an end, must accordingly yield". Thus, while it is true, as Lord Brown observes (para 84), that in Scott v Scott the House deprecated the hearing in camera of a matrimonial issue in the interests of decency or delicacy, the House there expressly recognised the possibility that such a hearing might be ordered if the alternative was "a defeat of the ends of justice" (p 439). Subsequent authorities discussing the principle of public justice and the circumstances in which it may be qualified include Department of Economics, Policy and Development of the City of Moscow v Bankers Trust Co  EWCA Civ 314;  QB 207, where the Court of Appeal recognised that the principle might require modification to respect the confidentiality inherent under English law in their choice of arbitration. Party agreement can therefore have relevance to the modification in the interests of justice of what would otherwise be basic principles.
Viscount Haldane was speaking in Scott v Scott in the context of publicity, where the inroad into ordinary process is of a different order to that involved in any closed material procedure. But a claimant who is told that, because of the defendant's possession of material which cannot be disclosed, the claim is and must be struck out as untriable is just as effectively told, in Viscount Haldane's words, that "justice cannot be done". The inference may or may not be that the material favours the defendant state which has the documents and could rely on them for what they were worth in any closed material procedure. The claimant must by definition have a properly arguable case without the documents, since otherwise his or her case would be susceptible to being struck out. I myself see no reason why the court should not in such circumstances be able and prepared to offer the claimant the chance, if he or she wished, to pursue the claim by a closed material procedure, during which his or her interests would be represented by a special advocate. Lord Brown suggests (paras 86-87) that such cases must go off to some specially constituted tribunal, which does not pretend to be deciding such claims "on a remotely conventional basis". I find this difficult to square with the fact that courts and judges can and do operate closed material procedures where there is statutory authorisation, and can and do also depart from otherwise basic common law principles in the special classes of case which Lord Dyson mentions in para 63. Further, if one assumes that cases such as Carnduff are a "rarity" (Lord Dyson, para 50), that does not make it any more palatable, to my mind or to the individual claimant, to be denied any access to justice at all, in circumstances when he or she wishes to accept a closed material procedure. I do not see why a court should, instead of permitting a closed material procedure, insist on washing its hands of the case in such circumstances.
In a public law context, statutory schemes, e.g. governing suspected terrorism, have given rise to repeated issues about the legitimacy in terms of the Convention rights of closed material procedures: see eg Secretary of State for the Home Department v AF (No 3)  UKHL 28;  2 AC 269 and Tariq v Home Office  UKSC 35. It is also possible, at least in theory, to conceive of ordinary civil cases where the material withheld is in the hands of the claimant. A claim to PII is a duty, not an option, on the part of the state. A claimant (who might in some circumstances not be the state itself: see the Asiatic Petroleum case, cited in para 110 above) might find itself, at one and the same time, wishing to pursue a claim, but bound to raise (or, in the case of a non-state claimant, faced with) a claim for PII in respect of material in its possession. Again, it must be assumed that the claimant could, without the material, show an arguable case. If the court concluded that the material favoured the claimant, then the interests of justice would be unlikely to require its disclosure. The problem arises if the court concluded that it favoured the defendant but that the public interest in its confidentiality outweighed any particular interest of the defendant and that as a result the case could not fairly be tried.
Would it then be open to the court, as an alternative, to avoid denying any access to the court to the claimant, to order that the defendant should accept a closed material procedure? In the case of a state claimant, the public interest which required the claim to PII might perhaps also be said to require the state to accept that it could not pursue certain claims which it would otherwise wish to pursue in the public interest. In the case of a non-state claimant, seeking to pursue an ordinary civil claim in his, her or its private interest, that consideration could not be deployed. Nevertheless, it is far from clear that the court could go to the length of ordering a defendant to undergo a closed material procedure, in order to enable a non-state actor to pursue an otherwise untriable civil claim. But I shall express no opinion upon any such case since it lies far outside the realm of the present, and should be considered on its own merits were it ever to arise.
I should however address two further possibilities raised by Lord Clarke's judgment in relation to cases like the present where the relevant material is in the defendant's possession. One is the possibility, which I understand to be left open in paras 160-165, that, a judge might be able to order a closed material procedure at the claimant's instance, even though the judge concluded that the case was triable without the disclosure of the material withheld. The other is the suggestion in para 179 that the judge, if he or she declined to strike the claim out, might none the less accede to an application by the defendant for a closed material procedure "based on necessity, namely that such a procedure would be necessary in order to permit a fair trial". Both, it seems to me, involve an inconsistency. If a judge declines both to order disclosure and to strike the claim out, that means that he or she is satisfied that a fair trial, albeit on incomplete or imperfect material, is possible. There cannot in such a case be any "necessity", in the strict sense emphasised in Scott v Scott, for any departure from basic principles of open and natural justice. Both suggestions would in reality involve modifying the conventional PII exercise to introduce a fourth possibility, additional to the existing three (trial with or without disclosure, or no trial at all). As I have already indicated, I see no principled basis for such a modification.
For these reasons, and confining myself to cases such as the present where the material withheld is in the defendant's and not the claimant's possession, I consider that the Court can safely decide that there is no general basis for modifying the well-established rules governing the nature of a conventional PII exercise. There is no scope for introducing a closed material procedure as an alternative to such an exercise. I would not rule out a closed material procedure as outside the court's jurisdiction in a strict sense. But, statutory permission aside, the only exception that I would presently accept is where, after a conventional PII exercise, the judge concludes that there should be no disclosure, and that the case is as a result untriable. Then I think that the court could adopt some form of closed material procedure, if the claimant consented, in order to avoid denying the claimant any form of access to the court.
As regards consent, there is, as I see it, nothing between my conclusion and that of the Court of Appeal. If the preliminary issue is treated as raising a question of jurisdiction in the strict sense, where there is no consent, then my view differs on the issue of jurisdiction in the strict sense from that of the Court of Appeal's. As to the exercise of any such jurisdiction, I am in full agreement with the Court of Appeal that it cannot be appropriate to contemplate a closed material procedure in lieu of a conventional PII exercise. I can only envisage the jurisdiction ever being exercised after a conventional PII exercise in circumstances where a claimant would otherwise be denied any access to justice at all. As to what those circumstances might be, I would express an opinion only in relation to cases like the present, where closed material is in the defendant's possession. If the claimant, in order to avoid his or her claim being struck out, consents to engage in a closed material procedure, it would and should be permissible in my opinion for a court to allow a closed material procedure. The contrary, as I have pointed out (para 113), was not argued by the respondent.
The appellants in this appeal are the Security Service and various other organs of the state. It is an appeal from a declaration made by the Court of Appeal (Lord Neuberger of Abbotsbury MR, Maurice Kay and Sullivan LJJ)  EWCA Civ 482,  3 WLR 1069 on 4 May 2010 allowing an appeal from an order made by Silber J ("the judge") on 18 November 2009  EWHC 2959 (QB). That order was made after the hearing of a preliminary issue, which had been ordered on 24 September 2009 but was varied by agreement in the course of the hearing.
The preliminary issue as so varied was in these terms:
Could it be lawful and proper for a court to order that a 'closed material procedure' (as defined below) be adopted in a civil claim for damages?
Definition of 'closed material procedure'
A 'closed material procedure' means a procedure in which
For the purposes of this definition, disclosure is contrary to the public interest if it is made contrary to the interests of national security, the international relations of the United Kingdom, the detection and prevention of crime, or in any other circumstances where disclosure is likely to harm the public interest.
It is interesting to note that, as the judge said at para 2 of his judgment, in its original form the preliminary issue contained the words "if satisfied that such a procedure is necessary for the just disposal of the case" after the word "damages" in the third line, but those words were deleted by agreement "as their meaning was not clear and they did not appear to add anything to the preceding words".
The judge granted the application for the declaration. He declared that it could be lawful and proper for a court to order that a "closed material procedure" (as defined above) be adopted in a civil claim for damages. He granted permission to appeal to the Court of Appeal, which allowed the appeal. It made a declaration that the court does not have power to order that a "closed material procedure" (as so defined) be adopted in an ordinary civil claim for damages. The appellants were granted permission to appeal by this Court. The underlying claims for damages were then settled and the question arose whether the Court should allow the appeal to continue. In very many cases the court would refuse to proceed in such a case but it decided to proceed with this appeal because the point of principle raised by the question whether the court has no power to make a declaration in the terms sought is of general public importance. Moreover, given the fact that the Court of Appeal made the declaration in the bald terms which it did, it seemed appropriate for the court to consider it and not leave the matter to a future leapfrog appeal.
One of the problems raised by the appeal is that the declaration is stated in absolute terms, without reference to the facts of a particular case. I am firmly of the view that it is in general undesirable to determine bare questions of law in this way. I would expect the court ordinarily to require the relevant legal question to be decided in a particular factual context.
The parties' cases in summary
The respondent's case is that the judges have no power, by developing the common law on an ad hoc case by case basis, to abrogate a number of the most fundamental features of the trial of a civil claim for damages. Ms Dinah Rose QC identifies a number of such features as follows:
the requirement that each party must plead its case, identifying to all other parties the issues which are in dispute;
the requirement that all written or oral evidence on which a party wishes to rely in support of its case must be disclosed to the other parties to proceedings;
the requirement that each party must be permitted to test the disputed oral evidence of other parties by cross examination;
the rules which apply to the disclosure of relevant documents in the possession of a party to proceedings, including the principles that govern claims of public interest immunity ("PII"); and
the requirement that a court must give a fully reasoned judgment, to be made available to all parties, so that each party knows why it has won or lost and can decide whether or not to appeal.
Ms Rose submits that these are requirements which have been developed and maintained over centuries in order to secure basic constitutional rights of fairness, open justice and equality of arms, as well as to maintain confidence in the integrity of the judicial system, and in order to balance those rights and interests against competing considerations, including the interests of national security. Ms Rose cites many cases which support these fundamental principles, which, subject to what follows, are not disputed by Mr Jonathan Crow QC. Those principles have been endorsed in ringing tones by Lord Kerr in his judgment in Tariq v Home Office  UKSC 35, which was heard at the same time as this appeal by the same court and in which judgment has been handed down at the same time. They have also been clearly and accurately set out by Lord Dyson at paras 10 to 17 above.
The appellants' case was summarised by Mr Crow in a short document as follows. The court's objective is to achieve real justice between the parties. Everyone is entitled to a fair trial. That right is absolute, but the means of achieving it are infinitely variable. The practice and procedure of the court are the means of achieving that objective. They should be the servants, not the masters, of justice. As a general rule, real justice and a fair trial can only be achieved by a process which allows open hearings, open disclosure, each side confronting the other's witnesses and open judgments. There are, however, no absolute, inflexible rules as to how real justice or a fair trial can be achieved. The requirements of fairness must always be responsive to the particular circumstances of the case. In particular, the courts have adopted procedures for hearings in private, restricted disclosure, the exclusion of one party from part of the proceedings and closed judgments, with and without the assistance of special advocates. The adoption of such unusual procedures is not confined to any limited class or category of case, whether involving children, wards of court, confidential information, patent actions, insolvency or anything else. While the adoption of such procedures may be more likely in such categories of case, the common law proceeds by reference to principle and not by a 'tick-box' system. If a particular procedure is necessary by reference to the circumstances of the particular case, it should be adopted whether or not it falls into a category in which such a procedure has been adopted in the past.
In the agreed statement of facts and issues, the issue is said to be whether, in the absence of a specific statutory power, a closed material procedure can ever be adopted in a civil claim for damages.
Defects in the order sought for the use of a "closed material procedure"
I will return below to the question whether the court would have power to make an order of the kind sought. However, I should state at the outset that I cannot conceive of circumstances in which the court could in fact properly make an order for use of a "closed material procedure" as defined. It contains no procedure setting out how the claim that disclosure of a particular document or class of document is contrary to the public interest should be made or determined. Is it a matter for the judge and, if so, on what material, with whose assistance and in accordance with what principles? These are critical questions, especially the last. As formulated it seems implicit in the procedure that there is to be no question of any balance between the public interests referred to and interests of the parties. Yet the various public interests sought to be protected are very different. They extend from national security to "any other circumstances where disclosure is likely to harm the public interest", which is very broad indeed. As I see it, very different considerations are likely to apply to each class of case. In these circumstances, it seems to me to be clear that it would never be appropriate to make an order in the bald terms apparently sought. It does not, however, follow that it would never be appropriate to make an order suitably tailored to the circumstances of the particular case which has the effect of limiting some of the common law rights identified by Ms Rose. The appellants' case was that there were circumstances in which it would or might have been appropriate to do so on the facts of this case.
The factual and procedural background
Although the issues between the parties have now been settled, it is appropriate to summarise the facts very briefly because it is rarely sensible to consider any legal principle in the abstract and because the facts here demonstrate some of the problems that can arise. I can take them, albeit in less detail, from paras 5 to 10 of the judgment of the Court of Appeal, which was delivered by Lord Neuberger MR.
The claimants were individuals some of whom were detained at Guantanamo Bay. They said that as a result of their detention and mistreatment while detained they had valid claims under at least some of the following heads, namely false imprisonment, trespass to the person, conspiracy to injure, torture, breach of contract, negligence, misfeasance in public office and breach of the Human Rights Act 1998. The claims were based on the contention that, to put it in broad terms, each of the appellants caused or contributed towards the alleged detention, rendition and ill-treatment of each of the claimants. The appellants then filed an open defence in which, while admitting that each of the claimants was detained and transferred, they put in issue any mistreatment which the claimants alleged and, in any event, denied any liability in respect of any of the claimants' detention or alleged mistreatment. It was said that there was material not pleaded in the open defence which the appellants wished to contend that the court should consider but which could not be included without causing real harm to the public interest. That material was said to be contained in a closed defence. The open defence made it clear that the appellants wished the case to proceed throughout on the basis that it included what may be characterised as a closed element.
Thus, at least on the face of it, during the period prior to trial, there would be parallel open and closed pleadings, parallel open and closed disclosure and inspection, parallel open and closed witness statements and parallel open and closed directions hearings. Similarly, at the trial, the hearing would be in part open and in part closed, no doubt with some documents and witnesses being seen and heard in the open hearing and others in the closed hearing (with some witnesses conceivably giving evidence at both hearings). After trial, there would be a closed judgment and an open judgment, which would be in substantially the same terms save that those passages in the closed judgment which referred to or relied on closed material would be excluded from the open judgment. In relation to the open elements of the proceedings, the claimants would be represented by their solicitors and counsel in the normal way, whereas, in relation to the closed elements, their interests would in effect be protected by special advocates.
The claimants objected to the course proposed by the appellants, contending for the normal approach in cases where the Crown or government emanations are parties and consider that they have relevant documents in respect of which PII might be claimed, and where the defendants could call relevant oral evidence which might not be able to be given on public interest grounds. The appellants accepted that the PII procedure was well established, but contended that a closed material procedure was permissible in any civil case, at least before a judge sitting without a jury, and that it might well be appropriate in this case, where there was a very substantial amount of potentially relevant material which would be subject to PII.
The evidence filed on behalf of the appellants suggested that there might be as many as 250,000 potentially relevant documents, and that PII might have to be considered in respect of as many as 140,000 of them. It was also said by the appellants that the PII exercise might take three years before the relevant ministers could conscientiously decide in respect of which documents PII could properly be claimed. The appellants argued that the effort, cost and delay involved in such an exercise might well justify a different approach, such as that presaged by the open defence. It was said in a witness statement served on the appellants' behalf in support of the application to use a closed material procedure in this case that the principal motivation for it was the enormous scale of the disclosure exercise and the impracticability of carrying it out.
The appellants had initially sought directions from the High Court for the determination of preliminary issues as to whether a court has the power to order that a "closed material procedure" be adopted in a civil claim for damages if satisfied that such a procedure is necessary for the just disposal of the case; and, if so, whether and what arrangements for a closed material procedure should be adopted in relation to the claims brought by the respondent (and others). The High Court accepted the respondent's (and his then fellow claimants') submission that the first issue should be considered alone because the question whether the power to hold a closed material procedure should be exercised in this particular case could itself only be answered by looking at closed material. The preliminary issue described above was subsequently ordered and, as varied, determined by the judge.
The common law
It is important to note that the issue between the parties is concerned only with the position at common law. There are now a number of circumstances in which powers have been conferred on the courts to make similar orders by statute.
Tariq v Home Office is an example of such a case. It concerns the permissibility and in particular compatibility with European Union law and with rights under the European Convention on Human Rights ("the Convention") of a similar closed material procedure authorised by certain statutory provisions. The issues there focus on the lawfulness and effect of those provisions and their compatibility, for example, with article 6 of the Convention, whereas in the instant appeal the court is concerned with the position at common law.
Mr Crow submits that the common law is in a permanent state of development and should not be allowed to stultify. Ms Rose submits, by contrast, that the fundamental common law rights summarised in para 126 above lie at the heart of the right to fair treatment before the law and should not be limited or abrogated in any way. She recognises that the common law has developed the principles of PII in order to cater for problems of the kind which concern the state, especially national security. She submits that the state is sufficiently protected by those principles and that there is no warrant for permitting any kind of closed material procedure. In these circumstances it is appropriate to consider how PII works and, to do so in the context of a case like this.
PII the principles
The principles of PII are a construct of the common law which were developed because it was appreciated that conflict may arise between the public interest and established rules of discovery and disclosure. They were developed having regard to the public interest in the administration of justice and other public interests which precluded or were said to preclude disclosure of materials which would otherwise be disclosable. As to disclosure, the general principle is that if the court "is satisfied that it is necessary to order certain documents to be disclosed and inspected in order fairly to dispose of the proceedings, then .... the law requires that such an order should be made": see eg Science Research Council v Nassι  AC 1028, 1071E-F.
The principles of PII have developed significantly over the years in a way which to my mind shows the flexibility of the common law. In Duncan v Cammell Laird & Co Ltd  AC 624 Viscount Simon LC (with whom the other six members of the House agreed) made it clear at p 629 that the question whether documents were subject to Crown privilege, which was the forerunner of PII, could arise, as in that case, in an action between private parties or, as here, in an action in which the Crown is a party. Crown privilege could be relied upon in respect of an individual document or a class of documents. The House of Lords held that it was for the minister personally to consider the question whether it or they should not be disclosed on grounds of public interest. Viscount Simon made it clear at p 642 that disclosure must not be withheld in order to avoid criticism or embarrassment or to avoid paying compensation. He said at pp 642-643:
In a word, it is not enough that the minister of the department does not want to have the documents produced. The minister, in deciding whether it is his duty to object, should bear these considerations in mind, for he ought not to take the responsibility of withholding production except in cases where the public interest would otherwise be damnified, for example, where disclosure would be injurious to national defence, or to good diplomatic relations, or where the practice of keeping a class of documents secret is necessary for the proper functioning of the public service. When these conditions are satisfied and the minister feels it is his duty to deny access to material which would otherwise be available, there is no question but that the public interest must be preferred to any private consideration.
Viscount Simon had said a little earlier that an objection properly taken by the minister was conclusive. Although he stressed that the ruling was to be made by the judge, not the executive, "the proper ruling" was to accept the minister's objection. The House thus held that, although the decision excluding such documents was for the court, it had no discretion in the matter. In short, Viscount Simon made it clear that a court could never question a claim to Crown privilege if the claim was made in the proper form.
However, in Conway v Rimmer  AC 910 a five member appellate committee of the House of Lords disapproved the approach taken in Duncan v Cammell Laird. It held, not only that it was for the court to decide whether Crown privilege should apply, but also that it was for the court, not the minister, to balance the competing public interests. Lord Reid said this at p 940:
It is universally recognised that here there are two kinds of public interest which may clash. There is the public interest that harm shall not be done to the nation or the public service by disclosure of certain documents, and there is the public interest that the administration of justice shall not be frustrated by the withholding of documents which must be produced if justice is to be done. There are many cases where the nature of the injury which would or might be done to the nation or the public service is of so grave a character that no other interest, public or private, can be allowed to prevail over it. With regard to such cases it would be proper to say, as Lord Simon did, that to order production of the document in question would put the interest of the state in jeopardy. But there are many other cases where the possible injury to the public service is much less and there one would think that it would be proper to balance the public interests involved. I do not believe that Lord Simon really meant that the smallest probability of injury to the public service must always outweigh the gravest frustration of the administration of justice.
A little later, at pp 951-952, after quoting the reference in Duncan v Cammell Laird to "the proper ruling" referred to above, namely to accept the minister's view in every case, Lord Reid said:
In considering what it is 'proper' for a court to do we must have regard to the need shown by 25 years' experience since Duncan's case, that the courts should balance the public interest in the proper administration of justice against the public interest in withholding any evidence which a minister considers ought to be withheld.
I would therefore propose that the House ought now to decide that courts have and are entitled to exercise a power and duty to hold a balance between the public interest, as expressed by a minister, to withhold certain documents or other evidence, and the public interest in ensuring the proper administration of justice.
Lord Reid then gave a number of practical examples of how the balance might be struck in different classes of case. He also said at p 953 that he could see nothing wrong with the judge seeing the documents without their being shown to the parties. In the event the House of Lords (or at any rate Lord Reid) inspected the documents and, the House having found (at pp 996-997) that there was nothing in them which was in any way prejudicial to the proper administration of the relevant police force or to the general public interest, directed that they be disclosed for use in the litigation.
It is common ground that the current state of the law on what is now called PII is set out in R v Chief Constable of West Midlands Police, Ex p Wiley  1 AC 274. In that case, the House of Lords held that there was no justification for a claim for immunity for the entire class of documents generated by an investigation into a complaint against the police.
I would accept the submission made by Ms Rose that the following principles correctly state the approach to PII as it has stood until now:
A claim for PII must ordinarily be supported by a certificate signed by the appropriate minister relating to the individual documents in question: Duncan v Cammell Laird per Viscount Simon at p 638.
Disclosure of documents which ought otherwise to be disclosed under CPR Part 31 may only be refused if the court concludes that the public interest which demands that the evidence be withheld outweighs the public interest in the administration of justice.
In making that decision, the court may inspect the documents: Science Research Council v Nassι at pp 1089-1090. This must necessarily be done in an ex parte process from which the party seeking disclosure may properly be excluded. Otherwise the very purpose of the application for PII would be defeated: see the Court of Appeal judgment at para 40.
In making its decision, the court should consider what safeguards may be imposed to permit the disclosure of the material. These might include, for example, holding all or part of the hearing in camera; requiring express undertakings of confidentiality from those to whom documents are disclosed; restricting the number of copies of a document that could be taken, or the circumstances in which documents could be inspected (eg requiring the claimant and his legal team to attend at a particular location to read sensitive material); or requiring the unique numbering of any copy of a sensitive document.
Even where a complete document cannot be disclosed it may be possible to produce relevant extracts, or to summarise the relevant effect of the material: Wiley at pp 306H-307B.
If the public interest in withholding the evidence does not outweigh the public interest in the administration of justice, the document must be disclosed unless the party who has possession of the document concedes the issue to which it relates: see Secretary of State for the Home Department v MB  UKHL 46,  1 AC 440 per Lord Hoffmann at para 51.
The question then arises, what, if any, use can be made of material which the court has held cannot be disclosed because of PII. Ms Rose submits that, if the court concludes that the public interest in withholding the evidence outweighs the public interest in the administration of justice, then the evidence cannot in any circumstances be admitted, or relied on by either party. She relies upon R v Lewes Justices, Ex p Secretary of State for the Home Department  AC 388, per Lord Simon of Glaisdale at p 407. He said there that Crown privilege is a misleading expression because it is not in truth a privilege. It refers to the rule that certain evidence is inadmissible on the ground that to adduce it would be contrary to the public interest. He recognised that PII has to be weighed against the public interest in the administration of justice in accordance with Conway v Rimmer. He then said that, once the former privilege is held to outweigh the latter, the evidence cannot in any circumstances be admitted. He added at p 407B-C:
It is not a privilege which may be waived by the Crown (see Marks v Beyfus (1890) 25 QBD 494 at p 500) or by anyone else. The Crown has prerogatives, not privilege. The right to procure that admissible evidence be withheld from, or inadmissible evidence adduced to, the courts is not one of the prerogatives of the Crown.
I will return to this point below.
PII the facts
In the course of the argument, Mr Crow referred the Court to the description of the PII certification process contained in the Attorney General's statement made to the House of Commons on 18 December 1996 following the Matrix Churchill affair (Hansard (HC Debates) cols 949-58) and the decision in Wiley. In it the Attorney General noted that PII is subject to the ruling of the court and that, in a criminal case, when government documents are in issue, the judge examines the documents and makes the actual decision on disclosure in the light of the facts of the case. The statement also said that in what it described as the new approach, ministers would focus directly on the damage that disclosure would cause, the former division into class and contents claims would no longer be applied and they would only claim PII when it was believed that disclosure of a document would cause real damage or harm to the public interest. The statement further said that the new approach was subject to the supervision of the courts and that it accorded with the view expressed by the then Lord Chief Justice that PII "should only be claimed for the bare minimum of documents for which the claim of serious harm can be seen to be clearly justified." Finally, the statement referred to the kinds of serious harm which might be involved.
Mr Crow described the certification process in a little more detail in this way. Lawyers consider material to see if it passes the threshold test for disclosure under CPR Part 31. In so far as it is prima facie disclosable, officials review material for potential to cause harm to the public interest. If harm to the public interest is identified, the department carries out a balance between harm caused by the disclosure on the one hand and injustice in the litigation on the other. It also considers whether it is possible to redact or gist the information or to make admissions of fact. Officials consider whether and to what extent the balance falls against disclosure in order to give advice to the minister as to whether to certify. If the minister, having considered the advice, decides that a certificate should be given, a PII certificate is prepared which includes a disclosable certificate or schedule describing the types of harm that might be caused to the public interest and a sensitive schedule as to why it is believed that disclosure of documents would cause real damage or harm to the public interest.
After the minister has signed a PII certificate, the balance between the relevant public interests must be made by the judge. In a simple case he will hear argument on both sides and reach a conclusion, often having looked at the documents. There will be no need for special advocates. The position may be very different in a case of complexity, especially a case of great complexity such as this was or would have been but for the settlement. The judge may need assistance in order carry out the balance. Such assistance will not of course be available from counsel for the non-state parties because they will not have seen the documents.
PII special advocates
It is common ground between the parties that it is in principle permissible for the court to approve the appointment of one or more special advocates to peruse the documents at an appropriate stage in order to assist the judge to decide how the balance should be struck between the public interest in disclosure and the public interest in non-disclosure. It is thus common ground that the court has the power to give such approval and, no doubt, to make other ancillary orders as to what the special advocate should do. In my opinion the court does have such a power at common law. Just as the House of Lords recognised in R v H  UKHL 3,  2 AC 134, that the court had power at common law to give such approval in the context of criminal proceedings, so I would accept that there is such a power in this context. This is to my mind a classic example of the measured development of the common law in confronting and solving new problems. As Lord Bingham put it extra-judicially in The Business of Judging, 2000, p 29, "the last quarter century has seen fundamental, Judge-made changes in the law relating to [PII]".
The most obvious role for a special advocate would be to look at the documents which are the subject of a certificate and consider how the balance should be struck. In a complex case like this the exercise would be close to impossible for the judge to do on his own without assistance. One or more special advocates would to my mind be essential. It may be that in such a case it would be sensible for the special advocates to be allowed to see the documents at an early stage in order to avoid or minimise delay, but that would depend upon the facts of the case. However that may be, as I see it the use of special advocates would be (or would have been) necessary here in order to enable the judge to carry out the balancing exercise.
Should the PII exercise be abandoned?
The appellants proposed that there should be a closed material procedure instead of the well recognised and accepted PII procedure, partly if not largely because of the enormous scale of the PII disclosure exercise and the impracticability of carrying it out. I am quite unable to accept that, at any rate in the absence of Parliamentary intervention, that was a sufficient reason for abandoning the established PII procedure and replacing it with some form of closed procedure at common law. As I see it, whatever procedure was adopted, it would have been necessary for the appellants to identify what documents were relevant and in principle disclosable under CPR Part 31. In addition it would have been necessary for the minister to decide which of those documents should not be disclosed in the public interest. That in turn would have required officials to identify which documents potentially came into that category in order to enable the minister to carry out the appropriate balance. A detailed review of the documents would have had to be carried out whether the procedure adopted was the PII procedure described above or the proposed closed procedure. In both cases it would have been necessary for the relevant documents to be identified and the balance struck.
The critical difference between the two approaches is not that part of the exercise but what happens thereafter. Since the first part of the exercise has to be carried out in any event, I can see no reason for abandoning the PII exercise.
Procedure after the balance is struck
In the ordinary case, where PII proceeds as outlined above and there is no closed procedure of the type sought, if the judge rules that some of the documents should not be disclosed because they attract PII, the position to date has been that identified by Lord Simon of Glaisdale in the Lewes Justices case, namely that they are not disclosed or produced to the non-state parties and are not admissible in evidence. In a criminal case, as explained in R v H  2 AC 134 and other cases, where the judge rules that they are necessary for the defence case, the prosecution must either disclose them or abandon the prosecution.
In a civil case of this kind Ms Rose submits that the effect of the PII rules is that
the court strikes the balance between the competing public interests, while duly acknowledging the opinion and expertise of the Secretary of State and those advising him;
the public interest in the administration of justice is given full weight; and
whatever the outcome of the exercise, the principles of natural justice and equality of arms are always preserved; both parties are equally treated in relation to the use which may be made of the sensitive material.
Mr Crow submits that that is far from a satisfactory result because in a case of this kind it may well be the position that the claimants are able to deploy their evidence in support of their claims, whereas, by reason of the inadmissibility of evidence which, but for the necessity to certify PII, would be admissible and upon which the appellants would wish to rely at the trial, the appellants are unable to defend the action, either at all or sufficiently. He submits that both parties have the right to a fair trial and that in such circumstances the trial would not be fair to both parties. He submits that it is in these circumstances that justice requires some form of closed procedure and that the court must have power at common law to permit it. There is to my mind considerable force in that submission. In Tariq v Home Office  UKSC 35, Lord Brown describes at para 84 the suggestion that, where the defendants cannot defend themselves at all, it matters not because they can simply pay up as preposterous.
Another possibility canvassed in the course of the argument is for the court to decide that it would not be appropriate for a closed procedure to be introduced for the reasons given by Ms Rose but that the correct course would be to stay or strike out the action because, by reason of circumstances beyond the control of both parties, it was not possible for there to be a fair trial and that the just course was not to have a trial. It was in this connection that the court was referred to Carnduff v Rock  EWCA Civ 680,  1 WLR 1786. In that case a registered police informer brought an action against a police inspector and his chief constable to recover payment for information and assistance provided to the police. The Court of Appeal struck out the action on the basis that a fair trial of the issues arising from the pleadings would require the police to disclose sensitive information which it would be contrary to the public interest to force the police to disclose. Laws LJ said at para 36 that "a case which can only be justly tried if one side holds up its hands cannot, in truth, be justly tried at all". It is perhaps noteworthy that the claimant complained to the European Court of Human Rights in Strasbourg on the basis that his right of access to the court under Article 6 of the Convention had been infringed. His complaint was rejected as unfounded: Carnduff v United Kingdom (Application No 18905/02) (unreported, BAILII:  ECHR 731) 10 February 2004.
I note in this regard that in Tariq v Home Office  UKSC 35 Lord Mance says (at para 40) that the striking out of the action for these reasons is not an option that the law should readily contemplate. By contrast, Lord Kerr says at para 110 that it is a more palatable course than to adopt a closed procedure. There is plainly a tension between these two approaches, which in my opinion should be resolved on the facts of a particular case. It is not quite clear to me what approach Lord Dyson would take to the question whether the action should be struck out on what may be called Carnduff principles in a case of this kind.
A possible solution
It appears to me that the way forward is or should be along these lines. After the PII process described above it should be for the parties to consider their respective positions and then to make representations to the judge as to the appropriate way forward. Depending upon the submissions advanced, the judge may wish to consider the three possibilities to which I have referred. They are
that the matter should proceed in the traditional way with the PII material simply being treated as both undisclosable and inadmissible and the trial proceeding on the basis of the disclosed and admissible evidence;
that the action should be stayed or struck out on the basis that through neither party's fault a fair trial is not possible; and
that there should be some form of closed procedure, involving special advocates, along the lines suggested by the appellants, but subject to the exigencies of the particular case.
The stance taken by the protagonists at that stage would no doubt depend very much upon the circumstances at the time. Thus the claimants might adopt the stance now taken by Ms Rose. On the other hand, they might conclude that there were advantages in some form of closed procedure, especially if their case was thought to depend to any significant extent upon documents in the possession of the defendants. The defendants might also adopt a different position depending upon the circumstances. However, as things stand at present, they might be expected to contend in every such case that the action should be struck out on the ground that, without the documents and/or evidence to which PII attached or would attach, a trial which was fair to both parties including the defendants would not be possible.
In that event, the claimants might well perceive it to be in their interest to consent to agree to a closed procedure of some kind as an alternative to their claims being struck out. I understood it to be common ground between the parties that there would be no objection in principle to such consent being given. It was not contended on behalf of the appellants that, if consent was given, such a procedure was contrary to principle. Indeed, it follows from Mr Crow's submission that it would be appropriate even without consent that, a fortiori, it would be appropriate with consent.
If a closed procedure were in principle a possibility, the precise nature of it would no doubt depend upon the circumstances of the particular case. It is only at the conclusion of the PII process that the question whether to direct some form of closed material procedure would arise. That will not of course happen in this case because it has settled. By the time the question comes to be decided in a specific case, Parliament may have intervened. It was suggested in the course of the argument that the common law should not consider inventing some form of closed material procedure but should leave it to Parliament. I agree that it would be better for the problems which arise in this class of case to be dealt with by Parliament. I understand that a green paper is in prospect and that thought is being given to possible Parliamentary intervention. The approach of Parliament would plainly be potentially critical to the appropriate way forward at the conclusion of the PII procedure.
However, it may be that the question what directions should be given may arise in a case in circumstances (as occurred for example in A v HM Treasury (Nos 1 and 2), as explained in the Court of Appeal: see  EWCA Civ 1187,  2 AC 534 at paras 59 and 64-65), where for one reason or another there has been delay and where Parliament has not decided whether or not to intervene or, perhaps, where it has decided not to intervene. In such circumstances the court would have to decide what, if any, novel procedure to adopt. As appears below the court has adopted novel procedures in not dissimilar situations, no doubt by way of development of the common law and pursuant to its obligation under the CPR Part 1 to act justly. In considering the relationship between the role of Parliament and the role of the common law, Lord Bingham said in The Business of Judging at pp 386-387 that regard should be had to any relevantly analogous statute and that he saw no reason "why statute and the common law should not feed and refresh each other".
I see no reason in principle why the common law should not be able to develop along these lines. There are many statements in the books of the flexibility of the common law. As Lord Slynn of Hadley memorably put it in Kuddus v Chief Constable of Leicestershire Constabulary  UKHL 29,  2 AC 122, at para 33, "the genius of the common law is its capacity to develop". See also, for example, Sir Frederick Pollock in The Genius of the Common Law (1912) at p 45 and Lord Denning MR in Comet Products UK Ltd v Hawkex Plastics Ltd  2 QB 67. In so far as such a development would be a development of the common law of PII, it would be no more than a further development of a process which, as Lord Bingham put it in 2000, has been taking place over the last quarter of a century.
In these circumstances I would not grant a declaration such as that granted by the Court of Appeal. The common law should in my opinion very rarely, if ever, say never. Moreover, the argument in this appeal has persuaded me that it is not appropriate to make a declaration such as that granted by the Court of Appeal in a vacuum. The court should consider what orders to make in the context of a specific case.
I turn to consider the terms of the declaration which was granted. In it the Court of Appeal distinguishes between "ordinary" or "normal" civil claims and those which involve "a substantial public interest dimension". I note in this context that in the declaration granted by the judge he referred to "a civil claim for damages", whereas the Court of Appeal added the word "ordinary". In its judgment at paras 59, 61, 63-67 and 71, it distinguished the various cases relied upon by the appellants on the ground that they were not ordinary civil claims for damages. Some were criminal cases: see R v Shayler  UKHL 11,  1 AC 247 and R v H. Others were civil cases of various kinds. They include Secretary for State for the Home Department v Rehman  UKHL 47,  1 AC 153; R (Roberts) v Parole Board  UKHL 45,  2 AC 738, R (Malik) v Manchester Crown Court  EWHC 1362 (Admin),  4 All ER 403; A v HM Treasury and R (AHK) Secretary of State for the Home Department (Practice Note)  EWCA Civ 287,  1 WLR 2049. It is true that none of those cases involved an ordinary civil claim for damages, by which it appears from paras 59 and 71 that the Court of Appeal meant a case in which the court was simply an arbiter.
The reasons for the Court's view can be seen from its concluding remarks at paras 68 to 71. It expressed concern at para 69 that a procedure which is said to apply only in exceptional circumstances often becomes common practice. It then expressed these conclusions at paras 70 and 71:
The importance of civil trials being fair, the procedures of the court being simple, and the rules of court being clear are all of cardinal importance. It would, in our view, be wrong for judges to introduce into ordinary civil trials a procedure which
We leave open the question of whether a closed material procedure can properly be adopted, in the absence of statutory sanction in an ordinary civil claim such as the present, where all the parties agree, or in a civil claim involving a substantial public interest dimension (ie where the judge is not simply sitting as an arbiter as between the parties). Both principle and the authorities relied on below seem to us to suggest that a different conclusion may well be justified in such cases, albeit only in exceptional circumstances, but that is an issue which should be considered as and when it arises.
Save that I am not persuaded that an appropriate procedure could not be devised which is consistent with the CPR, I entirely accept the fundamental principles summarised in para 70. However, I do not think that there is any principled distinction between an ordinary civil claim and any other civil claim for these purposes. In any event, if there is potentially such a distinction, this is surely a case in which there is a public interest in the issue involved, at any rate on the approach I have identified. The issue which has given rise to the question in this appeal is what is the correct approach to the fact that the appellants have what may be a valid claim to PII. The question is what to do when such a claim is made and, in particular, what to do when the minister issues a PII certificate. The public interest self-evidently plays a central part in the resolution of those questions. So, if there is a distinction between ordinary civil actions and those involving a substantial public interest dimension, this case falls on the latter side of the line.
On the question whether there is a principled difference between ordinary civil actions and others I would accept Mr Crow's submissions on behalf of the appellants, which were shortly to this effect. First, no such distinction has been drawn in the past. Secondly, the question in this appeal is whether a closed material procedure can ever be adopted in the case of a civil claim for damages. Distinctions between different claims for damages are irrelevant for the purposes of answering that bald question. The distinction may be relevant to the answer to the question whether an order should be made in a particular case. Thirdly none of the cases in which an exception has been made to the principle that all the material must be shown to both parties in every case depends upon this distinction. They all depend upon what the interests of justice demand. None of them depends upon a classification of the kind adopted by the Court of Appeal: see eg the wardship and children cases, including In re T (Wardship: Impact of Police Intelligence  EWHC 2440 (Fam),  1 FLR 1048, Chief Constable v YK  EWHC 2438 Fam and In re D (Minors) (Adoption Reports: Confidentiality)  AC 593, 615, per Lord Mustill. There have been other examples in the field of commercial confidentiality; see for example British Sky Broadcasting Group plc v Competition Commission  EWCA Civ 2,  2 All ER 907.
Fourthly, there is no principled basis for accepting that the court can properly adopt a closed material procedure (subject to appropriate safeguards) when government action is subject to judicial review but not when the same action is subject to a claim in tort for damages. The differences between the two procedures are irrelevant to this question. Both may involve identical questions of law and fact. Both may involve the examination of the same evidence and examination and cross-examination of witnesses: see eg R (Al-Sweady) v Secretary of State for Defence  EWHC 2387 (Admin),  HRLR 12. Fifthly, the boundaries of the two classes of case are often blurred. I note, for example that in R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs (No 1)  EWHC 2048 (Admin),  1 WLR 2579, the claims were in part for judicial review and in part for Norwich Pharmacal relief: see Norwich Pharmacal Co v Customs and Excise Comrs  AC 133. The natural process for the latter would be by ordinary action. It is of interest to note in passing that there were there both open and closed hearings and cross-examination by a special advocate, no doubt by consent between the parties.
In these circumstances, I would not accept the jurisdictional distinction that the Court of Appeal draws between ordinary civil claims and other civil claims. It appears from para 71 of the judgment quoted above that, subject to that distinction, the Court of Appeal accepted that both principle and the authorities relied on before the judge suggested that a different conclusion may well be justified in such cases, albeit only in exceptional circumstances, but said that was an issue which should be considered as and when it arose.
I appreciate that, as both Lord Dyson and the Court of Appeal have pointed out, the issues of jurisdiction were not argued in those cases. That was certainly true in the two cases referred to in which I played a part, namely A v HM Treasury and the AHK case. It is also true that, as the Court of Appeal pointed out, the decision in Ahmed was reversed by this court. The principles relevant to this case were not however commented upon adversely. They are summarised at para 58 of A and at para 17 et seq of AHK.
Both cases involved the appointment of special advocates without express statutory authority. They are to much the same effect. So I refer only to para 58 of A, where I said this:
There is so far no statutory power to appoint a special advocate in proceedings arising out of a [Terrorism Order]. However, as I see it, there is no reason in principle why a special advocate should not be appointed in a particular case. The authorities show that in an appropriate case the court would have power to authorise or request the use of a special advocate: see in particular the decision of the House of Lords in R (Roberts) v Parole Board  2 AC 738, where it was held that the court had power to do so even where it was not sanctioned by Parliament. Whether it should do so or not would depend on the particular circumstances of the case. It has very recently been held by the Divisional Court in .... Malik .... that the court has power to ask the Attorney General to appoint a special advocate, but that it should only do so in an exceptional case and as a last resort: per Dyson LJ, giving the judgment of the court, at paras 93-102, especially, at para 99. In these circumstances the court would have power to procure the appointment of a special advocate through the Attorney General.
I am not persuaded by Ms Rose's submissions that the principles do not remain sound and applicable in an appropriate case. However, all depends upon the circumstances of the particular case and I agree with the view expressed in the Court of Appeal in para 71 (quoted above) that the issue of their correctness and application to the circumstances should be decided when they arise in a particular case.
If they ever arise in a context like this, it will be in very different circumstances from any of the other cases. If they arise in a case of this kind it will be necessary to examine in detail the relationship between (a) PII and its application in this type of case, (b) the principle in Carnduff v Rock, including its correctness and its application and relationship to PII and (c) the prospect of fashioning a procedure which will on the one hand retain the principles of PII and on the other hand enable the action to proceed in a way which will provide a fair trial to both parties, while respecting their fundamental rights, including the fundamental rights of the claimants relied upon by Ms Rose. In reaching these conclusions I do not wish to understate the fundamental nature of those rights. It follows that any new procedure which affected them should only be adopted by the common law in most exceptional circumstances. Whether those circumstances exist will depend upon the facts of the particular case and should only be determined in that context.
Since writing the above, I have had the opportunity of reading Lord Dyson's judgment in draft. I entirely understand his concern (and that of the Court of Appeal) that a principle that a closed procedure could be used only in exceptional circumstances may become the thin end of the wedge. However, in my opinion the judges can be relied upon to ensure that that will not happen. The test would be one of necessity: see Scott v Scott  AC 417, per Viscount Haldane LC at pp 436-439.
As Lord Dyson himself recognises at paras 63 and 64, various exceptions to the fundamental principles he describes have been recognised by the common law. These show that, although fundamental, the principles are not absolute and must yield where it is necessary in the interests of justice that they do so. As Lord Dyson puts it at para 64 in the context of confidentiality, such claims by their very nature raise special problems which require exceptional solutions (his emphasis). If the judge concludes after carrying out the PII balancing exercise that it is necessary to have some form of closed process, that same principle would permit such a process at common law.
Thus, at the conclusion of the PII process it will be necessary for the judge to decide how to proceed. If he is persuaded that it is necessary in the interests of justice that some form of closed process should take place, I can see no reason why such a process should not be followed. If, as is common ground, it was open to the courts to develop the common law in children or wardship cases or in confidentiality cases on the ground that it was necessary to do so in the interests of justice, the same principle should apply here. I appreciate that this is at the end of what has hitherto been described as the PII process. However, what to do at that point can to my mind fairly be regarded as part of that process. It is certainly very closely related to it.
As stated above, at that stage the whole process and the way forward will have to be reviewed by the judge in the light of the submissions of the parties. Much will depend upon those submissions. In this regard it is important to note that those submissions will depend upon the facts of the particular case. Thus in a case of this kind, where the state organs are the defendants in a civil claim, the position may be very different from a case in which the state has asserted or is asserting rights against a non-state party and it is the latter which needs information in order to enable it to defend its position. In the former case, as explained above, if the judge concludes that, absent some closed procedure the claim should be struck out because a fair trial to both parties is not possible, the claimant may consent to a closed procedure in order to allow his claim to proceed. In that event the action will proceed on the basis of some appropriate closed procedure. No question of necessity will arise. On the other hand, if the judge declines to strike the claim out, the defendants might seek an order for such a procedure based on necessity, namely that such a procedure would be necessary in order to permit a fair trial. Whether this is a realistic possibility should in my opinion be left for decision in a concrete case.
A closed procedure might also be necessary in a case in which it is the non-state party which wishes to rely upon the material which would otherwise be subject to PII in order to defend itself in some way against the state. In such a case either party might seek an order for such a procedure based on necessity, namely that such a procedure would be necessary in order to permit a fair trial.
These considerations to my mind demonstrate the importance of leaving the question of what, if any, procedure is permissible, to be determined in the context of a specific case. The court should not in my judgment now rule out the possibility of some form of closed procedure in any case in circumstances in which it is not possible to predict what may happen in the future. All should depend upon the particular circumstances after the judge has struck what may be called the PII balance.
In this regard the decision in R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs (No 2)  EWHC 152 (Admin),  1 WLR 2653 is instructive. At para 34 Thomas LJ set out four questions that arise when carrying out the balancing test:
Is there a public interest in bringing the redacted paragraphs into the public domain?
Will disclosure bring about a real risk of serious harm to an important public interest, and if so, which interest?
Can the real risk of serious harm to national security and international relations be protected by other methods or more limited disclosure?
If the alternatives are sufficient, where does the balance of the public interest lie?
Thomas LJ then identified factors relevant to determining where the balance of public interest lies. He did so under four headings:
the public hearing;
making decisions and reasons public;
public justice, the rule of law, free speech and democratic accountability; and
the role of the media.
In discussing the need to uphold the rule of law, he said, at para 41 that "the more serious the alleged infringement of the rule of law, the more strongly that principle applies" and at para 46 he emphasised that "the provision of information .... which enables public debate to take place and democratic accountability to be made more effective is one of the bases on which democracy rests".
Although the Court of Appeal expressed some disagreement with the Divisional Court on the facts, while they do not say so expressly, as I read their judgments, they agreed with its general approach in principle: see  EWCA Civ 65,  QB 218, per Lord Judge CJ at para 51, Lord Neuberger MR at para 187 and Sir Anthony May P at para 290. At para 295 Sir Anthony emphasised that the balance is not to be struck by the Foreign Secretary but by the court. As I read the judgments in Mohamed in both courts, in addition to that principle, they support these further propositions. First, the rule of law and the democratic requirement that governments be held to account mean that the case for disclosure will always be very strong in cases involving alleged misconduct on the part of the state and, secondly, that the more serious the alleged misconduct on the part of the state, the more compelling the national security reasons must be to tip the balance against disclosure.
As I see it, the special advocates will have a significant role at that time because under the present system they will have played a substantial part in the consideration of which documents are relevant and which are (and which are not ) subject to PII. In doing so, they will have learned a good deal about the issues between the parties and will have been able to make submissions to the judge on the question how the balance should be struck by the judge as between disclosure and non-disclosure. This might include inviting the judge to consider the balance between the public interest in non-disclosure on the one hand and the public interest in disclosure having regard to the importance of each in the particular context, which would or might include the importance of ensuring that the state does not abuse its power or the rights of the citizen.
Equally special advocates would be likely to be valuable in the future, not only in the context of striking that balance, but also both in a consideration of the question which may arise at the end of the PII process as to whether a non-state claimant should consent to some form of closed procedure or, in the case of a non-state defendant, whether the circumstances meet the requirement of necessity such as to justify a closed evidence procedure. In order to carry out these functions effectively and to provide a substantial measure of procedural justice to the party, it may well be that in the future special advocates will need to be permitted to communicate with the party and the parties' representative. The Joint Committee on Human Rights has long advocated relaxation of the rule against communication between special advocates and non-state parties, most recently in its report "Counter-Terrorism Policy and Human Rights (Sixteenth Report): Annual Renewal of Control Orders Legislation" (2010) HL Paper 64, HC 395. This is essentially an issue for Parliament but it may also be an issue which, in the absence of Parliamentary intervention, a judge considering whether to order a closed procedure will wish to address, although proper safeguards would of course have to be put in place to protect the public interest. As ever, all would depend upon the circumstances.
Finally, I should add that I am not persuaded that there is anything in the CPR which prevents an appropriate process being developed. I respectfully doubt that the CPR require an adversarial process at every point. In the pursuit of the overriding principle of dealing with cases justly it may well be necessary to introduce inquisitorial elements, as for example where the judge looks at some evidence which both parties do not. Further, for my part, I would not accept the argument that there is no significant difference in this context between a criminal case and a civil case. The decision in R v Davis  UKHL 36,  1 AC 1128 is of importance but it was a criminal case, where I well understand that the importance of the defendant being able to confront his accusers may be critical. The same may be much less important in a civil case, especially where the state is the defendant and not the accuser.
In all the circumstances I would allow the appeal and set aside the declaration granted by the Court of Appeal. I would do so because I can envisage circumstances in which it might be appropriate to develop the common law by directing that some form of closed material procedure take place. Such a development would, as I see it, be a further step in the courts' approach to PII which has already changed step by step without statutory assistance since Duncan v Cammell Laird. For my part I would not replace the declaration granted by the Court of Appeal with the declaration granted by the judge. I would simply leave these issues to be further considered in the light of the facts of a specific case.
I have considered the judgments of Lord Clarke and Lord Dyson. I agree with Lord Brown that each makes a compelling case. That case relates, however, to a general issue which did not arise on the facts of this case. Could the common law ever permit a closed material procedure to be used in litigation involving a civil claim? That issue is now doubly academic as this litigation has now been settled. The preliminary issue with which the Court of Appeal was concerned was a narrower one.
Lord Clarke has set out that issue at para 123 of his judgment. It is important to appreciate the full import of the issue. What was proposed was an alternative to the manner in which public interest immunity ("PII") is dealt with under the conventional process of discovery. What the conventional exercise involves is summarised by Lord Clarke at para 145 of his judgment. What that conventional exercise was likely to involve on the facts of the case is set out by Lord Clarke at para 135.
The proposed scheme, as set out in the preliminary issue, was a procedure which was intended to replace the conventional exercise. It would not be necessary, in the case of each document, for the minister and the court to balance the damage that would be done to the public interest if the document were disclosed against the damage to the administration of justice if it were not. Instead, any documents disclosure of which would be contrary to the public interest, as broadly defined in the preliminary issue, would be dealt with by the "closed material procedure" as defined.
Lord Clarke comments at para 130 that he cannot conceive of circumstances in which the court could properly make an order for the use of "closed material procedure" as defined. And at para 152 he states that such a course would require Parliamentary intervention. I agree. The common law develops incrementally. The change envisaged by the preliminary issue would be fundamental. Only Parliament could bring about such a change.
This reasoning would have sufficed to enable the Court of Appeal to give a negative answer to the preliminary issue, and I consider that it would have been more satisfactory had the Court taken this course. In the event the Court of Appeal gave the following reasons of principle for giving a negative answer to the preliminary issue  3 WLR 1069:
In our view, the principle that a litigant should be able to see and hear all the evidence which is seen and heard by a court determining his case is so fundamental, so embedded in the common law, that, in the absence of parliamentary authority, no judge should override it, at any rate in relation to an ordinary civil claim, unless (perhaps) all parties to the claim agree otherwise. At least so far as the common law is concerned, we would accept the submission that this principle represents an irreducible minimum requirement of an ordinary civil trial. Unlike principles such as open justice, or the right to disclosure of relevant documents, a litigant's right to know the case against him and to know the reasons why he has lost or won is fundamental to the notion of a fair trial.
Later in its judgment the Court of Appeal ranged more widely. It considered at para 51 the use of a closed material procedure after the conclusion of a conventional PII exercise, observing at para 53 that this was "not the closed material procedure as defined in the preliminary issue".
It is not always easy to determine in some parts of the judgment of the Court of Appeal whether it was addressing the closed material procedure as defined in the preliminary issue, or a closed material procedure of some other description. This is particularly true of the last paragraph of its judgment:
We leave open the question of whether a closed material procedure can properly be adopted, in the absence of statutory sanction in an ordinary civil claim such as the present, where all the parties agree, or in a civil claim involving a substantial public interest dimension (ie where the judge is not simply sitting as an arbiter as between the parties). Both principle and the authorities relied on below seem to us to suggest that a different conclusion may well be justified in such cases, albeit only in exceptional circumstances, but that is an issue which should be considered as and when it arises.
I consider that the judgment of the Court of Appeal should be treated as applying only to the preliminary issue with which it was concerned. Whether the general principles applied by the Court of Appeal would necessarily preclude the use of a different closed material procedure, not as a substitute for the conventional PII exercise, but to mitigate the injustice that can occur when relevant evidence is excluded from disclosure because of PII, is a question that should be left open until it actually arises, just like the question left open by the last paragraph of the Court of Appeal's judgment.
It follows that I do not propose to go down the hypothetical road followed by Lord Clarke in that part of his judgment that begins at para 154. Nor do I propose to consider the merits of Lord Dyson's dissent from that part of Lord Clarke's judgment. I would dismiss this appeal on the narrow ground set out in para 192 above.
Lord Rodger, who died before judgment was given in this case, had indicated that he would have dismissed the appeal.
Jonathan Crow QC, Rory Phillips QC, Karen Steyn, Daniel Beard & Peter Skelton (instructed by Treasury Solicitors) for appellant.
Dinah Rose QC, Richard Hermer QC & Charlotte Kilroy (instructed by Birnberg Peirce and Partners) for respondent.
all rights reserved