House of Lords portcullis
House of Lords
Session 2003 - 04
Publications on the Internet

Judgments - Regina v. Connor and another (Appellants) (On Appeal from the Court of Appeal (Criminal Division))
Regina v. Mirza (Appellant) (On Appeal from the Court of Appeal (Criminal Division)) (Conjoined Appeals)


SESSION 2003-04
[2004] UKHL 2
on appeal from: [2002] EWCA Crim 1236 and [2002] EWCA Crim 1235




Regina v. Connor and another (Appellants) (On Appeal from the Court of Appeal (Criminal Division))

Regina v. Mirza (Appellant) (On Appeal from the Court of Appeal (Criminal Division))

(Conjoined Appeals)



The Appellate Committee comprised:

Lord Steyn

Lord Slynn of Hadley

Lord Hope of Craighead

Lord Hobhouse of Woodborough

Lord Rodger of Earlsferry




Regina v. Connor and another (Appellants) (On Appeal from the Court of Appeal (Criminal Division))
Regina v. Mirza (Appellant) (On Appeal from the Court of Appeal (Criminal Division)) (Conjoined Appeals) [2004] UKHL 2


My Lords,

The principal question:

    1.  These appeals raise the question whether evidence about the deliberations of a jury, which reveal a lack of impartiality on the part of the jury, is always inadmissible under the common law secrecy rule however compelling the evidence may be and however grave the circumstances of the lack of impartiality may be. Two examples will illustrate the point. A juror reveals after verdict that during the jury deliberations it emerged that some members of the jury were associated with a Neo-Nazi group and that they urged the conviction of the accused because he was a black immigrant. In the second example a juror reveals after verdict that a majority of the jury refused to deliberate and that the jury ultimately arrived at a verdict of guilty by spinning a coin. Nobody would seriously contest that such guilty verdicts would be the result of perverse processes by judicial tribunals. Neither example is fanciful or extreme: both were suggested to me by decided cases. Moreover, anyone versed in criminal practice would be able to match these hypothetical cases with other equally telling examples.

    2.  In the two appeals now before the House, viz Mirza [2002] EWCA Crim 1235; [2002] Crim LR 921; and Connor and Rollock [2002] EWCA Crim 1236, the Court of Appeal gave separate judgments on 13 May 2002. The thrust of the judgments given by the Vice President, Rose LJ, with the concurrence of Hunt and Keith JJ, was that the Court of Appeal was bound by the decision of a differently constituted Court of Appeal in R v Qureshi [2001] EWCA Crim 1807; [2002] 1 WLR 518, which had enunciated an absolute rule of the secrecy of deliberations of the jury after verdict. It follows that even in cases such as I have described the Court of Appeal would feel compelled by stare decisis to rule that it has no jurisdiction to examine a possible miscarriage of justice in this corner of the law.

    3.  It is inherent in the opinions of the majority delivered today that even in the exceptional cases postulated the evidence must always be excluded and the conviction upheld. That reflects the submissions of counsel for the Lord Chancellor who asserted "that the residual possibility (of a miscarriage of justice) is the necessary price for the preservation and protection of the jury system": Case para 53. In contradistinction counsel for the Director of Public Prosecutions, when taxed with the problem I identified in paragraph 1, acknowledged that the point is one of great difficulty. Ultimately, he did, however, support the reasoning which has been upheld by the majority. It is to the effect that in the interests of maintaining the efficiency of the jury system the risk of occasional miscarriages of justice may acceptably be tolerated. In other words one must accept some dubious verdicts, even in cases of the utmost gravity, as the cost to be paid for protecting the jury system. While I acknowledge that the problem is one of acute difficulty, I cannot assent to the austere conclusion reached by the majority.

The risk of miscarriages of justice

    4.  Nowadays we know that the risk of a miscarriage of justice, a concept requiring no explanation is ever present. In earlier times courts sometimes approached the risk of a miscarriage of justice in ways which we would not nowadays find acceptable. In 1980 the Court of Appeal denied the Birmingham Six the right to sue the police in civil proceedings. Lord Denning MR said about the possible innocence of the men: "This is such an appalling vista that every sensible person in the land would say: It cannot be right that these actions should go any further": McIlkenny v Chief Constable of the West Midlands [1980] QB 283, at 323D. The men stayed in prison. Some 12 years later their convictions had to be quashed. Together the miscarriages in the cases of the Guildford Four, the Maguire Seven and the Birmingham Six were described by Lord Devlin as "the greatest disasters that have shaken British justice in my time": "The Conscience of the Jury"(1991) 107 LQR 398. It led to the appointment of a Royal Commission on Criminal Justice which reported in July 1993: Report (Cm 2263). One of the key messages of that Report was that the Court of Appeal must be readier to examine possible miscarriages of justice. One of the recommendations was the creation of new and independent arrangements for identifying miscarriages of justice. This recommendation was implemented in 1995 by the setting up of the Criminal Cases Review Commission: section 8 of Criminal Appeal Act 1995. It is an independent body with extensive powers to investigate complaints of miscarriages of justice. There was also a more general change in legal culture. A good illustration of that is the decision in R v Secretary of State for the Home Department, Ex p Simms [2000] 2 AC 115 where, in the face of some 60 miscarriages of justice in the 1990s, the House of Lords set aside Home Office instructions denying prisoners access to journalists in their efforts to get their convictions overturned. The philosophy became firmly established that there is a positive duty on judges, when things have gone seriously wrong in the criminal justice system, to do everything possible to put it right. In the world of today enlightened public opinion would accept nothing less. It would be contrary to the spirit of these developments to say that in one area, namely the deliberations of the jury, injustice can be tolerated as the price for protecting the jury system.

    5.  The scope of the problem should be put in context. In only 1% of criminal cases is there a trial by a judge and jury. In all other cases there is no jury and the law is, and always has been, that the accused has an absolute right to a trial before an impartial tribunal: see article 6(1) of the European Convention on Human Rights. If there are substantial reasons to doubt the impartiality of the tribunal, the matter must be examined in order to determine whether there has been a breach of this fundamental guarantee. That is so even if the reasons for doubt about the impartiality of the tribunal only emerged after the verdict. If there has been a breach, the conviction cannot stand. All this is elementary law. But in respect of the 1% of cases where there are jury trials it is said to be essential to prohibit absolutely any examination of evidence about jury deliberations which suggests that the jury was not impartial. On this basis, and to this extent, the law would then subordinate the risk of a miscarriage of justice to the interests of protecting the efficiency of the jury system. And it has to be faced that jury trials involve the most serious cases in the criminal calendar resulting sometimes in lengthy determinate sentences or life imprisonment. One is not dealing with a cost/benefit analysis: a miscarriage of justice bears on real individuals, their families, and communities. If the law requires individual cases to be subordinated to systemic considerations affecting the jury system, one may question whether the law has not lost its moral underpinning.

The jury is a judicial tribunal

    6.  It is important to take account of the fact that a jury is a judicial tribunal, and is expected to conform to judicial standards. Lord Devlin (Trial by Jury, 1956) explained (at p 41):

    "As the jury changed its character from a body of witnesses into a body of persons who had to determine facts on the evidence placed before them, it became a judicial tribunal and fit to be invested with judicial attributes. The judges punished as misconduct any deviation by the members of the jury from judicial standards and as contempt of court any interference by outsiders with the discharge of their judicial duties. There is no code embodying this. The rules came into existence piecemeal during the long period in which the jury was changing its character.


    Jurymen are invested with judicial immunity. They have full judicial privilege and are not accountable for anything said or done in the discharge of their office, and any threats or abusive language directed towards them as jurymen is punishable as contempt of court."

A jury is not above the law. As a judicial tribunal it must comply with the requirements of article 6(1) of the European Convention on Human Rights. In a case of a grave departure by the jury or jurors from judicial standards, a judge may before verdict have to discharge the jury. A subsequently revealed infringement of judicial standards by the jury or jurors may require the Court of Appeal to quash the conviction. A lack of impartiality would be a classic case requiring such action to be taken.

    7.  Lord Devlin observed "that trial by jury is more than an instrument of justice and more than one wheel of the constitution: it is the lamp that shows that freedom lives": Trial by Jury, (1956), p 164. This was a reference to the candles that were lit in London in the windows of London houses following the acquittal of the seven Bishops in 1688: see Macaulay, The History of England from the Accession of James II, (1849) vol 2, at p 389.The jury is an integral and indispensable part of the criminal justice system. The system of trial by judge and jury is of constitutional significance. The jury is also, through its collective decision-making, an excellent fact finder. Not surprisingly, the public trust juries. What public opinion would not tolerate are jury verdicts arrived at by perverse processes.

    8.  Everything possible must be done to ensure that the system works both efficiently and justly. In this respect counsel for the Director of Public Prosecutions emphasized the many safeguards already in place to ensure that the jury render their verdicts in accordance with a judicial process. All this is, of course, well known and laudable. Unfortunately, however, it does not touch on the problem which I have mentioned. If one wants to address the particular problem, it is necessary to accept that the Court of Appeal has the power in exceptional cases to examine material regarding jury deliberations tending to show that the jury or some of them were false to their oath.

    9.  Undoubtedly, there must be a general rule that the deliberations of the jury must remain secret. In Gregory v United Kingdom (1997) 25 EHRR 577 the European Court of Human Rights observed (p 594, para 44):

    "The court acknowledges that the rule governing the secrecy of jury deliberations is a crucial and legitimate feature of English trial law which serves to reinforce the jury's role as the ultimate arbiter of fact and to guarantee open and frank deliberations among jurors on the evidence which they have heard."

Moreover, there is a strong rebuttable presumption that the jury was impartial: Rojas v Berllaque and HM Attorney General for Gibraltar, [2003] UKPC 76.

    10.  The question is whether the rule about the secrecy of jury deliberations is indefeasible in all circumstances, however extreme, and even in the face of evidence disclosed after a verdict demonstrating a real risk that the jury was not an impartial tribunal. This is ultimately a question of the jurisdiction of our appellate courts.

The decided cases

    11.  Judges of great experience and distinction have held that it is never permissible to admit evidence of what happened during the deliberations of the jury: R v Thompson [1962] 1 All ER 65, 66, per Lord Parker CJ; Ellis v Deheer [1922] 2 KB 113, 117 - 118, per Bankes LJ; at p 121, per Atkin LJ; Attorney General v New Statesman and National Publishing Company Ltd [1981] QB 1,10, per Lord Widgery CJ; R v Miah [1997] 2 Cr App R 12, 18 - 19, per Kennedy LJ; Roylance v General Medical Council (No 2) [2000] 1 AC 311, 324B, per Lord Clyde; R v Qureshi [2002] 1 WLR 518, per Kennedy LJ. The only exception is that where there has been, or may have been, an irregular occurrence of an extraneous nature, which may have compromised the impartiality of the jury the evidence may be admitted: Ras Behari Lal v King-Emperor (1933) 50 TLR 1; R v Hood [1968] 1 WLR 773; R v Brandon (1969) 53 Cr App R 466; R v Young (Stephen) [1995] QB 324. The position is similar in Scotland: Stewart v Fraser (1830) 5 Murray 166; Swankie v H M Advocate (1999) SCCR 1. Subject to differences as to the scope of the exception, a similar exclusionary rule has prevailed in Commonwealth countries: Canada: R v Pan; R v Sawyer [2001] 2 SCR 344; Australia: R v Andrew Brown [1907] 7 NSWSR 290; R v Medici (Court of Criminal Appeal, Victoria, 5 June 1995); New Zealand: R v Papadopoulos [1979] 1 NZLR 621.

The Qureshi case

    12.  It would not be helpful to review the case law in detail. But it is worth returning to Qureshi because it was the basis on which the Court of Appeal dismissed the appeals in the present case. A major complaint in Qureshi was the allegation of overt racism in the jury room. In a note on Qureshi [2002] Crim LR 62 Sir John Smith QC commented as follows:

    "If the allegations in the present case were true and if the court had been able to inquire into them, it seems likely that the conviction would have been held to be unsafe. As it is, we shall never know. Not a happy situation. . . ."

"We shall never know" fits in uneasily with modern conceptions of fairness and due process in the criminal justice system. Another perspective of Qureshi is provided by Professor J R Spencer [2002] 61CLJ 291 who commented, at p 293:

    "The nub of the decision in Qureshi, one suspects, is the remark that if the appeal were allowed it might lead 'to many such complaints, some perhaps owing their origin to friends or relatives of the defendant'. But this simply will not do. The fact that many allegations of this sort are false cannot justify ignoring all of them because, as Young [the ouija board case] so painfully reminds us, some of them regrettably are true."

Qureshi is a questionable foothold of the rule of absolute secrecy in all circumstances. Saying "never" is in law sometimes fraught with future difficulties. Absolutist judicial pronouncements frequently do not survive the gauntlet of experience. Lord Goff of Chieveley and Lord Bingham of Cornhill deliberately avoided this trap and have not closed the door completely: Lalchan Nanan v The State [1986] AC 860, at 872, per Lord Goff of Chieveley; R v Millward [1999] 1 Cr App R 61, at 65G, per Lord Bingham of Cornhill.

The rationale of the secrecy rule

    13.  The rationale of the rule that evidence relating to what occurs in the jury room is inadmissible has often been explained. Drawing on English and Canadian jurisprudence Arbour J, in giving the judgment of the Canadian Supreme Court in Sawyer, recently gave the reasons for the rule in terms which cannot be improved on. She said, at pp 373-375:

    "The first reason supporting the need for secrecy is that confidentiality promotes candour and the kind of full and frank debate that is essential to this type of collegial decision making. While searching for unanimity, jurors should be free to explore out loud all avenues of reasoning without fear of exposure to public ridicule, contempt or hatred. This rationale is of vital importance to the potential acquittal of an unpopular accused, or one charged with a particularly repulsive crime. In my view, this rationale is sound, and does not require empirical confirmation.

    "The Court of Appeal also placed considerable weight on the second rationale for the secrecy rule: the need to ensure finality of the verdict. Describing the verdict as the product of a dynamic process, the court emphasized the need to protect the solemnity of the verdict, as the product of the unanimous consensus which, when formally announced, carries the finality and authority of a legal pronouncement. That rationale is more abstract, and inevitably invites the question of why the finality of the verdict should prevail over its integrity in cases where that integrity is seriously put in issue. In a legal environment such as ours, which provides for generous review of judicial decisions on appeal, and which does not perceive the voicing of dissenting opinions on appeal as a threat to the authority of the law, I do not consider that finality, standing alone, is a convincing rationale for requiring secrecy.

    "The respondent, as well as the interveners supporting its position and, in particular, the Attorney General of Quebec, place great emphasis on the third main rationale for the jury secrecy rule - the need to protect jurors from harassment, censure and reprisals. Our system of jury selection is sensitive to the privacy interests of prospective jurors (see R v Williams [1998] 1 SCR 1128), and the proper functioning of the jury system, a constitutionally protected right in serious criminal charges, depends upon the willingness of jurors to discharge their functions honestly and honourably. This in turn is dependent, at the very minimum, on a system that ensures the safety of jurors, their sense of security, as well as their privacy.

    "I am fully satisfied that a considerable measure of secrecy surrounding the deliberations of the jury is essential to the proper functioning of that important institution and that the preceding rationales serve as a useful guide to the boundaries between the competing demands of secrecy and reviewability."

    It is now possible, against the background of the dicta in the decided cases and the reasons for the rule, to take stock of the nature and scope of the secrecy rule regarding jury deliberations.

The interpretation of the decided cases

    14.  The degree of consensus revealed by the case law in this jurisdiction and elsewhere is remarkable. All the decisions hold that, except in the case of extraneous influences on the jury, evidence about the deliberations in the jury room may never be admitted. On the other hand, judgments must be read not mechanically but with reference to the issues considered by the court. A dynamic interpretation of the case law is necessary. In none of the decisions, in England or abroad, did the court face and confront the far-reaching situations which I outlined in paragraph 1 of this opinion. In none of the judgments is there any clear indication that the courts would have been prepared to uphold an absolute rule excluding evidence about jury deliberations in cases where there is credible evidence, disclosed after verdict, showing that the jury was not impartial. In none of the decided cases has it been held that the court may in the interests of the efficient functioning of the jury system tolerate real identifiable risks of miscarriages of justice. In my view therefore as a matter of precedent there is nothing in the case law which prevents the House from ruling today that in the cases of the type postulated evidence impinging on jury deliberations may exceptionally be admitted. The allegiance that a judge owes to earlier precedent is attenuated where it is clear that the precise problem was not considered. This is particularly the case where there is cogent evidence suggesting a real risk that the jury reached their verdict by a fundamentally perverse process.

    15.  It must also be borne in mind that in cases involving extraneous influence on the jury the court may on established principle hear evidence about such issues. That may require the admission of evidence about how the extraneous influence was brought to bear on the jury in their deliberations. For example, adapting the facts of R v Young (Stephen) [1995] QB 324, it is accepted by counsel for the Director of Public Prosecutions, that if the jury brought an ouija board into the jury room and determined the issue by consulting it, the exclusionary rule would not apply. Evidence may then be led about what actually happened. It would, of course, be absurd not to allow such evidence to be led. On the other hand, counsel for the Director of Public Prosecutions felt constrained to argue that, if the foreman of a jury took a coin out of his pocket in the jury room, the evidence about the tossing of a coin in the jury room to obtain a verdict, was inadmissible. Such absurd distinctions do not reflect well on our jurisprudence. Rhetorically I would ask: What justification in logic, common sense and fairness can there then be for not admitting evidence about jury deliberations tending to establish that the verdict was the result of a fundamentally tainted process?

    16.  Public confidence in the legitimacy of jury verdicts is a foundation of the criminal justice system. And there must be a general rule making inadmissible jury deliberations. But it is difficult to see how it would promote public confidence in the criminal justice system for the public to be informed that our appellate courts observe a self denying rule never to admit evidence of the deliberations of a jury even if such evidence strongly suggests that the jury was not impartial. In cases where there is cogent evidence demonstrating a real risk that the jury was not impartial and that the general confidence in jury verdicts was in the particular case ill reposed, what possible public interest can there be in maintaining a dubious conviction?

    17.  The common law rule is a judge made rule. Where the reason for a judge made rule stops, it may be appropriate to qualify its reach. Where a new situation arises which was never previously considered the scope of the rule may be reconsidered. That is the case here. So far as judges have propounded a rule which, in the light of experience, is potentially productive of injustice, it is not beyond their power to put the matter right. I would hold that under our domestic law the court has the power to admit evidence about the deliberations of the jury in the circumstances I described in paragraph 1 of this opinion.

Article 6(1)

    18.  The European Court of Human Rights have considered issues regarding jury trials, notably in the context of the disclosure before verdict of matters reflecting on the impartiality of the jury: see Gregory v United Kingdom (1997) 25 EHRR 577; Sander v United Kingdom (2000) 31 EHRR 1003. These decisions turned on their facts and are not directly in point. If the problem which I have posed arises before it, the ECHR will in my view be bound to uphold the fundamental guarantee of a right to a trial before an impartial tribunal under article 6(1). It would have to act so as to give effective protection to that fundamental right. An indication of the likely approach of the ECHR is to be found in Remli v France (1996) 22 EHRR 253. It involved an alleged racist remark by a juror. The defendant raised an objection. The court declined to investigate the matter. The defendant was convicted. It was therefore a post verdict case and relatively close to the issues under consideration in the present case. The ECHR held:

    "It is not for the Court to rule on the evidential value of Mrs M's written statement or on whether the racist remark attributed to the juror in question was actually made. It notes merely that Mrs M's statement - which contained a serious allegation in the context of the case - was filed with the Assize Court by the applicant's lawyers, who asked the Court to take formal note of it. The Court dismissed their application without even examining the evidence submitted to it, on the purely formal ground that it was 'not able to take formal note of events alleged to have occurred out of its presence'. Nor did it order that evidence should be taken to verify what had been reported - and, if it was established, take formal note of it as requested by the defence - although it could have done so. Consequently, the applicant was unable either to have the juror in question replaced by one of the additional jurors or to rely on the fact in issue in support of his appeal on points of law. Nor could he challenge the juror, since the jury had been finally empanelled and no appeal lay against the Assize Court's judgment other than on points of law.

    Like the Commission, the Court considers that article 6(1) of the Convention imposes an obligation on every national court to check whether, as constituted, it is 'an impartial tribunal' within the meaning of that provision where, as in the instant case, this is disputed on a ground that does not immediately appear to be manifestly devoid of merit.

    In the instant case, however, the Rhône Assize Court did not make any such check, thereby depriving Mr Remli of the possibility of remedying, if it proved necessary, a situation contrary to the requirements of the Convention. This finding, regard being had to the confidence which the courts must inspire in those subject to their jurisdiction, suffices for the Court to hold that there has been a breach of article 6(1)."

This is, of course, an approach diametrically opposite to that which prevailed in Qureshi.

    19.  In my view it would be an astonishing thing for the ECHR to hold, when the point directly arises before it, that a miscarriage of justice may be ignored in the interests of the general efficiency of the jury system. The terms of article 6(1) of the European Convention, the rights revolution, and fifty years of development of human rights law and practice, would suggest that such a view would be utterly indefensible.