R v Spencer [1987] UKHL 2

House of Lords
Between
RAppellantSpencerRespondent
Date 24 June 1984
It is Ordered and Adjudged, by the Lords Spiritual and Temporal in the Court of Parliament of Her Majesty the Queen assembled, That in the cause of Regina against Spencer and others the said Orders of Her Majesty's Court of Appeal (Criminal Division) of the 7th day of November 1984, complained of in the said Appeals be, and the same are hereby, Set Aside, and That the convictions by Nottingham Crown Court of the 24th June 1984 be, and the same are hereby Quashed: That in the cause of Regina against Smails and others the said Orders of Her Majesty's Court of Appeal (Criminal Division) of the 7th day of November 1984 complained of in the said Appeals be, and the same are hereby, Affirmed and that the said Petitions and Appeals be, and the same are hereby, dismissed this House; And it is further Ordered and Adjudged, That in both causes the Certified Question be amended so as to read
"In a case where the evidence for the Crown is solely that of a witness who is not in one of the accepted categories of suspect witnesses, but who, by reason of

Judgment: 24.7.86

HOUSE OF LORDS

REGINA v. SPENCER AND OTHERS (APPELLANTS)

REGINA v. SMAILS AND OTHERS (APPELLANTS)

(ON APPEAL FROM THE COURT OF APPEAL (CRIMINAL

DIVISION))

(CONSOLIDATED APPEALS)

Lord Chancellor Lord Bridge of Harwich Lord Brandon of Oakbrook Lord Mackay of Clashfern Lord Ackner

LORD CHANCELLOR

My Lords,I agree with the disposal of the second appeal about to be recommended by my noble and learned friend, Lord Ackner, whose judgment on the merits I have had the advantage of reading in draft, and which is now available in print. I also agree with his proposed answer to the certified question.As regards the first appeal, (that relating to Spencer and others), I have no doubt that this appeal should be allowed on the ground that owing to irregularities in the trial, in relation to a particular juror, the verdict must be regarded as unsafe and unsatisfactory. I do not believe that, in the rather special circumstances of this case, the Court of Appeal were entitled to speculate as to what may or may not have occurred after the jury had retired or as to the circumstances which may or may not have influenced their verdict of guilty by a majority of ten to one after deliberating for five hours.Without such speculation the following facts are clear beyond peradventure.(1) The first is that, just before the jury were about to retire, and on an application by defending counsel to discharge the jury, the judge learned of circumstances which, in his opinion at least, made it unsafe to permit one juryman ("P.") to remain on the jury. The avowed reason when given

- 1 -

was that the juryman in question had, through his wife, a somewhat tenuous connection with the institution at which the accused were employed as nurses, but one is, I think, entitled to assume that the judge considered it arguable that, through his wife, the juryman might have been exposed to gossip prejudicial to the accused about the merits of the case. What is beyond doubt is that, even at this stage, and apparently throughout the trial, the juryman in question had unequivocally evinced a most unjudicial bias against the defence and this may well have acted as an additional factor in influencing the judge's mind. However I believe in these circumstances that the judge acted correctly in then declining to discharge the whole jury.The second fact beyond dispute is that when, on the following day, defence counsel renewed his application to discharge the jury, in refusing the application, the judge first of all misdirected himself as to the correct test to apply and secondly took into account factors which should have been disregarded in considering the result of the application. In considering such an application, the interests of justice should be paramount, and neither the inconvenience of a second trial nor the necessity which would have been involved in calling again as witnesses the victims of the alleged assaults, possibly to their detriment, should have outweighed the necessity of the accused receiving, and being seen to receive, a fair trial. The effect of these misdirections when the point was raised on appeal was to free the Court of Appeal in the use of their own discretion of any necessity to be guided by the judge's own use of his discretion.The third point which is beyond dispute is that, after discharging P. from the jury, the judge permitted him to remain behind in the precinct of the court in order to give a lift in his own car to three of the remaining members of the jury over a journey taking about half-an-hour. My own view is that the judge was mistaken so to do, but, whether this view be correct or not, the judge thought it imperative specifically to warn the three jurors who were to avail themselves of the lift not to discuss the case with P. when they were with him in the car. What the judge did not know on the renewal of the application on the following day, but what, by the time they heard the appeal, the Court of Appeal certainly did know, was that this warning had been disregarded. I do not believe that the judge could have exercised his discretion so as to refuse the defence application when it was renewed the following day, at least without diligent enquiry as to the nature of the discussion, had he been aware that his specific warning had been disregarded in this way.In the event, the Court of Appeal formed the view that there was "no realistic chance" of the discussion in the car (whatever it was), having influenced the verdict. Although they had available to them the statement prepared by the D.P.P. they seem to have made no additional enquiry as to what had happened, and though, of course, it may be true that nothing detrimental had taken place, I do not think

- 2 -

they had enough material on which to say that it had not. This was a case of total want of corroboration, a majority verdict of 10-1, and a wholly unauthorised and improper conversation between a juror who had been sent off the jury as not suited to remain there with three of the remaining jurors in the absence of the other eight and in disregard of an express warning by the judge against such conduct. In my view this constituted a serious breach of natural justice and the resulting verdict must be treated as unsafe and unsatisfactory. In such a case as this there can be no question of applying the proviso.As regards the certified question, I am content to accept what is about to fall from my noble and learned friend. But, in view of certain parts of the judgment of the Court of Appeal, there are certain additional comments I would wish to make. It is notorious that the law of Scotland does not admit of a conviction on the uncorroborated evidence of a single witness, but that the law of England does. There are a certain number of statutory exceptions to which I drew attention in Reg. v. Kilbourne [1973] A.C. 729 at pp. 739-740. In addition there are certain categories or classes of case which, by the time of the establishment of the Court of Criminal Appeal in 1907, had already hardened into firm rules of practice. In these, in the absence of a special warning from the trial judge, a conviction will normally be set aside as unsafe or unsatisfactory unless the court decides that it is safe to apply the proviso. I referred to these at p. 740 of the report. My reference to cases of alleged mistaken identity has now been overtaken by events, viz. by the Devlin Report (Report to the Secretary of State for the Home Department of the Departmental Committee on Evidence of Identification in Criminal Cases (1976) H.C. 338) and Reg. v. Turnbull [1977] Q.B. 224, and need not, therefore, be referred to further, but there are also references to what I said in Kilbourne, both in Archbold Criminal Pleading Evidence & Practice 42nd ed. (1985) p. 1136 and in the judgment of the Court of Appeal which, in my view at least, need correction. In Reg. v. Hester [1973] A.C. 296, which shortly preceded the hearing of Kilbourne, Lord Morris of Borth-y-Gest had said, at p. 315:
"The weight of the evidence is for the jury ... It is for the jury to decide whether witnesses are creditworthy. If a witness is not, then the testimony of the witness must be rejected. The essence of corroborative evidence is that one creditworthy witness confirms what another creditworthy witness has said. Any risk of the conviction of an innocent person is lessened if conviction is based upon the testimony of more than one acceptable witness."
When in Kilbourne I said at p. 746 "corroboration can only be afforded to or by a witness who is otherwise to be believed. If a witness's testimony fails of its own inanition, the question of his needing, or being capable of giving, corroboration does not arise", I was there expressly referring back to this passage in Lord Morris' speech in Hester, and the suggestion that this is an alternative approach to that of Lord Reid at p. 750 would, I think, have astonished Lord Morris and Lord Reid as much as it did me when first I read it and I said as much in Reg. v. Boardman [1975] A.C. 421, 454. The contrary view seems to me pure fantasy.

- 3 -

The other point on which I would wish to comment is the Court of Appeal's view on my reference in Kilbourne [1973] A.C. 729, 740 I added witnesses "of admittedly bad character" to the number of cases where a warning of some kind was required as to the danger of convicting without corroboration. I was, of course, using the phrase in the technical sense of witnesses who have been shown to be not of a character to make them worthy to be believed on their oath. In this connexion I must say that even if there were not authority to support this view, (and I believe there is plenty), I would regard it as a matter of sheer common sense that if a judge did not warn the jury of the possible danger of convicting an innocent man if they convicted solely on the disputed but uncorroborated testimony of such a person, his failure to do so would, apart from the proviso, make a verdict unsafe and unsatisfactory in the extreme. The fact is, I believe, that in the instant appeal, the Court of Appeal, confronted with the contrary decision in Reg. v. Bagshaw [1984] 1 W.L.R. 477 with which they rightly disagreed, were haunted by the spectre of Young v. Bristol Aeroplane Co. Ltd. [1944] KB 718 . However this may be, your Lordships were not troubled by this spectre, nor do I wish to pursue it in the present case, since I am fully persuaded by my noble and learned friend, that Bagshaw was wrongly decided and that, even on the view most favourable to the appellants in Bagshaw and in the instant appeals, Judge Hopkin gave a perfectly adequate warning to the jury of the danger of convicting on the uncorroborated testimony of the witnesses for the prosecution. He not only advised them of the necessity of proceeding with extreme caution, but warned them specifically of the dangers they would be incurring if they did not do so.The cases which were cited to the contrary, e.g. Reg. v. Price (Herbert) [1969] 1 Q.B. 541, were, as it seems to me, cases in which the particular dangers were not sufficiently spelled out in detail as the trial judge did here.The only other observation I would make on the certified question is that the modern cases, quite correctly in my view, are reluctant to insist on any magic formula or incantation, and stress instead the need that each summing up should be tailor made to suit the requirements of the individual case. c.f. per Lord Morris of Borth-y-Gest in Reg. v. Hester [1973] A.C. 296 at p. 309, per Lord Pearson at p. 321, and Lord Diplock at 325, 328. In particular, when as here, it is agreed that no corroboration exists, a disquisition on what can or could amount to such if corroboration were needed is emphatically not required and greatly to be discouraged (per Lord Diplock, loc.cit.) Speaking for myself, I even dislike the expression "categories" as applied to the cases. They are simply classes of case where the experience of the courts has gradually hardened into rules of practice, owing, as my noble and learned friend points out, partly to the inherent dangers involved, and partly to the fact that the danger is not necessarily obvious to a lay mind. The less juries are confused by superfluous learning and the more their minds are directed to the particular issues relevant to the case before them, the more likely they are, in my view, to arrive at a just verdict.

- 4 -

LORD BRIDGE OF HARWICH

My Lords,I have had the advantage of reading in draft the speeches of my noble and learned friends, the Lord Hailsham of St Marylebone L.C. and Lord Ackner. In regard to the first appeal I agree with the views expressed by my noble and learned friend on the Woolsack and would allow the appeal for the reasons he has given. I would dismiss the second appeal for the reasons given by my noble and learned friend Lord Ackner and answer the certified question as he proposes.

LORD BRANDON OF OAKBROOK

My Lords,I have had the advantage of reading in draft the speech prepared by my noble and learned friend, Lord Ackner. I agree with it, and for the reasons which he gives I would allow the appeals of Spencer, Ball and Mason, and dismiss those of Smails, Ball and White.

LORD MACKAY OF CLASHFERN

My Lords,I have had the privilege of reading in draft the speech prepared by my noble and learned friend Lord Ackner. I agree with it and with the orders he proposes in these appeals.

LORD ACKNER

My Lords,The appellants in these consolidated appeals were members of the nursing staff at Rampton Hospital, which is a secure hospital catering for patients suffering from mental disorders. The majority of such patients have been sent to Rampton as the result of court orders made under the Mental Health Act 1959 or the Mental Health Act 1983, by reason of their having been convicted of serious crimes. A television programme was shown in 1979 which made a substantial number of allegations of ill-treatment to patients by the nursing staff over a period of some years. This led to police inquiries and resulted in fourteen separate trials at the Crown Court at Nottingham. In five of those trials, the defendants were convicted and your Lordships are concerned with two of such trials. The first four of the fourteen trials were presided over by High Court judges, and the remaining trials, with

- 5 -

the approval of the Presiding Judges of the Circuit, were presided over by His Honour Judge Hopkin.On 24 June 1983 in the Nottingham Crown Court, the three appellants, Spencer, Ball and Mason, were convicted, as to Spencer on six counts of ill-treating a patient contrary to section 126 of the Mental Health Act 1959 and as to Ball and Mason, each on one similar count. Spencer was sentenced to 12 months' imprisonment on each count concurrent, suspended for two years. Ball and Mason were each sentenced to six months' imprisonment, suspended for 18 months. On 12 October 1983 in the same Crown Court, the appellants White, Smails and Ball again were each convicted of one similar offence contrary to the provisions of the same section of the Act of 1959. On 19 October 1983 they were each sentenced to six months' imprisonment suspended for 18 months. In both trials the convictions arose from substantially similar facts and the appeals were the subject of a single judgment in the Court of Appeal (Criminal Division) [1985] Q.B. 771. In each appeal the same question as to the adequacy of the judge's direction to the jury on the way in which they should treat the evidence of the complainants was raised. Hence the order for consolidation of the two appeals before your Lordships' House.In all the trials, the prosecution case against the nurses depended very largely upon the uncorroborated evidence of a single patient. I say, "very largely", because there were apparently some alleged incidents of violence which were witnessed by patients who were themselves complainants to other incidents. Judge Hopkin fully appreciated that in those circumstances very careful thought had to be given as to how he should direct the jury to approach the prosecution's case. In one of the earlier, if not the earliest, of the trials over which he presided, he informed counsel, in advance of his summing up, of the warning which he proposed to give to the jury. This warning was fully approved by leading counsel Mr. J. R. Roberts Q.C. for the prosecution, who has appeared before us on behalf of the Crown, and Mr. Wilfred Steer Q.C. who appeared for the defence in all or the majority of the trials, including the two the subject matter of this consolidated appeal. It was further agreed between counsel and approved by the judge, that for the greater protection of the accused, the judge would inform the jury in his summing up, that one patient could not be treated as supporting another, with the result that the prosecution case depended entirely upon the uncorroborated evidence of a single patient. This was but one example of the care that was taken to ensure complete fairness to the accused. The judge's direction as to the way in which they should approach the evidence of the complainants was in substance the same in each case. At an early stage in his summing up he gave them a general instruction. There is no real difference in what he said in either of the two appeals and I therefore quote the language which he used in the first of the two trials. He said:
"You must, ladies and gentlemen, approach the evidence of Mr. Hosein, Mr. Firth, Mr. Evan Glyn Hughes, Mr. William Hughes, Mr. Alldred, and Mr. Nugent with great caution. Why? Weil for three reasons, Firstly, because they are all persons of bad character. The law, in rules which are formulated over many years, requires me to tell you even if they were merely persons of bad character and nothing else,

- 6 -

that you must approach their evidence with great caution. It goes further than that of course In this case. The second reason is this, that at the time of these events they were all persons suffering from some form of mental disorder. Thirdly, they may of course have all conspired together to make false allegations. People make false allegations as we know for all sorts of reasons, some of which have been suggested in this case, but apart from that your own experience no doubt tells you that people do on occasions make false allegations. So therefore I must tell you that as far as all those patients are concerned you must approach their evidence with great caution. You would be wise to look for support for their evidence for those reasons. I tell you at once, Mr. Steer is right, and Mr. Roberts does not argue to the contrary, that there is no support for their allegations at all because as far as they are concerned even where you have two or three of them giving evidence upon one particular count, one witness of this type cannot support another. But, if, ladies and gentlemen, having seen them and having heard them and borne in mind the warning which I have given to you you come to the conclusion that you are sure in all or any of their cases they are telling the truth, then you may convict upon their evidence without there being any support at all. It is, as I say, for you to judge the facts. When I say the facts I mean also the importance which you attach to any particular witness and whether or not he is telling the truth."Having been given this warning, he then referred to the three defendants, emphasising that they were all men of good character, describing each in detail and referring to the testimonials which witnesses at the trial had given to each of them. He drew particular attention to the fact that the alleged incidents had occurred some years ago, that Rampton was part a prison and part a hospital, that in both parts there were outbreaks of violence which had to be quelled, and that those who looked after the patients had on occasions to resort to force. He referred in some detail to the absence of complaints by the patients, and the factors which the jury might consider taking into consideration when considering the significance of this aspect of the case. The judge, taking each count separately, then drew the jury's attention immediately to the character and characteristics of the complainant. He dealt in detail with his previous convictions, with his progress at Rampton, which had been the subject matter of medical evidence, and to the views of the psychiatrist employed at Rampton as to his personality defects, e.g. that he was prone to fabricate without any conscience and to pursue wrong without any feeling - he had a tendency to resentment against authority and to fly into a mood which might involve considerable violence (count one - Mr. Hosein).In view of my earlier reference to the judge having sought and obtained counsel's agreement to his proposed direction to the jury as to the way in which they should approach the evidence of the complainants, the circumstances in which his direction came to be criticised in the Court of Appeal is of some relevance. On 11 May 1982 Bagshaw, Holmes, and Starkey were found guilty at the Crown Court at Nottingham of ill-treating patients contrary to section 126 of the Act of 1959 and were sentenced by Judge

- 7 -

Hopkin. They applied to the Court of Appeal (Criminal Division) [1984] 1 W.L.R. 477 for leave to appeal against these convictions on the grounds that they were unsafe by reason of the inadequacy of the evidence, asserting that it was uncorroborated and unsupported in any material particular and that the patients who gave the evidence were inherently unreliable. We were told that in the application for leave to appeal it was stated In terms that:
"No criticism is or could be made of the summing up."
The Court of Appeal apparently thought otherwise, and during the hearing gave leave to the appellants to amend their grounds so as to question the adequacy of the judge's direction to the jury as to the treatment of the evidence of the complainants. A short adjournment was granted for this new approach to be considered, and after hearing further submissions the Court of Appeal reserved its judgment.The warning given by Judge Hopkin in the Bagshaw case was in similar, though perhaps stronger terms. O'Connor L.J., in giving the judgment of the court, said at an early stage in the judgment [1984] 1 W.L.R. 477, 479:
"We should like to say at once that the judge's summing up is a masterpiece of lucidity and fairness. He gave an impecable direction to the jury that they should treat the evidence of the complainants with the greatest caution. The question is whether these witnesses were such that a full warning was required, namely, that it was dangerous to convict upon their unsupported evidence."
The "full warning" to which O'Connor L.J. was referring is the warning as to the danger of convicting upon uncorroborated evidence, which by rule of practice has to be given if the prosecution is relying upon the evidence of an accomplice, or the victim of a sexual offence or the sworn evidence of a child.Having considered the well known cases of Reg. v Kilbourne [1973] A.C. 729 and Reg. v. Hester [1973] A.C. 296, the judgment continued at [1984] 1 W.L.R. 477, 484:
"Patients in hospital under the Mental Health Act 1959 are not a category like accomplices or complainants in sexual cases, nor would we wish to make them into an additional category. Patients detained in a special hospital after conviction for an offence or offences, even if they are not a category, may well fulfil to a very high degree the criteria which justify the requirement of the full warning in respect of witnesses within accepted categories. It seems to us that in such cases nothing short of the full warning that it is dangerous to convict on the uncorroborated evidence of the witness will suffice."
Thus, because Judge Hopkin had not used the words "it is dangerous to convict" the Court of Appeal concluded that the convictions were unsafe and they were therefore quashed. The court, however, in the final paragraph of the judgment commented in relation to the original grounds of appeal, that if they had stood alone they might or might not have sufficed adding at p. 484:

- 8 -

"We say that because we are conscious that in practice it would mean that the protection afforded to patients by section 126 of the Act of 1959 would be cut down to a large extent. It would be tantamount to saying that a conviction based on the uncorroborated evidence of such a complainant could not be safe. That would be to step outside the common law and usurp the function of Parliament."
When the appeals with which your Lordships are concerned came to be heard by the Court of Appeal, differently constituted, the first submission made on behalf of the appellants was that the court was bound by the decision in Reg. v. Bagshaw, and since that case could not be distinguished on its facts, the appeals should be allowed. On behalf of the Crown, Mr. Roberts submitted to the Court of Appeal that the court's decision in the Bagshaw case was reached per incuriam. In Bagshaw, Mr. Roberts had not come prepared to deal with the amendment which was made only at the suggestion of the court. Although following the short adjournment he sought to deal with the new ground as adequately as he could in his reply, he did not deal with the point as fully as he would otherwise have done, and in particular the court's attention was not drawn to an earlier decision of the Court of Appeal, namely, to Reg. v. Beck [1982] 1 W.L.R. 461. In that case the main ground of appeal was that the judge wrongly failed to direct the jury that it would be dangerous to act on the uncorroborated evidence of three witnesses, none of whom could be considered as participants or involved in the crime charged, but who "had a purpose of their own to serve in giving evidence, namely to cover up false representations made or acceded to by them in the insurance claim." In Beck's case the appellant's counsel based his contention that such a warning should have been given essentially upon the case of Reg. v. Prater [1960] 2 Q.B. 464. In that case a co-prisoner, who could have been considered an accomplice, gave evidence. The Common Serjeant did not give a warning in regard to his testimony and the danger of acting upon it unless corroborated. Edmund-Davies J. in the course of the judgment of the Court of Criminal Appeal said at p. 466:
"For the purposes of this present appeal, this court is content to accept that whether the label to be attached to Welham in this case was strictly that of an accomplice or not, in practice it is desirable that a warning should be given that the witness, whether he comes from the dock, as in this case, or whether he be a Crown witness, may be a witness with some purpose of his own to serve .... In the circumstances of the present appeal it is sufficient for this court to express the view that it is desirable that, in cases where a person may be regarded as having some purpose of his own to serve, the warning against uncorroborated evidence should be given. But every case must be looked at in the light of its own facts."
The Court of Appeal in Reg. v. Beck [1982] 1 W.L.R. 461, 468 having considered subsequent decisions in which Prater's case was criticised, concluded that the phrase in Reg. v. Prater [1960] 2 Q.B. 464, 466 "it is desirable that in cases where a person may be regarded as having some purpose of his own to serve, the warning against uncorroborated evidence should be given" is confined to

- 9 -

cases where witnesses may be participants or involved in the crime charged. The Court of Appeal further observed at [1982] 1 W.L.R. 461, 469 that:
"While we in no way wish to detract from the obligation upon a judge to advise a jury to proceed with caution where there is material to suggest that a witness's evidence may be tainted by an improper motive, and the strength of that advice must vary according to the facts of the case, we cannot accept that there is any obligation to give the accomplice warning with all that that entails, when it is common ground that there is no basis for suggesting that the witness is a participant or in any way involved in the crime the subject matter of the trial."
The phrase, "with all that that entails," requires perhaps further explanation. Where there is no corroboration, the rule of practice merely requires that the jury should be warned of the danger of relying upon the sole evidence of an accomplice or of the complainant in the sexual case, or upon the evidence of a child. The warning to be sufficient must explain why it is dangerous so to act, since otherwise the warning will lack significance. The jury are, of course, told that while as a general rule it is dangerous so to act, they are at liberty to do so if they feel sure that the uncorroborated witness is telling the truth. Where, however, there is evidence before the jury which they can properly consider to be corroborative evidence the position becomes less simple. The trial judge has the added obligation of identifying such material , and explaining to the jury that it is for them to decide whether to treat such evidence as corroboration. He should further warn them against treating as potential corroborative evidence, that which may appear to them to be such, but which is not so in law, e.g. evidence of a recent complaint in a sexual offence. Moreover where the prosecution are relying, as potential corroborative material, upon lies alleged to have been told by the accused, a particularly careful direction is needed. A special direction is also often needed where evidence of complainant's distress is relied upon by the prosecution in sexual cases as potentially corroborative material. The trial judge has further the additional obligation of directing the jury that accomplices, who are parties to the same charge, cannot corroborate each other.The Court of Appeal was, in my judgment, fully entitled to conclude that had the court in Reg. v. Bagshaw [1984] 1 W.L.R. 477 had the benefit of the full argument which they had had in these two appeals, and, in particular, had their attention drawn to Reg. v. Beck [1982] 1 W.L.R. 461, a different conclusion might have been reached. They accordingly concluded they were not bound by the decision in Bagshaw's case. I consider that they were entitled so to decide.In the submissions before your Lordships' House, there has been little, if any dispute, as to the relevant law. Counsel for the appellants has fully accepted the decision of the Court of Appeal both in Bagshaw's case and in these two appeals, that patients in hospital under the Mental Health Acts are not in a category like accomplices or complainants in sexual cases or young children. To create from them such a new category would clearly

- 10 -

Involve considerable problems of definition. What sort of patients, and patients with what sort of criminal records are to be included? The submission of the appellants, In essence, Is that without use of the word "danger", in any case analogous to those of the three established categories, and where the evidence of the only or principal witness relied upon by the prosecution is inherently unreliable, such a warning must be inadequate. I cannot agree. It has been said both in the Court of Appeal and in your Lordships' House, that the obligation to warn a jury does not involve some legalistic ritual to be automatically recited by the judge, or that some particular form of words or incantation has to be used and, if not used, the summing up is faulty and the conviction must be quashed. (See Reg. v. Russell (1968) 52 Cr.App.R. 147, 150 per Diplock L.J.). There is no magic formula which has to be used with regard to any warning which is given to juries (Reg. v. Price (Herbert) [1969] 1 Q.B. 541, 546 per Sachs L.J.). As this is no mere idle process it follows that there are no set words which must be adopted to express the warning. Rather must the good sense of the matter be expounded with clarity and in the setting of a particular case. (Reg. v. Hester [1973] A.C. 296, 309 per Lord Morris of Borth-y-Gest). The summing up should be tailored to suit the circumstances of the particular case. (Reg. v. Kilbourne [1973] A.C. 729, 741 per Lord Hailsham of St. Marylebone L.C.)To my mind the question raised by these appeals is both simple to define and simple to answer. Given that it is common ground that a warning was required as to the way in which the jury should treat the evidence of the complainants, the question is: was that warning: sufficient? Did it in clear terms bring home to the jury the danger of basing a conviction on the unconfirmed evidence of the complainants?In the three established categories where the "full warning" is obligatory, the inherent unreliability of the witness may well not be apparent to the jury. Hence the phrase often used in a summing up - it is the experience of the courts accumulated over many years etc. etc. Complainants of sexual assaults do on occasions give false evidence for a variety of reasons, some of which may not have occurred to a jury. Accomplices may have hidden reasons for lying, and this possibility may again not be apparent to a jury. Children who, although old enough to understand the nature of an oath and thus competent to give sworn evidence, may yet be so young that their comprehension of events and of questions put to them, or their own powers of expression, may be imperfect. All this needs properly to be spelt out to the jury. Hence the well established rule of practice.In other cases the potential unreliability of the sole or principal witness for the prosecution is obvious for all to see. These were such cases. The complainants were men of bad character. They had been sent to Rampton rather than to an ordinary prison, because they were mentally unbalanced. That they were anti-authoritarian, prone to lie or exaggerate, and could well have old scores which they were seeking to pay off, was not disputed. Notwithstanding that the possibility of their evidence being unreliable was patent, that it was clearly dangerous to prefer their evidence to that of the defendants, all men of good character on whose behalf witnesses had spoken in glowing terms,

- 11 -

the judge nevertheless told the jury In the clearest possible terms and repeated himself, that they must approach the evidence of the complainants with great caution. It is common ground that having given that warning, he then identified the very dangers which justified the exercise of great caution. He gave three reasons. Firstly, they were ail persons of bad character; secondly, they were all persons suffering from some form of mental disorder, and thirdly, they may have all conspired together to make false allegations. Thus the judge warned the jury of the dangers of relying on the complainants' testimony because, for the reasons which he gave, such testimony could well be unreliable. The judge, however, did not leave the matter there. As previously stated he pointed out, when dealing with each count, the details of the background of the complainant, his past criminal record, the nature of his mental disturbance and his history in the hospital, and perhaps most important of all, the hospital psychiatrist's view of the personality defects from which the patient suffered and of which I have already given a typical example. I agree with the Court of Appeal that he gave the emphatic warning which was required to meet the justice of the case. Indeed had this been one of the category of cases which required the "full warning" then the judge's direction would have been fully adequate.The certified point of law is in these terms:
"In a case where the evidence for the Crown is solely that of a witness who is not in one of the accepted categories of suspect witnesses, but who, by reason of his particular mental condition and criminal connection, fulfilled the same criteria, must the judge warn the jury that it is dangerous to convict on his uncorroborated evidence."
I would amend the question by substituting for the words "the same criteria" "analogous criteria." I would then answer the question in the affirmative, adding, for the sake of clarity, that while it may often be convenient to use the words "danger" or "dangerous," the use of such words is not essential to an adequate warning, so long as the jury are made fully aware of the dangers of convicting on such evidence. Again, for the sake of clarity I would further add that Reg. v. Beck [1982] 1 W.L.R. 461 was rightly decided and that in a case which does not fall into the three established categories and where there exists potential corroborative material, the extent to which the trial judge should make reference to that material depends upon the facts of each case. The overriding rule is that he must put the defence fairly and adequately.On 18 December 1984 your Lordships' House gave the appellants, Spencer, Ball, and Mason, leave to appeal against the decision of the Court of Appeal (Criminal Division) on an additional ground, namely "on the question whether incidents relating to the conduct of a member of the jury both before and after the juror was discharged were of such serious proportions as to render their convictions unsafe or unsatisfactory." The facts were these. During this trial it became clear both to the judge and counsel that one male juror, a Mr. Peet, had early formed a definite view of the case, which was hostile to the defendants, and on occasions clearly showed that he was biased against the defendants and in favour of the prosecution. For example, he

- 12 -

showed his impatience at having to listen to the cross-examination of one or more of the complainants. Both the judge and counsel were concerned at this situation, and after consultation, decided to take no action.However, a member of the jury informed a court usher that Mr. Peet's wife worked at the Eastdale Unit of Balderton Hospital, another mental hospital which figured in the evidence in the trial. It was an establishment to which some patients went as a half-way house before their ultimate discharge. Mr. Peet's wife apparently worked there as a cleaner. The judge was given this information during the short adjournment on the day before the trial ended and after the judge had begun his summing-up. The trial was then on its tenth day. The judge raised the matter with counsel in the absence of the jury. He informed them that he would have the jury back, seek from Mr. Peet confirmation of the facts which had been communicated to him, and if they were accurate, discharge him. He observed to counsel:
"It may well be that in the course of conversation with his wife, he has heard things which it would be better if he had not. I have considered whether in the circumstances it would be necessary to discharge the whole of the jury but I do not feel that It is such a case myself."
When the jury returned to court, Mr. Peet confirmed, in answer to the judge's question that his wife did in fact work at the Eastdale Unit. Mr. Peet asked the judge whether "it would be all right to wait in the ante-room "because I have got three to take back to Newark" - a half-an-hour journey from the court. The judge agreed and Mr. Peet left court. To the remaining eleven jurors the judge observed:
"Members of the jury, you realise of course, I am sure, how very careful we have to be. Who he is giving a lift back to Newark, I know not. If it is any of you I must enjoin you, please do not discuss this case with Mr. Peet any further."
By next morning defence counsel had had an opportunity of thinking rather more fully about what Mr. Peet had said the previous day, and had also learned that Mr. Peet had been in the habit of giving three of his fellow jurors a daily lift to and from their homes in Newark, which was near where he himself lived. In those circumstances the probabilities were that Mr. Peet would have aired his anti-defence opinions during those car journeys and possibly given his passengers other information prejudicial to the three defendants. Application was therefore made to the judge for the discharge of the remaining eleven jurors. The application was resisted by the prosecution. It was pointed out that if there were to be a retrial the witnesses would have to be recalled. Concern had already been expressed by doctors and others about the effect that the giving of evidence and the experience of being cross- examined had had upon the Rampton patients involved. Mr. Roberts submitted that the test which the judge should apply was that the jury should not be discharged unless it could be shown that there was "a very high risk" that the apparently biased juror had influenced any of his fellow jurors.

- 13 -

Mr. Roberts accepted, both in the Court of Appeal and before us, that he had expressed the test in terms which were too strict. The correct test is the one stated in Reg. v. Sawyer (1980) 71 Cr.App.R. 283, 285 namely, whether there was a "real danger" that the appellants' position had been prejudiced in the circumstances which I have outlined. In the course of giving his ruling,

the judge said:

"This is an anxious matter. I bear in mind that this case is now in its tenth day. I also bear in mind that so far as these defendants are concerned, through no fault of their own I hasten to add, through no fault of the prosecution, years have passed since they were first seen about these matters. I bear in mind also very much the interests of the patients who come to this court and of necessity are cross- examined about what no doubt to them, whatever be the truth of this case, are painful matters and which must have caused concern to those who have the medical care of them."
In the exercise of his discretion he refused to discharge the remaining jurors.Because of the manner in which Mr. Peet had conducted himself in the jury box, it was very properly decided by the Director of Public Prosecutions, following the conviction of the appellants, that a statement should be taken from Mr. Peet as to his involvement with the other three jurymen, in particular what happened following his discharge. In a. written statement, which was put before the Court of Appeal, Mr. Peet confirmed that after his discharge he had given three members of the jury a lift back to Newark, and that "during the journey we discussed the case and also how it came to light about my wife working at Balderton Hospital." Mr. Peet returned the next day to Nottingham, to take his wife and son shopping. Having left them, he went to the court, arriving about 12.45 p.m. However, the summing up had been concluded and the jury had retired at 10.35 a.m. He was told by a representative from Rampton that the jury were unlikely to be back before half-past-two, and in the course of the conversation he expressed the opinion that the defendants would be found guilty. He returned to the court about quarter-to- two, spoke to a lady who appears to have been a solicitor's clerk, and told her that he had been a bit upset at having been discharged from the jury. He told her that in his view the defendants would be found guilty. He stayed waiting in court until about 4.00 p.m. but was told that the verdict would not be until 5.00 p.m. He went back to his car because he had told his son that he would be in the car park at 4.30 p.m. He then returned to the court at 5.00 p.m. and waited until the jury gave their verdict at 5.17 p.m.The Court of Appeal having read Mr. Peet's statement quoted the description of the discussion during the journey back to Newark with the three members of the jury following Mr. Peet's discharge. May L.J. giving the judgment of the court then made this comment at [1985] 1 Q.B. 771, 788:
"Other parts of this statement made it quite clear that Mr. Peet had formed a clear view about the guilt of the

- 14 -

appellants but nothing to suggest that he had expressed this view to the three juror passengers in his car when he was giving them a lift home on that last occasion. Nevertheless, on all the material that there is before us we think we must conclude that In the course of that journey he did do just that."It is common ground that the judge applied the wrong test for the exercise of his discretion. Moreover he did not have the information as to what in fact had happened on the journey back to Newark which the Court of Appeal concluded, and in my judgment rightly concluded, involved the discussion of the case despite the judge's injunction to the contrary. It fell to the Court of Appeal, as it does to your Lordships, to ask themselves whether in ail the circumstances they were satisfied that the verdict was safe and satisfactory. The Court of Appeal were so satisfied, the substance of their judgment on this point being expressed in these terms at p.790:
"The three jurors to whom Mr. Peet gave the final lift home had been specifically warned by the judge immediately after Mr. Peet had been discharged that they should not talk to him about the case, just as no doubt they had been warned on earlier occasions. It is quite true that these jurors did not comply with that warning and direction from the judge and their failure to do so is to be deplored. Nevertheless we do not think that in the circumstances there is any realistic chance that those three jurors, or any of them, carried with them into the deliberations of themselves and their fellow jurors, when the judge finished his summing up the following morning, any prejudice or bias which they had acquired from Mr. Peet the afternoon before. We do not forget that the convictions in this case were by majority verdicts reached after the jury had been out considering their verdicts for over 5 1/2 hours. We think this is readily explicable by the nature of the evidence which the jury had heard, and by the type of witness by whom that evidence had been given, and about whom and their evidence the judge had so fully warned the jury in the course of his summing up."
It may well be that Mr. Peet, who apparently during the greater part of the trial had been expressing to his fellow jurors and in particular the three jurors whom he drove to and from court his firm view that the defendants were all guilty, did not in the conversation in his car after his discharge, add any further strength to the opinion which he had been expressing. However, in the circumstances of this case I find myself unable totally to discount the likelihood of such a possibility. Mr. Peet was clearly very concerned that the defendants should be convicted. As the trial judge appreciated, there was the real possibility that Mr. Peet in the course of his conversations with his wife may have heard things which it would be better if he had not. Mr. Peet was clearly frustrated by being discharged from the jury, despite the very courteous terms in which this was done by the judge. Knowing that he would no longer be able to influence the decision of the jury, he could well have used the journey home as the final opportunity to achieve the outcome of the trial which he thought was right. He may have done this in imparting information which

- 15 -

he had intended to reserve until the conclusion of the summing- up and after the retirement of the jury when all would be present or he might have only re-stated his views but with added emphasis. What the three jurors were discussing, or being subjected to, was the assertion by a biased, but an ex-fellow juror, of the necessity to convict.I therefore have a lurking doubt, that justice may not have been done, which makes me conclude that the verdict was unsafe. With some reluctance I find myself unable to agree with the Court of Appeal on this particular aspect of the first of the appeals. Accordingly, I would allow the appeal of Spencer, Ball, and Mason, and dismiss the appeals of Smails, Ball, and White.

- 16 -