O'Brien v Chief Constable of South Wales Police [2005] UKHL 26 (28 April 2005)

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom House of Lords Decisions

You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> O'Brien v Chief Constable of South Wales Police [2005] UKHL 26 (28 April 2005)
URL: http://www.bailii.org/uk/cases/UKHL/2005/26.html
Cite as: [2005] WLR 1038, [2005] 2 WLR 1038, [2005] 2 All ER 931, [2005] UKHL 26, [2005] 1 WLR 1038, [2005] 2 AC 534

[New search] [Buy ICLR report: [2005] 2 AC 534] [Buy ICLR report: [2005] 2 WLR 1038] [Help]

Judgments - O'Brien (Respondent) v Chief Constable of South Wales Police (Appellant)

HOUSE OF LORDS

SESSION 2004-05
[2005] UKHL 26
on appeal from: [2003] EWCA Civ 1085

OPINIONS

OF THE LORDS OF APPEAL

FOR JUDGMENT IN THE CAUSE

O'Brien (Respondent)

Chief Constable of South Wales Police (Appellant)

THURSDAY 28 APRIL 2005

The Appellate Committee comprised:

Lord Bingham of Cornhill

Lord Steyn

Lord Phillips of Worth Matravers

Lord Rodger of Earlsferry

Lord Carswell

HOUSE OF LORDS

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT

IN THE CAUSE

O'Brien (Respondent) v. Chief Constable of South Wales Police (Appellant)

[2005] UKHL 26

LORD BINGHAM OF CORNHILL

My Lords,

I have had the advantage of reading in draft the opinions of my noble and learned friends Lord Phillips of Worth Matravers and Lord Carswell, with which I am in complete agreement. For the reasons they give, I also would dismiss this appeal.

As the number of reported cases on the topic makes clear, similar fact evidence has proved a contentious and uncertain area of the law, particularly in criminal cases but also in civil cases like that before the House. But such evidence may be very important, even decisive. It is undesirable that the subject should be shrouded in mystery.

Any evidence, to be admissible, must be relevant. Contested trials last long enough as it is without spending time on evidence which is irrelevant and cannot affect the outcome. Relevance must, and can only, be judged by reference to the issue which the court (whether judge or jury) is called upon to decide. As Lord Simon of Glaisdale observed in Director of Public Prosecutions v Kilbourne [1973] AC 729, 756, "Evidence is relevant if it is logically probative or disprobative of some matter which requires proof ….. relevant (ie. logically probative or disprobative) evidence is evidence which makes the matter which requires proof more or less probable".

That evidence of what happened on an earlier occasion may make the occurrence of what happened on the occasion in question more or less probable can scarcely be denied. If an accident investigator, an insurance assessor, a doctor or a consulting engineer were called in to ascertain the cause of a disputed recent event, any of them would, as a matter of course, enquire into the background history so far as it appeared to be relevant. And if those engaged in the recent event had in the past been involved in events of an apparently similar character, attention would be paid to those earlier events as perhaps throwing light on and helping to explain the event which is the subject of the current enquiry. To regard evidence of such earlier events as potentially probative is a process of thought which an entirely rational, objective and fair-minded person might, depending on the facts, follow. If such a person would, or might, attach importance to evidence such as this, it would require good reasons to deny a judicial decision-maker the opportunity to consider it. For while there is a need for some special rules to protect the integrity of judicial decision-making on matters of fact, such as the burden and standard of proof, it is on the whole undesirable that the process of judicial decision-making on issues of fact should diverge more than it need from the process followed by rational, objective and fair-minded people called upon to decide questions of fact in other contexts where reaching the right answer matters. Thus in a civil case such as this the question of admissibility turns, and turns only, on whether the evidence which it is sought to adduce, assuming it (provisionally) to be true, is in Lord Simon's sense probative. If so, the evidence is legally admissible. That is the first stage of the enquiry.

The second stage of the enquiry requires the case management judge or the trial judge to make what will often be a very difficult and sometimes a finely balanced judgment: whether evidence or some of it (and if so which parts of it), which ex hypothesi is legally admissible, should be admitted. For the party seeking admission, the argument will always be that justice requires the evidence to be admitted; if it is excluded, a wrong result may be reached. In some cases, as in the present, the argument will be fortified by reference to wider considerations: the public interest in exposing official misfeasance and protecting the integrity of the criminal trial process; vindication of reputation; the public righting of public wrongs. These are important considerations to which weight must be given. But even without them, the importance of doing justice in the particular case is a factor the judge will always respect. The strength of the argument for admitting the evidence will always depend primarily on the judge's assessment of the potential significance of the evidence, assuming it to be true, in the context of the case as a whole.

While the argument against admitting evidence found to be legally admissible will necessarily depend on the particular case, some objections are likely to recur. First, it is likely to be said that admission of the evidence will distort the trial and distract the attention of the decision-maker by focusing attention on issues collateral to the issue to be decided. This is an argument which has long exercised the courts (see Metropolitan Asylum District Managers v Hill (1882) 47 LT 29, 31 per Lord O'Hagan) and it is often a potent argument, particularly where trial is by jury. Secondly, and again particularly when the trial is by jury, it will be necessary to weigh the potential probative value of the evidence against its potential for causing unfair prejudice: unless the former is judged to outweigh the latter by a considerable margin, the evidence is likely to be excluded. Thirdly, stress will be laid on the burden which admission would lay on the resisting party: the burden in time, cost and personnel resources, very considerable in a case such as this, of giving disclosure; the lengthening of the trial, with the increased cost and stress inevitably involved; the potential prejudice to witnesses called upon to recall matters long closed, or thought to be closed; the loss of documentation; the fading of recollections. It is, I think, recognition of these problems which has prompted courts in the past to resist the admission of such evidence, sometimes (as, perhaps, in R v Boardman [1975] AC 421) propounding somewhat unprincipled tests for its admission. But the present case vividly illustrates how real these burdens may be. In deciding whether evidence in a given case should be admitted the judge's overriding purpose will be to promote the ends of justice. But the judge must always bear in mind that justice requires not only that the right answer be given but also that it be achieved by a trial process which is fair to all parties.

His Honour Judge Graham Jones and the Court of Appeal were in my opinion right to regard the evidence which Mr O'Brien seeks to adduce as potentially probative, and so admissible. Mr O'Brien contends that, in the course of investigating the murder of Mr Saunders and prosecuting him and his co-defendants for that murder, named officers for whom the Chief Constable is responsible resorted to specific methods which were oppressive, dishonest and unprofessional. Accusations of such gravity must be clearly proved, and proof could never be easy. The primary evidence must relate to how Mr O'Brien and his co-defendants were treated. But if he were able to show that these same officers had, in the earlier cases of Griffiths and Ali, resorted to the same or similar methods in order to try and obtain admissions and convictions, his hand would be significantly strengthened: put technically, the matter which requires proof would be more probable.

The judge reviewed the evidence and considered that it should be admitted. In the absence of misdirection or demonstrable error that is not a judgment with which an appellate court should interfere, and the Court of Appeal was right not to do so save in a very limited way. Both courts, in my respectful view, reviewed the competing arguments in a careful and judicious way, and I do not think their conclusions can be faulted. I would add only this: that while, for purposes of pleading and disclosure, it was desirable and perhaps necessary to obtain a proleptic ruling in principle on the admission of this evidence, the final say, in relation to any particular item of evidence, should rest with the trial judge. That judge will recognise the need to be loyal to the ruling already made, but the ultimate responsibility for ensuring a fair trial and a just outcome rests with him. The trial judge cannot be deprived of all discretion, although the discretion should be exercised consistently with the ruling made by Judge Graham Jones and approved on appeal.

LORD STEYN

My Lords,

I have had the privilege of reading the opinions of my noble and learned friends Lord Bingham of Cornhill, Lord Phillips of Worth Matravers and Lord Carswell . For the reasons they have given I would also dismiss the appeal.

LORD PHILLIPS OF WORTH MATRAVERS

My Lords,

Introduction

This appeal requires your Lordships to consider the circumstances in which evidence of 'similar facts' can be admitted in a civil suit. Under English common law, in both criminal and civil proceedings that were not disposed of in summary fashion, the functions of the trial used to be shared between judge and jury. The judge was responsible for resolving issues of law, the jury issues of fact. This division of functions enabled the judge to control the evidence that was placed before the jury. The principles under which he exercised what must initially have been a discretion in relation to this control became over time, by the process of the common law, recognised as rules of law governing 'admissibility', a concept that I believe to be foreign to most civil law jurisdictions. In time Parliament intervened to codify, and sometimes to vary, the principles of admissibility applied by the judges.

For obvious reasons evidence has never been admissible if it has not been relevant to the issues arising in the proceedings. Rules of admissibility govern the circumstances in which evidence which is relevant is not admitted. Two policy considerations underlie the rules of admissibility with which this appeal is concerned. First, evidence should not be admitted if it is likely to give rise to irrational prejudice which outweighs the probative effect that the evidence has in logic. This consideration of policy carries particular weight where the tribunal is a jury, whose members are not experienced as are judges in putting aside irrational prejudice. Secondly, evidence should not be admitted if its probative weight is insufficient to justify the complexity that it will add to the trial. That is a consideration of general application.

The evidence whose admissibility is in issue on this appeal is known as 'similar facts' evidence. Issues in relation to such evidence normally arise in criminal rather than civil proceedings. Where a defendant to a criminal charge has a criminal record, his propensity to commit crime will normally have some relevance to the question of whether he committed the offence with which he is charged. As a general rule such evidence has none the less been held to be inadmissible on the ground that its prejudicial effect is likely to outweigh its probative value. Exceptions have, however, been made to this general exclusion. The nature and extent of those exceptions have proved a frequent preoccupation of the appellate courts and, on at least four occasions, of your Lordships' House. They are now to be found codified in sections 101 to 106 of the Criminal Justice Act 2003 ('the 2003 Act'), which were brought into effect in December last year.

Not infrequently a defendant in a criminal trial makes allegations of misconduct on the part of police witnesses in relation to the circumstances in which alleged admissions have been obtained, or the accuracy of those admissions. The defendant may wish to adduce 'similar fact' evidence designed to show that the police witness has been guilty of similar misconduct in the past. Once again the courts have restricted the circumstances in which such evidence can be adduced by a defendant. Those restrictions are now codified in section 100 of the 2003 Act.

On this appeal the issue of the admissibility of similar fact evidence arises in the context, not of a criminal prosecution, but of a civil suit. The claimant, Michael O'Brien, is suing the Chief Constable of South Wales Police for misfeasance in public office and malicious prosecution. In 1988 Mr O'Brien was convicted of murder. He served 11 years of his life sentence. His case was then referred to the Court of Appeal by the Criminal Cases Review Commission and his appeal was allowed. Mr O'Brien alleges that he was, in common parlance, 'framed' by the police for a murder which he never committed. He identifies as primarily responsible a Detective Inspector Lewis but alleges that his superior officer, Detective Chief Superintendent Carsley, gave express or tacit approval to at least some aspects of the misconduct alleged against DI Lewis. Mr O'Brien gave notice of his intention to adduce evidence designed to demonstrate that DCS Carsley had behaved with similar impropriety on one other occasion and that DI Lewis had done so on two other occasions. At the Case Management Conference the admissibility of this evidence was challenged. His Honour Judge Graham Jones ruled that it was admissible, but that he had a discretion to exclude it as a matter of case management. He declined to exclude it, with the exception of one incident alleged against DI Lewis. On appeal the Court of Appeal held that the judge's ruling on admissibility had been correct, as had the exercise of his discretion in not excluding the evidence. The court held, however, that he should also have admitted the evidence of the incident that the judge had excluded.

The case advanced on behalf of the Chief Constable is that there is a rule of law which prevents the admission of similar fact evidence in a civil trial unless it has an enhanced probative value. Whether there is such a rule and, if so, the degree to which the probative value must be enhanced before the evidence becomes admissible are the issues of law raised by this appeal.

The background facts

On the night of 12 October 1987 Mr Philip Saunders, the owner of three newspaper kiosks in the centre of Cardiff, was attacked and robbed on his way home. His skull was fractured as a result of severe blows to the head, probably from a spade. He died of his injuries five days later. In July 1988 three men were convicted of Mr Saunders' murder at Cardiff Crown Court. They were Darren Hall, Ellis Sherwood and his brother-in-law, Michael O'Brien. Mr Hall and Mr Sherwood were then aged 19 and Mr O'Brien aged 20. All three defendants gave evidence at the trial. Mr Hall had tendered a plea to manslaughter, which had not been accepted. He gave evidence that he had acted as look-out while Mr Sherwood and Mr O'Brien attacked and robbed Mr Saunders. Mr Sherwood and Mr O'Brien admitted that they had been in Mr Hall's company on the night of the murder, but denied that any of the three had been involved in the murder or present when it occurred. They said that they had been looking for a car to steal. The three defendants were sentenced to custody for life.

Applications by the defendants for permission to appeal against conviction were refused. 11 years later, however, the Criminal Cases Review Commission referred their case to the Court of Appeal. Mr Hall now contended that the evidence that he had given at the trial was untrue. The appeal was allowed on 25 January 2000 and the convictions were quashed. The terms of the judgment delivered by Roch LJ called in question the propriety of the conduct of DI Lewis, who had led the investigation into Mr Saunders' murder on a day-to-day basis.

Mr O'Brien applied to the Home Secretary for compensation for the 11 years that he had spent in prison. He received an award under section 133 of the Criminal Justice Act 1988 of approximately £670,000. That amount has been the subject of appeal to the Administrative Court and the Court of Appeal, but his entitlement to a substantial award on the ground that he has been the subject of a miscarriage of justice is not in issue. He is, however, not satisfied with the compensation that he will receive. It is his case that the police acted with deliberate and flagrant impropriety in framing him and his co-defendants for Mr Saunders' murder. The object of his current claim is to secure aggravated and exemplary damages which will augment the compensation to which he is entitled.

The misconduct alleged

The misconduct alleged by Mr O'Brien was summarised by Brooke LJ in the Court of Appeal as follows:

… It is said that Detective [Chief] Superintendent Carsley gave express or tacit approval to the malpractice relied on. "

Similar facts in R v Griffiths

The first batch of similar fact evidence on which Mr O'Brien wishes to rely relates to the conduct of the same police force in relation to an investigation in July 1982 into a series of explosions in Cardiff and elsewhere. This led, in the autumn of 1983 to what became known as 'the Welsh Bomb Trial' at Cardiff Crown Court. At the end of that trial Mr Griffiths and three other defendants were acquitted. Mr O'Brien seeks to adduce evidence that the police, and in particular DI Lewis, were guilty of impropriety towards the four defendants who were acquitted that was similar to that which he, Mr O'Brien, and his co-defendants experienced, including the bullying of witnesses and the fabrication of statements. It is Mr O'Brien's case that DCS Carsley was complicit in the impropriety that occurred and, in particular, that he supplied a false alibi for DI Lewis for the time that he was engaged in sustained off the record questioning of one of Mr Griffiths' co-defendants. Among the witnesses that he intends to call will be Mr Griffiths himself.

Similar facts in R v Ali

The second batch of similar fact evidence on which Mr O'Brien wishes to rely relates to the investigation carried out by the same police force in February 1990 into the murder in Cardiff of Karen Price. This led to the trial for murder of Idris Ali and Alan Charlton at Cardiff in 1990. Mr Ali was convicted of murder, but that conviction was quashed by the Court of Appeal in November 1994. A fresh indictment was preferred and Mr Ali was subsequently convicted of manslaughter. The basis of this conviction is not known.

Mr Ali was a 24 year old of limited intellectual capacity. It is Mr O'Brien's case that the police improperly subjected Mr Ali to an oppressive series of interviews throughout the whole of one night and that they either obtained by oppression, or fabricated, confessions that were false. He contends that Mr Lewis, who was the investigating officer and the officer in the case, was in the police station at the material time and must have been complicit in what was taking place.

Counsel for each of the parties spent a considerable time exploring before the House the cogency of the similar fact evidence and the extent to which this provided support for Mr O'Brien's case as to the manner in which he and his co-defendants were treated. This exercise could only be material should your Lordships be persuaded that the Court of Appeal applied too lenient a test of admissibility and that it was appropriate for this House to apply the correct test to the material facts. In so far as the result in the Court of Appeal turned on a review of the exercise of discretion by the trial judge, I do not believe that your Lordships would be minded to conduct a further review.

The test applied by the Court of Appeal and the challenge to this

The Court of Appeal held that, in civil as opposed to criminal proceedings, the judge has to proceed in two stages when deciding whether to admit evidence. First he has to decide whether the evidence is admissible. If it is, he has to decide, as a matter of discretion whether he will permit the evidence to be led. The test of admissibility is that propounded by your Lordships' House in Director of Public Prosecutions v P [1991] 2 AC 447. The exercise of discretion as to whether admissible evidence should be permitted to be led involves the approach that the judge should bring to case management in accordance with the Civil Procedure Rules (CPR).

The Court of Appeal thus applied to the test of admissibility in civil proceedings the approach adopted in criminal proceedings. Mr Freeland QC for the appellant also relied extensively on jurisprudence in the criminal field. His primary submission was largely founded, however, on the decision of this House in a civil appeal, Metropolitan Asylum District Managers v Hill (1882) 47 LT 29. That decision, he submitted, supported the proposition that similar fact evidence was not admissible unless likely to be 'reasonably conclusive' of a primary issue in the proceedings. Mr Freeland's alternative case was that, to be admissible, similar fact evidence had to have enhanced relevance, so as to be of substantial probative value. This was the test for admission of evidence of bad character by the defence in a criminal trial recommended by the Law Commission in their Report on Evidence of Bad Character in Criminal Proceedings (Law Com No 273) of October 2001 (Cm 5257).

The criminal cases

The criminal courts have always applied a general rule that evidence that an accused has committed criminal offences other than those with which he is charged is inadmissible. This rule was explained by counsel for the appellants in Makin v Attorney General for New South Wales [1894] AC 57 as being justified, not because the evidence was wholly irrelevant, but because it was 'inconvenient and dangerous'. Such evidence would tend 'both to confuse and unduly to prejudice the jury'. In that case the appellants were convicted of murdering an infant that they had received for adoption. They complained, unsuccessfully, that evidence had been admitted of the discovery, buried in their garden, of the bodies of five other children. In explaining why this evidence was admissible, Lord Herschell LC said, at p 65:

For nearly 100 years this passage was repeatedly cited by courts to justify the admission of similar fact evidence, notwithstanding that, as pointed out by Mr L H Hoffmann, the author of Similar Facts after Boardman (1975) 91 LQR 193 at p 200, "the Makin rule affords no real guidance on whether similar fact evidence is admissible or not".

In reality, judges tended to let in similar fact evidence, and appeal courts to approve their doing so, when, as a matter of logic, the evidence pointed cogently to the conclusion that the accused had committed the offence with which he was charged. In R v Boardman [1975] AC 421 your Lordships' House tried once again to explain the principles governing the admission of evidence of past conduct by an accused similar to the conduct with which he was charged - in that case, acts of buggery by a schoolmaster with boys in which the accused was the passive partner. Their Lordships expressed the test in different ways: "a really material bearing on the issues to be decided" per Lord Morris of Borth-y-Gest at p 439; "a strong degree of probative force" based on the "striking similarity" of the material facts per Lord Wilberforce at p 444; "such an underlying unity between the offences as to make coincidence an affront to common sense" per Lord Hailsham of St Marylebone at p 453, quoting Lord Simon of Glaisdale in R v Kilbourne [1973] AC 729, 759; "evidence which would point so strongly to his guilt that only an ultra-cautious jury, if they accepted it as true, would acquit in face of it" per Lord Cross of Chelsea at p 457; "the similarity would have to be so unique or striking that common sense makes it inexplicable on the basis of coincidence" per Lord Salmon at p 462.

Mr Hoffmann, in his article, saw this decision as 'a great advance' in that it recognised that the admissibility of similar fact evidence depended 'simply and solely upon its probative strength'. It is clear, however, that the probative strength had to be substantial, before the evidence became admissible. If it was not it would be excluded because its prejudicial effect on the accused risked being greater than its relevance justified.

Some of the language in Boardman suggested that the probative strength of similar fact evidence had to be very high indeed before it could be admitted. This led, in some instances, to the exclusion of evidence that was of considerable probative weight. An example of this was the decision of the Court of Appeal in DPP v P (1990) 93 Cr App R 267. The defendant was charged with a number of specimen counts of rape and incest against each of his two daughters. The trial judge refused an application that there should be separate trials in respect of the offences alleged against each daughter. The defendant was convicted. The Court of Appeal allowed his appeal and quashed the conviction on the ground that the judge had erred in refusing separate trials. Lord Lane CJ held that the court had looked in vain for features of similarity that was striking or that went beyond "the incestuous father's 'stock in trade'" that were considered necessary if the evidence of offences against one daughter was to be admissible in relation to the offences alleged against the other.

The prosecution appealed to this House. The certified question of law asked whether there had to be 'striking similarities' before similar fact evidence could be admitted in a case of alleged sexual abuse by a father of a daughter. Lord Mackay of Clashfern LC, who gave a speech with which the other members of the House agreed, advanced principles which, thereafter, were rightly treated by courts as being of general application. After considering at length the speeches in Boardman he propounded at p 460 a simple test of admissibility:

Reg v Boardman

Lord Mackay went on to say that while such probative force may be derived from the striking similarity of the similar fact evidence this was not a precondition of admissibility, pp 460-461:

The test of admissibility advanced by Lord Mackay in DPP v P still requires similar fact evidence to have an enhanced relevance or substantial probative value before it is admissible against a defendant in a criminal trial. This is because such evidence usually shows that the defendant is a person of bad character and thus risks prejudicing a jury against the defendant in a manner that English law regards as unfair. Instead of applying Lord Mackay's simple test, a trial judge now has to apply his mind to the matters set out in sections 101 to 106 of the 2003 Act. These preserve, however, by rules of some complexity, the requirement that the similar fact evidence should have an enhanced probative value.

It is not obvious that the test in DPP v P was one that it was appropriate for the Court of Appeal to apply in this case. Here there is no defendant at risk of conviction of a criminal offence rendering him liable to imprisonment. The claimant was so convicted and he wishes to use similar fact evidence against the police who he alleges improperly procured his conviction. There is a much closer parallel between this situation and that where, in the course of a criminal trial, a defendant seeks to establish that evidence given against him by police witnesses is untruthful. The test of admissibility in such a case received consideration by the Court of Appeal in R v Edwards [1991] 1 WLR 207.

In Edwards the appellant had been convicted of robbery while in possession of a firearm and sentenced to 14 years imprisonment. The evidence against him included police evidence that he had made confessions in interview. The appellant had challenged the veracity of the interview notes, alleging that the police officers concerned had 'fitted him up'. After the trial those representing the appellant discovered that one of the senior officers concerned had, two months before the trial, been reprimanded for certifying interview notes in another trial when these, to his knowledge, had been wrongly rewritten. The fact that this officer was facing disciplinary proceedings should have been disclosed to the defence. In the judgment of the court allowing the appeal Lord Lane CJ considered at length the use to which the defendant could have put of evidence of the police officer's previous misconduct, had he been aware of this.

Lord Lane started his consideration of the law at p 214 with the following proposition:

After citation of a case dealing with cross-examination as to credit Lord Lane continued, at p 219:

Harris v Tippett

Lord Lane then referred to exceptions to the rule that evidence to contradict answers on cross-examination as to credit will not be permitted. One such exception was that evidence could be introduced to show that a witness was "biased or partial in relation to the parties or the cause". Lord Lane then considered the existence of a further possible exception: evidence to show "that the police are prepared to go to improper lengths to secure a conviction". He held that such evidence could only be introduced if relevant to an allegation of bias.

Lord Lane remarked that the appeal before the court raised two problems, first, what questions could properly be asked in cross-examination, and secondly, whether evidence to contradict the answers would have been admissible. He went on to deal with the first question, drawing a line between complaints of misconduct, upon which there had not been adjudication, and disciplinary charges found proved. Questions in cross-examination could be asked about the latter, but not about the former. Lord Lane then turned to the question, particularly relevant in the present context, of evidence that suggested that the testimony of a police witness appeared to have been disbelieved in a previous trial. After consideration of authority, Lord Lane summarised the position as follows, at p 217:

After referring to two previous cases that exemplified the above proposition, Lord Lane turned to the question of whether, if the facts put in cross-examination were denied, evidence could be led to counter such denial. At p 220 he concluded that it could not:

This conclusion reflects, I believe, "the necessity of keeping the criminal process within proper bounds and avoiding the pursuit of side issues which are only of marginal relevance to the jury's decision", to which Lord Lane had earlier referred at p 219. It is not, however satisfactory. Evidence which indicates that a police officer has fabricated admissions in a previous case is not evidence 'as to credit alone', if it is alleged that the same officer has fabricated evidence in a subsequent case. The position is now governed by section 100 of the 2003 Act which renders admissible, with the leave of the court, evidence of the bad character of a person other than the defendant if, and only if, it has substantial probative value in relation to a matter which is in issue in the proceedings and is of substantial importance in the context of the case as a whole.

The Court of Appeal in the present case dismissed an argument by Mr Freeland that the judge should have applied the approach laid down in Edwards on the ground that

I have reservations about this conclusion. The issue in both Edwards and the present case is the admissibility of evidence to prove police misconduct. In each case, there is a need to exert disciplinary control to avoid unbalancing the proceedings by the adducing of evidence of only marginal relevance. This appeal raises the question of whether a rule of admissibility, such as that now to be found in section 100 of the 2003 Act, should apply in civil proceedings. It is time to turn to the authorities that deal with similar fact evidence in the civil context.

The civil cases

Metropolitan Asylum District Managers v Hill 47 LT 29 involved an action for nuisance brought by the owners of land adjacent to a smallpox hospital in Hampstead against the managers of the hospital. Their case appears to have included an averment that a smallpox hospital was a nuisance per se because, even if the hospital had been managed with due care, the disease of those within would escape to infect those living in the vicinity. The action was tried by a jury and the answers that the jury gave to the questions left to them led the judge to rule that the hospital was a nuisance per se. On appeal a new trial was ordered, on terms as to costs, on the ground that the jury's findings had been against the weight of the evidence. The issue before this House was whether this order was properly made. A side issue was, however, canvassed. The plaintiffs had sought to adduce statistical evidence of the incidence of smallpox in the vicinity of two other smallpox hospitals, in an attempt to demonstrate that this was greater than in areas where there was no such hospital. Lord Selborne LC commented at p 30 that if evidence could be given of similar facts from which the effect, or absence of effect, of other hospitals on the surrounding neighbourhoods could either positively or approximately be ascertained, it would be admissible and material.

Lord O'Hagan and Lord Blackburn were both concerned at the effect that the admission of such evidence might have on keeping trials at Nisi Prius within a practicable and manageable compass, and envisaged that it might be open to the judge to refuse to admit it for this reason. Lord Watson declined to rule on the admissibility of the evidence in question, but made the following general observations at p 35:

Mr Freeland submitted that this passage from Lord Watson's speech supported the proposition that similar fact evidence is not admissible unless it is likely to be reasonably conclusive of the issue to which it relates. I believe that he misinterpreted the passage. Lord Watson was expressing the opinion that evidence of collateral facts should not be admitted unless the evidence was likely to be reasonably conclusive of the collateral facts. The collateral facts themselves had merely to raise a 'reasonable presumption or inference' as to the matter in dispute to which they were said to be relevant. More to the point, it seems to me that Lord Watson was contemplating that the trial judge would have a discretion to shut out evidence of collateral facts in the interests of keeping the trial manageable. He was not propounding an inflexible rule of admissibility.

I propose to jump nearly 100 years to an observation of Lord Denning MR in Mood Music Publishing Co Ltd v De Wolfe Ltd [1976] Ch 119. This was a breach of copyright case involving music and the issue was whether the judge had properly admitted similar fact evidence showing that the defendants had published music resembling material protected by copyright in the past. The defendants urged that the approach of this House in Boardman should be applied. Lord Denning held at p 127:

Reg v Boardman

In two subsequent cases, Sattin v National Union Bank Ltd (21 February 1978), a decision of the Court of Appeal and Berger v Raymond Sun Ltd [1984] 1 WLR 625 it was held that the test in Makin should be applied in civil cases. It seems to me that that test afforded no more precise guidance in civil cases than it did in criminal.

Of greater relevance to this appeal are two decisions of the Court of Appeal in actions in which claims of misconduct were brought against the police. In each the issue of admissibility of evidence arose in the context of applications for discovery. The first is Thorpe v Chief Constable of Greater Manchester Police [1989] 1 WLR 665. The plaintiff was arrested at a demonstration, charged with obstructing the highway and convicted before the magistrates. His conviction was quashed by the Crown Court on appeal. He sued for assault, unlawful arrest, false imprisonment and malicious prosecution. He sought discovery of documents showing any convictions of or adverse disciplinary findings against the police witnesses. The Court of Appeal held that he was not entitled to this as the evidence would not be admissible. Dillon LJ held that the observations of Lord Denning in Mood Music did not apply to a civil jury trial. The principles in Boardman should be applied. Neill LJ reached a similar conclusion, and Mustill LJ agreed with both.

In Steel v Commissioner of Police of the Metropolis (18 February 1993) the claimants were suing for wrongful arrest and malicious prosecution by three police officers. They had been convicted of conspiracy to rob and served sentences of 3 years imprisonment. Their convictions were subsequently quashed. At the heart of the claimants' case was the allegation that interview records of confessions had been fabricated at the particular instigation of a Detective Sergeant Day. They sought specific discovery of documents that they believed would establish that he had behaved in similar fashion in other cases. In giving the leading judgment Beldam LJ applied the test in DPP v P rather than Boardman, which had been applied in Thorpe. In ruling that the claimants were entitled to the discovery sought, he said:

Discussion

In giving the judgment of the Court of Appeal, Brooke LJ said

I am inclined to think that, far from this test being too lenient a test of admissibility in civil proceedings, it was too restrictive. The test of admissibility of similar facts against a defendant in criminal proceedings, as propounded in DPP v P and in the 2003 Act, requires an enhanced relevance or substantial probative value because, if the evidence is not cogent, the prejudice that it will cause to the defendant may render the proceedings unfair. The test of admissibility builds in protection for the defendant in the interests of justice. It leads to the exclusion of evidence which is relevant on the ground that it is not sufficiently probative. So far as evidence of bad character that the defendant wishes to adduce against a police witness, the test of admissibility in both Edwards and section 100 of the 2003 Act requires an enhanced relevance in order to ensure that the ambit of the trial remains manageable.

I can see no warrant for the automatic application of either of these tests as a rule of law in a civil suit. To do so would build into our civil procedure an inflexibility which is inappropriate and undesirable. I would simply apply the test of relevance as the test of admissibility of similar fact evidence in a civil suit. Such evidence is admissible if it is potentially probative of an issue in the action.

This is not to say that the policy considerations that have given rise to the complex rules of criminal evidence that are now to be found in sections 100 to 106 of the 2003 Act have no part to play in the conduct of civil litigation. They are policy considerations which the judge who has the management of the litigation will wish to keep well in mind. CPR 1.2 requires the court to give effect to the overriding objective of dealing with cases justly. This includes dealing with the case in a way which is proportionate to what is involved in the case, and in a manner which is expeditious and fair. CPR 1.4 requires the court actively to manage the case in order to further the overriding objective. CPR 32.1 gives the court the power to control the evidence. This power expressly enables the court to exclude evidence that would otherwise be admissible and to limit cross-examination.

Similar fact evidence will not necessarily risk causing any unfair prejudice to the party against whom it is directed. It would not have done so in Metropolitan Asylum District Managers v Hill. It may, however, carry such a risk. Evidence of impropriety which reflects adversely on the character of a party may risk causing prejudice that is disproportionate to its relevance, particularly where the trial is taking place before a jury. In such a case the judge will be astute to see that the probative cogency of the evidence justifies this risk of prejudice in the interests of a fair trial.

Equally, when considering whether to admit evidence, or permit cross-examination, on matters that are collateral to the central issues, the judge will have regard to the need for proportionality and expedition. He will consider whether the evidence in question is likely to be relatively uncontroversial, or whether its admission is likely to create side issues which will unbalance the trial and make it harder to see the wood from the trees. He will have well in mind the considerations that concerned this House when contemplating the effect of the admission of the disputed evidence in Metropolitan Asylum District Managers v Hill.

For these reasons I would reject the appellant's submission that similar fact evidence is only admissible in a civil suit if it is likely to be reasonably conclusive of a primary issue in the proceedings or alternatively if it has enhanced relevance so as to have substantial probative value.

The result in this case

Mr Freeland realistically did not suggest that the evidence that Mr O'Brien seeks to introduce in this action is not relevant or probative. He argued that it was not relevant or probative enough to pass the test of admissibility. Applying the correct test of whether the evidence is potentially probative, the answer is plainly that it is.

Mr Freeland also emphasised that the evidence in question would be hotly contested, so that it was likely to add greatly to the length and complexity of the trial. This is obviously cause for concern in a case such as this. It was of concern to Judge Graham Jones. He directed himself that it was necessary to keep the case within proportionate and manageable bounds and to ensure that the jury was not distracted from its central task. The judge gave careful consideration to the extent to which the evidence would add to the length and complexity of the trial and concluded that it would not do so to an extent that called for the exclusion of the evidence.

The Court of Appeal concluded that the judge had been seriously over-optimistic as to the extent to which the trial judge would be able to control the amount of evidence that would be called on the similar fact issues. Accordingly it proceeded to exercise its own discretion. It had regard to the importance of the case for Mr O'Brien, who alleges that the improper actions of the police officers resulted in his spending 11 years in prison for a murder that he did not commit. It had regard to the difficulty facing a claimant in Mr O'Brien's position in establishing his account on the basis of his word against that of a number of police officers, and to the impact that the similar fact evidence could have in assisting him to overcome that difficulty. Finally the court commented that the evidence was of malpractice that, if proved, struck at the heart of the administration of a fair system of criminal justice.

Mr O'Brien has now made it clear that he is not seeking a jury trial. Quite apart from this, there is nothing that would justify this House in conducting a review of the exercise of discretion by the Court of Appeal. For the reasons that I have given I would dismiss this appeal.

LORD RODGER OF EARLSFERRY

My Lords,

I have had the advantage of considering your Lordships' speeches in draft and I agree with the analysis of the law which they contain.

I also agree that, since there is no demonstrable error in the approach adopted in the courts below, it would not be appropriate for the House to differ from the conclusions which they have reached that the similar fact evidence should be admitted. I am bound to say, however, that in the case of R v Ali the alleged connexion of Mr Lewis with the supposed misconduct in interviewing Mr Ali is somewhat indirect and, if exercising the initial judgment myself, I would have been inclined to exclude that evidence on the ground that its potential significance would not justify the time and expense of exploring it at the trial.

The probative value of the evidence relating to R v Griffiths is more readily apparent. None the less, as the hearing of the appeal went on and counsel for both parties explained the issues and counter-issues which would be explored, it seemed to me that there was a considerable risk that they would add very greatly indeed to the length of the trial and might even come to dominate it.

Therefore, while I see no sound basis for interfering with the decision of the Court of Appeal, I should not wish the decision in this case to be seen as authority for the view that there may not be sound reasons in case management terms for excluding or limiting similar fact evidence which is likely to be as extensive as the evidence in this case.

LORD CARSWELL

My Lords,

I have had the advantage of reading in draft the opinions of my noble and learned friends Lord Bingham of Cornhill and Lord Phillips of Worth Matravers. I agree with the conclusions reached by them and with the reasons which they have expressed, and wish to add only a few observations of my own.

A significant part of the difficulty which has been found in articulating with proper clarity the rules governing the admission of similar fact evidence has arisen from the failure to appreciate the need to preserve a sufficiently clear distinction between criminal and civil cases and between the two stages of the juridical process of consideration of the admission of evidence of similar facts.

It is helpful in any consideration of the topic to keep distinct these two stages, as there has been a tendency in many of the decided cases to elide them. The first stage is common to both criminal and civil cases, the requirement that the evidence which it is proposed to adduce is relevant to one or more issues in the trial. The second stage is the application of the control test, which is different in civil cases from that which is applied in criminal trials. In the latter the second stage is commonly incorporated with the first to make a composite rule of law, but they do nevertheless reflect distinct reasoning processes.

The test in the first stage is that of relevance, whether the evidence is logically probative or disprobative of some matter which requires proof: Director of Public Prosecutions v Kilbourne [1973] AC 729, 756, per Lord Simon of Glaisdale. On this issue I respectfully agree with what Lord Bingham of Cornhill has said in para 4 of his opinion. The matters which require proof in the context of the present appeal are that Detective Inspector Lewis committed the acts alleged against him in the pleadings and that Detective Chief Superintendent Carsley knew of irregular behaviour in the investigation and failed to take any action.

Mr Freeland QC for the appellant submitted that the test of relevance which I have set out should be expressed in more stringent terms. He propounded a number of suggested limitations, which may be summarised as follows:

(i) mere propensity to behave in the manner alleged in the material averments is insufficient;

(ii) isolated examples of alleged misconduct in other cases are insufficient;

(iii) evidence of similar facts must, in order to be admissible, be highly or strongly probative of an issue in the substantive case; an alternative mode of expressing this is that it must have enhanced relevance;

(iv) the probative strength of the allegation which the party seeks to adduce as evidence of similar facts, and hence its admissibility, will be affected by the extent to which that allegation is a proven fact: complex, unsubstantiated allegations will not generally suffice and the evidence should be reasonably conclusive.

In some of the cases the two stages may have been elided and the control mechanism imported from the second stage into the test of relevance. It was suggested by the appellant's counsel that the decision in Metropolitan Asylum District Managers v Hill (1882) 47 LT 29 supported the proposition that similar fact evidence was not admissible unless it was reasonably conclusive of the issue to which it related. For the reasons given by Lord Phillips of Worth Matravers in para 46 of his opinion I think that the passage from Lord Watson's speech on which counsel relied has been misinterpreted and that it means only that the collateral facts should be established by reasonably conclusive proof, which relates rather to Mr Freeland's fourth proposed limitation. It is right to say, however, that the two stages in considering the admission of similar fact evidence are not kept clearly distinct in the speeches either of Lord Watson or Lord O'Hagan. In the same way, the speeches of their Lordships in R v Boardman [1975] AC 421 do not maintain a distinction between the stages. This is understandable in a criminal case, where in the second stage the probative force of the similar fact evidence has to be balanced against the degree of prejudice to which it may give rise, so that this rather than pure relevance is the important issue for decision and the tests are commonly compressed into a single principle.

In my opinion the correct approach is to keep the stages separate and to reject the more stringent tests of relevance as a condition of admissibility. This proposition is borne out by modern authority. Lord Denning MR expressed it in Mood Music Publishing Co Ltd v De Wolfe Ltd [1976] Ch 119, 127 in a passage which clearly expressed the correct approach:

I do not myself find it helpful to dismiss certain types of similar fact evidence by giving them the pejorative label of evidence of "mere propensity" or "disposition". As my noble and learned friend Lord Steyn pointed out in R v Randall [2004] 1 WLR 56, at para 26:

A case in point is R v Straffen [1952] 2 QB 911, in which the similar fact evidence could be described as evidence of pure propensity to commit crimes similar to that with which he was charged. The probative strength of the evidence may be a material factor in balancing the factors in the second stage of the process, not only in criminal trials, but in civil cases, as the Court of Appeal pointed out in para 71 of its judgment in the present case; cf L H Hoffmann, (1975) 91 LQR 193, 205. It should be kept firmly in mind, however, that it is not such a factor in the first stage.

Mr Freeland sought to draw support for his second proposed limitation, that isolated examples of similar facts will not suffice, from the speech of Lord Morris of Borth-y-Gest in R v Boardman [1975] AC 421, 439, where he said that, to be admissible, evidence must be related to something more than isolated instances of the same kind of offence. This statement was in the context of a criminal case, in which, as I have said, the probative strength of the similar fact evidence is a material factor in considering the second stage of the test of admissibility. Even then one must keep in mind the observations of Lord Hobhouse of Woodborough in R v Z [2000] 2 AC 483, 508:

A parallel may be seen with circumstantial evidence, which Pollock CB in R v Exall (1866) 4 F & F 922 at 929 compared to a rope comprised of several cords. He went on to say:

The limitation which Mr Freeland propounded with most vigour was his third, that the similar fact evidence must have enhanced relevance, that is to say, it must be strongly probative. Again he called in aid statements of Lord Wilberforce, at p 444 and Lord Hailsham of St Marylebone, at p 454 of R v Boardman [1975] AC 421. Again, however, it has to be borne in mind that they were considering the test in criminal cases: see the observations of Lord Phillips of Worth Matravers at para 33 of his opinion in the present appeal. There is no good reason in principle to require that evidence of similar facts in civil cases must, to be admissible, be strongly probative or have enhanced relevance. Nor is there in my opinion any authority which supports such a proposition, for the cases cited on behalf of the appellant, when properly analysed, do not bear it out. On the contrary, the clear statement of Lord Denning MR in the Mood Music case is firmly against it. I accordingly agree with the conclusion expressed by Lord Bingham of Cornhill in para 4 of his opinion, that in the first stage of the enquiry admissibility turns only on whether the evidence proposed to be adduced is probative. In a criminal trial, as Lord Phillips of Worth Matravers pointed out in para 52 of his opinion, it may be necessary to look for enhanced relevance or substantial probative value, for that may be necessary to offset the degree of prejudice caused, but that is a matter for the second stage.

The appellant's fourth suggested requirement, that evidence of the allegations proposed to be adduced as similar facts will be admitted only if they are proven facts, is in my view wrong both in principle and on authority. It is refuted by the analysis which I have quoted of Lord Hobhouse of Woodborough in R v Z of the cumulative strength which may be built up from a number of relatively frail strands. It is inconsistent with the remark of Lord Mackay of Clashfern LC in R v H [1995] 2 AC 596, 605 that the judge is not to be held to have accepted that the evidence is true. It is also inherent in the decision in Director of Public Prosecutions v P [1991] 2 AC 447 that the allegation was unproven, as both incidents in that case were the subject of the trial of the defendant. Moreover, section 109(2) of the Criminal Justice Act 2003 expressly recognises that the truth of the allegation may not have been formally established. The strength of the allegations, which may be evidenced by their having been established as proven facts, may come into the scales in the second stage, but it is not necessary in the first stage to require that they be so proven.

When the court has decided that the evidence proposed to be adduced as similar facts is logically probative of the matter or matters in the substantive case to which it is desired to relate it, then it must in the second stage of the exercise consider the limiting or control factors. I fully agree with Lord Bingham of Cornhill's description of these factors in paras 5 and 6 of his opinion and do not wish to add more than a couple of thoughts. First, I would not accept the appellant's submission —which may be said to relate to both stages—that the two sets of allegations must be so similar that coincidence can effectively be ruled out. Secondly, the lengthening of the trial and increase of costs to which the calling of similar fact evidence will give rise must not be disproportionate. Thirdly, the application of the principles governing the limiting factors may differ according to whether the mode of trial is with a jury or by a judge alone.

His Honour Judge Graham Jones, sitting as a judge of the High Court, in a thorough and careful judgment reviewed the matters proposed to be adduced as similar facts and concluded that the aspects of the evidence which he listed were relevant as being probative. The Court of Appeal agreed with these conclusions, but also added the further matter which was the subject of the cross-appeal. I consider that all these matters satisfied the test appropriate to the first stage and that the decision of the Court of Appeal was plainly right. It was urged upon the House by the appellant's counsel that the evidence against DCS Carsley was too flimsy to be allowed in, but that will be a matter of proof: if established on the evidence, the allegations are relevant and potentially admissible.

The judge went on to make an exercise of his discretion in accordance with the factors material to the second stage, which he correctly identified. The Court of Appeal upheld the exercise of that discretion in respect of the allegations in the Ali case. In respect of the Griffiths case the Court of Appeal considered that the judge had under-estimated the potential lengthening of the trial if the allegations made in that case were adduced as similar fact evidence. It accordingly proceeded to exercise its discretion afresh and, having done so, reached the same conclusion. I see no ground on which to interfere with the conclusions of the judge or the Court of Appeal, both of whom expressed the principles correctly and gave proper consideration to the issues.

I would therefore affirm the decision of the Court of Appeal and dismiss the appeal.

BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/UKHL/2005/26.html

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	United Kingdom House of Lords Decisions
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> O'Brien v Chief Constable of South Wales Police [2005] UKHL 26 (28 April 2005) URL: http://www.bailii.org/uk/cases/UKHL/2005/26.html Cite as: [2005] WLR 1038, [2005] 2 WLR 1038, [2005] 2 All ER 931, [2005] UKHL 26, [2005] 1 WLR 1038, [2005] 2 AC 534