PC Appeal Nos 0088, 0089, 0094 of 2013

IpsofactoJ.com: International Cases [2015] Part 10 Case 3 [PC]


(from the Court of Appeal, Bermuda)



- vs -


Lord Kerr

Lord Wilson

Lord Hughes

Lord Toulson

Lord Hodge

6 OCTOBER 2015


Lord Hughes

  1. The Board has heard together three appeals against conviction, challenging decisions of the Court of Appeal of Bermuda. They raise similar, although not identical, questions concerning the admissibility and proper ambit of evidence as to the existence and practices of gangs and the defendant’s connections with them.

  2. Each of the three cases concerned a deliberate shooting by someone who had clearly sought out the victim in order to shoot to kill. Two of the victims were shot dead; the third (in the case of Brangman) was hit but survived. The issue in each case was identity. In two of the trials (Myers and Cox) the Crown case was that the shooting was part of a long-standing feud between two rival gangs, and was triggered by an incident shortly beforehand in which there had been either insult to, or attack on, someone associated with one of the gangs. The shooting was in those two cases said to be a rapid and fatal retaliatory attack, not on the perpetrator of the earlier incident, but simply on some random member of the opposing gang. In these cases the gang evidence was part of the prosecution case from the outset. In the third trial (Brangman) the gang evidence was not originally part of the Crown case. It was admitted in consequence of cross examination of the victim, not to prove the existence of a feud between identifiable gangs, but to prove intra-gang loyalty between the defendant and an associate of his who had suffered an insult or attack. Thus in each of the cases the gang evidence was admitted primarily to demonstrate that the defendant had a motive to kill the victim.

  3. In all three cases the principal gang evidence was given by a police officer, Sergeant Rollin. His evidence was admitted at each trial against objection taken on behalf of the defendants and, as will be seen, there was also objection to related evidence from other prosecution witnesses. Before the Court of Appeal several grounds of appeal against conviction were advanced, including that this evidence ought not to have been admitted. The Court of Appeal rejected that ground of appeal in each case, although by a majority in the cases of Myers and Cox. In those two appeals, Auld JA disagreed on that point, although he concurred in upholding the convictions on the grounds that the proviso to section 21(1) of the Court of Appeal Act 1964 (no substantial miscarriage of justice) applied. Before the Board, the only surviving ground of appeal which the defendants have permission to pursue relates to the admission of the gang evidence.

  4. The appeals raise the following broad issues:

    1. whether the gang evidence was inadmissible as irrelevant and/or as impermissible proof of no more than bad character, or was admissible to prove motive;

    2. whether the gang evidence was admissible under the principle enunciated in R v Pettman (unreported 2 May 1985) (“explanatory evidence”);

    3. if gang evidence is admissible, what is its proper extent and content;

    4. if admissible, whether gang evidence can be given by a police officer who has made a special study of the gangs concerned, as well as of gang culture generally;

    5. if gang evidence is admissible, to what extent (if any) may the witness rely on information gathered from, or researched by, others;

    6. if such evidence is prima facie admissible, what ought to be the approach to the application to it of section 93 of the Police and Criminal Evidence Act 2006 (prosecution evidence which ought not to be admitted because its effect on the trial would be unfair); and

    7. if such evidence is admissible, what ought to be the practice relating to the advance notice, form and presentation of it.


  5. On the night of Friday 4/Saturday 5 December 2009 there was a party at the Devonshire Recreation Club. At about 03.30 there was an argument between Neika Daily and David Cox. Daily threw a beer bottle at Cox, who threw a drink over her. Daily was pushed out of the club. Kumi Harford was also at the club with several friends. Sometime after 04.30 his group left and returned in various combinations to St Monica’s Road, where one of the girls in the group lived. By just after 05.00, Harford had left her home and was driving his car away when he was shot dead at the wheel. Two firearms had been used. None of those facts was in dispute.

  6. The Crown sought to prove, chiefly through Sergeant Rollin, that there was a violent feud between, on the one hand, the Parkside and Middletown gangs (associated with each other) and, on the other, the 42nd gang and its associate, the MOB gang. Myers was said to be a member of Middletown. Neika Daily was said to be closely associated with Middletown. Her home was one of its principal haunts. Her son (Juju Williams) was said to be a member, whilst her niece Rogernae, who lived partly with her, was a girlfriend of Myers. On the other side, David Cox and the deceased Harford were said to be members of the 42nd. The Crown’s case was that the shooting of Harford was a further episode in this violent feud, and an instant retaliation by Middletown against the 42nd for the insult done to Neika Daily about an hour and a half beforehand in the Devonshire Club.

  7. Sergeant Rollin gave evidence that he was a member of a small police unit charged with targeting gangs in Bermuda. He regularly patrolled the streets where gangs congregated. He saw and frequently spoke to their members, most of whom he knew by name. He had studied their territories, and the markings which they put on walls within them, and their structures. He had undergone specialist training in gang monitoring and study from the FBI, both locally and in the USA and he was in communication with that institution’s head and field offices in Washington DC. There were three other officers in his unit, also patrolling regularly in the areas concerned. They pooled their information, sightings and like material upon a shared database. The Crown put him forward as an expert in gang culture in general and the Bermuda gangs in particular. The judge treated him as such.

  8. The form of Sergeant Rollin’s evidence was fourfold. Firstly, he described the two relevant gang groupings, namely Middletown/Parkside on the one hand and 42nd/MOB on the other and explained their territories. Secondly, he gave evidence of the violent feud between them, referring to acts of violence and murder in general and to the murder by shooting of one of the 42nd named Kenwandee Robinson in particular. Thirdly, of a number of named people, including the defendant, he gave evidence that he considered them members of one or other of these gang groupings, stating inter alia the allegiances set out in para 6 above. He described the defendant as a high-level member of Middletown. And fourthly, he produced three photographs which showed assembled groups of Middletown/Parkside members, including the defendant, identified them by name in most cases, and described as the gang sign the “M” symbol which the defendant could be seen making with his hand in two of the photographs, and which another member was also making. When challenged on the basis that his information derived from others, he said that there was a unit database to which all members of his team contributed, but that only a very small part of what he had given in evidence was other than his first-hand information. His evidence was that he had known the defendant very well for some eight years, having seen him almost daily.

  9. However, when identifying named persons as members of one or other gang, Mr Rollin’s witness statement simply said that he “considered” X to be a member of the relevant gang. He did not set out the basis for his belief, beyond saying that he was very familiar with the streets. He did not, in his witness statement, give information of particular, or even unparticularised, sightings of the individual concerned in significant places or in significant company, although in oral evidence he did say, in relation to the defendant, that he was habitually to be found with other members in or near the house of Neika Daily, which was the common meeting point of the Middletown gang, whilst the photographs were further evidence of this association.

  10. Both in his witness statement and in his oral evidence, Sergeant Rollin added the information that the Middletown gang (and the other gangs also) were known to him to be involved in the sale of controlled drugs, in robberies and in acts of violence, including with firearms. Moreover, when describing the feud between the gangs, he asserted that the cause was unknown, but that it appeared to revolve around drug dealing. Of the defendant, he added his evidence that he had never known him to be in employment.

  11. The evidence of a firearms expert Mr McGuire established that two weapons had been used to gun down Harford. One was of .40 calibre and the other 9 mm. His evidence, together with that of Detective Constable Burgess, whose job it was to record the findings of ballistics experts in shootings, demonstrated that the .40 gun had been implicated in six other attacks, some before and some after the present one, and the 9 mm gun in two, both after it. All but one of those offences had occurred in the St Monica’s Road area, the territory of the 42nd gang. In anticipation of the several victims of these shootings being named by DC Burgess, Sergeant Rollin had identified all but two of them seriatim as associated with 42nd/MOB. One of them was Kenwandee Robinson. DC Burgess gave it as her evidence that gangs appeared to keep guns for ready use by their members. The inference invited was that these guns were available to Parkside/Middletown. The witness confirmed in cross examination that that was her understanding; no suggestion was put to her that she was wrong.

  12. Objection was taken to the admissibility of the gang and gun use evidence given by Sergeant Rollin, Mr McGuire and Detective Constable Burgess. The objection was principally on the basis that it amounted to impermissible evidence of bad behaviour by the defendant other than the offence charged. The Crown contended that it was admissible as “background evidence”, in particular as showing motive, and relied upon R v Pettman. The judge held that it was admissible as relevant to motive. Having thus ruled, the judge separately also declined to exclude the evidence that the Middletown gang generally, and the defendant in particular, were concerned in the supply of illegal drugs, along with the evidence that the defendant had no record of employment.

  13. Gang evidence apart, Edwin Darrell gave evidence that shortly after hearing the gunshots he had encountered Myers, whom he knew well. Darrell said that the defendant had told him that he had just shot Harford, and that he had asked Darrell to help dispose of his clothing. Darrell said that he had refused, and that Myers had then asked the whereabouts of a mutual friend, Andrew (Sykie) Laws. Later that night, according to Darrell, he had seen Myers and some friends burning some clothes near to Neika Daily’s house until interrupted by the arrival of the police, at which they fled. Darrell further said that in the ensuing weeks Myers had threatened him with a gun and warned him to stay quiet about what he knew. Andrew Laws also gave evidence, to the effect that he had heard a friend of the defendant remonstrating with him for not having got Laws to dispose of the clothes. (Laws was the father of one of the Middletown members identified by Sergeant Rollin, Jahkiel Samuels.) There was evidence from the police officers who had come upon the fire that they had saved from it some jeans, two shoes and a sock which were shown to have on them DNA which matched that of the defendant and which were admitted at the trial to belong to him. The jeans and sock also bore traces of gunshot residue.

  14. The evidence of Darrell and Laws was hotly contested at the trial. The suggestion was made that both had concocted their evidence in pursuit of a reward. Darrell’s bad character as a self-confessed drug addict who had served ten years for drug importation was adduced and the additional suggestion made that his evidence might be the product of drug-induced paranoia. Laws’ regular drug misuse was adduced. The defendant called a gunshot residue expert to suggest the possibility of contamination in police hands of the jeans and sock, and his mother to assert that around half an hour before the murder he was seen on a couch at her home watching television, although the timings and distances involved were such that the latter would not, if true, necessarily have prevented his being responsible. He did not give evidence himself.


  15. On the night of 8 August 2010 at about 23.45 Julian Washington was shot in the Mid Atlantic Boat Club. The Crown case was that the Club was a regular haunt of the 42nd gang and that Washington was shot in the course of an indiscriminate attack on those present by one or more people who arrived and sprayed gunshot about. Washington was hit in the stomach, but was taken to hospital and survived. Within about three quarters of an hour, at around 00.25, Troy (“Yankee”) Rawlings was shot dead in the doorway of the Spinning Wheel Club in Court Street by two armed men, one of whom was said to be the defendant Cox. Cox was said by the Crown to be a member of 42nd whilst Rawlins was said to be an associate of the rival gang Parkside, via an earlier-formed but connected gang called Frontline. The killing of Rawlings was said to be an immediate reprisal for what had happened to Washington.

  16. The only witness called by the Crown to prove the antecedent shooting of Washington was Detective Constable Edmonson. She had arrived at the Boat Club some two and a half hours after the event and although she was the co-ordinator of the police file relating to the incident she had no first-hand knowledge of anything at all. The Crown was induced by objection to her evidence to accept that she could not relate what people at the club had told her, as had plainly originally been intended. She was nevertheless invited by the Crown to tell the jury about the shooting of Washington, and moreover what theory the police had about it. She gave evidence that it was thought to have been a targeting of the 42nd’s known place of recreation. In that form, this was plainly not admissible, and should not have been permitted unless by agreement, which might well have been appropriate on the facts of this case but which it is clear never occurred. However, it made no difference, because the defendant’s interviews with the police certainly were admissible and he both told them that he had been there and made it sufficiently clear what had happened. Washington was a close friend of his. The defendant had borrowed his motorcycle to get to the club. Washington had only come to the club to retrieve it, but had stayed to have a drink. Cox himself had run when the shooting started, and had taken refuge in the Ladies’ lavatory. He asserted that he had not seen what had happened. He knew, however, that Washington had been taken to hospital. He might, he said, easily have been shot himself. He had left immediately, in a jeep driven by a friend. He asserted that he had spent the rest of the night in the home of a girl he had met, but whom he declined to name. He told the police that he came from the 42nd district and that “they” were targeting the 42nd all the time. Somebody had, he said, “just shot up the Boat Club”. He said that all the shooting that there had been on the Islands was “crazy” and that he had himself lost four friends, Minors, Kumi Harford (the victim in Myers’ case) Kenwandee Robinson and Perry Puckerin (both referred to in Myers’ case). Asked whether he would go to the area of Parkside (Court Street) he said carefully that he would not “just to hang out”. Asked if there was a war, he said that he was no he-man. In due course he gave evidence himself at his trial to broadly similar effect. He said that although he came from 42nd area he was not himself a member of the gang.

  17. Sergeant Rollin gave evidence of his posting to the Gang Targeting Unit, and his training, more or less as he had in the case of Myers (see para 7 above). He did not in this case give evidence of the pooling of information between himself and other officers. His evidence about the persons involved was given as deriving from his personal observations and the interpretation of photographs which he produced. He was challenged in cross examination that anything he said about the shooting of Washington at the Mid Atlantic Boat Club was second-hand hearsay, but it was not suggested that his evidence about the individuals of whom he spoke suffered from the same defect, and it did not on its face do so.

  18. Of Washington, Sergeant Rollin said that he had known him for about five years, and that he saw him almost daily in the St Monica’s area, in particular at the house known as the ‘Dublin residence’ which was a haunt of the 42nd gang. He produced a number of photographs in which Washington appeared, and in which he could be seen wearing a necklace with a pendant with the numbers “42” on it, and in which he was in the company of another man with a similarly numbered necklace. He gave evidence that such necklaces were common to members of the 42nd gang. In two other photographs which he produced he identified Washington “throwing up” (making a sign) of four and two fingers, which sign he identified as a mark of the gang. Next, he gave evidence that in patrolling the area he had seen members of the 42nd congregating at the Mid Atlantic Boat Club at all hours, but had never seen a member of the rival Parkside gang there.

  19. Of Rawlings, Sergeant Rollin said that he had known him for some nine years, and saw him more or less daily in the area of Court Street, which was the haunt of the Parkside and Middletown associated gangs. (It was also where the Spinning Wheel Club lay.) He had seen him regularly in the company of other members of those gangs. Rawlings had been a member of the Frontline gang which had formed in the late 1990s or early 2000s and which was associated with Parkside; many of its members had now grown out of the association, but others had not. Rawlings was referred to on the street by others as an “OG” (original gangster) and was accorded respect in consequence.

  20. As to Cox, Sergeant Rollins said that he had known him for some nine years having seen him habitually in the St Monica’s Road area, which was the home of the 42nd gang. He knew Cox to wear the necklace with a “42” pendant, and to have a tattoo on his arm reading “Coxy 42nd holla”. He produced a photograph of the tattoo.

  21. Sergeant Rollin gave evidence of an ongoing feud between the 42nd gang on the one hand and Parkside/Middletown and their associates on the other. His evidence was that the culture of the gangs was such that an insult to a member of one would provoke retaliation against some member of the other side, not necessarily against the perpetrator of the insult. High-ranking members (“shot-callers”) would direct lower rankers to carry out such retaliatory attacks.

  22. Ballistics examination by Mr McGuire of bullets and spent cartridge cases from the scene demonstrated that two firearms had been used in the murder. One was a .38 revolver and the other a 9mm auto-loading pistol. The police had recovered the former from the St Monica’s area. It had been used in three other shootings on the Islands. The unrecovered 9mm weapon had been used in five other such shootings. DC Burgess produced the records relating to the other shootings. They had all taken place in the Pembroke area frequented by the Parkside gang. There was in this case no evidence adduced by the Crown about any of the victims. Counsel for the Crown led DC Burgess to agree with the propositions that these findings appeared to show that “gang members share the same weapon” and do not have access to multiple weapons.

  23. Sergeant Rollin’s evidence also included the propositions that the gangs he mentioned were active in the sale of controlled drugs, in firearms offences and in violence. He added that the origins of the feud which he described were unknown but that it appeared to revolve around the sale of drugs.

  24. Objection was taken to the evidence of Sergeant Rollin, and to that of Burgess and McGuire so far as it related to previous use of the guns on the grounds that it constituted inadmissible evidence of bad character and/or mere propensity to offend, contrary to Makin v Attorney General for New South Wales [1894] AC 57. Secondly it was contended that Pettman did not render it admissible since the case could perfectly well be understood as an attack with guns without the evidence of the suggested gang background. The judge admitted the evidence, treating Sergeant Rollin as an expert in gangs.

  25. Gang evidence apart, the Crown case against the defendant consisted of the following.

    1. Michael Parsons, who had known the defendant for many years as a friend, gave eye-witness evidence that he recognised him as one of the two gunmen; he was standing near him when the attack occurred and in the well-lit doorway of the Spinning Wheel Club; the gunman in question was wearing a zipped up and hooded rain jacket but he said that he recognised him by his eyes; the defendant accepted that there was no bad blood between Parsons and himself.

    2. CCTV showed that the gunman in question was wearing a peaked baseball-type cap (or similar) under his hood; outside the Club after the shooting the police found such a cap bearing DNA which matched the defendant and also gunshot residue; this was moreover in an area which the defendant did not ordinarily frequent.

    3. An eye witness said, and the CCTV also showed, that the gunmen appeared to be wearing latex gloves; they escaped in a stolen car which was found at about 07.00 next morning close to a yard which the defendant admittedly frequented; in the yard were four latex gloves, on one of which was found DNA matching the defendant and gunshot residue, whilst two of the others also bore gunshot residue. In interview the defendant first denied any use of latex gloves and then asserted that he had used some in Pembroke whilst mending his motorcycle to keep oil off his hands; there was, however, no oil on the recovered gloves.

    4. Close to the yard the police recovered one of the guns used (by the other gunman) in the murder.


  26. At about 02.15 on the night of Friday/Saturday 13 February 2010 Nathan Darrell was sitting in a car near his flat, rolling a reefer and speaking on the telephone, when a man on a motorcycle came up to the car and peered in at him, before moving away. To see what was going on, Darrell moved the car closer to his flat. He saw the motorcycle and its rider standing beside it. He parked in the yard of his flat. The person he had seen came up to the car, now with the helmet visor up, paused as if to make sure that Darrell had seen him, and then shot him through the closed window. It later transpired that at least three shots were fired from a 9mm Glock semi-automatic pistol. Darrell escaped through the passenger door, but had been shot in the chest, neck and buttocks. He eventually reached help and was taken to hospital. They were able to save him. His evidence was that he had recognised the defendant, whom he had known well for years, as the gunman. The defendant was charged with attempted murder.

  27. The Crown case thus depended essentially on the recognition evidence of Darrell. As support, it relied also on the following:

    1. Two police officers said that at about 02.40 on the Saturday afternoon, about twelve hours after the shooting, they saw the defendant in Somerset Village, not far away. He was wanted, having by then been named by Darrell. When he saw the officers, he ran away, scaling a fence and evading their pursuit.

    2. The defendant went voluntarily to the police station about three hours later that afternoon. According to PC De Silva, when being placed in a cell, he was asked whether he was still seeing Nakisha Robinson and replied that that had been over some time ago. However, later that evening the police searched the Cambridge Hotel (also near Somerset Village) and found that Miss Robinson had taken a room for the Friday and Saturday nights, in which they recovered a phone camera containing photographs of herself and the defendant enjoying themselves and sharing the bed.

    3. In the same room the police found a face mask, a pair of gloves and a Nolan motorcycle helmet. One glove had on it DNA matching the defendant with a very low possibility of random match. There were gunshot residue particles on one glove and the face mask, of a composition not unique to the ammunition used in the shooting. Two gunshot residue experts differed in their assessments of the likelihood of particles of this particular composition deriving from the ammunition used. The helmet was of a reasonably common make, but the same as Darrell said the gunman had worn. There was no evidence whether or not the gunman had worn a face mask at any time, and at the time of shooting his face had been uncovered.

  28. The defendant did not give evidence and had not answered any questions about the shooting when interviewed by the police. The case put on his behalf was that Darrell’s identification was wrong, either through error or malice. It was suggested on his behalf that the police officers had not chased him, but another person. It was put to PC De Silva that the conversation about Nakisha Robinson was invented. The other supporting evidence was said to be inconclusive.

  29. The Crown did not initially propose to call evidence of gang associations or relationships. Darrell’s evidence was that the gunman had acted as if he wished to make sure that his victim saw who it was who was shooting him. He had made a witness statement setting out his connection with the defendant and suggesting that the motive for the shooting was revenge arising from personal animosity. His statement recounted a number of hostile encounters between the two men, when they had, in effect, faced each other down or treated each other with disrespect. Two of them had been in the week or so before the attempted murder. He had also related that there was rumour “on the street” that he had shot a gun at the house of a man called Wellman, who was a cousin of the defendant, and with whom he (Darrell) had had past dispute; this, he suggested, might have contributed to the animosity and be part of the motive for the shooting. In his evidence in chief, he told of the animosity, added that there was “war” between himself and the defendant and others, and made no bones about the fact that he had given as good as he had received. He said, however, that the war was not a matter of gang feuding. In asserting the animosity, he related the hostile encounters between the defendant and himself but did not refer to (and was not asked about) the rumoured shooting at Wellman’s house.

  30. Counsel for the defendant introduced the Wellman connection in cross examination. He had already established from Darrell that he sometimes carried a gun (ostensibly to demonstrate that his original description of the weapon as a revolver rather than a semi-automatic pistol would have been informed by knowledge, with a view to undermining the reliability of his observations). He then elicited the street rumour and suggested in terms that immediately after he had been shot, Darrell had asked himself who might have done it, and answered himself that it must be to do with the rumour. He went on to suggest to Darrell that the latter knew that Wellman had been arrested shortly before the shooting, and that in consequence he believed that it could not have been Wellman who had shot him. The next stage of the suggestion put was that, excluding Wellman for this reason, Darrell had fastened on the defendant as the culprit. The last stage was to suggest that actually Wellman had been released on bail and so could have done it.

  31. A second line of cross examination of Darrell was to suggest that he had no shortage of enemies. Counsel elicited the fact that Darrell had told a police officer that he believed that a local garage owner had paid someone else to kill him; Darrell added that the someone was, as he believed, the defendant. Moreover, a particular incident was put to Darrell when, after a football match, he had responded to perceived insult by standing in the middle of the pitch “pumping your chest to people like Yankee Troy Rawlings and Parkside, saying ‘Come bring your guns; come and get me’”. He accepted that this had indeed occurred.

  32. The Crown called ballistics evidence from Mr McGuire. It adduced evidence of the type of gun used, that it appeared to have been a single gun, and as to what could be inferred about the proximity of the gunman from the fact that an ejected casing was found inside Darrell’s car. Counsel for the defence then adduced in cross examination that the same gun which fired the bullets at Darrell had been used in two other shooting incidents.

  33. In consequence of these aspects of cross examination, the Crown applied to adduce additional evidence from Sergeant Rollin and from Mr Belton, a consultant to the local police in gang-related matters. Sergeant Rollin’s evidence was contained in a statement made during the trial. It was tendered to show a close gang association (MOB) between Wellman and Brangman. The object was to support the argument that even if Wellman would himself have been angered by any shooting at his house, the defendant’s gang loyalty would have led him to see it as a reason for reprisal also by him. The judge ruled that the import of the cross examination had plainly been to suggest a motive for Wellman rather than the defendant to shoot Darrell, and that it had in this way opened the door for the additional Rollin evidence. The judge also admitted, against objection, the evidence of Belton, to the effect that the other shootings involving the same gun, to which McGuire had referred in cross examination, were consistent with that gun being used on different occasions by associated members of a gang. Before the Board, the complaint on behalf of the defendant is confined to the admission of the Rollin evidence.

  34. Sergeant Rollin gave evidence of his posting and experience to the same effect as he had in the cases of Myers and Cox. Of both Brangman and Wellman he said that he considered them members of the MOB gang. As to Brangman, he said that he had known him for about three years. He based his conclusion on Brangman’s association with known members of the gang and on locations at which he had been seen. As to Wellman, he said that he had known him for about six years mainly from his patrols of the western end of the Island. He frequented gang locations and associated with other members of the gang. He produced a photograph which showed both Brangman and Wellman at a party. Brangman was making a W sign with his fingers, which was a sign associated with the MOB gang, deriving from the western end of the Island, which he said was its territory. Wellman was making a gun sign with his hand. Sergeant Rollin said that Wellman was higher up in the gang than Brangman. He also gave evidence about the places which this gang frequented. Further he explained that gang loyalty was such that anything untoward done to one member was treated as done to the entire gang. More senior members were referred to as “shot callers” and junior ones as “soldiers”.

  35. Sergeant Rollin said that his unit collected intelligence about gangs on the Island. He had, he said, a number of sources which were “entrenched within” the gangs, by which he must have meant informers. However, he refuted the suggestion put to him in cross examination that his conclusions were based simply on what other people had told him. He had seen a lot of what happened for himself, and he himself spoke to the gang members on a daily basis. He said that his evidence was based largely on his own conversations and observations, and his personal interaction with gang members and their families, although he acknowledged that some information came to him from others.

  36. Describing the gang, Sergeant Rollin referred to it as “known to be involved in criminal acts, which will encompass acts of violence as well as weapons offences”, by which he said he meant bladed instruments and firearms. Shortly afterwards he appeared to relate this to inter-gang violence and the defence of territory. The judge stopped further questioning about weapons known or believed to be in the possession of the gang. Although Mr Rollin’s witness statement had referred to the gang being active in drug sales, this evidence was not adduced.

    Admissibility: relevance, bad character and propensity

  37. The starting point is that evidence is not admissible unless it is relevant. It is relevant if, but only if, it contributes something to the resolution of one or more of the issues in the case. It may do so either directly or indirectly.

  38. The second important proposition is that not all relevant evidence is admissible. At common law, relevant Crown evidence falls to be excluded if, in the judgment of the trial judge, its admission will be unfair to the defendant, in the sense, as it is conventionally put, that its prejudicial effect exceeds its probative value: see Noor Mohammed v The King [1949] AC 182, 192. This rule is now given statutory effect in most common law jurisdictions and is to be found in Bermuda in section 93(1) of the Police and Criminal Evidence Act 2006:

    In any proceedings the court may refuse to allow evidence on which the prosecution proposes to rely to be given if it appears to the court that, having regard to all the circumstances, including the circumstances in which the evidence was obtained, the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it.

    Although often referred to in submissions and elsewhere as conferring a discretion on the judge, this rule imports more accurately an exercise in judgment, rather than one of indulgence. It is, however, plainly true, that on many facts it may be possible, and wholly legitimate, for different judges to arrive at different judgments on the assessment of the balance between probative value and unfairness which is central to the statutory question.

  39. Whilst the situations in which Noor Mohammed/section 93 may fall for decision are infinitely variable, the common law confronted one frequently recurring circumstance as long ago as Makin v Attorney General for New South Wales [1894] AC 57, namely evidence which shows that the defendant has been responsible for some criminal offence, or for some reprehensible behaviour, other than that charged. Such evidence cannot be said to be irrelevant. Evidence which shows that the defendant has a propensity to offend or behave badly may well be very relevant. But it is normally to be excluded on grounds of fairness, unless there is some reason to admit it beyond mere propensity. Lord Herschell’s famous formulation of the principle is, as Lord Hailsham of St Marylebone subsequently explained in R v Boardman [1975] AC 421 at 451, an application of the unfairness rule in due course to be adumbrated in Noor Mohammed. Lord Herschell said this, at p 65:

    It is undoubtedly not competent for the prosecution to adduce evidence tending to shew that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried.

    On the other hand, the mere fact that the evidence adduced tends to shew the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused.

    The statement of these general principles is easy, but it is obvious that it may often be very difficult to draw the line and to decide whether a particular piece of evidence is on the one side or the other.

    The first sentence encapsulates the general rule that mere propensity to behave badly is to be excluded as unfair. The second states the principle that if evidence of bad behaviour has probative value beyond mere propensity to misbehave it may be admissible, and it gives examples, not in any sense exhaustive, of when it may have such value. The third states an important truth.

  40. In jurisdictions, of which Bermuda is one, where there is no statutory modification of the common law to allow the admission of propensity alone in defined circumstances (such as sections 98-113 of the Criminal Justice Act 2003 do in England and Wales), the general rule in Makin stands. Mere propensity to behave badly is to be excluded as unfair. Admission requires justification beyond such mere propensity. An example of such justification is so-called similar fact evidence (which was in question in Boardman, and see now DPP v P [1991] 2 AC 447); in such a case the justification arises because the evidence is sufficiently compelling to have real value in controverting innocent coincidence. Another example is the kind of case where there has been a course of violent dispute between the defendant and the victim; there the evidence may be admissible (inter alia) to show either who was responsible for the last (charged) occasion, or the intention with which the defendant acted on that occasion, or to explain the reactions of the two parties. Likewise, in a case of alleged sexual abuse, the history and nature of a relationship said to have been abusive will often be relevant to proving a particular incident charged, even though it also shows prior misbehaviour by the defendant. It is impossible to catalogue every situation in which such justification may be present. But unless it is, evidence of misbehaviour unconnected with the offence charged is not admissible. As Lord Hailsham explained in Boardman at 453F, relying on mere propensity as evidence of guilt is an inadmissible chain of reasoning. If the inadmissible chain of reasoning is the only purpose for which the evidence is to be adduced, the evidence is inadmissible. If there is some other justification for the admission of the evidence, the jury usually needs to be warned not to pursue the inadmissible chain of reasoning.

  41. Moreover, to respect the general rule, where such justification does exist the evidence which is admitted ought normally to be restricted to that which is within the justification. The justification is the measure of the admissibility. The existence of such justification does not generally create “open season” to adduce any evidence of the defendant’s bad character or misbehaviour.

  42. Gang evidence, such as was tendered in the present three cases, will almost always involve implications of bad behaviour other than that charged on the part of those labelled gang members, and often the defendant. It is of high importance that in such cases the decision as to admissibility starts with Makin and the limitation just set out in para 41.


  43. In a case of murder or attempted murder, as in most criminal cases, evidence of motive is relevant but not necessary. Often the Crown may be able to prove what happened, and who did it, without knowing why. But where there is evidence that the defendant had a motive to kill the victim, that goes to support the case that it was him, rather than someone else, and/or that he did it with murderous intent, rather than accidentally or without intent to do at least grievous bodily harm. It may equally be relevant to rebut asserted self-defence or provocation. Admissible evidence of motive may sometimes necessarily involve showing bad behaviour by the defendant on occasions other than that charged. If that is the case, this is an example of the second sentence of Lord Herschell’s principle in Makin; that the evidence relevantly proves motive may be a justification for its admission notwithstanding that it also shows bad behaviour.

  44. In the cases of Myers and of Cox, the murderous intention of the gunmen could not be, and was not, in dispute, so the evidence of motive did not go to that issue. But the evidence that there existed a feud between gangs was relevant to identity, which was the core issue in dispute. It went to show that those two defendants had a motive to kill the victims. It showed that they were members of a group which was likely to have felt aggrieved and, moreover, to have reacted by targeting the deceased on grounds of his membership of the opposing association. In each case, the evidence contributed to the proposition that it was the defendant who had done it, by supporting the other evidence that it was he who was responsible.

  45. As will become clear, the Board shares some of the concerns voiced by Auld JA in the appeals of Myers and Cox as to the dangers of gang evidence. But it does not agree with his proposition, on which those two appellants rely, that in order to be admissible motive must be harboured uniquely by the defendant. Evidence of a shared motive can be just as relevant. In the case, for example, of a feud between neighbouring families, the motive may well be shared by several members of a family, of whom the defendant is one, but it is still relevant to show that he had a reason to do what is alleged. It does not become irrelevant simply because others had the same motive.

  46. The fallacy in this submission made on behalf of Myers and Cox is the implication that the motive evidence must prove the case against the defendant all by itself before it can be admitted. Even if motive may occasionally do this, much more often it is but one strand in a case, together with either circumstantial or eye-witness evidence. In these two cases there was both, and the shared motive supported the other evidence. It thus makes it more probable that the defendant was responsible. So to say is not, with great respect to Auld JA’s judgment in Myers’ case, in any way to water down the rule that the guilt of a defendant in a criminal trial must be proved to the criminal standard and not simply as a matter of probability. The rule that the jury must be satisfied so that it is sure on all the evidence does not mean that no single piece of evidence is admissible unless it is capable by itself of discharging that burden. Nor does R v Tirnaveanu [2007] EWCA Crim 1239, [2007] 1 WLR 3049, to which Auld JA referred, say anything different. It was certainly a case in which the issue was whether it had been the defendant or some other person who had committed the offences (of smuggling illegal immigrants) and the disputed evidence, of a similar fact/specific propensity kind, went to this issue, and fell to be considered under the new bad character rules in the English Criminal Justice Act 2003. But the observation by Thomas LJ, as he then was, that the evidence went to whether the culprit was the defendant or someone else, cannot be read as carrying the implication that in all cases it is irrelevant to show that the defendant is one of a relatively small group of people with either the motive or the opportunity to commit it.

  47. Put another way, the evidence in these two cases rebutted the argument “Why on earth should this defendant, who has no proven connection with, or dispute with, the deceased, have taken it into his head to shoot him?”.

  48. For the same reasons, the ballistic evidence of the connection between the gun(s) used in these offences and in other shootings which could be inferred to have been committed in pursuit of gang feuds was admissible evidence in support of the motive attributed by the Crown to the defendants. Further it contributed to the conclusion that the weapons were accessible to gangs of which the defendants were members, and thus to them.

  49. In both these cases, it was an important strand in the rope of evidence that there had occurred a trigger event which would have created a grievance in the gang of which the defendant was a member. If that is a necessary part of the Crown case, it is for the Crown to prove the trigger event. This was properly done in Myers’ case by adducing evidence from someone who had seen what happened to Neika Daily at the Devonshire Recreation Club and who could say that the dispute had been with David Cox. That logically laid the ground for the evidence that Daily was central to one gang and Cox a member of the other. However, in the later case where Cox himself was the defendant, the Crown failed to prove the alleged trigger event at the Mid Atlantic Boat Club, except to the fortuitous extent that the defendant’s police interviews did so (see para 16 above). The evidence of DC Edmonson was rank hearsay. That something may well turn out to be not in serious dispute is no reason to adduce inadmissible evidence of it. It is necessary to draw attention to this, because it may indicate that the Crown is not always scrupulous to avoid hearsay evidence. Where gang evidence is adduced, careful attention to the rule against hearsay is likely to be necessary: see below.

  50. Brangman’s case was not put as one of reprisal in the course of a gang feud, nor did the Crown advance the case as depending on a plainly identifiable trigger event. Nathan Darrell’s witness statement had disclaimed gang warfare or a turf war as the reason for his shooting. Whilst of course if there really has been a gang turf war members of both gangs may deny the fact, the Crown did not put the case that way. The gang evidence of Sergeant Rollin became admissible because of the cross examination which suggested that Wellman was likely to have been responsible for the attack. True it is that Darrell had in any event given evidence of hostility between himself and Brangman, and that that arguably provided Brangman with a motive in any event. But that did not mean that it was not legitimate for the Crown to counter the specific possibility advanced that Wellman might have had the stronger motive, by demonstrating that if any shooting at Wellman’s house was indeed the cause of the attack on Darrell, then Brangman had, via the gang association, as powerful a motive as Wellman had. The Board concludes that the judge was entitled to come to the conclusion that this evidence was admissible in consequence of the line taken in cross examination. Certain it is that, before admitting it, the judge needed to consider whether it would be unfair to admit it, having regard to the level of its probative value and the potentially damaging effect of evidence of the defendant’s gang association. But in doing so, he was entitled to have regard to the extent to which the conduct of the defence had, no doubt advisedly, involved introducing Darrell’s familiarity with firearms and his public challenge to (in particular) Parkside, which reference was unlikely to be lost on a Bermudian jury. Also relevant was the introduction by the defence of the fact that the gun appeared to be a shared one; that made it relevant that the defendant was a member of a group such as might do the sharing. It may be that if the only factor had been the Wellman cross examination, some judges might have concluded that the prejudicial effect of the gang evidence exceeded its probative value, but it was not.

    “Explanatory evidence”: R v Pettman

  51. A further example of justification within Makin for the admission of evidence which shows a defendant’s bad behaviour or propensity may be afforded where the evidence is relevant to proof of the charge, and the bad behaviour unavoidably comes with it. A simple example is a trial of an allegation of violence between prisoners; the fact that the defendant is in prison will unavoidably emerge. R v Pettman (unreported) in the Court of Appeal (Criminal Division) in England and Wales, 2 May 1985, is often cited as a statement of this principle. Purchas LJ put it in this way:

    where it is necessary to place before the jury evidence of part of a continual background or history relevant to the offence charged in the indictment, and without the totality of which the account placed before the jury would be incomplete or incomprehensible, then the fact that the whole account involves including evidence establishing the commission of an offence with which the accused is not charged is not of itself a ground for excluding the evidence.

    However, the kinds of case touched on in para 40 above, where a course of conduct or the history of a relationship is relevant to proof of offences charged, may also sometimes be analysed in these terms. Examples include R v Sawoniuk [2000] 2 Cr App R 220 at 234; R v Williams (Clarence) (1987) 84 Cr App R 299 and R v Underwood [1999] Crim LR 227. In either case Pettman is an example of the principle set out above, namely that departure from Makin must be justified.

  52. The Pettman proposition, valid as it is, needs cautious handling if it is not to become a token excuse for admitting the inadmissible. Claims by prosecutors that the evidence is necessary to understanding of the case, or, as is sometimes asserted, to discourage the jury from wondering about the context in which the events discussed occurred, need to be scrutinised with care. It is only where the evidence truly adds something, beyond mere propensity, which may assist the jury to resolve one or more issues in the case, or is the unavoidable incident of admissible material, as distinct from interesting background or context, that the justification exists for overriding the normal Makin prohibition on proof of bad behaviour. Moreover, admissibility is subject to the power to exclude under Noor Mohammed or, now, section 93.

  53. The facts of Pettman itself are an illustration. The Crown case was that the defendants were armed robbers. The charges were of conspiracies to rob. Some raids were said to have been executed and others to have remained in the planning. In relation to a planned wages delivery robbery in London which was not in fact carried out, apparently because the security van did not arrive, there was observation evidence suggesting that the robbers were on hand to execute it on three Fridays and were using a particular motor car. The Crown was permitted to call evidence that Pettman had been followed in that same car, the night before the first Friday, down to Brighton where he and an accomplice had burgled a house. Moreover, on the second Friday the users of the car had bought petrol using a credit card stolen from the house in Brighton. The burglary was not before the jury, but with other similar offences was the subject of a separate indictment, to be tried later if necessary. The evidence of the defendant’s connections with the car was clearly admissible, and important, on the indictment for conspiracy to rob. The fact that that evidence also proved the commission of the burglary was, in the circumstances, an unavoidable incidental consequence, and did not render it inadmissible. Given the use of the credit card, it would not have been practically possible for the evidence of the Brighton observation to be given without including the burglary. The argument in the appeal was confined to whether the judge erred in not excluding the observation evidence on the Noor Mohammed (now section 93) principle, and it was held that he had not. The Board would observe that, but for the use of the credit card, it might well have been possible to prove the Brighton observation, and thus the use of the car by Pettman, without including the evidence of the burglary, and that this kind of limitation upon incidental evidence of bad behaviour, where it does not involve objectionable artificiality or risk incomprehensibility, is both customary and correct. Of course, even in that event, it would have been unavoidable for the jury to learn that police officers were keeping the defendant under observation. Such is an oft-encountered example of the unavoidable incidental effects of admissible evidence and is normally accompanied by a suitably neutral warning by the judge tailored to the facts, for example that the jury should avoid speculation and/or should not assume that any suspicions apparently harboured by the police were accurate.

  54. In the present three cases the justification for the admission of the gang evidence was not Pettman or any residual principle that unavoidable incidental evidence of bad behaviour may be admissible. In the present cases, the gang evidence was admissible not because it was incidental to evidence otherwise admissible, but because in itself it significantly advanced the Crown case by demonstrating motive. If it had not done this, it would not have been admissible under Pettman, for it would then simply have been gratuitous evidence that the defendants were members of criminal gangs, and would have proved nothing relevant to justify departure from the general rule in Makin.

  55. The dissenting judgments of Auld JA, in particular in the case of Cox, helpfully highlight the manner in which over-easy invocation of expressions such as “necessity” or “explanatory context” can lead to the unfair admission of evidence in reliance on Pettman. The same warning was sounded by the English Court of Appeal in R v Dolan [2003] EWCA Crim 1859, [2003] 1 Cr App R 281 at p 285, where evidence of a tendency to bad temper with inanimate objects was not admissible as similar fact on a charge of murder of a child, nor was there then any basis on which mere propensity could be adduced, and it followed that Pettman could not be used to “smuggle in” the inadmissible. It is necessary to emphasise that the basic rule in Makin is the starting point, that departure from it requires the evidence in question to be justified as proving something on which the jury may legitimately rely to resolve one or more issues in the case, and that in any event it is not to be admitted unless in all the circumstances of the case this can be done without unfairness to the defendant.

    Ambit of gang evidence

  56. It follows from the principles set out above that the ambit of gang evidence will depend, in any particular case, on what legitimate role it may have in helping the jury to resolve one or more issues in the case. It is not possible to lay down general rules for gang evidence beyond that. However, it follows also that the measure of admissibility is the extent to which the evidence justifies departure from the starting point of Makin. Some of the evidence in the three cases before the Board was not thus justified. The evidence adduced from Sergeant Rollin that the gangs were involved in drug trafficking (see paras 10 and 23 above in relation to Myers and Cox respectively) was simply irrelevant, and offended the Makin rule. It was relevant that there was a violent feud between the gangs, but it was irrelevant whether their activities were otherwise lawful or not, as it was whether the cause of their mutual animosity was drug rivalry or football rivalry. Likewise, it was relevant for Sergeant Rollin to give evidence that the rival gangs were in the habit of wreaking serious violence or death upon each other, but insofar as he went further, as he did in the case of Cox, and gave evidence that they committed other offences of violence, that failed the Makin test: see para 21 above. Nor is it clear, when it was adduced from him that Myers had never been in employment (para 10 above), whether this had any legitimate justification or was no more than an unspoken invitation to the jury to assume that he had an illegitimate source of income. This is important to record, even though in none of these cases are the convictions in the end unsafe (see below). The ruling of the judge in the case of Myers, in which he specifically admitted these pieces of additional evidence (para 12 above), was wrong.

    Police Officers as expert witnesses

  57. Police officers have been accepted as expert witnesses in several different contexts and in many common law jurisdictions. A simple example is the police officer who has special training (and considerable experience) in the investigation and reconstruction of road traffic accidents. He may well be as much an expert in his field as any consulting accident engineer, and he is accepted as such routinely in both civil and criminal trials: see for example R v Oakley (1980) 70 Cr App R 7, where Lord Widgery CJ held in emphatic terms, at p 9:

    … we would like to make it quite clear straight away that there is no question of a police officer being prevented from giving evidence as an expert if the subject in which he is giving evidence as an expert is a subject in which he has expert knowledge, and if it is restricted and directed to the issues in the case.

    In such a case the expertise in question is as to the behaviour of motor cars, the effects of road surfaces and the like. In other cases, however, a police officer has been permitted to give expert evidence about criminal behaviour. An example is evidence of the customary practices of drug users, in relation to such matters as packaging, methods and quantities of usage and supply, and prevailing price: see R v Hodges [2003] EWCA Crim 290, [2003] 2 Cr App R 247, in which this type of evidence was held admissible. Evidence of the practices, mores and associations of gangs, whether general or particular, is in a similar category. It has been received in several jurisdictions and there can in principle be no objection to it being given by a police officer, providing that the ordinary threshold requirements for expertise are established, and providing that the ordinary rules as to the giving of expert evidence are observed.

  58. Those two provisos, however, are of some importance. It is not necessary here to traverse the different expressions which have been used in different jurisdictions to describe the threshold of expertise which must be demonstrated (such as the two contrasting tests propounded in different US states in Daubert v Merrell Dow 509 US 579 (1993) and Frye v United States 293 F 1013 (1923) or the Canadian analysis in R v Mohan (1994) 89 CCC (3d) 402). It is enough to refer to R v Rangzieb Ahmed [2011] EWCA Crim 184, paras 56-57, essentially adopting R v Bonython [1984] SASR 45, subject to the refinement set out in R v Dallagher [2002] EWCA Crim 1903, [2003] 1 Cr App R 195, para 28. The particular issues which may arise when a new scientific theory is advanced do not arise here. But the officer must have made a sufficient study, whether by formal training or through practical experience, to assemble what can properly be regarded as a balanced body of specialised knowledge which would not be available to the tribunal of fact. The police witness in Oakley had passed an examination and had attended and analysed thousands of road accidents. The officer in Hodges had analysed prices over a period of 12 years and checked every drugs file in his area. In both cases the necessary expertise was easily established. But care must be taken that simple, and not necessarily balanced, anecdotal experience is not permitted to assume the robe of expertise. An example is given by the Canadian case of Sekhon v The Queen [2014] 1 SCR 272, where in the course of otherwise perfectly permissible expert evidence as to drug practices, there was adduced from the officer the fact that he had never encountered an innocent courier. That was clearly not a balanced, tested or researched proposition as to the methods of drug importers, but simply his personal experience. It was not admissible and indeed proved nothing about the particular defendant on trial.

  59. Secondly, whilst a police officer may be an expert, by training or experience or both, if he is then he comes under the same duties to the court as does any other expert. Those duties were helpfully set out by Cresswell J in the commercial case of The Ikarian Reefer [1993] 2 Lloyds Rep 68 at 81 but apply equally in the criminal context: R v Harris [2005] EWCA Crim 1980, [2006] 1 Cr App R 55, paras 271-272, where they were summarised as including the following:


    Expert evidence presented to the court should be and seen to be the independent product of the expert uninfluenced as to form or content by the exigencies of litigation.


    An expert witness should provide independent assistance to the court by way of objective unbiased opinion in relation to matters within his expertise. An expert witness in the High Court should never assume the role of advocate.


    An expert witness should state the facts or assumptions on which his opinion is based. He should not omit to consider material facts which detract from his concluded opinions.


    An expert should make it clear when a particular question or issue falls outside his expertise.


    If an expert’s opinion is not properly researched because he considers that insufficient data is available then this must be stated with an indication that the opinion is no more than a provisional one.


    If after exchange of reports, an expert witness changes his view on material matters, such change of view should be communicated to the other side without delay and when appropriate to the court.

  60. Compliance with these exacting standards can be difficult for a police officer who is effectively combining the duties of active investigator (if not of the current case) with those of independent expert. It is particularly important that such a witness should fully understand that once he is tendered as an expert he is not simply a part of the prosecution team, but has a separate duty to the court to give independent evidence, whichever side it may favour. In particular a police expert needs to be especially conscious of the duty to state fully any material which weighs against any proposition which he is advancing, as well as all the evidence on which he has based that proposition. When considering an application by the Crown to adduce the evidence of a police expert, it is incumbent upon the judge to satisfy himself that these duties are recognised, and discharged.

  61. In the present cases a good deal of Sergeant Rollin’s evidence was plainly based on his own personal observation. His status as an expert was not necessary to the admissibility of that evidence. But his interpretation of the signs shown in photographs or of Cox’s tattoo was clearly a matter of expertise based on a study of the gangs generally. Similarly, his evidence of the gang culture of reprisal taken against any member of the rival organisation was grounded in expertise, whilst the evidence of territories, the history of violent gang conflict and the associations between individuals was a mixture of expertise and personal experience. The Board is satisfied that he sufficiently demonstrated expertise in these areas to give evidence about them.


  62. Insofar as Sergeant Rollin gave evidence as an expert, the question arises how far such a witness is affected by the rules of evidence as to hearsay.

  63. It is well established that an expert is entitled, in giving his evidence, to draw upon the general body of knowledge and understanding in which he is expert, notwithstanding that some (or even all) of the material may have been assembled by other students of the subject. So in R v Abadom [1983] 1 WLR 126 a forensic scientist had compared the refractive index of glass fragments found on the shoes of the defendant with control samples of broken glass from the crime scene. He was held entitled to give evidence not only that the two matched, but also that the match was unlikely to be a random innocent match. The latter proposition depended on records of other examinations of glass submitted to forensic science laboratories up and down the country and upon statistical analysis of the frequency with which the refractive index in question had been encountered. Those other examinations, and the statistical analysis, had been carried out by others, frequently in different laboratories. Similarly, in English Exporters (London) Ltd v Eldonwall Ltd [1973] Ch, 415, Megarry J held that an expert valuer was entitled to inform his valuation not only by his own personal experience of transactions but also by such materials as textbooks, journals, auction reports, exchanges with co-professionals and the like, some related to particular transactions and others more general. He might have added files on other transactions unknown to him personally but kept by others in his office. In a case very close to the present ones, the Court of Criminal Appeal for South Australia adopted the same approach in R v Cluse [2014] 120 SASR 268 in holding that a police officer in a similar position to Sergeant Rollin was entitled in giving evidence of gang culture, characteristics and rivalry to draw upon information gathered from a number of sources: see paras 48-49 and the approval of the similar approach to evidence of drug use and culture at paras 5-6.

  64. It does not, however, follow that because the witness is an expert he is immune from all inhibition on hearsay. He is not. In each of the three cases cited in para 63 a similar point was made. In Abadom it was made clear that the scientist could not give hearsay evidence of the comparison between the glass from the defendant’s shoes and from the scene of the crime; first hand evidence of that was required. In Eldonwall Megarry J refused to allow the valuer to prove the precise terms of transactions advanced as comparables without first hand evidence of them. And in Cluse, the court drew the line at evidence from the officer of particular incidents of violence between the two gangs; that was held to require first hand evidence, or (so at least it was said) evidence from an officer who had himself investigated the incident. Thus all these cases show that there is a limit to the extent to which an expert may advance hearsay material.

  65. In some cases the dividing line is between opinion evidence, which may be informed by hearsay information (admissible) and specific evidence of observable fact, which has to be proved in a manner which satisfies the ordinary rules of evidence. That was the distinction drawn by Megarry J in Eldonwall and it is readily applicable in the case of a valuer, since his evidence of value is necessarily opinion. In Cluse Kourakis CJ referred (at para 2) to the same distinction between opinion and fact. But experts often give evidence of observable fact and such evidence may legitimately be, and very often is, informed by the accumulated body of knowledge collected by others as well as by the witness’ own experience. The expert in the social and political conditions in a foreign country, who is a witness before an immigration judge gives evidence of fact, and much of it will not be his personal experience. The consultant radiologist who says that a brain scan shows a particular type of haemorrhage is giving evidence of fact, and so is the electronics engineer who says how a complex piece of apparatus works, but it may well be that their evidence relies on the work of researchers and on published data, rather than on their first-hand knowledge. The forensic scientist in Abadom was not simply giving it as his opinion that the match was unlikely to be innocent. He was telling the jury that only 4% of glass examined in the laboratories shared its index and that was evidence of fact entirely gleaned from others; moreover he was giving evidence which effectively asserted the accuracy of the measurement of the indices in all those cases. The police officer’s evidence in Cluse about the existence, culture and rivalries of particular gangs was not evidence of opinion, but could nevertheless be given on the basis both of personal experience and data collected by others.

  66. The test of whether evidence based upon hearsay material can be given is better seen to be whether it ceases to be the expounding of general study (whether by the witness or others) and becomes the assertion of a particular fact in issue in the case. The first is expert evidence, grounded on a body of learning or study; the second is not, even if it may be given by someone who is also an expert. The line between the two is case-specific, but it will usually be possible to discern it. Moreover, it may well be that the application of section 93 will yield the same answer to the question whether the witness ought to be permitted to give second-hand evidence.

  67. In the present cases the evidence of Sergeant Rollin appears to have been very largely based on his own observations. To the extent that his evidence of places of association, the culture of the gangs, or the signs which they used was supported by the observations of others, that appears to have been general evidence based on the accumulated information collected by his unit, from a multitude of sources, and legitimately given. If in another case the assertion that a particular person, and especially the defendant, was a member of a gang, were to depend on particular sightings of him in the company of other known members, that might in some instances pass the point at which it was no longer a matter of general study or accumulated knowledge, but required proof according to the ordinary rules. That is one reason why it is essential that a witness such as Sergeant Rollin sets out from the beginning the sources on which he has relied: see below. Moreover, if there is a challenge to one or more particular observations of a defendant, fairness is likely to require that the first-hand evidence of them is called, so that they can properly be tested. In such a case, section 93 would be likely to lead to the exclusion of non-direct evidence in any event. Evidence of a specific alleged trigger event or events is another instance of something which is not part of a general body of learning, but specific to the case; hearsay evidence is not admissible, and the case of Cox affords an example: see paras 16 and 49 above. No doubt if an assertion of fact falls on the particular, rather than the general, side of the line, it might be provable by admissible hearsay under section 75 or 76 of the Police and Criminal Evidence Act, if the conditions required by those provisions are met.

    The presentation of “gang evidence”

  68. In its judgment in the case of Cox, the Court of Appeal offered at para 40 some valuable advice as to the presentation of evidence of this kind. The Board endorses this helpful approach, and would expand a little upon it. As part of the duty of an expert witness to the court, a police officer tendering the kind of evidence called in these cases must make full disclosure of the nature of his material. His duty involves at least the following.

    1. He must set out his qualifications to give expert evidence, by training and experience.

    2. He must state not only his conclusions but also how he has arrived at them; if they are based on his own observations or contacts with particular persons, he must say so; if they are based on information provided by other officers he must show how it is collected and exchanged and, if recorded, how; if they are based on informers, he must at least acknowledge that such is one source, although of course he need not name them.

    3. In relation to primary conclusions in relation to the defendant or other key persons, he must go beyond a mere general statement that he has sources of kinds A, B and C, but must say whence the particular information he is advancing has come; an example would be observations of a defendant in the company of others known to be members of a gang.

  69. If a witness statement tendered for a trial does not meet these standards, the judge can be asked to direct that it be expanded in whatever particulars he judges necessary. Such an application should not be left until the beginning of the trial but should be made well beforehand. If directions given are not complied with, that will be relevant both to whether the witness has established a proper basis for giving expert evidence and to whether his evidence ought to be excluded under section 93.

  70. In the present cases, Sergeant Rollin’s witness statements adequately established his credentials and qualifications. In several respects they correctly made it clear that the witness had had personal dealings with the defendants and others of whom he spoke and what they had amounted to. But they contained a number of bare assertions, unsupported by the basis for them, such as “I consider X to be a member of the B gang”, or “The known areas used by the C gang are …”. He did not sufficiently distinguish between assertions based on his own observations and contacts and those to which others had contributed. The defence must not be left to explore what the sources were in speculative cross examination before the jury. Nor must the jury be left with over-generalised assertions, such as that “very little” of the evidence given derives from other people.

  71. In the case of Myers, Sergeant Rollin’s witness statements did not disclose the evidence which he could give (and did give, in chief) as to the gang affiliations of the people who had also been shot with the guns used in the murder of Harford (see para 11 above). No complaint was made about it at the time, although leading counsel for Myers remarked on it in passing in cross examination, and it may thus be that sufficient informal notice had been given. But this kind of evidence may be, and was in this case, of no little significance. It ought to be signalled in advance and the basis for each assertion shown.

    Section 93: Police and Criminal Evidence Act

  72. As is apparent from the foregoing, gang evidence of the kind tendered in these cases should be tested under section 93 on each occasion it is advanced. Gratuitous assertions as to the illegal activities of a gang, such as drug trafficking, are inadmissible, for the reasons given above, and the application of section 93 does not arise. But it is necessary for the judge in every case to look carefully at the overall effect of gang evidence and to reach a judgment as to the balance between legitimate probative value and unfair prejudicial effect. When assessing that balance, a highly relevant consideration is the ability of the defendant to test the evidence. It is likely to be unfair for the witness to state a bald conclusion such as “I consider X to be a member of the M gang”. The defence cannot be expected to embark upon speculative cross-examination as to the basis for such a conclusion, at the risk of inadvertently eliciting either inadmissible or unfair information.

    Conclusions: Myers

  73. The gang evidence of culture, feud and the defendant’s membership was admissible to show motive. Sergeant Rollin was qualified to give it. His evidence did not detail as clearly as it should have done which part depended on sightings or observations by others and which on his own personal experience, nor did he give advance notice, in his witness statement at least, of the evidence which he was to give as to the gang affiliations of the other victims of the guns used. But the evidence of Myers’ membership was cogent, including the photographs and Sergeant Rollin’s personal knowledge of him which involved seeing him most days over a period of years. By the end of the case leading counsel for Myers realistically made the express concession before the jury that the defendant was a member of the gang. No objection was taken to the oral evidence of the affiliations of the other victims. That the guns used had been used in multiple shootings was also admissible, since the inference was justified that they were weapons shared by the gang of which Myers was a member and that he could thus be expected to have access to them; this ought however to have been left as an inference for the jury to draw rather than being adduced as an assertion from DC Burgess (see para 11 above). The gang evidence went beyond the admissible in asserting the drug trafficking habits of the gang collectively. But the conduct of the defence inevitably put before the jury the drug trafficking habits not only of the Crown witnesses Darrell and Laws, but also of the defendant, and this last was expressly put by counsel to Sergeant Rollin in cross examination as having been not simply a matter of gang association, but of personal supplying. In those circumstances the admission of the inadmissible gang evidence of drug dealing had no significant impact on the trial. In any event the unanimous conclusion of the Court of Appeal that no miscarriage of justice occurred was plainly justified, given the powerful other evidence. The Board will accordingly humbly advise Her Majesty that Myers’ appeal against conviction ought to be dismissed.

    Conclusions: Cox

  74. The gang evidence of culture, feud and the defendant’s membership was admissible as in the case of Myers, and again Sergeant Rollin was qualified to give it. In this case his evidence appears to have been based on his own observations (see paras 17 and 20), and was supported by the cogent photograph of the tattoo. The evidence of the multiple use of the guns, in the area of the opposing gang, was likewise admissible, although once again the derived proposition that the gang appeared to share weapons ought to have been a matter of inference for the jury rather than of evidence by DC Burgess (see para 22 above). The gang evidence went beyond the admissible in the assertion of habitual drug trafficking by the gang, and, to the extent that this is what was said, in the assertion of habitual violence other than in the course of the feud. The impact of that excessive evidence was, however, on the facts of Cox’s case, likely to have been relatively slight, and in any event the other evidence linking him to the crime plainly justified the unanimous decision of the Court of Appeal that no miscarriage of justice had occurred. In his case also, the Board’s humble advice to Her Majesty will be that his appeal against conviction should be dismissed.

    Conclusions: Brangman

  75. The judge was entitled to reach the conclusion that the cross examination justified the addition of Sergeant Rollin’s evidence to the Crown case (see para 50 above). The evidence went to the nature of the association with Wellman, rather than to gang feud, but it was nevertheless for that reason relevant to motive. Rollin was qualified to give the evidence he did and was not disabled from doing so by his position as a police officer. To the extent that he relied on information from other officers, or informants, he was entitled to do so; that was not the source of his evidence of Brangman’s membership of the gang but contributed only to the general background formed by the accumulated body of knowledge about the gang. The photograph was cogent evidence of membership and did not depend on hearsay. The evidence of generally nefarious gang activity was marginal, was restrained by the judge, and added nothing significant to what had been adduced in cross examination. The evidence of gang membership was of course damagingly prejudicial, but given the course of the (no doubt advisedly undertaken) cross examination by counsel then appearing for Brangman the judge was entitled to hold that the balance of fairness did not lead to the exclusion of the evidence. In his case also, the Board will humbly advise that his appeal against conviction ought to be dismissed.


John Perry QC & Elizabeth Christopher (instructed by Simons Muirhead & Burton) for appellant, Myers.

Joel Bennathan QC (instructed by Saunders Law LLP) for appellant, Brangman.

John Perry QC & Larry Mussenden (instructed by Simons Muirhead & Burton) for appellant, Cox.

Howard Stevens QC, Rory Field, Director of Public Prosecutions, Carrington Mahoney, Deputy DPP Cindy Clarke, Deputy DPP (instructed by Charles Russell Speechlys LLP) for respondent.

all rights reserved