Yaryare & Ors v R.

On 10 May 2018 at Leicester Crown Court before Judge Mooncey and a jury, Liban Yaryare, Khalid Hassan and Yahaya Osman were convicted of the various offences described below, for which they were sentenced on 12 October 2018.

Liban Yaryare (now aged 24) was sentenced on count 1 (violent disorder at the High Cross Shopping Centre, contrary to section 2 Public Order Act 1986) to 12 months’ imprisonment; on count 2 (conspiracy to commit violent disorder at Lower Brown Street, contrary to section 1 Criminal Law Act 1977) to 12 months’ imprisonment, to be served consecutively to the sentence on count 8; on count 6 (a like conspiracy to count 2, an offence at the Bede Park) to 2 years’ imprisonment; and on count 8 (attempted murder) to 13 years 6 months’ imprisonment. The total sentence, therefore, was 14 years 6 months’ imprisonment, to be served consecutively to a sentence he was already serving of 3 years’ imprisonment. On 25 January 2019, following an Attorney General’s Reference ([2019] EWCA Crim 78), the sentence on count 8 (attempted murder) was increased to 17 years 6 months’ imprisonment. The other sentences were unchanged and the sentence on count 2 remained a consecutive sentence to that imposed on count 8. The total sentence, therefore, following the Reference, was 18 years 6 months’ imprisonment (to run consecutively to the sentence of 3 years).

Khalid Hassan (now aged 26) was convicted on counts 6 and 8 (respectively conspiracy to commit violent disorder at Bede Park and attempted murder) and he was sentenced to 2 years’ imprisonment on count 6 and 13 years’ imprisonment, to be served concurrently, on count 8.

Yahaya Osman (now aged 24) was convicted on counts 2 and 6 (conspiracy to commit violent disorder at Lower Brown Street and Bede Park respectively), and he was sentenced to 1 year’s imprisonment on count 2 and two years’ imprisonment, to be served consecutively, on count 6. As a result of the commission of these offences, the appellant was in breach of a Suspended Sentence Order of 12 months’ imprisonment for robbery, which was activated in full. His total sentence, therefore, was 4 years’ imprisonment.

They stood trial with various co-accused, who, as relevant, are referred to in the summary of the facts below.

Yaryare appeals against his conviction by leave of the single judge, limited to ground 1 (the judge’s refusal to exclude the recognition evidence of DC Bee pursuant to section 78 Police and Criminal Evidence Act 1984 (“section 78”)) and ground 2 (the judge’s refusal to direct the jury as to the dangers identified by the Court of Appeal in R v Smith and Others [2008] EWCA Crim 1342; (2009) 1 Cr App R 36 and to correct a suggested false impression created by DC Bee as to the lack of an obligation to take notes). He renews his application for leave to appeal against conviction on ground 3 (the submission of no case to answer). On 31 March 2020 the full court refused Yaryare’s renewed application in respect of ground 4 in which he sought to leave to adduce fresh evidence ([2020] EWCA Crim 661).

Hassan appeals against conviction by leave of the single judge on ground 1 (which mirrors ground 1 as advanced by Yaryare). He renews his application for leave to appeal against conviction on ground 2 (the submission of no case to answer).

Osman appeals against conviction by leave of the single judge, limited to grounds 1 and 2 (which mirror grounds 1 and 2 as advanced by Yaryare). He renews his application for leave to appeal against conviction on ground 3 (the admission of bad character evidence).

In summary, therefore, the single judge granted leave on two grounds: first, whether the judge erred in admitting the recognition evidence of DC Emma Bee and, second, whether the judge erred in failing to direct the jury as to the dangers identified by the Court of Appeal in R v Smith and in failing to correct a suggested false impression created by DC Bee, to the effect that there was no obligation to take notes in a recognition case such as the present. During the hearing of the appeal this latter ground was, with the leave of the court, argued by all three appellants, it having not been included as one of Hassan’s grounds of appeal.

Additionally, Yaryare and Hassan renew their applications for leave to argue that the judge should have upheld their submissions of no case to answer and Osman renews his application for leave to argue that the judge erred in admitting bad character evidence.

During Thursday 30 April 2015 there were a number of violent altercations between two opposing groups (identified at trial as Group 1 (“Green Group”) and Group 2 (“Blue Group”)), which culminated in the attempted murder of Gideon Buabeng, who was stabbed 11 times in Bede Park,

Group 1/Green Group consisted of men mainly of Somali descent, amongst whom were Liban Yaryare, Abdulhakim Yassin, Khalid Hassan, Abass Hudur, Soubane Ismail, Anwar Mire, Khalid Abdirahim, Yahaya Osman, Ahmed Abdullahi, Sakariye Garane and Lydon D’Costa.

Group 2/Blue Group consisted of men of either African or Asian heritage and included the twin brothers, Alfred and Wilfred Yimbo, the brothers Ahmad Siyar Hakimi and Ahmad Sayam Hakimi, Aslam Sakha, Vahees Nagulesparan, Abdullah Imran, Jermain Felix, Goncalo Teixeira, Joel AduBosompem, Mozes Junior Banjo, Mahad Ahmed and Anthony Ekundayo.

It is necessary to explain the structure of the indictment. It contained 8 counts, which reflected the entirety of the violence:

The first event, therefore, related to a violent disorder at the High Cross

The precise cause of the disturbance remains unknown. CCTV footage showed a confrontation between the two groups which culminated in a brief but intensely violent fight. Security staff intervened and separated those involved. The members of Group 1 came off worse and some of them made contact with their associates via their mobile telephones, and shortly afterwards other young men arrived. By that time Group 2 had left the scene.

Following the events at the High Cross Shopping Centre, call data evidence suggests there was contact between the two groups and arrangements were made to meet. These exchanges, in the main, involved Joel Adu-Bosompem, Goncalo Teixeira and Ahmed Abdullahi. An attempt by members of the two groups to gather at Bede Park at about 6.30 pm failed, albeit CCTV footage showed some of Group 1 in the area of Bede Park when two vehicles arrived at the scene. The first was a dark blue series 5 BMW containing Liban Yaryare, Abass Hudur, Sharmarke Yaryare and two unidentified males. Shortly afterwards, a grey BMW series 1 arrived, driven by Sakariye Garane with Abdulhakim Yassin and Lydon D’Costa as passengers. They had planned to meet with Goncalo Teixeira who, for reasons that are unclear, failed to appear at the time arranged. Instead, he was caught on CCTV footage with Anthony Ekundayo, Joel Adu-Bosompem, Gideon Buabeng and two unidentified men walking back towards Buabeng’s house.

There was further contact between the two groups and a meeting was arranged once members of Group 1 started to threaten the safety of Goncalo Teixeira’s mother. At about 7.20 pm the groups came together on Lower Brown Street in Leicester. By this time each group had grown in number and there were, overall, in excess of 30 people now involved. Call data confirmed that Goncalo Teixeira made a call at 6.59 pm to Jermain Felix to request the help of him and his friends, all of whom had been involved in the earlier incident at the High Cross Shopping Centre.

Elizabeth Rutherford saw Mahad Ahmed and Goncalo Teixeira pass a large kitchen knife between them as she stood in the street waiting for a friend. There were a number of calls to the police. Officers attended but the defendants and their associates left, some on foot and some in cars. No one was detained. Among those present from Group 2 were Anthony Ekundayo, Goncalo Teixeira, Joel Adu-Bosompem, Mahad Ahmed, Mozes Junior Banjo, Jermain Felix, Alfred Yimbo, Wilfred Yimbo, Ahmad Siyar Hakimi, Ahmad Sayam Hakimi, Aslam Sakha, Vahees Nagulesparan and Abdullah Imran. From Group 1 there were Liban Yaryare, Abdulhakim Yassin, Abass Hudur, Yahaya Osman, Sakariye Garane, Ahmed Abdullahi, Lydon D’Costa, Soubane Ismail and Anwar Mire.

As regards counts 6 – 8, from about 8.00 pm members of Group 1 started to congregate in Bede Park. There were a number of vehicles at the scene, some of which were not stationary. There were a number of calls to the police, starting at 9.01 pm. A police vehicle drove along Briton Street and along Western Road near Tarragon Road. The officers then left the area.

Buabeng was called by his friend, Joel Adu-Bosompem. Adu-Bosompem asked him if he could join them to help him sort out a problem but he was not given any details. Call data showed that Group 1 had planned to meet members of Group 2. The members of Group 2 that attended Bede Park at this time were Anthony Ekundayo, Goncalo Teixeira, Joel Adu-Bosompem, Mahad Ahmed, Mozes Banjo and two unidentified males. Upon arrival they were met by a large group of Somalian males. They initially retreated, but Teixeira made the decision that they had to go back into Bede Park to sort out the problem. As they walked into the Park, more Somalian males joined them, with the result that there were 30-40 Somalian males in total.

Liban Yaryare, Abdulhakim Yassin, Khalid Hassan, Abass Hudur, Yahaya Osman, Sakariye Garane, Khalid Abdirahim, Ahmed Abdullahi, Soubane Ismail and Anwar Mire, all from Group 1, were in Bede Park. Call data showed that the two defendants most involved in arranging the meeting were Ahmed Abdullahi (from Group 1) and Ekundayo (from Group 2). A large number of those in Group 1 remain unidentified.

Buabeng testified that as they got nearer to the opposing group on Bede Park it was clear they were brandishing weapons, and as a result they decided to leave. Although the members of his group ran away from the Park back over the bridge, towards Briton Street, Buabeng ran into the Park chased by members of Group 1, who were brandishing weapons. He was pushed to the ground and attacked with multiple weapons by many individuals. He received 11 stab wounds to his body.

At trial, in addition to the testimony of Buabeng, the prosecution relied principally on four strands of evidence. First, given each of the incidents was captured by CCTV recordings, recognition evidence from the officer in the case, DC Emma Bee. Second, evidence from a facial mapping expert, William Platts, which provided support for DC Bee’s identifications. Third, evidence showing the movements of some of the relevant mobile telephones and the details of contact between some of those involved. Fourth, various exhibits seized from the respective defendants.

The defence of these three appellants was that they had been incorrectly identified by DC Bee.

The prosecution sought, therefore, to rely upon the evidence of DC Bee who had recognised a number of suspects from the CCTV footage of these incidents, which she stated she had studied for many hours (“hundreds/thousands”). From the outset the appellants accepted that DC Bee, as a matter of law, was entitled – subject always to the particular circumstances of the case – to give evidence as an expert on the CCTV footage. This included giving evidence as to those she recognised as allegedly participating in the relevant offences.

However, there were substantive objections to the admissibility of this evidence, founded on the circumstances of the recognition. Most notably, it was highlighted that DC Bee first provided a witness statement setting out why she had arrived at her conclusions as to the identification of the suspects considerably after she commenced viewing the CCTV footage and she had failed to maintain a satisfactory log.

In her first relevant statement dated 25 May 2017 (served on 2 June 2017), DC Bee indicated:

will provide a more detailed explanatory statement regarding how I have identified the defendants on the footage, and how I compiled the CCTV compilations in due course.”

On 1 August 2017, she dealt with the issue in more detail:

have viewed this footage hundreds of times since starting this investigation, over many hours. Some of the footage I have watched over and over again. It is difficult to quantify the hours that I have watched the footage over the last 2 years, but I have watched the footage hundreds of times in detail, scrutinising the various people featured on the footage and looking at their actions and their movements. As a consequence of this viewing I have become familiar with the people it features and I am able to recognise and identify the various people as they move from one piece of footage to another.

She annexed photographs, inter alia, of the individuals who she said matched these three appellants from the CCTV footage.

Written applications were served to exclude her evidence, and this led to a further short statement from DC Bee dated 26 February 2018 in which she explained that over the course of the investigation she had completed 21 investigation workbooks and 3 interviewing workbooks. She exhibited extracts from this material that related to her consideration of the CCTV footage (served on 6 March 2018).

On 1 March 2018 (three days after the trial had commenced), she completed a further statement in which she set out in greater detail the grounds for her identifications (also served on 6 March 2018). She rehearsed that she had been involved in the investigation for nearly 3 years as the sole investigating officer. She had viewed the CCTV on multiple occasions and, working with experts, she had compiled an edited CCTV compilation of each of the defendants, which she continually reviewed. She said that it was difficult to quantify the exact number of hours she had spent watching the footage, but it amounted, at least, to hundreds of separate occasions. She considered each relevant individual, focussing on their appearance, action and movements. She became familiar with those featured and was able to recognise some but not all of those involved as they moved from one section of footage to another. Additionally, her work with the telephone analyst and interviewing the suspects (which involved reviewing the footage), resulted in further detailed consideration of the footage.

She obtained photographs and CCTV footage of the appellant Yaryare from a number of different sources, including from his Facebook page, CCTV footage of him in the front enquiry office at Mansfield House Police station and footage whilst he was in custody. She worked from a compilation of the still photographs from which she had identified Yaryare at the separate incidents. As already indicated, she interviewed Yaryare, and took him in detail through the relevant parts of the CCTV footage. She explained her reasons for identifying him, namely by reason of height, age, ethnicity, walk and posture (“stands with his shoulders rounded and his hips forward”), hairline, shape of his lips, ear position and skin colour. Subject A had been wearing the same clothing in all the footage. In evidence DC Bee referred also to the shape of his eyes.

She interviewed Hassan , but he was not shown the footage during this process, as he had not been identified by that stage. There is no CCTV evidence placing Hassan at the Lower Brown Street incident. She was in possession of photographs and CCTV footage of Hassan from a number of sources, including from a Facebook page (“Kh Al id”) and his Cash Convertors account. DC Bee looked at footage of Hassan at the front desk enquiry office at Beaumont Leys Police Station and footage of him whilst in custody. She worked from a selection of still photographs from which she identified Hassan as being present at the incidents, and she provided a compilation of the relevant CCTV footage. She noted his height (approximately 5’ 10”), that he was of medium to large build, with “wider hips and posterior”. DC Bee took account of the similarities between Hassan and subject C, as regards age, ethnicity, and skin tone. She remarked on the similar distinctive features of shape, build and posture which were reflected in the way the appellant and the suspect moved and their style of walking. When standing, the two appeared to “lean” on one side and their walk is described as a prominent “waddle”. Against that background, DC Bee suggested that when comparing these features with subject C in the Highcross and Bede Park footage, the “stance and walking style have distinct similarities”. DC Bee observed that on the Highcross footage, notwithstanding the fact that the suspect is wearing a hood, the viewer is afforded “glances” of his face, which reveal the same shaped mouth as Hassan, with dark shadowing on the upper lip. She opined that this is likely to be dark short facial hair, the same as seen on Hassan. For the Bede Park footage, subject C had been wearing a hood, but on occasions the viewer got glimpses of his hair and hairline as the hood moved back. Again, this matched Hassan’s hair and hairline. The clothing was identical throughout the footage. Finally, there were strong other links between Hassan and some of his co accused.

DC Bee interviewed Osman and she charged him when he answered to his bail. She looked at Yaryare’s Facebook page entitled “Liban Ya”, where there were photographs of him along with Osman. She also obtained a photograph of him from his Cash Converters account. As with the other two appellants, she reviewed CCTV footage of Osman at the police station (albeit in relation to a different investigation). She created a compilation video. Osman was approximately 6’ tall. He and suspect J shared the same slim build, they were of the same age group and ethnicity, and had the same skin tone. They each had a prominent hooked nose, which was indistinguishable as between the two men. Throughout the relevant footage of the incidents, J was wearing the same top with a blue stripe down the shoulders and the arms (the only difference being that on the Bede Park footage he was also wearing a body warmer). Osman attended Keysham Lane Police Station wearing a distinctly similar garment. DC Bee concluded that Osman and J were the same person, and that the top worn at the police station was “extremely similar” to the one viewed in the footage.

As set out above, the defence were served with DC Bee’s workbooks which contained her contemporaneous notes or jotting concerning aspects of the investigation. They have the semblance of hasty entries, that are far from complete. They appear to represent a handwritten record of particular matters that attracted the officer’s attention or which for some reason were thought to require a brief record. They did not contain any record as to why, at any particular stage in this process, the officer concluded a particular individual shown on the CCTV footage was one of the defendants, and they did not amount to contemporaneous notes made by her when watching the CCTV. She indicated in cross-examination that she relied to a significant extent on her memory (“it’s all up here”).

The applications to exclude the evidence of DC Bee pursuant to section 78 were made on 26 March 2018.

Before analysing the judge’s approach to this issue, it is necessary to have in mind that this court has previously considered the circumstances when an identification takes place broadly in line with the procedure adopted by DC Bee. In AG Ref. No.2 of 2002 [2002] EWCA Crim 2373; [2003] 1 Cr App R 21 the court at [19] stated as follows:

A police officer asked to view a CCTV is not in the same shoes as a witness asked to identify someone he has seen committing a crime. But, as the prosecution accepted, safeguards which the code is designed to put in place are equally important in cases where a police officer is asked to see whether he can recognise anyone in a CCTV recording. The mischief is that a police officer may merely assert that he recognised someone without any objective means of testing the accuracy of such an assertion. Whether or not Code D applies, there must be in place some record which assists in gauging the reliability of the assertion. In cases such as these, there is no possibility of comparing the initial observation of a witness, as recorded in a contemporaneous note of description or absence of description, who purports to make a subsequent identification. The police officer can hardly be asked to record his recollection of a description of a particular suspect before he has picked that suspect out from the CCTV recording.

Absent any such check as would be available had a witness described the commission of an offence and recollected his description of the offender, it is important that the police officer's initial reactions to the recording are set out and available for scrutiny. Thus if the police officer fails to recognise anyone on first viewing but does so subsequently those circumstances should be noted. The words that officer uses by way of recognition may also be of importance. If an officer fails to pick anybody else out that also should be recorded, as should any words of doubt. Furthermore, it is necessary that if recognition takes place a record is made of what it is about the image that is said to have triggered the recognition.

Absent any such record, it will not be possible to assess the reliability of the recognition. We were told that a protocol is being prepared for such cases. With the increasing use of CCTV recognition it is vital that a protocol is prepared which provides the safeguard of measuring the recognition against an objective standard of assessment. Only by such means can there be any assurance that the officer is not merely asserting that which he wishes and hopes, however subconsciously, to achieve, namely the recognition of a guilty participant.”

Whether the person knew or was given information concerning the name or identity of any suspect.

What the person has been told before the viewing about the offence, the person(s) depicted in the images or the offender and by whom.

How and by whom the witness was asked to view the image or look at the individual.

Whether the viewing was alone or with others and if with others, the reason for it.

The arrangements under which the person viewed the film or saw the individual and by whom those arrangements were made.

Whether the viewing of any images was arranged as part of a mass circulation to police and the public or for selected persons.

The date time and place images were viewed or further viewed or the individual was seen.

The times between which the images were viewed or the individual was seen.

How the viewing of images or sighting of the individual was controlled and by whom.

Whether the person was familiar with the location shown in any images or the place where they saw the individual and if so, why.

Whether or not on this occasion, the person claims to recognise any image shown, or any individual seen, as being someone known to them, and if they do:

the reason

the words of recognition

any expressions of doubt

what features of the image or the individual triggered the recognition.

the person who views the image or sees the individual and makes the recognition.

the officer or police staff in charge of showing the images to the person or in charge of the conditions under which the person sees the individual.”

In paragraph D:3.35 it is stated that, so far as possible, the showing should also “follow the principles for video identification if the suspect is known, see Annex A”. Annex A states at paragraphs D10 and D11:

Finally, in the notes for guidance, at D:3A it was provided that save for one exception (immaterial for these purposes), “a police officer who is a witness for the purposes of this part of the code is subject to the same principles and procedures as a civilian witness”.

The Association of the Chief Police Officers’ Practice Advice 2011 on the use of CCTV in criminal investigations, supports the instruction that viewing logs are always to be kept (section 6.6, page 48):

Document what has been seen in the footage;

Describe the actions of individuals (especially victims and suspects) in a neutral manner.

Furthermore, the Advice (appendix 2, page 82) summarises the relevant part of the judgment in Smith as follows:

Any initial reactions to seeing the CCTV images;

Where a police officer fails to recognise anyone on the initial viewing but does so at a later date;

Where a police officer fails to recognise anyone at all;

Anything that an officer may say with regard to any doubt;

Where there is recognition, any factors relating to the image that caused that recognition to occur.

On this issue, the judge ruled as follows:

It is unarguable that DC Bee has “acquired special knowledge” as envisaged in the authorities cited. In that regard the authorities do not have a useful role in this application. Essentially, the defence concern is that DC Bee has not made viewing notes that are contemporaneous, spontaneous, setting out her reasonings and analysis of what she is watching. She has not articulated her thoughts and responses as she made her observations. The defence say this deprives them potentially of materials on which to base cross examination of her reliability.

The prosecution submit that there is more than ample material that has been provided by DC Bee about her work in this case. The quality of her work and the reliability and matters of that ilk are matters that can be dealt with in the trial process. If it is felt she is lacking in any part of her role, the “trial process” provides for those issues to be dealt with. It is an issue that goes to the weight of the evidence and therefore it is materials that should be left for consideration by the jury with the usual directions that would be given.

Having carefully considered the matters, I am of the view that DC Bee’s evidence can be introduced. Overall, I find favour with the submissions made by the prosecution. […] the authorities cited all find favour with the prosecution being able to call witness whose familiarity with materials allows them to give “expert” evidence of this nature.

Except for the case of SMITH, none of the authorities deal with the issue raised by the defence about the nature of the notes that should or must be made by a witness in the position of DC Bee. In Smith, at paragraph 67, the court identifies the mischief that has to be guarded against, namely “that a police officer may merely assert that he recognised someone without any objective means of testing the accuracy of such an assertion”. The court made reference to say there “must” be in place some record of what was done. Apart from observing that that case had features which are wholly different to the present case – here there are work books and statements clarifying her work. It is to be noted by paragraph 73, the court in SMITH observed the evidence was not confined to that evidence of the purported identification and recognition and therefore did not think the verdict was unsafe. In this case, the jury will have additional materials to consider too.

Even if it is accepted that the case of Smith led to changes in the guidance offered to the investigators, it is to be noted that it is still “advice” and at 6.6 of the ACPO guidance, it advises that a viewing log “should” be kept and a higher requirement than that is not made. Finally, from a practical level, it may be impossible to note every thought that one has when watching a piece of video footage. In the presentation of this case, the prosecution have played clips a number of times so that what is viewed becomes familiar. More viewings reveals more details. It is difficult for instance to say it was on say my third viewing that I noticed a particular feature of the recording. What the defence seek may be counsel of perfection.

For those reasons, I conclude, DC Bee’s evidence is admissible. The mischief identified by the court in previous cases can easily be met with guidance given to the jury in the usual way. Appropriate directions can be given to the jury. The defence can have a fair trial.”

At trial the judge gave the jury the following relevant directions in law:

This ground of appeal is argued on behalf of the appellants on the basis that, although it is accepted that an officer could give evidence of the kind provided by DC Bee, there were fundamental difficulties with her evidence. It is contended she had insufficiently explained why she had identified the various suspects. She did not at any stage indicate when or why she first recognised any of the appellants; instead, as late as 1 March 2018 she set out the significant individual characteristics that she claimed she had observed in relation to each of them. It is highlighted that her statement of 1 August 2017 was completed more than 2 years after the incident, and the later statements were at even greater remove from the incident. It is complained that it is unclear whether this was an “identification” or a “recognition” case.

It is suggested that Code D clearly applies and that there should have been a contemporaneous written record to protect against an officer merely asserting that he or she recognised an individual without any objective means of testing the accuracy of the assertion. It is complained that there is no record of any spontaneous or genuine reaction to the CCTV footage, which could have been recorded in viewing logs (which in the instant case are said in any event to have been wholly inadequate in their content). The appellants underscore, therefore, that the logs failed to a) record any initial reactions to viewing the CCTV images; b) explain the developing position as regards recognition; c) record any areas of doubt; and d) identify the image or images (the still or stills) that led to the recognition. The appellants contend that if the procedure had been properly followed, it would have been possible to follow, over time, the key moments in the process of recognition.

Against that background, the appellants argue that the judge was wrong to rule that the steps they suggested should have been followed by DC Bee amounted to a “counsel of perfection” (see [45] above); instead, the judge should have determined that the prosecution had acted substantively in breach of the requirements of Smith and Code D, as analysed below.

Turning to the discrete submissions advanced by the three appellants, Mr Bhatia Q.C., on behalf of Yaryare, highlights that two other officers who had viewed a still from the CCTV footage had suggested to DC Bee that subject A might be Yaryare, prior to her viewing the footage. Yaryare emphasises that he challenged the suggestion that he was subject A during the interviews, pointing out at that stage certain suggested differences between his appearance and that of the suspect. It was submitted on his behalf that there was a marked absence of other evidence implicating him. The clothes recovered from his home address did not match those worn by subject A and there was a lack of any relevant cell site evidence.

Hassan emphasises that subject C, at an early stage of the investigation, had been identified as a man called Sade Ibrahim who was interviewed and who shared some physical characteristics in common with the Hassan. DC Bee positively recognised Hassan as subject C on 25 November 2016, although she had made the link before he was interviewed. It is not entirely clear what prompted this significant change of recognition from Ibrahim to Hassan, albeit she stated she was certain it was the latter of the two men. Mr Murphy, on behalf of Hassan, submitted that the record, as a minimum, should have recorded when the officer viewed the CCTV footage and thereafter her reaction when she saw Hassan in interview. Otherwise, it should have included the kind of material included in the 1 March 2018 statement ([32] above).

Mr Witcher, on behalf of Osman, submits the log, inter alia, should have reflected i) the dates when the footage was viewed; ii) the exact sections of the footage then under consideration; iii) a summary/review of the critical developments in the officer’s assessment (e.g. when she made a link with one of the subjects and any suspect); iv) the particular features, as noted by the officer, of the appearance of the subject and the suspect whenever something emerged that affected her appraisal; and v) the moment when she made a link between a subject and a suspect. Mr Witcher emphasises the extent to which DC Bee was uncertain as to how long she had spent viewing this footage.

Mr Jarvis, on behalf of the respondent, submits that this was not a case “where Code D strictly applied”. The basis for this submission is that DC Bee was not instructed to sit down to view the CCTV footage in order to identify the suspects. Instead, she was tasked to follow the movements of the individuals involved in these offences, given some of them could be seen at one or more of the various locations. During this process of tracking their whereabouts, she started to realise that she might be able to recognise some of them. It is highlighted that in certain instances her ability to do so only emerged over time, whilst watching the relevant footage, whereas in other situations she recognised a suspect as soon as she saw them in person. However, it is accepted by Mr Jarvis that although Code D, as then formulated, did not – as he contends – strictly require DC Bee to make an immediate note of her recognition, on the basis of a broad view of the jurisprudence

In conclusion, Mr Jarvis submits that the key question is whether DC Bee’s recognition of the suspect was reliable and whether the jury were in a position to test the reliability of the recognition during the course of the trial.

Following the judge’s ruling concerning DC Bee, the appellants submitted the judge should direct the jury in accordance with the relevant part of the decision in Smith (see [39] above) and, on a linked issue, that he should correct a suggested false impression in DC Bee’s evidence, when she suggested that the Police and Criminal Evidence Act Code D did not apply to her identifications and that she was not obliged to keep notes.

The judge declined to take either of these steps. Instead, he directed the jury as set out at [46] above.

In essence, Mr Bhatia submits that the judge should have directed the jury in accordance with Smith, and in particular that the law requires that a record is kept of i) any initial reactions to seeing the CCTV images; ii) any failure initially to remark on someone who is recognised at a later stage; iii) any failure by an officer to recognise a particular individual at any stage; iv) any expressions of doubt as regards an identification; and v) the factors that led to a recognition. Otherwise, Mr Bhatia submitted that in accordance with Smith at [72] the judge should have directed the jury that by reason of the inadequacies in the procedure and the record of recognition, there was no objective standard or record against which to measure the assertions made by DC Bee.

The appellants rely on the following passage in the judgment of this court in R v Fergus (1994) 98 Cr App R 313 at 318:

It is submitted that these observations highlight the need for the judge to explain in detail the potential failures highlighted in Smith.

Mr Jarvis contends that the passages from the summing up, cited above, demonstrate that the judge directed the jury as to the matters that they needed to bear in mind as regards the identification evidence (e.g. the need for caution and a direction that confident witnesses can be mistaken), save that he failed to indicate that the law requires that a sufficient log should have been kept, broadly in accordance with the decision in Smith. We are reminded that the judge gave a full summary of DC Bee’s evidence, including the criticisms made of it.

On the basis that there was no other evidence against Yaryare, it is submitted that if DC Bee’s evidence should have been excluded, then as an inevitable consequence the submission of no case to answer ought to have been upheld at the close of the prosecution case. Without DC Bee, there was no case against Yaryare on which the jury could be sure.

Mr Jarvis submits that the case against Yaryare was strong, based on the extent of the recognition evidence. DC Bee had had extensive opportunities to study his appearance and there were a number of sightings of person A on the footage which provided a clear basis for making comparisons. Furthermore, Mr Platts, the imaging analyst, found “strong support” for the suggestion that Yaryare and person A were the same individual.

Additionally, there was unchallenged evidence from the security staff at Leicester College (Mullins and Bains) that the appellant had an association with a number of the offenders, and in particular Hudur, Felix, Teixeira, Yimbo (Wilfred), Yimbo (Alfred), Hakimi (Siyar), and Hakimi (Sayam). There was also evidence from his mobile phone contacts list that he had close contact with the offenders D’Costa, Osman, Abdullahi, Ismail, and Mire. Finally, he lied in his interviews under caution about his friendship and association with Hudur. In fact Hudur had been with the appellant at the time of the offences and he was convicted by the jury.

The Crown accept there were points that could properly be made on Yaryare’s behalf. Mr Zjalic, the defence imaging analyst, found there was only “some support” for the conclusion that they are the same person. A jacket with the distinctive markings seen on person A was not found either at the appellant’s address when it was searched two months later or amongst the images on his Facebook account. Furthermore, there is no relevant cell site evidence indicating the use of his mobile telephone at the scene of these events. However, it is contended that these were all points properly for evaluation by the jury, and there was a clearly sufficient basis for him to be convicted as charged, given the unequivocal evidence of DC Bee, supported by Mr Platts.

The appellant relies on the criticisms of DC Bee set out above, with the result that the evidence should have been excluded or, alternatively, treated as inherently weak. Mr Platts’ conclusions on the imagery provided no more than limited support for the proposition that Hassan and subject C were the same person. Otherwise, the evidence against Hassan is said to have been extremely limited, namely i) three unanswered telephone calls that involved his telephone and the telephones of other defendants at times relevant to these events; ii) a bank paying-in slip dated 6 months after the incident bearing Hassan’s name and account details which was found in a codefendant’s grey BMW (that of Sakariye Garane, a member of Group 1, who was driving one of the convoy vehicles), which was said to show continued association with a member of a group of individuals who were alleged to have been involved in these offences; and iii) his having driven a white Audi 9 months after the offences that was connected with the offences in question.

Mr Jarvis relies on the evidence of DC Bee, but he also stresses the potential significance of the three telephone calls. Different members of Group 1 tried to contact Hassan, and of these the last two calls were particularly noteworthy given they were from individuals involved in the High Cross Shopping Centre incident (Abdirahim and Hudur).

At the conclusion of the cross-examination of DC Bee by Mr Witcher, the prosecution foreshadowed an intention to apply to adduce Osman’s bad character under section 101 (1) (g) Criminal Justice Act 2003 (“CJA 2003”):

The convictions relied on by the Crown in this regard were:

16 June 2012: possessing an offensive weapon in a public place (a formal reprimand, when a youth). He took a knuckleduster to school.

23 September 2014: battery (a conviction). After being sent off in a football match, he headbutted a match official and then spat at him and at another official.

24 April 2015: robbery (a conviction). At a hotel, with three other men carried out a knifepoint robbery of two hotel guests, stealing their mobile phones

Initially, this application was unopposed, albeit in due course Mr. Witcher resisted the application. The basis for the application, and the judge’s reasons for admitting the evidence, were set out in his ruling on the issue:

There is evidence before the jury that the police keep records of their investigations in various sources, Crime Logs, Custody Records and the like. DC Bee has also kept her own “Workbooks” of which there are some 20 plus. These Workbooks as with other materials have been the subject to review for disclosure including by Independent Counsel not instructed to appear as an advocate in this case. The defence had been given extract of some of the Workbooks under the Disclosure principles. Essentially, DC Bee’s position is that: I have spent a very long time looking at the footages and materials, I have not kept a precise log as to how many hours but when I look back over the 3 years, “I would say it’s more accurate to say 1000s not 100s of hours spent watching this material” [she said in cross-examination by Mr. Bhatia for Liban YARYARE]. She readily accepts she used the phrase “hundreds of hours” in her statements. DC Bee is steadfast in saying that in using these expressions, she has not lied or exaggerated the amount of work she has put into this case by way of examining the imagery.

Mr. Witcher explored this issue further in his cross-examination and when nearing the end of his cross-examination, the following exchanges took place:

The prosecution submits that the cross examination by Mr. Witcher went so far enough to engage the Bad Character provisions. The prosecution asked me to consider the manner and tone of the cross examination but I do not regard that as being of significance. I did detect some exasperation on the witness part but nothing in my view turns on that. However, I do regard the content of the exchange as very significant. In my view:

to say: “you are anything but independent” equates logically to saying you are partial or biased.

to say: “you have invested a lot in this case and you want a result” suggests that her motivation is to simply

to say: “you have been at pains to exaggerate how many hours you have spent working on the footage because before you stepped into the witness box, you had done 100’s of hours…it becomes 1000’s when you know you are about to be challenged about how good you are at your job? is suggestive of deliberately misleading (at pains to exaggerate). This is followed by her motivation being that she fears her professionalism is being challenged – “good at your job”.

Cumulatively, the exchange does suggest she has not spent that long on viewing the footages to gain expertise she claims to have, that she is cavalier about her identification of Osman, that she is intent on getting a conviction and keeping her reputation intact. That in my view clearly demonstrates an attack on DC Bee’s character. This went far beyond suggesting simply that she was mistaken. The questioning was clear as to what impact it would have. The suggestion by Mr. Witcher that he can in the presence of the withdraw the question about “at pains” in my view would not resolve matters.

In arriving at my decision, I have considered the issue of how many hours DC Bee says she has spent on viewing footages. As a matter of record, she has used expression hundreds and then thousands. She has never specified exactly how many hours she has spent on viewing the footages. Whether it is an exaggeration, deliberate or otherwise, or whether it is just an endeavour by her to convey she that she has over 3 years watched the footages so many times as to lose track on measuring it will be a matter for the jury – logically hundreds and hundreds does then take the measurement in to the thousands.

In arriving at my conclusion, I have considered as a whole the cross examination of DC Bee. I have also considered any adverse effect in can have on the fairness of the trial. The questioning I have referred to above leaves me with no choice than to accede to the prosecution’s application. My ruling is that the Bad Character evidence in relation to Yahaya OSMAN can go before the jury.”

The judge directed the jury as follows on this issue:

On behalf of Osman, it is submitted that he had not alleged that DC Bee’s identification had been brought about by a dishonest intent. Instead, it is suggested the defence had explored whether DC Bee was i) a mistaken witness, who ii) lacked independence, and was iii) unable to recognise subconscious bias on her part, and it was suggested she had a natural and understandable desire to be right. It was accepted, however, that in the robust way in which these points had been developed in questioning there had been an attack on the officer’s integrity. Critically, however, Mr Witcher contends that the evidence should have been excluded under section 101 (3) CJA 2003:

The central submission is that “the evidence of the bad character of the defendant had little if any bearing on his credibility in being the maker of the attack” and that there was no benefit to the proceedings in admitting this evidence, and instead only manifest prejudice. It is suggested that the appellant was not positing that he “knew” she either lacked independence or was unable to recognise that she had a desire to be right or that she was subconsciously biased. It is contended there was nothing coming from the defence that required an assessment of the appellant’s credibility. Instead, the jury needed to assess the credibility of DC Bee.

In the respondent’s notice, it is suggested that the appellant, through his counsel, made a number of serious attacks on the character of DC Bee by implying that she had lied on oath during the trial, that she had deliberately exaggerated the true position in the course of her evidence and had acted in bad faith in order to secure the conviction of the applicants. On that basis, the judge was correct in not excluding this evidence.

Counsel have relied on six authorities, in addition to those considered above, which have potential bearing on this appeal, and which are summarised below.

In R v Abnett (Gary) [2006] EWCA Crim 3320, the officer in the case viewed the relevant CCTV footage and a number of still images. On the basis of particular features and characteristics he determined that the appellant and the suspect were the same person. The court concluded:

In R v Chaney [2009] EWCA Crim 21; [2009] 1 Cr App R 35, as summarised in the judgment at [14] and [21] – [25], an officer who knew the defendant viewed CCTV stills, and on the basis particularly of his hairline, facial features and stature, concluded that the man in the stills and the defendant were the same. The court had the officer’s initial reactions to the images (he indicated in an email that he had identified the defendant). It was highlighted that the only justifiable complaints were, first, that it had been suggested to the officer in advance of the viewing that the man in the CCTV may be the defendant; second, the officer failed to identify at the time which features of the man depicted in the photographs led him to the conclusion that it was the appellant; and, third, the judge failed to give any warning in the summing up as to the dangers of identification evidence. However, as the court observed, the relevant evidence in this regard was before the jury, and they were able to consider the images when evaluating the reliability of the officer’s evidence. The images were relatively clear and sufficiently so for it to have been open to the jury to conclude that the officer’s identification was accurate. The court was seemingly unconcerned that the officer did not at the time identify the features in the photograph that led to his identification of the defendant. On this issue, the court observed at [24]:

Although this was a case of recognition rather than identification, the court concluded it would have been appropriate for the judge, in his summing up, to have referred to the need for caution, although a

In R v Henry McGrath [2009] EWCA Crim 1758 this court upheld a conviction based substantially on an officer’s recognition of the accused from CCTV footage, when there was no contemporaneous record of the procedure during which the officer made her recognition (no notes were made at the time) and the officer had prior knowledge of the appellant as a suspect [15]. The judge had given a clear and full direction to the jury on the issue.

In R v Moss [2011] EWCA Crim 252 the officer who viewed the CCTV in a relatively informal setting made only the most cursory record of his identification at the time in his pocketbook, simply stating he was sure that the perpetrator was the accused. 6 months later he made a statement setting out in somewhat greater detail the basis of his recognition. This court stated as follows:

In our view in a case of this kind it is important not to lose sight of the essential principles. PC Osmond reported the recognition to his superior, he made a note of it when he returned to work, he gave a description of his previous contact with the appellant, and thus an explanation of his ability to recognise him, all of which provided a basis on which the jury could judge the reliability of his evidence.

The judge had a discretion to exclude the evidence of PC Osmond but declined to do so. In his summing-up he gave the jury a clear warning of the dangers inherent in recognition evidence and of the risk that an honest witness who says that he recognised someone known to him may nonetheless be mistaken. He also gave them a clear warning of the need for special caution and of the danger of a witness digging his heels in when challenged. In our view the judge was right to allow the evidence to be given. He directed the jury correctly in relation to it and this ground of appeal fails.”

R v JD (supra at [40]) was another case in which the case against the accused depended entirely on an officer having recognised him when viewing the CCTV footage of the incident. There were the following uncontested features relied on by the appellant:

The identifying officer had visited the appellant at his home (between 2002 and 2006, when the incident occurred in 2011) and he had interviewed him for a number of minutes on two occasions in 2010. He looked at the CCTV three times.

Against that background, the court concluded:

The court recognised that whether the breaches of the Code in this context (including asking the officer whether he could pick a particular named individual) will render a verdict unsafe, will turn, for instance, on the whether there was other evidence against the appellant (see [27] and [28]).

In R v Lariba and others [2015] EWCA Crim 478, it was held that although there had been breaches of Code D as regards recognition evidence by police officers viewing CCTV footage (such as the absence of a contemporaneous recording of the reason for the identification, the words used, any expressions

We highlight that the following conclusions, relevant to this ground of appeal, are to be drawn from this jurisprudence.

First, at the time of this investigation, following, inter alia, the decisions in Smith, JD and Chaney, the germane provisions of Code D as then in force should have been followed (whether or not Code D was directly applicable, given there is some tension in the authorities on this issue), and particularly as regards creating a contemporaneous record that would have assisted in testing DC Bee’s recognition evidence. The current version of Code D (in force since 23 February 2017) deals with a closely related situation, in that D:3.34 (“Recognition by controlled showing of films, photographs and images”) is introduced with the words:

Second, whether a failure to follow Code D renders the verdict unsafe will depend on the particular facts of the case, and the court will need to consider the extent and significance of any breaches of the Code and any consequential unfairness that have been caused (see JD at [28]).

Third, although the impact of the breach or breaches of Code will, therefore, vary between cases, two notable strands are to be discerned from the authorities. On the one hand, there are cases such as Smith and JD in which no contemporaneous record was kept and the recognition evidence was inherently poor. In Smith the recognition was based on no more than his stature and his clothing “it’s everything, it’s not one particular thing, it’s the whole really” but not including recognition of his face (see the judgment at [64] and [65]). In JD, the officer who suggested he recognised the appellant gave no details as to what features led him to this conclusion, and instead simply stated that he was in no doubt that the man in the green T-shirt was the appellant (see the judgment at [7]) having viewed the footage 3 times. On the basis that the evidence should have been excluded, the conviction in Smith would have been quashed had there not been additional material implicating the appellant and in JD the conviction was quashed. On the other hand, in cases such as Chaney and Lariba, notwithstanding the failure to apply Code D (including in Chaney promptings by other officers that the defendant may be in the stills or CCTV footage), if a detailed explanation is given of the basis for the recognition, particularly when the jury is in a position to view the relevant material itself, it may – depending always on other factors – be fair to admit the recognition evidence.

In the present case, during her evidence DC Bee set out that, whilst viewing the footage over the course of hundreds or thousands of hours, she observed the finer details of the movements and other relevant characteristics of those involved, and over time she became familiar with some of the people who feature in this case. Further, she began to recognise some, but not all, of them. She viewed the footage on a variety of different types of equipment, from ordinary laptops to “high tech machines”. All of the CCTV material and the visual imagery evidence relied on by DC Bee was exhibited and adduced at trial, and the jury would have been very familiar with the images – both stills and moving images – which were said by DC Bee to provide the link between

We consider there is some force in the judge’s observation at the end of his ruling ([45] above), that from a practical point of view it may be unrealistic to expect an officer to note all of his or her passing thoughts whilst watching CCTV footage time and again. Any conclusions in a case such as the present are likely to emerge incrementally, and the fine detail of an improving or changing recognition may be difficult to record in a log. However, that said, the officer should record, in accordance with the approach established in Code D, at the least, the “Red Letter” events as described by Mr Jarvis – the moments, for instance, when they first begin to note similarities with a particular individual, along with any significant features that occur to them during the process of viewing. They should also note any factors that tend to indicate the suspect does not match a particular individual who is being considered. Otherwise, whether or not it is directly or entirely applicable in this particular context, D:3.34 – 37 of the present Code, which match the provisions then in force, sets out the steps (as relevant) that should be followed.

For each of the appellants, DC Bee explained the reasons for her recognition of them by reference to the relevant features of their appearance, and the jury, as in Lariba, were able, from the footage and the stills, to see exactly what the witness saw from these permanent images. Moreover, given the relative good quality of the images, they were able to assess whether a recognition made from them was one on which they could rely even though they were not making an identification of their own. They would have been well aware for Yaryare that other officers had suggested he may be one of the suspects and that for Hassan the recognition changed from Ibrahim to Hassan. It is to be regretted that DC Bee failed to apply the relevant parts of Code D and the guidance given in Smith, and we accept that the appellants and the jury were put at a disadvantage as a result, but given the extent of the opportunities to test and assess the reliability of the recognition evidence, and in the particular circumstances of this case, we consider that this expert evidence which it is accepted was prima facie admissible did not fall to be excluded under section 78. The evidence did not have such an adverse effect on the fairness of the proceedings that the court ought not to have admitted it.

It has been of some concern to the court that the judge failed to direct the jury that the law required DC Bee to maintain a sufficient log whilst she was viewing the CCTV footage, broadly in accordance with the decision in Smith (in this regard it is immaterial whether Code D was directly applicable). Similarly, the court failed to correct DC Bee’s assertion that the approach provided in Smith and Code D did not apply in this case.

The judge, however, did remind the jury of the main criticisms that were made of DC Bee’s failings in this regard, as set out above: the lack of detail as to her initial reactions when she individually recognised the suspects; the particular features that then struck her; and, overall, the lack of a contemporary record to enable the jury to follow the progress of her identification of the appellants. The judge invited the jury to make appropriate allowances if they considered there was force in those contentions. Although the judge’s directions would have had undoubted additional force if the jury had been told that these were legal requirements, once the judge – clearly correctly – concluded that the images were of sufficient quality to permit the evidence to be given, it was the task of the jury to assess whether they could be sure that the recognition based upon it was reliable. The jury had the advantage of having seen exactly the same material as DC Bee – the CCTV footage and the stills – and they were able to assess the reliability of her suggested recognition.

We do not underplay the importance of the jury receiving accurate and full directions as to the law, but given the jury had heard the clearly articulated criticisms of the approach adopted by DC Bee, criticisms that were summarised by the judge in the summing up, we do not consider that the verdicts are unsafe as a result of this failure by the judge to direct the jury that the law required DC Bee to maintain a sufficient log whilst she was viewing the CCTV footage or to correct DC Bee’s assertion that the approach provided in Smith and Code D did not apply to the work she undertook as regards recognition. In assessing the officer’s reliability, the jury would have undoubtedly focussed on whether the notably detailed reasons she provided for recognising each of the appellants was undermined, principally, by the absence of a contemporaneous record. Given the overall quality of the images and the footage, and the detail provided by DC Bee, we do not consider this to be “poor” recognition evidence (see Turnbull).

Moreover, in each case there was some other evidence. For Yaryare, Mr Platts, the imaging analyst, found “strong support” for the suggestion that Yaryare and person A were the same individual. For Hassan, Mr Platts suggested the imagery provided limited support for the proposition that Hassan and subject C were the same person. Perhaps more significantly, the evidence included i)

We accept Mr Jarvis’s submission that there was a case against Yaryare that was of sufficient strength to go to the jury, based on the detail of the recognition evidence. As described above, DC Bee had had extensive opportunities to study his appearance and there were a number of sightings of person A on the footage which provided a clear basis for making comparisons. Mr Platts, the imaging analyst, found “strong support” for the suggestion that Yaryare and person A were the same individual.

The evidence of association, summarised in detail earlier ([63] above), was relevant background evidence. This was not poor identification evidence; instead there was a clear and sufficient basis for him to be convicted as charged, given the unequivocal evidence of DC Bee, supported by Mr Platts.

As set out above, it is argued by Mr Witcher that the evidence of the bad character of the appellant had no real bearing on his credibility as the author

In our judgment, Mr Witcher is seeking unjustifiably to limit the purposes for which bad character evidence can be introduced when an attack of this kind is made and his submissions fail to take into account section 106 CJA which is in the following terms:

he adduces evidence attacking the other person's character,

he (or any legal representative appointed under section 38(4) of the Youth Justice and Criminal Evidence Act 1999 to cross-examine a witness in his interests) asks questions in cross-examination that are intended to elicit such evidence, or are likely to do so , or

evidence is given of an imputation about the other

on being questioned under caution, before charge, about the offence with which he is charged,

on being charged with the offence or officially informed that he might be prosecuted for it. (2) In subsection (1) 'evidence attacking the other person's character' means evidence to the effect

has committed an offence (whether a different offence from the one with which the defendant is charged or the same one), or

has behaved, or is disposed to behave, in a reprehensible way; and 'imputation about the other person' means an assertion to that effect.

It is clear from this section that the introduction of the accused’s bad character is not dependent on either the defendant giving evidence or having “personal knowledge” of the matters that constitute the attack. Instead, it is sufficient that the defendant’s advocate asks questions in order to elicit evidence that the witness has behaved, or is disposed to behave, in a reprehensible way. That is what happened during the course of Mr Witcher’s cross-examination of DC Bee and it is useful to have in mind the nature of the suggestions that were made. The questions were not framed on the basis that she had simply made an unintended mistake in her identification of Osman. Instead, as the judge summarised the matter in his ruling, Mr Witcher suggested, first, that DC Bee had not made a dispassionate and impartial assessment of the evidence (“you are anything but independent”); second, that she had not told the truth about the time she had spent viewing the footage (“you have been at pains to exaggerate how many hours you have spent”); and that she did not care whether she had made a correct identification (“you have invested a lot in this case and you want a result”). In summary, the suggestion was that she was intent on getting a conviction and keeping her reputation intact.

That was a serious attack on the integrity and the honesty of DC Bee, to the effect that she had taken improper steps to secure Osman’s conviction and that her identification of him was either false or based on deliberately exaggerated testimony, albeit we wish to stress that these were entirely proper questions asked by Mr Witcher on Osman’s behalf, no doubt on the basis of his instructions.

Although all cases will turn on their own facts (and subject to section 103

Although many of the authorities in this context relate to appellants who gave evidence at trial, and therefore the discussion of admissibility tended to focus on his or her credibility as someone who testified in the proceedings, the decisions do not limit the ambit of section 101 in the way urged on us by Mr Witcher. In R v George [2006] EWCA Crim 1652, a trial which concerned the kidnapping and murder of an elderly victim, the appellant sought to blame his brother for the attack on the victim and for persuading him to engage in inappropriate sexual behaviour. One of the grounds of appeal related to the directions given by the judge on the issue of the appellant’s bad character, and in the course of his judgment Moses LJ reflected on the reasons why this evidence had been introduced:

This was described in a similar way in R v M (E) [2014] EWCA Crim 1523; [2014] 2 Cr App R 29, a case which involved an attack by the defendant on the character of the victim in a murder trial. Pitchford LJ dealt with the present issue as follows:

In R v Johnson [2019] EWCA Crim 1025 the appellant did not testify and in the course of his interviews following his arrest suggested that he had been the victim of a sustained attack by the deceased, and that the latter had met his death, having produced a knife, as the result of an accident. This court upheld (at [25]) the following direction, amongst others, by the judge to the jury:

The judge’s direction in that case was clearly based, at least in part, on the suggested wording for a judicial direction in this context as set out in the Crown Court Compendium at [12.9], which is in the following terms:

In conclusion on this issue, if an accused advances a relevant attack on the character of another person, it is not in any sense determinative of an application to introduce his or her bad character whether the attack was based on the defendant’s personal knowledge of the facts or his or her own testimony, or whether it is simply the way in which he or she chose to run the case (as a consequence, for instance, of a favoured interpretation of the evidence). Whatever the basis, it is potentially open to the judge to admit bad character evidence so that the jury, as the judge put the issue in Johnson, are not left in ignorance of the character of the defendant making those accusations. Mr Witcher’s contentions in this regard would impose a limitation on section 101 that is not to be found within the statutory provisions and in support of which he has not advanced any authority.

Section 101 (3) provides an important protection (see [71] above). However, we consider the judge was entitled to conclude that the admission of this evidence would not have such an adverse effect on the fairness of the proceedings that the court ought not to admit it. These were three offences with convictions in 2012, 2014 and 2015 which were relevant to the appellant’s character and his general credibility as the person making these allegations against DC Bee (respectively, possession of an offensive weapon, battery and robbery). Whilst providing assistance to the jury in the context of the attack on the police officer, they would not have had such a profound prejudicial effect on the fairness of the trial that the judge should have excluded this evidence.

The judge impeccably summed up the relevance of this evidence for the jury, and there has been no complaint in this regard.

For the reasons set out above, these appeals are dismissed.

The representation orders were limited by the Registrar to one counsel for each appellant (Mr Bhatia for Yaryare and Hassan and Mr Witcher for Osman). At the hearing of the appeals Ms Nedelcu, Mr Murphy and Ms Thornber appeared in addition for Yaryare, Hassan and Osman respectively. We were asked at the end of the hearing to extend the representation orders to cover their involvement in the appeal.

It is important in this context to emphasise that decisions as regards representation orders should be made, save exceptionally, well in advance of the hearing of an appeal. This is pre-eminently a matter for the Registrar, albeit the issue can be put before the presiding Lord or Lady Justice, or the Vice President of the Court of Appeal (Criminal Division), in advance of the hearing if there is a substantive basis for appealing her decision. These decisions require careful evaluation of the criteria set out in Regulation 18(2) of the Criminal Legal Aid (Determinations by a Court and Choice of Representative) Regulations 2013 (SI 2013/614 as amended by 2013/2814. This should not be left, without good reason, until the day of the hearing because it requires a ‘forward-looking’ assessment of whether the case can be adequately presented without two counsel. The court needs to consider variously (depending on the application) the exceptional condition, the counsel condition and the prosecution condition. Although it is not impossible to make this judgment ex post facto, as a matter of principle it should be dealt with in advance of the hearing. Although we are grateful for the assistance of the three juniors and we recognise they have undertaken work on behalf of the appellants pro bono, we are not at this stage prepared to amend the representation orders. In our judgment, given the limited ambit of the grounds of appeal, these arguments could properly be made on behalf of each appellant by a single advocate, in this case a Q.C. for two of the appellants and a junior for the third, without the assistance of juniors.