Shillibier v R

On 18 December 2000 at Cardiff Crown Court, after a trial before Aikens J and a jury, the appellant, Marc James Shillibier, was convicted by a majority of 11 to 1 on a count of murder and was sentenced to life imprisonment.

An application for leave to appeal was lodged soon after conviction. Since then the case has been before the full court on a number of occasions, as issues have been raised and have then fallen away. It is unnecessary for us to go into the procedural history in any detail. We mention it simply to explain the length of time that has elapsed since the conviction.

We are left with two matters. First, there is an appeal against conviction on one ground on which leave was granted by the full court in October 2003. The issue is whether the judge erred in admitting into evidence the police questioning of the appellant that was conducted at a time when he was treated as a volunteer rather than as a suspect and he was not given a full caution.

Secondly, there is an application to add a further ground of appeal based on fresh evidence. This concerns the reliability of expert evidence given at trial concerning similarities between mud stains on seat covers from the appellant’s car and samples of mud taken from the area where the body of the deceased was recovered.

The victim of the killing was a young woman called Rebecca (or Becky) Storrs, who was 18 years of age. Her semi-naked body was found partially submerged in the River Ogmore behind a cash and carry store in Bridgend at about 11 a.m. on 6 March 1999. She had suffered a serious sexual assault and had been strangled.

There was evidence from two young men, Gareth Jones and Mark Hill, that in the early hours of 6 March they had gone with Rebecca to a party at the flat of a young woman called Katrina Wallace on the Wild Mill estate. They had left the flat some time after 3.20 a.m. because they were fed up and Rebecca wanted to look for drugs. Outside the flat - Jones said it was on the landing, Hill could not remember where it was - they met a man called “Shilbe” or something like that. Rebecca asked the man for drugs and he invited her to go with him, which she did. Jones then phoned for a taxi, a call which was timed at just before 4.00 a.m., and the two men went off in the taxi.

Jones said that before leaving them Rebecca had given him her telephone number. The following day he tried on three occasions to get hold of her, and on the third occasion the call was answered by the police, which was how they came to learn of this evidence concerning Rebecca’s movements in the early hours of the morning.

In due course it was established, in circumstances described below, that the man with whom Rebecca had gone off was the appellant. The Crown’s case at trial was that the appellant was the last person to see her alive and was responsible for her death. The case was based on circumstantial evidence, alleged confessions by the appellant, and lies told by the appellant in the course of his police interviews.

The appellant’s case was that he had met Rebecca outside Katrina Wallace’s flat and she had asked him for drugs, so they set off to his car which was parked at the address where he was a lodger. En route she greeted a ginger haired male. They sat in his car and had a cigarette, then he supplied her with drugs which he kept behind the sun visor of his car. She left the car and walked off in the company of the ginger haired male. The appellant spent the rest of the night in his car, which remained where it had been parked.

The Crown alleged that the appellant had invented the presence of the ginger haired male, basing him partly on Gareth Jones and partly on the appellant himself.

A number of prosecution witnesses gave evidence about the appellant’s movements and conduct on the morning in question. There was also evidence directed to whether his car had been moved or not, and evidence on a variety of other issues relevant to his credibility. We need only pick out a few items in order to give a flavour of the case.

Adrian Jones, who lived at the house where the appellant was a lodger, said that he awoke at about 7.45 to 8 a.m. As he was reading the newspaper, the appellant came into the house and said he had spent the night in the car because he did not want to waken the household.

Stuart Udraufski gave evidence that the appellant called on him at 5.30 a.m. and said he had had a rough night. The appellant called again later that morning wearing different clothes and suggested that Udraufski should not say they had seen each other earlier.

Katrina Wallace originally told the police that Rebecca, Gareth Jones and Mark Hill had left her flat at about 4 a.m. and had gone off together in a taxi. She subsequently changed her statement about their going off together in a taxi. She adhered to that changed position at the trial, where she also gave evidence that she had been getting ready for bed and watching television when the appellant knocked at her door, between about 6 and 7 a.m. He was clean and smartly dressed: normally he looked dirty. He said he needed help: he had “fucked and fucking killed someone”, who he said was Becky. He threatened her not to tell the police or she would be next.

Both Stuart Udraufski and Katrina Wallace were cross-examined by reference to inconsistencies between their statements and their evidence to the jury and on the basis that they had been influenced by the possibility of a reward for giving information.

On 9 March the appellant was questioned by the police in circumstances to which we will need to return at some length. A partial, but not full, caution was given. The prosecution relied at trial on lies told by the appellant in the course of that questioning, including a denial that he had had any dealings with Rebecca that morning. Later the same day the appellant was arrested, following which he was questioned further under full caution and gave an account along the lines of his case at trial.

On 12 March he was released without charge. After his release he went to visit a friend by the name of Aaron Latham. Latham gave evidence of conversations in which, he said, the appellant made remarks amounting to a confession of Rebecca’s killing. The appellant denied having made such remarks and relied on inconsistencies between Latham’s statement and his evidence to the jury.

There was no scientific evidence to link the appellant with Rebecca’s murder, save for one disputed matter which arose in this way. There was evidence that the appellant’s jacket had got wet on the night of the murder. The prosecution suggested that this had happened when the appellant deposited Rebecca’s body in the river. The only possible scientific link between the appellant and the river or the riverbank came from two small mud stains found on detachable seat covers from his car. The covers themselves had been taken off and were found in the boot of Adrian Jones’s car. For the prosecution, Professor Pye gave evidence as to similarities between those traces of mud and samples of mud from the riverbank. The defence expert, Dr Jeans, disagreed. The reliability of the evidence given by Professor Pye is the subject of the fresh evidence application.

The appellant himself did not give evidence in the voir dire concerning the police questioning of him before a full caution was given, but he did give evidence before the jury in support of his case at trial.

At an early stage of the police investigation the senior officers in charge, Superintendent Cooper and Chief Detective Inspector Kemp, laid down a policy for the investigation. Decisions were kept in a policy file. Entries made in the file on the afternoon of 7 March referred to three relevant categorisations:

It was decided that Gareth Jones and Mark Hill were to be treated as significant witnesses. Both of them were questioned on the evening of 7 March and witness statements were taken from them as the questioning progressed. At some point before or during their questioning each was also nominated a TIE individual. In any event, in the case of each of them a full caution was administered at the start of, or soon after the start of, questioning.

On the same evening, 7 March, Katrina Wallace herself was asked to attend the police station as a volunteer. She was questioned and a witness statement was taken from her. She too was given a full caution.

Warrants were also obtained to enable the police to search the homes of Jones, Hill and Wallace.

As Aikens J observed in the ruling to which we will need to refer more fully in a moment, the information which the police had at that time as a result of the witness statements of Jones, Hill and Wallace was contradictory. On the one hand, Jones and Hill said they had left Wallace’s flat with Rebecca, and that she had gone off with the unknown man while they then went off to find a taxi. On the other hand, the information given by Wallace was that Rebecca had gone off in a taxi with Jones and Hill. It was only through subsequent investigations that it was established that the two young men had indeed gone off in a taxi, without Rebecca.

At 2.00 pm on 8 March a decision was taken to treat the unknown male referred to by Jones and Hill as a TIE individual. The stated reason was that he was the “last person to see the deceased alive and in her company”; though, as the judge observed, the police did not actually know whether the unknown male was the last person to see her alive since at that point they had two conflicting accounts.

An entry in the file later that afternoon recorded, in relation to this TIE individual: “The requirements are as follows: full statement to include movements during relevant times; obtain clothing of subject; obtain intimate samples of subject; take possession of any vehicles of subject and forensic examination of same; home address of subject to be searched; movements of subject to be verified by way of witness statements ….”

An entry on the file at 10 a.m. on 9 March stated that the appellant was to be treated as a TIE individual, because of “research undertaken by intelligence cell”. It seems that the connection was made through the similarity between the appellant’s known nickname, “Shilly”, and the name or names mentioned by Jones and Hill when describing their encounter with the unknown male.

In the afternoon of 9 March two police officers went to the address where the appellant was lodging and asked him to go to the police station to assist them with their enquiries. He agreed to go with them.

At about the same time a warrant was obtained to search the address where he had been lodging. The information for the warrant stated:

“(ii)

there is material on the premises which is likely to be of substantial value to the investigation of the offence …: clothing, footwear, sharp instruments and vehicles, likely to prove or disprove as below

(iii)

the material is likely to be relevant evidence …: likely to prove or disprove involvement in the commission of the offence …

(v)(d) the purpose of the search may be frustrated or seriously prejudiced unless a constable arriving at the premises can secure immediate entry to them …: vital evidence could be destroyed.”

At the police station the appellant was questioned and a witness statement was taken from him by Detective Sergeant James and Detective Constable Dawe.

DS James took a deliberate decision as to the form of the caution he should give the appellant. He knew that he had to give some form of caution because that was the policy laid down in relation to TIE individuals. But he said that he was concerned to ensure that the person questioned did not feel oppressed or a suspect, and he thought that the full wording of the caution under paragraph 10.4 of Code C might have that effect because of the words “it may harm your defence if you do not mention when questioned something which you later rely on in court”. He also stated in cross-examination that in a briefing on 8 March he had heard it said that the full form of caution had resulted in difficulties with communication with other volunteer witnesses.

In the result, what he said at the beginning of the first session of questioning was as follows:

“Marc, in fairness to you, I have to caution you that you do not have to say anything unless you wish to do so. Anything you do say may be given in evidence. I am also duty bound to remind you of free and independent legal advice, and tell you that you are entitled to this at any time whilst you are a volunteer attender, that you are not under arrest and that you are free to leave at any time. You are entitled to contact a solicitor by telephone, and if you want a delay to speak to a solicitor, just say ….”

The appellant was then asked if he wanted a solicitor, to which he replied that he did not and that his solicitor was in Plymouth and could not get there anyway. He was told that if he wanted to speak to his solicitor a room would be made available for him to speak on the telephone.

DS James also read out the standard declaration at the top of the witness statement. When asked if he understood it, the appellant replied that he did, and there occurred the following exchange:

“DS James: What does it mean mate?

Appellant: It means, give you any sort of lies, any sort of shit, and you’re up for prosecution for it.

DS James: “Okay, that’s if, if it’s tendered in evidence. Yeah that’s right.”

The appellant was then questioned in detail and a witness statement was taken as they went along. The questioning took up six tapes, one of which was faulty. At the beginning of each tape a caution was given in the same form as we have described.

During the sixth tape DS James went out of the room and told Superintendent Cooper and DCI Kemp that in his opinion the appellant was lying as to his movements in the early morning of 6 March. What he had been saying was contrary to other information that the police had. A decision was then taken by Superintendent Cooper to arrest the appellant on suspicion of the murder of Rebecca Storrs. That was done at 8.45 pm on 9 March.

At the end of the sixth tape, when asked about the way the interview had gone so far, the appellant said there was nothing he wanted to say about it, and “I know it’s done fair and square”.

The following day, 10 March, there took place the first interview after the appellant’s arrest as a suspect. A solicitor was present and a full caution was given. The statement made in the course of the previous questioning was read over to the appellant, who indicated that he wanted to add to what he had said to the police previously. He said he had left out one thing: “I’ve left it out for one reason: fear, and I ain’t got a fucking alibi”. It transpired that what he had left out was that, on his account, he had sat in his car with Rebecca and had supplied some amphetamine to her, and that he had then slept in his car rather than returning home.

At the trial the prosecution sought to rely on that and other lies alleged to have been told by the appellant in the course of questioning before he was arrested and interviewed under full caution. That led to an application to Aikens J to exclude from evidence, under 78 of the Police and Criminal Evidence Act 1984, the entirety of the questioning of the appellant prior to his arrest.

The judge rejected the defence application for reasons set out in a detailed and lucid ruling. His reasons were in summary as follows.

First, he rejected the submission that the appellant should have been treated as a suspect at the time when he was taken to the police station as a volunteer. The police had conflicting statements about what had happened to Rebecca. Nor was there any certainty that the unknown male was in fact the appellant. The known facts, taken together, could not reasonably make him a suspect of the offence of murder. The fact that the police obtained a warrant to search the house where he lived did not prove that he ought to have been treated as a suspect. To obtain the warrant the police had to be satisfied that there was material at the house which was likely to be of substantial value to the investigation; but the police were trying to investigate so as to eliminate from their inquiry people who were at the scene at the relevant time but, upon investigation, could not be suspects. As the information stated, the search was likely to prove or disprove the involvement of any inhabitant of the house in the offence.

Thus judge held that there did not exist reasonable grounds to suspect the appellant of the murder and that there was therefore no requirement that he be cautioned as a suspect in accordance with paragraphs 10.1 to 10.4 of Code C.

Secondly, the judge held that DS James’s deliberate decision not to use the full form of the caution did not involve a breach of the Code. There was no obligation under the Code to caution TIE individuals at all. It was likely that the senior officers would have assumed that the paragraph 10.4 form of caution would be used, but a failure to conform with an informal policy laid down for a particular investigation did not amount to a breach of the Code.

Thirdly, the judge held that even if, contrary to his view, the failure to use the paragraph 10.4 form of caution did amount to a breach of the Code, it was not a “serious and substantial” breach in the circumstances of the case. He was satisfied that DS James acted in good faith. A form of caution was used which made it clear that what the appellant said as a volunteer could be used in evidence in court. That, together with the declaration at the top of the statement, put him on notice that he had to tell the truth. It was reasonable for DS James to consider that in relation to a witness, as opposed to a suspect, a reference to harming his defence if he failed to mention something when questioned might appear oppressive and threatening and might make the witness think he was indeed a suspect. Moreover the appellant was given all the other information mentioned in paragraph 10.2 of Code C, i.e. that he was not under arrest and was free to go. Overall, the appellant was adequately made aware of his rights as a witness.

Finally, the judge considered what might have happened if the full form of caution had been used. He was satisfied that the appellant would have carried on exactly as he did. He noted that when the appellant came to explain why he wanted to add to the story he had given, he did not say that it was anything to do with what he had been told earlier by DS James, but that it was because he had been afraid and he lacked an alibi; which suggested that he knew perfectly well at the earlier stage that what he was saying was important, yet he was still prepared to go ahead and say it.

For all those reasons the judge concluded that the evidence should be admitted.

Mr Glen QC submitted that the judge was wrong so to rule. He accepted that he must show some legal misdirection or that the judge’s decision was Wednesbury unreasonable. In essence, his case was that it was not reasonably open to the judge to conclude as he did that there did not exist reasonable grounds at the time to suspect the appellant of the murder.

The submissions proceeded from the uncontentious premise that the Police and Criminal Evidence Act 1984 and the Codes are intended inter alia to limit the circumstances in which evidence can be obtained from suspects by questioning, and that the regime provides the guarantees required by article 6 of the European Convention on Human Rights. We were reminded of what was said in R v Hasan [2005] UKHL 22, [2005] 2 Cr App R (S) 22, as to the role of section 78 of the 1984 Act in securing compliance with article 6, given its unrestricted capability to avoid injustice by excluding evidence obtained by unfairness.

Mr Glen submitted that the policy and purpose of Code C and the relevant part of the Act are the protection of accused persons from oppression and self-incrimination except in accordance with the regime prescribed. At the heart is the caution under paragraph 10.1, which is triggered by the existence of grounds to suspect. In the form in which it existed at the material time, it stated:

“A person whom there are grounds to suspect of an offence must be cautioned before any questions about it … are put to him regarding his involvement or suspected involvement in that offence if his answers or his silence … may be given in evidence to a court in a prosecution.”

It was not in dispute that “grounds to suspect” means reasonable grounds. Once such grounds exist, it is not simply a matter of being given a caution before questioning. The suspect will usually be arrested and will receive the full protection of the custody officer regime and the “detention clock”. He will be aware of the true nature of the questioning and can make meaningful decisions as to whether to avail himself of legal advice and how to respond to questions. If he obtains legal advice, the solicitor will demand disclosure so as to advise the suspect whether to answer questions at all.

Mr Glen submitted that the Code distinguishes between those who are questioned as mere witnesses and those who are questioned as suspects. In the case of the former, a caution is not required; in the case of the latter, it is. The normal witness declaration about telling the truth or being liable to prosecution for perjury is not the same as the caution about the risk of harming your defence.

Reference was made to paragraph 3.15 (now 3.21) of Code C, which deals with persons attending a police station voluntarily:

“Any person attending a police station voluntarily for the purpose of assisting with an investigation may leave at will unless placed under arrest. If it is decided that he should not be allowed to leave then he must be informed at once that he is under arrest and brought before the custody officer …. If he is not placed under arrest but is cautioned in accordance with section 10 below, the officer who gives the caution must at the same time inform him that he is not under arrest, that he is not obliged to remain at the police station but if he remains at the police station he may obtain free and independent legal advice if he wishes …”

That provision contemplates that there may come a point during the questioning of a witness when there are reasonable grounds to suspect him of an offence. At that point the paragraph 10.1 requirement is engaged and he must be cautioned as a suspect. Prior to that, however, no caution is required.

Mr Glen criticised the TIE policy as an impermissible half-way house which is not recognised by the Act or Codes and is alien to the regime. In his skeleton argument he described the TIE policy as “no more or less than a method of dealing with suspects”. Its effect was to circumvent the detention and questioning procedures under the Code. Its application was bound to involve a “significant and substantial” breach of the Code so as to demand exclusion of evidence so obtained.

It was suggested that the appellant was misled as to his status when he was asked to provide a witness status as a volunteer. The cut-down version of the caution used in his case had no basis in the Code and was not an available form of caution in the circumstances. It failed to signal his true status. Obtaining evidence in this way inevitably had an adverse effect upon the fairness of the proceedings. As soon as he was fully cautioned he requested a solicitor and repaired the omission in his previous account. That indicates that he would have reacted differently had the full caution been given in the first place. As it was, the answers he gave before the full caution was administered were relied on heavily against him by the prosecution at trial.

Mr Glen submitted that the following circumstances inevitably gave rise to suspicion necessitating the appellant’s arrest and caution: (i) the evidence that he was the last person in company with the deceased; (ii) the fact that there were sufficient grounds to apply for and obtain search warrants for premises connected with him, and the terms of the information in support of the application for the warrants; (iii) the taking of intimate samples (though Mr Glen was unable to direct us to any factual material relating to the taking of such samples); and (iv) the fact that, as was known to the police, the appellant had been acquitted a few months previously of a murder in Bath.

For those reasons it was submitted that, whether or not the police did in fact view the appellant as a suspect, he ought reasonably have been so regarded and the judge erred in holding otherwise.

Mr Glen’s submissions tended to elide two separate questions. The first concerns the circumstances in which Code C requires a caution to be given. The second is whether those circumstances existed on the particular facts of this case.

As to the first question, we agree with Mr Glen (and Mr Lewis QC, for the Crown, did not dispute) that the essential distinction under the Code is between those who are questioned as suspects within paragraph 10.1 and those who are questioned as non-suspects. A caution is required in the case of the former but not in the case of the latter; and where a caution is required, it must be in the terms of paragraph 10.5 as it now is (formerly 10.4).

We do not see the TIE policy (which we understand to be a national policy) as cutting across or undermining that essential distinction. As explained in evidence to Aikens J, it applies to persons who are not at that stage regarded as suspects. What appears to have happened, though we have limited information about it, is that the police have adopted, for operational purposes, a number of categorisations in respect of those whom they wish to question. Suspects form one category. Non-suspects are divided into a number of categories, including “TIE”, “significant witness” and, no doubt, mere witnesses. There can be no objection in principle to such a course. The adoption of those categories does not affect the requirement under the Code to caution suspects or the absence of a requirement to caution non-suspects. Whether any difficulty is occasioned by the practice of cautioning non-suspects even in the absence of a requirement to caution them is a point to which we will come back.

Given the essential distinction drawn by the Code between suspects and non-suspects, the crucial question in this case, as it seems to us, is whether there existed grounds to suspect the appellant of the murder of Rebecca Storrs at the time when he was questioned on 9 March.

That, of course, was the first question addressed by Aikens J in his ruling. In our judgment he addressed it unimpeachably, directing himself correctly, examining the evidence before him, and reaching a conclusion on it which was reasonably open to him.

It should be stressed that in the voir dire there was no suggestion on the appellant’s behalf that the police were acting in bad faith in treating him as a TIE individual rather than as a suspect. It was only later, when he gave evidence before the jury, that the appellant suggested that he had been deliberately misled about his status. It is plain in any event that the judge accepted the good faith of the police in the matter; and he did so expressly on the separate issue where good faith was raised, concerning DS James’s decision to use a cut-down version of the caution rather than the full caution.

Thus the judge was examining the state of knowledge of the police at the time in order to see whether they should have reached a different conclusion. In relation to that issue he was entitled to place weight on the existence of conflicting evidence as to whether the unknown male had gone off with Rebecca Storrs, and on the lack of certainty that the unknown male was in fact the appellant. He was entitled to conclude that the known facts, taken together, could not give reasonable grounds to suspect the appellant of the offence of murder.

We do not accept Mr Glen’s submission that the obtaining of the search warrant, or the terms of the information used for that purpose, meant that the appellant fell necessarily to be treated as a suspect, any more than the obtaining of search warrants in respect of the homes of Jones, Hill and Wallace meant that they all fell to be treated as suspects. The criteria for obtaining a search warrant are different from the criterion under paragraph 10.1 of Code C; and, as the judge pointed out, they relate to the existence of material of potential value to the investigation and they apply as much to the elimination of persons from the inquiry as to proving their involvement in an offence. We accept that the fact that search warrants are obtained is a relevant consideration, but we reject the contention that there is an automatic read-over such that the person in relation to whom a warrant is obtained must be treated as a suspect.

The fact that the police knew that the appellant had recently been acquitted of the Bath murder cannot assist the appellant at all. It was not advanced on his behalf before the judge and, in our judgment, is not something that the police could properly have taken into account as a ground for suspecting the appellant of the murder of Rebecca Storrs. Not only was this a previous acquittal, rather than a previous conviction, but there was in any event no similarity between the two murders (the allegation in the Bath case being that the appellant had murdered the victim during a homosexual encounter and had then set light to the victim’s flat in order to destroy evidence).

If, as we have held, the judge was entitled to conclude that the appellant did not fall to be treated as a suspect within paragraph 10.1, it follows from what we have already said that there was no requirement under the Code to caution him at all.

Can the appellant derive any assistance from the fact that the police, in their local application of the TIE policy, decided that cautions would be given to TIE individuals, or the fact that full cautions were given to Gareth Jones and Mark Hill (whether as significant witnesses or as TIE individuals) whereas only a cut-down caution was given to the appellant? In our view, no. We do not think that he can complain about being given an unnecessary caution, even if its wording was not as strong as that of the caution given to other persons similarly categorised in the same investigation. A caution, even in cut-down form, could be said to be more advantageous to the witness than no caution at all. In any event, neither the decision to give a caution under the policy nor the uneven implementation of that decision can generate a breach of the Code in circumstances where the Code does not require a caution to be given at all.

Nevertheless it does seem a little strange, in relation to a person who is not a suspect, to precede the questioning with the full form of caution, including “it may harm your defence, etc.”; and it is understandable why DS James reached the decision he did not to use the full form of caution when questioning the appellant. The apparent unsuitability of the full caution in these circumstances does prompt the question why the decision was taken that it should be used. A possible answer it that it was done in order to improve the chances of the interview being admitted in evidence should it later be held that the person concerned ought to have been treated as a suspect: that is to say, as it was put in the course of argument, that it involved some hedging of bets by the police. If so, we do not think that that renders the approach objectionable as such. We would, however, stress that it should not allow attention to be deflected from the important question of whether a person should be treated as a suspect. Moreover, if there were any tenable suggestion that the TIE categorisation was being used in some way to circumvent the protections afforded to suspects under the Code, issues of bad faith would arise that are wholly absent on the facts of the present case.

For the reasons given, we reject the appellant’s case on the caution ground. We turn to consider the additional ground, relating to mud, on which leave is now sought.

The effect of the evidence at trial on the issue of mud was summarised as follows by the judge. Having pointed out that there was no forensic evidence to link the applicant with the killing, “other than possibly that relating to the mud on the seat covers” of the appellant’s car, he continued:

“Now, on the mud issue the prosecution ask you to prefer the evidence of Professor Pye to that of Dr Jeans. The prosecution asks you to conclude that there is a match of the mud samples to the mud found on the two patches on the seat covers.

What is the evidence on this? Both the experts on this subject are distinguished in their field. Professor Pye possibly has more experience in dealing with this subject in court than Dr Jeans, but that is not a particularly important point, you may think. Professor Pye’s evidence was that the two samples from the seat covers, one from the upright front seat, one from the rear seat of the car, when tested, showed, on his own scale, that there was an 8 out of 10 likelihood that the type of mud from those seat covers was the same as that of samples from the riverbank where Rebecca Storrs’ body was found.

Dr Jeans’ evidence was that in his opinion there was no match in the characteristics. He identified four areas of difference. First he said that the colour didn’t match. Secondly he said the grain size didn’t match. Thirdly he said there were different charcoal counts, and lastly he said the chemistry of the samples – one from the seat covers, the other from the riverbank – was different. Dr Jeans particularly relied on the last two factors, the different charcoal counts and the chemistry difference, for his conclusion that there was no match between them.

Professor Pye was cross-examined on the differences that Dr Jeans identified. Professor Pye said that the difference in colour, difference in grant size, were insignificant. He said that the charcoal count differences were accounted for in the sample size and the particle size distribution found in the seat cover sample. He said that the chemical differences were not of any significance.

Well, members of the jury, it is for you to decide which opinion you prefer, but you may think that even if you preferred the opinion of Professor Pye, it doesn’t get you a great deal further forward. Even if there is a strong similarity between the characteristics of the two mud matches and the samples taken from the riverbank, you may think that is only one strand in the whole weave of the evidence that you have to consider.”

As to those last observations, Mr Glen pointed out that the jury also had before them a written memorandum prepared by the judge to help them follow what he said in his summing up. In that memorandum he set out a conventional direction about the status of expert evidence, and went on:

“In relation to one area only – the ‘mud’ evidence – there is a conflict of view between Professor Pye (called by the prosecution) and Dr Jeans (called by the defence). You must decide whose evidence, and whose opinions you accept, if any. But this evidence relates only to one issue in the case. Deciding whether the mud on the car seat covers in the Renault matches samples from the area where RCS’s body was recovered may help you in reaching your verdit, but you must reach your verdict having considered all the evidence” (original emphasis).

Mr Glen suggested that the memorandum did not play down the mud issue as much as the judge did in his summing up. That may be so, but in our view it did not make much of the issue; and the memorandum emphasised in paragraph 1 that it did not replace the summing up. It seems to us that the overall message that would have been conveyed to the jury by the judge was that mud was not an issue of great significance in the case.

Mr Lewis drew our attention to an application for leave in December 2002 in which the appellant’s own counsel described the scientific evidence relating to mud from the riverbank as being “of limited value on any view since the appellant lived near the scene of the murder”. Mr Lewis also informed us that the defence dealt with the issue in the same way at trial. He submitted that this approach was reflected in the judge’s observation that even if the jury accepted the evidence of Professor Pye “it doesn’t get you a great deal further forward”.

The new ground involves a substantial change of tack. The appellant seeks to argue, by reference to further evidence, that Professor Pye’s methodology was unreliable and his conclusion incorrect, and that this undermines the safety of the conviction.

The further evidence relied on is twofold: (1) a witness statement from Dr Jeans, the defence expert at the trial, and (2) a witness statement (or preliminary report, as he describes it) from Dr Moncrieff, an expert geologist who was not involved in the trial. We were also provided with the transcript and tape of a “Rough Justice” television programme in which criticisms were made of Professor Pye’s evidence in another murder trial. In his skeleton argument Mr Glen said that this was put in “for illustrative purposes”, but in his oral submissions he made clear that he was not relying on it. That was very wise of him. The material in the form presented to us was inadmissible and, in the absence of full details of the trial in question, of no possible assistance to us. We can therefore confine our attention to the witness statements of Dr Jeans and Dr Moncrieff.

The witness statement of Dr Jeans deals with the following matters:

Some of the language used by Dr Jeans is surprisingly intemperate. Dr Moncrieff’s provisional report is much more measured in its tone and is based on an assessment of the evidence given by both experts at the trial. It contains the following summary:

“2.

Professor Pye and Dr Jeans disagreed on the matter of whether the colour of the car seat cover mud was the same as mud from the body recovery site. Professor Pye said it was; Dr Jeans said it was not. Dr Jeans washed the mud from the seat cover before making a comparison, in theory giving him a better chance of assessing the colour accurately.

3.

Professor Pye gave evidence to the effect that the clay mineralogy of the body recovery site was not common. He presented no data to back up that evidence.

4.

Dr Jeans’ grain size comparison showed that the mud on the car seat cover was finer grained than that at the body recovery site. Professor Pye did not work on the grain size of the seat cover mud, but expressed the view at trial that similar fine-grained mud would have been present at the body recovery site. This is possible but not certain and Professor Pye did not test the matter by a re-examination of the site. This would have been the only method of deciding the matter.

5.

The seat cover mud can be distinguished from the body recover site mud because the latter has a far higher charcoal content. Professor Pye dismissed this distinction for a number of reasons. I have checked the methodology used to assess the charcoal content and consider it valid. This result suggests very strongly that the seat cover mud did not come from the body recovery site.

6.

A considerable amount of data were produced regarding the chemistry of the various muds. Owing to a number of factors including contamination of the samples, it is not possible to reach unambiguous conclusions on the basis of these data.

7.

To arrive at the conclusion that the seat cover mud must (rather than could) have come from the body recovery site, it is necessary to look further afield to determine how common that type of mud is. It was implicit in Professor Pye’s conclusions that satisfactory control data had been considered …. The control locations given by Professor Pye are few and restricted to the immediate vicinity of the body recovery site. This approach is unscientific and might have left a seriously misleading impression.

8.

Professor Pye presented no evidence to the Court to support his conclusions. His approach was to write a brief statement giving his conclusions with no data volunteered. He offered no data in support while giving evidence during the trial. It appears that when data are requested by the defence they are presented in a manner that is difficult to untangle and understand with no discussion given as to how the conclusions are reached from the data. This approach is unscientific and led to the situation in which no basis for Professor Pye’s conclusions was ever presented to the Court either in statements or while giving evidence” (original emphasis).

In the course of his detailed exposition Dr Moncrieff comments that it was excessive to assign a value of 8 on a scale of 1 to 10, and that it should have been considerably lower: given the charcoal/coal evidence, the figure should effectively be 1. He also notes that the size of the sample recovered from the seat covers was extremely small (0.0066 gr) and that, as a matter of principle, he is uncomfortable about reaching firm conclusions in a matter of this gravity on such a tiny sample. Several pages of his report are taken up with a discussion of a murder trial in 2001 in which he and Professor Pye were opposing expert witnesses. He makes numerous criticisms of Professor Pye’s evidence in that case.

In response to that material, Professor Pye himself has prepared a very detailed further report in which he takes issue with the criticisms directed at him and his scientific method, and asserts that no new scientific evidence of relevance to the appellant’s case has been produced.

Pursuant to directions previously given by the full court, the various experts attended court in readiness to give oral evidence should we think it appropriate for them to do so. In the event, Mr Glen did not press the court to hear from them even de bene esse and Mr Lewis positively opposed their being called. In agreement with counsel, we took the view that in the circumstances of this case no useful purpose would be served by hearing oral evidence. The issues could be resolved satisfactorily on the basis of the written statements.

The question whether leave should be granted to the appellant to pursue this additional ground depends on whether the further evidence of Dr Jeans and Dr Moncrieff should be received by the court under section 23 of the Criminal Appeal Act 1968.

Before we consider that question, however, we should get a few incidental points out of the way. First, Mr Glen made clear that he does not rely on any suggestion made by Dr Jeans that Professor Pye was dishonest or deliberately misled the court.

Secondly, we have noted above a concern expressed by Dr Jeans about the judge’s summing up. But there is no ground of appeal, either existing or proposed, relating to the summing up, and Mr Glen made clear that he does not criticise the summing up.

Thirdly, in so far as Dr Jeans and Dr Moncrieff refer to other cases in which Professor Pye has appeared as an expert witness, Mr Glen accepted that he could not derive any assistance from such material. There is nothing to show that Professor Pye was discredited or that his professional standing was undermined in any of those cases. It would be neither appropriate nor possible for this court to conduct a detailed examination of such cases in order to see whether the criticisms now made of Professor Pye’s evidence in them are justified. Nor is the present appeal an inquiry into Professor Pye’s general professional competence. The court must concentrate on the evidence given at the trial of the appellant and the safety of the appellant’s conviction.

Fourthly, in the course of his oral submissions Mr Glen took a point which did not feature in the further ground of appeal or in his skeleton argument, to the effect that Professor Pye should not have been permitted to give in evidence his subjective assessment of the degree of match as 8 out of 10. He based this on observations made by the court in R v Gray [2003] EWCA Crim 1001 (as quoted in R v Gardner [2004] EWCA Crim 1639 at para 44) doubting whether expert witnesses in the field of facial mapping or imaging should ever express subjective opinions as to the degree of support that comparison of facial characteristics provided for the identification of a defendant as the offender. Those observations, however, were based on the absence, in the particular field of facial mapping or imaging, of any database or accepted mathematical formula from which such conclusions could safely be drawn. The court was not laying down any general rule against the giving of opinions of this kind by expert witnesses, though such opinions must of course always have a proper factual basis to them and must be presented in a way that does not mislead the jury or cause undue weight to be attached to them. We note that in the present case Dr Moncrieff himself refers to the use of a scale as “a necessary measure when presenting technical issues to a jury”, though he would prefer the use of a phrase rather than a number, and that he indicates where on Professor Pye’s scale he considers the degree of match to fall. Whether Professor Pye had a proper factual basis for his scale does not raise an issue of principle but brings us back to the facts of the case and to the status of the further evidence.

We return to the question whether the further evidence of Dr Jeans and Dr Moncrieff should be received by the court. Section 23 of the Criminal Appeal Act 1968 provides:

“(1)

For the purposes of an appeal under this Part of this Act the Court of Appeal may, if they think it necessary or expedient in the interests of justice –

…

(c)

receive any evidence which was not adduced in the proceedings from which the appeal lies;

(2)

The Court of Appeal shall, in considering whether to receive any evidence, have regard in particular to –

(a)

whether the evidence appears to the Court to be capable of belief:

(b)

whether it appears to the Court that the evidence may afford any ground for allowing the appeal;

(c)

whether the evidence would have been admissible in the proceedings for which the appeal lies on an issue which is the subject of the appeal; and

(d)

whether there is a reasonable explanation for the failure to adduce the evidence in those proceedings.”

We can proceed on the basis that the evidence is capable of belief and, to the extent that Mr Glen relies on it, it would have been admissible at the trial on the mud issue. We focus our attention on the two remaining criteria in subsection (2). In our judgment, both of them pose a serious obstacle to the receipt of this evidence.

On the issue of reasonable explanation for the failure to adduce the evidence at trial, it seems to us that there is nothing in Dr Jeans’s statement that could not have been said by him at trial. Indeed, the substance of it was said at trial. Dr Jeans had had an opportunity to study all Professor Pye’s working papers, which were disclosed to the defence before trial. In his evidence he set out his conclusion that the samples did not match and the basis (including the specific measurements) on which that conclusion was based. Thus the nature and extent of his disagreement with Professor Pye were laid before the court and must have been clear to the jury. They were adequately summarised in the judge’s summing up. Dr Jeans’s further witness statement does not add materially to the position at trial.

Dr Moncrieff was not a witness at the trial, but he could have been. The fact is that the defence chose to instruct Dr Jeans, but it is not suggested that Dr Moncrieff could not have been identified with reasonable diligence or would not have been available had the defence wished to instruct someone other than Dr Jeans. What the appellant is really seeking to do in this case is to deploy a second expert to reinforce the evidence given by the expert instructed at trial. That is not a legitimate use of the appeal process. The point is highlighted by Mr Glen’s suggestion that Dr Jeans may not have made as good an impression on the jury as Professor Pye since he is less self-confident and more academic in his approach. Even if that were true (and, as explained below, we have no reason to believe that the jury in fact preferred the evidence of Professor Pye to that of Dr Jeans), it would plainly not be a proper basis for inviting the court to receive the evidence of a different expert on an appeal.

We note that in certain respects Mr Moncrieff goes further than Dr Jeans went in his evidence at trial, notably by his review of the methodology used to assess the charcoal content of the samples. Again, however, the matters raised could all have been covered at trial. This is not a case in which recent scientific developments or insights throw new light on the issues at trial. It is in large measure an attempt to fight the same battle as at trial, but with reinforcements.

We turn to consider whether the further evidence might afford a ground for allowing the appeal.

We have referred already to the relative lack of significance attributed to the mud issue both by the defence and in the judge’s memorandum and summing up to the jury. It did not form an essential part of the prosecution case and, in the context of the evidence as a whole, it was in our judgment very much a subsidiary issue. Because of the weight of the other evidence against the appellant it is impossible to tell from the jury’s verdict whether they accepted Professor Pye’s evidence or not. Equally, however, the weight of the other evidence was such that the mud issue cannot in our judgment have been a decisive factor in the jury’s deliberations. Thus, even if the jury did accept Professor Pye’s evidence and even if they would not have accepted Professor Pye’s evidence had they had the further evidence (in particular of Dr Moncrieff) before them, there is no real possibility that their verdict would have been different. To put it another way, the further evidence does not cause us to doubt the safety of the conviction.

Accordingly, we do not consider that the further evidence would afford a ground for allowing the appeal.

Having regard to the considerations to which we have referred under section 23(2)(b) and (d) of the 1968 Act, it could not in our view be said to be necessary or expedient in the interests of justice to receive the further evidence upon which the appellant relies.

It follows that the basis of the application to add the further ground of appeal falls away and leave must be refused.

We have rejected the appellant’s case on the one ground of appeal on which he has leave and we have refused him leave on the further ground that he seeks to advance. His appeal against conviction is therefore dismissed.