R v Justin Plummer

On 15 May 1998 in the Crown Court at Luton, the appellant, who was a prolific burglar, pleaded guilty to five counts of burglary (counts 2 and 4 to 7 of the relevant indictment) and on 2 December 1998 he pleaded guilty to a sixth offence of burglary (count 3).

On 16 December 1998 at the Crown Court at St. Albans, before Gage J and a jury, he was convicted of the murder of Janice Cartwright‑Gilbert (count 1 of the same indictment). In essence, the Crown's case was that the killing occurred during an unsuccessful burglary. The appellant was said to have stamped on the face of the deceased with one of a pair of trainers which he had worn during the course of at least four of the above burglaries. On 17 December 1998 he pleaded guilty to 21 further offences. He was sentenced to imprisonment for life for the murder, with a minimum term (per Lord Bingham CJ) of 16 years, with concurrent terms of imprisonment of varying lengths for the other offences.

On 23 September 1999, the single judge refused leave to appeal his conviction. On 17 January 2000 the full court refused his renewed application for leave. The ground of appeal then advanced was that the judge had erred in admitting the appellant's pleas of guilty to the six burglary offences set out above. The court endorsed the reasoning of the judge in his ruling on the application to exclude this evidence, as follows:

On 1 March 2000 the appellant made an application to the Criminal Cases Review Commission ("CCRC"), which was declined, for his case to be referred back to this court.

On 20 November 2017 the appellant made a further and, on this occasion, successful application to the CCRC to refer the case to this court. In consequence, he appeals against his murder conviction under section 9 of the Criminal Appeal Act 1995 on the basis that an expert report by Padraig O'Shea (that analyses the footwear evidence) and a report from Lee Parkes (addressing changes in the field of forensic footwear analysis which had occurred since 1998, particularly in relation to footprints on dead bodies), may undermine the prosecution expert evidence given at trial and the safety of the conviction. The appellant's applications to rely on fresh evidence pursuant to section 23 of the Criminal Appeal Act 1968 has also been referred to the full court. We grant this latter application.

Janice Cartwright‑Gilbert was killed on 28 February 1997. She and her partner, Roderick Cove had been together for 15 years and jointly ran a security company Argus Limited. At the end of 1995 the couple had moved to East End, Wilden where they lived in a mobile/static home whilst their house was being built. There were two containers on the plot in which electrical equipment and tools relating to their business, along with general building materials were stored. The contents were worth £2,000 to £3,000.

Mr Cove gave evidence that on 27 February 1997 Miss Cartwright‑Gilbert's car window had been broken whilst she attended a school governor's meeting during an attempt to steal the vehicle's spare wheel. At 7.30 am on 28 February 1997 he had left to travel to London for work and he last spoke to the deceased at 12.35 pm on the telephone. He was informed of her death whilst on his way home.

Various witnesses gave evidence about the contact that they had had with Miss Cartwright‑Gilbert on 28 February 1997, the last known communication seemingly being a fax she sent at 1.10 pm. No one answered the door at about 1.33 pm when an attempt was made to make a delivery to the mobile home. It was noted that the curtains and the blinds were drawn.

At about 2.00 pm it was noticed that the mobile home was on fire. The fire brigade was called and one of the neighbours, Mr Stacey entered the caravan and discovered the lifeless body of the deceased. He attempted to pull her out of the caravan, but a length of flex that was tied around her neck was anchored to the structure. The perpetrator had used the flex to inflict injuries to her neck and voice box. A kitchen knife and a pair of scissors had been used to stab Miss Cartwright‑Gilbert respectively in her right side and at the midline of her neck. These two items were still in her body. She had a black eye and heat damage to her face. There was a stab wound above and slightly forward of her left ear and her nose had been badly fractured. Bruising on her forehead was consistent with stamping. A different knife to that which caused the neck injuries had been used to inflict the eight stab wounds to her chest which resulted in punctures to her heart and lungs. These latter injuries caused her death. This knife was never recovered. There were other stab wounds around her body.

The appellant was arrested by the police on 25 March 1997 for burglary and inter alia his Nike Air Screech right trainer was seized. He was interviewed by the police on 28 April 1997 following his arrest for the murder of Miss Cartwright‑Gilbert. When asked what he had been doing on 28 February 1997, he said he could not remember.

As already indicated, it was the prosecution's case that the murder occurred during a burglary that went wrong. The Crown relied substantially on the suggestion that the appellant's Nike Air Screech right trainer had made the relevant marks when it was used to stamp on the deceased's face. We have set out above that the appellant had been convicted by his guilty pleas of six burglaries between 25 January 1997 and 27 February 1997, three of which were domestic. The last burglary was at a garage on 27 February, one‑and‑a‑half miles from East End, Wilden. At the scene of four of the burglaries, footprints had been left which were "similar" to that found on the deceased's face. It was admitted by the appellant that he had worn the Nike Air Screech trainers at those four burglaries. The prosecution relied additionally on lies the appellant had told in interview about his movements on the day of the murder and his initial denial of responsibility for the garage burglary on the night of 27/28 February 1997. What was said to have been the appellant's false alibi was backed up, argued the Crown, by witnesses including Adam Betts, Jamie Street and Michelle Sneddon who, it was suggested, had given a false account.

The defence case therefore was one of alibi. It was contended the appellant had not been at the scene of the murder and although he had initially lied about the burglary of the garage, he testified that he genuinely could not remember when interviewed where he had been that day. It was highlighted by the appellant's counsel at trial that there was no evidence to connect him to the scene of the crime and nothing had been stolen. The frenzied attack suggested a more personal motive than a failed domestic burglary and other suspects had not been properly investigated because the prosecution had been "blinkered" in their approach by the shoe marks. It was his case that the footwear which had left marks on the deceased's face was not his and that the scientific evidence was inconclusive. It was argued to be of note that a witness, Steven Miller gave evidence that he had seen a man at about 2.13 pm walking along a road nearby to the caravan and he had seen a car parked in a gateway. At the identification procedure which took place at which Mr Miller attended he picked out someone other than the appellant.

Photographs, including ultraviolet photographs, were taken of the marks on the deceased's face and image enhancing procedures were applied to them. For the prosecution, Dr Nathaniel Cary, a forensic pathologist, testified "There was a distinct area of patterned intradermal bruising on the forehead with what appeared to be a letter 'A'". He said: "It is a relatively flat surface and supported by bone of the skull, and the result is that an imprint on that surface will leave a pattern close in size and scale to the imprinting object. There is little, if any, room for distortion." The marks, he observed, were likely to have been caused by a stamping shod foot. In cross‑examination, he said that it was possible that no blood had got on the trainers or blood could have come off following contact with water. He had not been surprised therefore that there was no blood on the shoe.

The evidence of Dr Kevan Dunnicliffe, a Home Office Forensic Scientist and a footwear expert was read. He concluded that there was "strong evidence to support the proposition that a size 6 trainer had made the ('A') mark." He could not rule out the possibility that a different shoe or a different sized shoe had been used.

Dr David Lewin, a retired dentist, had since 1998 been a forensic expert in the "art of matching weapons and teeth marks". This was his first case involving a shoe, but the technique he suggested was the same. He had produced the overlays from a life size scan of the appellant's right trainer which enabled him to demonstrate similarities with the patterns on the face of the deceased. There was evidence of multiple stamping. From an ultraviolet photograph it was clear that the overlay coincided with a tick and a hole on the sole of the trainer. On this basis, he concluded (there was, as the judge summarised his testimony "no doubt in his mind") that the heel had been the cause of the pattern marks on the forehead and the left side of the face. It was, he said: "One of the clearest marks he had ever seen. No other shoe could have caused it."

Professor Peter Vanezis, a forensic pathologist, prepared a video and used reconstruction and superimposition techniques. He concluded that the appellant's right trainer had caused some of the injuries.

Dr Dennis Bouch, a forensic pathologist, gave evidence for the defence. He agreed he was not an expert in the analysis of marks left by footwear and he had not seen the body. It was his opinion that it was only a remote possibility that the pin prick mark on the appellant's right trainer would leave a corresponding mark on the face of the deceased and, if it had, the edges of the deeper ridges surrounding it would similarly have left a mark. He thought it equally only a remote possibility that a shoe had made the mark of three dots on the deceased's face which, instead, he was of the view could have been from contact with objects on the ground. He was unable to say whether the trainer had made the "A" mark. Overall, he said: "it had not been proved to him that (the right trainer) had made the so‑called unique marks."

Mr Geoffrey Oxlee, an imagery analyst, had dealt previously with two cases of marks on the face. He concluded that he could not rule out the possibility that the right trainer had made the marks on the face, but he suggested this was not demonstrated beyond reasonable doubt. He did not accept that the two marks relied on by the prosecution were highly probably unique. He suggested there was a real difficulty in aligning two‑dimensional pictures with a three‑dimensional surface and that these difficulties were compounded by the angles of the camera taking the images. In his view the shoe could have been a size either side of a size 6. Therefore, he suggested it was impossible to say that the right trainer was the only shoe which could have made the relevant mark or marks.

Mr Roger Blackmore, a chartered chemist and forensic scientist, had been involved in footwear analysis for eight years. He disagreed with the evidence of Dr Lewin and Professor Vanezis to the extent that he could not exclude the possibility that shoes other than the appellant's right trainer had been used. He did not accept that there was a link between one of the unique marks on the right trainer (a tick mark) and the marks on the face of the victim.

For the purposes of this appeal it is unnecessary to review the appellant's evidence in any great detail. As already set out above, he suggested he had not been at the scene of the murder and had been uninvolved in the death of Miss Cartwright‑Gilbert. He admitted committing daytime domestic burglaries. Although when interviewed he had been unable to recall where he had been at the time of the murder, as part of his defence at trial he provided alibi evidence for 28 February 1997. He said that on the day of the murder he had met a woman called Margaret to whom he had sold cigarettes which he had stolen from the Wootton Garage the previous evening. He had travelled to give her the cigarettes at about midday, thereafter returning to his garage, which was about 50 yards from his home, in order to work on one of his three cars. He saw his friend Adam Betts about an hour later. He arranged for Mr Betts to sell some of the cigarettes for him and the pair went back to Mr Plummer's home. Later that day he went to Nottingham to pick up his son from his former partner, Michelle Sneddon. We have already referred to the several alibi witnesses who gave evidence as to his whereabouts. He accepted, again as we have already rehearsed, that he had lied to the police that he had not committed the garage burglary on 27/28 February 1997.

Against that background we turn to the criticisms summarised by the CCRC that are advanced by the forensic scientist Padraig O'Shea who has been involved in conducting footwear mark comparisons since 2002. He is a reporting scientist and an expert witness in this field. Mr O'Shea agrees with one of the prosecution experts at trial, Dr Dunnicliffe, that there was no evidence to suggest that the appellant's right trainer and no other made the marks on the victim's face and head. He is critical of the assertion by Dr Lewin that there was no difference between the techniques of matching bite marks and footwear marks. He observes, and the Crown does not dispute, that there is a clear difference in the methods of comparison, along with a difference in the complexities of the comparisons, that lead to evaluations in these two fields. The techniques Dr Lewin used in this case were neither validated nor suitable for footwear mark comparisons, either in the late 1990s or today. The method used by Dr Lewin of scanning the heel using a computer and then transferring the result onto a transparent film was not accepted practice at the time of the trial and is not currently accepted. Instead, test marks are conventionally prepared for comparison by powdering the sole and then walking/placing the sole onto a sticky transparent acetate and covering it with a transparent film. This enables fine details to be reproduced in the test mark.

Dr Lewin's methodology was entirely different and the scanning technique would have failed to reflect that the sole of a training shoe is rarely fully flat and in this case the soles curved upwards at the toe and the heel. Furthermore, scanning (even with today's improved machinery) does not capture all the fine details. Dr Lewin drew an unhelpful and incorrect comparison between the soles of the trainers and vehicle tyres and he wrongly focused on the comparison of individual marks rather than considering the "whole pattern". Mr O'Shea was uncertain whether the "letter A" is present as suggested by Dr Lewin, and Mr O'Shea was unable to accept that there was damage or characteristic features on the sole of the shoe that were reproduced in the marks on Miss Cartwright‑Gilbert's forehead and face.

Other criticisms are made of the evidence of Dr Lewin by Mr O'Shea who in the event disagreed that there was a conclusive link between the appellant's trainer and the injuries to Miss Cartwright‑Gilbert's face. This was the first comparison of this kind undertaken by Dr Lewin and Mr O'Shea observes, again without contradiction by the Crown, that he should not have been treated as an expert in this particular field in these circumstances.

Turning to Professor Vanezis, Mr O'Shea observes that there is nothing to suggest that the former, as a forensic pathologist, has appropriate qualifications to act as an expert in the field of footwear mark comparisons. He suggests that Professor Vanezis was commenting on matters which were outside his field of expertise. He also used a method which was inappropriate, namely overlaying images of the heel over images of the face of the victim when the scales of the images being compared were uncertain. Mr O'Shea disagreed with Professor Vanezis' conclusion that there were fine details present in the marks on the face of the victim and the injuries observed on her face.

Mr O'Shea agreed broadly, although not entirely, with the conclusions of Mr Oxlee (who as indicated had been called by the defence at trial). However, he questioned his credentials as an expert giving evidence as to footwear marks and he disagreed with some of his conclusions such as the assumption that marks on the face of the deceased had been caused by blows or stamping. Mr O'Shea essentially agreed with Mr Blackmore, again called by the defence at trial. Finally, Mr O'Shea observes that Dr Bouch is not a marks comparison expert and was commenting on results which were outside of his field of expertise.

Lee Parkes, who joined the Forensic Science Regulation Unit in 2018, summarised his findings as follows:

The likelihood ratio, not used by the prosecution experts in this case, as agreed amongst the Association of Forensic Science Practitioners, is reflected in the following scale: Inconclusive, limited/weak, moderate, moderately strong, strong, very strong and extremely strong. On this scale Mr O'Shea is of the view that the evidence provides moderately strong support for the proposition that the marks on Miss Cartwright‑Gilbert's face had been made by the heel of right trainer belonging to the appellant.

Section 23 of the Criminal Appeal Act 1968 governs the admission of fresh evidence. It provides that the court may receive evidence if it is thought necessary or expedient in the interests of justice to do so. In considering whether or not to take this step, the court shall 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 from 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.

The Privy Council considered the position in New Zealand regarding the admission of fresh expert evidence in R v Lundy [2013] UKPC 23 (a jurisdiction we note which applies a similar approach in this regard as in England and Wales). At paragraph 120 the Board stated as follows:

We consider this represents the approach to be taken in this case. Although it will sometimes be necessary to consider inter alia the position at trial and to inquire why the evidence, if available, was not adduced at that stage (see for instance R v Erskine [2009] EWCA Crim 1425, [2009] 2 Cr.App.R 29) this is not necessary here. The errors were errors by the prosecution and not the defence, in that witnesses were relied on who stepped outside the curtilage of their expertise. Although the appellant's lawyers could have objected, any fault on their part is clearly secondary to that of the prosecution or the experts who the prosecution called.

The Crown accepts that the evidence of Mr O'Shea is credible and reliable as to the true significance and weight of the evidence of a match between the foot mark left by the killer on the head of the deceased with the sole of the right trainer. The respondent accepts that the evidence of Dr Lewin and Professor Vanezis was in error in so far as either witness stated that the match could be regarded as conclusive. The respondent accepts Mr O'Shea's criticisms of the methods and the manner of the presentation of the results at trial of Mr Lewin and Professor Vanezis.

The jury was in effect directed that its verdict would turn on whether it accepted the evidence of either or both Dr Lewin and Professor Vanezis, witnesses who on this occasion had given fatally flawed evidence. The concessions by the Crown that we have set out above were, in our judgment, correct and unavoidable. Together with the evidence of Mr Parkes, they render the verdict in this case unsafe and we allow the appeal and quash the conviction.

Turning to the issue of a retrial, we are conscious that this case occurred in the region of 23 years ago. The test under section 7 of the Criminal Appeal Act 1968 is whether a retrial is in the interests of justice. This case involved the violent, indeed particularly savage, murder of someone who was seemingly simply spending time in the vicinity of, or within, their own home. Notwithstanding the gap in time, there is a clear public interest in the resolution of grave crimes of this kind, in a sense that the relevant suspect against whom there is a case to answer is tried by a jury.

We have borne in mind, therefore, the age of this case, along with the fact that the appellant has served a longer period in prison than the tariff set by the then Lord Chief Justice, Lord Bingham. He has never applied for parole because he has continued to assert his innocence, albeit the Parole Board considered his case on 19 December 2018 and on 26 January 2021, with further consideration due in December 2022.

The respondent suggests that notwithstanding the change in approach to the shoe mark evidence, there is compelling evidence based on the remaining available evidence, some of which was not called in 1998. It is not for us to make decisions as to the admissibility of evidence at any future trial, save that we will ignore any evidence in relation to which, in our judgment, there is no reasonable prospect of it being admitted.

The principal evidence on which the prosecution can credibly seek to rely is as follows:

In our judgment this evidence, along with certain other material to which we have not referred, provides at the very least a credible case against the appellant. Although Christopher Dunne, John Stacey and Adam Betts have died, the prosecution have undertaken to ensure that the evidence from the latter witness will be admitted into evidence in a convenient format if a request is made of them by the appellant's lawyers. Dunne's evidence is potentially admissible as a hearsay account, depending on any judicial ruling on this issue. If it is admitted, we have no doubt it will be the subject of strong judicial warnings.

The test to be applied was set out by Lord Bingham in R v Graham [1997] 1 Cr.App.R 302 at page 318B:

In particular, we must ensure that there should be no unfairness or oppression if a retrial is ordered.

In summary, the case is exceedingly serious given the nature of the murder. There is a credible case, as we have just observed, against the appellant. There are outstanding disclosure issues as regards the confession evidence, particularly as regards Dunne and whether he was given some form of inducement by the prosecution, but even if this happened, it would not necessarily be determinative of whether his evidence is admissible and his account may in any event carry some legitimate weight.

There are no reasons, in our judgment, why the appellant cannot have a fair trial, particularly given the judge will be able to provide the jury with appropriate directions as to how they should approach the gap in time between the trial in 1997 and the trial either this year or next.

It is correct for Miss Thorne QC to observe that certain exhibits were not retained and cannot now be subjected to forensic examination, but this in our view can be dealt with by the judge appropriately by highlighting any disadvantages under which the appellant will labour at trial. The courts frequently give directions to juries in this context.

We have borne in mind the length of time since this offence was committed, but we consider that there is a strong public interest in the allegation against the appellant being resolved by a jury, notwithstanding the length of time he has spent in custody as a Category A prisoner ‑ a period which is considerably in excess of the tariff period.

It follows that although they were persuasively advanced, we are unpersuaded by Miss Thorne's submissions as regards the punishment he has already received, the suggested disproportionality of re‑trying him at this stage and the argument that it would be unfair for him to be tried in 2021 or 2022 for an offence committed in the late 1990s. That is not to suggest that the appellant is deprived of the opportunity to raise admissibility and fairness issues with the judge, for instance under section 78 of the Police and Criminal Evidence Act 1984 or indeed by way of a submission that the case should be stayed. We are not in a position to investigate, for instance, what is said by Miss Thorne to have been the wholesale loss of exhibits, close to some 3,000 items, of which we were only informed today and which came to the attention of Mr Plummer's legal team in the relatively recent past. This can, indeed in all probability should, be given careful and anxious consideration by the trial judge along with other matters, to some of which we have already referred.

The appellant should be arraigned on the trial count of murder and counts 19 and 20 from Indictment 1 from 1997/1998, namely offences of common assault on Adam Smith on 29 April 1997 and common assault on David Bails on the same day. That has the consequence that we lift the order that those two offences should lie on the file. It will be for the trial judge to determine whether these two latter charges should in the event be tried along with the allegation of murder.

There is to be no publication of this judgment or the submissions advanced on this appeal until the conclusion of the retrial, save that copies of it will be distributed to the respondent and to the appellant and a copy will be made available to the trial judge. The prosecution are to inform the Registrar when the Crown Court trial is concluded in order for this judgment to be made publicly available.

Mr Sells Q.C., who led for the Crown at trial, should be informed that the court anticipates that he will assist in providing the parties with the benefit of any further recollection he may have as regards the suggested inducement to Dunne.

We direct that the fresh indictment be served on the Crown Court Officer not more than 28 days after this order. We direct that the appellant be arraigned on the fresh indictment within two months. We direct that the venue for the retrial should be determined by a Presiding Judge of the South Eastern Circuit. We direct that for the time being the appellant is to be held in custody and that any issues concerning bail should be directed at an appropriate judge sitting in the Crown Court (a High Court or Crown Court judge).