Underwood, R v

This is a reference by the Criminal Cases Review Commission.

On 16th March 1995 in the Crown Court at Exeter the appellant was convicted of murder and sentenced by Tuckey J to life imprisonment.

The Single Judge (Macpherson J) refused leave to appeal describing the case against the appellant as being overwhelming. On 19th November 1996 this court (Hobhouse LJ as he then was, Bracewell and Curtis JJ) refused his renewed application for leave to appeal against conviction.

In July 1994 the deceased, Terry McPherson, was killed in his flat in Teignmouth, by the infliction of repeated blows to the head. He was aged 54 and lived alone in the flat, on the second floor of 28 Northumberland Place. He suffered from schizophrenia and depression and was described as a loner. He was unemployed and in receipt of invalidity benefit of £70 a week, his rent being paid directly to the landlord. On 21st July he had signed a Housing Benefit claim form stating that he had £460 in bank accounts and £930 in cash. The last payment of benefit to him had been on 15th July when orders to the value of £588 were cashed. After his death £120 in notes was found in a pocket on the back of the door, and there was about £12 in loose change. He was last seen alive on 20th or 21st July, visiting a local shop, and his body was found on the afternoon of Monday 25th July. The prosecution case was that the appellant killed him in the course of robbing him, on the night of 23rd/24th July.

The appellant lived on his own in a ground floor flat at 28 Northumberland Place, and admitted to the police when interviewed that he knew the deceased and knew something of his illness, but said that he did not see him often. He had once been to his flat to look at a dripping tap. The appellant was also unemployed. He was in receipt of benefit of about £73 a fortnight, his rent being paid direct to the landlord. His last girocheque was received on 14th July. He did work from time to time for his landlord who had paid him a maximum of £2,200 by cheque or cash, between 10th October 1992 and 24th June 1994, when he gave him a cheque for £75, which included a loan of £30. It was admitted by the appellant that in the period immediately before 23rd July 1994 he owed nearly £300 on his electricity and water bills, and that he was borrowing odd amounts from friends in the Queensbury public house, where he was a regular customer. The barmaid there said that he seemed very short of money at that time. He told the police in interview that over the last 9 to 12 months he had been putting money away, including his earnings from his landlord, in order to go to Holland, and that he had saved £1,200. He said he borrowed money from time to time, but never touched these savings.

On Saturday 16th July 1994 Micky Lewis, a good friend of the appellant’s, came from Birmingham to stay with him in Teignmouth. On Tuesday 19th July the appellant’s cousins Fred Dickens and Sharon Underwood, with five children, came from Birmingham to see him, and pick up Micky Lewis. The appellant wanted to go back with them but there was not enough room in the car. They all spent the night at his flat and left the next day, Micky Lewis and Sharon, to whom it seemed that the appellant was much attracted, saying that they would come back at the weekend.

On Saturday 23rd July the appellant was in the Queensbury public house from about 1 pm onwards. He drank 4 or 5 pints of cider, and one or more vodkas with orange (but it was not suggested that he was drunk). He went out several times to telephone Birmingham, to see what had happened to Micky and Sharon, and seemed increasingly depressed that they had not arrived. Between 9 pm and 10 pm he went out. He asked someone to keep an eye on his unfinished drink, saying that he would be back in a minute, but he did not return.

It was admitted that at 12.50 am on what would now be Sunday 24th July he was in the Brunswick Street car park taking a taxi from Teignmouth to Exeter. (A map was produced showing that the car park, the Queensbury and 28 Northumberland Place, were all very close together.) The police, when they interviewed the appellant, asked what he had done between leaving the pub and then. He said that he had walked home, which took about two minutes. His flat door was ajar which was not usual but he had not thought anything of it. He told the police that he collected money (in about the sum of £1,200) which he had hidden under the floorboards of the cupboard under the stairs; he said he then closed the door but did not lock it, and went to the car park to get a taxi. He agreed that that did not account for the whole period and that he must have done something else, but said he could not remember what it was.

At about 1.30 am he took a taxi from the taxi rank in Exeter to the Gordano Services area on the M5, arriving there at about 2.40 am. The driver said that he was carrying a rolled up carrier bag with not much in it. At 3.30 am (not now carrying anything) he took a taxi from there to Birmingham city centre. At 5.30 am to 6 am he arrived at the house of Fred Dickens, who was very surprised to see him. He said that he was paying a courtesy visit, and had left his bag at a service station. He wanted to go for a drink, saying that he had about £1,000 which he had won on the horses. Later he went and bought some new trousers because he said that the zip on his jeans was broken. From about 11 am onwards he spent the day in various pubs with various relations who joined him. He gave some of them a total of £400 to buy themselves drinks, saying (again) that he had won money on the horses. At trial that was admitted to be a lie. He did not tell people that he had come to Birmingham by taxi, and he gave the impression to witnesses that he had had a lift in a van or lorry. Micky Lewis said in terms that he told him that he had had a lift in a lorry and had left his spare clothes in the cab.

Next day the appellant bought a car for £375, and then went to various pubs with his companions. That afternoon, as a result of the landlord of the Queensbury having reported the appellant missing and the police having gone to his ground floor flat, the deceased’s body was discovered in the second floor flat. It became known to the people in Birmingham that the police wanted to see the appellant about it. His brother told him and his reaction was one of shock. He said, then, and to other people over the next few hours, that it was nothing to do with him. He said he tried to contact the police but without success.

Micky Lewis gave evidence that later that evening, when he and the appellant were alone in a pub car park, he asked the appellant if he was responsible and the appellant answered “I did not mean to go that far”. Then the others came out and nothing more was said. Lewis said that he had not mentioned this in his first statement to the police out of loyalty to his friend. The two of them spent that night in the sitting room at Sharon’s house and when they were on their own the appellant said “I am facing 20 years for this aren’t I?” Lewis told him that they would sort it out in the morning.

The police came to the house at 6 am the next day. The appellant was found hiding in a cupboard under the stairs. He had on him £556. Before he was taken back to Devon, Sharon visited him and he told her that he had put £1,000 behind the bathroom sink for her and her boyfriend. This was recovered by the police. (The money spent by the appellant since 12.50 am on 24th July, together with the money in his possession, amounted to some £2,566.)

The appellant was taken to Torquay and interviewed over the next two days. On the Tuesday evening, 26th July, he was seen by a police surgeon who found an injury to the knuckles of his right hand, consistent with his having used the clenched fist to strike a blow or blows to the face, although there was no bruising to his hands. He had told his friends and family in Birmingham that this injury had been caused by punching the punchbag in the Queensbury. There was evidence that when he was in the Queensbury on the Saturday afternoon he had punched the punchbag a few times with his bare hands, and then again wearing gloves, but no-one saw any resultant damage to his hands. The prosecution claimed that his account of the injury was a lie. Evidence was given by a forensic scientist who punched the punchbag repeatedly as hard as he could, without sustaining any such injury. Nor did he find in the boxing gloves any trace of the blood which would have been there if the appellant had used them with injured knuckles.

The police officers who had gone to the appellant’s flat at 3 pm on Monday 25th July, found the door closed but not locked. There was a calor gas bottle behind it which had to be moved when the door was opened. The key was on a shelf inside. (There were no keys on the appellant when he was arrested.) The kitchen at the back of the flat had a door into an enclosed yard. This door was open with the keys in it, and there was a fixed ladder in the yard on to the roof which provided a means of getting out of it. On the hall floor was found a cord which it later appeared was from the deceased’s dressing gown. It had been knotted several times and was heavily bloodstained. DNA evidence indicated that the blood was that of the deceased. At 3.20 pm the police officers went up to the deceased’s flat and found his body. The dressing gown, without its cord, was on the back of the door. A copy of the TV Times was open on a table showing the programmes for the evening of Saturday 23rd July, which the prosecution said gave some assistance as regards the time of death.

No identifiable fingerprints were found anywhere. A patch of blood on the stairs leading to the second floor could not have come from the appellant or the deceased. Three towels in the appellant’s flat appeared to have blood on them. That on the first could not be analysed; that on the second could have been the appellant’s but not the deceased’s; and that on the third could not have come from either of them. No blood was found on the appellant’s clothing when he was arrested (except for a very small spot on his T-shirt which was probably his own), or on a sample of the banknotes recovered from Sharon’s bathroom. The prosecution suggested that the absence of blood on the appellant’s clothing was explained by the fact that by the time to got to Birmingham he was not wearing the same clothes as he had been seen wearing at the Queensbury on the Saturday evening, and must have disposed of them.

Dr Fernando, the pathologist, carried out a post mortem on the evening of Monday 25th July. He considered that death had occurred more than 24 hours before. It was difficult to say how much before, but he thought one or two days. The deceased had died as a result of a gross head injury consisting of extensive fracture of the skull, and laceration and contusion of the brain. The injuries were consistent with having been inflicted with a heavy blunt instrument such as a claw hammer or a spanner. They included a black eye and broken nose, which were consistent with punches, but could also have been caused by the blunt instrument. There was also a mark round the neck. The prosecution suggested that this was caused by the dressing gown cord’s having been used in an attempt to strangle the deceased, which would explain how it came to be bloodstained.

As we have noted, in interviews with the police the appellant said he had about £1,200 stashed away in his flat. He had been putting money away which he had earned from his landlord for doing bits of work here and there. But the evidence was that at least £2,560 was either spent by him or left in Birmingham. He said that he took about 15 minutes between leaving the Queensbury and getting a taxi. In fact about 3 hours elapsed and, as we have said, the appellant accepted he could not account for that period of time. He described the taxi journeys and said he had inadvertently left a bag of shoes and trousers at the Gordano service station. He denied he had changed his clothes. He said he had been wearing a yellowish T-shirt and blue jeans (the clothes he had been wearing on arrival at Birmingham). But the witnesses said that on Saturday evening he had been wearing black trousers and a white shirt in the Queensbury public house.

The appellant did not give evidence. The defence case was that he had given his account to the police in interview and had nothing to add. There might be suspicion but there was no proof of his guilt. There was no forensic evidence to connect him to the crime. His trip to Birmingham could be attributed to his feelings for Sharon, and the jury could not be sure that the money, which he had, had not been saved, or at any rate that it had come from the deceased. Someone else had the opportunity to kill the deceased and could have left via the appellant’s flat, dropping the cord as they went. Micky Lewis’s evidence of a confession was challenged, on the basis that (1) he did not mention it in his first statement to the police, and (2) the statement, in which he did then mention it, was inconsistent with his evidence in court as regards the point in the evening when the alleged conversation took place.

The reasons given by the Single Judge for describing the case against the appellant as overwhelming were that the cord found in the appellant’s flat was deadly evidence as were remarks which he made later. That must be a reference to the remarks made to Lewis which we have set out above.

In the course of their investigations the Commission discovered that the witness Micky Lewis had 54 previous convictions, 39 of them involving dishonesty. When making his witness statements he, Lewis, had given a false middle name and date of birth, so that his previous convictions were not discovered. It was likely that he deliberately concealed his identity from the police, as there was an outstanding arrest warrant against him. The Commission concluded that he may well have been motivated to fabricate important evidence against the appellant in order to have a bargaining chip, in the event of the police discovering his true identity. Moreover, the fact that he had in mind that he would give evidence against the appellant in return for the warrants against him not being pursued, was demonstrated by representations falsely made on his behalf to the Magistrates’ Court in 1997, when he said that the police told him in 1995 that the warrants would not be pursued if he gave such evidence.

Mr Atkinson QC and Mr Farmer now appear for the appellant and have submitted that the conviction should be quashed for that reason. They also sought leave to call fresh evidence from a newly instructed pathologist in relation to the time of death and other matters and relied on new but uncontroversial DNA evidence. We will consider these matters in reverse order.

This was to the effect (1) that after further extensive tests, of a more sensitive kind than could be done in 1994 and 1995, none of the appellant’s DNA was found at the flat of the deceased and none of the deceased’s DNA was found at the appellant’s flat, apart from that on the dressing-gown cord; (2) that the cord contained DNA of a third party (unidentified) as well as that of the deceased and (3) that blood, whose DNA did not match that of either the appellant or the deceased, had been found on the staircase wall of the deceased’s flat. We received this evidence in written form without objection from the Crown; Mr Atkinson relied on these points as makeweights rather than points which, on their own, established the unsafety of the conviction. We regard the points as entirely neutral.

We thought it right to hear the evidence of Dr Cary and having heard it decided that it should be admitted even though there was no very satisfactory evidence as to why it was not called at the trial. The defence had had the services of a pathologist at the trial, Dr Hunt, but in the event decided not to call him; Mr Atkinson agreed it would be right for the Crown and the court to see Dr Hunt’s report. In the event, it emerged that he had agreed with the oral evidence given by Dr Fernando for the Crown that the deceased must have been dead for at least 24 hours by the time he was examined by Dr Fernando at 7.40 pm on Monday 25th July. Both Dr Fernando and Dr Hunt agreed, further, that it was very difficult to say how long before 7.40 pm on Sunday 24th July he would have died. As Dr Hunt said:-

“A longer time is more difficult to assess, especially as the head injuries were such that he may not have died for a little time.”

Dr Fernando’s view was that it could have been up to 1 to 2 days earlier than the 24 hour period before his examination.

It appeared from Dr Cary’s written evidence that he might be saying that the time of death was between 24 and 36 hours before the examination at 7.40 pm on Monday 25th July. That would be important, if right. Since 36 hours before that date meant the killing would have taken place at or after 7.40 am on Sunday 24th July when the evidence established the appellant was already in Birmingham.

It transpired, however, that this was not Dr Cary’s oral evidence at all. He said that all he could say was that death happened earlier than 24 hours before Dr Fernando’s examination. “Anything beyond 24 hours is a reasonable possibility” he said. Again, “it must be more than 24 hours; whether it is 1 or 24 hours more, there is no science other than the absence of flies’ eggs. But that is vulnerable.” The reference to the absence of flies’ eggs on the corpse was established by Dr Fernando’s original report. That is a help to a pathologist in relation to the time of death and encouraged Dr Cary to take the view that the time of death might be nearer 24 hours than 48 hours before Dr Fernando’s examination. He accepted that it was a “vulnerable” criterion because there was what he agreed was “hindered access”, by reason of the fact that the only access was through the transom of a window which was itself shielded by both a net curtain over the window and drapes of curtains down the side of the window. He concluded that he could not pin the time of death any more closely than to between 24 to 48 hours before 7.40 pm on 25th July. That seemed to us not to change the position of the evidence in any material way from what it had been before the jury; it does not advance the appeal.

Dr Cary also gave evidence that blood on the appellant’s knuckles was consistent with his account of having punched a punchbag and that the marks in the photographs of the deceased’s neck were not necessarily those of the cord of the dressing gown found in the appellant’s flat. This evidence did not take the case anywhere either. On the cause of the blood on the appellant’s knuckles, it was entirely a matter for the jury whether it was likely or possible that the appellant had incurred the injury in the way he described. They had evidence from Dr Fernando (as well as the forensic evidence we have already mentioned) that he had actually punched the relevant punchbag while it was in the exhibits room at Exeter Crown Court and had sustained no injury to his knuckles since it was a flat surface, although he did succeed in spraining his shoulder. The point about the cord of the dressing gown was not so much its use as a weapon for attacking the deceased (it was not suggested he died from the use of the cord as a ligature) but that it was found drenched with the blood of the deceased, in the flat of the appellant. Mr Atkinson recognised that these two further points made by Dr Cary were small makeweight points. They certainly cannot, of themselves, affect the safety of the conviction and we turn, therefore, to Mr Atkinson’s main argument.

There is no question of any fault on the part of the Crown. When Lewis made his statements he deliberately gave the police a false middle name and a false date of birth. Routine checks did not reveal that Lewis had, as we have already said, 54 previous convictions, 39 of which were for offences involving dishonesty. It seems that the appellant knew Lewis had previous convictions and had so informed his solicitors at an early stage. That may not have been communicated to trial counsel but the real reason why trial counsel was not informed was that Lewis had successfully deceived the police. The appellant himself had 25 convictions for dishonest offences, 4 convictions for assault, one for wounding and one for arson. These convictions would, of course, have emerged if Lewis’s credit had been attacked and if the appellant had given evidence. In fact the appellant did not give evidence and so it is a fair inference that, if the defence team had known of Lewis’s previous convictions, they would have used them to attack Lewis’s credibility; the decision that the appellant was not to give evidence would have been taken on the first instead of the third day of the trial. The trial took place at a time when juries were directed not to hold against a defendant the fact that he had not given evidence. It is clear to us that, if the previous convictions had been known, counsel would have conducted the defence differently and would, no doubt, have cross-examined Lewis more effectively than he was in the circumstances able to do if, indeed, Mr Lewis had been called at all. We agree, therefore, with the Commission’s conclusion at para 9.61 of the reference that if the defence had known of Lewis’s convictions they would have decided to use the information in cross-examination, on the basis that the appellant would not give evidence.

We were referred to the authorities of R v Smith 4th June 1998 Unreported, Vasilou 4th February 2000 [2000] Crim L Rev 845 and Farrell 20th March 2000 Unreported. In all these cases convictions were quashed because the convictions of an important witness, who gave evidence about the central matters in issue in the case, were not disclosed. We take the law from paragraph 7 of Farrell where Lord Bingham of Cornhill CJ said this:-

“7.

Thus we have a clear and simple case in which the convictions of the prosecution witness were not disclosed when they should have been as a result of inadvertence or oversight. What is the effect of such non-disclosure if a defendant is convicted and evidence of convictions on the part of the prosecution witness then comes to light? There is no simple and straightforward answer to that question. The answer will depend on the weight of evidence in the case, apart from the evidence of the witness whose convictions have not been disclosed. The greater the weight of the other evidence the less significance, other things being equal, the non-disclosure is likely to have had. The answer will also depend on the extent to which the credibility and honesty of the prosecution witness whose convictions have not been disclosed is at the heart of the case. If, as here, the prosecution witness whose convictions have not been disclosed is the only witness against a defendant, and his credibility and honesty are squarely in issue, and the jury are led to believe that that witness is of good character when such is not the case, then there is strong ground for contending that the conviction is unsafe. That is the conclusion to which we find ourselves driven on this appeal and we accordingly feel constrained to allow the appeal and quash the conviction.”

Mr Atkinson submitted that the credibility and honesty of Lewis did, indeed, go to the heart of the case. What, he asked, was more at the heart of the case than a confession made by a killer to someone whom he thought to be a friend? He buttressed the argument by saying that Lewis, who gave his evidence at the end of the first day and must have thus left a resounding impression on the jury at a critical time of the case, was the only live witness of consequence; the other evidence was all circumstantial and the conviction was, therefore, inevitably unsafe.

We disagree. The submission overlooks the Lord Chief Justice’s observation that there is no simple and straightforward answer to the question of what is to happen if evidence of previous convictions of a prosecution witness come to light after the trial. As he said:-

“The greater the weight of the other evidence the less significance, other things being equal, the non-disclosure is likely to have had.”

Accordingly it cannot be the case that if an important witness’s convictions are not disclosed, the conviction must inevitably be quashed. As Lord Bingham said later in R v Pendleton [2002] 1 WLR 72, para 19, it is for this court to determine whether the conviction is unsafe and in any case of difficulty to test its provisional view by asking whether the new evidence (here the previous convictions) if given at the trial might reasonably have affected the decision of the trial jury to convict.

The critical fact that makes this case different, in our judgment, from the authorities cited is that in those cases the witness in question gave evidence about the central facts in issue in the case. Here Lewis had nothing to say about the central facts in issue, circumstantial though they are. His evidence, persuasive if believable, related to a subsequent confession and left the central facts in issue untouched. The question therefore is whether, without Lewis’s evidence, the jury would have convicted. We are convinced that they would and that evidence about Lewis’s convictions could have made no difference to the verdict.

We have already set out the evidence given at the trial that leads us to that conclusion but the salient and telling features are as follows:-

(1)

The appellant’s sudden departure from Teignmouth at ten to one in the morning;

(2)

His journey to Birmingham in the middle of the night by an expensive form of transport, viz. taxi;

(3)

The fact that the deceased’s dressing gown cord drenched in the deceased’s blood was found in the appellant’s flat;

(4)

The fact that a block, in the form of a calor gas container, had been put inside the door of the deceased’s flat by someone who then left inconspicuously by the back door and by ladder over the rooftops;

(5)

The appellant’s arrival in Birmingham without the carrier bag which he had with him on leaving Devon but with over £2,500 in cash when the deceased was known to have in his flat cash which was missing and the appellant was known to have been in debt before he left Teignmouth;

(6)

The appellants admitted lies

(a)

to his relations and friends in Birmingham, that he had won his money on horses;

(b)

to the police, that he had taken £1,200 when he had, in fact, taken over £2,500;

(7)

The appellant’s unadmitted but proved lies

(a)

to those in Birmingham, that he had left a kitbag on a lorry in which he had got a lift;

(b)

to the police, that he had been wearing on Saturday the yellowish shirt and blue jeans in which he arrived in Birmingham, when the evidence of those in the Queensbury public house was that he had been wearing a white top and black trousers;

(c)

to the police, that injuries sustained to his knuckles were the result of hitting a punchbag at the Queensbury when the evidence was that the boxing gloves which the appellant put on bore no traces of blood;

(8)

The absence of any explanation on the part of the appellant for

(a)

his activities during the 3 hour period elapsing between his departure from the Queensbury and his first taxi ride;

(b)

the presence of the deceased’s dressing gown cord in his flat;

(c)

the contrast between his impecuniosity in Teignmouth and his suddenly acquired wealth in Birmingham

(9)

The implausibility of supposing that an unknown murderer could have battered the deceased to death, decided to exit the house through the appellant’s flat, dropped the deceased’s dressing gown cord there and, before leaving, have decided to deposit a gas canister behind the door of the flat.

For all these reasons, in our judgment the prosecution had a completely compelling case, even if Lewis had been successfully discredited and even though the evidence can fairly be described as circumstantial.

The critical paragraphs of the reference from the Criminal Cases Review Commission state as follows:-

“9.68

The Commission considers that had Michael Lewis’ convictions been disclosed, there is a realistic possibility that the defence would have cross-examined him about them. This would inevitably have had an impact on the credibility of his evidence, evidence described by the single judge at the appeal stage as ‘deadly’. The prosecution case against Mr Underwood was strong but the only direct evidence against him was that of the confession to Michael Lewis. If the defence could have cast sufficient doubt on the reliability of Mr Lewis’ evidence, the prosecution case would have been wholly circumstantial.

9.69

The question for the Commission is therefore whether there is a real possibility that the Court of Appeal would consider that the jury might have acquitted Mr Underwood had Mr Lewis’ convictions been disclosed. In the light of the above, the Commission considers that there is such a possibility.”

We naturally respect the conclusion of the Commission that there is a real possibility that the jury would have acquitted the appellant if Lewis’s convictions had been disclosed. But it seems to us to amount to a view that, however strong the circumstantial evidence may be, any verdict that depends upon circumstantial evidence alone is arguably unsafe. Having now heard full argument that the verdict is unsafe, we are sure for the reasons we have given that it is not unsafe. On no view was it just Lewis’s evidence that was “deadly” evidence against the appellant. In these circumstances, our duty is to say that the verdict is a safe one and the appeal will be dismissed.