England and Wales Court of Appeal (Criminal Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CRIMINAL DIVISION)
ON APPEAL FROM CROWN COURT AT BIRMINGHAM
Mr Justice Gross and a jury

		Royal Courts of Justice Strand, London, WC2A 2LL
		3rd May 2005

Lord Justice Judge:

1. This is an appeal by Chaha'oh Niyol Kai-Whitewind against her conviction on 10th December 2003 at Birmingham Crown Court before Gross J and a jury of murder. The deceased was the youngest of her three children, Bidziil, who was born on 9th May 2002 and died on 1st August 2002.
2. The issue at trial was whether the child was indeed the victim of a deliberate killing, as the Crown alleged (and the jury found), or died from natural, even if unexplained, causes.
3. There are two distinct grounds of appeal. First, it is argued that the conviction, said to be "entirely based" on conflicting expert opinions, cannot survive the decision of this Court in R v Cannings [2004] EWCA Crim 1, [2004] 2 CAR 7 p. 63. Second and alternatively, it is submitted that fresh evidence serves to undermine the safety of the conviction: if this evidence had been available to the jury, its decision might reasonably have been different (R v Pendleton [2002] 1 WLR 72).

Sudden Infant Death Syndrome

4. In criminal trials in which the death of an infant or infants is being investigated, terms like "cot" or "crib" death, or sudden infant death syndrome (SIDS) or an "unascertained" or "unexplained" or "undetermined" cause of death are frequently used. Unfortunately those who use these words (even very highly regarded experts) do not always intend to convey precisely the same meaning, or understand the word in the same way. This uncertainty of definition can give rise to misunderstanding. In the CESDI SUDI Study (1993-1996) into Sudden Unexpected Deaths in Infancy this problem is highlighted in a section entitled "Inconsistencies in the definition of SIDS". The text which follows is self-explanatory:

"The definition of SIDS allows considerable scope for inconsistency … Two pathologists making the same findings may not agree on whether they constitute an adequate explanation for death: for example, if there are signs of respiratory infection one might give this as the cause of death, while the other might regard it as coincidental and classify the death as SIDS. In addition, pathologists and coroners may vary in their readiness to accept SIDS as a registered cause of death, sometimes preferring terms such as … "unascertained". … These variations may give rise to inconsistency in the reporting of SIDS in different places in the United Kingdom. Inevitably, there will be similar inconsistencies between different countries, so that international comparisons of SIDS rates should be interpreted with caution … The exact meaning of the lay term "cot death" ("crib death" in the USA) which was coined by Barratt in 1954, is still the subject of debate. Some people restrict it to deaths meeting the Beckwith criteria for SIDS, while others give it the wider meaning of any infant death that occurs suddenly and unexpectedly, regardless of whether or not a cause is subsequently determined. Besides this impression, the term can be misleading by its implication that death always occurs in the cot, which, although usual, is not invariable.

Finally, the grouping of the deaths in the Confidential Enquiry of Stillbirths and Deaths in Infancy, CESDI [study], which was given the acronym SUDI (sudden unexpected deaths in infancy) does not correspond with any single ICD or ONS classification and includes various other unexpected deaths as well as SIDS."

(ONS is the Office for National Statistics: ICD is the International Classification of Diseases.)

5. In Byard's Sudden Death in Infancy, Childhood and Adolescence (2nd edition – 2004) to which our attention was drawn, precisely the same point was made.

"SIDS appears destined to continue to be a difficult, contentious and emotive term that can, unfortunately, be used very easily as a "diagnostic dustbin" to disguise incomplete investigations and inaccurate conclusions … It is a rather disappointing fact that debate continues about the most appropriate definition of SIDS … Our understanding of the pathogenesis of SIDS is still incomplete, and this is reflected in the vast number of often contradictory papers that have been published in recent years."

6. In Cannings itself the Court endeavoured to explain:

"The convenient acronym SIDS requires a little amplification, particularly in relation to the last "S", which stands for syndrome. Treating the problem as a syndrome tends to obscure the fact that sudden unexplained infant deaths occur in different circumstances, and some may be multi-factorial, the result of a coincidence of processes which, taken in isolation, would not necessarily cause death. No underlying condition for every death categorised as SIDS has been identified. The critical point of each such death is that it is indeed unexplained, and that its cause or causes, although natural, is or are as yet unknown. SIDS does not apply to deaths, or if already attributed to SIDS, ceases to apply to deaths which are clinically explicable or consequent on demonstrable trauma. In each SIDS case the mechanism of death is the same, apnoea, loss of breath or cessation of breathing. In the true SIDS case we do not know why the particular infant's breathing stopped. All we know is that for some unexplained reason it did. One obvious reason for loss of breath is smothering or some deliberate interference with the infant's normal breathing process. However the same process, with the same result, also occurs naturally."

7. This analysis was based on the evidence then before the Court. However, it will become apparent when we examine the second ground of appeal that cases are sometimes categorised as SIDS, notwithstanding that an explanation, for example, accidental suffocation caused by overlaying (when an infant is sleeping with an adult) is available. Some experts regard such an accident as falling within the SIDS classification: others do not. Yet whether the infant has been accidentally or deliberately smothered, the post mortem changes will be similar. These differences of view can be crucial. For the present, however, we simply highlight the importance both to courts, and to expert witnesses giving evidence by reference to the results of research in this field, of identifying precisely what the individual expert or researcher has or had in mind by the description of a "cot" or "SIDS" death.
8. We shall not attempt any compendious redefinitions. Where appropriate, however, we shall comment on the way in which individual terms are used by the expert whose evidence is being analysed.

The Facts

9. The appellant was married to Kim Whitewind. He was the father of her eldest child, who was four years old at the relevant time, but not of her daughter, then two years old. According to the appellant he was not the father of Bidziil, who was conceived in the course of a rape.
10. Bidziil was born by Caesarean section, a healthy little boy. His birth weight was 4340 grammes at 39 weeks gestation. Although there was a history of maternal diabetes, the baby's heart structure was sound, and his death was not linked to his mother's condition.
11. On 21st May, when Bidziil was twelve days old, the appellant was visited at her home at 2 Maas Road in Birmingham by a health visitor, Ms Hannaford. It was an introductory visit. During the course of the conversation, the appellant told her about the circumstances of Bidziil's conception and that he had been an unwanted baby. His father was not her husband, Kim, who she told Ms Hannaford, was nevertheless supportive. She also told her that she was suffering from depression, and seeing the community psychiatric nurse, but that she was not taking anti-depressants because she was breast feeding her baby. As summarised by Gross J, she told Ms Hannaford that she had once felt like placing a pillow over her daughter's face and also, that there was a fleeting moment when, for no particular reason, she felt like killing Bidziil. She had dealt with that by putting the baby down and walking away. When Ms Hannaford asked how she would deal with such a situation if it happened again, she said that she did not know.
12. Ms Hannaford said that it was not unusual to hear mothers say that they felt like killing their baby and she placed no particular reliance on the comment, but she agreed that she had never before heard a mother express thoughts as specific as placing a pillow over her baby's face. She felt it appropriate to revisit the issue of anti-depressants with the community nurse. She opened a supplementary file on the case "because of what the defendant had said to her, her depression, and her feelings towards Bidziil with regard to the manner of his conception". Subject to those concerns, she felt that the appellant was a caring mother, and Bidziil appeared in good health. Arrangements were made for Ms Hannaford to meet again with the appellant.
13. In the morning of 6th June the appellant telephoned NHS Direct, and reported that while Kim Whitewind was changing the baby's nappy, Bidziil was "kicking about so that my husband took hold of his legs and opened them so that he could clean the baby. And he heard a click of the right hip. There was a clicking noise, he put the nappy on the baby, left the baby on the bed and he hasn't moved his leg since". She said that the baby was now up against her shoulder and had gone to sleep. She said that he was being breast fed, and that as she was "on medication so he is getting a small amount of sedative". She was advised to take the baby to hospital immediately.
14. Bidziil was admitted to Birmingham Children's Hospital suffering from a spiral fracture of the right femur and bruising to his right elbow. The leg was put in plaster. No rib fracture was diagnosed. The right leg was put in plaster. Kim Whitewind was subsequently arrested and admitted responsibility for the injuries, asserting that they were accidental. During the investigation and treatment of Bidziil's injuries, nothing which indicated that the baby was subject to a blood clotting, or indeed any other blood or natural disorder, was discovered.
15. Social Services intervened. When Bidziil was discharged from hospital, arrangements were made for him to live with the appellant's mother, Ann Rofique (now deceased) at her home in Stirchley. He lived with his maternal grandmother until shortly before his death. Kim Whitewind was allowed supervised visits. The appellant continued to live with him at 2 Maas Road. She expressed milk so that the baby could be fed with a bottle. While there, Bidziil was visited on two occasions by the health visitor. He appeared to be thriving, and there were no concerns about his health. The health visitor believed that the appellant was coming to terms with her new baby and bonding well with him. She had no concerns about the quality of care he was receiving, but she continued to be troubled by the appellant's depression.
16. On 26th July Bidziil left his maternal grandmother's home, and returned to live with the appellant. The health visitor spoke to the appellant on the telephone, and there was a discussion about the baby's feeding. The appellant wished to start breast feeding him again, but the baby was reluctant. This issue was later raised by the appellant in her interview with the police. She placed considerable reliance on the importance of breast feeding, believing it was a way in which she might more easily bond with her children, which was something she admitted to finding difficult. The telephone call on 26th July was the last communication between the appellant and the health visitor.
17. After Bidziil's death on 1st August, the appellant's earlier comments to the health visitor assumed much greater significance, and were plainly damaging to her defence. It was argued on her behalf that they should be put in proper context. Mr Cooper suggested that the evidence of the appellant's remarks should be counterbalanced by the evidence that she was a caring and loving mother, and accordingly they should not be held against her. He reminded us of Ms Hannaford's evidence that the appellant appeared to be bonding with the baby, and of her apparent distress on 1st August itself. He referred to the evidence of Mrs Woodstock, Mrs Nesbitt and Miss Hampshire. Miss Hampshire saw the baby with the appellant when he was living with his grandmother, saying that he received "lots of continuous attention from both of them". Mrs Woodstock never saw the appellant with the baby, and only spoke to her twice on the telephone at the end of July or beginning of August about the use of a breast pump. Mrs Nesbitt saw the appellant with Bidziil on a single occasion. When she saw her with her other children she described her as an apparently well-organised loving and caring mother.
18. At this stage of his short life, Bidziil had lived in his mother's home for less than one month. No problems, let alone problems which could be associated with his death, nor indeed any suggestions of ill health occurred while he was in hospital, nor when he was being cared for by his grandmother in her home. As we have noted, before he went into hospital, the appellant confessed to having entertained, even if fleetingly, the thought of killing him, and referred to another occasion when she felt like using a pillow to suffocate a child. Shortly after he left his grandmother's home, he was apparently unwell again. Within a week of his return to live with the appellant, he was dead.
19. On 27th July, the appellant telephoned NHS Direct and reported two episodes of vomiting by Bidziil. The first had taken place on the evening of 26th July, and the second on the morning of 27th July itself. She said:

"Yesterday he was coughing and he brought up green bearing in mind he is on SMA White he shouldn't be bringing up any green and my mum says not to worry about it it's probably just a cold, but today just before I phone you he brought up a whole load of brown water that's the only way I can describe it."

When the nurse from NHS direct rang her back the appellant said:

"Yesterday he had a bit of a cough and … at the end of his feed he was a bit sick and he was bringing up green. Now because I am feeding him on SMA White …"

In answer to a question from the nurse, the appellant confirmed that Bidziil had vomited just the once, when he had brought up green vomit. She went on to say that he had diarrhoea three times in the night and he was bit wheezy in the morning, and that he had been sick "now three times" and on each occasion had brought up a brown fluid like a brown water. In answer to subsequent questions, the appellant said that it was "brown fluid … almost transparent with bits of brown in it."

20. Advice was given to the appellant to carry on with the feeding, and it was suggested that the appellant should ring back if things were getting worse, or the baby was refusing his food.
21. When later interviewed by the police the appellant said that on the first occasion Bidziil brought up what she described as green mucus, "…. More like over-boiled cabbage than green mucus, it was more solid than mucus". There was not a lot of it, "just one lump". On the second, he brought up what looked to her like dark brown water. "It was a transparent brown … a lot of brown liquid". She was given advice by NHS Direct.
22. The matter rested there. There was no further communication by the appellant with NHS Direct, nor any suggestion then, or later in interview, that there had been any further recurrence of vomiting or that the baby had not fully recovered from any tummy upset. Indeed he was described as fit and well on 31st July by his aunt, the appellant's sister, with whom he spent the day, and a "picture of health" on 1st August itself when he was seen at his grandmother's home by another witness.
23. On 29th July according to her account given in the subsequent police interview, the appellant and Bidziil were passengers in a taxi, when Bidziil's pushchair fell over while he was sitting in it. At trial the defence did not suggest that any significance whatever attached to this incident, and it ceased to have any relevance to the issues before the jury.

1st August

24. The appellant took Bidziil to her mother's home for a supervised visit by Kim Whitewind to the appellant's three children. The visit lasted from about 9.40 am until 11.30 am. Bidziil was laughing, and appeared to be in good health. When her mother left for work, the appellant took her children home. She was in sole charge of Bidziil from then until he died. The only accounts that we have of events between then, and just after 5.30, came from the appellant herself. These were not given in evidence at trial, but in the course of telephone calls with NHS Direct and later police interviews.
25. At about 3.00 pm the appellant attempted to feed Bidziil. He would not feed. She noticed specks or spots of blood on the roof of his mouth. She bottle-fed him. She then walked a short distance to a branch of Boots. She took Bidziil with her in his pushchair. She made a purchase timed at 3.23. They then went home. At some stage while they were out, Bidziil developed a nosebleed. This is an important feature of the evidence for at least two reasons. First, spontaneous nosebleed in an infant is extremely rare: second, from the findings at post mortems it emerged that by the time when bleeding started in Bidziil's nose, he was already in deep distress and death was imminent.
26. Shortly before 4.50 the appellant telephoned her sister, and said that Bidziil was not well. She was concerned about the spots of blood on the roof of his mouth and, for present purposes, importantly, she reported the nosebleed. She described the blood as a "trickle": it was not pouring. Her sister advised her to contact her GP or NHS Direct.
27. The appellant made two telephone calls to NHS Direct, the first at 4.49 pm. She spoke to a call operator, and said that her baby was conscious and breathing, but had worked himself into a state during breast feeding. She noticed that there were blood speckles in the roof of his mouth. She told the operator that they had left the house for about twenty minutes and that he had had a nosebleed while they were out. She reported that he was now asleep in his pushchair. In truth, he was probably already dead. The call operator told her that a nurse would telephone back. When NHS Direct attempted to telephone the appellant, there was no answer.
28. The appellant telephoned again at 5.10 pm. She spoke to a nurse. The conversation was tape recorded. She repeated that the baby had worked himself up into a state when she was feeding him earlier and that she "noticed that the top of his mouth, the roof of his mouth was speckled with blood". This did not seem to be excessive, and was not bothering the baby, so she "let him carry on with his feed". Then they went out for a bit, for about twenty minutes, and "when we came back, I was about to get him out of his pushchair and noticed that he had had a nosebleed." She was asked whether the baby was "talking, sorry crying and responding to you normally?" She said "Yes". She was asked whether there were any new breathing problems. She said that there were not. He was not going blue but was a lot paler than usual. He had not been injured and he was no longer bleeding. He was pale, but not cold and clammy. She was asked about the bleeding from the nose and said that "It wasn't bleeding when we left the house and when we got back twenty minutes later it was."
29. She resumed the conversation with a nurse and reported the vomiting incidents a few days earlier. The nurse asked how the baby was, and she said that he was asleep and that she had not tried to rouse him. The nurse asked her to try and wake the baby and she said that she would just go and do that. She then said that she had taken him out of his pushchair and that she was holding him and he was not opening his eyes. The nurse asked whether he was breathing. She responded "Oh God" and then was heard on the phone saying "I can't find a pulse, I can't find a pulse" and then crying, telling herself to calm down, and in something of a panic. She was told that the ambulance was on its way and when asked whether the baby was breathing said that she had found a "very, very weak pulse", but that the baby was "very very cold".
30. It was alleged that the appellant then made a telephone call to Mrs Woodstock, the National Childbirth Trust volunteer, to cancel their meeting to discuss the possible use of a breast pump, because "the baby was being taken to hospital in an ambulance". If this evidence was accurate, the telephone call was made after the second call to NHS Direct, but before the police arrived. In the result, given the particular circumstances, it would be wrong to draw any adverse inferences against the appellant if she did in fact make that particular call at the time alleged. On the other hand, there was powerful evidence that on the account given by the appellant to NHS Direct, Bidziil had been dead for some time before the first telephone call to NHS Direct, and considerable reservation about the accuracy of her description of him as conscious and breathing.
31. The police arrived at the scene at just after 5.30. Bidziil was propped up on the sofa. The appellant told the police that the child had stopped breathing. One officer tried to help, but was unable to find a pulse. He tried mouth-to-mouth resuscitation. The baby felt very cold and looked pale and slightly blue. He noticed dried blood in the baby's right nostril. After trying mouth-to-mouth resuscitation, the officer attempted resuscitation using the palm of his hand by giving Bidziil five chest compressions. There was no reaction from the baby. He tried the same sequence again, yet again without success. The police officer agreed that it would have been better to use two fingers rather than the palm of his hand, but he was doing the best he could to save the baby's life. He applied a degree of force, but the pressure was gentle. The other officer was also aware that it would be more appropriate for fingers to be used for this purpose, but this was a traumatic incident. The attempts at resuscitation failed. They were waiting for the ambulance to arrive while the appellant was panicking. Eventually the ambulance arrived.
32. A paramedic ran into the house and ran back out, carrying the baby to the ambulance. In the brief time necessary for this purpose, it was observed that the baby was limp and cyanosed, blue around the mouth. Bidziil was there connected up to a heart monitor. There was no reaction. Compressions and ventilation were tried. There was no pulse and no response. The senior paramedic thought the baby was already dead. He was very white and very pale, and very clean. He observed blood staining on his tongue. Just because the patient was a baby, attempts at resuscitation continued.
33. Bidziil arrived at Birmingham Children's Hospital at 5.45 pm. He was not breathing. There was no heart beat. He was intubated and attempts were made to resuscitate him. Adrenalin and sodium bicarbonate were administered to assist his circulation. There was a complete lack of response. Bidziil was pronounced dead at 5.52 pm.
34. At death Bidziil was described as a well-nourished twelve week old infant, without any external sign or mark of injury, save of course the unusual feature for an infant of his age, the dried blood in the nose, which continued to evidence the earlier nose bleed.

Police Investigation

35. During the course of the police investigation, what was described as the T shirt worn by the appellant on the afternoon of 1st August was obtained. Blood staining was found, predominantly on the left side of the front of the T shirt. The blood in the region of the left shoulder was possibly a little diluted, but the staining appeared to be the result of drip staining from above, and was consistent with the appellant holding Bidziil while he was bleeding from his nose. The trial, and the appeal before us proceeded on the basis that the blood was Bidziil's blood, and the appellant never suggested that it came from some other, different source.
36. On 2nd August the appellant attended the police station at Bourneville Lane. She was not under arrest. An appropriate adult was present together with a representative of her solicitors. Kim Whitewind also attended the police station, and he, too, was interviewed. During the course of the interview the appellant gave a narrative account of events.
37. She explained how the baby had woken up in the afternoon. She had tried to breast feed him, but "He wouldn't take it and he looked quite cross, so I gave him his bottle and he had about 5 ½ ounces". She then took him out to his pushchair, and after a child-like lack of co-operation by her two older children, they went together to Boots. She left the baby in his pushchair outside the back door with the back door open, and took the other two children into the house. After seeing to them she "went to get Petee [Bidziil] out of his pushchair and … noticed that he had had a nosebleed." She telephoned her sister, and then NHS Direct. She left the baby in the pushchair because she "didn't want to have him lay down too much and have his nose start bleeding again." When the nurse asked if she could wake the baby she went to the pushchair and tried to do so. She got him out of the pushchair, spoke to him, shouted at him, scratched his legs but "he didn't move at all". She then reported that she couldn't wake the baby. She was asked if she could find a pulse. She tried "his face, throat, and couldn't find one. Tried his wrist and couldn't find one", and when she calmed down, she put the back of her hand against his bare chest inside his vest and although to start with she couldn't find one, "I could feel a very faint weak heartbeat".
38. The appellant said nothing which could be construed as an admission of responsibility for Bidziil's death, and she asserted that she had done nothing to cause it. In particular she denied smothering or stopping him from breathing. When she was asked the direct question whether she thought there was anything that she may have done which contributed to her son's death, she replied:

"In my own head making him cry when I was trying to breast feed him. But I know that couldn't have caused his death because he bottle fed afterwards. So no."

When asked when she was last satisfied that the baby was alive she said, "for definite", he was alive before they went out, and that she was not concerned until she noticed the nosebleed. She put the timing of that at about 4 pm, but in view of the till receipt timed at 3.23, her timing was plainly wrong. She repeated that she noticed the nosebleed when she brought him back into the house after her visit to the shops. She confirmed that apart from her other children, no-one else was present in the house during the afternoon.

39. Much later, on 24th October 2002, a further short interview took place. The appellant declined to answer any questions, save to complain that the interviewing officer had mispronounced the child's name. The judge directed the jury to disregard the appellant's response, and to draw no adverse inference against her.

Post mortem examination

40. At about 7.30 pm Bidziil's body was examined by Dr Debelle, a consultant paediatrician at the hospital. He noticed some blue/purple discolouration to parts of Bidziil's body as well as signs of rigor mortis around his neck. In the result these findings were not relied on by the Crown, and we disregard them. However Dr Debelle was also struck by the account of a nosebleed. Unlike adults, spontaneous nosebleed in an infant is exceptional. In all his experience he could not recall such a case in a baby of Bidziil's age. He was concerned that this was an unnatural death.
41. The first post mortem was conducted on 1st August at 11 pm by Dr Edmund Tapp, a Home Office pathologist. He found bloodstained fluid emerging from the nose and mouth without any injury to account for it. He also noted bloodstained fluid in the airways and lungs, which were congested and haemorrhagic. He did not carry out any microscopic examination of the lung tissue sections. On the basis of his examination he thought that a cot death, or natural death, was "quite possible", but on reading the later report of Dr Cox, he thought that the appropriate description of the cause of death was "unascertained" rather than SIDS, and he agreed that this would not rule out natural causes.
42. Legitimate criticism was directed on the appellant's behalf at this first post mortem. For example, Dr Rushton argued that the proper protocol for the conduct of autopsies on infants of Bidziil's age had not been followed. Further virology and bacteriological tests should have been carried out. Insufficient samples were taken fully to test for the presence of infection. Later, Professor Rutty said that he would have considered the skeletal x-rays before post mortem because they would have shown the soft tissue of the lungs and highlighted any relevant pathology. He was also troubled at the failure to take a blood sample from Bidziil's heart for the purposes of coagulation and haematological study.
43. We note these criticisms. They are not directed to the findings actually made by Dr Tapp, and indeed so far as those findings, and his opinion about them were concerned, they were relied on by the defence. The criticisms are directed to the absence of material for checking whether any further, natural explanation for Bidziil's death might have become apparent. It is not suggested that they would have done so: simply that without the additional checks and tests their presence (if they were present) would not have been detected. For example, criticism of the failure to examine the skeletal x-rays prior to conducting the post mortem advances the defence no further because, when they were examined, no relevant abnormality was detected.
44. The second post mortem was conducted on 6th August 2002 by Dr Philip Cox, a consultant perinatal pathologist specialising in infant deaths. This post mortem was carried out in the presence of Dr David Rushton, a paediatric pathologist attending on the appellant's behalf. Dr Cox carried out a microscopic examination of the lung tissue sections. He found extensive fresh bleeding into the air spaces of all five lobes of the lung and fresh bleeding into the solid interstitial lung tissue. In addition he found iron laden microphages, widely distributed in the lungs, which would have taken at least two days to form, consistent with an episode of bleeding into the lungs at least two days before Bidziil's death.
45. Dr Cox also found evidence of minor peribronchial inflammation around the major air passages in the lungs, but these, like the enlargement of Bidziil's thymus gland, are common findings without special significance. Dr Cox agreed that testing for viruses would have been helpful, but he ruled out viral infection as a possible cause of death. The histological examination produced nothing to suggest any natural cause of death. If the bleeding which caused Bidziil's death had been caused by a virus, then the cellular reaction observed by Dr Cox in the histological examination would have been much greater. He could find no evidence to indicate a natural cause of death. There was no suggestion of infection, heart failure or abnormality, or fluid overdose. The immediate cause of death was lack of oxygen, resulting in asphyxiation, and Dr Cox could not find any alternative explanation, either for the new or the old blood in the lungs. His evidence was consistent with two distinct episodes of upper airway obstruction, the first at least two days before Bidziil's death, evidenced by the iron laden macrophages in his lung, and the second evidenced by the nosebleed on the day of his death, which resulted from the fresh bleeding found in the lung.
46. A blood specimen taken at the time or shortly before death, and the post mortem stomach contents of the infant, were analysed. Nothing by way of drugs, or alcohol, or antidepressants were found. Three further samples from Bidziil were analysed for microbiology purposes. These came from the accident and emergency department rather than post mortem. They were analysed for bacterial pathogens. The first sample was taken while Bidziil was in hospital receiving treatment for the fractured leg. No pathogens were found in the faeces. The other samples were taken very shortly after Bidziil's death. No pathogens were found in a nasopharyngeal aspirate, and although coaglase negative staphylococci were isolated from a blood culture, they were irrelevant to the infant's death. Analysis was further made into the issue whether Bidziil had inherited any metabolic disorders. The investigation revealed nothing unusual. It was suggested that in the absence of any suspicion of infection, testing for viruses was probably pointless.
47. These findings were examined as a whole by Professor Risdon, consultant paediatric pathologist, and head of a department, at Great Ormond Street Hospital for 17 years, where he, or his department, investigated roughly one third of all infant deaths in the United Kingdom. He was particularly concerned about findings of blood emerging from the infant's nose, with blood or bloodstained fluids in the airways and lungs, together with the findings of haemosiderin laden macrophages. In his experience, and this was common ground among all the professional witnesses, a spontaneous nosebleed in a baby is extremely uncommon. The most likely explanation is bleeding in the lungs. In the case of Bidziil this diagnosis was reinforced by the evidence of fresh blood in the airways and lungs. The iron laden macrophages demonstrated an earlier episode of bleeding, at least 48 hours before death. This was an extensive bleed and haemosiderin was present in all three lobes on the right hand side and both lobes on the left hand side of the lungs. Fresh blood was seen in all sections and airways. Professor Risdon was convinced that there were two distinct episodes of bleeding into Bidziil's lungs.
48. He considered whether either or both episodes of bleeding could be attributed to natural causes. He rejected the possibility of heart failure (there was no abnormality in the heart); chronic arrhythmia (there was no history or supporting evidence); bleeding complications, unlikely unless there was some inherited condition; resuscitation efforts (inconsistent with the old blood in the lungs, and in relation to the time of the nosebleed as it related to Bidziil's death, inconsistent with the absence of circulation); lung inflammation (insignificant in the absence of other signs suggesting infection); viral pneumonia (similarly); inflammation of the meninges (irrelevant to the blood found in the lungs); virus infection (no evidence, where some consistent finding would be expected).
49. Professor Risdon also addressed:

(a) Infection.

Even if the haemosiderin laden macrophages were caused by infection (itself not a common complication of infection) the infection would have to be severe, and the child would be obviously unwell, and the site of infection would be visible somewhere in the body or in the lungs themselves.

(b) Bleeding complications

Bleeding complications were unlikely suddenly to produce blood in the lungs, and any abnormalities would "almost certainly" have emerged when the infant was treated for the fracture of the femur. Although bleeding would have affected all parts of the lung, it was not sufficiently extensive for the child to have drowned in blood. The immediate cause of death was lack of oxygen to the brain.

(c) Vomiting

If the infant had inhaled vomit, that would have been apparent at post mortem or on histological examination. This vomiting was not associated with haemorrhage. The vomit episodes did not suggest lung infection or fatal gut infection on 1st August.

(d) Meninges

There was no evidence that Bidziil's brain was swollen, and the inflammation of the meninges would have had nothing to do with the blood found in his lungs.

(e) Petechial haemorrhages

The absence of petechial haemorrhages did not undermine his conclusion, nor did the absence of evidence of oxygen starvation when the brain was examined using conventional techniques. The first episode had not been of sufficient severity to produce changes to the brain: it was somewhere in the middle of the scale. There had been insufficient time for changes in the brain to develop immediately prior to death.

(f) Breast feeding

He dismissed an accident through the course of breast feeding.

50. A number of other matters were raised with him on the appellant's behalf. Professor Risdon rejected them as irrelevant. These included the weight of the lungs, the temperature of the baby on admission to hospital, and the finding of rigor mortis, and the time of day, which he regarded as too non-specific to be helpful.
51. Professor Risdon rejected SIDS or cot death. The findings at post mortem of fresh and old blood in the lungs were highly significant. His conclusion was that obstruction of the airways was "far and away the most likely" cause of bleeding into the lungs. The overall findings meant that he could not think of any alternative explanation to asphyxia. It was not for Professor Risdon to assert that the asphyxia was non-accidental, but if the obstruction to the airways was indeed accidental, it would have happened by accident twice.
52. Dr Bonshek is a consultant ophthalmic pathologist. He examined both Bidziil's eyes, received some five days after death. There was recent haemorrhage in the left eye, and the lenses of both eyes showed signs of swelling. Attention was drawn to gliosis, evidence that repair was taking place in the cells of the retina. This was consistent with deprivation of oxygen, of potential significance to hypoxia. The pattern of bleeding, however, was not symptomatic of hypoxia, which was more associated with spot-like bleeding. In any event these findings were not diagnostic of oxygen deprivation. Gross J was later to observe that he had found, and the jury too might find that this evidence had been difficult to follow. Perhaps none of them was helped by the fact that there were three attempts at cross-examination, and two at re-examination.
53. At the close of the prosecution case it was submitted to Gross J that there was no case to answer. Our summary of the available evidence demonstrates why it was right for him to reject it. Indeed if he had allowed this submission, he would have wrongly usurped the function of the jury.

Defence Case

54. The appellant elected not to give evidence. The judge conducted the appropriate investigation in open court in the presence of the jury. It was conceded, and rightly, that the judge would be justified and was certainly entitled to give the jury an appropriate direction about the possible impact of her absence from the witness box. In the result the jury had no evidence from the only adult with whom Bidziil spent his last hours, and who was the only witness to the events during the fateful afternoon of 1st August. The direction given in due course was impeccable.
55. Dr David Rushton, consultant paediatric and perinatal pathologist, with many years' experience as a consultant, who conducted the second post mortem on Bidziil with Dr Cox, gave evidence for the defence.
56. He was critical of Dr Tapp's first post mortem, pointing out, as was accepted by the Crown, that the protocol for the conduct of autopsies on infants of that age had not been followed. He further criticised the absence of insufficient samples to test for the possible presence of infection. In his view, further virology and bacteriological tests should have been carried out. There was a growing body of literature which suggested that a combination of viruses and bacteria operating together contributed to sudden infant deaths.
57. Dr Rushton agreed that there was evidence of two distinct episodes of bleeding in Bidziil's lungs. Although this bleeding was consistent with asphyxia as a possible cause of death, it should be treated as unascertained because the absence of an alternative explanation did not necessarily mean that there was none. He also agreed that a nosebleed in a baby would be extremely uncommon, but he suggested that there was no sufficient detail about the nosebleed from which it would be safe to conclude that a smothering had taken place.
58. He thought that the vomiting may have been significant, and that what was described as brown vomit may have been blood. If the vomit had aspirated into the lungs, and had contained blood that could account for the macrophages found there. In his view, contrary to that of Professor Risdon, and later Professor Rutty called for the defence, the pattern of bleeding in the lungs would also have been unusual for a death caused by asphyxia. The structure of the macrophages was unusual. They were multiple nuclei, which suggested that something unusual was going on. The inflammation found in the lungs and the unusual structure of the macrophages suggested an infective process.
59. When cross-examined, Dr Rushton agreed that there was evidence of extensive bleeding in the lungs which could be linked to Dr Tapp's findings of blood in the throat and bronchii. Blood had come out of Bidziil's nose and mouth and dripped onto the appellant's top, and most significant, the blood from Bidziil's nose must have emerged after his lungs had started bleeding. "This is a very rapid, very extensive episode involving bleeding in the lungs, leading to the collapse of lungs and blood flowing up and out of them into the throat, out of the mouth and nose, and the almost instantaneous death of the baby".
60. Dr Rushton accepted that asphyxia was a possible cause of death, but just because he could not identify an alternative possibility, it did not follow that Bidziil had been asphyxiated. A great deal about the death of young babies was still unknown, and if he knew of a possible mechanism which had caused the bleeding in Bidziil's lungs immediately before his death, he would, of course, have described it. Dr Rushton believed that a virus could explain the earlier episodes of vomiting and the bleeding in the lungs, and that bacteria could have been generating toxins which, in combination with a virus, might have caused heart failure which itself led the lungs to bleed until they collapsed. The effect would be similar to cases of toxic shock syndrome.
61. Dr Rushton's evidence on this topic exemplified what the judge in his summing up described as "number of unfortunate difficulties" with his evidence. Despite the emphasis on toxic shock syndrome, no earlier mention had been made of it in his report, and indeed Professor Risdon was not asked any questions whatever on the topic during cross-examination. Another problem arose from Dr Rushton's evidence about the size of Bidziil's thymus, which led him to tell the jury that evidence recently to hand from the Netherlands had suggested provisionally that an enlarged thymus may have some impact on the immune system of the infant. However when he was cross-examined, he observed that his remark on the subject was an aside, not intended to apply to Bidziil's death. He was then asked his source for the aside. Initially he said that he could not remember the journal in which he had found it. In the end it turned out that he had never read the article in question. He had read a summary of it in a medical newspaper, and he was unable to discover the location either of the newspaper or of the journal itself. Then again, in the context of the notorious case of Sally Clarke, he referred to botulism and tetanus, later underlining that he was not saying that either botulism or tetanus had any relevance to Bidziil's death. They were used only for the purposes of illustration. There were, indeed, a number of features of Dr Rushton's evidence which would have entitled the jury to approach his evidence with legitimate reservations.
62. His overall view, as summarised by the judge went as follows: (1) the virus in the body which could have been anywhere, not necessarily in the lungs, could explain the vomiting. (2) The vomiting could explain the first episode of bleeding in the lungs. (3) Bacteria in the body could have generated toxins. (4) The potency of the toxins could have been enhanced by the virus acting in combination with the bacteria. (5) These lethal toxins could have resulted in heart failure. (6) The heart failure could have caused the lungs to bleed, indeed such failure could have given rise to a fulminant haemorrhagic pulmonary oedema; that is, a very rapid, very extensive episode involving bleeding and fluid in the lungs leading to the collapse of the lungs, blood flowing up and out of them into the throat, out of the mouth and nose and the almost instantaneous death of the baby. This was bleeding episode 2. Dr Rushton postulated that episode 2 was or could have been a single dramatic incident involving the collapse of both lungs with the baby dying more or less literally on the spot. The toxins could have been staphylococcal toxins and their impact was identified as toxic shock syndrome.
63. Dr Rushton agreed that Bidziil's heart showed no sign of fault or structural defect. He also acknowledged, but was undeterred by the absence of any sign of aspiration or vomit in the histology or gastric contents in the microscopic examination of the lungs. An alternative mechanism to heart failure, was that the baby drowned in blood. He accepted that otherwise healthy babies would normally cough out matter they had inhaled, but this did not follow invariably.
64. On the basis that his thesis could explain what had happened, he subscribed initially to the view that the cause of death was unascertained, but in the end suggested that death by natural causes was more probable than unnatural death.
65. Professor Rutty, a professor of forensic pathology and a consultant pathologist to the Home Office also gave evidence on behalf of the appellant.
66. He was critical of the post mortem conducted by Dr Tapp. If he had been responsible, he would have looked at Bidziil's clinical history and x-rays. His personal practice would have been to take a blood sample from the heart of the deceased, to conduct coagulation and haematological studies, together with samples for virology and bacteria. He suggested that there were a large number of possible causes for bleeding in the lungs, some natural, some unnatural. On the basis of the autopsies, the cause of death was unascertained. He could not say whether Bidziil had died of natural causes or not.
67. When cross-examined, Dr Rutty agreed that there had been two distinct episodes of bleeding into the lungs, one 48 hours or more before death. The macrophages did not, of themselves, assist whether the cause of such bleeding was natural or unnatural, and similarly with the blood or bloodstained fluid emerging from the infant's nose and mouth, which could be attributed to either natural or unnatural causes. He did not disagree that there had been two distinct episodes of bleeding into the lungs, and that the source of the nosebleed was blood emerging from the lungs upwards through the throat and mouth, and that nosebleeds in a very young infant were most unlikely to be spontaneous. He could not identify any source for the blood found on the top the appellant was wearing other than the baby's nose or mouth or a combination of the two, and on the basis of the appellant's own account, the nosebleed could not have resulted from resuscitation or hospital procedures, all of which came much later in the narrative of events. He also said:

"The distribution of blood itself is important as bleeding into the airspaces themselves might suggest a natural bleeding disorder but when bleeding occurs in the interstitial tissues, although this can occur under natural circumstances, it does raise the high suspicion of raised intra-thoracic pressure, hence imposed airway obstruction. All of these features are present in this case."

68. He also stood by the conclusion expressed in his first report:

"In the absence of any provable natural disease, then the findings in this case entirely support a death due to asphyxia due to mechanical obstruction of the airways: that is, a smothering. And again, in the absence of any natural disease, the two bleeding episodes would have had a common cause, and the first episode could be described as a near miss asphyxiation."

69. He agreed that the distribution of bleedings in the lungs extending to all five lobes supported Professor Risdon's view that the cause of death was airways obstruction, and in disagreement with Dr Rushton, he thought that this bleeding was typical of asphyxiation. However he disagreed with Professor Risdon on the issue of testing into blood clotting or blood disorders. He had carried out this form of investigation for some years, but agreed that it was not done routinely. He accepted that Bidziil had shown no signs of bleeding disorder, and that none was discovered when he was treated for his fractured femur. If the blood in the lungs was the result of spontaneous bleeding caused by a bleeding disorder then he did not believe that a blood disorder could be excluded merely because the blood had spread throughout the lungs but nowhere else in the body.
70. He reaffirmed that spontaneous nosebleeds in infants of Bidziil's age were very rare indeed, and he could not think of any occasion when he had personally come across the phenomenon.
71. The summing up presented the jury with a detailed analysis of the relevant evidence given by all the professional witnesses, in particular Professor Risdon, Dr Rushton, and Professor Rutty. Towards the end of his summing up the judge decided to "recap on some of the highlights". In truth, this was a careful analysis in summary form of the rival contentions.

"Professor Risdon relied on the clinical history, the findings of Dr Tapp and Dr Cox, the presence, distribution and extent of blood in the lungs and the evidence as to two episodes of bleeding into the lungs. In effect, the tests which were not done would have been pointless. That was his view. Dr Gray, it could be said, supported him in large measure. Professor Risdon, having dismissed that long list of natural causes of bleeding into the lungs, what remained was an unnatural cause, asphyxia. Like Dr Cox, Professor Risdon could not think of any other explanation for the presence of the old and new blood in the lungs. The previous episode of bleeding involved a near miss asphyxia. On 1st August asphyxia was both the cause of bleeding into the lungs and the cause of death by depriving the brain of oxygen, the cause known as hypoxia. The presence of blood in the lungs could not determine whether the asphyxia was accidental or deliberate.

Professor Rutty, whom I take next, would agree with Professor Risdon that the bleeding, notwithstanding the absence of some commonly found features, pointed towards asphyxia and that the two bleeding episodes were likely to have a common cause, subject – but this is a most important subject – to the elimination of natural causes. It is that point which gives rise to the point of departure between Professors Risdon and Rutty.

Professor Rutty was not prepared to accept that natural causes of bleeding into the lungs could be eliminated. Here he was critical of the failures or gaps in the post mortems, especially the first conducted by Dr Tapp. Virology, bacteriology and blood testing were mentioned in particular. He could not say that the tests which were not undertaken would have been significant, but, unlike Professor Risdon, he thought they should be undertaken. He disagreed with Professor Risdon as to whether after death blood testing could be undertaken.

For these reasons Professor Rutty was not prepared to go further than to accept that there was blood in the lungs and that there had been two episodes of bleeding into the lungs. The cause of death, he said, was properly regarded as unascertained.

While Dr Rushton was minded to go further than Professor Rutty and to opine that the probable cause of bleeding into the lungs was the result of natural causes, he otherwise agreed with Professor Rutty in the result but not, in significant part, in the reasoning. He, too, was critical of the tests not done but he alone also said that he did not think that the evidence of bleeding and the macrophages were typical of asphyxia. He contemplated that the first and second episodes of bleeding into the lungs could have had two distinct causes; the first might be attributable to vomiting, the second involving a catastrophic single incident including heart failure was or might be attributable to toxic shock syndrome, that unholy alliance between a virus and a bacteria."

72. The summing up is not criticised, and indeed it is not open to criticism. The directions of law were clear and accurate. The summary of the evidence was fair and balanced, and the appellant's case was fully developed and explained to the jury, which convicted her.

Grounds of Appeal

R v Cannings

73. We must first address the decision of this Court in R v Cannings.
74. Mr Cooper argued that this conviction is unsafe on the basis of what he identified as the three stage test found in paragraph 178 of the judgment. First, there is a serious disagreement between reputable experts about the cause of death; second, a body of such expert opinion concludes that natural causes, whether explained or unexplained, cannot be excluded as a reasonable (and not fanciful) possibility; third, there is no additional cogent evidence, extraneous to the expert evidence, which tends to support the conclusion that the child was deliberately harmed. Although he accepts that this approach was said to apply to situations where more than one unexplained infant death occurred in the same family, he argued that there is no reason why such a test should not apply to a single, sudden, unexplained infant death. We understand that Cannings is being deployed in many cases by the defence as authority for different arguments running along the lines that whenever there is a genuine conflict of opinion between reputable experts, the prosecution should not proceed, or should be stopped, or that the evidence of the prosecution experts should be disregarded. If so, the single passage found in part of paragraph 178 in Cannings, taken in isolation, is being asked to sustain an unforeseen, and as we shall explain, inappropriate burden.
75. We must begin by revisiting the factual and legal context, which is critical to an understanding of Cannings and what was said in the penultimate paragraph of the lengthy judgment of the Court. Mrs Cannings was convicted of the murder of two of her children, namely Jason (who at the time of his death was 7 weeks old, in 1991) and Matthew (who was 4½ months old, in 1999). In addition, the trial considered evidence relating to the death of her first child, Gemma, in 1989, and various Acute or Apparent Life Threatening Events (ALTE) involving lack of or difficulty in breathing which did not lead to death, and which affected Jason, nine days before he died, Jade, a surviving child, born in 1996, when she was about 11 weeks old, and, albeit of less gravity, Matthew, some nine days before his death. At post mortems following the deaths, there were no pathological findings in relation to Gemma, and no significant unequivocal such finds in relation to Jason. Their deaths were unexplained, and both initially classified as SIDS. There were no relevant findings in relation to Matthew. However Matthew was the third infant death in the Cannings family.
76. The case for the Crown depended on specialist expert evidence about the conclusions to be drawn from the history of three infant deaths and ALTEs in the same family, which were said to be, and evidently are, extremely rare (para 129). Professor Sir Roy Meadow, called by the Crown, said of Jason's death that the fact that a previous child had died in the family was relevant because "that combination of circumstances, that sort of story is one that is very typical of a child who has died as a result of smothering". As a result, his diagnosis in his case was "probably smothering". When dealing with Jade, he said that the ALTE at the age of 3 months was "very unusual" and went on to observe that one of the reasons for an ALTE was smothering and "in the context of the family as a whole it is of importance". Finally, in relation to Matthew, he said:

"Firstly, the investigations and the pathologists did not find a reason for him dying. For me, the unusual feature is death so soon after being seen well, the fact that there had been previous deaths in the family and the fact that he had had an episode of some sort only nine days before he died that caused him to be assessed in hospital, because those features are ones that are found really quite commonly in children who have been smothered by their mothers. So the diagnosis for me, the clinical diagnosis, would be that this was characteristic of smothering… One then goes on to say 'Well, is it possible it is a condition that is not yet understood by doctors or described by them?', and that must be a possibility, but nevertheless as a doctor of children I am saying these features are those of smothering." (para 133)

This approach was supported by Dr Ward Platt, the thrust of whose opinion was that although the rarity of three infant deaths in the same family was not of itself a compelling reason to conclude that harm was deliberately inflicted, the pattern revealed by the history as a whole was compelling (para 137).

77. The Crown's approach therefore was to consider whether it was possible to identify one or other of the known natural causes of infant death and, if it was not, to conclude that the rarity of such incidents in the same family raised a powerful inference that the deaths must have resulted from deliberate harm – "lightening does not strike three times in the same place" (paras 10 and 11). In the result therefore the case depended on inferences drawn by expert witnesses, based on the multiplicity of incidents, each of which was unexplained. However, putting to one side the features relied on by the Crown to support the inference, as well as the general background which suggested the opposite, nothing, or virtually nothing at post mortem revealed factual material which assisted the Crown's experts and supported their view.
78. Summarising the submissions of the appellant, the Court noted:

"It is of course possible to smother a baby without leaving physical signs discernible on medical examination or at post mortem. Nevertheless, given that all four children were said by the Crown to have been subjected to violence sufficient to cause death, the absence of any physical signs of injury was somewhat surprising. There was no fresh copious bleeding in the lungs of the dead children and no petechial haemorrhage. There were no pressure marks to show as reddening in the area of the mouth and nose, nor blood or bloodstained fluid in the nose. No bruises were discovered on the outer skin surface or indeed subcutaneously. The fraenulum, in each case, was undamaged." (para 160)

In view of one of Mr Cooper's submissions in the context of this particular appeal, it is perhaps worth pausing to note that this passage demonstrates that physical signs of injury falling within the phrase, "indications of violence", referred to in paragraph 10 of the judgment, includes evidence beyond the kind of external injuries which a jury would readily understand, such as, for example, a fractured skull or a cut lip, and extends to internal findings within the body of the deceased such as fresh copious bleeding within the lungs, or blood or blood stained fluid within the nose, or petechial haemorrhages which would normally be identified, and explained to the jury by expert witnesses.

79. In summary, in Cannings the Crown's approach was to be contrasted with the reputable alternative view, that although three unexplained deaths in the same family are rare, if there was nothing to explain them, they remained unexplained and still, despite the known fact that some parents do smother their infant children, possible natural deaths (para 12).
80. In the light of the fresh evidence adduced on appeal in Cannings, the Court rejected the Crown's approach. Two paragraphs in the judgment demonstrate the reasons:

"12. … Nevertheless a degree of caution was necessary to avoid what might otherwise have been the hidden trap of taking the wrong starting point. If, for example, at post mortem it was positively established that Matthew's death had resulted from natural causes, the situation reverted to precisely where it stood before he died. The concerns which would have arisen as a result of his death – as the third in the sequence – would have been dissipated. There would have been a positive innocent explanation for the death, which would no longer be a SIDS, and might help to confirm that the earlier deaths were indeed natural deaths. Equally, if there were unequivocal evidence that one of these deaths, or even one of the ALTEs, had resulted from deliberate infliction of harm by the appellant, that would be likely to throw considerable light on the question whether the other deaths, or ALTEs, resulted from natural or unnatural causes. If, after full investigation, the deaths, or ALTEs, continued to be unexplained, and there was nothing to demonstrate that one or other incident had resulted from the deliberate infliction of harm, so far as the criminal process was concerned, the deaths continued properly to be regarded as SIDS, or more accurately, could not properly be treated as resulting from unlawful violence.

13. … Whether there are one, two or even three deaths, the exclusion of currently known natural causes of infant death does not establish that the death or deaths resulted from the deliberate infliction of harm. That represents not only the legal principle, which must be applied in any event, but, in addition, as we shall see, at the very least, it appears to us to coincide with the views of a reputable body of expert medical opinion."

81. As there was effectively nothing in Cannings to establish unnatural as opposed to natural death, and the basis of the case against her depended on inferences by one group of experts which were refuted by another reasonable body of medical opinion, the Court observed:

"… We have received significant and persuasive fresh evidence, which was not before the jury, some of it the result of further research, or research published post trial, into the problem of SIDS generally, and some specific to Mrs Cannings and her extended family. The expert evidence was absolutely critical to these convictions. In our judgment the fundamental basis of the Crown's case, based on the extreme rarity of three separate infant deaths in the same family, and the pattern of events in this particular family is, for the reasons we have given, demonstrably undermined." (para 175)

"… Throughout the process great care must be taken not to allow the rarity of these sad events, standing on their own, to be subsumed into an assumption or virtual assumption that the dead infants were deliberately killed, or consciously or unconsciously to regard the inability of the defendant to produce some convincing explanation for these deaths as providing a measure of support for the Prosecution's case. If on examination of all the evidence every possible known cause has been excluded, the cause remains unknown." (para 177)

The Court finally came to the passage relied on by Mr Cooper:

"… [F]or the time being, where a full investigation into two or more sudden unexplained infant deaths in the same family is followed by a serious disagreement between reputable experts about the cause of death, and a body of such expert opinion concludes that natural causes, whether explained or unexplained, cannot be excluded as a reasonable (and not a fanciful) possibility, the prosecution of a parent or parents for murder should not be started or continued, unless there is additional cogent evidence, extraneous to the expert evidence … which tends to support the conclusion that the infant, or where there is more than one death, one of the infants, was deliberately harmed. In cases like the present, if the outcome of the trial depends exclusively or almost exclusively on a serious disagreement between distinguished and reputable experts, it will often be unwise, and therefore unsafe, to proceed." (para 178)

82. A preliminary point is that these observations were not directed to the conduct of the trial by the judge. Initially they are directed to the prosecution process, but they would plainly be relevant to an argument that there was no case to answer, or indeed to that rare occasion when the judge considers whether to withdraw the case from the jury at the close of all the evidence. They provide an emphatic criticism of the prosecution's theory that the rarity of three deaths, not specifically identified as natural, raises an overwhelming, or virtually overwhelming inference, that the deaths resulted from the infliction of deliberate harm. This reasoning was flawed, and "for the time being" should not be applied. The words, "for the time being", make it abundantly clear that Court was not suggesting a fresh proposition of law of general effect, but rather focusing on the proper approach to cases like Cannings in the light of the state of contemporary medical knowledge. This is emphasised by the introductory words to paragraph 178 itself that "the trial, and this appeal, have proceeded in a most unusual context", followed by the observation that, "with unexplained infant deaths, however, as this judgment has demonstrated, in many important respects we are still at the frontiers of knowledge". In such circumstances, in the absence of any other cogent evidence beyond the inferences drawn from coincidence, the jury would lack any reasoned basis for preferring the opinions of one reputable set of experts to the other in an area where the true knowledge, even of the experts themselves, is necessarily limited. The judgment as a whole, and paragraph 178 itself demonstrates that the Court had in mind "cases like the present", that is, Cannings itself, which depended on the inferences based on coincidence, or the unlikelihood of two or more infant deaths in the same family, or one death where another child or other children in the family had suffered unexplained ALTEs.
83. Although there were disputes between reputable experts about the significance of some of the findings made at post mortem, paragraph 178 of Cannings has no direct application to the present appeal. The allegation against the appellant arose from a single death. No-one could have suggested, and no-one did, that any inference should be drawn against the appellant from any previous incident involving any of her other children, or that there was a pattern of ALTEs from which adverse inferences could be drawn. Moreover it is plain as can be that if the only evidence consisted of proof that Bidziil had died while his mother was caring for him, a prosecution would not have been contemplated. Natural infant death is a well-recognised, if not wholly understood phenomenon. Like Cannings this case involved an allegation of filicide, but there were here significant findings that took the evidence well beyond mere proof that Bidziil had died while in his mother's care.
84. In reality, the problem with the argument based on reading paragraph 178 of Cannings outside its context, is that carried to its logical conclusion, the submission would mean that whenever there is a conflict between expert witnesses, the case for the prosecution must fail unless the conviction is justified by evidence independent of the expert witnesses. Put another way, the logical conclusion of what we shall describe as the overblown Cannings argument is that where there is a conflict of opinion between reputable experts, the expert evidence called by the Crown is automatically neutralised. That is a startling proposition, and it is not sustained by Cannings.
85. In Cannings there was essentially no evidence beyond the inferences based on coincidence which the experts for the Crown were prepared to draw. Other reputable experts in the same specialist field took a different view about the inferences, if any, which could or should be drawn. Hence the need for additional cogent evidence. With additional evidence, the jury would have been in a position to evaluate the respective arguments and counter-arguments: without it, in cases like Cannings, they would not.
86. Mr Cooper suggested that there was virtually no additional cogent evidence in this case. He would have the jury disregard evidence about Bidziil's condition found on post mortem examination, and in particular, the congestion of the lungs, the widespread haemorrhage in the terminal air spaces, the presence of oedema and haemosiderin laden macrophages, and the blood stained fluid on the face and in the airways. He agreed that the nosebleed constituted extraneous evidence, but in effect, as its significance depended on expert evidence, the importance to be attached to it was significantly diminished. In our view, however, evidence of this kind is evidence of fact notwithstanding that it is discovered and given by experts, and it was precisely this kind of material which was sought and could not be found in Cannings.
87. Mr Cooper accepts that certain medical facts, such as a finding, for example, of bruising around the face (not of course relevant in this case), may be extraneous to the expert evidence on the basis that it does not require medical expertise to reach conclusions about the cause of bruising. He argues, however, that other medical facts are meaningless without expert interpretation and so intrinsically bound up in opinion as not to be extraneous to it. He gives the example of the inferences to be derived from the presence of haemosiderin. We understand this argument, and acknowledge that there will be occasions when the line between fact and opinion may become blurred. The extent to which any particular fact will be freestanding, absent the necessary explanatory expert opinion, may well differ from case to case. However the presence of haemosiderin is a fact. Standing on its own it may be neutral. If present in great quantities, that is another fact, which may not be neutral. In this case however, it is an additional fact that Bidziil suffered an unexplained, spontaneous nosebleed in the afternoon when he died: it is further fact that both old and fresh blood were found in his lungs. It is yet another fact that no natural illness explained the presence of old and fresh blood in the lungs or the nosebleed in the afternoon, which itself is subject to the further fact that the post mortem conducted by Dr Tapp was open to serious criticism. All these facts, and many others, were put before the jury. The dispute between experts about the interpretation of the findings at post mortem did not extinguish the findings themselves.
88. It was therefore for the jury to evaluate the expert evidence, taking account of the facts found at post mortem, bearing in mind in addition, for example, that the findings related to a infant whose mother had spoken about killing him and had made a comment about smothering another child; who had had difficulties in achieving a bond with him in the way in which she thought appropriate; who may have delayed reporting his death; and who had elected not to give evidence.
89. In the context of disputed expert evidence, on analysis, what was required in this case was no different to that which obtains, for example, when pathologists disagree about the cause of death in a case of alleged strangulation. An argument whether the hyoid bone was fractured before death (supporting the conclusion of strangulation) or whether it occurred post mortem, perhaps during the course of the autopsy itself (which would discount strangulation), is commonplace. More important, it does not alter the fact that the hyoid bone was fractured. And even if the experts disagree about whether it was indeed fractured, that is a question for the jury. Cannings does not produce the result that it follows from an argument between experts that the issue whether the fracture occurred before or after death, or whether there is a fracture at all, is not appropriate for the jury's consideration. Evidence of this kind must be dealt with in accordance with the usual principle that it is for the jury to decide between the experts, by reference to all the available evidence, and that it is open to the jury to accept or reject the evidence of the experts on either side.
90. Although the observations in paragraph 178 of Cannings are directed primarily to prosecutors, they plainly impact on the task of the trial judge. Mr Cooper has not argued that the judge should have upheld his submission that there was no case to answer. As we have already indicated, the judge's decision was plainly right. Whenever the judge has to consider the argument based on Cannings, he will obviously bear in mind that there can be no sufficient case if it is based on the fallacious premise exploded by the Court in Cannings. This need for care, and Cannings generally, does not reveal any new principle of law or approach: it is no more than an example of the judge's general obligation to ensure that the case, however it is put, proceeds on a logically justifiable basis, and is not left to the jury unless the evidence, taken at its highest, is such that the jury properly directed could convict: see the well known test in Galbraith 73 Cr App R 124.
91. Analysed in this way, that there is no room for a mechanistic application of the three stage test derived by Mr Cooper from paragraph 178 of Cannings.
92. Having examined the material that was before the jury at the end of the trial, and in the light of the admirable summing up, in our judgment, there was ample evidence before the jury to justify the verdict. The conviction was safe.
93. We must now examine the second ground of appeal.

Fresh Evidence

94. Under s 23 of the Criminal Appeal Act 1968 ("the 1968 Act"), Mr Cooper sought to call five expert medical witnesses, none of whom was involved at the trial nor, indeed, instructed until after its conclusion. As explained in his skeleton argument, he was anxious to invite the Court to consider it:

"… in the context of their assessment generally of the expert evidence called at the trial, and in particular the lengthy and adverse, although perhaps not unjustified, observations of the learned trial judge on Dr Rushton and the way he gave his evidence… Dr Rushton made a number of perhaps unwise observations on peripheral matters in the course of his evidence. Given that, it is submitted that the perception of the jury of his evidence was undoubtedly coloured. Whilst this is a factor which perhaps unfortunately is an unavoidable side-effect of the adversarial jury system, the perception of the defence case it is submitted was open to substantial enhancement had the further evidence been available at the time."

95. It is not suggested that Mr Cooper was limited in any way in his approach to expert evidence at the trial, and it is obvious that if instructed, each of these witnesses would have been available. For various reasons, provided in a statement by the appellant's solicitor, they were not approached. Some of the evidence was in fact adduced, although not directly from them. Thus, the article by Becroft et al, Archives of Diseases in Childhood 2001.85.116-20, was summarised in Dr Rushton's report and is the basis of the proffered evidence from Professor Edwin Mitchell, a co-author. Similarly, although differing in emphasis and presentation, much of the evidence given by Professor Rutty at trial is repeated in the so-called 'fresh' evidence provided by Professor Milroy.
96. The Crown do not assert that once evidence had been given by a witness who was not as persuasive as the defence would have liked, or hoped, it would never be open to this Court to receive further evidence under s 23 of the Act. S 23(1)(c) permits the court to "receive any evidence which was not adduced in the proceedings from which the appeal lies". However the Crown does rely on two of the specific matters to which the court must have regard when considering whether to receive "evidence" under s 23, first, that such evidence would not afford grounds for allowing the appeal, and second, that there is no reasonable explanation for the failure to adduce it at trial. Very fairly, Mr William Davis QC, for the Crown, does not suggest that evidence which might afford a ground for allowing the appeal should not be heard and admitted. He would not seek to resist an appeal on a wholly technical basis. We agree that points which may legitimately be taken on behalf of the appellant which serve to undermine the safety of the conviction should, in the interests of justice, be considered on its merits. In the particular circumstances of this case, we heard all the evidence tendered before us de bene esse.
97. We should not re-write, and we are not re-writing s 23. The fact that the expert chosen to give evidence by the defence did not give his evidence as well as it was hoped that he would, or that parts of his evidence were exposed as untenable (as, certainly on one view, occurred with Dr Rushton) thereby undermining confidence in his evidence as a whole, does not begin to justify the calling of further evidence, whether to provide "substantial enhancement" of the unsatisfactory earlier evidence, or otherwise. Where expert evidence has been given and apparently rejected by the jury, it could only be in the rarest of circumstances that the court would permit a repetition, or near repetition of evidence of the same effect by some other expert to provide the basis for a successful appeal. If it were otherwise the trial process would represent no more, or not very much more than what we shall colloquially describe as a "dry run" for one or more of the experts on the basis that, if the evidence failed to attract the jury at trial, an application could be made for the issue to be revisited in this court. That is not the purpose of the court's jurisdiction to receive evidence on appeal.
98. We can deal comparatively briefly with three of the experts on whom Mr Cooper now seeks to rely. Professor Andrew Nicholson, a consultant histopathologist, considered the presence of a single granuloma or cluster of cells found on one of the slides taken from one of the sections removed from Bidziil's lungs: it was of no clinical significance. He was also concerned to consider whether haemosiderin found on the post mortem might have been the result of idiopathic pulmonary haemosiderosis, an auto-immune disease of unknown cause. The features evident from the slides which he examined were not characteristic of this disease. He therefore believed that it was unlikely to have been present. Mr Cooper accepted that little could be derived from this evidence other than the fact that haemosiderin could be present for natural reasons: as Mr Davis observed, given that Professor Nicholson excluded that the likelihood of another possible natural cause, the evidence did not advance the appellant's case at all. Whether or not this issue should have been investigated at trial, this evidence would afford no ground for allowing the appeal. Accordingly we shall not admit it.
99. Dr Mark Rosenthal, consultant in paediatric respiratory medicine was asked to consider whether the fact that the appellant was diabetic and had suffered an episode of diabetic ketoacidosis during pregnancy, might have been a factor in causing Bidziil's death and, secondly, to comment on a chest x-ray taken of Bidziil following the femoral fracture in June. Both these matters could and probably should, have been addressed before the trial. In the event he concluded there was no evidence to link maternal diabetes to Bidziil's death and no feature on the x-ray which gave cause for concern for his future respiratory health. On the issue he was asked to consider, he provided no relevant evidence.
100. In the course of a joint experts' telephone conversation, however, Dr Rosenthal expressed concern about the appellant's report to NHS Direct of green vomit on 27th July 2002 (see paragraphs 19 - 22 above). He considered that green vomit, that is likely to be containing bile, was a very unusual symptom and always pathological: thus it required immediate referral to hospital. In evidence he said that the symptom was wholly non-specific but nevertheless it required attention. Although no abnormality was revealed at the post mortem, he considered that not all the tests had been done and that, in any event, medical knowledge may not be sufficiently advanced to determine cause of the "green" vomit. When asked his reaction if he was told that such an incident had happened a few days earlier, he accepted that without further recurrence, it need not be indicative of anything. He would then advise that if it happened again the baby should then be referred to hospital.
101. It was well-known to every doctor who gave evidence at trial that Bidziil's earlier vomit had been green in colour. None ascribed any significance to it or suggested that it would repay further investigation. Although not focussing on the green vomit, Dr Rushton specifically reported the vomit of green mucus as part of the history and commented on vomiting both in his report and at trial. In evidence, he was reminded of Bidziil's vomiting episodes on 26th and 27th July and asked about its significance in relation to his findings within the lungs; his response was to express concern about the brown vomit.
102. We mean no disrespect to Dr Rosenthal, who rightly confined his evidence to his area of expertise, but on fair analysis all that he was able to say is the green vomit was a non-specific symptom of ill health in a baby who, nevertheless, appeared to be positively healthy a few days later. He did not suggest any link between this incident and what happened on 1st August, and in particular, the pulmonary bleeding. Mr Cooper submitted his evidence might have affected the jury's view whether it was safe to conclude that the baby was healthy when he died. In reality that would be wholly speculative. There is no reason why this evidence was not available for trial, but in any event it would afford no ground for allowing the appeal. Accordingly, we shall not admit it.
103. Dr Brendan MacDonald, a consultant neuropathologist, re-examined the tissue sections of Bidziil's eyes. His further evidence fell within a very narrow compass, and related solely to gliosis. In effect he was unable to confirm Dr Bonshek's diagnosis of gliosis and said that he would not be happy to make it. He said that interpretation was difficult because of artefactural changes following death. He agreed that one explanation for its presence, if established, would have been oxygen deprivation, but said that there could be other explanations although he could find none. His statement was similarly cautious: he said that he "was not convinced" that retinal gliosis could reliably be diagnosed.
104. In her statement, the appellant's solicitor explained that Dr Rushton too had taken the view that the examination of the eyes was hampered by artefactural changes caused by the delay in removing them. This rendered the findings uncertain: given his expertise as a pathologist, value was placed upon his observation. This led to the cross-examination of Dr Bonshek and his acceptance of the need for care in interpreting what was seen. It was not specifically put to him that the effect of the changes would be to render uncertain all possible findings, and it does not appear that he gave evidence to that effect.
105. As a neuropathologist Dr MacDonald is not in any better position than Dr Rushton as a perinatal and paediatric pathologist to comment on the evidence of ophthalmic pathology given by the particular specialist, Dr Bonshek. In effect therefore, the evidence which Dr MacDonald gave was available from Dr Rushton. There is no satisfactory explanation for the failure to call this evidence. In any event, the issue to which it related was of very limited significance at trial. Bearing in mind that Dr MacDonald's evidence does not challenge Dr Bonshek directly, but is limited to saying that he could not confirm Dr Bonshek's diagnosis of gliosis, and that Dr Bonshek did not suggest that the gliosis he identified was diagnostic of oxygen deprivation, it is quite impossible to conclude that this evidence would afford a ground for allowing the appeal.
106. We must next address the two witnesses whose evidence took rather longer than the previous three. Professor Christopher Milroy, professor of forensic pathology is a specialist whose expertise is similar to, if not identical with, both Dr Rushton and Professor Rutty. The appellant's solicitor justifies the reception of his evidence on two grounds. First, it is said that although the standard of the first post mortem performed by Dr Edmond Tapp was covered in detail at the trial, the rules by which such criticism was levelled were those emanating from the Confidential Enquiry for Stillbirth Death in Infancy (CESDI) Protocol, the local protocol in Birmingham, where Dr Rushton practises, and Professor Rutty's working standards. After trial, in September 2004, following a report of a working group convened by the Royal College of Pathologists and the Royal College of Paediatrics and Child Heath (chair: Baroness Helena Kennedy QC) a multi-agency protocol relating to Sudden unexpected death in infancy was published. Professor Milroy confirmed that although the protocol recommended a number of ancillary tests, essentially, the CESDI protocol was reconfirmed. He pointed out that these ancillary tests were not detailed in Dr Tapp's report and are "partially detailed" in the report of Dr Cox at the time of the second post mortem. More significantly, he does not suggest that if the new protocol had been followed, further flaws over and above those concealed and considered in detail at trial would have emerged. There is thus, in any event, no new or fresh material contained within this protocol which might serve to undermine the safety of the conviction.
107. The second basis upon which it is suggested that the evidence of Professor Milroy should be received is that he refers to more recent articles on the subject of haemorrhage and haemosiderin in the lungs. These were not referred to by Dr Rushton or Professor Rutty prior to trial and so, it is suggested, "their existence was unknown". In fact, Professor Milroy cites two papers by Hanzlick and others, published in the American Journal of Forensic Medicine and Pathology in 2000 and 2001, and the textbook written by Byard referred to in paragraph 5. This refers to a paper by Yukawa et al in 1999 quoted by Professor Risdon in his report, together with papers by Becroft and Lockett in 1997 and Professor Milroy himself in 1999.
108. Although Byard's work, in its second edition, was published after the trial, all the articles referred to by Professor Milroy were already in existence, and therefore either fully within the knowledge of these experts in the specific areas covered by them, or easily ascertainable on a professional database. If they add to the weight of any opinion which Dr Rushton or Professor Rutty wished to advance, there is not only no reasonable explanation for the failure to do so, there is no explanation at all. There is thus no justification for admitting these articles, in themselves, as fresh evidence.
109. In his skeleton argument, Mr Cooper advanced the first of these reasons, but not the second to justify the reception of Professor Milroy's evidence. He did however add a third in these terms:

"Professor Milroy also raises an issue which was not considered at trial, namely, the effect that the time the samples were taken for histology may be of significance when considering the observations in relation to haemosiderin in the lungs and haemorrhage.

Had the … question of the timing of the histology samples been explored further, it is submitted that on this issue also the jury might reasonably have taken a different view of the evidence."

By the time Professor Milroy came to give evidence, however, he had understood that the tissues had been taken at the first autopsy, preserved and then tested histologically. There is thus nothing in this point and it was not pursued further.

110. This was an inauspicious start, for which, again, Professor Milroy cannot be personally criticised. He repeated the criticisms made of Dr Tapp's work, and like Professor Rutty observed that although samples had been taken to exclude metabolic disorders, it was not possible to exclude them all. In his statement he had referred to "evolving" knowledge on haemorrhage and haemosiderin in the lungs, and that although upper airway obstruction was one cause of the presence of haemosiderin laden macrophages and pulmonary haemorrhage, these findings could not be considered diagnostic and had been identified in cases categorised as SIDS.
111. However he accepted that the Hanzlick paper in 2000, Pulmonary Haemosiderin in Deceased Infants revealed data consistent with the proposition that the presence of such haemosiderin suggested a cause of death other than SIDS, and that the 2001 paper Pulmonary Haemorrhage in Deceased Infants established that this was a common finding. Professor Milroy pointed out that the more robust the attempt at resuscitation, the more pronounced any haemorrhage was likely to be. This was not a fresh point. Professor Rutty had made it. Professor Milroy also said that the pulmonary haemorrhage may have been a function of the delay in the post mortem. None of these possibilities however addressed the fact of Bidziil's spontaneous nosebleed which resulted from blood coming from his lungs. Much of this material was essentially repetitious of evidence given by others at trial, and available for consideration by the jury. It does not give rise to a legitimate argument that the conviction is unsafe. We shall not admit it.
112. At the very end of his evidence, Professor Milroy did add one further unanticipated detail. Having, since 1992, been carrying out autopsies in suspected SIDS cases where the police have concerns, he had recently seen one case of pulmonary haemorrhage as extensive as evident in Bidziil, although no haemosiderin was found. The case concerned a child who was reported to have been winded and who after being placed on his back in a Moses basket was found lying dead in the same position the following morning. Professor Milroy concluded that this was a cot death. This was fresh evidence, but given the absence of haemosiderin, and the absence of any further details it is difficult to see how a true comparison can be made with the facts of this case, or how the individual case referred to by Professor Milroy served to undermine the broad thrust of the expert evidence by the Crown. In fairness, Mr Cooper did not suggest that, taken on its own, this individual example would serve to undermine the safety of the conviction. In reality, Professor Milroy's evidence takes the appellant no further. Apart from the late reference to the single individual case, as his evidence covered ground already covered by Dr Rushton and Professor Rutty, this is not surprising. This evidence takes the case no further.
113. We come finally to the evidence of Professor Edwin Mitchell, professor of Child Health Research at the University of Auckland. He gave evidence over a video link from New Zealand. He is, essentially, a paediatric epidemiologist who with Dr David Becroft, a paediatric pathologist, has published papers examining the epidemiology of pathological findings found in SIDS. In particular, they published a paper, Nasal and intrapulmonary haemorrhage in sudden infant death syndrome in Archives of Disease in Childhood, (2001.85.116-120). That paper was summarised in Dr Rushton's report, dated 21st April 2003, in some detail. He then reached the conclusion, which he repeated during the course of his evidence:

"It is therefore clearly unsafe to conclude that the presence of nasal and pulmonary haemorrhage are necessarily indicative of smothering since both patterns of haemorrhage are common in SIDS and the presence of nasal haemorrhage in such cases is associated with more extensive intra pulmonary haemorrhage. Certainly, it does not indicate beyond reasonable doubt that death was due to asphyxia be it accidental or non-accidental."

114. The evidence provided by Professor Mitchell consists of the detail that formed the basis of the quoted report together with an elaboration of that study. On the face of it, this, again, falls foul of the requirement that it is the evidence that is fresh; this is not a case, for example, of an additional number of alibi witnesses coming forward to support an alibi because what they saw constitutes fresh evidence. Rather, it is once again a different expert providing much the same evidence to that given at trial, although with considerable elaboration and in much greater depth.
115. The appellant's solicitor seeks to deal with this problem on the basis that what she called the "centrality" of the nosebleed did not become apparent until the trial judge commented at the end of the summing up that it was of first importance to the prosecution case. What she did accept, however, was that the nosebleed had a direct correlation to pulmonary haemorrhage which itself was a critical feature of the evidence. Professor Risdon and Professor Rutty both emphasised the fact of the nosebleed, together with two episodes of bleeding into the lungs were pointers to asphyxia, and, indeed, even Dr Rushton concluded that the findings of upper and lower respiratory tract haemorrhage, the presence of iron laden macrophages in the lungs and the possible petechial haemorrhage in the mucosa of the palate were consistent with but not diagnostic of asphyxiation, whether accidental or otherwise. The presence of a nosebleed was said to support that conclusion; it was always a most important part of the case for the Crown and there is not the slightest justification for concluding that its importance was not fully appreciated by the defence, or that they failed to give it the attention it merited.
116. The final ground put forward in the solicitor's statement for not seeking Professor Mitchell's assistance at an earlier stage is that his work was expressed to be directed to cases of SIDS and that all experts at trial agreed that this was not a SIDS death; this explanation is echoed in Mr Cooper's skeleton argument. Quite apart from the reference by Dr Rushton to the work of Professor Mitchell, this submission misses the point made earlier in this judgment that SIDS is a name given by different experts to describe different circumstances in which death occurs. Thus Professor Risdon's categorisation of an infant death as SIDS requires that nothing of significance should be found at post mortem; Professor Mitchell extends his categorisation of SIDS to many cases which would not be so described by Professor Risdon. In some cases described by him as SIDS, there were signs of an asphyxial death, many of them with babies who were not put to sleep on their back, and others where the baby was sleeping with one or two adults, either in a bed or on a sofa. The issue at trial was not how Bidziil's death should have been characterised, but rather whether there was material sufficient to justify the conclusion that it was caused by obstruction of the airways, and, if so, whether that obstruction was accidental or deliberate.
117. In reality, Professor Mitchell's evidence is, again through no fault of his own, very limited. He is not a pathologist and has not himself seen any of the histology arising from the investigation of Bidziil's death: he was frank enough to make it clear that he would not be able to interpret it, even it was provided. Neither was he expressing any opinion as to the cause of Bidziil's death. Rather, he was providing statistical data arising out of a study of 485 deaths in New Zealand each of which came within the Beckwith definition of SIDS ("sudden death of an infant or young child unexpected by history and where a thorough autopsy fails to demonstrate an adequate cause of death"). Thus, he can say what proportion of his population demonstrated this or that symptom: he cannot say whether they do so for the same or different reasons.
118. We start with nasal haemorrhage. He reported that, of 385 cases (28 days to first birthday) in which this information was recorded, 60 (15.6%) reported nasal haemorrhage. He said that factors found to be associated with nasal haemorrhage were infants less than 13 weeks old (about 40-45%), being placed to sleep in a non-prone position, that is, on the side or back, and sharing a bed with an adult. This was to be contrasted with the presence of frank blood in the mouth nose or on the face in 39% of 70 children whose parents were convicted of causing their death (Meadows, 1997) and 29% of 38 children with ALTEs in which suspicion of abuse was confirmed by covert video (Southall et al, 1997). He concluded that nasal haemorrhages in infants are more frequent in infants dying from non accidental injury than SIDS but that SIDS was more common than deaths from abuse so that the majority of dead infants with nasal haemorrhages will be SIDS cases. It is this that gave rise to Dr Rushton's conclusion.
119. The problem with this conclusion is that like is clearly not being compared with like. Thus, Professor Mitchell relied for his conclusion of nasal haemorrhage on a single answer in a structured interview of parents who had suffered this tragedy. They answered the question whether there was blood around the nose or mouth. There was no explanation or room to expand the answer which in any event was dependent on what the parent understood by the question or, indeed, meant by his or her answer. The appellant's solicitor herself was careful to make it clear that she would include pink frothy fluid in the definition of nosebleed and in Professor Mitchell's cases, it is not even clear that blood had to emanate from the nose, rather than the mouth. Professor Mitchell could not say whether it was reported that any of these infants under 13 weeks of age suffered spontaneous nosebleed without its cause being ascertained: that information was not sought.
120. That is not the only difficulty with this statistic. Professor Mitchell was not saying that none of the sixty cases in which some form of nasal haemorrhage was identified had not suffered obstruction to the airways. Twenty-eight of the sixty were bed-sharing: accidental airways obstruction while asleep against or overlain by an adult is a well-recognised cause of infant death. Thus not only is Professor Mitchell's study not able to discriminate between cases of the type of nose bleeds described in this case, and a subjective assessment of blood around the nose or mouth, but he does not say that any of those cases in which that symptom was described are not the consequence of airway obstruction: to that extent, this aspect of his evidence was less helpful to the appellant than was the evidence of Dr Rushton himself. As we have explained, Professor Risdon would simply not regard such a death as a SIDS death.
121. Moving on to pulmonary haemorrhage, Professor Mitchell said that it was common in SIDS and may be intra-alveolar and/or into the respiratory passages. In his study, Dr Becroft examined histologically lung sections from 115 SIDS cases from two regions. He assessed the severity of haemorrhage into the alveoli and the airways separately on a semi-quantitative scale, 0-3. Pathologically significant intra-alveolar haemorrhage was found in 47% of the 115 cases and 8 had moderate or severe intra alveolar haemorrhage (i.e. more than 5% of the lung affected); 6 of these 8 also reported nasal haemorrhage (as defined above) in which cases the intra-alveolar haemorrhage was more severe and there was more airways haemorrhage. Professor Mitchell concluded that this indicated that the degree of intra alveolar and airway haemorrhage seen in Bidziil's lungs was not uncommonly seen in SIDS cases.
122. If that conclusion could be justified by reference to the evidence, this might indeed be relevant but it cannot. The extent of the bleeding seen in the histology relating to Bidziil was far in excess of the 5% limit used in the study. Professor Mitchell simply could not tell us how the slides seen by Dr Becroft compared with Bidziil's slides: the exercise has not been carried out. Thus, he cannot say whether the degree of intra alveolar and airway haemorrhage visible from the histology taken from Bidziil's lungs is seen commonly or not. In any event, at the risk of repetition, SIDS within Professor Mitchell's definition includes cases that involved airways obstruction.
123. Moving on to the intra-alveolar iron laden macrophages, again, Professor Mitchell agreed with Professor Risdon that their presence indicated a previous episode of intrapulmonary haemorrhage. However he considered that there could be a natural cause on the basis that airway obstruction may be part of the final pathway of SIDS; he did so on the basis that 5 of the 115 cases in his study had such a finding. All that can be said of the comparison, however, is that in the five cases, diffuse indicative staining was found. He was not able to say how widespread or extensive the macrophages were: in Bidziil's case, this was abundant and it was simultaneously associated with extensive interstitial haemosiderin.
124. Having, at our request, checked through his papers after he had given oral evidence, Professor Mitchell reported that of the five cases of intra-alveolar macrophages that might cause suspicion of previous airways obstruction, only one had associated nasal haemorrhage in the sense in which the term was used during his research. This infant also had intra-alveolar pulmonary haemorrhage which was in the moderate-severe category and interstitial haemosiderin. Indeed Professor Mitchell observed:

"Thus this infant had exactly the same pathological findings as the infant in this case. There were no suspicious circumstances surrounding the death of our subject. The clinical features were that the baby was 5 weeks of age, co-sleeping and non-prone."

125. Mr Cooper argued that this example demonstrates a weakness in the prosecution theory that the presence of all these features in Bidziil's case would properly lead the jury to conclude that he had died of airway obstruction. We do not agree. It is not clear that the infant referred to by Professor Mitchell had "exactly the same pathological findings" as Bidziil. For example, we do not know what the parents of this child meant when they answered that there was blood around the nose or mouth of the child. Again, we equally do not know whether the pulmonary haemorrhage, or the interstitial haemosiderin were as extensive as they were in Bidziil's case. There are two further points. First, it is not clear that this individual case provided by Professor Mitchell did not involve a case of airway obstruction: the baby was co-sleeping, and in such circumstances, accidental airway obstruction is not excluded. Second, it was repeatedly pointed out at trial that the prosecution were not claiming that these findings were in themselves diagnostic of deliberate airways obstruction. The way in which the prosecution put its case was that the combined effect of the facts revealed at post mortem, combined with the remaining evidence, including the sequence of events of the afternoon of 1st August should properly drive the jury to the safe conclusion that Bidziil died as a result of deliberate airways obstruction.
126. At the end of his evidence, Professor Mitchell was asked whether possible significance might attach to the fact that Bidziil had been placed in a buggy on 1st August. He speculated (his word) on the basis of recent research into babies placed in car seats which showed that a poorly supported child could flop forward and suffer some obstructive of his breathing. This would usually arouse the child, but if the child lacked a perfect arousal mechanism, then he might suffer a nosebleed. However Professor Mitchell made it clear that none of the experiments had produced bleeding from the nose of a child in the car seat. For obvious ethical reasons, it was impossible to conduct research involving (a) deliberate obstruction of the child's breathing mechanism and (b) deliberately preventing or delaying the child's recovery by continuing to hold the head in a position to maintain obstruction of the airways. Although Mr Cooper prayed this material in aid during the course of his final submissions, we cannot attach any weight to it, not least because Professor Mitchell himself made it absolutely clear that he was doing no more than speculating, because he was asked to do so, about a possible mechanism which would have damaged Bidziil when he was in his buggy.
127. This epidemiological evidence does not advance the appellant's case. First, the research was described and set out in Dr Rushton's report. Second, there is no reasonable explanation for failure to adduce it at trial: a simple enquiry would have revealed that airways obstruction was not excluded by the definition of SIDS applied during this research. Third, and in any event, on analysis, in our judgment it does not provide any basis for allowing the appeal, or concluding that the verdict of the jury was or might be unsafe. Without any discourtesy to Professor Mitchell for whose candid and clear evidence we were grateful, although we heard his evidence, we shall not in the result admit it.

Conclusion on Fresh Evidence

128. We can express our conclusion very simply. Nothing in the so-called "fresh" evidence advances the appellant's case. It does not, we observe in addition, provide a basis for suggesting that the judge should not, or would not after all, have given the direction he did give about the way in which the jury would be entitled to approach the absence of the appellant from the witness box. We cannot see that if this evidence had been available to her it could or would have made the slightest difference to her decision. The single question for her was whether she wished to give evidence about events in Bidziil's daily life, and on 1st August itself: she could not have commented or made observations about issues covered by expert evidence.
129. In the result, although a great deal of effort was made after conviction to produce additional evidence with which to undermine the verdict of the jury, we have found nothing in any of it which leads us to doubt the safety of the conviction, or putting the same point in another way, which suggests that if this evidence had been deployed before the jury, it would on proper analysis, have created sufficient doubt to lead to an acquittal.
130. For these reasons, this appeal must be dismissed.

Infanticide

131. The conviction of this mother for murder stands.
132. There was no evidence to sustain the defence of diminished responsibility, and none to demonstrate infanticide. These issues did not arise for consideration at trial or on appeal, and just because they did not, we shall add some observations about infanticide.
133. We must briefly recap. The appellant was a woman of good character with two children. She had apparently given them natural maternal love and affection before she gave birth to Bidziil. He was conceived in the course of an alleged rape. She resisted suggestions that her pregnancy should be terminated. Immediately after the birth she underwent some unspecified level of depression, but was anxious about the possible implications of taking anti-depressants on the baby's feeding. Within a very short period, for understandable reasons arising from an injury to Bidziil for which the appellant herself was not responsible, after a period in hospital, he was cared for by her mother, and not by her. All this inevitably weakened the natural bonding process. When the baby returned home, the appellant was unable to breast feed the baby successfully, something to which she attached great importance. She seems to have believed that as a result of the difficulties, they were not bonding properly. It was in those circumstances that the baby was killed, less than three months after the appellant had given birth to him. Following conviction, the inevitable mandatory sentence of life imprisonment was imposed on her.
134. Section 1(1) of the Infanticide Act 1938 (which replaced the Infanticide Act 1922) provides:

"Where a woman by any wilful act or omission causes the death of her child being a child under the age of 12 months, but at the time of the act or omission the balance of her mind was disturbed by reason of her not having fully recovered from the effect of giving birth to the child or by reason of the effect of lactation consequent upon the birth of the child, then, notwithstanding … that but for this Act the offence would have amounted to murder, she shall be guilty … of infanticide."

Under s 1(2) provision is made for infanticide to be an alternative verdict available to the jury trying a mother for murder of her infant child. It does however require evidence that the "balance of her mind was disturbed" either because the mother has not recovered from giving birth to the child, or the effect of lactation on her. No other circumstances are relevant.

135. Just under twenty years later, the Homicide Act 1957 identified diminished responsibility as a general defence to murder. Infanticide remains both as a defence to murder, available to the mother who causes the death of her infant before it reaches twelve months, but simultaneously, as an offence in its own right where the mother has killed the baby, and it is nevertheless accepted that the balance of her mind was disturbed.
136. For many years now there has been some considerable discussion about infanticide, its definition, and indeed whether it should continue to be an offence, or alternatively, a defence to murder. Thus the Report of the Committee on Mentally Abnormal Offenders (Cmnd 6244 Chairman: Right Hon. Lord Butler of Saffron Walden (1975)) believed that assuming, contrary to its own "decided preference" the mandatory life sentence for murder were retained, diminished responsibility should be re-worded so as to apply where the defendant was "suffering from a form of mental disorder … such as to be an extenuating circumstance …". Infanticide would then cease to be of any practical value.
137. The Report pointed out at paragraphs 19.23 that "the medical principles on which the Infanticide Act is based may no longer be relevant. The theory behind the Act was that childbirth produced an hormonal disorder which caused mental illness. But …. the operative factors in child-killing are often the stress of having to care for the infant, who may be unwanted or difficult, and personality problems …". In the next paragraph the Committee quoted at length from "perhaps the most impressive evidence" received on the subject from the Governor and staff of Holloway Prison. "The disturbance of the "balance of mind" that the Act requires can rarely be said to arise directly from incomplete recovery from the effects of childbirth, and even less so from the effects of lactation … A combination of environmental stress and personality disorder … are the usual aetiological factors … and the relationship to "incomplete recovery from the effects of childbirth or lactation" specified in the Infanticide Act is often somewhat remote".
138. In its Fourteenth Report on Offences Against the Person, Cmnd. 7844 (1980), the Criminal Law Revision Committee suggested that infanticide should be retained, but recommended that it should be extended to cases where the balance of the mother's mind was disturbed by "environmental or other stresses". Clause 64(1) of the Draft Criminal Code Bill provided that a mother is guilty of infanticide:

"… if her act is done when the child is under the age of 12 months and the balance of her mind is disturbed by reason of the effect of giving birth or of circumstances consequent upon the birth."

The Report of the Select Committee of the House of Lords on Murder and Life Imprisonment (HL Paper 78) (1989) did not recommend that there should be a change in the law, but suggested that it should be "further considered".

139. The issues raised in these cases are delicate and sensitive. In October 2004 the Home Office announced a comprehensive review of the law of murder. The recently published Law Commission, Ninth Programme of Law Reform, anticipates that it will be involved in and contribute to this review. The public interest requires that the problems arising from and connected to the offence of infanticide should be included in any review. We shall highlight two particular areas of concern. The first is whether, as a matter of substantive law, infanticide should extend to circumstances subsequent to the birth, but connected with it, such as the stresses imposed on a mother by the absence of natural bonding with her baby: in short, whether the current definition of infanticide reflects modern thinking. The second problem arises when the mother who has in fact killed her infant is unable to admit it. This may be because she is too unwell to do so, or too emotionally disturbed by what she has in fact done, or too deeply troubled by the consequences of an admission of guilt on her ability to care for any surviving children. When this happens, it is sometimes difficult to produce psychiatric evidence relating to the balance of the mother's mind. Yet, of itself, it does not automatically follow from denial that the balance of her mind was not disturbed: in some cases it may indeed help to confirm that it was.
140. The law relating to infanticide is unsatisfactory and outdated. The appeal in this sad case demonstrates the need for a thorough re-examination.