Appellant: Shead, Nicolson; John Pryde & Co., Edinburgh

Respondent: W.J. Wolffe, Q.C., A.D.; Crown Agent

26 May 2011

Introduction

[1] On 7 April 2006 after a trial at the High Court in Edinburgh which ran for more than five weeks the appellant was convicted of the following charges (as amended):

"(1) on various occasions between 14 October 1982 and 20 December 1982 both dates inclusive, at [various locations in Edinburgh] you did assault J.S., born 14 May 1982, now deceased...and did cause her upper airways to be obstructed by means unknown, restrict her breathing, all to the danger of her life;

(2) on 20 December 1982 at [a location in Edinburgh] you did assault J.S., born 14 May 1982, residing there, and did obstruct her upper airways by means unknown, restrict her breathing, suffocate her and kill her"

[2] Charge (3) on the indictment was withdrawn by the Advocate depute. The appellant was sentenced to an extended sentence of fifteen years imprisonment comprising a custodial term of ten years and an extended period of five years. On appeal that sentence was reduced to one of seven years imprisonment. An appeal against conviction was not pursued.

[3] On 5 February 2010 the Scottish Criminal Cases Review Commission referred the appellant's case to the High Court. She lodged grounds of appeal which included contentions (1) that the trial judge had misdirected the jury by failing to set out for them the expert evidence which they would require to accept in order to convict and (2) that in the circumstances of this case there was an infringement of the appellant's right under Article 6(1) of the European Convention to a reasoned judgment. These contentions are inter-related.

[4] The appellant and her husband married in 1981. There were three children of the marriage of whom J.S. was the middle child. At the time of her death on 20 December 1982 she was somewhat over seven months of age. Prior to her final admission on 20 December she had been admitted to the Royal Hospital for Sick Children in Edinburgh on five separate occasions, the first of these being on 14 October. On each occasion the appellant, who was alone in charge of the child, had reported that she had suffered a "turn" where she went blue and stopped breathing or had convulsions. A post mortem examination was carried out by Dr Bain, a leading paediatric pathologist who had died before the trial, and (the then Doctor, subsequently Professor) Anthony Busuttil, who gave evidence at the trial. The initial cause of death was recorded by Dr Bain as "Sudden Infant Death Syndrome". This was subsequently amended by him to "SIDS explained, bronchiolitis and bronchopneumonia". Tissue blocks were retained and were available for the criminal investigation which led to the appellant's indictment in 2005. A number of pathological findings were undisputed at the trial: intra-alveolar haemorrhage (fresh bleeding in the lungs), haemosiderin-laden macrophages (iron deposits in the lungs, indicating previous episodes of bleeding there), aspiration of food in the lungs, bronchopneumonia "starry-sky" appearance in the thymus and pneumococcus in the heart blood. The post mortem conclusion of bronchiolitis was disputed.

[5] There was no direct evidence that on any occasion the appellant had assaulted the child. The Crown case at trial turned substantially on medical inferences that the child had on each occasion, including the last which led to her death, been subjected to deliberate obstruction of her airways. Given the passage of time many of the witnesses who spoke to the child's admissions to hospital and her condition while there were reliant on her medical records. A log kept in 1982 by Mrs Susan Brown, a social worker attached to the hospital, was also available at the trial.

Hospital admissions

14 - 18 October 1982
[6] On 14 October 1982 PC Stephen Macrae attended the appellant's home, following an emergency call about breathing difficulties with the child. He found her unattended in her room, tried to clear her airways and moved her into the recovery position. Mrs Brown's log recorded that he had found her lying on her back, having swallowed vomit, had held her by her heels and had smacked her, whereby she revived and began to vomit. She was taken to hospital by ambulance. Dr Alison James, a registrar at the hospital in 1982, spoke to her admission. The history provided was that the child had stopped breathing for 20 minutes. Dr James gave evidence that this was not possible. The child presented as developmentally normal. An inner ear infection was diagnosed. Other tests were normal. Her temperature settled, she fed well and was discharged on 18 October.

18 - 29 October 1982

[7] The child was re-admitted as an emergency on the same day. The history provided was that she had turned blue around three hours after being fed and had suffered vomiting and diarrhoea. A neighbour had performed mouth to mouth resuscitation. She was examined by Dr Aysha Cockshot, then a senior house officer, who noted a low grade temperature, slight swelling to the left cheek, a slightly accelerated heart rate and rapid breathing. However, ears, nose and throat were normal; heart and breath sounds were normal. These findings were consistent with an infection, aspiration of stomach contents, or with an otherwise stressful event. A further examination was carried out by Dr James. The child was sleepy, her liver edge palpable, her tendon reflexes slightly fast and her plantar reflex abnormal. In a sitting position she was floppy. Dr James gave evidence that the most common reasons for transient floppiness in a child were a very high temperature or experience of a fit. Not breathing could also result in floppiness. Aspiration could have a neurological element; some of the symptoms exhibited were consistent with a neurological problem. Various other tests were done, including an EEG which showed no abnormality. Dr James considered that the child might have experienced a fit. A urinary tract infection was diagnosed. Antibiotics were prescribed. The child's nursing records narrated: that her apnoea alarm sounded once on 22 October but there was no sign of her having stopped breathing; that jerking was observed during feeding on 23 October; that significant vomiting was noted on 26 October; and that she was floppy on 28 October. She was discharged on 29 October. Although not recorded in the nursing records, Dr James had noted that "during the EEG ... [J] was noticed by the nurse to become extremely upset and to have a short breath holding attack, during which she went to poor colour but not cyanose, not blue. At this time she had a small nose bleed."

1 - 9 November 1982
[8] The child was re-admitted as an emergency on 1 November. The history provided was that she had been irritable and that the appellant had noticed her to be blue around the mouth and face. No abnormal signs were noted and no signs of infection presented upon investigation. She was discharged on 9 November.

13 - 16 November 1982

[9] The child was re-admitted on 13 November. The history provided was that she had stopped breathing, turned blue and that blood was observed from both nostrils. On examination, she exhibited no abnormal symptoms and there was no evidence of a nose bleed. However, the medical notes recorded that, during an EEG test, a nurse observed the child become extremely upset, suffer a short breath holding attack, turn a poor colour, but not cyanose, not blue, and suffer a small nose bleed. Dr James gave evidence that a breath holding attack was not uncommon in babies who were extremely upset: it might cause the child to turn blue briefly before breathing re-started. Nursing records recorded two periods of vomiting before and after feeds on 15 November. The child was discharged on 16 November.

[10] The child was seen in the Outpatient Department on 26 November. No further turns were reported at that time.

12 - 19 December 1982

[11] The child was re-admitted on 12 December. The history provided was that she had suffered two convulsions of between 5 and 10 minutes and one of around 15 minutes, all involving her face turning blue, her limbs being stiff and her eyes rolling. She was examined by Dr Dugald Glen, then a junior doctor. He noted a bloodstained discharge from her nose. He also noted crackles in her left lung, which were consistent with an infection or pneumonia. Her white blood cell count was raised, indicating infection, and her chest x-ray was abnormal. An ear infection was noted, but was unlikely to have explained the white blood cell count. A raised temperature was noted on 14, 16 and 17 December. Feeding problems, including vomiting, were noted on 16, 17 and 18 December. On 18 December the child was examined by Dr Jacqueline Mok, then a senior paediatric registrar, and at the time of trial a consultant paediatrician and lead clinician for child protection with a post graduate doctorate in respiratory infections. A blocked nose and redness to the ear and the throat were noted, leading to a diagnosis of an upper respiratory tract infection. On 19 December the child was discharged. Her white blood cell count had lowered, her temperature stabilised and she seemed well. Dr Mok accepted, but doubted, that she might have missed signs of pneumonia when examining the child, or that she might have succumbed to an overwhelming pneumonia between 19 and 20 December. She acknowledged that crepitations noted in the child's chest were suggestive of a pneumonic process taking place. Nevertheless, the child ought to have been able to respond to infection. Her opinion was that something dramatic had happened to her.

Events prior to death

[12] On the evening of 19 December 1982 the appellant's then general practitioner, Dr Janet MacFie, attended at the appellant's home address following a report of blueness around the child's lips. The child's father testified that he and the appellant had been in the livingroom of their home when the appellant had run across to the pram in which the child was lying. On Dr MacFie's arrival she examined the child, but noted no abnormal symptoms. She was contacted again later that evening, it being reported that drips of blood had been observed coming from the child's nose. She again attended and examined the child. She noted a brief period of slowed breathing, but was not concerned. Joyce Hall, the appellant's aunt and a qualified nurse, was present. Ms Hall thought that the child appeared to be in some pain, perhaps due to colic, but there did not appear to be anything seriously wrong with her.

[13] The child's father gave evidence that she appeared well when he left for work on the morning of 20 December. In contrast Mrs Brown's log suggested that he had observed her turn very pale and was concerned. The appellant contacted Dr MacFie's surgery at around 8.00 am in a distraught state, reporting that the child was cold. An ambulance was called. Ms Hall attended. She could detect no pulse from the child and performed cardiopulmonary resuscitation. The child was dead on arrival at the hospital shortly after 9.10 am. Her body was examined by Dr James and another doctor. There were no external signs of injury or of petechial or retinal haemorrhaging. The child's frenulum (the fold of skin in the floor of the mouth) was undamaged.

The appellant's police interview

[14] The appellant was interviewed by the police in October 2001. She indicated that the child seemed fine from birth until about four months of age, when she began experiencing spells of turning blue. From about six months of age she had held her breath a lot. The night prior to her death the child seemed to have bad colic. The following morning she was still irritable. As the appellant was feeding her, the blood drained from the child and she turned grey. She was still breathing when the ambulance arrived. The appellant told the police that she had seen blood coming from the child's nose and that she had been blue.

The expert evidence
[15] A wide range of expert evidence was led, both from the Crown and from the defence, interpreting the clinical findings. It is appropriate to summarise that in the order in which it was given at the trial.

The Crown case
Haematology
[16] Dr Angela Thomas, a consultant paediatric haematologist, had examined the blood counts and bacteriological results for the child. She noted that the neutrophil (a subdivision of white blood cells) count could indicate an infection or be secondary to stress, exercise or adrenalin release. The white count and the neutrophil count were raised in response to two known infections (ear and urinary tract). On 12 December there was a significantly raised neutrophil count at a time when no obvious infection was present.

Radiology
[17] Dr Maeve McPhillips, a consultant in paediatric radiology, had noted that the x-ray taken on 12 December had shown some increase in shadowing. She considered that it represented some inflammatory change, probably as a result of infection. She did not consider that a post mortem chest x-ray would have been of assistance.

Radiology
[18] Dr Karl Johnson, a consultant paediatric radiologist, explained that an x-ray, believed to have been taken on 12 December, showed changes in the child's lungs, suggesting her breathing was shallow. This was consistent with infection, but also with the child being fractious or the radiographer inexperienced.

Paediatric respiratory medicine
[19] Dr Neil Gibson, a consultant in paediatric respiratory medicine, distinguished between two broad types of apnoea (stopping breathing) in infants: "central apnoea" due to a failure of the messages to make breathing efforts going out from the brain, and "obstructive apnoea" where the correct messages are sent and received but there is some obstruction or blockage preventing a breath being taken. Central apnoea was excluded by the clinical history. The most likely cause of death was suffocation by deliberately imposed airways obstruction. That could lead to aspiration of stomach contents, causing pneumonia. The reports of blueness prior to the admissions on 14 and 18 October could not be explained by the infections diagnosed. The convulsions described by Dr Glen on 12 December suggested that anoxic seizures had taken place. He rejected bronchiolitis and a single aspiration episode secondary to a gastro-oesophageal reflux as causes. The findings of petechial haemorrhages, haemosiderin, fresh haemorrhage and food in the lungs suggested several episodes of significant hypoxia and finally a severe anoxic event. He accepted that pneumococci were the most common cause of bacterial pneumonia, that the nosebleed observed during the EEG was unusual and potentially significant, and, with regard to the floppiness observed, that abnormal muscle function might affect breathing.

Endocrinology
[20] Dr Christopher Kellnar, a consultant endocrinologist, testified that the child's glucose level was raised during each of her admissions prior to death. That was consistent with her being unwell or particularly stressed. It would be very unusual for an ear infection to raise the level to that noted on 14 October. On 18 October the level was that commonly seen in the context of a fit or convulsion. On 1 November the levels were inconsistent with an ordinary apnoeic attack or with infection (unless severe). He accepted that there were some individuals who had a routinely higher blood glucose level and that this child could be one of them. However, the levels recorded on 14 October and 1 November indicated a "very stressful episode".

Pathology
[21] Professor Busuttil, by now emeritus Professor of Forensic Medicine, testified that the most likely diagnosis of the cause of death, particularly in light of the extensive pulmonary haemorrhage and the haemosiderin, was acute imposed upper airways obstruction. In 1982 he had acted as the second pathologist at the child's post mortem dissection but had not been involved in writing the death certificate or in any examination conducted after the dissection was completed. Aspiration could result from imposed upper airways obstruction. If someone had placed their hand over the child's nose and applied pressure, this might explain the nosebleeds observed. He did not believe that the lung tissue indicated bronchiolitis or that this, along with bronchopneumonia, was the cause of death. The presence of pneumococci in the heart blood was unusual, but might be explained by contamination or by the bacteria being in an early stage of a descending infection. A severe lung infection would have been apparent to the doctors who examined the child when alive, if that had been the cause of death. There was no sign of septicaemia. The level of bleeding in the lungs was too high to have been caused by previous infection or inhalation. The lungs appeared to be congested (engorged with blood) suggesting that the blood vessels there were under pressure to the extent that they were disrupted. On the pleura there were tiny pinprick areas of haemorrhage (petechial haemorrhages); these might be due to either cot death (they were commonly found in such deaths on the thymus gland, the lungs or the heart, albeit their mechanism was not understood) or the child having suffered from a lack of oxygen (asphyxia) for any reason. The presence of haemosiderin in the lungs might indicate past exposure to asphyxia. Over-exertion of the lungs was the most likely explanation for bleeding into the lung tissues. Such over-exertion might be the result of some kind of blockage. That might be natural - as with infections such as bronchitis - but someone dying from pneumonia would have very serious clinical findings. Imposed upper airways obstruction might also cause haemorrhaging. There was evidence of acute inhalation of food. This might be due to imposed upper airways obstruction, though there were other possible causes. Air sacs had collapsed which suggested that something other than recent infection and inflammation had obstructed the airways. Professor Busuttil did not consider that the child's death should have been attributed to "SIDS" or to "SIDS, explained bronchiolitis and bronchopneumonia". The lung tissue did not indicate bronchiolitis. The bruise on the child's head might indicate a bump caused while she was struggling for air. One would not expect to find external evidence of imposed upper airways obstruction unless the hand was applied directly. Professor Busuttil accepted that with aspiration and inflammation one would expect a measure of haemorrhage. It took some time, a matter of weeks, for haemosiderin to form. No single cause other than episodes of imposed upper airways obstruction concluding in a final episode leading to death could account for all the findings.

Paediatric pathology

[22] Dr Alan Howatson, a consultant paediatric pathologist, testified that suffocation, whether imposed or due to natural causes, was the likely cause of death. It was consistent with all of the pathological findings. Aspiration had taken place on more than one occasion and the resultant irritation might explain the bronchopneumonia, but the pathological findings were inconsistent with it being the cause of death. The appearance of the thymus suggested acute stress in the hours prior to death, which might be related to an inflammatory process or to acute upper airways obstruction. The presence of pneumococci was relevant, but was dismissed as the cause of death: it was inconsistent with the observations in the child's final hours and infective bacteria would have been identifiable in the lungs and airways. It might be explained by contamination, or by translocation following resuscitation efforts. The breath holding and jerking incidents noted in the nursing records were accepted as significant: they raised the possibility of a natural cause of death, such as epilepsy and long QT syndrome. He accepted that extrajudicially he had observed that "this was a case full of problems" and that "a number of issues had to be addressed".

Medical statistics
[23] Dr Angela Wade, a senior lecturer in medical statistics, offered some criticism of the statistical conclusions which Professor Morris, a defence expert, had drawn in his report.

Microbiology
[24] Dr Pota Kalima, a consultant microbiologist, was of the view that the child had had an upper respiratory tract infection on her final admission to hospital. Based upon the white blood cell count, it also appeared that a bacterial infection had taken place. The child's clinical presentation suggested that this was not serious. However, the findings on 12 December of a markedly raised white blood cell count and a raised temperature, without an obvious source, could be associated with occult (not clinically detected) pneumococcal bacteraemia (presence of bacteria in the bloodstream). That might also be consistent with her presentation on 19 December, although the necessary investigations had not been carried out. The presence of pneumococcus in the blood was potentially significant and could have been the result of bacteraemia. However, in the absence of supporting histopathological evidence, it was hard to conclude that it played a significant part in death, particularly as it had been found at only one site. If an organism was not consistent with the clinical picture in life, it may well be a contaminant.

Paediatric pathology
[25] Dr Jean Keeling, a retired consultant paediatric pathologist, gave evidence that one had to consider imposed upper airways obstruction as a cause of death. SUDI deaths ("sudden unexpected or unexplained deaths in infancy") were uncommon in children over six months. The bleeding in the lungs and the haemosiderin strongly suggested hypoxic stress. There was no convincing evidence of bronchiolitis. There was no evidence of any natural disease processes which could cause such stress. The bronchopneumonia did not relate to infection, but to the inhalation of gastric contents which had occurred on more than one occasion. The appearance of the thymus suggested a stressful event some hours or days before death, which was consistent with imposed upper airways obstruction. The pneumococcus might be a contaminant, but that would be unusual. Had it been the cause of death, the child would have been visibly unwell beforehand. Deaths from such infection did not occur suddenly. She accepted that there might be neurological causes of aspiration, but that the necessary examinations, particularly that of the brain, had not been carried out. She also accepted that aspiration could inflame the lungs and facilitate bacterial infection. She agreed that there was "no satisfactory answer to the cause of death".

Paediatrics
[26] Dr Helen Hammond, a consultant paediatrician specialising in child protection, prepared a report interpreting the clinical evidence. Her overall opinion was that the circumstances of the child's admissions to hospital and her death were highly suspicious and consistent with repeated upper airways obstruction. The pattern of reported apnoeic attacks, in particular the fact that no such attacks were observed at the hospital, suggested fabrication or induced causation. Bleeding from the nose was known to be associated with imposed upper airways obstruction and was an important indicator. The breath-holding incident during the EEG examination was potentially significant, but had not been well documented: it was not clear if the appellant had been present and accompanied prior to its onset. She dismissed the possibility of a life-threatening pneumonia: this would have been spotted by medical staff and the child's family. She did not accept that the child had bronchiolitis, nor a history of aspiration. She did not accept that hypoxia (lack of oxygen) might have been caused by a seizure, neurological problems or cardiac problems.

The defence case
[27] The evidence of Drs Myerscough and Reid and that of Professor Morris was interposed within the evidence of Dr Gray.

Pathology
[28] Dr Elizabeth Gray, a consultant paediatric pathologist, agreed with Dr Bain's post mortem conclusion that the cause of death was bronchiolitis and bronchopneumonia. The presence of pneumococci, as a single isolate, in the heart blood was extremely significant. As well as infectious pneumonia one could also get chemical pneumonias which included aspiration pneumonia. In her view chemical pneumonia, caused by repeated episodes of aspiration, had weakened the child's lungs leaving them open to a secondary pneumonia which had an infective cause. Her view was that this would explain the pulmonary bleeding, the haemosiderin and the appearance of the thymus. The most common cause of aspiration was gastro-oesophageal reflux, although it might also be related to neurological conditions. If there had been a neuro-pathological examination this might have provided an explanation for the floppiness observed. Aspiration and reflux were not regularly associated with smothering, which she felt was a very complex explanation for death, although she could not altogether rule it out.

Paediatrics
[29] Dr Elizabeth Myerscough, a consultant paediatrician specialising in child protection, accepted that a number of features of the case were suspicious, but indicated that natural causes could not be ruled out. The medical notes did not adequately record whether the onset of the incident during the EEG was independently witnessed. If it had been it would have been significant. The circumstances of the death were consistent with pneumonia being its cause: infants could become unwell and die over a matter of hours. The floppiness observed in the child could have been associated with seizures and a central nervous system abnormality. This may have caused hypoxia. Gastro-oesophageal reflux, possibly accompanied by aspiration, may also have caused hypoxia. Children with low-tone were particularly prone to such reflux. Developmental delays in the child were also potentially significant. In the absence of an ECG, the possibility of heart problems causing hypoxia could not be discounted.

Microbiology
[30] Dr Thomas Reid, a consultant clinical microbiologist, suggested that the most likely cause of death was acute toxaemia resulting from invasive pneumococcal infection. The presence of pneumococci, a common source of which was a middle ear infection such as had been diagnosed in the child, provided a credible explanation for sudden infant death. It might initially go undetected, but once it entered the blood a very rapid deterioration might result. It was not an organism which he associated with post mortem contamination.

Histopathology
[31] Professor James Morris, a consultant histopathologist, concluded that there were four possible causes of death. In order of the most likely these were: pneumococcal pneumonia and pneumococcal septicaemia (the two together causing death); recurrent aspiration of gastric contents leading to recurrent Apparent Life Threatening Episodes (ALTE) and eventually sudden death; another natural but unknown cause; and smothering. The presence of bronchopneumonia in the lungs and pneumococci in the heart blood as well as the appearance of the thymus were all suggestive of an infection. The pneumonia noted was likely to have been caused by a bacterial infection. That could cause the sudden death of a child in apparently good health. The study suggesting that nosebleeds in infants was indicative of smothering had since been discredited. The child's previous admissions were most likely related to a natural disease, probably reflux. He indicated that it would be unusual were the pneumococci discovered to have been a contaminant. The presence of haemosiderin in the lungs did not, in his view, point to smothering as its cause.

Subsequent procedure at the trial
[32] After the close of the defence case the Advocate depute addressed the jury. She described the case as very complex, with very complex medical evidence. She stated that she was not going to go over that evidence or even to summarise it as that would carry a real risk that she would get it wrong or that she might summarise it in a way that was not helpful to the jury. She founded the Crown case substantially upon what she described as a pattern of events involving the child being admitted to hospital on a number of occasions but on investigation there being found to have no or only minor ailments. There were no significant alarms about the child while she was in hospital care. The pattern, when read with the post mortem findings of haemosiderin and food contents in the lungs, pointed to aspiration, not as a result of natural disease but consequential on the deliberate obstruction of the child's airways. The jury should conclude that it had been established beyond reasonable doubt that the child had been repeatedly assaulted by the appellant, the last assault resulting in her death.

[33] Counsel for the defence told the jury that they had "a huge quantity, a mountain, maybe a swamp, of professional opinion to assess". He was not, he said, going to go over that evidence in detail since, like the Advocate depute, he might get it wrong. He emphasised that twenty five years had passed since the child's death and that there were significant sources of medical information which, as a result of practices at the time, were not available for the jury's consideration. The thrust of the Crown case (that there had been repeated induced asphyxiation of the child by the appellant) was negatived by at least one instance where the breathing difficulties had occurred when the appellant was not alone with the child. There had also been a significant gap in time between the fourth and fifth admissions. The evidence in favour of the death being the result of natural causes should be preferred. The Crown had not proved its case beyond reasonable doubt.

[34] The trial judge then directed the jury. As a result, it seems, of a technical failure, the whole of these directions was not duly recorded. However, the trial judge has produced a reconstruction of his charge, the accuracy of which was not disputed by either party. He gave to the jury directions which were in the circumstances adequate as to how they should approach expert testimony. He advised them more than once that he was not going to rehearse that evidence and did not do so. He reminded the jury that the Crown's case as to the commission of the crimes charged was based on circumstantial evidence and that the Advocate depute had outlined the circumstances upon which she relied. He was not, he said, going to rehearse these circumstances and did not do so. He told the jury that they should consider everything that had been said by either counsel.

[35] The jury deliberated for more than a full day. They then returned a majority verdict of guilty on both charges.

Submissions on behalf of the appellant

[36] The appellant's primary ground of appeal was that there had been misdirection by the trial judge. Mr Shead submitted that, standing the complex nature of the case and the divergence in the expert opinion, the trial judge ought to have analysed the evidence in greater detail and offered more guidance to the jury (cf Shepherd v HM Advocate 1996 SCCR 679, per Lord McCluskey at page 684). A number of English authorities provided helpful guidance. In complex cases directions should provide structure, focussing the issues and any material disputes which had to be resolved: one could not rely on the speeches of counsel (R v Amado-Taylor [2000] 2 Cr App R 189). The requirements of a fair trial demanded that a jury's decision be reached on a logically justifiable basis. Thus, in cases involving expert medical evidence, they should, where relevant, be directed on the need before convicting to reject any realistic alternative explanation for the cause of death and on the need for caution in relation to developing medical science (R v Henderson [2010] EWCA Crim 1269, per Moses LJ at paras 200 - 202 and 213 - 221). An analogy could be drawn with the trial judge's duty, in ensuring a fair trial, to give legal directions independently of the manner in which parties presented their case (Ferguson v HM Advocate 2009 SCCR 78, per Lord Osborne at para [30] - [35], adopting the dicta in R v Coutts [2006] UKHL 39; (2006) 1 WLR 2154).

[37] In an associated ground of appeal Mr Shead submitted that, in the circumstances of the present case, the failure of the jury to provide reasons for their decision had denied the appellant a fair trial in terms of article 6 of the European Convention on Human Rights. Beggs v HM Advocate 2010 SCCR 681 could be distinguished: that was not a particularly complicated case (Opinion of the Court, delivered by Lord Eassie, at para [208]). Moreover, it was decided prior to both the Grand Chamber decision in Taxquet v Belgium [2010] ECHR 1806 and recently published research into jury trials by Professor Cheryl Thomas (Are Juries Fair?, Ministry of Justice Research Series, 1/10). Taxquet highlighted the need for safeguards to protect against arbitrariness and to ensure that the reasons for a verdict could be ascertained (at paras 90 - 92). These safeguards might include more detailed directions from the trial judge, preferably in written form according to Professor Thomas' research, or the posing of specific questions to the jury. In the present case, no such safeguards had been adopted.

[38] Viewed separately or in conjunction these grounds of appeal supported the contention that the appellant had been denied a fair trial and that there had been a miscarriage of justice.

Submissions of behalf of the Crown

[39] The Advocate depute submitted that there had been no misdirection by the trial judge. He had a discretion as to the extent to which he directed the jury on the relevant issues, having had the benefit of hearing all the evidence (M'Phelim v HM Advocate 1960 JC 17, per Lord Justice Clerk Thomson at pages 21-2). Shepherd v HM Advocate merely provided guidance on the approach to be adopted in directing a jury, and did not fetter that discretion in any way. There might be cases in which it was incumbent upon a trial judge to give specific directions, but that was not so in the present case (cf Black v HM Advocate 1974 JC 43, per Lord Cameron at page 49). While it had involved a number of experts, they had given evidence over five weeks and that evidence had been tested by senior counsel, who had then addressed the jury: in the circumstances there had been ample opportunity for the exposition of the material issues which had to be resolved. Indeed, in a complex case it was often counter-productive for a trial judge to revisit the evidence or focus on particular points; to do so could complicate matters and distract a jury's focus from other relevant evidence. Reference was made to Hamilton v HM Advocate 1938 JC 134, per Lord Justice General Normand at pages 143-4.

[40] It was submitted that little guidance could be gleaned from English authorities. The system to which they related was not comparable. In particular it was there expected that a trial judge would revisit the evidence in his summing up, whereas in Scotland the duty of the judge was focussed more on the provision of legal guidance. Moreover, the need for a direction that a jury had to reject any realistic alternative explanation for the cause of death was questionable: the real question was whether the case had been proved beyond reasonable doubt (cf R v Henderson).

[41] It was accepted that, if the trial judge had materially misdirected the jury by omitting to give necessary directions, there would have been a miscarriage of justice.

[42] As regards the associated ground of appeal, it was submitted that there was nothing in the Grand Chamber's decision in Taxquet which undermined Beggs v HM Advocate. It made clear that a trial by jury was not incompatible, per se, with the Convention (at para 83) and that it was not necessary for a jury to provide reasons for its decision (at para 90). The safeguards to which it referred were merely examples, and were not definitive. In the present case, the evidence had been heard in public, recorded and rigorously cross-examined. The indictment made clear the facts which had to be proved and the jury were aware that the onus lay on the Crown in that regard. Moreover, the right of appeal ensured due consideration that there had been a logically justifiable basis for the verdict returned. In all of the circumstances it could not be said that the appellant had been denied a fair trial.

Ground (6)
[43] Both counsel addressed us on a separate ground of appeal (ground (6) - alleged misdirection by the judge in relation to express directions given by him in relation to culpable homicide) but in light of our decision on the grounds mentioned above it is unnecessary to rehearse the arguments in relation to that separate ground.

Discussion
[44] In Shepherd v HM Advocate Lord McCluskey, delivering the Opinion of the Court, said at page 684:

"The general principles which govern the treatment of evidence by the judge in his directions to the jury are well-established. The primary duty of the judge is to direct upon the law; and it is usually necessary for him to refer to the evidence to which the relevant law applies. It is, however, a matter for his discretion, in the light of the whole conduct of the trial, to determine the extent to which it is appropriate to refer to the evidence. There is no general obligation on him to attempt to summarise it. In certain types of cases - long, complicated cases, for example - a judge might consider it appropriate to remind the jury of much of the evidence upon which the parties seek to rely. In others - for example, short cases in which there are few issues and they are sharp issues of fact upon which the questioning and the speeches have focused sharply - the trial judge may consider it appropriate not to canvas the evidence at all. He may, in appropriate circumstances decide to refer to parts of the evidence to illustrate the applicability of particular rules of law - for example, rules about hearsay evidence and about fairness as a test of admissibility. He may respond to an invitation made by a party's advocate in a speech to assist the jury in their recollection of a particular body of evidence. He may consider it necessary to amplify or correct some reference to the evidence which has been made in a speech to the jury if the judge considers that the evidence as described in the speech has not been accurately or fairly dealt with. These are only some examples of circumstances which the trial judge will have regard to in determining the extent to which it is appropriate to refer to the evidence. The judgment, however, is clearly one that he is best qualified to make in the context of the trial over which he has presided. He is in the best position to understand what have been the real issues of fact canvassed at the trial and the extent to which the jury might benefit from further clarification."

[45] Lord McCluskey then quoted a sentence from a passage at page 144 in the Opinion of Lord Justice General Normand in Hamilton and Others v HM Advocate. The full passage in which that sentence occurs is as follows:

"The primary duty of the presiding judge is to direct the jury upon the law applicable to the case. In doing so it is usually necessary for him to refer to the facts on which questions of law depend. He may also have to refer to evidence in order to correct any mistakes that may have occurred in the addresses to the jury, and he may have occasion to refer to the evidence where controversy has arisen as to its bearing on a question of fact which the jury has to decide. But it is a matter very much in his discretion whether he can help the jury by resuming the evidence on any particular aspect of the case. I am confident that the anxious willingness of judges to give every help that they can to the jury may be relied on to guide them aright on the extent to which they ought to review the evidence in the case, and I am little disposed to attend with patience to the criticism that more should have been said about the evidence by way of exposition or explanation than the judge thought proper at the time. The Court of Appeal is not in a position to review this discretion of the presiding judge on matters which concern the best way of conducting the case before him."

[46] That the primary duty of the presiding judge in Scotland is to direct the jury upon the law applicable to the case is not in doubt. It is also not in doubt that the judge has no duty when directing the jury to rehearse the evidence which they have heard, that is, to summarise in chronological order the testimony of the witnesses. Not even in England, where judges in practice tend more to enter upon the evidence than in Scotland, is the duty expressed in these terms. In R v Farr, 10 December 1998, unreported but quoted by Henry LJ in R v Amado-Taylor at para 9, Rose LJ is noted as having said in relation to a 13 day trial:

"It cannot be too strongly emphasised that the judge is under no obligation when summing up, to rehearse all the evidence or all the arguments."

It is well established in England that the "fundamental requirements are correct directions in points of law, an accurate review of the main facts and alleged facts and a general impression of fairness" (McGreevy (1973) 57 Crim App R 424, per Lord Morris of Borth-y-Gest at page 430). In short trials it may in that jurisdiction be unnecessary to review the facts at all (Wilson, 14 May 1991 at page 5D, as quoted in R v Amado-Taylor).

[47] The passages cited above from Shepherd and Hamilton might suggest that, in Scotland, there is no obligation on a trial judge in any circumstances to address the evidence when directing the jury; that whether he does so or not is a matter for his unfettered discretion, a discretion with which an appeal court will not interfere. But that would be a misunderstanding. The overarching responsibility of the judge is to ensure that there is a fair trial. A fair trial imports that any verdict, and in particular a verdict of guilty, is not arbitrary but proceeds upon a reasoned basis. The nature of jury trial is such that there is not imposed upon the jury the discipline of disclosing the reasons which have led to their verdict. But that makes it all the more important that they should be provided by the trial judge with a framework which allows them to proceed to a verdict by a reasoned process. Lord McCluskey in Shepherd said: "There is no general obligation on [the trial judge] to attempt to summarise [the evidence]". That sentence, particularly when read with the following sentence, imports that in some circumstances, and in particular in long complicated cases, there may be an obligation to refer to some of the evidence - so that the jury may be focused on the critical issues for their decision. In Black v HM Advocate at page 49 Lord Cameron said:

"In considering the adequacy and sufficiency of a judge's charge to a jury it is at all times necessary to keep in view the circumstances of the particular case under review, its simplicity or complexity, and the legal issues which require to be brought to the attention of the jury. Now great elaboration of these issues can be as confusing to the members of the jury as failure to deal with those which are relevant and require to be explained can be misleading. ... As decisions on matters of fact and judgments on evidence are essentially the province of the jury, we do not think that it is incumbent on a judge - or even desirable that he should, in every case and as a matter of practice do so - to embark upon a detailed rehearsal or analysis of the evidence. In many cases - of which the present may well be one - the evidence lies within such short compass and has so recently been led before the jury that to enter upon a recapitulation of it would be wholly superfluous and might even be confusing.

At the same time there are cases in which the form of the indictment, the nature and complexity of the charge or charges themselves or of the evidence relating to them demand detailed directions in law and some balanced examination of the evidence in order to focus accurately for the jury the legal context within which their consideration of the evidence and the determination of their verdict have to be conducted."

[48] Cases involving the deaths of infants allegedly at the hands of a parent or other carer are amongst the most difficult, and potentially the most complex, of all cases coming before the criminal courts. In many such cases, and the present is such, there will be no direct evidence of criminal conduct by the accused towards the child. The case will largely, if not exclusively, depend on inferences to be drawn from medical testimony. In this field, while knowledge advances, there remain many uncertainties. Establishing the cause of a sudden infant death may be very difficult and in some cases may not be possible. If criminal liability is to be brought home to the accused it will be necessary to exclude not only any natural explanations for the death suggested in the evidence, but also any realistic possibility of there being an unknown cause. In R v Henderson (a series of cases concerned with "shaken baby syndrome") Moses LJ said at para 217:

"... a realistic possibility of an unknown cause must not be overlooked. In cases where that possibility is realistic, the jury should be reminded of that possibility. They should be instructed that unless the evidence leads them to exclude any realistic possibility of an unknown cause they cannot convict. In cases where it is relevant to do so, they should be reminded that medical science develops and that which was previously thought unknown may subsequently be recognised and acknowledged. As it was put by Toulson LJ, 'today's orthodoxy may become tomorrow's outdated learning' (R v Houldsworth [2008] EWCA Crim 971 at [57]). In cases where developing medical science is relevant, the jury should be reminded that special caution is needed where expert opinion evidence is fundamental to the prosecution."

[49] Where there is on the evidence a realistic possibility of there being an unknown (currently unknowable) cause of death, as there was on Professor Morris' evidence, the jury should be reminded that this must be excluded before they can convict. Also, of course, the jury will require, before they can convict, to exclude any natural cause of death which is positively suggested in the evidence. This may be obvious, but the jury should be reminded of it. The way in which a jury approaches its deliberations upon the issues before it will be a matter, ultimately, for them to decide; but in a case of the present kind it will be incumbent on the judge to give them guidance. Even where counsel for the prosecution and for the defence have in their speeches fully explored the technical evidence (and they did not do so here), it will be for the trial judge to provide a succinct, balanced review of the central factual matters for the jury's determination. Where natural causes for the death are suggested in evidence, an appropriate starting point may be to remind the jury what these are, giving a brief explanation of the evidential basis for each of them, and directing them that, if they are persuaded of them or if any evidence with respect to them gives rise to a reasonable doubt about the accused's guilt, they are bound to acquit.

[50] In the present case one expert led by the defence (Dr Reid) testified that the most likely cause of the child's death was acute toxaemia resulting from invasive pneumococcal infection. This was based on the presence of pneumococci in the heart blood, as detected at the post mortem examination. The child's records indicated that she had inflamed eardrums. One of the most common causes of middle ear infection was pneumococcus. Dr Reid had no difficulty with the fact that the child had been discharged from hospital the previous evening and had been seen twice by the family's general practitioner twice later that evening but had died suddenly the following morning. On the other hand, he acknowledged that he was not an expert in the effect of smothering on the lungs. He offered no opinion on the evidence that there had been prior bleeding in these organs.

[51] Professor Morris concluded that there were in this case four possible causes of death. In order of the most likely first, these were (i) pneumococcal pneumonia and pneumococcal septicaemia, (ii) recurrent aspiration of gastric contents, (iii) another natural but unknown cause and (iv) smothering. While he could not exclude the last, the simplest and most likely explanation, in his view, was recurrent bouts of aspiration and finally aspiration plus pneumococcal pneumonia causing death. He thought the most likely explanation for recurrent episodes of aspiration was gastro-oesophageal reflux. He thus offered a possible explanation for there being previous affection of the lungs.

[52] Dr Myerscough also offered gastro-oesophageal reflux with aspirations as a possible explanation for the child's presentation on 19 December. She raised the possibility of neuro-developmental problems as the origin of that reflux. She pointed to aspects of the child's medical history which might point to such problems. Imposed upper airways obstruction was a possibility but not in her view the only possibility.

[53] Dr Gray's view as to the cause of death coincided with that expressed by the principal pathologist who had carried out the immediate post mortem examination - bronchialitis and bronchopneumonia. She attributed the aspiration to gastro-oesophageal reflux, which might also be related to neurological conditions. The finding of pneumococci was, in her view, significant. Although she could not altogether rule out smothering as the cause for the aspiration, she concluded that the cause of death was purulent bronchopneumonia.

[54] The views of the experts led by the defence varied - in their particulars and in their emphases - but all afforded natural explanations for the death (and in most cases also for the earlier alarms leading to hospital admissions) which the jury would have to reject, and have a proper basis for rejecting, if they were to be satisfied beyond reasonable doubt that the accused had repeatedly obstructed the child's upper airways, on the final occasion causing her death. It was, in our view, incumbent on the trial judge to focus the jury's attention on these explanations to allow them to approach their task in a logical and rational way and so to reach a reasoned verdict. Simply to leave the technical evidence at large for the jury involved, in our view, in the circumstances of this case, a misdirection.

[55] If the trial judge was to identify and succinctly to describe the natural explanations advanced by the defence (as in our view he was), he would in fairness have required to conduct a like exercise in relation to the evidence upon which the Crown relied for the propositions that the child had been assaulted and ultimately killed by the accused. The technical evidence which the Crown led was wide-ranging and involved a number of distinct disciplines but it all tended, to a greater or lesser extent, towards the proposition that the accused had on repeated occasions over the period between 14 October and 20 December 1982 restricted the child's breathing and on the last occasion suffocated her. The pattern of events whereby the child was on many occasions admitted into hospital having reportedly suffered difficulty in breathing but on being admitted found to have no problem other than, on some of these occasions, apparently minor ailments was, certainly on one view, suspicious. In the view of Dr Hammond the circumstances were, for the reasons she gave, highly suspicious and consistent with repeated upper airways obstruction. Dr Gibson, an expert in paediatric respiratory medicine, was equally definite. Professor Busuttil and Dr Kellnar gave positive support to the Crown case, as did Dr Howatson and Dr Keeling. A succinct reference to the bases for each of their views would have been appropriate.

[56] The task which it was, in our view, incumbent on the trial judge to perform was not an easy one. There was a risk that, in seeking to identify the position of each of the principal experts, errors might be made. But the fact that the task was difficult does not mean that it was unnecessary. The failure to undertake that task in this case amounted in our view to a material misdirection. It is not disputed that, if we came to that view, the result was that there had been a miscarriage of justice and that the verdict would require to be quashed. We accordingly allow the appeal and quash the verdict.

[57] In these circumstances it is unnecessary to address in detail the ground of appeal based on an alleged infringement of the appellant's Article 6 right to a determination by a reasoned judgment. It is quite clear that, notwithstanding that a jury's verdict is not "transparent", jury trial in Scotland is not by that reason inconsistent with Article 6 (Beggs v HM Advocate, at paras [207]-[208], where account was taken, of among other authorities, the judgment of the Second Section of the European Court of Human Rights in Taxquet v Belgium, Application No.926/05, 13 January 2009). Since then the Grand Chamber of the Court has delivered its judgment in Taxquet (16 November 2010), in which it held that the institution of the lay jury cannot be called into question in the context of an Article 6 challenge. A decision to the like effect was reached by the Second Section of the Court in the context of Scottish criminal proceedings in Judge v United Kingdom, Application No.35863/10, 8 February 2011.

[58] It is central to these decisions that the jury's verdict is not given in isolation but in a framework which includes the presiding judge's charge to the jury. It accordingly follows that, if in the circumstances of the particular case that aspect of the framework is materially inadequate, the verdict may be vulnerable to challenge under Article 6. As we have already found that at common law there was a miscarriage of justice, it is unnecessary to reach a firm conclusion on the Article 6 point. We would, however, be inclined to hold that, given the complexity of the issues at this trial and the failure of the judge to give to the jury adequate guidance as to how they might address them, there was a breach of Article 6.

[59] As earlier stated, a further ground of appeal was argued but in the circumstances it is unnecessary to deal with it.