JS (a minor), Re

On 25 February 2011, a baby boy, whom I shall call J, then aged 15 weeks, was admitted to Gloucestershire Royal Hospital because of concerns about his excessive head growth. CT scans revealed that he had sustained subdural haematomas and he was transferred to the Frenchay Hospital in Bristol, where he underwent two operations in which the subdural collections were drained. Examination by ophthalmologists also revealed the presence of bilateral retinal haemorrhage.

Very fortunately J has made a complete recovery from these conditions and it is not anticipated that he will suffer any long term consequences, but the presence of subdural haematomas and retinal haemorrhage raises the question whether he has been the victim of a shaking, or shaking plus impact, assault, and the local authority and police embarked upon an investigation. Upon discharge from hospital, J was placed in the care of the local authority and, specifically, foster parents, where he remains to this day. The police have yet to reach any final conclusion in their investigation, but the local authority started care proceedings, arguing that J's intracranial injuries were inflicted non-accidentally by one or other of his parents, whom I shall refer to as his mother and father in order to preserve their anonymity. Although it was initially the local authority's case that the injuries were inflicted by one or other of his parents, in final submissions the local authority withdrew the allegation against the mother.

The case was transferred to the High Court and listed for fact finding hearing before me over 15 days in March 2012. This judgment is delivered at the conclusion of that hearing.

J's mother was born in 1973. She had a troubled childhood and her medical records reveal that she had a history of depression, bulimia and poor self image. She has two older children by previous relationships, whom I shall refer to as L, born [a date in] 1994 and therefore now aged 17, and X (I choose that initial because he shares the same initial as J), born [a date in] 2003 and therefore now aged eight. X suffers from cerebral palsy as a result of complications that occurred during his mother's pregnancy. The experts instructed in these proceedings have reviewed X's medical history and concluded that it has no relevance to the analysis of the injuries sustained by J. Both L and X have had regular contact with their respective fathers. The relationship between the mother and L's father remains cordial, but there have been ongoing difficulties between the mother and X's father. The mother suffered from post-natal depression following the births of both L and X. On each occasion she took an overdose whilst suffering from that depression.

J's father was born in 1969 and has worked as a driver for most of his adult life. His only other significant relationship prior to meeting the mother was with a woman some years older than he. That relationship lasted four years before they drifted apart, according to the father because he wished to have children. Following the breakdown of that relationship, at the point when he was heavily in debt, the father moved back to live with his mother in a prefab bungalow in Brockworth near Gloucester. The mother, L and X live next door in an adjoining bungalow. The parties met when the father offered to do some gardening and odd jobs around the house for the mother. Out of this they started the relationship, although they never lived together save for a few weeks at the end of 2010 in circumstances described below.

According to the mother, from an early stage in the relationship it was characterised by violence perpetrated by the father. In December 2009, the mother rang the police following an argument, but when the officers arrived she refused to speak to them and left the property. The police were involved in a further incident in April 2010, but again the mother declined to pursue her complaint. By this point the mother was two months pregnant and she claims that the reason for the argument was that the father was demanding that she have an abortion. That evening the father was admitted to hospital suffering from alcohol poisoning. He claims that this was an unusual incident for him and that he does not normally drink to excess. In September 2010 the mother attended the Accident and Emergency Department at the local hospital complaining of pain to the right ribs, saying that she had fallen five days previously in the kitchen and struck the sink. She denied that the injury had been sustained as a result of domestic violence. This incident has been the subject of evidence at this hearing which I shall consider below.

On 10 October 2010 the father was found on a motorway bridge saying he was depressed as a result of his problems, including debts. He was referred for medical help and in subsequent sessions with his GP complained that he was experiencing stress at home as a result of his debts, strains in his relationship and the fact that his girlfriend was pregnant. He saw the GP on a series of sessions in which he reported that he was continuing to experience stress. On one occasion his mother reported that he was demonstrating "bouts of anger" towards her and the mother which were "increasing in severity". Meanwhile, the mother herself was experiencing difficulties in the later stages of her pregnancy, suffering from a low-lying placenta and an irritable uterus.

On 11 November 2010 the mother gave birth to J at 37 weeks by caesarean section. Medical examination of the baby revealed no complications. His head circumference was recorded at 33.8 centimetres which, allowing for the fact that he was three weeks premature, placed him approximately on the 50th centile. According to the father, a day or so after the birth a doctor remarked that the plates in J's had not fused and that, as a result, the mother panicked, fearing a repetition of the problems she had experienced with X.

J was discharged from hospital on 14 November and he and his mother went to stay with the father at the paternal grandmother's home next door to the mother's property. The care of J was thereafter shared by the parents, including the night-time feeds. On 24 November the health visitor, BF, made her first visit to see the baby. All seemed well. She measured his head circumference and recorded in his notebook the measurement as 35 centimetres, also on approximately the 50th centile. Six days later the health visitor visited the baby again and recorded no problems. It seems that no head circumference measurement was taken

Meanwhile, the father's problems with stress were continuing. A record in his GP notes for 17 December state that on that date he attended complaining that he was still suffering from anxiety and feeling stressed, that he was finding the new baby difficult to cope with and that his bankruptcy was no further forward.

On 23 December the mother and J moved back next door to the mother's property. As they were living next door, there was a great deal of direct communication between the parties, but in addition they communicated very frequently by text message. A large number of texts sent by the mother to the father have been produced at this hearing and disclose that their relationship was extremely volatile. No text messages sent by the father to the mother during this time have been made available, although text messages sent by the father to the mother after J’s accommodation in foster care have been produced. The mother alleges that on 24 December the father forcibly removed J from her care. The following day, Christmas Day, the mother (with the assistance of a lift from the paternal grandmother) took J to her mother's house. In text messages over Christmas, the mother alleged that the father had been violent to her again. In addition, over the Christmas period, there was an altercation between the father and his brother outside the properties when the brother attempted to take father's car keys, fearing that he was unfit to drive.

Thereafter, J spent the majority of his time with the mother, but on occasions was cared for by the father. The volatile relationship between the parties continued. The mother alleges that, during one argument on 4 January, the father punched her on the face, causing her to suffer a black eye. As a result she did not answer the door to the health visitor when she visited the following day. The father alleges that, when he was looking after J on 8 January, an incident occurred in which J, in a sleeping bag, started to slip off the sofa and, to prevent his falling, the father grabbed him by the arm. As a result, J sustained a small bruise which the mother claims she did not see until the following day.

On 13 January, it is alleged by the mother and L that a further incident took place in which the father was violent not only towards the mother but on this occasion towards L. It is alleged that he grabbed L by the throat and pinned her towards the wall causing damage to the wall. The police were called, but once again the mother refused to give any details of the incident and asked the officers to leave. The details of this incident have been a matter of contested evidence which I shall consider below.

On 17 January, the health visitor examined J again at home. She noticed appropriate weight gain and that J was starting to smile, fix his gaze and follow with his eyes. She also noticed, however, that his head circumference had increased to 41.5 centimetres, which was on or around the 98th centile. As the rest of the examination was apparently normal, the health visitor took no further steps about this head circumference measurement at that stage.

It has been the father's case at the this hearing that when J was staying with him overnight on 21 January an incident occurred while he was feeding J in the middle of the night in which he appeared to have a spasm and his breathing was affected. The father later said that he thought that J had died. He summoned his mother, who was sleeping in the property. She thought that J had suffered a reaction to feeding, but he started to recover shortly afterwards, although was not his normal self. Neither the father nor the paternal grandmother sought medical attention at this stage. There are a number of disputes about this incident, including an issue as to the date on which it occurred. As already stated the father asserts now that it occurred on 21 January, but he has told a number of people earlier in the investigation that it happened when J was between five and seven weeks old, that is to say between about 16 and 30 December. There is also a dispute between the mother and the father as to what the father told the mother about this incident at that stage. I shall consider the evidence about this incident later in the judgment.

On 24 January, J was taken to the GP's surgery by his parents for what was intended to be his eight week check, although in fact it was taking place when he was about 11 weeks old. The paternal grandmother also attended. Neither the father nor the mother nor the grandmother mentioned the incident which is said to have occurred three days earlier. J appeared to be making good progress, although the GP, Dr. B, also noticed, as had the health visitor, that his head circumference was now on the 98th centile. The GP therefore arranged for the health visitor to repeat the measurement in four weeks. According to the mother's text messages for a few days after the examination J suffered from a cold with diarrhoea. He was unwell again for a few days in the second week in February. During this period he stayed with the father overnight on several occasions.

On 12 February, according to the father, a further incident took place while he was looking after J in the middle of the night. Once again, J recovered so the father did not seek medical attention. Once again, there is an issue about the date of this incident and about what the father said about it to the mother.

The parents allege that a further incident occurred a few days later. Again, there is uncertainty about the date. On this occasion both the father and the mother were present. J was being fed by the mother and became very lethargic for a short while. The mother states that the father observed that this was what had happened in the earlier incidents. Again, no medical advice was sought after this third incident.

On 22 February the health visitor visited the home again in accordance with the GP's directions. On this occasion J's head circumference was measured at 45 centimetres, which was considerably above the 99.6^th centile. The health visitor therefore referred the parents back to the GP.

On the following day, Dr. B received a letter from J's grandmother stating:

Concerned that these three episodes might represent convulsions Doctor B informed the paediatric department at the Gloucester Royal Hospital and J was admitted to the hospital on the afternoon of 25 February. Following admission a CT scan was performed which appeared to show bilateral subdural haemorrhages of different ages suggestive of non-accidental injury. In the small hours of 26 February, J was transferred to the Frenchay Hospital in Bristol. Later that day an MRI was carried out and reviewed by Dr Likeman, Consultant Neuroradiologist. He reported that the MRI revealed a marked discrepancy in the size of the skull and brain resulting in a wide extra-axial space. In addition, he observed large collections of subdural haemorrhage over both cerebral convexities. The collection was larger on the left hand side, where there was evidence of two phases of bleeding. A vascular neo-membrane was observed within the collection. Dr Likeman advised that the presence of the membrane indicated bleeding which was older than two weeks prior to the MRI. There was also evidence of more acute bleeding behind the left parietal and right anterior frontal lobe. Dr Likeman concluded that the findings were highly suspicious of non-accidental injury caused by shaking. He added, however, that there was a history of a large head circumference so that a benign external hydrocephalus resulting in craniocerebral disproportion was a possible cause of the intracranial bleeding. He advised that an ophthalmological examination and a skeletal survey should be carried out.

On the following day J was examined by Dr Emma Bradley, Consultant Community Paediatrician, in the presence of the parents, from whom she took a history in which they gave an account of the three "strange episodes" to use the phrase used to describe them in the medical notes by Dr Bradley. J was also examined by Mr Mike Carter, Consultant Paediatric Neurosurgeon, in the presence of his parents. Mr Carter observed J to be well-nourished with no obvious signs of injury. There was no tonic or clonic signs in his limbs, but his anterior fontanelle was full and quite dense and he showed signs of sunsetting, that is to say the phenomenon where the irises in the eyes fall below the lower eyelids, said to be a sign of raised intracranial pressure. As a result of concerns that J was suffering from raised intracranial pressure, Mr Carter performed an operation on J's head in which an ultrasound-guided percutaneous aspiration of the subdural collection was carried out. On the left side, two different collections of fluid were encountered during the operation, the more superficial containing very heavily blood-stained fluid, but the more deeper containing much less heavily stained fluid which was slightly brown in colour. A total of 40 millilitres of fluid were aspirated from the left side collection. An attempt was made to aspirate fluid from the right side subdural space, but was unsuccessful. CT scans taken after the procedure demonstrated quite marked reduction in the size of the collection and again suggested the presence of enlarged extra-axial spaces.

On 28 February, J was examined by an ophthalmological registrar who identified bilateral pre-retinal haemorrhage with a vitreous haemorrhage in the left eye. On 2 March, J was examined by Dr Williams, a Consultant Paediatric Ophthalmologist. She concluded there were bilateral retinal haemorrhages and pre-retinal haemorrhages on the right side. A skeletal survey was carried out around the same time, but revealed no further injury.

On 4 March a further percutaneous aspiration of the left sided subdural collection was performed by Dr Carter. Once again fluids of two densities were encountered and on this occasion a total of 35 millilitres was removed.

At a strategy meeting on 8 March, supervising paediatrician Dr Bradley advised that, on a balance of probabilities, the likely cause of J's injuries was non-accidental. On the following day, the parents were informed about this advice given to the strategy meeting. As a result, the father agreed to J being voluntarily accommodated, although the mother was very upset and did not talk to the social worker. This disagreement between the parents is said by the mother to have led to a further violent episode. Later that day the mother took an overdose of 16 Anadin tablets and was admitted to hospital.

On 11 March, J was transferred back to Gloucester Royal Hospital and three days later discharged from hospital into foster care. X went to live with his father and L went to live initially with her grandparents but returned after a while to the mother's home. The three children are still living at these separate addresses. There is a complex programme of contact between the siblings themselves and between the siblings, the two younger children and the mother and between J and his father.

Thereafter, the police carried out an investigation into the cause of J's intracranial and retinal bleeding. Both parents denied in interview that they had shaken J. In her interview the mother made no reference to the father's acts of violence on her. She described him as a good person who was just a bit loud sometimes. Transcripts of the interviews have been prepared and put before the court.

On 22 March, Gloucestershire County Council applied for a care order in respect of J. The matter was transferred to the county court. On 26 April, J was admitted to the hospital with a febrile illness, being discharged some three days later. During this admission his head circumference was measured at 45.5 centimetres, at that stage being between the 91^st and 98^th centile. Meanwhile, the parties' turbulent relationship had continued, as described in further text messages passing from the mother to the father and father to the mother. Eventually on 20 May the mother applied for a non-molestation injunction without notice, which was granted by District Judge Singleton in the Gloucester County Court. On 9 June the injunction was extended by that district judge for a period of one year. A few days earlier, at a hearing on 3 June, Her Honour Judge Darwall-Smith had given directions for these proceedings, including the commissioning and timetabling of experts' reports from a variety of disciplines. The proceedings were transferred to the High Court and listed before me initially for directions in July, at which point the case was timetabled through to a final hearing in March 2012.

Following the voluntary accommodation of J, as stated above, residence of X was transferred in the interim to his father. X continued to have contact with his mother twice a week at the home. Because of concerns about the risk to X as a result of J's father living next door, X's father sought an extension of the injunction against J's father. A without notice injunction was granted by this court, but in the event its extension was not pursued.

About three weeks before the date of the fact finding hearing before me, the mother disclosed to her solicitors for the first time a lengthy handwritten journal in which she set out in considerable detail allegations of domestic violence said to have been perpetrated by the father and also certain statements by the father which, on one view, suggested he had shaken J. The mother had been re-interviewed, as had the father, in May and June 2011 and in the course of her interviews the mother had referred to allegations of domestic violence, but the journal prepared by the mother was not produced at that stage and not produced, as I have said, until about three weeks before this hearing. Fearing that production of the journal would precipitate violent behaviour from the father the mother applied for a further injunction seeking an exclusion zone preventing the father from coming within a certain distance of her home. The effect of such an injunction would have been to prevent the father remaining in occupation of his mother's home next door to the mother's property. Having heard argument on this application I declined to make the injunction sought on behalf of the mother.

The central issues to be determined at this hearing are; first, whether the injuries sustained by J were inflicted non-accidentally, secondly, if so, who was responsible for inflicting them and thirdly, was there one or more than one incident giving rise to the injuries. If the injuries were inflicted it is not suggested by any party that the perpetrator was anyone other than the mother or father. Indeed, at the conclusion of this hearing the local authority withdrew any suggestion that the perpetrator could have been the mother. The father does not to seek to contend that the mother has injured J. Thus, if the injuries were inflicted the only person suggested as a perpetrator for those injuries is the father. There are, as always, subsidiary issues and matters to be considered, including the current state of medical knowledge about intracranial bleeding; the credibility of the family witnesses; the provenance and reliability of the mother's so-called journal; the relationship between each of the parents and the baby; the dynamics of the relationship between the family members, and the extent of domestic violence in the home and its effect upon the children.

A large volume of paper evidence and other written material has been put before me, comprising inter alia; the local authority threshold document and the parties' response thereto; statements made by clinical professionals and family members for the purposes of these proceedings; evidence collected by the police, including long transcripts of the extensive interviews carried out of the parents in March, May and June 2011 and again in January 2012; J's medical records, including GP records and hospital records; the medical records of both parents; reports from medical experts specifically instructed for these proceedings, all leading experts in their fields, namely Professor Michael Patton, Professor of Medical Genetics and Consultant Medical Geneticist at St Georges Hospital, London, Professor Sally Kinsey, Consultant Haematologist at St James University Hospital, Leeds, Miss Gillian Adams, Consultant Ophthalmic Surgeon at the Moorfields Hospital in London, Dr Philip Anslow, Consultant Paediatric Neuroradiologist at the John Radcliffe Hospital in Oxford, Mr Peter Richards, Consultant Paediatric Neurosurgeon at the John Radcliffe, and Professor Peter Fleming, Professor of Infant Health and Developmental Physiology at the University of Bristol and Consultant Paediatrician at the University Hospitals in this city, plus written answers to supplemental questions and a transcript of a joint telephone conference between some of the experts, accompanied by a number of published articles concerning medical issues arising in the case and, finally, contact recordings setting out details of contact visits between the parents and J. I was also shown a DVD of a recording of J taken by the father at various stages between his birth and admission to hospital in February. I also listened to a recording taken by L on her mobile phone of two instances of shouting at the mother.

Finally, I had the benefit of written material prepared by counsel comprising position statements and skeleton arguments and finally written submissions. At this point I wish to express my thanks to all counsel, leading counsel and junior counsel, for their assistance in this case. The quality of advocacy and representation at this hearing has been uniformly high. I also wish to thank their instructing solicitors and the representatives of their firms for their work during this hearing. The careful preparation by all lawyers in these proceedings has been of very considerable assistance to me.

The following witnesses gave oral evidence at the hearing; DC Vincent, one of the police officers in the case; Miss Gillian Adams; Mr Mike Carter, the Consultant Neurosurgeon who carried out the two operations removing fluid from the subdural collections; Kate Upton, the Social Worker for J; Dr Philip Anslow, Consultant Paediatric Neuroradiologist instructed as an expert witness; Dr B, the GP; Dr Bradley, the Consultant Community Paediatrician who was one of the doctors who examined J upon admission to hospital; BF, the Health Visitor; Mr Peter Richards, Consultant Paediatric Neurosurgeon instructed as an expert witness; the maternal grandmother; Professor Peter Fleming, Consultant Paediatrician also instructed as an expert; the mother's sister S; L; the maternal grandfather; the paternal grandmother; the mother and, finally, the father.

I mention at this point that the video recording taken by the father showing J at various stages between birth and admission to hospital in February, whilst available to all parties for several months, was not for some reason shown to any of the experts or clinicians, save for Professor Fleming, who made observations about it in the course of his evidence.

In determining the issues at this fact finding hearing I apply the following principles. First, the burden of proof lies with the local authority. It is the local authority that brings these proceedings and identifies the findings they invite the court to make. Therefore the burden of proving the allegations rests with them.

Secondly, the standard of proof is the balance of probabilities (Re B[2008]UKHL 35). If the local authority proves on the balance of probabilities that J has sustained non-accidental injuries inflicted by one of his parents, this court will treat that fact as established and all future decisions concerning his future will be based on that finding. Equally, if the local authority fails to prove that J was injured by one of his parents, the court will disregard the allegation completely. As Lord Hoffmann observed in Re B:

Third, findings of fact in these cases must be based on evidence. As Munby LJ, as he then was, observed in Re A (A Child) (Fact-finding hearing: Speculation)[2011] EWCA Civ 12:

Fourthly, when considering cases of suspected child abuse the court must take into account all the evidence and furthermore consider each piece of evidence in the context of all the other evidence. As Dame Elizabeth Butler-Sloss P observed in Re T[2004] EWCA Civ 558, [2004] 2 FLR 838 at 33:

Fifthly, amongst the evidence received in this case, as is invariably the case in proceedings involving allegations of non-accidental head injury, is expert medical evidence from a variety of specialists. Whilst appropriate attention must be paid to the opinion of medical experts, those opinions need to be considered in the context of all the other evidence. The roles of the court and the expert are distinct. It is the court that is in the position to weigh up expert evidence against the other evidence (see A County Council & K, D, & L [2005] EWHC 144 (Fam); [2005] 1 FLR 851 per Charles J). Thus there may be cases, if the medical opinion evidence is that there is nothing diagnostic of non-accidental injury, where a judge, having considered all the evidence, reaches the conclusion that is at variance from that reached by the medical experts.

Sixth, in assessing the expert evidence I bear in mind that cases involving an allegation of shaking involve a multi-disciplinary analysis of the medical information conducted by a group of specialists, each bringing their own expertise to bear on the problem. The court must be careful to ensure that each expert keeps within the bounds of their own expertise and defers, where appropriate, to the expertise of others (see observations of King J in Re S[2009] EWHC 2115 Fam).

Seventh, the evidence of the parents and any other carers is of the utmost importance. It is essential that the court forms a clear assessment of their credibility and reliability. They must have the fullest opportunity to take part in the hearing and the court is likely to place considerable weight on the evidence and the impression it forms of them (see Re W and another (Non-accidental injury)[2003] FCR 346).

Eighth, it is common for witnesses in these cases to tell lies in the course of the investigation and the hearing. The court must be careful to bear in mind that a witness may lie for many reasons, such as shame, misplaced loyalty, panic, fear and distress, and the fact that a witness has lied about some matters does not mean that he or she has lied about everything (see R v Lucas[1981] QB 720).

Ninth, as observed by Hedley J in Re R (Care Proceedings: Causation)[2011] EWHC 1715 Fam:

Finally, when seeking to identify the perpetrators of non-accidental injuries the test of whether a particular person is in the pool of possible perpetrators is whether there is a likelihood or a real possibility that he or she was the perpetrator (see North Yorkshire County Council v SA[2003] 2 FLR 849. In order to make a finding that a particular person was the perpetrator of non-accidental injury the court must be satisfied on a balance of probabilities. It is always desirable, where possible, for the perpetrator of non-accidental injury to be identified both in the public interest and in the interest of the child, although where it is impossible for a judge to find on the balance of probabilities, for example that Parent A rather than Parent B caused the injury, then neither can be excluded from the pool and the judge should not strain to do so (see Re D (Children)[2009] 2 FLR 668, Re SB (Children)[2010] 1 FLR 1161).

As explained by the Court of Appeal in the case of R v Harris and Others[2005] EWCA Crim 1980, the medical diagnosis of non-accidental head injury has for some years been based on what is called the triad of intracranial injuries consisting of encephalopathy, (defined as the disease of the brain affecting the brain's function), subdural haemorrhage and retinal haemorrhage. For many years the coincidence of these injuries in infants was considered to be the hallmark of non-accidental head injury. As was made clear by the Court of Appeal in the Harris case, however, the presence of this triad is a strong pointer to, rather than diagnostic of, non-accidental head injury. In each case the court must carry out a thorough investigation of the facts. In many cases the child will have suffered other injuries that are often associated with non-accidental injury, for example bone fractures. In other cases some of the elements of the triad may be missing. But having examined all the evidence the court may still conclude that the child has suffered a non-accidental head injury.

As the Court of Appeal stated some years ago in Harris, determining findings in these difficult cases requires medical evidence from experts from a number of different disciplines, interpreting often very small signs within the complex structures of the infant brain and surrounding tissue. Typically, in my experience, these disciplines will include a paediatric neurologist, in some cases a paediatric radiologist, a neurosurgeon, an ophthalmologist, a haematologist and a paediatrician. If the child has died a pathologist and/or a neuropathologist will be required. Each of these experts brings an essential contribution to the very complex analytical process. Whilst the courts always have to be vigilant to guard against the proliferation of experts in family proceedings, the court must, in my judgment, always have available to it the necessary expertise to make the right findings in these important and difficult cases.

I start with some basic anatomy and physiology derived principally from Dr Anslow's and Mr Richards' reports. Enveloping the brain and the central nervous system is a system of membranes known as the meninges which consist of three layers; the pia, which follows the contours of the brain, the arachnoid and, nearest to the skull, the dura. Between the pia and the arachnoid there is a space, the sub-arachnoid space or extra-axial space, which is filled with cerebrospinal fluid. Between the arachnoid and the dura there is no actual space but rather a potential space. (In a variation of the anatomical description normally presented to the courts Dr Anslow in these proceedings described the dura and arachnoid as being one continuous but asymmetric membrane which separates along an easy plane of cleavage into an inner arachnoid layer and an outer dura layer.)

A collection of blood in the potential subdural space is called a subdural haematoma. Subdural haematomas do not occur as spontaneous events in children. Something has to cause them. In the absence of a bleeding disorder, some pre-disposing malformations and certain very rare metabolic disorders, subdural haematomas are strongly associated with trauma. The ultimate mechanism of the formation of subdural haematomas is thought to be rupture by shearing forces of the small unsupported bridging veins which pass between the brain and the skull. These veins arise within the brain and pass through the pia, where they run into the sub-arachnoid space for some centimetres. They then travel through the arachnoid and dura before emptying into a venous site. If a vein ruptures while passing through the potential subdural space a subdural haematoma is created. Fresh bleeding in the spaces causes an acute subdural haematoma. Such bleeding normally disappears without trace, other than some changes visible through a pathologist's microscope. However, in certain individuals the acute blood breaks down and becomes more watery, exciting an inflammatory response within the subdural space which leads to the development of fragile subdural membranes. Instead of the blood disappearing the subdural space fills up with progressively more watery fluid and intimate episodes of spontaneous re-bleeding which may lead to the volume of fluid in the subdural space becoming greater over time. This dynamic situation develops over a period of at least two weeks and is known as chronic subdural haematoma.

In the infant the accumulation of chronic subdural haematomas is usually accompanied by the expansion of the skull, which is possible because the growth plates in the skull have not yet fused. This allows a significant volume of fluid to collect without any clinical symptoms becoming apparent, other than enlargement of the head. Mr Richards observed that appreciating that a head is enlarging rapidly can be difficult for the everyday observer and is usually identified by someone who has not seen the child for some weeks or, alternatively, by a routine surveillance examination of the head circumference. The origin of a chronic subdural haematoma is, therefore, an acute bleed into the subdural space some time in the child's life, usually at least two weeks before presentation and maybe longer.

There are a large number of medical conditions which are known to cause spontaneous subdural haematoma. In this case, however, all appropriate investigations have been conducted and ruled out all such medical causes. In the absence of such a cause the remaining cause is injury. In the case of an infant, who is not ambulant, the possible causes of such an injury are birth, accidental trauma and non-accidental trauma.

Recent research, specifically three studies by Looney and others, Whitby and others and Rooks and others, has shown that subdural haematomas occur much more frequently at birth than was previously recognised. The Rooks paper indicated that 46 per cent of neonates had subdural haematomas seen by MRI within 72 hours of delivery and that subdural haematomas were seen after all modes of delivery, including caesarean section. Most of these subdural haematomas had resolved within one month and all had resolved by three months. In his oral evidence Mr Richards made this observation:

Dr Anslow and Mr Richards were unable to think of any medical or scientific reason why birth related acute subdural haematomas should evolve differently from other acute subdural haematomas. As Mr Richards put it:

Thus, the expert opinion put before me has refused to rule out the possibility that a chronic subdural haematoma found in an baby older than three months might be attributable to birth, but the preponderance of expert opinion was that it was unlikely. Professor Fleming concluded that the presence of bilateral, large and apparently expanding subdural haematomas, accompanied by or leading to a rapid increase in head circumference around three months, is a very unlikely consequence of subdural haematomas occurring at birth.

In many cases involving non-accidental head injury much attention is devoted to the question of whether a subdural haematoma is caused by a shake or by a shake plus impact. No party has introduced this element into this inquiry. Some evidence was given as to the degree of force required to cause a subdural haematoma. The consensus of opinion amongst the experts was that, although the minimum force required to cause an acute subdural is not known, clinical experience would suggest that accidental events have to be at a level higher than encountered in normal everyday life and would normally only be expected as a result of accidents memorable to parents.

A central question that has arisen in this case is whether the extra-axial spaces, that is to say the subarachnoid space, in J's head were enlarged and, if so, whether this was a cause or consequence of the subdural haematoma. I shall consider this issue in more detail below.

If the extra-axial spaces were enlarged before any subdural bleeding, a condition known variously as craniocerebral disproportion or benign external hydrocephalus, it is suggested that the degree of force required to cause subdural haematoma might be less than would be needed to cause subdural haematoma in the absence of such enlargement. Mr Carter, the neurosurgeon who carried out the two operations to drain J's subdural collections, stated that the presence of enlarged extra-axial fluid spaces is known to be a risk factor for spontaneous subdural bleeding or subdural bleeding after minimal trauma.

Cross-examined by Mr Storey QC on behalf of the mother, Mr Carter said that the possible explanations for this were that the friable blood vessels running across the area between the skull and the brain had further to travel and/or that movement of the brain in a large scale might cause additional forces which could fracture the vessels and/or that there are fluid currents set up by the movements which could add to the forces on the vessels. Cross-examined by Mr Ekaney QC on behalf of the father Mr Richards accepted that the increased extra-axial spaces may lead to increased risk of bleeding. Whilst there was no proof of this it was, he said, found that it was proposed in the medical literature. In his report Professor Fleming stated that the presence of a wider than normal subarachnoid space is relatively common and has been reported as leading to an increased risk of haemorrhage occurring in the presence of a relatively minor injury. Given that the condition of enlargement of the subarachnoid space is relatively common, whilst that of subdural haemorrhage is much less common, the consensus of professional opinion, according to Professor Fleming, has been that the level of trauma required to produce subdural haematomas in the presence of an enlarged subarachnoid space, whilst less than would be required in the absence of such enlargement, is still unlikely to occur as a result of normal or even relatively robust handling.

Listening to the expert evidence about subdural haematomas in this case, I have been struck by the measured way in which all of the experts expressed their opinions. Manifestly there has been considerably development in medical knowledge in this area in the last decade. The most striking example is the research into birth induced subdurals. There is also greater understanding about the presence and behaviour of venous membranes within chronic subdural haematomas and the propensity for the veins in the membranes to fracture, causing re-bleeding to occur within chronic collections. In addition, Mr Richards and Dr Anslow confirmed that subdural haematomas can cross from one hemisphere of the cranium into the other. Indeed, Dr Anslow demonstrated from the MRI that this appeared to have occurred in J's case. In surgery both Mr Carter and Mr Richards have seen perforations in the falx, that is to say the fold in the dura that descends into the inter-hemispheric fissure, which would permit communication between the potential subdural spaces in the convextus above the two hemispheres. Given these developments, it was not surprising to hear Dr Anslow observe that there has been an evolution of understanding about subdural haematomas over the last few years and that he is sure there is more to be discovered. Mr Richards observed frankly of this area:

The retina is a structure of multi-layered light sensitive tissue at the back of the eye that plays a vital role in the process of vision. Bleeding in the retina is known as retinal haemorrhage. It is characterised and described by reference to whether it is unilateral or bilateral, that is to say in one or both eyes, whether it is multi-layered or limited to only one or a few of the layers of the retina and whether it extends to the periphery of the retina or is confined to the central area known as the posterior pole and peri-papillary areas.

There are various causes for retinal haemorrhage, as explained by Miss Adams in the appendix to her report. The mechanism suggested for the pathogenesis of retinal haemorrhage, excluding direct injury to the eyes (which was not reported in this case), includes raised intravascular pressure, which may be produced by raised intracranial pressure, increased central venous pressure, which can occur with intrathoracic pressure due to impaired venous return to the heart, such as may occur in chest compression or strangulation, and an increase in central retinal vein pressure, such as may occur in certain medical conditions, or interaction with the vitreous jelly on the retina at its points of attachment to the retina caused by the effects of acceleration/deceleration during shaking. In this case the possible causes of the retinal haemorrhages were raised intracranial pressure and trauma.

In her report Miss Adams referred to research carried out by the Welsh Child Protection System Review Group which considered inter alia the distinguishing characteristics of retinal haemorrhages found in abusive head injury compared to accidental head injury. Miss Adams' reported that the research concluded:

It is important to note that, whilst the particular appearance of retinal haemorrhages may be associated with abusive or non-abusive head trauma, the appearance of the retinal haemorrhages is not diagnostic of a particular cause. Thus, for example, although retinal haemorrhages attributable to a non-abusive head injury are more likely to be non-extensive and restricted to the posterior pole, it does not follow that retinal haemorrhages that are so restricted are necessarily non-abusive in origin. Indeed, the Welsh study found that in 37 per cent of cases where the retinal haemorrhages were abusive in origin they were restricted to the posterior pole and peri-papillary areas.

Miss Adams advised that it is well recognised that retinal haemorrhages occur at birth. Research shows, however, that in most cases such bleeding clears within six weeks. In her oral evidence Miss Adams explained that the timing of retinal haemorrhages was based mainly on the study of birth related haemorrhages which were detected and followed. The study of birth related haemorrhages indicated that superficial retinal haemorrhage cleared within between one day and one week, but deeper retinal haemorrhages were all clear within four weeks, that pre-retinal haemorrhage can take three to four months to clear and that vitreous bleeding can take up to a year to disappear.

I turn to the specific medical evidence in this case.

The measurements of J's head circumference were as follows:

In his evidence Mr Richards emphasised that the measurement of head circumference contains a subjective element, a phenomenon he is able to demonstrate with medical students who will typically produce a range of measurements of the same patient's head circumference. Nonetheless, in this case the middle three measurements, that is to say on 24 November, 17 January and 24 January, were taken by the same very experienced health visitor BF, who described her practice of taking such measurements using a special measure. I find therefore that I can have a degree of confidence about the accuracy of the measurements in this case.

It was claimed by the parents in the course of these proceedings, although withdrawn by the mother in cross-examination by Mr Sharp QC on behalf of the local authority, that J had been noted by medical staff at birth to have a large head. There is no record of such an observation. What was noted was that he had slightly wider sutures between the skull bones, which in the neonate are separate before fusing as the infant grows. In his report Professor Fleming stated that what he described as the "wide metopic suture" observed in J's skull after birth was one of the normal lines of junction between the bones at the front of the growing skull and, in the absence of any other signs of abnormalities of the skull bones, or of raised intracranial pressure, minor variations in the size of the sutures were of no clinical significance. The widened sutures were clearly visible in the video recording taken a few days after birth which was shown to Professor Fleming in the course of his evidence. At one point Professor Fleming pointed out in the video how the parietal bone slightly protruded above the frontal bones, an appearance which he described as entirely normal. The fact that the frontal bones were at a marginally lower level indicated, according to Professor Fleming in cross-examination by Mr Sharp, that there was nothing imposing upward pressure on the skull bones. Although Professor Fleming was unable to say conclusively from the video whether or not there was benign external hydrocephalus present immediately after birth, there was no evidence that it was present and the slight protrusion of the rear bone of the skull was an indication that it was not present. Looking at the video recording Professor Fleming concluded that at this point, that is to say a few days after discharge from hospital, J's head looked entirely normal.

The latter part of the video recording was taken when J was about 10 weeks old, that is to say at the end of January or beginning of February. Professor Fleming observed that it clearly showed that the frontal part of the skull was bulging, the head was proportionately larger than in the earlier part of the video. Cross-examined by Mr Sharp he said that whilst babies' heads do change the change in this case was "quite marked". In other words, the video confirmed the significant growth in J's head between birth and shortly before his admission to hospital in February. Mr Richards described this growth, crossing two centile lines, as pathological.

In his report Dr Anslow gave a succinct and clear account of the radiological imaging carried out in this case and incorporated reproductions of the scans within the text of his report. He said that the CT scans carried out on 25 February in Gloucester demonstrated on the right side of J's head widened subarachnoid spaces and on the left a large low density subdural haematoma, but no fresh blood. He said in oral evidence that on some of the CT scans he could detect a faint line on the left side indicating the border between the extra-axial spaces and the subdural haematomas. In one of the scans he noted that a large low density subdural was now visible posteriorly. He then analysed the MRI carried out the following day. These images clearly displayed the left side subdural haematomas, which had been of single density on the CT scans, were of two distinct signals separated by a membrane. One image showed the deep subdural haematoma crossing the mid-line between the left and right hemispheres, which Dr Anslow described as an unusual finding. In another image what Dr Anslow described as a contour change over the left subdural haematoma indicated it was of longstanding. In another image the two collections and the extra-axial fluid were all easily discernible in the intracranial spaces. Dr Anslow noted that the superficial subdural haematomas continued over the surface of the skull to the posterior parafalcine region. Further images demonstrate what Dr Anslow describes as tiny flecks of sub-acute blood, one of the arachnoid surfaces of the subdural haematoma in the occipital area, another relating to a bridging vein traversing the subdural haematomas and a third closely related to the subdural haematomas in the parafalcine region. A final image inserted into Dr Anslow's report shows what he describes as the spreading nature of the subdurals and the complex way they relate to each other. The chronic subdurals are two distinct collections, confirmed surgically, separated by a membrane. In Dr Anslow's opinion the radiology indicates that the collections are probably more than a month old.

As demonstrated in his analysis and confirmed in his oral evidence, the tiny effects of sub-acute blood are all associated with the membrane or the bridging veins. Dr Anslow advised that such bleeds may occur spontaneously or after trivial trauma and do not necessarily indicate a more recent severe or excessive trauma. Dr Anslow was of the view that there had been one traumatic event which caused the acute subdural, which over time evolved into a chronic subdural characterised by membranes from which small re-bleeds occurred, causing the collection to persist and enlarge and contain flecks of sub-acute blood. Clearly, however, it is possible that the acute blood is attributable to a second episode of trauma. The presence of acute blood in the subdural space is consistent with both a re-bleed and a further episode of trauma.

When did the first traumatic episode leading to the chronic subdural occur? One possibility is that it happened after birth. Dr Anslow thought that it might have been birth or some other traumatic event for which there is no clinical history, meaning, as I understand it, in the case of a non-ambulant baby, an act of non-accidental injury. He was not able to say which was more likely, but agreed with Mr Storey, on behalf of the mother, that he could not say that there was no connection between the chronic subdurals and birth. Mr Richards was also not prepared to rule out the possibility that J sustained subdurals at birth that developed into a chronic subdural. His clear view, however, was the probable cause of J’s chronic subdural was that it followed the development of an acute subdural caused by an incident of inflicted head injury that included a shaking episode.

On behalf of the local authority, Mr Sharp submits that an origin related to birth must be rejected on the evidence. He contends that the chance of an elective caesarean section in a child of 37 weeks with no abnormalities during pregnancy giving rise to a subdural are very low and the chances that this same child should be the first ever seen example of a chronic subdural haematoma caused at birth are, to use his word, infinitesimal.

With respect to Mr Sharp, I consider that he puts his case too high. I remind myself of the wise words of Mr Richards and Dr Anslow about the development of knowledge about subdurals in recent years and their awareness that there is much still to learn. Mr Sharp rightly stresses that I must make findings on the basis of the evidence. On the basis of the expert evidence in this case I find that it is unlikely that the chronic subdurals originated at birth, but not impossible. Manifestly, however, the most likely explanation based on the medical evidence is that the subdural haematomas originated in an episode of inflicted injury that included an element of shaking.

Mr Ekaney submits on behalf of the father that the court should give due weight to the likelihood of craniocerebral disproportion/benign external hydrocephalus having contributed to the subdural haematomas. Were it the case that J had a large head at birth there might be some support for the theory that he was born with craniocerebral disproportion, which in turn might have made him more vulnerable to subdural haematomas. Cross-examined by Mr Storey, Mr Carter said that he thought it likely that J had started with enlarged extra-axial spaces before any other factor had intervened. On the other hand, Dr Anslow, in his written contribution to the joint expert discussion, said he thought that the large skull was a consequence of the subdural haematomas, not the other way round. Mr Richards observed that whereas the sub-arachnoid spaces in J were prominent, they were not significantly enlarged given his age. Although he cannot completely dismiss the possibility that craniocerebral disproportion contributed to the development of J's subdural haematomas, Mr Richards thought it on balance unlikely. He agreed with Mr Sharp that the craniocerebral disproportion was attributable to the subdural haematomas. That disproportion had continued after the aspiration procedures because, as Mr Richards reminded us, "nature abhors a vacuum" so that when the fluid was extracted from the subdural collections the sub-arachnoid spaces with the cerebral spinal fluid expanded in size. The preponderance of medical opinion is therefore that it was the subdural haematomas that caused the enlargement of the extra-axial spaces and not vice versa.

Thus the expert evidence points to an episode of inflicted head injury.

When did it occur? In this case the child's head size increased significantly at some point between the end of the second week after birth, 24 November 2010, and the 10^th week, 17 January. This period coincides with the probable dating of the oldest subdurals. Mr Richards has advised that it takes at least two weeks for the presence of acute subdural blood in the subdural space to be capable of enlarging the head. He therefore suggests that the acute event that led to the development of the chronic subdural haematoma was "at least two weeks before the eight week check" which was in fact conducted on 24 January at 11 weeks. This places the acute event around or before 10 January. Professor Fleming suggests that it seems likely that the rising in circumference observed by the health visitor on 17 January was a consequence of the bilateral subdural haematomas and that, if this is the case, "it is highly likely that the haemorrhage responsible for this rise in head circumference would have occurred at least two or three weeks prior to this time". This would place the acute event at a point prior to 3 January. As already noted, Dr Anslow advises that the radiology suggests that the subdural haematomas are probably more than a month old at 25 to 26 February, which suggests an event before the end of January.

Having considered all the medical evidence I find that if the subdural haematomas originated in an act of inflicted trauma it occurred at some point between the end of November and 10 January.

On 28 February J's eyes were examined by a senior registrar who observed bilateral pre-retinal haemorrhage and vitreous haemorrhage on the left side. The registrar drew a diagram illustrating the location of the haemorrhages on the medical notes. On 2 March J's eyes were reviewed by Miss Williams, a Consultant Paediatric Ophthalmologist. She observed bilateral retinal haemorrhages and three retinal haemorrhages at the posterior pole of the right side.

Miss Williams arranged for pictures of the retinas to be taken and copies were supplied on a DVD to Miss Adams for the purpose of her expert assessment. Miss Adams reported as follows:

Miss Adams was asked to comment on the dating of the haemorrhages. She reported that one would usually anticipate superficial retinal bleeding clearing within one week of causation and deeper intra-retinal haemorrhage clearing within four weeks. Her view of the pictures did not identify any superficial retinal haemorrhages, but she did observe retinal haemorrhages. She therefore concluded that the haemorrhages were sustained at some point within four weeks of the taking of the pictures on 2 March. She was therefore able to rule out the possibility that J's retinal haemorrhages had been caused at birth.

Miss Adams' overview of the pictures confirmed the bleeding was all within one zone of the retina, which she would not regard as extensive in area. There is no disc swelling, nor any superficial retinal haemorrhage. She confirms that there was retinal haemorrhage obscuring the underlying retinal detail. In addition, there appeared to be some intra-retinal haemorrhage.

Miss Adams concluded that, on the basis of the appearance of the retinal bleeding alone, it was difficult to be definitive about etiology. In this case she observed that the bleeding could have been due to non-accidental injury or could have developed secondary to raised intracranial pressure caused by the presence of large subdural haematomas in a child with enlarged extra-axial spaces. As to intracranial pressure, Miss Adams commented:

At the experts’ telephone conference, Miss Adams and Mr Richards debated the issue of whether the retinal haemorrhages could have been caused by raised intracranial pressure. Miss Adams reiterated her view that raised intracranial pressure was a possible cause and cited in particular the sunsetting observed in the eyes. Mr Richards expressed the view that sunsetting was very specific to hydrocephalus, where there is a downward pressure on the upper brain stem from the third ventricle and was not seen with generalised raised intracranial pressure. Miss Adams rightly deferred to his neurological expertise on this point. Mr Richards concluded that it was not a picture of seriously significant raised intracranial pressure, either acute or chronic, but rather a gradual low level increase in intracranial volume that was the concern in this case. Whilst he accepted Miss Adams' point that raised pressure could cause retinal haemorrhages he was not convinced in this case that there was significantly raised intracranial pressure of the type that would cause such haemorrhages, although he could not rule out this explanation completely.

In oral evidence Miss Adams conceded to Mr Sharp that a slow and steady rise in intracranial pressure would not cause retinal haemorrhage. It would require an acute rise in such pressure to cause such bleeding. She accepted that chronic subdural haematomas would not lead to raised intracranial pressure of a type which would lead to retinal haemorrhage "unless it gets to a point where there is a sudden spike". However, she adhered to the view that raised intracranial pressure was a possible explanation for J's retinal haemorrhages and in answer to Mr Storey identified a number of signs of raised intracranial pressure in this case which supported that view, not merely the sunsetting of the irises but also the bulging anterior fontanelle, two episodes of surgical aspiration being required and the separation of the cranio-cerebral sutures.

In his oral evidence, cross-examined by Mr Sharp, Mr Richards accepted that there must have been raised intracranial pressure sufficient to expand the skull, but said there did not seem to be any episode which would have indicated a sudden increase in intracranial pressure. There was, for example, no episode of loss of consciousness of the sort one would expect to be sufficient to generate raised intracranial pressure leading to retinal haemorrhages. He agreed with Mr Sharp that the note of the registrar's examination on admission to hospital "alert, active, playful" was not a description of child who was neurologically compromised.

In his oral evidence, Professor Fleming, who alone of the expert witnesses was able to view the video recording of J, pointed out in the latter part of the recording the presence of sunsetting in J's eyes. Professor Fleming confirmed that this sign is a common feature of hydrocephalus and may be a sign of raised intracranial pressure, but the absence in J of any swelling of the optic disc, a common and relatively early sign of raised intracranial pressure, at the time of the admission to the Frenchay Hospital makes the presence of significantly raised intracranial pressure over a prolonged time period before the hospital admission less likely. He accepted that raised intracranial pressure was a possible cause of the retinal haemorrhages sustained by J, but added that it was unusual to have retinal haemorrhages attributable to raised pressure in small children. In his experience with children with raised intracranial pressure being watched in intensive care, retinal haemorrhages are unusual. His view therefore was that it was compatible with raised intracranial pressure, but not common.

On the basis of the expert evidence I find that it is possible that the retinal haemorrhages, described by Miss Adams as mild, were caused by raised intracranial pressure, but that it is more likely that they were attributable to inflicted trauma.

The third component in the so-called triad which together constitute a strong pointer for non-accidental head injury is encephalopathy. In this case there is uncertainty as to whether J was at any point encephalopathic. Mr Carter thought that the reported history of several incidents of unresponsiveness in the preceding weeks suggested episodes of encephalopathy, which would be likely to appear with subdural haematoma occurring as a result of a shaking injury but "far less likely in the event of a haemorrhage occurring from minimal trauma associated with enlarged extra-axial spaces". He added however that J had had several episodes of unexplained reduction in responsiveness whilst on the ward not associated with any obvious injury or therapeutic intervention. He was uncertain as to the nature of these events and commented that "given that they sound similar to the events that occurred previously, it is difficult to ascribe these prior events as likely to have been due to trauma."

In his report Mr Richards expressed the view that, given the very rapid recovery from the episodes viewed in hospital and reported by the parents, it was probable that the episodes did not represent encephalopathy. He thought it more likely that they represented seizure activity or reflux. On the balance of probabilities he thought that seizures were the most likely cause but he could not be certain. The cause of such seizures could be spontaneous following an episode of trauma or as a result of blood in the subdural spaces irritating the cerebral cortex. In cross-examination Mr Storey showed Mr Richards the father's explanation of the first "funny turn" in his statement for these proceedings in which the father describes how he thought J had died. He agreed with Mr Storey that the description of the baby was consistent with what might happen following a shaking episode.

In oral evidence Professor Fleming said that J's behaviour, as described by the mother following the funny turns, was non-specific. That behaviour would be consistent with, though not diagnostic of, a traumatic episode that might account for the intracranial bleeding.

The medical evidence therefore points to the following conclusions:

As stated above the court does not consider the medical evidence in isolation. The evidence in these cases is not evaluated and assessed in separate compartments. I must therefore assess the medical evidence in the light of the other evidence, in particular about life in the parents' family leading up to and after J's birth and the character and personalities of the family members. It is to those matters that I now turn.

I had the benefit of hearing oral evidence from several members of J's family, namely his parents, both grandmothers, his maternal grandfather, maternal aunt and his half-sister. Of these I focus on the evidence of his parents, paternal grandmother and half-sister.

I start with his half-sister L. She was a most impressive witness, only 17 years of age but remarkably mature, sensible, intelligent, measured and balanced. She is a credit to her mother. Although manifestly supporting her mother, she was able to acknowledge positive features of J's father's conduct, for example his treatment of her other brother X. I accept her evidence without hesitation. It was her account that the father could be very charming and then switch suddenly. Small things could turn him quickly and unpredictably.

In contrast to L, I found the paternal grandmother a most unsatisfactory witness. She was manifestly biased in favour of the father and her evidence was littered by examples of minimising the father's conduct. In every virtually every answer that she gave during a persistent but fair cross-examination by Mr Sharp she adhered to this position. I do not consider her to be a reliable witness in many respects.

The principal witnesses from the family of course were the parents themselves. Neither of them was a satisfactory witness and the court is therefore faced with a real difficulty as to how much weight should be attached to their testimony. I agree with Mr Ekaney's submission that the mother has a complex personality which the father may have found puzzling. Even on the father's own case, however, she has been the victim of domestic abuse over a period of time. The mother was in a state of considerable distress throughout the hearing and this extended to her oral evidence. She was highly anxious, timid and at times confused. She plainly found the process of giving evidence very difficult. She often failed to answer the question and retreated into repeating a mantra "I thought it was medical" meaning that she thought there was a medical explanation for J's injuries. It was only right at the end of her evidence, in answer to the very last question from Mr Storey in re-examination, that she expressed the view that the injuries must have been inflicted by the father. I found much of her testimony self-serving and unconvincing. Despite her ostensible vulnerability I find that she is capable of a degree of calculation in the way she has given evidence.

One example of her unreliability as a witness is that she has given conflicting accounts of when she was told about the funny turns. Her oral evidence on this was hard to follow. In the police interviews she asserted that she had been told about the fits on the following day. She maintained in cross-examination by Mr Ekaney that she had been referring to the bruise during the police interviews, that is to say the bruise sustained on 8 January, not the fits. In oral evidence she maintained that her account to the police was based on what the father had said to her in hospital. This cannot be right and I am driven to conclude that she is seeking to distance herself from the incidents and from culpability for failing to seek medical attention.

Another example of her capacity to mislead is the so-called journal. This document, running to over 100 pages of close written manuscript, purports to be a contemporaneous record of the period from the time she discovered she was pregnant in early 2010, through the birth of J, his admission to hospital and placement into care, up to the date on which she obtained injunctions against the father. The main themes of the document, almost to the exclusion of anything else, are her devotion to her children and the father's violent and aggressive character. It contains a number of poems devoted to these subjects.

In a characteristically penetrating cross-examination Mr Sharp demonstrated that the journal was, to say the least, not wholly reliable. There are entries that show that it was, at least in part, not a contemporaneous document but rather constructed at a later date. A considerable number of the dates are changed. There is much use of Tippex. A variety of different ink is used. Some text has obviously been added later. Some loose pages, apparently from other notebooks, have been stuck into the journal. Asked to explain why she had not produced the journal at an earlier stage, for example in support of her application for an injunction in May 2011, the mother said that it was a private document not intended to be shared with others. This cannot be true because various people, for example her sister, are identified by name and then by their relationship to her in brackets in a fashion that would obviously be unnecessary in a private document. In at least one instance the journal describes an incident that never took place, namely a visit by the health visitor which is said to have happened on 17 February. The entry appears in the middle of what appears to be a continuous passage from 5 to 19 February. If the 17 February entry is wholly fictitious that calls into question the veracity of the whole of those pages.

In cross-examination, Mr Sharp put it to the mother that the journal was wholly fictitious. In his final submissions, however, he rowed back from that position. He invited the court to accept that to have constructed the whole volume predicated a degree of sophisticated planning which seems improbable given the mother's abilities, and that therefore the journal should be regarded, at its core, as a genuine document which has nevertheless been added to at a later stage. This interpretation leaves a number of questions unanswered, not least why the mother failed to produce it until a very late stage in the proceedings, but on balance I have concluded that Mr Sharp's ultimate analysis is correct. I do not think that the mother has the capacity to construct a 100 plus page document as a complete work of fiction. I find that some entries were written contemporaneously and others added later, but I find it impossible to identify with confidence those passages that are original and those that are later embellishments. In those circumstances I do not think it fair or safe to rely on the document as evidence of specific events.

It does, however, demonstrate a number of general points. First, in so far as she has made substantial alterations and additions to the document it shows, as I have said, that she is capable of a degree of calculation that calls into question her reliability as a witness. I remind myself of the Lucas principle, but nonetheless the production of the journal leads me to be cautious about accepting her evidence in other respects. Secondly, on the other hand, having heard her give evidence, I am satisfied that by and large it is the mother's voice speaking in the book. By that I mean that in describing her devotion to her children and her fear and anguish at the father's aggression she is describing her genuine feelings. Thirdly, however, I find that this document is powerful evidence that, notwithstanding her devotion to her children, she was incapable of protecting them from the emotional harm caused by the father's aggression.

In addition the court has seen a large number of text messages sent by the mother to the father. They appear to be one half of a daily ongoing conversation between the parties. It is important to note, however, that only limited texts from the father have been produced so only one side of the conversation is available for much of the time. Nevertheless, unlike the journal it is not suggested the texts have been fabricated after the event and I accept that they amount to a contemporaneous record. In many instances the mother makes reference to the father's aggressive and, on occasions, violent behaviour. In contrast to the journal I find that these texts do provide reliable evidence of the level of conflict between the parents and the degree of violence suffered by the mother to which the children living in the home were then exposed.

If the mother was a timid and anxious witness, the father was almost the polar opposite. He had a confident, at times arrogant, manner and answered questions with little hesitation. Despite that his evidence was just as unsatisfactory as the mother's. On his behalf Mr Ekaney carefully laid the ground for a position in which it was accepted that the father had been loud, aggressive, intimidating, violent in the sense of pushing her, controlling, frightening and obsessive about the house being clean and tidy. Despite this the father maintained a position in his oral evidence of defiance. Like the mother he too had a mantra which he repeated at several points during his evidence when confronted with some unavoidable evidence about his behaviour "I'm not proud of myself". In my judgment, however, he failed to show genuine remorse for his behaviour and repeatedly minimised the extent of his violence.

As with the mother, a document was produced that shed revealing light on the father's true character. That document was a recording which L made on her Blackberry. She transcribed the recording and played it in the course of the evidence. In it the father is heard shouting in an extremely loud, aggressive and frightening manner at the mother on two occasions. The transcript reads as follows:

The second part reads:

It is not often that the court is presented with direct evidence of domestic abuse of this sort. L said that this was typical of his behaviour and, to be fair to the father, he admitted that he had behaved in this manner on other occasions, but he stopped short of admitting that he was violent to the mother to the extent claimed by the mother and L.

There are a number of specific incidents of violence alleged against the father which he denies. In September 2010, when the mother was about 30 weeks pregnant, an argument took place about the failure of the mother and L to clean a large paddling pool in the garden. In the course of this argument the mother sustained bruising to her side. L's evidence is that she witnessed the father push the mother, causing her to fall over the coffee table. When L challenged the father about this at the time he replied "You saw, she tripped". The mother sought medical attention for the bruising, claiming she had fallen over. Both parents maintained this account in their first police interviews in March 2011. It was only during the mother's further police interviews in May 2011, when shown L's account, that she accepted that the father had pushed her. The father maintains his version of the incident, but I accept L's account supported, as it now is, by the mother herself.

At Christmas 2010, when J was six weeks old, the mother wished to have all her children together on Christmas Eve. After an argument the father insisted on keeping not only J but also it seems all the family Christmas presents in his property overnight. He admitted in evidence that he had done this and again said that he was not proud of himself, but to my mind showed no real understanding of the emotional cruelty of this conduct.

Over the Christmas period, after another argument with the mother, the father got involved in an altercation with his brother who, according to the mother, tried to take his car keys away fearing that he was not in a fit state to drive because of his temper.

In early January 2011, it is alleged that the father assaulted the mother by punching her in the face, causing a black eye. In a text message sent the following day the mother refers to having not answered the door to the health visitor because she did not want her black eye to be seen, an account that she repeated in her oral evidence. The health visitor confirmed that when she called on 5 January no-one answered the door. In his evidence the father maintains that the mother inflicted the black eye on herself by hitting her face with her hands. I find that an absurd suggestion and reject it without hesitation. Equally, I reject the suggestion that the explanation for the text messages sent by the mother to him, in which she referred to an alleged act of violence he had perpetrated on her, was that the allegations were false.

A few days later on 30 January, a very alarming incident occurred. It is accepted by the father that following an argument L accidentally sent a text message, complaining about his conduct, to him instead of its intended recipient, one of her friends. The father went round to the mother's property to have it out with her. He accepts that he got angry and shouted at her and that he pushed her. She maintains that he grabbed her by the neck and pinned her against the wall in the hall, causing a crack in the wall. At the same time she says that he pushed the mother to the floor, causing her to fall. This account is corroborated by the mother. In cross-examination the father came fairly close to accepting this account, which he had previously denied, saying that he had grabbed or taken hold of L by the collar bone. I unhesitatingly accept L's account. It follows therefore that this father has assaulted a girl then aged 16 and continues to deny the extent of his actions.

I have referred to the crack on the wall caused in this incident. On another occasion the father punched a hole in a wall in the mother's property. On yet a further occasion another hole was caused in another wall, this time in the boys' bedroom, caused, according to the father, when he kicked one of X's toys so hard against the wall that it caused this damage.

The incidents described above, and others in the evidence to which I find it unnecessary to refer in any detail, illustrate and, in my judgment, prove that at this point, (which is of course the time when on the medical evidence the intracranial bleeding, if caused traumatically, was likely to have occurred), the father was a man who had great difficulty controlling his temper, was regularly aggressive and was on occasions violent to vulnerable members of his family.

The context of the father's behaviour is found in other evidence, in particular concerning his mental state. No psychiatric evidence is available, but it is accepted that in October, that is to say after the paddling pool incident, he was found on a bridge apparently contemplating suicide and subsequently sought medical advice on a number of occasions when he complained of feeling stressed and angry and identified a number of problems, including his debts and after J was born the fact that he was struggling to cope with the baby. There was talk of a referral to anger management, but nothing came of this. An indication of the extent of his problems is that his own mother visited his GP to express concern about his bouts of anger which were increasing in severity.

In contrast there is no evidence that the mother has ever demonstrated any propensity to violence. On the contrary, the evidence is that she lives for her children. The evidence of L on this point was particularly persuasive. The father himself does not suggest that the mother could have injured J. He said he could not contemplate any situation in which this could have occurred. The only person who suggested it in the course of the hearing was Mr Sharp in the course of cross-examination when, very properly, testing the mother's evidence and exploring all possible explanations for J's injuries and the reasons for her actions, but having asked the question Mr Sharp drew back in final submissions. On behalf of the local authority he indicated that the authority was no longer seeking a finding that the mother was the perpetrator or a possible perpetrator of J's injuries.

Against this background I turn to consider the evidence about the three funny turns which J is said to have experienced.

There is a conflict of evidence as to when the incidents occurred. In the initial accounts given to the hospital the first incident was said to have occurred when J was between five and seven weeks old, i.e. between 16 and 30 December, the second incident when he was about 10 weeks, i.e. about 19 January, and the third at 11 weeks, i.e. about 26 January. In his first statement in these proceedings the father maintains that the first incident occurred on 21 January, the second 12 February and the third on 19 February. Subsequently, however, the mother has asserted that the third instance, which she is said to have witnessed, occurred on 11 February.

The father's description of the first incident in his statement in these proceedings reads as follows:

This account is partially corroborated by the paternal grandmother, who recounts in her statement to the police how the father had come running into her room and was quite panicked because J appeared to be limp and not breathing. She immediately thought that J had choked on some milk, so she rubbed J's back and he seemed to be okay. The father described how J had gone rigid, but she thought it was a feeding or wind issue and nothing more serious. In her oral evidence she said she did not think the baby had stopped breathing, it was just that the father was panicking. In the course of his police interview the father described how J had screamed when he woke. The paternal grandmother, who described herself as a light sleeper, denied hearing any screaming.

There is a conflict of evidence as to the course of J's recovery after this incident. In her first police interview in March the mother described J, on returning to her care after the first incident, as "very grouchy, crying, screaming" and said she thought "My God, shall I get the doctor? What's wrong with him?" She said that "He wouldn't feed. He was just really screaming the house down, which he never does." Subsequently, however, he had calmed down, gone to sleep and managed to feed. Mr Sharp reminds me that the account of a child unsettled in this fashion is consistent with the evidence given by Professor Fleming as to the typical behaviour demonstrated by a child following a traumatic incident that causes intracranial bleeding. In her first statement in these proceedings, however, the mother said that J "seemed his usual self" when he returned to her care on the following day after the first episode was said to have occurred.

The father's account of the second incident is as follows in his statement filed in these proceedings:

Significantly on this occasion, however, the paternal grandmother does not recall this incident at all and in oral evidence she said that she would have remembered it if the father had been panicky again in the night.

The third incident took place in the presence of both parents and the paternal grandmother. Whilst the mother was feeding J he became momentarily lethargic and the father jumped up and said that this was what had happened on the previous occasion. It seems clear, however, that J quickly recovered on this occasion, there was no arching of the back, no rigid spells, no lifelessness. This third incident therefore seems to be of a different character from the earlier two as described by the father.

An important feature of this case on which the local authority relies is that no medical attention was sought for J on either occasion and, further, neither parent mentioned any of these incidents to the health visitor at any stage or to the GP at the check up on 24 January. On the father's own evidence these incidents were frightening. He thought the baby had died. In such circumstances it is extraordinary that he did not seek medical assistance. His case is that he was reassured by his mother. It is just conceivable that an inexperienced father might be reassured by his own mother on the first occasion. That cannot, however, explain his inaction on the second incident, especially if, as I find, the paternal grandmother is telling the truth on this point when she says that she was not present. After two incidents any reasonable father, observing the incidents described by this father, would seek medical attention. Instead the evidence is that he discouraged the mother from going to the doctor, against her better judgment. Similarly, his explanation for failing to mention the first episode at the GP check on 24 January, that he arrived late and thought the mother had mentioned it, is to my mind incomprehensible. The clear inference is that he had something to hide.

I found the father's account of the first two incidents in his written statement in these proceedings and his oral evidence to be implausible. It is manifestly clear, in my judgment, that he has not told the whole truth about these incidents. I find that these two incidents have occurred, that elements of the father's story are true, including that J had a spasm and then went limp, and that he thought that J had died, but there is, in my judgment, a crucial element missing from his account.

Although in his direct evidence to the court he has denied shaking the baby, there is evidence that on two occasions he came close to admitting that he had done so. First, in the course of one police interview he accepted that he may have shaken J in panic. Secondly, the evidence of the mother and the maternal grandfather is that in a conversation the father showed them how he might have done it. The maternal grandfather's account of this is particularly compelling:

As Mr Richards said in oral evidence, the sudden behaviour demonstrated by J, as described by the father in his statement in the passage I have read concerning the first incident, is consistent with what would be expected of a child in an acute phase after a shaking injury.

Having regard to all the evidence, including the medical evidence as to subdural haematomas and retinal haemorrhages, the evidence of the father's explosive temper and his propensity to violence to the mother and on one occasion to L, and taking account of the possibility that injuries may have unknown aetiology, I find that the reason for J's collapse on the two occasions to which the father has referred was that he had been assaulted by the father in a fashion that included an element of shaking.

I find that the first assault occurred on a date around the turn of the year, that is to say in the last days of December or the first week in January, and the second in the first week in February. The father looked after J overnight on several occasions during these periods. I cannot make a finding as to the precise dates, although I note that J sustained a bruise on 8 January whilst staying with the father and, according to the mother, the father discouraged her from obtaining medical advice about that. It is possible that the first incident occurred on that occasion.

As to the precise circumstances in which the assaults occurred, I cannot make more detailed findings because the only person present, the father, has not told the whole truth. Mr Ekaney submits that, if this is a shaking injury, the trauma is at the mild end of the spectrum and the court can safely find that it occurred due to a momentary loss of control in circumstances of high pressure and anxiety for the father. Mr Sharp suggests to the court that the reason for the assault may have been because the father was obsessed with J finishing all his milk when, as all parents know, babies sometimes refuse to co-operate in this way. Both these submissions are made on conjecture. I cannot make any further finding as to the circumstances in which these incidents occurred. To do so would be to cross over into speculation. On the evidence, however, I am satisfied on the balance of probabilities that J was shaken by his father on two occasions.

It follows that I absolve the mother of any blame for inflicting injury on J. However, I do find that she culpably failed to protect her son in a number of ways. On her behalf Mr Storey warns me that there is tendency in some courts simply to make a finding of failure to protect without giving sufficient thought to the meaning of that phrase. In this case I have given very careful thought to how the mother has failed to protect her son and I make the following findings:

Firstly, knowing of the father's violent and ungovernable temper she exposed J to his care overnight on a regular basis. It is clear from her evidence that throughout the period from J's birth to his admission into hospital in February the father was undergoing a personal crisis in which his temper was deteriorating and he was demonstrating worrying signs of instability. Mr Storey sought to argue in final submissions that to find a mother who is herself a victim of domestic violence culpable of failing to protect her child would be unfair. In my judgment, it would be manifestly unfair to a child not to make findings in such circumstances. It is a basic obligation on any parent to protect their child from physical harm. I recognise that circumstances may make that very difficult and that in this case the unusual housing arrangements posed a particular problem. Nevertheless, this mother ought to have realised that J would be in danger. I find that she did realise that he would be in danger, but that she lacked the strength to take action to protect him, just as she failed to take steps to protect L following the assault on 13 January.

Secondly, the mother is also culpable for failing to protect all of her children from the emotional harm caused by being exposed to the father's ranting and explosive temper. It is well recognised nowadays that domestic violence causes emotional harm to children. The mother lacked the strength to protect her children from that harm. Her journal, whenever it was written, contains entries in which she does recognise that she has failed to protect her children in that respect.

Thirdly, I find that the mother failed to seek medical help after the two episodes. I acknowledge that it is her case that the father withheld from her the fact that J had seemed to stop breathing. Nevertheless, even on the more limited account given to her she plainly should have sought medical advice, particularly if J was not his usual self on the following day after the first incident, as she maintained in her first police interview. It is even more surprising that she failed to contact the doctor after the second episode. I note the exchange in the first police interview in which she said that she and the father had discussed whether they should tell someone about the fits, that the father had said that they need not do so, but she thought "it was something wrong". If this is correct the reason for her failing to inform medical staff about the episodes may have been because she was unable to act against the influence of the father. Whatever the reason, however, I find that her failure to seek medical advice was a culpable failure to protect.

Fourthly, I find that the mother failed to give any information about the "funny turns" to either the health visitor or the GP during routine visits. The first incident occurred before the health visitor's home visit on 17 January and the GP's check on the 24^th. The mother had plenty of opportunity on those occasions to pass on information about the incidents and, given the attention that was given on those occasions to the head circumference, it must have been in her mind to do so. Why did she fail to do it? Was it because she suspected that father had injured the child or was it because she was concerned not to upset the father? I cannot say. Whatever the reason the failure to mention the episodes to professionals when she had a clear opportunity to do so was yet another example of failure to protect her son.

Fifthly, the mother seriously misled the police, Social Services and the medical staff by failing to disclose the extent of the father's history of violence during the initial investigation after J's admission to hospital. In doing so she withheld relevant information and thereby jeopardised the safety of her children.

I realise that these findings against the mother have to be taken in the context of her weak character and the father's intimidating and threatening behaviour towards her. I also acknowledge that in many other respects she has been a good mother to L and to her son X with his significant disability. Those factors suggest that in due course it may be possible for her to care safely for her children again, but any assessment of her capacity must address the fact that she failed to protect her children from this man and, as a result, contributed to the harm that they have suffered.

I therefore find, on the balance of probabilities, that between 24 December and 2 February J sustained bilateral subdural haematomas and bilateral retinal haemorrhages as a result of inflicted trauma on two occasions in the form of assaults which included an element of shaking perpetrated by the father.

I have found that J and indeed L and X were exposed to domestic violence perpetrated by the father on the mother and upon L and, as a result, suffered emotional harm.

I have further found that the mother failed to protect J from this physical and emotional harm.

These are serious findings and will form the basis for future assessments of the parents to see what future role they should play in their son's life.

In conclusion, however, I record that despite my findings there are positive signs about the qualities of these parents. I have already referred a moment ago to the mother's qualities and it is right to note the positive comments made about the father, in particular, his tenderness towards X and J, acknowledged by the mother and by L. Although at one point he tried to persuade the mother to have an abortion and in a later text he said that he did not want J, I do not think it is established that he lacked empathy with his son. I accept Mr Ekaney's submission that J's birth was indeed a joyous event for everyone concerned, including the father. Thereafter, however, there was little joy in the family until the parties separated.

This man came to fatherhood later in life than many men and as he told the GP he struggled to cope. He was obsessed with a tidy house and could not tolerate the mess that invariably comes with looking after children. He was going through a crisis of his life. Some elements of that crisis are clear, others as yet are not. Clearly he is a man with a serious temper problem, but it may be that this will be responsive to treatment. If so, it is by no means impossible that he will have a role to play in the life of his son.