England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE FAMILY COURT AT MEDWAY
Deputy High Court Judge Clare Ambrose
ME19C01777

		Royal Courts of Justice Strand, London, WC2A 2LL
		3 June 2020

Covid-19 Protocol: This judgment was handed down remotely by circulation to the parties' representatives by email, release to BAILII and publication on the Courts and Tribunals Judiciary website. The date and time for hand-down is deemed to be at 10:30am on Wednesday, 3 June 2020.

Lord Justice Peter Jackson:

Introduction

1. At the end of a remote hearing on 28 May 2020 we informed the parties that this appeal would be allowed. This judgment contains my reasons for reaching that conclusion.
2. The appeal arises from care proceedings about a little boy, T, who was born in July 2019. At the age of around two weeks he was taken to hospital, where he was found to have bruising to his face and was admitted for tests. Four days later, on 12 August, there was an incident when the father was seen to be mishandling the baby in a hospital cubicle. The local authority obtained an emergency protection order and T was placed in foster care in mid-August. An application for a care order was issued and on 20 August an interim care order was made. After proceedings that became unfortunately protracted, a seven day remote final hearing took place in April/May 2020 before Ms Clare Ambrose, sitting as a Deputy High Court Judge. The local authority's case was that at the relevant date T was suffering and was at risk of suffering physical harm, emotional harm and neglect, the most significant allegation relating to the incident in the hospital.
3. In a judgment given on 1 May, the judge made detailed findings of fact but she found that those facts did not satisfy the threshold for intervention under s.31 Children Act 1989 (CA 1989) and she dismissed the proceedings. In a separate judgment given on 7 May, she refused the local authority's application for a holding interim care order under s.40 CA 1989, with the consequence that T would have had to be returned immediately to his parents unless they agreed to him being accommodated for a transitional period. She also refused the local authority's application for permission to appeal and for a stay, with the consequence that an urgent application for a stay was made. That was granted that evening by Lady Justice King and on 18 May I gave permission to appeal.
4. The outcome of the appeal is that the proceedings revive on the basis of the substituted threshold findings set out below and the interim care order that was previously in force is restored. The matter is remitted to the Family Court for a welfare decision and an early case management hearing will take place before the Designated Family Judge on 4 June.

The course of the proceedings

5. Before embarking on the substance of the matter, I note that the state of the proceedings up to the point of the hearing was very far from satisfactory. The original intention had been for there to be a single final hearing, but two 10-day listings had been vacated and the local authority, whose social worker had become ill, had not filed the necessary assessments. By 1 April, its position was nonetheless that T should not be placed with his parents but with an aunt under a special guardianship order. It recognised that it was not ready for the welfare stage of the proceedings and the hearing conducted by the judge was therefore limited to fact-finding and determination of the threshold. These difficulties in the preparations for the hearing were compounded by the need for the hearing to take place remotely during the lockdown.
6. It is to the credit of the parties and the judge that the hearing was able to be completed, but it need hardly be said that it is troubling that a child who had been taken into care at the age of 3 weeks was still waiting for a decision about his future 9 months later. The situation was also extremely difficult for the parents, who are on any view vulnerable individuals. It is clear that T has not been well-served by the proceedings so far, and it is important that every attempt is now made to make up for lost time.

The background

7. This is relatively simple and can be taken from the judge's first judgment, from which I remove identifying information:

"13. Turning to the factual background, the following findings are based on the evidence before me including the parents' accounts of events, and are largely uncontroversial. The parents are young, the mother is 24, the father is 23. T is their first child, they have been together since around the middle of 2018.

14. The father is autistic and has adult ADHD. He was removed from the care of his parents due to neglect at age 2 and placed initially in foster care and then with his grandmother and then he moved in with his father aged 16. Both parents suffer from anxiety. The cognitive assessments place the mother in the 45^th percentile for IQ and the father in the 18^th percentile. Neither parent uses drugs nor abuses alcohol. This is credit to them as both have grown up in an environment where there has been significant abuse of alcohol. The father's father has been a chronic alcoholic for many years with very serious drinking habits, and the mother's mother works as a carer but has a history of heavy drinking.

15. Prior to T's birth the mother was diagnosed with gestational diabetes. T was born on [date] July 2019. Following the birth the mother had sepsis and both mother and child were put on antibiotics for around five days. They stayed in hospital until they were discharged on 30 July. On discharge the parents and T returned to their home, a flat that they had shared as a couple for a year or so with the paternal grandfather. A health visitor had attempted a visit to the home on 7 August 2019. She reported that no one was in and the parents reported that she had called at the wrong flat in the building.

16. The parents' neighbours in the flat below, Mr and Mrs G, had offered to look after T for that night so the parents could catch up on some sleep. The parents took T down to their flat at around 6, stayed briefly and then left T with them at around 6.30. T did not settle, and Mrs G and the mother continued to exchange messages over the evening until around 1 am when Mrs G said he was too unsettled and the mother went to collect him. At that point the maternal grandmother was present in the parents' flat as she had come to visit and another relative had also visited.

17. On seeing T the maternal grandmother thought he was wheezy, and the mother telephoned 111 and then the relative took both parents and T to A&E and they arrived shortly before 2am. On arrival at the hospital T was seen in triage by Nurse W and was then referred to the paediatric registrar, Dr K. Nurse W recorded that initially Dr K thought that the bruises on his face were birth marks, but he then agreed they were bruises and admitted the baby to a paediatric ward at around 4am for safeguarding. He reported seeing blue bruises on the left side of T's face.

18. Neither parent reported having seen the bruises before although they had noted a birth mark on his temple and the back of his head. T was seen by the paediatric consultant Dr M and staff nurse B on the ward round the next morning. Dr M made a body map and recorded two bruises on T's check, one that was .5 centimetres by .5 centimetres and another larger bruise marked as a very faint blue bruise that had less demarcation. Dr M immediately made a plan for T to undergo a skeletal survey, CT scan, and ophthalmology assessment. These were done later and revealed no injuries.

19. The social worker statement shows that Social Services and the police were involved immediately and had a strategy meeting by telephone. PC L and Miss McK, the social worker, visited the parents' home shortly afterwards on 8 August and also visited the home of neighbours, Mr and Mrs G, who had offered to take the couple and T to stay in their home. T remained in hospital for tests and was being cared for by his father on the night of 11 August 2019. At around 5.50am in the morning of 12 August Nurse P, a staff nurse on the night shift on the ward, reported that she had seen the father turn T in a 360 degree turn.

20. It was common ground that T had been turned 360 degrees by his father but there was an issue as to the nature of the turning. A further strategy meeting took place at around 11am that day and a DC B attended. The Local Authority applied for an Emergency Order on 12 August and on 13 August T was placed with the foster carer. The strategy meeting records show that following this incident the paediatricians considered that further scans should be avoided but T was examined on 15 August by a consultant ophthalmologist who reported no signs of injury.

21. On 13 August DC B interviewed the father in a non-custodial interview of which a transcript was made available, and he had Mr G with him as an appropriate adult. In that interview the father gave an explanation of the turning and denied having thrown the baby up but accepted that he had rotated him but kept him supported throughout. T was then placed with a foster carer. The parents moved in with the Gs for a month and then moved in with the maternal grandmother, where they remain. On around 4 September the foster carer raised concerns regarding T's eyesight and T was seen by a consultant ophthalmologist in September; later in October the foster carer reported that he was able to hold a gaze and track objects and fortunately his sight appears normal."

The flipping incident

8. I will return to the other allegations made by the local authority, but it was common ground that the most serious one concerned what has been called 'the flipping incident'. The judge heard from Nurse P and from the father. She summarised Nurse P's evidence in this way:

"46. Nurse P was a staff nurse overnight on 11/12 August 2019 on T's paediatric ward. She reported that at around 5.50am on 12 August she had heard T crying for around five minutes and she was due to give him observations. At that point she had approached his cubicle and had bent down to observe from the outside through the bottom glass window of the door to the cubicle. She accepted that it was awkward to lean down but this was what she did sometimes rather than pulling the blind back on the top window of the door. She reported seeing the father flip T into the air causing him to rotate 360 degrees in the air before the father then caught him.

47. The father had one hand on the baby's lower legs, and she could not clearly recall seeing where the second hand was, although thought it was supporting the baby. The father had one hand on the baby's lower legs when he threw T into the air, letting go of him so he flipped over and then caught him. At first she thought that the father was throwing a doll into the air but by the second time she observed it through the window she realised it was a baby. Just after she walked into the cubicle he did this motion a third time. She emphasised that she had not seen the father swing the baby or reported that the baby was swung by the ankles as had been noted in one of the police reports. She reported that baby T was crying throughout this incident. Her explanation corresponded with the action she demonstrated in evidence with a doll.

48. She told the father that he should not do that to a baby. He said that the father was calm, and he responded that the baby would not settle. She took the baby from the father to give him a walk and asked the father whether a feed was due and suggested he prepare the feed. She then took T out and reported the incident to the nurse in charge and the doctor present. At that stage the father walked out of the ward and she did not see him again. Her evidence was that she was quite shaken and burst into tears immediately following this incident. Shortly afterwards she spoke to Nurse B about the incident and then left her shift."

9. The judge recorded the father's account of the incident:

72. The father's account of the events on the morning of August 12 was that T had woken and was due a feed, but he was grizzly and unsettled. The father tried settling him with cuddling and cradling him and putting him on his chest and using the dummy with no success. At that point he was sitting on the bed and he tried to rotate T in a 360 motion as a last resort with the baby's head supported throughout. He demonstrated the action with the baby doll, and it involved turning the doll completely upside down and having his hand over the baby's face to support it as it went over. He reported that he was at the time trying to make sure his hands were moving in the right place to support T's body.

73. He said that it seemed to work as the crying seemed to lessen and he did it a second time. He reported that T started to stop crying and that after the motion he was silent, and smiling, more happy and settled. He then put T down on the bed and he was undoing his buttons when Nurse P came in. The father's evidence was that when Nurse P came in she told him that he should not be swinging his child like that. She asked if T needed a feed and offered to take him for a walk. The father went to get a feed but when he returned there was no one in the cubicle so he left the ward."

10. The judge noted the difference in the two accounts:

"55. … What Miss P reported was very different to what the father reported so the discrepancies were unlikely to be due to poor vision or a mistake in perception. The father demonstrated a rather slow, awkward turning motion where he was carefully moving his hands at each stage to support the baby in a slow rotation, whereas a flip is a quicker motion where the child is let go and caught again."

11. The judge also considered evidence about the father's previous handling of T. On the morning of 12 August, the mother was asked about it by Nurse B:

"44. … She asked the mother if the father had done anything like this before and the mother had said that he always turns him. When asked what this meant the mother made a roly-poly motion. The mother had reported to Miss B that T liked the motion of this turning and when asked what she does when the father does this turn she replied that she did not do anything when he did. Miss B's evidence was straightforward, and I accept it."

12. The mother was also spoken to by DC B:

"45. … She reported that the mother was upset and crying. When DC B asked her if she had seen the father flip the baby she said she had never seen him throw the baby in the air but had seen him spin [him] and had demonstrated with a doll, turning him over 360 degrees in a supported manner. She said she had seen him do it around three times. The last time she had seen him was on the night T went into hospital. The mother had reported to DC B that although she does not do it, it seemed to settle the baby. … DC B's evidence was straightforward, and I accept it."

13. When interviewed by DC B, the father's account of previous handling was this:

"75. … It was his evidence that he had only done the motion he described once previously in the flat and the mother had seen him do it. She had told him not to do it as she was worried about shaken baby syndrome. He had looked up shaken baby on the internet and considered that it was only a risk if the baby was violently shaken or fell. …"

14. The judge's finding seems to be that there had probably been just one previous incident where the father had rotated the baby, though not involving flipping of the kind seen by Nurse P:

"68. For all these reasons I consider that the mother's precise accounts she gave on 12 August of previous turning incidents and her response to them are not reliable. I accept that she had not seen the father throw the baby up in the air at home and the Local Authority did not allege this in closing. I also accept that she had broadly seen him handling the baby gently and trying to comfort him in various ways, including bouncing and the 360 degree supported motion. She may have thought his manoeuvres were more adventurous than she would undertake herself and I accept that she may have warned him of shaken baby syndrome."

At a later stage she remarked:

"118. …To the extent it was relevant I considered that the flipping had not previously occurred, and the father had flipped the baby in a genuine but ignorant and misguided attempt to settle him. …"

Finally, she said this:

"124. … For reasons set out above I accept that the father turned the baby in a supported rotation probably on one occasion before and the mother had probably seen this and warned the father of the risks of shaken baby syndrome."

15. At all events, the judge considered the evidence about the 12 August incident itself very fully over the course of no less than seven or eight pages of the judgment. The parents had strongly challenged Nurse P's evidence on a range of grounds but the judge found it reliable and she accepted it. She accepted that the father had been motivated by a desire to settle T and that he had not acted in anger or distress or with any intent to cause harm. She remarked:

"84. The turning incident took place over a matter of minutes but has been deeply significant for both the parents and T. It was the father's own evidence that he felt upset and destroyed in the minutes after the incident and needed to go outside to clear his thoughts. In the hours, days and months that have followed the consequences of those few minutes have been revisited, not least each time the parents have come back to court.

85. I take into account the father's current acceptance that the rotation manoeuvre he says he did was an unusual and unwise one, and his evidence that it involved carefully turning the baby completely upside down and putting his hands over his face to support him and moving his hands continuously over the baby to support him. I considered that this rotation was a highly unusual and awkward manoeuvre to carry out and unlikely to settle a baby, even if carried out on a previous occasion at home. On the father's own evidence it was an inappropriate manoeuvre that could cause harm to T even if not significant harm.

…

88. I do not consider that he was deliberately attempting to cover up the truth, and at this stage after all the retelling he may remain genuinely convinced that his version is right. However, taking all the available evidence into account I consider that his account of what took place was not reliable and Nurse P's account was to be preferred."

16. These being the judge's findings about this incident, she addressed its relevance to the threshold, again at length, between paragraphs 109 to 125. She started from the position that she had found that:

"The father flipped T in the air, providing no support for his head, and undertook this movement three times."

17. A consultant paediatrician, Dr Rahman, had been instructed to carry out a paediatric overview. Among other issues, he was asked questions in writing and orally about the risks involved in this form of handling. When giving his oral evidence he was shown a recording of the demonstration given by Nurse P, using a doll, when she had given her evidence. The judge summarised Dr Rahman's view on this issue:

"111. Dr Rahman considered that the manoeuvre demonstrated by Miss P, what was described as flipping T in the air, was risky. His evidence more precisely was:

"It is a risky manoeuvre. If he dropped him it could have caused injury. The manoeuvre itself could have caused injury but not significant injury if done like that."

However, he maintained the view expressed in his statement that there was a risk that T could have suffered retinal bleeds, brain injury and intracranial bleeds from the action and also maintained that he could have been dropped or slipped and fallen, leading to a more serious injury such as fracture to skull or long bones. … His evidence was that if it was done gently it would present less risk. Again, the level of risk to T from this manoeuvre depended on how often the manoeuvre took place and also how much force was used in turning T over."

18. As to that, the judge observed:

"116. Dr Rahman's conclusion that gently flipping a baby up and over was risky but not going to cause significant injury may be surprising to a layperson who might be horrified to see a newborn baby flipped up in the air. I have looked critically at his opinion as I [am] assessing whether the action caused harm and also the likelihood of harm to T at the time or in the future. The fact that the baby escaped unharmed on this occasion is only of limited weight if there was a real possibility that significant harm would be suffered then or on another occasion.

117. I am not bound by Dr Rahman's opinion as to what establishes the required likelihood of significant harm, not least because the question I have to answer requires the correct application of the legal test under Section 31. I also take into account that the legal test requires more than a risk, it requires that significant harm is likely in the sense of a real possibility of harm rather than harm being more likely than not. I take into account the Guardian pointing out that the father had attempted the manoeuvre even though the mother had already told him not to rotate the baby, and that in his evidence he still struggled to accept there were any risks associated with the rotation movement other than dropping T."

19. The judge then stated her conclusion on this aspect of the threshold. In fairness to her, I must set it out in full:

"118. However, in determining threshold the Court is not concerned with actual intent or blame and I have to assess the risk objectively. To the extent it was relevant I considered that the flipping had not previously occurred, and the father had flipped the baby in a genuine but ignorant and misguided attempt to settle him. Although with DC B he had initially resisted accepting that the manoeuvre was wrong he had then conceded that it was inappropriate. Overall taking account of his evidence and his conduct since the incident when there has been no question of flipping the baby or turning him inappropriately I considered that the father had properly acknowledged the risks of flipping the baby or rotating him.

119. I took account of DC B's impression that his actions reflected a lack of understanding. I took careful account of Dr Rahman's evidence in his statement that a child being thrown up 360 degrees placed T at risk of retinal bleeds, injury to brain and intracranial bleeds as well as a fracture and that such risk applied even if the parent was sitting on the bed. I took careful account of the Guardian's concern regarding the risk to T and also Nurse P's alarm and distress in having seen the incident and the hospital's reaction to it.

120. They are well informed observers, although the Guardian is acutely aware of both parents' vulnerability. Miss P was right to step in to address the incident. The father's conduct also justified the concern of the doctors and the Local Authority as it was an unusual manoeuvre that took place in circumstances where there was unexplained bruising, and T's home conditions were also placing him in a vulnerable situation. The late development of his eyesight further justified these concerns.

121. However the test for state intervention is not that a parent's conduct is alarming to an observer. Indeed the test is an objective one and the Local Authority submitted that it does not take into account the vulnerability of the parent. The test requires more than a risk of significant harm, it is necessary to show that the significant harm is likely and is attributable to the parents' care falling below what it would be reasonable to expect. I considered that it was appropriate to adopt Dr Rahman's assessment of the risk of harm taking account of Nurse P's demonstration of what happened.

122. Paediatricians as well as most parents are aware of the fortunate resilience of small babies to hazardous handling. They will know that many parents will, even with the best intentions and some knowledge of childcare, have handled their newborn baby at some point, usually a relatively one-off occasion, in a way that could certainly be described as risky or alarming to a more experienced parent, and in a way that entails a risk of injury. This is commonplace and there is a large variety of ways in which parents can do this.

123. Paediatricians see the cases where the baby is injured, and Dr Rahman was well placed to assess the risk of injury. He was entitled to conclude that what was demonstrated should not have caused any significant injury as it was done gently. He was obviously influenced by the gentle manner in which the flip was demonstrated and correctly emphasised that the risk of brain injury was directly related to the force used. Similarly, he suggested that the frequency of the action was relevant. These are clearly relevant considerations in assessing the likelihood of significant harm. A situation where a newborn baby is thrown up vigorously or on repeated occasions represents a very different risk to that arising out of a single occasion where that baby is thrown up gently by a well intentioned parent, especially where the baby has come out unscathed. Overall I conclude that the facts found did not establish that T was likely to suffer significant harm.

124. The Local Authority alternatively alleged the supported 360 rotation that the father described created a risk of significant harm. The Local Authority pointed out that even if Miss Pentecost's evidence was not accepted, the father had accepted that he had rotated the baby on at least two occasions. For reasons set out above I accept that the father turned the baby in a supported rotation probably on one occasion before and the mother had probably seen this and warned the father of the risks of shaken baby syndrome. The Local Authority relied on Dr Rahman's evidence to suggest that T was at risk of significant physical harm from this manoeuvre even if it was a gentle one. Dr Rahman suggested that this manoeuvre was not an ideal thing to do and was risky, presenting similar risks to the flip but noted it was done gently.

125. I considered that the manoeuvre when T was supported at all times was more unusual than the flipping but less risky because the baby was supported. For similar reasons to my conclusions on the flipping I do not accept that it placed T at risk of significant harm, or that the father's care of him in this respect fell below that which would be reasonably expected."

20. The judge therefore found that the local authority had failed to prove that aspect of its case.
21. The other threshold allegations concerned: lack of adequate knowledge of basic childcare skills or routines; the facial bruising; parental vulnerability to domestic abuse; neglect leading to severe nappy rash and oral thrush; and unsafe and unhygienic home conditions.

Lack of basic knowledge

22. The judge's conclusion about the parents' lack of knowledge and skill was this:

"95. For the avoidance of doubt, I consider that the allegation that the parents lacked adequate childcare knowledge was not proven. They were young first time parents and their child was removed when he was not yet three weeks old. It was common ground that they had limited knowledge of caring for small babies and were vulnerable individuals.

…

97. However, taking the overall evidence into account and assessing their skills objectively I consider that their knowledge was within what would be reasonable to expect and presented no risk of significant harm. …"

The bruising

23. By the time of the hearing, the local authority did not contend that these were inflicted but it argued that T could not have caused the bruises himself, that a reasonable parent should have noticed them and been able to explain how they had occurred, and that this was relevant to their capacity to protect T from future physical harm. The judge considered the concerns of the local authority to be legitimate and noted Dr Rahman's advice about the great rarity of bruises in non-mobile children. She stated that bruises on a new-born baby are evidence of physical harm. The existence of facial bruising on T together with a lack of any explanation for the bruising was troubling and warranted investigation, including consideration of neglect, even if there was no evidence of deliberate or inflicted injury.
24. This was the judge's conclusion on this issue:

"107. It was Dr Rahman's evidence that it was unlikely that a lot of force was inflicted to produce the bruises, and the likely degree of pain would have been moderate. The bruises could have been sustained accidentally while handling the child. I conclude that the most probable cause of the bruising was that a parent or the grandfather accidentally bumped into something while holding the child or dropped or knocked something near the baby's face and this followed Dr Rahman's evidence. Although this sort of accident is attributable to the parents' care of T it was not attributable to the care falling below that it would be reasonable to expect.

108. The bruising was noticeable to the professionals, but I accept the parents' evidence they had not noticed it prior to coming into hospital. This was reasonable since they were less alert to bruises than a professional or an experienced parent. Indeed they were inexperienced parents trying to make sense of their very newborn baby with his own special markings and I take into account the lighting was poor in their flat. I conclude that the Local Authority has not shown significant harm or risk of physical harm attributable to the bruising or the parents' failure to notice the bruising."

Vulnerability to domestic abuse

25. The local authority pointed to risk to T arising from the parents being vulnerable to abuse by associates. It referred to the police being called to the home twice in the year before proceedings were issued, and on other occasions. The judge rejected this part of the case, saying that vulnerability of this kind will rarely be a threshold fact and that the local authority had not pleaded its case properly. She concluded:

"139. … I am not satisfied the alleged conduct and vulnerability evidences a risk of significant harm and the Local Authority failed to establish that the parents would be unable or unwilling to protect T from exposure to domestic abuse or aggressive altercations."

26. Shortly before the hearing, the local authority had produced a social work statement that recorded a conversation in which the mother had alleged domestic abuse by the father towards her. However, the local authority's approach, apparently not challenged at the time by the other parties or the judge, was that this issue could be considered at the welfare stage, and it does not therefore feature in the judgment. This somewhat curious situation was the result of the lack of preparedness of the local authority's case.

Neglect and poor hygiene

27. The judge approached this allegation in this way:

"141. … In making findings of fact I must take into account the overall picture and the existence of moderate to severe nappy rash and thrush would be relevant to allegations of neglect due to poor hygiene and childcare skills and could form part of the jigsaw of facts justifying a conclusion that a baby is at risk of significant harm from neglect."

However, in the light of Dr Rahman's advice about the common and spontaneous nature of nappy rash and thrush, she found that significant physical harm had not been established, nor had T's condition been proved to be attributable to lack of care.

Home conditions

28. The judge described these in this way:

146. The allegation was that the parents neglected T's physical need for hygiene in that the home conditions were cluttered, unsafe and unhygienic. Rabbits were in the property with rabbit droppings throughout. There was a large hole in the ceiling and there was no accessible kitchen. It was said that the bedroom T shared with his parents was dirty and cluttered.

…

149. The rabbits were let out throughout the property although the parents generally preferred to have them in their bedroom as it was safer for the rabbits. There had been a problem with the electrics and the main ceiling light did not work in the bedroom although sockets did so they could plug in the television and use that for lighting. They washed up the baby's bottles in the bathroom. The kitchen tended to be full of rubbish such that they had to climb over the rubbish to get to the sink, although the parents used a kettle near the door.

150. There had been an infestation of fleas and mice in the flat that was still an issue. The grandfather invited people over to drink and would stay up sometimes all night and sometimes disturbing the parents in the middle of the night. He was controlling about the use of the hob and the washing machine. Obviously the flat was unsuitable for T and poverty was the main reason preventing the parents finding a better place to live."

29. The judge dealt with the situation in the home, again at some length. She found that the parents minimised their responsibility for giving T clean living conditions and that they did not keep their bedroom in a fit state for a new-born baby. However, Mr and Mrs G had offered to take them in and she found that the parents had resolved to move out by the time T was admitted to hospital. Her conclusion on this issue was this:

"162. I have made findings regarding the parents' contribution to the unhygienic conditions that T was faced with. It is important the parents understand their responsibility as parents. I am acutely aware of the eccentricities of parenting that must be tolerated by the state and are not caught by Section 31. Keeping rabbits roaming around a new-born's bedroom and leaving their droppings probably does not amount to significant harm. The state has to tolerate children being raised in eccentric ways. However, I considered that the Local Authority had very legitimate concerns about the home conditions that were attributable to the parents' conduct. It remains the case that poor standards of hygiene will expose T to harm, even if it does not meet the threshold of Section 31.

163. I conclude that the Local Authority did not establish that the home conditions as at 12 August met the threshold, and overall the Local Authority has not met the threshold for intervention under Section 31."

30. That being the final threshold allegation, the application for a care order was dismissed.

The subsequent hearing

31. The judgment, given ex tempore on Friday 1 May, finished at 5 pm and a further hearing was fixed for consideration of consequential orders. There followed a series of exchanges between the parties and the court about plans for T in the context of the local authority's application for permission to appeal. The situation was complicated by the pandemic, which had led to contact ceasing since 23 March. The local authority sought an order under s.40 CA 1989 to hold the ring pending the appeal decision, and a stay.
32. The matter came back before the judge on 7 May, and she declined to make these orders. There was discussion about transition plans. The local authority's care plan following the dismissal of its application had been for there to be an assessed transition with a view to T returning progressively to his parents' care over the course of a few weeks, with a placement with the aunt being considered if the progression to the parents was unsuccessful. The parents sought T's immediate return. The Guardian supported a longer, careful transition as T had formed a strong attachment to his foster carers.
33. The judge gave a second judgment on 7 May. She expressed disappointment at the stance taken by the parents:

"41. The parents' current position somewhat reinforced my view that the Local Authority have legitimate concerns about the risk of harm to T. However, it does not change my view on the threshold issues which were not dependent on the parents' willingness to accept professional support.

42. I urged them to take on board the advice of the Guardian who has taken a truly neutral stance in these proceedings. She correctly sees a risk of harm to T in how he transfers his attachment. This is important to him now. It goes to whether he avoids distress over the next few days and weeks and it may have a long term impact on how he forms attachments in his later life.

43. Although I have expressed my views strongly, I do not appear to have any jurisdiction or oversight to ensure that T has a gentler staged transition. It was common ground that, upon a ruling that threshold is not met, the care proceedings finish and there is no power of oversight or to prevent T's immediate return.

44. It seems surprising that, in making an Interim Care Order, the Court must ensure it puts the child's welfare at the centre of the decision yet, in making a decision that the order will be discharged, the Court is not empowered to make directions to protect the child from harm caused by an abrupt transfer of care. It was not suggested that I had any general power to protect T in this way and I acknowledge that, in the absence of any other order, the proceedings finish and the Court has no power of oversight over T's welfare."

34. Following the judge's decision, the parents agreed to T being accommodated for one week for his transfer to their care be arranged. However, as described above, a stay of the judge's order was granted that evening by this court. Since then, contact has been taking place five days a week.

Section 40

35. The terms of this provision are set out in the last section of this judgment.
36. The judge accepted (contrary to submissions made on behalf of the parents) that s.40 can apply where the threshold has not been met, and that the Human Rights Act 1998 does not lead to a different view. Her analysis included these passages:

"50. … The purpose of section 40 is to enable the court to protect a child for a very limited period, the appeal period, in circumstances where the judge considers that his decision may be wrong in dismissing the application for a care order (or discharging a care order).

…

55. The real issue is as to whether to make an Interim Care Order under section 40, and I accept that Article 8 rights are engaged in that decision. In a similar way to making any decision to grant an Interim Care Order, the Court must make sure that any interference with Article 8 rights is proportionate and weigh the risks to the child against the interference with the parents' and the child's family rights.

56. In circumstances where the court's decision on threshold may be wrong and the court considers that it is in the child's interests to maintain the status quo for a limited period, then Article 8 does not block protection of that child. Instead, the Court has to balance whether interference is justified pending the appeal. The fact that section 40 is not in common use suggests that it is to be applied cautiously. The existence of a risk of harm if the judge has got a decision wrong, would it not in itself justify an order since otherwise orders would be common.

57. It would be unusual for a judge to make a section 40 order where she did not consider that the appeal had a real prospect of success. Section 40 is intended to enable a care order to be made or continued. It is not intended to enable the Court to exercise control over the manner in which the child returns to his parents when a care order is discharged. It would not be correct use of the power to make directions for a staged transition of care so as to mitigate risk from an abrupt return, and the Local Authority did not suggest that this would be an appropriate order under section 40.

58. Here, I do not consider that the appeal has a real prospect of success. Further, based on my detailed findings of fact and also the evidence as to the parents' conduct since 12 August 2019, the risk of harm to T, even if I am wrong on threshold, is not so great as to justify separating him from the care of the parents for another significant period. Accordingly, I dismiss the section 40 application."

The grounds of appeal

37. The local authority, through Ms Grief QC and Mr Barnes (neither of whom appeared below), advanced four amended grounds of appeal, which I label and distil in this way:

1. The flipping incident: the judge was wrong to find that the serious risks involved in the father's behaviour did not amount to a risk of significant harm, particularly where the father had (a) done it before, (b) been warned by the mother and (c) had denied what he had done.

2. Bruising: the judge did not apply an objective 'reasonable parent' test, instead focusing on and excusing the parents' inexperience, and did not consider the obvious inferences from the finding about how the bruising had been caused, namely that there had been a lack of care or supervision.

3. Home conditions: the judge failed to properly analyse the risks to T at the relevant date, arising from the state of the property and the behaviour of the grandfather, risks which were only averted by the imposition of protective measures, namely the application for an emergency protection order on 12 August.

4. Section 40: when approaching the exercise of this power, the judge applied the wrong test and was wrong to place any weight on her refusal of permission to appeal.

38. The local authority did not pursue a further ground of appeal that criticised the judge for accepting its own proposal not to litigate the issue of domestic abuse between the parents at the hearing. Nor does it challenge the judge's rejection of its threshold case about the parents' level of knowledge, neglect through nappy rash, and domestic violence victimhood. It does however, seek to reserve the right to refer at the welfare stage to the concerns said to have been expressed by the mother about abuse of her by the father. This was the cause of controversy at the hearing before us.
39. Where there is a normal composite hearing, the court will find all the facts that are relevant to its inquiry, determine whether the threshold is crossed and, if so, make its welfare decision. Where there is a split hearing, the first limb being for fact-finding and determining threshold, the court should identify the issues for that hearing. In a case of the present kind, the local authority will normally be required to present its core case in relation to the family history, and, so far as it concerns the threshold, will necessarily focus upon harm or likely harm. It is therefore to be expected that the court will make its historic findings in the course of that hearing, including but not necessarily limited to issues of past harm or risk of future harm. Those findings will significantly inform later assessments and decisions. Importantly, in both the private law and public law spheres the court is required to identify and assess any issues of domestic abuse that may bear upon the welfare of a child or the capacity of a parent.
40. At the welfare stage, the court must have regard to all the circumstances, and in particular to the wider range of factors in the welfare checklist. That will lead to the court making a range of other findings to underpin its final decision. However, for obvious reasons of fairness and good order, it is not usual for the local authority to seek to 'bulk up' its case on harm at the welfare stage by making factual allegations of harm arising from the same family history that would better have been made at the fact-finding and threshold stage. But the governing considerations for the court are the paramount welfare of the child and the fairness of the proceedings.
41. In the present case, the issue of whether there had been domestic abuse between the parents was unfortunately not considered at the hearing, apparently because it arose in the course of an assessment directed to the welfare stage, for which the local authority was unprepared. The judge will now need to consider whether and to what extent this issue should be litigated at the welfare stage, balancing welfare and fairness, and acknowledging the need to take allegations of domestic abuse seriously.

The respondents' position

42. Mr Arnot and Ms Cox for the mother, and Ms McGrath QC and Mr Chippeck for the father, sought to uphold the judge's conclusions on these main grounds:

1. There should be appellate restraint in relation to findings of fact and judicial evaluations where the judge has correctly stated the law and had a unique chance to assess the witnesses.

2. The local authority failed to demonstrate how the minor accidental bruising crossed the threshold.

3. Based on all the evidence, and particularly the demonstrations by Nurse P and the father and the evidence of Dr Rahman, the judge was entitled to find that there was no likelihood of harm arising from the father's handling of T.

4. The home conditions were outside the parents' control and by 12 August they had decided to move.

5. If the appeal succeeds, this court should not substitute its own findings but should remit for rehearing.

6. The judge was right not to make a s.40 order, and the argument that there can be no such order where the threshold has not been made out is repeated in a Respondent's Notice provided by the mother.

43. On behalf of the Guardian, Ms Cook QC (who did not appear below) and Mr McCormack support the appeal. They submit that on the basis of its combined findings regarding the flipping and the bruising, the court was bound to find the threshold crossed. They urge that future planning for T is rapidly progressed.

Conclusion: the threshold

44. I acknowledge the care with which the judge approached her task and firmly remind myself that we are not entitled to depart from her evaluations unless they were not open to her. I am nevertheless amply satisfied that her decision to dismiss the proceedings was wrong for the following reasons.
45. There are three elements to the harm required by the threshold condition in s.31(2) CA 1989: it must be actual or likely; it must be significant; and it must be due to parenting that is not reasonable. In my judgement, the judge's specific findings about the flipping incident at hospital and the previous incident at home, and the unexplained accidental facial bruising in such a young baby, in each case taken in the context of her general findings about the parents' personal circumstances and the home in which they had been living, lead inexorably to the conclusion that all three of these elements were satisfied. The father's bizarre handling of a baby of T's age plainly gave rise to a real possibility of future harm that could not sensibly be ignored; the harm that might result was not merely significant but serious; the treatment of the child was undoubtedly not what was reasonable for any parent, however inexperienced. In my view the incident in hospital alone was sufficient to cross the threshold, and the judge should have so found. It was not just the incident itself, but the implications for T's future safety of being in the care of a parent, however well-intentioned, who could even consider such handling to be appropriate, let alone act it out. Taken together with the unexplained bruising, the inevitable conclusion that the threshold for protective intervention was crossed is reinforced.
46. Next, instead of looking at the whole picture that her findings painted, the judge treated each individual finding in a compartmentalised manner. Threshold allegations are separated out for forensic purposes, but there is only one threshold and the court must measure the effect of all of its findings against it. To take a crude example, the threshold may not be crossed in a case of a parent who has weak parenting skills or in the case of a parent who is an alcoholic, but it may well be crossed where the same parent has both characteristics. Each piece of information affects the calculation of risk. This is quite different to the position of findings of primary fact, where unproven facts cannot be aggregated to form proven facts. As ever, the position has never been better expressed than it was by Lord Nicholls of Birkenhead, who said this in Re H and R (Minors) (Child Sexual Abuse: Standard of Proof)  [1996] AC 563:

"94. Thus far I have concentrated on explaining that a court's conclusion that the threshold conditions are satisfied must have a factual base, and that an alleged but unproved fact, serious or trivial, is not a fact for this purpose. Nor is judicial suspicion, because that is no more than a judicial state of uncertainty about whether or not an event happened.

95. I must now put this into perspective by noting, and emphasising, the width of the range of facts which may be relevant when the court is considering the threshold conditions. The range of facts which may properly be taken into account is infinite. Facts include the history of members of the family, the state of relationships within a family, proposed changes within the membership of a family, parental attitudes, and omissions which might not reasonably have been expected, just as much as actual physical assaults. They include threats, and abnormal behaviour by a child, and unsatisfactory parental responses to complaints or allegations. And facts, which are minor or even trivial if considered in isolation, when taken together may suffice to satisfy the court of the likelihood of future harm. The court will attach to all the relevant facts the appropriate weight when coming to an overall conclusion on the crucial issue."

Here, the judge found that T's parents are young and inexperienced, that they had been living in conditions that were entirely unsuitable for a new-born baby, that he had already suffered some harm in the form of worrying and unexplained bruising, and that his father had handled him both at home and in hospital in a manner that was obviously unsafe, despite being warned about it. She found that each of these matters individually fell below the threshold. What she did not do was to stand back and look at the whole picture.

47. Further, the exact nature of the judge's conclusions regarding the flipping incident and the bruising are not clear. Although she directed herself clearly at paragraph 121, it is unclear from paragraphs 118-123 whether she found that the harm that might be caused by flipping was not significant, and/or that it was not likely to be repeated, and/or that it was parenting that did not fall below what was reasonably to be expected. She certainly did not make any finding, express or implied, that supports Ms McGrath QC's submission to us that the father's behaviour was "a one-off" so that the prospect of recurrence could sensibly be ignored. Nor does the reference to Dr Rahman's evidence give any support to the submission that any harm that might be caused would not be significant. As to the cause of the bruising, the only statement about that is at paragraph 107, where the judge says:

"I conclude that the most probable cause of the bruising was that a parent or the grandfather accidentally bumped into something while holding the child or dropped or knocked something near the baby's face and this followed Dr Rahman's evidence."

The way in which this is expressed allowed Mr Arnot to make the startling submission that at the end of a seven day hearing in proceedings lasting nine months the court had made no finding at all about this original presenting issue. I reject that and prefer to read it as the judge finding that accidental bumping was more likely than not to be the explanation (and not just the most likely explanation) for the bruising, but it is not clear why she did not view T's bruises in the context of his chaotic living conditions and, in particular, his father's approach to handling him. Bruising, even accidental, in a new-born baby has a quite different significance to bruising in a mobile child, and the judge was bound to consider the likelihood of repetition where no satisfactory explanation had been given.

48. Then, the judge correctly directed herself as to the relevant date for her assessment, 12 August, but she did not address what T's situation would have been had protective proceedings not been taken. I accept the parents' submission that the judge found that they had already decided not to return from hospital to the grandfather's flat, and that the specific risk of harm from living with him in those premises had come to an end. However, that does not impact upon the risks arising from the father's notions of handling T. As the judge said at paragraph 118:

"Although with DC B he had initially resisted accepting that the manoeuvre was wrong he had then conceded that it was inappropriate. Overall taking account of his evidence and his conduct since the incident when there has been no question of flipping the baby or turning him inappropriately I considered that the father had properly acknowledged the risks of flipping the baby or rotating him."

The interview with DC B took place the day after the EPO. The father's account of the incident, rejected by the judge, was given nine months later. It is impossible to see how the issues of handling and bruising could be regarded as being in the past at the time protective measures were taken. The fact that the father may have adjusted his views to some extent by the time he gave evidence is not relevant to the question the judge had to answer.

49. Finally, I regret to say that this inquiry into the relatively simple issues that were ultimately before the court became over-legalised and overcomplicated. While the judge was strictly correct to say at paragraph 121 that the test for state intervention is not that a parent's conduct is alarming to an observer, that observation does not take one very far where the witness, a specialist nurse, was so shaken by what she had seen that she burst into tears, and where the doctors and social workers were united in their level of concern for T.
50. Also, the judge herself expressed surprise at paragraph 116 at some of Dr Rahman's evidence as she understood it. Yet at paragraph 121 she adopted as her own what she described as "his assessment of the risk of harm taking account of Nurse P's demonstration of what happened". The simple fact is that a court does not need expert evidence on the question of whether a two-week baby, flipped in the air through 360 degrees three times by an inexperienced parent, is likely to suffer significant harm. The presence of an expert for one reason does not mean that he or she has to be asked about everything. In this case, Dr Rahman's advice was needed in relation to the bruising and T's overall health status, but he was also asked about the possible medical consequences of handling of this kind. His broad answer concerning the risks from shaking or dropping could scarcely be controversial, but he then became drawn into a detailed discussion about different kinds of handling. Those were matters that are well within the competence of a specialist court. The consequence of involving an expert unnecessarily is shown in this case, where the judge in her decisive paragraph 123 analysed the risk of harm through the lens of opinions expressed by Dr Rahman, whose own source of information was a remote viewing of a video clip of the witness doing her best to re-enact the incident with a doll. The judge had accepted the evidence of Nurse P in preference to the evidence of the father and she should have applied her own assessment to it. I have little doubt that the elaborate thought process that instead occurred contributed to a situation in which the wood was lost behind the trees.
51. For these reasons I would allow the appeal and set aside the judge's threshold determination. There is no purpose in remitting an issue with only one answer, nor would that be in anyone's interests. I would therefore substitute the following determination:

"The court finds that the threshold condition under s.31(2) Children Act 1989 is satisfied because, T was at the relevant time (namely when the application was made for an emergency protection order on 12 August 2019) likely to suffer significant physical harm attributable to the care given to him by his parents not being what it would be reasonable to expect, in that (in each case in the context of the facts found by Ms Clare Ambrose at paragraphs 13-21 of her judgment of 1 May 2020):

(1) he had been subject to inappropriate handling by his father at home by being rotated 360 degrees in a supported manner, following which the mother had warned the father that this was inappropriate and the father had researched shaken baby syndrome on the internet;

and

(2) he had a few days later been subject to inappropriate handling by his father in hospital, by being flipped into the air while he was crying three times in succession, with the father released him and causing him to rotate 360 degrees before catching him, the incident having been seen and stopped by hospital staff;

In each case, such handling, which occurred in a misguided attempt to sooth T, gave rise (dependent upon the precise degree of force) to a real risk of significant injury to the brain and eyes, and (if T was dropped) of significant injury to the head and limbs;

and

(3) because he had suffered two accidental facial bruises that the parents had not observed and for which they could offer no explanation."

Conclusion: Section 40

52. The final issue concerns the proper interpretation of s.40. This provision is rarely encountered, probably because it is uncommon for multi-issue care proceedings to be dismissed and also because the remedy of a stay provides a ready alternative when permission to appeal is being sought. At all events, this appeal provides an opportunity to consider the proper use of the power.
53. Section 40 is in these terms:

"40 Orders pending appeals in cases about care or supervision orders.

(1) Where—

(a) a court dismisses an application for a care order; and

(b) at the time when the court dismisses the application, the child concerned is the subject of an interim care order,

the court may make a care order with respect to the child to have effect subject to such directions (if any) as the court may see fit to include in the order.

(2)Where—

(a) a court dismisses an application for a care order, or an application for a supervision order; and

(b) at the time when the court dismisses the application, the child concerned is the subject of an interim supervision order,

the court may make a supervision order with respect to the child to have effect subject to such directions (if any) as the court may see fit to include in the order.

(3) Where a court grants an application to discharge a care order or supervision order, it may order that—

(a) its decision is not to have effect; or

(b) the care order, or supervision order, is to continue to have effect but subject to such directions as the court sees fit to include in the order.

(4) An order made under this section shall only have effect for such period, not exceeding the appeal period, as may be specified in the order.

(5)Where—

(a) an appeal is made against any decision of a court under this section; or

(b) any application is made to the appellate court in connection with a proposed appeal against that decision,

the appellate court may extend the period for which the order in question is to have effect, but not so as to extend it beyond the end of the appeal period.

(6) In this section "the appeal period" means—

(a)where an appeal is made against the decision in question, the period between the making of that decision and the determination of the appeal; and

(b) otherwise, the period during which an appeal may be made against the decision."

54. The section provides a short-term safety net. It allows the court to preserve the position of a child who has been subject to an interim care order (ss.1) or an interim supervision order (ss.2) or a care order (ss.3) during the appeal period (ss.4 and 6).
55. There are just two previous reported decisions, both by this court and predating the Human Rights Act. Neither sheds any light on the submissions we have heard.
56. Croydon London Borough Council v A (Note) (No 2) [1992] 1 WLR 984; 2 FLR 348 established that an appeal court itself has no power to make an order under s.40 when the lower court had declined to do so, but Hollings J held that it could instead make an interim care order under s.38.
57. In Re M (A Minor) (Appeal: Interim Order) (No 1) [1994] 1 FLR 54 the trial judge (Ward J) found that the threshold criteria were met on a reduced basis and he did not make a care order. Faced with an application for permission to appeal by the local authority, he made an interim care order under s.40 and left it to this court to decide whether to extend it during the appeal proceedings, which it did. The basis for its decision was that it was not in the child's interests to be moved twice if the local authority's appeal succeeded. Although Butler-Sloss LJ rejected the mother's submission that the order should not have been made because the appeal was hopeless, it is clear to me that the court was in reality applying a welfare test and not a test based solely on the prospects of success of the appeal.
58. Mr Arnot submitted to us that s.40 is not available where the threshold has not been crossed, and that any order made in those circumstances would be unlawful on human rights grounds. Like the judge, I disagree. There are several reasons why the submission is hopeless:

(1) It ignores the obvious purpose of the section, reflected in its title, which is to allow for a holding position where an appeal may be in prospect.

(2) There is nothing in the plain words of the section to suggest any distinction between care proceedings that are dismissed with and without threshold findings. The submission requires the unwarranted addition to ss.(1)(a) of the words "… after the s.31(2) threshold has been found to be crossed."

(3) The submission that the use of the section in a case where the threshold is not crossed is not compliant with Article 8 impermissibly privileges the rights of the parents at the expense of the rights and welfare of the child.

(4) When asked how his human rights submission squared with the established practice of granting a short stay pending a decision on permission to appeal by this court, Mr Arnot had no answer.

The argument contained in the Respondent's notice therefore fails.

59. I turn then to the judge's own reason for declining to make a s.40 order pending appeal. I note her perplexity about the position that had been reached on 7 May, even to the extent that the parents' position "somewhat reinforced [her] view that the Local Authority have legitimate concerns about the risk of harm to T." She was right to say that she had no general power to require a gentle staged transition once proceedings have been dismissed. She was also right to say that the very purpose of the section is to enable a court to hold the status quo for a very limited period pending appeal, but she was wrong to add in paragraphs 50 and 56 the words "in circumstances where the judge considers that [the] decision may be wrong." That was to conflate the power under s.40 with the refusal of permission to appeal. The power under s.40 is, like the power to grant a stay, a protection pending appeal in case the court has inadvertently gone wrong. Of course, if the proffered basis for appeal seems entirely groundless, that will speak against a s.40 order or a stay, but that was not the position here. Unfortunately, the judge had apparently been told by counsel then acting for the local authority that a stay would not normally be granted when permission to appeal was being refused (see paragraph 46); that is not correct and it may have led her to make a similarly incorrect link between a s.40 order and her refusal of permission to appeal.
60. When considering whether to make a s.40 order, the court will consider the welfare of the child, and the rights of the parents and the child, in the particular circumstances of the individual case, taking account of its findings and the appeal rights of the parties. In this calculation there may be factors pointing both ways, but a prominent consideration will be whether an abrupt cessation of the previous order may have undesirable welfare consequences so that a limited delay until the application for permission to appeal can be urgently considered is sensible and proportionate. Such an order entails no significant breach of rights as these applications can be considered by this court on an urgent basis in a matter of days at most. In Re M, Ward J made a s.40 order until this court could consider the matter, which it did twelve days later. The same could have happened here, and did indeed happen when a stay was sought and granted. Had the judge applied the correct test, she would have been likely to have granted either a short-term order under s.40 or, to the same effect, a short stay. Apart from being correct in principle, that would have resolved the practical difficulties arising from the stance taken by the parents on 7 May.
61. In concluding that the appeal should succeed, I say nothing about the welfare decision that now remains to be taken, except that it is now urgent.

Lady Justice Simler

62. I agree.

Lady Justice Nicola Davies

63. I also agree.