BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales Court of Appeal (Civil Division) Decisions

You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> R (A Child) [2019] EWCA Civ 895 (24 May 2019)
Cite as: [2019] EWCA Civ 895

[New search] [Context] [View without highlighting] [Printable PDF version] [Help]

Neutral Citation Number: [2019] EWCA Civ 895
Case No: B4/2019/0725

Her Honour Judge Anderson

Royal Courts of Justice
Strand, London, WC2A 2LL
24 May 2019

B e f o r e :



R (A Child)


Ashley Lord (instructed by Crockett & Co Solicitors) for the Appellant Intervener
Simon Wilkinson (instructed by Leeds City Council) for the Respondent Local Authority
Written submissions were received on behalf of the other Respondents (the Mother, the Father and the Children's Guardian).

Hearing date: 22 May 2019



Crown Copyright ©


    Lord Justice Peter Jackson:


  1. On 25 February 2019, Her Honour Judge Anderson, sitting as a deputy judge of the High Court, handed down a fact-finding judgment after a hearing lasting 25 days in proceedings concerning five children from two related families. The focus of the hearing had been to establish how two young children from one of the families had come by inflicted injuries at different points in time. Those children were LM, now five, and LR, now rising two. Also under consideration was the effect on the children of the general lifestyle of the parents (LR's father was acting as LM's stepfather at the time she was injured).
  2. As will be seen, this appeal relates to just one of the judge's conclusions, and I will start by recording findings that are not subject to appeal. In the first place, the judge found that the parents' use of drugs, and the prevalence of domestic abuse, and the father's volatile and aggressive behaviour, exposed any child in their care to a real risk of significant physical and emotional harm.
  3. The judge also made significant findings about LM, who had been removed from her parents' care in May 2016 and is now subject to a special guardianship order in favour of her paternal grandmother. The judge found that when LM was aged 21 months, she had suffered extensive bruising to her head and face as a result of at least 4 forceful blows. She rejected the parents' account of the fall downstairs. She could not say which parent was responsible, but found that they both knew the truth and had persistently lied about it, including to the court at a previous fact-finding hearing when the court had reached the same conclusion. Attempts by the parents to escape that earlier decision by giving a supposedly new account had led to the circumstances of LM's injuries being reheard within the present proceedings. As I have said, there is no appeal from these findings, which are amply substantiated by the evidence.
  4. This appeal instead relates to injuries suffered by the younger child, LR, in November 2017, when she was 11 weeks old. The circumstances in were unusual. Because of the family history, the local authority had issued proceedings at the time of her birth. An interim care order was made and the child was placed with a paternal aunt and her partner. The appellant, who is LR's paternal grandmother (I will call her "the grandmother"), lived nearby, helped with the daily care and, as the judge found, had sole care on some occasions. At the outset, the aunt had agreed to supervise parental contact at her home, but after a short time contact moved to a contact centre. The local authority's case was that the parents were also allowed to have unauthorised contact at other times.
  5. On 9 November 2017, during the course of contact at the contact centre, the mother noticed bruising to LR. She was examined at hospital and bruising was found on her abdomen and back. Subsequently, x-rays were taken and on 20 November they were reported as showing that LR had three fractured ribs. The medical evidence suggested that the injuries were due to at least two occasions of squeezing and that the timeframe was within about 3 weeks of the x-rays. The bruising was considered to be fresh when seen on 9 November.
  6. The judge found that the people with the opportunity to cause these injuries were: the aunt and her partner, the parents, the grandmother, the child's maternal grandmother and, in relation to the rib injuries, the foster carers to whom LR had been moved when the bruising was discovered. Having heard all the evidence, the judge concluded that she could not identify the perpetrator or perpetrators. She went on to find that there was no real possibility that the injuries had been caused by the aunt or her partner, or by the foster carers, or by the maternal grandmother. Her conclusion was that the mother and/or the father and/or the grandmother were responsible. The grandmother now challenges her inclusion in the pool of perpetrators.
  7. What the grandmother does not challenge is a further and damning series of findings that she had failed to protect LR from her parents by facilitating unauthorised contact on more than one occasion between 30 October and 9 November, despite knowing that they presented a risk of harm, or a finding that she had lied to professionals and the court when denying that she had allowed contact and that she had put protection of the parents ahead of protection of her granddaughter.
  8. The welfare stage of the proceedings concerning LR is due to return to the judge for IRH on 28 June and final hearing on 15 July. In the light of the serious findings of failure to protect, the grandmother has realistically withdrawn her claim to care for LR in the future. However, the judge's conclusion that she is a possible perpetrator of some or all of the child's injuries is a serious one with possible consequences for someone with eight grandchildren and no past history of any alleged mistreatment of children. It must therefore be independently sustainable.
  9. The judgment

  10. The judgment shows the methodical care with which the judge approached her task. In circumstances where a child had been seriously injured despite being placed in what was intended to be a protective family placement, she paid close attention to the nature of the family relationships and the complicated cross-currents between the adults. The judgment runs to 95 pages, over half of which relate to LR. It is comprehensive, but not a page too long. In the course of it, the judge made a careful assessment of the credibility of all of the witnesses. She found the parents to be thoroughly unconvincing and untruthful. She found the aunt, who gave evidence for three days, to be a very impressive witness who was concerned to help the court to establish how the injuries had occurred. She found the grandmother not to be a credible witness, but one who was concerned to protect her son, the father. From this standpoint, the judge embarked upon a close study of the period of weeks before the injuries were discovered. At the outset of that study, she said:
  11. "230. An analysis of these circumstances has assisted me in trying to identify how the injury to LR was caused, and by whom. It has been necessary for me to assess the reliability of the evidence of witnesses, particularly where evidence is inconsistent or contentious, and to establish whether it provides me with any insight into the likelihood of those who are in the potential pool of perpetrators behaving in ways that they have denied. In this case, that includes not only by denying that they have injured LR but also by denying that they have allowed contact that was not authorised with a person who has injured LR."

    This approach was entirely correct.

  12. It is unnecessary to chart the detailed chronology in the way that the judge did: it is enough to note that she declined the local authority's invitation to make certain findings adverse to the family (and the aunt in particular) and to the grandmother. But relevantly, she found that the grandmother had looked after LR and one of the aunt's own children for several hours on 8 November in the aunt's absence, and that it was likely that she had allowed contact to take place with the parents whenever she was caring for LR in this way. The judge found that the aunt had not caused LR's injuries and that they had been caused some time shortly before the morning of 9 November. She concluded that she could not identify the perpetrator of the injuries and went on to consider the position of each relevant individual. She considered the mother and the father individually and together, and concluded that there was a real possibility that each of them had caused the injuries. At a later stage, she carried out a similar exercise in relation to the aunt and her partner, and excluded them.
  13. The judge's consideration of the grandmother appears at paragraphs 339-341:
  14. "The grandmother as a possible perpetrator
    339. The grandmother as a trusted carer of LR had the opportunity to injure LR. She is a grandmother of 8 children and has been closely involved in their care. There is no evidence that she has caused harm to her own [four] children or any of their children in the past. However, once LR was placed with the aunt, and the father became unhappy with the arrangements for his own contact, the grandmother became conflicted. I am confident that the father will have made his unhappiness with the altered arrangements clear to his mother. He had stressed to the authorities and to his mother that he wanted his mother to show him how to look after LR. She will have been under a great deal of pressure. As a result, it is likely she allowed unauthorised contact to the father and, I find, to the mother. In doing so, the grandmother, conflicted and under pressure, behaved in a manner which is at odds with the history of how she had cared for children in the past. She has failed to put LR first by allowing the unauthorised contact and then by lying about her son's visits, and the unauthorised contact she has offered, since.
    340. On the evidence, it appears that the amount of time the grandmother was spending caring for LR did increase over the time LR was at the aunt's home LR spent short periods of time in her grandmother's sole care, to assist the aunt. On 8th November 2017, the grandmother was in charge for up to 6 hours. I am not satisfied that she was assisted by [her other daughter] for any significant time, and in any significant way, on that day. I consider it likely that the grandmother facilitated unauthorised contact to the parents. The grandmother's ability to calm the father down when he loses his temper and becomes aggressive has been evidenced but I cannot be satisfied that she would be able to calm him down in the event of him becoming agitated and frustrated in the presence of LR, or if there was an altercation between him and the mother in the presence of LR, or that she would not lose control herself."
  15. It can be seen that, except for the last eight words, these paragraphs provide significant support for a finding of failure to protect, rather than for a conclusion that the grandmother was a possible perpetrator. So it was that those representing the grandmother appropriately sought clarification in the form of two pertinent questions:
  16. a) What evidence has the court relied upon when concluding that it cannot be satisfied that the grandmother would not lose control of herself?
    b) What weight has the court given to the submissions made on the grandmother's behalf in respect of the likelihood of someone other than the parents having inflicted LR's injuries, given that the court has found that one or both of those individuals inflicted injuries on LM?
  17. In relation to the latter question, the judge replied:
  18. "I have considered this submission made on behalf of the grandmother. However, I have also taken into account that I've not been able to identify the perpetrators of the injuries inflicted on LM; that the injury suffered by LR occurred to a child of a very different age and in a completely different placement. I take into account that, sadly, injuries can sometimes be inflicted by individuals who have no history of having inflicted injuries to children (or more generally), but who find themselves unable to cope. I have in relation to each party/intervener considered whether there is a real possibility that he/she caused the injuries."
  19. Returning to the first question, the judge identified a combination of five factors in the evidence that underpinned her conclusion that she could not be satisfied that the grandmother would not lose control herself:
  20. That, then, is the judgment and the clarification.
  21. The grounds of appeal and submissions

  22. Leaving aside one ground that was not pursued, the grounds of appeal are that:
  23. 1. The conclusion that the grandmother was a possible perpetrator is not one that was open to the judge on the evidence. She was wrong:
    (a) to rely on the grandmother's mental health without any real evidence about it, either by way of report, medical records, or questioning during the course of the evidence.
    (b) to rely on the grandmother being an emotional person, or questioning her ability to look after two small children, when she had found the grandmother to have been closely involved with her eight grandchildren without any evidence of her harming them.
    (c) to treat as relevant an isolated instance of loss of temper with her adult son.
    2. The judge gave insufficient weight to her finding that the parents were responsible for injuring LM.
  24. The appeal has been argued by Mr Ashley Lord for the grandmother and Mr Simon Wilkinson for the local authority. Both made submissions of very high quality – concise, focused, realistic and fair. The other parties lodged written submissions in which the Guardian supports the local authority's position, as does the mother, while the father supports the appeal.
  25. There was agreement at the hearing before us that, although the judge did not say so in terms, her overall findings strongly pointed to her having concluded that the injuries to LR were caused on or around 8 November, which was a date when she spent several hours in the grandmother's care.
  26. Mr Lord accepts the high hurdle that he faces in challenging a conclusion of this kind, but argues that he is not seeking to go behind the judge's assessment of the witnesses. The errors in the judgment are apparent on its face. The judge's initial reasons for the pool finding were non-existent and the clarification does not withstand scrutiny. Her detailed assessments of the grandmother as a witness and a possible perpetrator at paragraphs 196-205 and 339-341 go entirely to the finding of failure to protect. Nor does her conclusion take proper account of the parents' history of child abuse and deception.
  27. Mr Lord then addressed the points made by the judge in clarification. He notes that in the first, third and fourth bullet points set out above the judge refers to the grandmother's history of depression and anxiety. He argues that this amounted to no more than speculation and that no reasonable inference could be drawn from that evidence. As to the second bullet point, the evidence gave no reason to believe that the care of two small children was beyond the capacity of an experienced parent and grandparent. As to the final bullet point, the description of the grandmother as someone who "lives on her emotions" simply arose from the evidence of the aunt that her mother can be very emotional when talking about her grandchildren and that she is not shy about expressing her emotions. The other evidence quoted in this bullet point arose from a long account of an incident at the hospital given by the grandmother herself in a police interview. We have been taken to that passage, which appears at page 285. It is difficult to see how it could on any interpretation be taken to heighten the possibility that the grandmother had injured LR. There was nothing sinister in her description of patting or jiggling the baby, nor did the judge suggest that. The time that the grandmother burst into tears came at a later point when she realised that the child was going to be taken into care. Of more significance, the occasion on which the grandmother said that she had lost her temper arose because the father was behaving so badly at the hospital – abusing the social worker and causing a fracas in the waiting room – that he was convicted of public disorder in December 2017. The grandmother's account of that in her police interview was this:
  28. "[The father] had come back and was abusing the social workers and so I handed the baby to [the mother]. I got hold of [the father], shoved him out the door. Obviously, lost my temper with [him] because he was being abusive and, like I says to him, "This is not helping the situation at all." [The aunt] was on the floor crying as she was saying, "I'm sorry, they're going to take her." So we got [the father] out, who was kicking a few bins and throwing a few leaflets about in the waiting room, and we got him… it was like a secure door, so I got him out of there… But then when we went back, they locked the door of the social workers and [social worker] said she was in fear of her life, or something, and it just got really messy."
    Not surprisingly, the judge heard evidence about this incident because of the light that it threw on the father's character. The grandmother was not asked about her description of losing her temper during the course of her evidence and nobody, including the social workers who were present, were in any way critical of her behaviour on that occasion. Accordingly, Mr Lord argues that there was nothing remarkable about that evidence that could support the possibility that the grandmother had injured the child.
  29. In relation to ground two, Mr Lord submits that the judge ought to have taken into account the broad spectrum of evidence, including her findings in relation to LM, but there is no sign that she attached any weight to them. The individual reasons given in clarification do not withstand scrutiny. The fact that it is not known which parent injured LM cannot make any difference. The fact that LM was injured at the age of 21 months and LR at the age of 11 weeks can scarcely be a real distinction. The accepted fact that anyone can injure a child does not take the matter any further.
  30. In an equally well-judged response, Mr Wilkinson did not seek to uphold the judge's conclusion on the basis of her reasoning alone. Rather, he argues that her conclusion should stand on the basis of a broader view of the matter and he emphasises four points:
  31. 1. The judge was uniquely placed to assess the credibility of the witnesses and the wide canvas of the evidence. There is a need for appellate restraint, epitomised by the Supreme Court in Re B (A Child) [2013] UKSC 33, where Lord Neuberger stated that:
    "53. ... where a trial judge has reached a conclusion on the primary facts, it is only in a rare case, such as where the conclusion was one (i) which there was no evidence to support, (ii) which was based on a misunderstanding of the evidence, or (iii) which no reasonable judge could have reached, that an appellate tribunal will interfere with it."
    Here, the judge was considering a complex case, which she had case-managed throughout. She understood the complex family dynamics and the link between credibility and perpetration. She herself said (at paragraph 24) that hearing the oral evidence had been of the greatest assistance to her. The court should be slow to interfere with her conclusions unless her reasoning was so wrong as to be unsafe.
    2. The judgment must be evaluated as a whole. It includes the very serious finding that the grandmother was not a credible witness and had given directly untruthful evidence about her son's presence in her home, and evidence that there were additional pressures on the grandmother around the time of the injuries, culminating in a long period of care on 8 November. So, says Mr Wilkinson:
    "The learned judge was thus left in a position where the appellant had not simply opportunity – the sole care of the child – but opportunity in different circumstances than which she had previously found herself, and against a background of an assessment of the appellant as a dishonest witness who had misled the court in respect of a number of crucial and important factual matters."
    3. The judge received careful submissions on the law and applied it correctly. She did not equate opportunity with real possibility, or rely on the absence of alternative explanations. Nor did she make all the findings sought by the local authority. Instead, she carried out her own analysis. She considered all the evidence and was entitled to include within it the grandmother's mental health history.
    4. The judge made a reasonable assessment of the probabilities, including the probability of LR being injured by her grandmother rather than by those who had injured LM.
  32. In the course of submissions, it became clear that the second of these four points represented the heart of Mr Wilkinson's argument. That argument places very significant weight on the judge's credibility finding in circumstances where the judge herself had not made reference to it. However, says Mr Wilkinson, she can scarcely have been immune to its influence. But he nonetheless accepted that if this was a significant aspect of the judge's thinking, she had not given herself a Lucas direction, by asking herself whether the grandmother's lies were relevant to perpetration as well as to failure to protect. There is no doubt that the judge was mindful of the Lucas principle because she referred to it twice when considering the parents' lies in relation to LM.
  33. Overall, says Mr Wilkinson, the judgment is fundamentally sound and logically coherent, and the judge's conclusions in relation to the grandmother should stand in full. However, he concluded with the sensible concession that if the appeal was allowed, the local authority would not seek a rehearing on this one issue.
  34. Conclusion

  35. This was a complex and unusual case in which, despite being placed in an apparently safe environment under an interim care order, a child suffered serious injuries that may have been caused by the very individuals from whom she should have been protected. I begin by paying tribute to the judgment as a whole. The submissions we have heard focus on one small but important corner of a decision that is not challenged in any other respect. The quality of the judgment, taken together with the proper deference that this court pays to evaluations of this kind, speaks strongly in favour of upholding the conclusion now under appeal. It can also be said that even if the grandmother's appeal succeeds, it does not relieve her of the findings of culpable failure to protect her granddaughter from serious harm, findings that are particularly heavy when she was in a position of trust. At the same time, in fairness to all concerned, the judge's conclusion about the grandmother as a possible perpetrator must stand or fall on its own merits.
  36. Given the state of the evidence, it is not surprising that the judge was unable to positively identify a perpetrator on the balance of probabilities. This court has recently considered the proper approach in such a case in Re B (Children: Uncertain Perpetrators) [2019] EWCA Civ 575. It emphasised that a decision to place a person in the pool of perpetrators is not a conventional finding of fact, but a conclusion that nevertheless has to be positively proved, and that there can be no reversal of the burden of proof. That is the way in which the judge correctly approached the matter here. The question is whether, taking the evidence as whole, her decision to move the grandmother from the list of people with an opportunity to have caused the injuries into the pool of possible perpetrators was one that was properly open to her. In considering that question, it is appropriate for this court to take the broadest view of the judge's command of the case and to resist isolated arguments based upon logic or semantics, or criticisms of the precise way in which a small part of a lengthy judgement has been expressed.
  37. Approaching matters in that way, I have nevertheless concluded that the judge's identification of the grandmother as a potential perpetrator of injury went further than the evidence allowed. Mr Lord's analysis of the limitations in the evidence relied upon by the judge is convincing. None of the five matters could properly be said to add significant support to the conclusion, and to that extent irrelevant or insubstantial matters were unduly brought into account. On the other side of the scale, the judge does not appear to have given any real weight to her findings about the character of the parents or their behaviour in relation to LM. Taking these matters together, the judgment does not in my view give adequate reasons for a conclusion of this seriousness. Mr Wilkinson's argument, however persuasive, for upholding the decision on significantly different grounds requires the matter to be considered at an unduly high level of generality; it also depends heavily on the grandmother's lack of credibility being deployed in a way that the judge did not herself deploy it, and without the safeguards that would be necessary before that could be done.
  38. For these reasons, I would allow the appeal and vary the judge's order of 25 February 2019 to the limited extent of removing the grandmother's name from paragraphs 6 and 10(d) of the second Schedule. The rest of the order remains in full effect.
  39. The test for permission to appeal

  40. I finally refer to a procedural issue. In this case, the application for permission to appeal correctly set out the relevant test under CPR r.52.6(1), namely that permission to appeal may be given only where –
  41. a. the court considers the appeal would have a real prospect of success; or
    b. there is some other compelling reason why the appeal should be heard.
  42. The same criteria govern appeals within the Family Court and commentary on the equivalent provision (FPR 30.3(7)) appears in the Family Court Practice 2018 at p.1909. When seeking permission to appeal, Mr Lord cited that commentary, which is in these terms:
  43. "Real prospect of success – There are two conflicting authorities on the meaning of a 'real prospect of success'. In NLW v ARC [2012] 2 FLR 129, FD, Mostyn J held that the 'real prospect of success' meant it was more likely than not that the appeal would be allowed at the substantive hearing: "anything less than a 50/50 threshold could only mean there was a real prospect of failure". Moor J, however, has held that a 'real prospect of success' is one that is realistic rather than fanciful, and does not mean a greater than 50/50 chance of success. … The weight of current first instance authority follows the approach of Moor J. "
  44. Several years on, this divergence of approach continues to be referred to in applications for permission to appeal to this court and to the High Court. This appeal represents an opportunity to resolve any remaining doubt. The test for the grant of permission to appeal on an application to the Court of Appeal or to the High Court or Family Court under the first limb of the relevant sub-rule is that the appeal would have a real prospect of success. As stated in Tanfern v Cameron-MacDonald (Practice Note) [2001] 1 WLR 1311 CA at [21], which itself follows Swain v Hillman [2001] 1 AER 91 CA, there must be a realistic, as opposed to fanciful, prospect of success. There is no requirement that success should be probable, or more likely than not.
  45. Lord Justice Baker:

  46. I agree.

BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII