S (A Child) , Re

This is an appeal brought by an applicant local authority against the findings of fact made at a hearing in public law children proceedings brought under the Children Act 1989 [CA 1989] which took place on 10 September 2013 in the Cambridge County Court. The proceedings concern a 12 month old baby girl who I shall call ’S’. The local authority was Cambridgeshire County Council and the other parties were S by her children’s guardian and S’s parents who were separately represented.

The precipitating circumstance for the proceedings was the clinical presentation of S on her admission to hospital on 30 April 2013. By the time the hearing took place the detail of that presentation was not in dispute. There was un-contradicted medical evidence that S had been brought to the hospital with a serious head injury. That injury can be described in simple terms as being two skull fractures, one at each side of the back of S’s head, one of which was stellate, with associated brain haemorrhage and swelling.

I have called the harm an injury for reasons which will become clear, but as the identification of the key issues in the proceedings and the terminology that was used are critical to the appeal, it needs to be understood that this was a case in which the court had to decide among other questions whether facts existed sufficient to satisfy section 31(2) CA 1989 (known as the threshold criteria) namely, whether S was suffering significant harm, whether that harm was attributable to the care given to her and whether that care was not what it would be reasonable to expect a parent to give to her. Section 31(2) reads as follows:

“31(2) A court may only make a care order or supervision order if it is satisfied -

(a)

that the child concerned is suffering, or is likely to suffer, significant harm; and

(b)

that the harm or likelihood of harm, is attributable to -

(i)

the care given to the child, or likely to be given to him if the order were not made, not being what it would be reasonable to expect a parent to give to him;

or

(ii)

the child's being beyond parental control.”

The local authority’s case in writing was that the injury had been caused by two separate blunt impacts to the child’s head or by the force occasioned in a bilateral crushing mechanism. That was colloquially described in a schedule of findings of fact as ‘non accidental injury’ and it was alleged that the injury had occurred while S was in the care of her parents. It was also alleged that one of the parents caused the injury and that the other failed to protect S from that harm. It was not alleged that both parents caused the injury nor was any other fact put in issue that would have been relevant to the question of whether the harm suffered was attributable to the care given by the parents, for example, any detail relating to the alleged perpetrator, the failure to protect or the family’s upbringing of S.

The judge concluded that S suffered significant harm while in the care of her parents and that the harm was caused by an injury. He was not satisfied that either of the parents had deliberately inflicted the injury. The local authority appeals that conclusion on the basis that in an untranscribed exchange after the judgment the judge is said to have declined to exclude an accidental cause. That discussion was not part of the judgment and has not led to any further reasons being given by the judge. In their grounds of appeal, they assert in summary that:

Each of the grounds of appeal is misconceived for the reasons I shall explain. The judge was not inconsistent in his findings which were not inconsistent with the expert medical opinion. He did not find that the injury was accidental (which would have been a contradiction in terms) nor did he find that the parents had not injured their child. The judge cannot be said to have been plainly wrong in the findings he made nor in the inferences he drew or declined to draw and his reasoning was sufficient given that he had the benefit of hearing the oral evidence of the adults who were present when the child was injured and no part of that evidence as accepted by the judge has been demonstrated to this court to be logically inconsistent with the judge’s findings. In short, the local authority pursued a case, some of which they failed to establish or have yet to establish. At the conclusion of the appeal hearing the appeal was dismissed and the proceedings were remitted to the allocated judge for a further hearing where any outstanding issues of fact that may need to be decided can be determined in the same hearing that the welfare and proportionality evaluations are conducted by the court.

This is an appeal in care proceedings in which the background problem is a failure by the parties to identify the key issues which needed to be determined by the court and then how those issues should be determined. Before this court, the local authority’s understanding of their own case and the principles to be applied where significant harm has been suffered by a child and where the mechanism for that and / or its alleged perpetration is one of the facts in issue in the proceedings, can only be described as worrying. The hearing that gave rise to the findings which are the subject of this appeal was what is known as a ‘split hearing’ where discrete findings of fact are made in a hearing which is separate from the final hearing at which the welfare of the child is evaluated and determined. It is by no means clear why that form of case management was adopted in this case. It was an inappropriate way of dealing with necessarily inter-related issues of fact and welfare.

It is common ground before this court, as can be seen from the judge’s judgment, that there were other social care issues that may be relevant to welfare which are yet to be determined. Given that the judge in the county court may be asked to make further findings, this court will of necessity be brief and somewhat circumspect in its description of the family. All that this court needs to record is that the parents had difficulty finding suitable accommodation for themselves and their child. At the time the injury occurred they were living with the paternal grandmother and her partner in the grandmother’s accommodation which was a one bedroom flat. They used the living room as their home and they negotiated the logistics of grandmother’s night shifts, father’s daytime employment and everyone’s need to have access to the facilities available in the flat.

Dealing then with the circumstances of the harm that was suffered, the judge found that S had been vaccinated by a health professional on 29 April 2013 when nothing of concern was noted. Having regard to the medical evidence about the timing of the harm, he concluded un-controversially that whatever happened it was shortly before S was taken to hospital on the evening of 30 April 2013. He also found un-controversially that S was taken to hospital by mother and grandmother’s partner who had been in the flat at the time of the injury and that the child’s father and grandmother had also been present in the flat when the injury occurred.

Given the adults who were present in the flat at the material time and in the circumstance that there was no reliable direct evidence about the mechanism or perpetration of the harm, the theoretical pool of perpetrators if the harm was not accidental included all of the adults. It was either agreed between the parties or the local authority itself chose to limit that pool so as to exclude grandmother and her partner based solely upon the fact that the parents sought to exclude them. There may have been a better legal (i.e. evidential) or assessed social care basis for that decision, but if there was, this court is unaware of the same. Given that it appears to be conceded, and in any event the judge found as a fact, that all of the adults told deliberate untruths about the circumstances in which the harm occurred and the split hearing was limited to an examination of the responsibility of the parents alone, the limitation on the pool was not tested in evidence before the judge and is an unreasoned position which on its face calls for an explanation.

It was a conceded fact before the judge that a false history of how the harm occurred had been given at the hospital. The false history was subsequently repeated in various degrees of detail to suggest that the child had been in a bouncer suspended from a door in the flat at a height of about half a metre from the ground when a strap broke, the bouncer fell and the child hit her head. That false history was discarded by the family when the child’s father voluntarily went to the police and said that the history previously given was not true. It was also discovered that the strap had been cut (it is said by grandmother) to make it look like it had snapped. A new history was then proffered by the family which was given to the judge in evidence. The evidence was that father was standing holding S in the kitchen behind mother and grandmother when S wriggled and in some way fell to the floor.

With the possible exception of grandmother’s partner, who was not said to have witnessed anything of relevance, the judge found the family’s account of what happened on the night in question including the timings to be wholly unconvincing. He regarded the grandmother as a particularly unconvincing witness and was satisfied that father, mother and grandmother had conspired to attempt to exculpate father from what had occurred.

The judge’s ultimate finding was that the grandmother and the parents had lied to the hospital and again to the police and the court i.e. both histories were untrue. He found as a fact that the court had not been told the truth about what had occurred. He found as a fact that something happened to S in the care of her parents (i.e. that the harm was not a true accident) but did not find that either of the parents deliberately injured their child. Having read and heard the evidence he was not satisfied on the balance of probabilities that either parent had deliberately inflicted the injury. That was not a positive finding that exculpated the parents or indeed any other adult. The case management order which purported to record the findings is erroneous if it was intended to suggest otherwise.

The judge’s conclusion left open the question of how the injury occurred. The judge put down a marker to himself for the future welfare hearing that what had happened “may well involve negligence”. It is clear from the terms of the judgment that he did not find that the parents or any of the adults had been negligent but he opened up that possibility no doubt for future examination in the light of any other evidence that the court may consider at a subsequent hearing.

Nothing in this judgment or the analysis above (which is restricted to the particular facts of this case) should be taken to detract from the principles of binary fact finding as explained by Lord Hoffman and Baroness Hale of Richmond in In re B (Children) [2008] UKHL 35 at [2], [13] and [32]: a fact in issue is treated as either having happened if it is proved to the civil standard of proof or not having happened if it is not.

It may be obvious to the interested bystander that there was a theoretical range of possibilities relating to what had happened to S: from accidental harm through to deliberate infliction of injury, but that was not the way the case developed in evidence. The neuroradiologist who gave evidence to the court gave his opinion about the range of possibilities in the following terms:

“The skull fracture can occur as a result of accidental injury. In this context fracture most commonly arises from a fall from a carer’s arms and results in an un-displaced unilateral fracture most often of parietal or occipital bone. Skull fractures may also occur as a result of impact in the context of non-accidental head trauma. If the injury is non-accidental, a fracture may also be displaced, comminuted, involves (sic) more than one bone or if there are multiple fractures in the same bone, it is more likely that the injury is non-accidental in origin”

The local authority chose to pursue deliberate infliction of injury with the witnesses at the hearing and both at that hearing and before this court equated the term non-accidental injury with infliction. The local authority submitted that infliction was not necessarily deliberate infliction, but it is by no means clear that anyone else defined the terminology in that way and that characterisation of the evidence ignored the statutory formulation that the local authority needed to prove. This court was helpfully taken by counsel to the questions put to the witnesses which appear to demonstrate that the local authority assumed that ‘non-accidental injury’ if proved would be deliberate. They did not adequately explore the circumstances of the adults' care other than to cast doubt on both of the histories that had been given nor did they explore what has come to be known as the attributability element of section 31(2) of the Act. Whether there had been reckless or negligent acts and omissions or what the neuroradiologist meant by the terminology that he had used or the causative mechanisms he had described remain in doubt.

This court has sympathy both with the judge and the advocates in a situation where the direct evidence from those who were present is false and the local authority rightly assumed the burden of demonstrating that. In doing so, however, they appear to have lost sight of the rest of the case and of the statutory formulation in section 31(2) of the Act.

The term ‘non-accidental injury’ may be a term of art used by clinicians as a shorthand and I make no criticism of its use but it is a ‘catch-all’ for everything that is not an accident. It is also a tautology: the true distinction is between an accident which is unexpected and unintentional and an injury which involves an element of wrong. That element of wrong may involve a lack of care and / or an intent of a greater or lesser degree that may amount to negligence, recklessness or deliberate infliction. While an analysis of that kind may be helpful to distinguish deliberate infliction from, say, negligence, it is unnecessary in any consideration of whether the threshold criteria are satisfied because what the statute requires is something different namely, findings of fact that at least satisfy the significant harm, attributability and objective standard of care elements of section 31(2).

The court's function is to make the findings of fact that it is able on the evidence and then analyse those findings against the statutory formulation. The gloss imported by the use of unexplained legal, clinical or colloquial terms is not helpful to that exercise nor is it necessary for the purposes of section 31(2) to characterise the fact of what happened as negligence, recklessness or in any other way. Just as non-accidental injury is a tautology, 'accidental injury' is an oxymoron that is unhelpful as a description. If the term was used during the discussion after the judgment had been given as a description of one of the possibilities of how the harm had been caused, then it should not have been; it being a contradiction in terms. If, as is often the case when a clinical expert describes harm as being a ‘non-accidental injury’, there is a range of factual possibilities, those possibilities should be explored with the expert and the witnesses so that the court can understand which, if any, described mechanism is compatible with the presentation of harm.

The threshold is not concerned with intent or blame; it is concerned with whether the objective standard of care which it would be reasonable to expect for the child in question has not been provided so that the harm suffered is attributable to the care actually provided. The judge is not limited to the way the case is put by the local authority but if options are not adequately explored a judge may find a vital piece of the jigsaw missing when s/he comes to look at all the evidence in the round.

This court has not been addressed about the volume of guidance in materials issued by professional bodies (including the medical Royal Colleges) and Government which makes reference to ‘non-accidental injury’. Indeed, counsel for the local authority went so far as to submit that none of that material provided a generally accepted medical or legal definition of the term. Whether that is right or not, it is not necessary for this court to analyse that material because all that is required in a case of this kind is for the court, legal practitioners and experts to have regard to the statutory formulation with which the court is concerned. If other terminology is used in evidence its meaning should be precisely ascertained so that the court knows what is being alleged and advised.

In this case, the judge was careful to sever the question of whether the harm was an example of a true accident i.e. a chance happening that is by definition unexpected and unintentional, from the question of attributability which in the circumstances of this case was said to include perpetration of harm and lack of protection from harm. The local authority’s case was of (deliberate) infliction by one or the other parent and that was the case they put. They did not succeed in establishing that case. No other possibilities were sufficiently examined to enable the judge to make conclusions upon them. The medical expert had left open the possibility of an accidental cause, albeit that it was unlikely and in that circumstance anything between accident and deliberate infliction must also have been possible. Accordingly, there was no inconsistency between the findings and the evidence and the local authority’s description of the findings in the grounds of appeal are misconceived.

The judge examined what were conceded before him to be the false explanations for the injury given by those who took S to the hospital which were subsequently repeated by other members of the family and also what was said to be the true history of the events of the night in question which he decided was also false. He directed himself to consider that there may be innocent explanations or explanations that do not attract responsibility for telling an untruth about a fact in issue: the so called Lucas direction (R v Lucas [1981] 1 QB 720). Far from failing to consider drawing an inference of attributability from the repeated lies that were told, the judge overtly considered the options and having given himself a Lucas direction concluded that the lies were more likely to be related to their inexperience as parents. Having heard the parents in evidence the judge declined to draw the inference that they had deliberately inflicted injury but did draw the inference that they had failed to protect S by delaying the obtaining of medical treatment.

In its review of findings of fact which are challenged, the appellate court’s approach is to consider whether they are plainly wrong. That test has recently been re-affirmed by McFarlane LJ in Re A (Children) [2013] EWCA Civ 1026 at [34] and by Sir James Munby P giving the judgment of the court in Re B-S (Children) [2013] EWCA Civ 1146 at [77]. That test accords with the judgments of the Supreme Court in In the matter of B (A Child) [2013] UKSC 33: see, for example, Lord Kerr at [108]: “unless the finding is unsupportable on any objective basis it will be immune from review”. In this case, the judge’s findings and inferences cannot be said to be plainly wrong. They are conclusions based in evidence heard and in the judge’s evaluation of credibility and reliability - with which this court will be hesitant to interfere (see Piglowska v Piglowski [1999] 1 WLR 1360 at 1372 per Lord Hoffman).

For these reasons, the judge’s findings ought not to be interfered with by this court. Aside from the local authority’s misunderstanding about the nature and extent of the findings they sought and those made by the court, the case raises yet again issues of case management relating to split hearings which ought to be addressed given that the social care context was missing from the consideration of the pool of perpetrators and from any consideration of factors that may have caused secondary facts to be found from which an inference of primary fact could have been made.

It is by no means clear why it was thought appropriate to have a ‘split hearing’ where discrete facts are severed off from their welfare context. Unless the basis for such a decision is reasoned so that the inevitable delay is justified it will be wrong in principle in public law children proceedings. Even where it is asserted that delay will not be occasioned, the use of split hearings must be confined to those cases where there is a stark or discrete issue to be determined and an early conclusion on that issue will enable the substantive determination (i.e. whether a statutory order is necessary) to be made more expeditiously. The reasons for this are obvious: to remove consideration by the court of the background and contextual circumstances including factors that are relevant to the credibility of witnesses, the reliability of evidence and the section 1(3) CA 1989 welfare factors such as capability and risk, deprives the court of the very material (i.e. secondary facts) upon which findings as to primary fact and social welfare context are often based and tends to undermine the safety of the findings thereby made. It may also adversely impinge on the subsequent welfare and proportionality evaluations by the court as circumstances change and memories fade of the detail and nuances of the evidence that was given weeks or months before.

I ought to emphasise for the avoidance of doubt that although parallels can be drawn between the use of fact finding hearings in public and private law children proceedings, the appropriate and measured use of fact finding hearings in private law proceedings which are often safety cases, for example involving recent domestic abuse between parents, are not the subject of this court’s consideration in this judgment. An example of this court’s guidance in relation to those proceedings can be found in In the matter of C (Children) [2009] EWCA Civ 994. In private law proceedings it is the court that is defining an aspect of parental responsibility in its determination of the arrangements that are put in place for the child and findings of fact are appropriate, where necessary, to inform that process by reference to the factors in section 1(3) of the 1989 Act and in particular where safety issues have arisen which justify the court’s interference with the article 8 ECHR rights of the family members. In public law children cases where a care order is in issue, the court is being asked to sanction an agency of the state, namely the local authority, being permitted to exercise parental responsibility for a child. The jurisdiction in the court to undertake that task has to be based upon the existence of facts (primary and / or secondary) that satisfy the threshold in section 31 CA 1989. Accordingly, concessions or findings of fact relevant to the threshold question will always be necessary in public law cases alongside such further findings of fact as are necessary to inform the welfare evaluation.

It ought to be recollected that split hearings became fashionable as a means of expediting the most simple cases where there was only one factual issue to be decided and where the threshold for jurisdiction in section 31 CA 1989 would not be satisfied if a finding could not be made thereby concluding the proceedings (see Re S (A Child) [1996] 2 FLR 773 at 775B per Bracewell J). Over time, they also came to be used for the most complex medical causation cases where death or very serious medical issues had arisen and where an accurate medical diagnosis was integral to the future care of the child concerned. For almost all other cases, the procedure is inappropriate. The oft repeated but erroneous justification for them that a split hearing enables a social care assessment to be undertaken is simply poor social work and forensic practice. The justification comes from an era before the present Rules and Practice Directions came into force and can safely be discounted in public law children proceedings save in the most exceptional case.

Social work assessments are not contingent on facts being identified and found to the civil standard (see, for example Oldham MBC v GW & PW [2007] EWHC 136 (Fam), [2007] 2 FLR 597 and Re S (Sexual Abuse Allegations: Local Authority Response) [2001] EWHC Admin 334, [2001] 2 FLR 776 per Scott-Baker J at [34] and [35]). That is the function of the court not a social worker (Dingley v Chief Constable of Strathclyde Police [2000] UKHC 14 per Lord Hope of Craighead at [120] and [122]). Social work assessments are based upon their own professional methodology like any other form of professional risk assessment. In care cases, an appropriate social work assessment and a Cafcass analysis should be undertaken at the earliest possible opportunity to identify relevant background circumstances and context. In so far as it is necessary to express a risk formulation as a precursor to an analysis or a recommendation to the court, that can be done by basing the same on each of the alternative factual scenarios that the court is being asked to consider (see, for example, In the matter of W (Children) [2009] EWCA Civ 644 at [33]).

It may be helpful to highlight the fact that a decision to undertake a split hearing is a case management decision to which Part 1 of the Family Procedure Rules 2010 [FPR 2010] and Pilot Practice Direction 12A ‘Care, Supervision and Other Part 4 Proceedings: Guide to Case Management (the PLO)’ apply. A split hearing is only justifiable where the delay occasioned is in furtherance of the overriding objective in rule 1 FPR 2010, that is:

On the alleged facts of this case, there was no discrete issue which was appropriate for trial without its social or welfare context and delay was the inevitable consequence of the decision to have a split hearing. Given that by rule 1.3 FPR 2010 the parties have a duty to help the court to further the overriding objective, it is all the more surprising that one of the submissions made to this court was that a split hearing was inappropriate. That professional analysis should have been offered to the court below. The benefits and detriments of such a course, if proposed, should be analysed by the children’s guardian. In future, a decision to undertake a split hearing should be reasoned in court at the case management hearing and the reasons should be recorded on the face of the Case Management Order alongside what has always been the good practice of the court which is to settle the issue to be tried on the face of the order.

I agree.

I also agree.