In a Family case seeking to identity a perpetrator, in circumstances where children had suffered significant harm as a result of alleged ill-treatment, the Court of Appeal, Civil Division, provided some guidance as to the correct approach to the identification of a perpetrator as a result of alleged ill-treatment from a pool of perpetrators. It also allowed the father’s appeal due to the lack of proper analysis of the evidence.

[2019] All ER (D) 49 (Apr)

*Re B (children: uncertain perpetrator)

[2019] EWCA Civ 575


Court of Appeal, Civil Division


King, Lindblom and Peter Jackson LJJ


4 April 2019


Family proceedings – Evidence – Pool of perpetrator 

When three girls, Z aged 10, A aged 8 and M aged 5, in a family of four children, were found to be infected with gonorrhoea, the police exercised their powers of protection in relation to all four children and placed them in foster care, where they remained. Both parents were arrested on suspicion of sexual assault and neglect. Each denied the allegations against them. The local authority issued proceedings and obtained emergency protection orders. Interim care orders were later made. In a hearing to determine the identification of the perpetrator, the judge held that the father was within a pool of possible perpetrators with other unknown males who might have had access to the children, or at least one of them, including the two young men in the family home. The judge finally held that although it could not be said that the father was responsible for the infection of the children, he was not able to be excluded and there had to remain a real possibility of him having caused the infection in some way. The father appealed that finding arguing that the judge had applied the wrong legal test, that the decision was internally inconsistent, that there were significant gaps in the evidence, that proper consideration had not been given to such evidence as there was, and that in the end, the finding that the father was ‘in the pool’ was unfair and has no meaningful forensic value for the welfare stage of the proceedings.

Appeal allowed.

Issues and decisions

(1) What was the proper approach to the identification of a perpetrator in circumstances where children had suffered significant harm as a result of alleged ill-treatment from a pool of perpetrators.

The concept of ‘the pool’ did not arise in the normal run of cases where the relevant allegation could be proved to the civil standard against an individual or individuals in the normal way. The concept of ‘the pool’ sought to strike a fair balance between the rights of the individual, including those of the child, and the importance of child protection. It was a means of satisfying the attributable threshold condition that only arose where the court was satisfied that there had been significant harm arising from ill-treatment and where the only ‘unknown’ was which of a number of persons was responsible (see [46] of the judgment).

A decision to place a person within the pool of perpetrators was not a finding of fact in the conventional sense. The person was not a proven perpetrator but a possible perpetrator. That conclusion was then carried forward to the welfare stage, when the court would consider the strength of the possibility that the person was involved as part of the overall circumstances of the case. At the same time, it would keep firmly in mind that the persons in the pool had not been shown to be responsible for the child’s injuries. That conclusion presented the court with a particularly difficult problem. Experience bore that out particularly where a child had suffered very grave harm from someone within a pool of perpetrators (see [47] of the judgment).

The starting point was s 31(2) Children Act 1989 (ChA 1989), which contained the threshold conditions for statutory intervention. The concept of the pool of perpetrators was to encroach only to the minimum extent necessary upon the general principles underpinning ChA 1989, s 31(2). It did not alter the general rule on the burden of proof. Where there were a number of people who might have caused the harm, it was for the local authority to show that in relation to each of them there was a real possibility that they had done. No one could be placed into the pool unless that had been shown. That was why it was always misleading to refer to ‘exclusion from the pool’ (see [38], [48] of the judgment).

The court should first consider whether there was a ‘list’ of people who had the opportunity to cause the injury. It should then consider whether it could identify the actual perpetrator on the balance of probability and should seek, but not strain, to do so. Only if it could not identify the perpetrator to the civil standard of proof should it go on to ask in respect of those on the list was there a  likelihood or real possibility that A or B or C was the perpetrator or a perpetrator of the inflicted injuries’.  Only if there was should A or B or C be placed into the ‘pool’ (see [49] of the judgment).

The concept of a pool of perpetrators was a permissible means of satisfying that the threshold was forged in cases concerning individuals who were ‘carers’. The extension did not stretch to ‘anyone who had even a fleeting contact with the child’ in circumstances where there was the opportunity to cause injuries. Nor had it extended to harm caused by someone outside the home or family unless it would have been reasonable to expect a parent to have prevented it (see [50] of the judgment).

Where there were two, three or four known individuals from whom any risk to the child had to have come. The position of each individual was then investigated and compared. To assess the likelihood of harm having been caused by A or B or C, one needed as much information as possible about each of them in order to make the decision about which if any of them should be placed in the pool. So, where there was an imbalance of information about some individuals in comparison to others, particular care needed to be taken to ensure that the imbalance did not distort the assessment of the possibilities. The same might be said where the list of individuals had been whittled down to a pool of one named individual alongside others who had not been similarly identified. That might be unlikely, but not impossible. Here it had to be shown that there genuinely was a pool of perpetrators and not just a pool of one by default (see [51] of the judgment).

As part of the court’s normal case-management responsibilities it should  ensure: (i) that a list of possible perpetrators was created, and (ii) that directions were given for the local authority to gather (either itself or through other agencies) all relevant information about and from those individuals, and (iii) that those against whom allegations had been made had been given the opportunity to be heard (see [52] of the judgment).

NH v A County Council; Re D (children) (non-accidental injury) [2009] All ER (D) 30 (Jun) considered.

(2) Whether the decision had been correct.

There had been fundamental shortcomings in the present case as the very serious allegations had not received the thorough investigation that was required and faced with that unsatisfactory state of affairs, the judge’s analysis had fallen short. The matter had to be reheard (see [53], [61] of the judgment).

Firstly, the judge had wrongly started from the position, and effectively with the presumption, that the father was in the pool of perpetrators and that the question was whether he should be removed from it. That had effectively put the burden on the father to show that there was no real possibility that he had abused his children. The principle was that judges should not strain to identify a perpetrator, not that they should not strain to exclude a person from the pool. By ‘not straining to exclude’ the judge was making the father’s task in extricating himself from the pool all the harder (see [54] of the judgment).

Secondly, the judge had not required the local authority to make out a positive case before reaching a conclusion in the light of all the circumstances. Rather, he had reached his decision on the narrow but important basis that the children had the infection and the father had the opportunity. He had failed to give weight to evidence in favour of the father and had not made any adverse finding about the father’s credibility. On an issue of such importance, the court had to assess a key witness’s evidence and, if it was not accepted, explain why that was so. The judgment had not told the father, who had given ostensibly credible evidence that he had neither had gonorrhoea nor abused his children, why he was apparently disbelieved on both of these matters (see [55] of the judgment).

Thirdly, the assessment of the evidence was unsatisfactorily constrained by the gaps in the evidence and the artificial decision to focus on the issue of infection without considering other aspects of the asserted threshold. The case had included allegations against other occupants of the property but it had not been evidentially pursued. In terms of art 6 of the European Convention on Human Rights, that was presumptively unfair to them (see [56] of the judgment).

Fourthly, there was a possible unexplained inconsistency between the judge’s acceptance that the initial infection had been transmitted sexually and his acceptance that the father might not have infected all the children directly. That conclusion either raised a doubt about the finding of adult sexual transmission or it supposed the possibility of sexualised behaviour between the children, for which there was no direct or indirect evidence. If the court had been minded to accept the father’s evidence, it might have been open to reviewing any preliminary conclusion it had reached about the method of transmission and whether it had been sexual or not (see [57] of the judgment).

Sixthly, the Judge included within the pool of perpetrators ‘other unknown males who may have had access to the children’. An unknown (or, more accurately, an unidentified) person could not be placed in the pool of perpetrators. In the case, there was no positive evidence to support the real possibility that the infection was transmitted by an unknown person and the judge had not explained how he had drawn an inference to that effect. The basis for this open-ended expansion of the pool of perpetrators was unclear, even assuming for the present that it was permissible in principle (see [59] of the judgment).

Finally, the judge’s threshold finding in the case had a very low forensic value and would be unusually difficult to interpret at the welfare stage. The judge had been aware of the gaps in the evidence but he had not given any weight to them (see [60] of the judgment).

Chris Stevenson (instructed by Avadis and Co Solicitors) for the appellant father

Sally Bradley (instructed by London Borough of Islington) for the respondent Local Authority

Saiqa Chaudhry (instructed by Steel and Shamash) for the respondent Mother.

Caroline Budden (instructed by TV Edwards Solicitors) for the respondent children through their Guardian.

Tara Psaila   Barrister.