WHERE care proceedings are brought in respect of a child who is the victim of non-accidental injury inflicted by a parent, a care or supervision order may be made even where it cannot be established which parent was responsible. The child is suffering significant harm attributable to his not receiving the care it would have been reasonable for a parent to give to him (Lancashire C.C. v. B. [2000] 2 A.C. 147). But where does this leave a sibling or half-sibling also at risk? The Children Act 1989, section 31(2) provided expressly for the protection of unharmed children. It incorporated a new element in the threshold for state intervention where the child concerned is likely to suffer significant harm. However the House of Lords has pronounced that, in relying on this prospective limb of the statutory threshold, the local authority must prove the allegations on which it is relying. It must establish that there is a real possibility of such future harm by proving the key allegations to the ordinary standard of balance of probabilities (Re H. (Minors) (Sexual Abuse) [1996] A.C. 563; Re B. (Children: Care Proceedings: Standard of Proof) [2008] UKHL 35, [2009] 1 A.C. 11). In Re S-B. (Children) [2009] UKSC 17, [2010] 1 A.C. 678, Lady Hale made remarks to the effect that the threshold will not be crossed in relation to a second or subsequent child unless the identity of the perpetrator of the first child's injuries is proved.
Re F. (A Child) [2011] EWCA Civ 258 concerned a baby boy. His parents were engaged to be married, but the father had a history. In 2004 a care order was made in relation to an earlier child of his by a different mother. That child had been injured either by the mother or by him but it had not been established which it was. Now in his new relationship, the father did not seek to conceal his past. When his new partner fell pregnant, the couple notified the relevant local authority of this and of the 2004 proceedings in which the father was found to be in the “pool of perpetrators” viz a possible perpetrator but not a proven one.
The authority indicated that it would apply for an interim care order after the baby's birth. Its plan was that the child should live with the mother; the father should leave and undergo a risk assessment. This was not acceptable to the parents and, within a day or two of the birth, the authority applied unsuccessfully for an interim care order. It relied solely on the previous proceedings giving rise to reasonable grounds for believing that the baby was likely to suffer significant harm if left in the care of the parents. The judge gave permission to appeal. The authority, acknowledging that its appeal must fail on the state of the authorities, sought to persuade the Court of Appeal to grant permission to appeal to the Supreme Court. Refusing permission, Wilson LJ declined to hold that there should be a relaxation of the principle of proven factual foundation in uncertain perpetrator cases. For him the requirement was a “bulwark against the state's removal of a child from his family” which he considered “very precious”.
All child protection cases involve a delicate balance between striving to ensure safety of children while not violating rights of families to be free from unjustified state intrusion. It might be argued that the interests of parents in the pool of perpetrators can be adequately protected at the welfare stage of care proceedings when, the threshold having been crossed, the court must decide what order to make. So, the argument goes, full account can be taken of the evidence concerning the possibilities of perpetration at that stage. But, at the threshold stage, possible perpetration should be sufficient because the second child is indubitably still at risk. Precisely what weight should be given at the welfare hearing to a parent being in the pool in relation to an injured child is something which Wilson L.J. admitted had exercised him for some years. But this issue becomes yet more complex when the question is what weight should be given to that fact in relation to a second child where the threshold is crossed for other reasons in relation to that child.
The Court of Appeal declined to give guidance on this and the issue must surely be canvassed before the Supreme Court. This will not now happen in Re F. as the authority has since obtained an interim care order on the quite different basis that the parents were abusing drugs while caring for the child. But return to the Supreme Court in another case it must, especially in the light of that Court's ruling in Re S-B. that, highly desirable though it is to have a clear finding on perpetration, judges should not strain to make it. Neither should they comment in percentage terms about the relative chances of perpetration if they genuinely cannot make an identification.
Despite the overwhelming desire not to leave children at risk, there are formidable objections to abandoning the requirement of proof of perpetration when that alone is the basis of the proceedings. Illogical though it may seem, a distinction may need to be made between parents who remain together and those who separate and, more often than not, between siblings and half-siblings. Let us suppose that the parents stay together. A younger sibling is already in the household or is born into it. In these circumstances the threshold should be crossed. What we have here is proof that one of two parents, both still in the household, has injured a child. It should be possible to rely on this despite no finding of which one it was. The real possibility of harm to the second child does have a factual foundation. What if instead, as in Re F., the parents separate and, say, the mother acquires a new partner? She has another child, half-sibling to the injured child. In this situation the certainty of a risk has been replaced by the possibility of one. There are sound policy reasons why the threshold should not be crossed in these circumstances if the only finding against the mother is that she was a possible perpetrator. Must she forever run the gauntlet of care proceedings for her second, third, fourth or fifth child? This is surely not a proportionate response to risk, especially if we postulate that she was indeed the “innocent” party, leaving aside any issues of failure to protect. The present position, that the authority must have other concerns which enable it to cross the threshold, is defensible.
The current law is causing consternation but this may be an over-reaction. First, in the great majority of cases it will be possible for a judge to identify the perpetrator on the balance of probabilities. Secondly, as Re F. demonstrates, it will be a comparatively unusual case in which the only fact relied upon is a parent being in a pool of perpetrators. There will usually be other substantial causes for concern. It is right that the authority should have the onus of demonstrating what they are.
