Children and Families Bill

Debate between Lord Lloyd of Berwick and Lord Brown of Eaton-under-Heywood

Monday 21st October 2013 Grand Committee

[Lord Lloyd of Berwick]

My Lords, I said that the two questions posed by Section 31 should be capable of being answered without too much help from lawyers, but that has not proved to be the case. In the 27 years since 1986, the section has been considered on no fewer than eight occasions in our highest court, and on two occasions already this year. On one of those occasions it was said:

“This is unfortunate, especially in an area of the law which has to be applied on a daily basis by courts at all levels, and in which clarity is therefore of particular importance”.

I think noble Lords would say amen to that. Happily, there is one word in the section on the meaning of which everyone is agreed, and that is the word “likely” in subsection (1). It does not mean more likely than not. It means only that there is a real possibility of harm to a child or, as one judge put it, a possibility that cannot sensibly be ignored.

Perhaps I may paraphrase Section 32(1) again. It states that a child may be taken into care only if, first, it is suffering significant harm or, secondly, there is the real possibility that it will suffer significant harm. I shall not repeat all the wording. In other words, there are two separate conditions which may trigger the threshold, one relating to the present and the other relating to the future. When the case is put on the basis of present harm—for example, that the child is being physically or sexually abused by its father—it will be necessary to prove that fact on a balance of probability. Anything less than that would be unfair on the father and, indeed, on both parents. This is so also where the case is that the child has suffered serious injury and it is uncertain whether the injury was inflicted by the mother. It will be necessary to prove on a balance of probability that the child has indeed been injured and that the injury was inflicted by either the father or the mother, or both, but it will not be necessary at the threshold stage to decide which parent it was. That will be decided, if it can be decided at all, at the welfare stage on all the evidence which will then be available. The same principle of the threshold test also applies in relation to any unharmed child of the family.

So far, all is plain sailing. The difficulty arises when the parents separate. Let us suppose that the father goes off to live with another woman who already has a child of her own of the same age as the injured child. Is the threshold satisfied in relation to that child? Common sense would suggest that it is. There is a 50% chance on the proved facts that it was the father who injured the first child, who we will call child A. There must be at least a serious possibility that he will also injure child B—a possibility which, I repeat, cannot sensibly be ignored. If so, the threshold would be satisfied in relation to child B as well as to child A. However, the Supreme Court has held in a very recent case, Re J (Children), that that is not so. The Court has held that a serious possibility that it was the father who inflicted the injury is not enough. In order to satisfy the threshold in relation to child B, it will be for the local authority to prove on a balance of probability that it was the father and not the mother who injured child A. Since on the assumed facts that could not be done, child B would remain at risk.

I suggest that this cannot have been what Parliament intended when enacting Section 32(1), otherwise why did Parliament include the word “likely” as the alternative ground on which the threshold may be satisfied? The matter can be tested in this way by assuming that the mother is also now living with another man and has had a child which we shall call child C. Does child C also have to remain at risk because it cannot be proved on a balance of probability that it was the mother rather than the father who inflicted the harm on child A? The noble and learned Lord, Lord Nicholls, who gave the leading judgments in the three initial cases on Section 31 in the House of Lords, described such a result as,

“grotesque because it would mean the court would proceed on the footing that neither parent represents a risk even though one or other of them was”,

responsible for the harm in question.

How then, one may ask, has the Supreme Court in Re J arrived at a different conclusion that, as the noble and learned Lord has said, on the face of it is “grotesque”. How can two of the judges in the Supreme Court have held that the injury to child A in such a case is logically irrelevant in deciding whether child B and child C are at risk and must therefore be disregarded altogether?

The answer to how the Supreme Court can have reached that conclusion is to be found in the judgment of Lord Justice McFarlane in the same case but in the court below. Lord Justice McFarlane is one of our most experienced Family Division judges, and is the author of one of the leading textbooks in this field. It was he who gave the leading judgment in the Court of Appeal in Re J. He went through all the House of Lords and Supreme Court decisions going back to 1996, and he showed that a clear distinction is drawn in the cases between those where the question is whether any harm has been proved at all and those cases where harm has been proved but the perpetrator of the harm is uncertain. That is the very distinction drawn by the noble and learned Lord, Lord Nicholls, in the House of Lords case that I have already quoted.

Somehow though, that distinction was overlooked in later cases. Lord Justice McFarlane makes no secret of the fact that he favours the approach of the noble and learned Lord, Lord Nicholls, and would therefore have allowed the appeal in Re J if he could. However, subsequent decisions in the Supreme Court meant of course that his hands were tied. So the Court of Appeal took the unusual course of dismissing the appeal but itself giving leave to appeal to the Supreme Court, thereby, one might think, inviting the Supreme Court to have another look at this problem. Unfortunately, as I see it, the Supreme Court simply came up with the same answer again.

Lord Justice McFarlane’s judgment is long and detailed but his conclusion is clear, concise and very relevant in this context. It is contained in a single page of the Law Reports, which I have had copied, and the Committee may find it helpful to read his conclusion when considering this amendment. Copies are available on the table by the door.

I come to the decision of the Supreme Court itself in Re J. Is it open to us to take a different view? If so, is it wise for us to do so? To both those questions I would answer yes, for three reasons. First, three of the judges in the Supreme Court were themselves attracted by the argument that the approach in these cases has become much too complicated and that this is having unfortunate consequences. Secondly, the decision in the Supreme Court has been subjected to a hail of criticism in lengthy articles by Professor Mary Hayes and Stephen Gilmore, appearing in Family Law. There is not the slightest reason to doubt that, as they point out, the decision is causing real concern, if not consternation, among social workers and local authorities who have to apply Section 32 in practice. Thirdly, the decision in the Supreme Court hardly does justice to Lord Justice McFarlane’s decision in the Court of Appeal; indeed, it is scarcely even mentioned.

There is another reason for accepting this amendment. I am not seeking to amend the wording of Section 32(1) itself; the wording is fine and has stood the test of time. It is only the interpretation of that section that needs correcting, and that is what the amendment seeks to do. Its intended purpose is to clarify, and above all to simplify, the approach in cases of the kind that I have described where the harm has been proved on the balance of probabilities but the court cannot make a finding on the evidence whether it was the father or the mother who inflicted that harm. A judge of great experience in the Family Division said that that is the sort of case that occurs very often—“commonplace”, I think he said—in practice. In such cases, if the amendment were accepted, both parents would be placed in what is called a pool of possible perpetrators, thus enabling the case to proceed to the next stage, the welfare stage, where a decision could be made.

I refer to a “pool” because that is the term used by those who read these cases, or a “list”, as it is called in the amendment. Why does one have to have a pool or a list? The reason is quite simple: in one case, which has actually occurred in practice, there was a third possible perpetrator. In addition to the parents of the child in question, there was a childminder who also had a child of her own of about the same age. In such a case, it obviously makes sense that the childminder should be included in the pool of possible perpetrators, thus enabling that child to be protected should it become necessary. I hope that this has done something to clarify the purpose of the amendment and I beg to move.

[Lord Brown of Eaton-under-Heywood (Non-Afl)]

My Lords, I support the amendment as strongly as I may. The critical consideration to keep in mind here, as the noble and learned Lord, Lord Lloyd, has explained, is that what we are concerned about today is a threshold provision. The amendment would mean simply that in a small but very important additional category of cases, the court would have the jurisdiction and the power to investigate the case in depth and to consider whether in all the circumstances it should then make a care order or supervision order for the child’s protection. The small category of additional cases—again, the noble and learned Lord has explained this—is where it is established that some other child has already suffered significant harm, perhaps has even been killed, but the local authority concerned about some other child can demonstrate only the possibility, rather than the actual probability, that the perpetrator of that harm was someone who is now caring for the child in question—the child, that is, whose safety is presently under consideration.

As it happens, I was not in any of the string of cases in which the question of the true interpretation of Section 31(2) of the Children Act 1989 has arisen in recent years. Whether in the original House of Lords case I should have agreed with the majority view or with the dissenting minority view of the noble and learned Lord, Lord Browne-Wilkinson, and indeed of my noble and learned friend Lord Lloyd, does not matter. It is unnecessary to decide now which was the better interpretation of the language that Parliament originally enacted in 1989.

What is clear, as again the noble and learned Lord, Lord Lloyd, has explained, is that several judges who have had to grapple with this point, even if they felt bound by the original majority’s decision, have expressed serious misgivings about the consequences of that interpretation. In the case last year, Re J, to which again my noble and learned friend has referred, both the noble and learned Lord, Lord Judge, then the Lord Chief Justice and now a Member of this House, and the noble and learned Lord, Lord Neuberger, then Master of the Rolls and now President of the Supreme Court, agreed with Lord Justice McFarlane. His judgment expressed his trouble with the interpretation given to this section and described it,

“as a cause of concern amongst child protection agencies”.

What is certain is that the clause as originally enacted was not clear enough as to what Parliament then intended. The amendment of the noble and learned Lord, Lord Lloyd, or some comparable draft, would make it plain. It would solve the real and recurrent difficulty that this vitally important part of the law has got itself into, and it would produce a result that for my part I believe we should be striving for, which is to open the gateway to the court.

I repeat, this is only a threshold provision which would apply whenever a child is found to be at risk of being harmed, as must surely be the case when one of the caring parents is shown to have been a possible perpetrator of serious harm in the past. To anybody who is concerned that the court, following this amendment, would too readily take children away from a parent who only might have harmed some other child, I would say this is absolutely not the case. To quote subsection (2)(2A) of the proposed amendment,

“to infer that a child is likely to suffer significant harm”,

is to infer no more than that there is a risk of that child being harmed as surely there is if there is a real possibility that its carer has significantly harmed some other child. Crucially, it would then remain for the court, looking at all the facts of the case, to decide whether, under Section 1 of the 1989 Act, the child’s welfare is indeed best served by making a care or supervision order. I support the amendment.