Introduction

In order to examine the decision of the appellate Supreme Court holistically, it is necessary to set out the policy objectives which govern the law on child protection in the United Kingdom (U.K.). The policy interplay can thus be summarised as follows: before the state can intervene in family life to protect children, it must not only establish that the intervention is in the interest of the child’s welfare but that it is necessary due to the risk of harm to the child.² Therefore, in child protection disputes, a threshold of intervention³ is necessitated before the welfare principle⁴ is applied.

The applicable legislation is Section 31(2) of the Children Act 1989 which states that:
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 rationale for this can be gleaned from Luckock et al. who state that the ‘notions of family privacy and the minimal State inform the role given to the courts by the Children Act 1989’.⁵

However, Fortin has argued that this policy is misguided because ‘the courts have allowed a strict legalism to creep into the “proof” stage of care proceedings, at the expense of the child’.⁶ Thus, many decisions⁷ at the appellate level on the interpretation of s. 31(2) Children Act 1989 appears to demonstrate an abysmal trend towards child protection and Re J⁸ is no exception.

Before analysing the decision in Re J,⁹ it is expedient to set out the lacuna in the law regarding the interpretation of the threshold conditions. It is submitted that these areas have caused difficulties in judicial interpretation and will be considered in the following order – the standard of proof, the meaning of significant harm and finally the unknown perpetrator cases which Re J¹⁰ indubitably falls under.

Standard of Proof

The starting point in any analysis in this area of child protection is to look at the judicial interpretation on the standard of proof required. The purpose of this inquiry is to demonstrate that there is no clear guiding principle on the standard of proof to be applied until most recently. This lacuna is a harbinger that would set the stage for the judicial dilemma experienced in Re J.¹¹

At first glance, the standard of proof required to satisfy the threshold can be drawn from Re H,¹² where the meaning of ‘is likely to suffer significant harm’¹³ was interpreted to be a ‘real possibility’ that cannot ‘sensibly’ be ignored.¹⁴ This made it relatively easy to satisfy the threshold in relation to a risk of future harm to which Herring has commented that ‘this is a remarkably pro-child protection stance of the law to take’.¹⁵ However, it is contended that Herring makes the assertion only in isolation as the ‘real possibility can only be established by facts proved to the civil standard of balance of probabilities’.¹⁶ The second part of Re H¹⁷ potentially made it even harder to prove facts¹⁸ by requiring compelling evidence for serious allegations¹⁹ but that was clarified in Re B²⁰ where Lady Hale stated that the standard of proof in finding facts necessary to establish the threshold ‘is the simple balance of probabilities’.²¹

While Bendall argues that Re B²² tipped the balance in favour of child protection once more,²³ it is submitted that the current standard of proof is flawed because the requirement of proven facts means that where the ‘gravest suspicions of actual abuse’²⁴ exists, the child cannot be protected. Thus, critics such as Hoyano and Keenan have argued that this approach ‘is fundamentally wrong’ as the child protection legislation is ‘purported’ to be based on the best interest of a child.²⁵ As will be demonstrated, the standard of proof has led to serious deficiencies at appellate level decisions and the apprehension is that this may set an undesirable precedent for future cases presenting similar facts.

Significant harm

A recent Supreme Court case which appeared to prioritise child protection against parental autonomy is Re B (a Child).²⁶ Although the decision shifted the focus back on the child’s welfare, the decision is arguably inconsistent because the court allowed potential harm arising from ‘deficits relating to parents’ character rather than to the quality of their parenting’²⁷ to satisfy the meaning of significant harm.

The first issue here is that this appears to be a shift from the standard of proof required (discussed above) as it based a likelihood of harm on a ‘mere’ possibility that the parent’s character will cause them to parent the child defectively.²⁸ Thus, this appears to be basing a possibility on another possibility, which would offend the cardinal principle that the real possibility can only be established by ascertained facts proved on a balance of probability.

The second issue is that this interpretation of the Supreme Court would suggest that it follows the previously agreed guiding principles on the interpretation of ‘significant harm’. The interpretation of ‘significant harm’ can be gleaned from two sources;

Thus, it appears on the outset that these decisions suggest a very liberal interpretation of the terms ‘significant harm’.

However, the interpretation of significant harm has not always been liberal. For instance, a very troubling decision is Re MA³² where the witnessing of violence towards the elder ‘mystery’ child by the younger children was not regarded as significant enough to remove them from the family home. Accordingly, this decision has been subjected to criticism as the way the ‘mystery’ child had been treated by the parents showed a ‘capacity for cruelty’³³ and so gave rise to a real possibility that they would harm their own children.

In order to explain the rationale behind the differing approaches, one might glean from the explanation afforded by Ward LJ in Re MA,³⁴ who states that Article 8³⁵ (right to respect for family life) is relevant when determining the significance of the harm.³⁶ By including competing interests of family life as part of the balancing test, it makes it even harder for the threshold to be satisfied because it would appear that the interference with family life would prevent any intervention by the state.

On the other hand, the decision in Re B (a Child)³⁷ was particularly liberal; Lord Wilson held that the interference with Article 8 occurs only at the welfare stage and not the crossing of the threshold.³⁸ By separating these competing interests, it would allow the threshold to be crossed at the outset and any interference with family life can be determined at the later ‘welfare’ stage. This is submitted to be not only a neater solution but a right attitude to take in terms of prioritising the promotion of the child’s welfare over just solely risk management.³⁹

Accordingly, Hayes et al.have condemned the decision in Re MA⁴⁰ and state that ‘any reference to the value placed by society on “diversity and individuality” cannot be used to justify child cruelty or exposing young children to the risk of significant harm’.⁴¹ Indeed, Fortin has agreed that ‘the significant harm formula may itself encourage a superficial and legalistic approach to child protection work’.⁴² An appropriate balance must thus be struck between the decision of Re B (a Child)⁴³ and Re MA.⁴⁴ The writer’s opinion is that a hard and fast legal rule might not be appropriate here especially when everything hinges on the fact-specific nature of such cases.

Attribution

In Lancashire CC (LCC) v B,⁴⁵ the House of Lords considered the situation in which it was unclear who had caused significant harm to a child. In order to circumvent the standard of proof which required past facts to be proved on the balance of probabilities, the court held that the phrase ‘attributable to’ in respect of s. 31(2)(b), is not to a parent but the care given by all the carers.⁴⁶

While Lord Nicholls recognised that ‘parents who may be wholly innocent… will face the possibility of losing their child’,⁴⁷ he was satisfied that this was the right decision to take in order to prevent further injuries being inflicted on the child.⁴⁸ Logically, this decision would be laudable, but as Hall has demonstrated, Lord Nicholls was being overly interventionist at the parents’ expense while setting the threshold too high in Re H.⁴⁹ Perhaps this ‘is indicative of the difficulty in finding an acceptable balance between, on the one hand, respecting the integrity of the family, and on the other, protecting the child’.⁵⁰ However, the decision may be justified on the basis that the child had already suffered proven significant harm. If so, why did the court not extend the attribution element to include non-carers?⁵¹ One submits that when significant harm has been proven, it should not matter who the harm came from as it is illogical to distinguish between a childminder and a non-carer.⁵²

Subsequent unharmed children

Although the House of Lords decision in LCC v B⁵³ demonstrates a considerably strong child protection stance, it was not given the opportunity to consider what should happen to the childminder’s unharmed child. One must fall back on the Court of Appeal’s worrying decision in LCC v B⁵⁴ where the threshold was not crossed because it had not been proven on the balance of probabilities⁵⁵ that the child minder was the perpetrator of the other child’s injuries.⁵⁶ As Perry put it succinctly, this approach means that ‘where one is dealing with likelihood of harm, not only must the harm itself be proven, but it must also be beyond doubt who was the cause of that harm’.⁵⁷ This is irrational as it denies any possibility of protecting the child even though the childminder ‘was one of three potential perpetrators of serious abuse suffered by a child who was in her care’.⁵⁸

This decision is to be contrasted with a much more liberal approach taken in Re CB,⁵⁹ in which one of two children was harmed by either of the parents, but it was unclear who the actual perpetrator was. As to the unharmed child, Wall J held that the circumstances gave rise to a real possibility that the unharmed sibling would also suffer significant harm.⁶⁰ This child-focussed approach was also followed by Hedley J in Re K, ⁶¹ who held that it was not necessary to ‘go the final step and decide as between the two’ who killed the child. ⁶²

Recent cases

Nonetheless, these liberal approaches to child protection in Re CB⁶³ and Re K⁶⁴ appear not to have survived three recent rulings: Re S-B,⁶⁵ Re F⁶⁶ and Re J.⁶⁷ These cases confirm that where there is a finding that the elder child has been abused but the court is unable to do more than establish that a parent of the unharmed sibling is in the pool of possible perpetrators, that state of affairs is insufficient on its own to establish a finding that the sibling is likely to suffer significant harm.⁶⁸

While Wilson LJ in Re F⁶⁹ appeared to recognise the unnecessary risk imposed upon the child in holding that the threshold had not been crossed, his justifications that ‘the requirement of proven factual foundation is a bulwark against the state’s removal of a child from his family, which [he] consider[s] very precious’⁷⁰ is highly disturbing. Clearly, a risk of harm to a child is significant enough for some form of limited intervention.

Nonetheless, in Re SB⁷¹ and Re F,⁷² the facts are such that the harmed child had only suffered injuries but these injuries were not serious enough to have caused death and so that could be a possible justification for not holding the threshold crossed. Possibly, the risk to the unharmed child may not materialise and the apparent overarching state interest is the promotion and preservation of the new family nucleus. However, such tolerance and accommodation of parental autonomy surely cannot be the overarching consideration in Re J,⁷³ where the severity of the injuries inflicted on the harmed child resulted in death.⁷⁴ In spite of that, it is somewhat surprising that the Supreme Court was not minded on this point.

The liberal approaches in Re CB⁷⁵ and Re K⁷⁶ were also criticised as failing ‘to distinguish between the degree of likelihood required by the word “likely” and the factual findings required to satisfy the court of that likelihood’.⁷⁷ However, Gilmore disagrees as he states that ‘there was no engagement with the careful reasoning… nor with the policy’⁷⁸ leading to that conclusion. Furthermore, Hayes has argued that the need to establish facts on the balance of probabilities as per Re H⁷⁹ was ‘stated in the context of proof of actual and likely significant harm’ and should not apply where a court cannot determine who in a pool of possible perpetrators was responsible for the proven harm.⁸⁰

The court, thus, ‘made a judgement that concentrates on black-letter, judge-made rules relating to proven facts’.⁸¹ The effect of Re J⁸² as illustrated by Herring, is that ‘if a child dies but it is unclear whether the mother or father killed the child and a finding is made that it was one or other of them and so both are possible perpetrators, what happens if they separate and both have children with a new partner?’⁸³ As a result of Re J,⁸⁴ the local authority cannot even obtain a supervision order to monitor the child’s welfare.⁸⁵

The only aspect which Re J⁸⁶ contributes to protecting vulnerable children, as illustrated by Bainham, is to confirm the approach that ‘where both or all potential perpetrators continue to reside in the same household the threshold should be crossed’.⁸⁷ Here, as Lord Wilson puts it, ‘the fact is that somebody in the child's proposed home did perpetrate injuries to another child’.⁸⁸

Opposing view

On the contrary, there are factors which could help to mitigate the sting of Re J.⁸⁹ For instance, the mother was very young when the child died; this was a new relationship with a respectable person and she was now much more mature.⁹⁰ Notwithstanding this, Baroness Butler-Sloss acknowledged that the fact ‘she had colluded with the man in the first case’,⁹¹ to conceal facts as to who perpetrated the injuries, was not taken into account. If this was taken into account, it would most probably have allowed the threshold to be crossed. Moreover, in B (Children),⁹² the court did not permit the separation of possible perpetration and a failure to protect the child from harm. Thorpe LJ held that ‘the two issues inevitably infect each other’⁹³ and ‘adverse findings in relation to failure to protect were inevitably relevant to the findings as to perpetration’.⁹⁴ Hence, it leaves one to wonder why the Supreme Court in Re J⁹⁵ proceeded on a single issue of possible perpetration when it had the authority to include other relevant factors for consideration. Correspondingly, with the decision in Re SB⁹⁶ and Re J,⁹⁷ it seems that a failure to protect may not be as authoritative as it once used to be. Therefore, this demonstrates that evidence is a crucial aspect in care proceedings. The recent attempt to reduce care proceedings to twenty six weeks⁹⁸ due to fears that delay may be detrimental to a child,⁹⁹ is a cause for concern. It now takes about thirty six weeks for care proceedings to be heard in court.¹⁰⁰ If such a reduction compromises on resources to gather evidence and present them effectively, then, as Holt and Kelly have stipulated, the reduction may not be as beneficial as anticipated.¹⁰¹

Possible other approaches

What should the court have done in Re J?¹⁰² A less galling approach would be to allow cases of possible perpetrators to be determined under the liberal ‘attribution’ element of the threshold test instead of the ‘likelihood’ element which invokes the difficult standard of proof. Downplaying the enormity of the threshold test, the court in Re C and B,¹⁰³ held that merely because the threshold is crossed does not mean a care order would be an appropriate response, as the court still needs to consider the welfare checklist.¹⁰⁴

The grant of a supervision order in lieu of a care order would also demonstrate a greater compliance with the principle of proportionality at the disposal hearing.¹⁰⁵ However, Fortin has argued that the use of supervision orders is ‘deeply unpopular [as] the local authority gains no real control over the case’,¹⁰⁶ to which Gillespie proposes a third order to address the shortcomings of the supervision order and the excessiveness of the care order.¹⁰⁷ It remains to be seen if this proposal would be adopted. Nonetheless, one submits that the use of supervision orders is simply better than doing nothing at all.

Moreover, in Re O and N¹⁰⁸ where the court took a liberal approach to child protection, Lord Nicholls said that where the facts suggest that a parent was a perpetrator of proved harm, ‘it would not be right for that conclusion to be excluded from consideration at the disposal hearing’.¹⁰⁹ Support for this proposition can be gleaned from Bainham who recognises that at the threshold stage, ‘possible perpetration should be sufficient because the second child is indubitably still at risk’.¹¹⁰ As for the interests of the parents in the pool of perpetrators, these rights ‘can be adequately protected at the welfare stage of care proceedings’ when the court has to decide what order to make.¹¹¹

Furthermore, in Re J,¹¹² Lady Hale had been criticised for holding that a consignment of a person to a pool of possible perpetrators was insufficient on its own but when taken with other factors would enable the threshold to be crossed.¹¹³ This position is illogical; as Lord Wilson explains ‘if, for the purpose of the requisite foundation, X’s consignment to a pool has a value of zero on its own, it can, for this purpose, have no greater value in company’.¹¹⁴ It has to be the case that if a person is deemed to be part of a pool of possible perpetrators and there are sufficient evidences to prove that this is so, then clearly it is not right for the court to attribute no value at all to such a ‘fact’. Therefore, it is submitted that one should not ignore the sole fact of past possible perpetration at any stage. So how should the enigma be resolved? Lord Wilson has provided extremely useful guidance in reconciling unproven allegations with the welfare needs of a child at the disposal stage,¹¹⁵ which is submitted to be a better safeguard against disproportionate interference into family life. Consequently, the argument that the threshold is in place to protect parents is unpersuasive as they can be guaranteed at the disposal stage.¹¹⁶ This would address the criticism which Fortin raises that ‘the case law increasingly suggests that the current law is more concerned with protecting parents from false accusations of child abuse than fulfilling children’s rights to state protection’.¹¹⁷

Reform

The approach suggested above is not binding and is unlikely to be applied by any courts in the near future. Thus, reforms may be necessary. Notwithstanding that the threshold cannot be crossed, Cobley and Lowe propose legislating to permit courts to take some action to safeguard children possibly at risk; by requiring short-term periodic reviews by an Independent Reviewing Officer with a view to bringing the matter back to court if the circumstances necessitate it.¹¹⁸ However, if the court determines that the threshold cannot be crossed, it means that it recognises that a continuing risk to a child is absent. It is illogical to conclude one way and then seek to use a backdoor to circumvent this finding. Therefore, when there is a real risk of serious injury necessitating a care order, this solution may not provide adequate protection for the child, as the lower courts will be bound by the decision in Re J.¹¹⁹ The only real solution, as Hayes has identified, is for Parliament to change the law.¹²⁰ An amendment to the Children and Families Bill¹²¹ has been laid before Parliament¹²² and would have the effect of permitting judges, in an appropriate case, to find likely harm from the sole fact of past possible perpetration, thus reversing the effect of Re J.¹²³

Conclusion

In conclusion, this essay has demonstrated that vulnerable children may no longer be sufficiently protected following recent case law, such as Re J.¹²⁴ The standard of proof requiring past facts to be proved on the balance of probabilities can sometimes produce very absurd outcomes. This is attributable to the court perceiving its primary role as protecting family life against disproportionate state interference. Legislation might work towards mitigating the sting of Re J; however the amendments put forward by Lord Lloyd in the House of Lords during the 4th sitting of the Committee on the Children and Families Bill did not make it to law when the Children and Families Act was passed in 2014.¹²⁵ Re J, thus, survives as good law.

However, where the continuing risk of harm to a child is patently obvious, it is imprudent to restrain cases at the threshold stage on mere specificities, as this means that some form of protection cannot even be proffered. This is akin to sweeping problems under the carpet where the risk of significant harm persists and it is submitted that this reflects a very dangerous approach towards protecting vulnerable children.