In the Matter of ZZ, AZ, FA, ARA, KA and ASA (Children)

This is a father’s appeal against a decision of District Judge Aitken sitting in what was, at the time, the Principal Registry of the Family Division. In accordance with the then practice, the appeal was made to, and the hearing before me on 12 March 2014 took place in, the Family Division of the High Court. In accordance with articles 2 and 3(1) of The Crime and Courts Act 2013 (Family Court: Transitional and Saving Provision) Order 2014, SI 2014 No. 956, the proceedings have continued on and after 22 April 2014 in the Family Court as if they had been issued in that court. It is accordingly in the Family Court that I now sit to give judgment.

The appeal arises out of care proceedings relating to six children: ZZ born in 1997, AZ born in 2000, FA born in 2005, ARA born in 2006, KA born in 2008, and ASA born in 2010. The mother of all six children is ZN. The father of the four younger children is SA. The maternal grandparents are MN and SP. AM and RM are the maternal great uncle and great aunt. SS is a maternal cousin.

The proceedings commenced on 19 July 2012. The local authority’s ‘threshold’ statement asserted that all six children had suffered significant harm. In support of that assertion the local authority relied on nine specific allegations. The first was that SA had assaulted the two elder children, ZZ and AZ: the next three were that SA had assaulted ZZ; the fifth was that ZN had failed to protect ZZ; the other four were that SA had assaulted ZN. The threshold statement concluded with the assertion that, as a consequence, all the children suffered significant emotional harm and were at risk of suffering physical and emotional harm. A fact finding hearing in relation to the nine specific allegations took place before District Judge Aitken over ten days in December 2012. The District Judge handed down a reserved judgment on 14 January 2013. She found all nine allegations proved.

Subsequently, ZN and ZZ purported to retract their complaints about SA. The question of what should be done came before the District Judge on 13 August 2013. SA sought a rehearing of the entire fact finding before a different judge. ZN sought a rehearing of the first four allegations before a different judge. The local authority accepted that certain findings should be discharged. The District Judge refused the applications for a rehearing, discharged certain of her findings, and gave directions for a final hearing on 30 September 2013 with a time estimate of five days. She refused SA’s application for permission to appeal.

SA’s renewed application for permission to appeal came before Parker J on 25 September 2013. The local authority sought to cross-appeal. Parker J allowed the appeal and cross-appeal and remitted the matter for a further fact finding hearing before the same District Judge. I shall return below to deal with Parker J’s reasoning and the precise terms of the order she made. Suffice for the moment to say that she was referred to and sought to apply In re B (Minors) (Care Proceedings: Issue Estoppel) [1997] Fam 117, a decision of Hale J (as she then was) which I shall refer to as Re B, and Birmingham City Council v H and others [2005] EWHC 2885 (Fam), a decision of Charles I which I shall refer to as Birmingham (No 1). These two decisions lie at the heart of the controversy on the principal issue that I have to determine.

Following the further fact finding hearing, which in the event lasted eight days, the District Judge gave a further judgment on 9 October 2013. I shall return below to deal with her reasoning, but her conclusion was expressed as follows:

“I do not find the retractions by either ZZ or ZN to be credible … I do not discharge any of the findings made following the fact finding hearing.”

She went on to find that threshold was established on the basis that:

“AZ and ZZ have suffered physical and emotional abuse and the younger children … have suffered emotional abuse.”

Her ultimate decision, embodied in orders dated 9 October 2013 and 23 October 2013, was that there should be residence orders in respect of the two elder children in favour of MN and SP, a special guardianship order in respect of the youngest child in favour of SS, and special guardianship orders in respect of the other three children in favour of AM and RM.

SA again sought permission to appeal. His application came before Baker J on 15 January 2014. Baker J gave permission to appeal. He also gave permission to amend the grounds of appeal (see below) to raise what he suggested was the real issue in the appeal, whether the approach in Birmingham (No 1) is correct in a case such as this.

Before proceeding any further it will be convenient to consider the case-law.

I start with Birmingham (No 1).

Birmingham (No 1), like Re B before it, was a case where a local authority sought to rely upon findings made in earlier, concluded, proceedings relating to a different child. In Re B the current care proceedings related to two children G and D; the local authority sought to rely upon findings against their father made in earlier care proceedings relating to two other children. In Birmingham (No 1) the current care proceedings related to S; the local authority sought to rely upon findings against her mother by Bracewell J, upheld subsequently by Kirkwood J, in earlier proceedings relating to another of her children, P. In accordance with the order made by Charles J in Birmingham (No 1), the final hearing of the proceedings relating to S took place before McFarlane J (as he then was): Birmingham City Council v H and others [2006] EWHC 3062 (Fam), 95 BMLR 159, which I shall refer to as Birmingham (No 2).

In the present case, of course, the forensic context is quite different. Here, as in Re M and MC (Care: Issues of Fact: Drawing of Orders) [2002] EWCA Civ 499, [2003] 1 FLR 461, the issue arises in the context of care proceedings which have not yet concluded. One of the questions I have to consider is whether that forensic difference justifies a different approach.

In Birmingham (No 1) Charles J said this, para 55:

“In my view the approach [of the family court to earlier findings] has three stages. Firstly the court considers whether it will permit any reconsideration or review of, or challenge to, the earlier finding (here referred to by the parents as a review). If it does the second and third stages relates to its approach to that exercise. The second stage relates to, and determines, the extent of the investigations and evidence concerning the review. The third stage is the hearing of the review and thus it is at this stage that the court decides the extent to which the earlier finding stands by applying the relevant test to the circumstances then found to exist.”

McFarlane J specifically agreed with that analysis in Birmingham (No 2), para 42(ii).

Before proceeding any further with Charles J’s analysis it is convenient to return to Re B. In that case Hale J was concerned primarily with the first stage. She identified, pages 128-129, a number of factors to be borne in mind by the court when deciding whether or not to allow any issue of fact to be tried anew. I need not set them all out. For present purposes what is important is this:

“Above all, the court is bound to want to consider whether there is any reason to think that a rehearing of the issue will result in any different finding from that in the earlier trial. By this I mean something more than the mere fact that different judges might on occasions reach different conclusions upon the same evidence … The court will want to know … whether there is any new evidence or information casting doubt upon the accuracy of the original findings.”

In relation to the third stage, a matter which she made clear had not been addressed before her, Hale J said that the trial judge:

“will no doubt wish to consider whether there appears to be some real reason to cast doubt upon the earlier findings.”

In Birmingham (No 1) Charles J was not concerned with the first stage for, as he explained, para 57, that had been dealt with at an earlier hearing by Holman J. Nor was he directly concerned with the third stage. As he said, “We are at the second stage.” Nonetheless, he went on to consider the third stage. In a crucial passage, paras 62-63, he said this:

“62

At this stage, as well as at the earlier stages, the strong public interest in finality in litigation plays an important part. It supports the conclusion that before earlier findings are set aside, or are not to be treated as binding, a high test has to be passed.

63

At this stage it seems to me that analogies can be drawn from the approach taken by the Court of Appeal:

a)

in respect of appeals against a finding of fact where great weight is given to the decision of the judge whenever, in a conflict of testimony, the demeanour and manner in which relevant witnesses gave evidence are material elements (see for example the notes to RSC Order 59), and

b)

the approach of the Court of Appeal in the recent cases of [In re U (A Child) (Department of Education and Skills intervening), In re B (A Child) (Department of Education and Skills intervening) [2004] EWCA Civ 567, [2005] Fam 134] and [2004] 2 FLR 263 (see in particular paragraphs 88 to 91 and 151 to 152) and [In Re Uddin (A Child) [2005] EWCA Civ 52,] [2005] 1 WLR 2398 (in particular paragraphs 21 to 23) where the importance of the findings of the judge as to the credibility of the mother in that case (which were not appealed) were given weight in refusing permission to appeal and to reopen the appeal in the case of U, and in dismissing the appeal in the case of B.

In my view the approach taken by the Court of Appeal in these circumstances supports the view that the parents have to satisfy a high test to prevent the local authority and the court from relying on the earlier findings of Bracewell J, as to which firstly the Court of Appeal refused permission to appeal and Kirkwood J found that, if anything, it was reinforced by the result of the further enquiries in 2001.”

He added, para 66, that “each case is fact specific.”

Before Charles J all the parties had agreed that it was necessary to wait and see how things developed at the final hearing. He said, para 68:

“I agree and in those circumstances I am of the view that I should not try to define the test or approach at the third stage. The trial judge will formulate and apply the test or approach he adopts. His aim will be to achieve justice having regard to the competing (and common) private and public interests involved”.

He added, paras 78-79:

“The detail of [the] arguments both as to the test to be applied and its application are matters for the trial judge … the argument at the third stage would be as to whether given the nature and extent of the changes and their relevance and place in the reasoning of Bracewell J (i) her overall finding of inflicted injury should stand, and if not (ii) what, if any, finding should replace it.”

He made clear, para 44, that at the end of the day the burden of proof lay on the local authority to establish threshold in relation to S.

The fullest analysis of the court’s function at the third stage is to be found in the judgment of McFarlane J in Birmingham (No 2). In an important passage that needs to be quoted in full, he said this, paras 42-45:

“42

… Save for one matter of fine tuning to which I shall turn in a moment, there is agreement that the approach to be adopted to the burden of proof is as follows:

(i)

The burden of proving the CA 1989, s 31 threshold criteria with respect to S is upon the local authority and remains upon them throughout;

(ii)

The role of issue estoppel in CA 1989 proceedings has been adapted by the family courts. The classic statement of the law remains that of Hale J (as she then was) in Re B … I agree with the analysis made by Charles J … in this case ([2005] EWHC 2885 (Fam) at [55]) where he indicated that there were three stages in such cases. At the first stage the court considers whether it will permit any reconsideration or review of, or challenge to, the earlier finding. The second stage relates to and determines the extent of the investigations and evidence concerning the review. The third stage is the hearing of the review;

(iii)

Questions of issue estoppel are primarily to be considered at the first stage. Once those who seek a review have passed that stage, issue estoppel is unlikely to be directly relevant either to the investigation process or to the hearing itself;

(iv)

At the review hearing, the previous finding is the starting point of the local authority’s evidence in relation to threshold. A finding of fact is strong evidence and should be followed in the absence of compelling evidence against it. To this extent those who challenge the finding bear ‘an evidential burden’ in the proceedings. The legal burden of proof rests upon, and remains with, the local authority throughout (emphasis added):

(v)

An ‘evidential burden’ is an imprecise, non-legal term applied to the forensic reality faced by a party who seeks to challenge a presumption that otherwise arises in favour of another party by operation of law, previous finding or apparently conclusive evidence. It is no more than the burden of adducing evidence.

[43] In the present case, by adducing the evidence presented at this hearing the parents have discharged the evidential burden of putting up a case to challenge the previous finding. That being accomplished, it is for the court to conduct the process of evaluating that evidence. The legal burden of proof of maintaining the findings made by Bracewell J remains throughout upon the local authority.

[44] The remaining difference of emphasis that still exists between the parties is not without importance and it relates to whether or not any continuing ‘burden’ is placed on the parents once the hearing has actually commenced. The parents submit that, once they have discharged the evidential burden of showing that they have an arguable case sufficient to get through the ‘gateway’ of the court door and start the hearing, there is no continuing burden. The local authority submit that throughout the process priority should be given to the original finding of fact; that finding, they submit, is not simply reduced to the level of evidence in the case alongside any fresh evidence that is called.

[45] Given that I am clear that the extent of any burden upon the parents is limited to an ‘evidential burden’, and given that such a burden has no effect upon the legal burden of proof that remains with the local authority, I consider that the difference that remains on this point is more appropriately to be viewed in relation to the standard of proof, rather than in terms of burden. The forensic reality remains that throughout the hearing the parents have had to make the running to present evidence that challenges the original finding. The evidential burden is no more than that; a description of its effect does not go to the respective weight or priority that may be afforded to the previous finding.”

There is a further important passage, paras 55-56:

“55

… once the hearing in a review process of this nature has commenced, the duty of the court is to look at the matter afresh. Issues of public policy relating to the priority to be given to earlier court decisions is in play at ‘stage one’ of an application to review a finding, when the question of estoppel is considered. Once that stage is passed, and the court is involved in evaluating the evidence itself, it would be wrong to afford some sort of priority to the evidence given and findings made at the earlier hearing that are to be considered within the review. Indeed it would be difficult to do so in any formal way. The starting point will be that a finding has been made. In a case such as this, that result can only have occurred because there already exists ‘strong’ evidence in support of it. Thus any fresh evidence, to get anywhere in achieving the aim of those who call it, must be sufficiently robust to challenge this hitherto ‘strong’ material (emphasis added).

[56] In analysing the evidence at this hearing I have readily adopted the approach suggested by Charles J at para [79] of his judgment. I have considered the fresh evidence alongside the earlier material (such as medical records) upon which it is based. I have taken as fixed points those parts of the detailed findings and judgment of Bracewell J which are either not challenged or remain intact despite the challenge that is being made on the question of the mother’s credibility.”

In relation to the passages in paras 42 and 55 that I have emphasised, it is apparent from what McFarlane J also said, para 46, that he had in mind the famous passage in the speech of Lord Nicholls of Birkenhead in In re H and others (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, 586 (“The more improbable the event, the stronger must be the evidence that it did occur before, on the balance of probability, its occurrence will be established”), which has since been given its quietus by the House of Lords in In re B (Children: Care Proceedings: Standard of Proof) (CAFCASS intervening) [2008] UKHL 35, [2009] 1 AC 11. I shall return to this point below.

In the present case, of course, the forensic context is quite different. The care proceedings here are part heard. Although there has been a separate fact finding hearing, the split hearing “is merely part of the whole process of trying the case. It is not a separate exercise. And once it is done the case is part heard”: In re B (Children: Care Proceedings: Standard of Proof) (CAFCASS intervening) [2008] UKHL 35, [2009] AC 11, para 76. The findings at a fact finding hearing are not set in stone so as to be incapable of being revisited in the light of subsequent developments as, for example, if further material emerges. Until the final decision is made, the judge must be able to keep an open mind and is entitled at any time to reconsider his earlier findings, at least if fresh evidence or further developments indicate that an earlier decision was wrong: In re A (Children: Judgment: Adequacy of Reasoning) [2011] EWCA Civ 1205, [2012] 1 WLR 595, para 21, In re L and another (Children) (Preliminary Finding: Power to Reverse) [2013] UKSC 8, [2013] 1 WLR 634, paras 33-35.

In Re M and MC (Care: Issues of Fact. Drawing of Orders) [20021 EWCA Civ 499, [2003] 1 FLR 461, there had been a fact finding hearing in care proceedings. Subsequently, and before final disposal of the proceedings, the mother confessed. His Honour Judge Hamilton, who had conducted the fact finding hearing, refused to allow the matter to be re-opened:

“Having heard the submissions … and the request for a further hearing to consider the findings I made, I have to say I am not convinced that that is either necessary or, indeed, appropriate … my conclusion is that it would not be appropriate to seek to have these matters reopened by having another finding of fact hearing because I have to say I doubt whether the conclusions that I could arrive at would, in fact, be necessarily any different.”

Allowing the appeal, and sending the matter back for further consideration by the same judge, Thorpe LJ, with whom Neuberger J (as he then was) agreed, said this, paras 13-14:

“13

… plainly trial judges have to be firm in not permitting the court’s important duty to investigate and establish past fact to be derailed or diverted by what may be simply strategic manoeuvring in response. Particularly, courts must be guarded in acceding to applications for yet another trial of an issue of fact in what should be the relatively brief period between the preliminary hearing of disputed facts and the subsequent hearing to dispose of the outstanding application for care orders.

14

So the notion that the process … should be torn up as though it had never happened simply because one of the adults had subsequently made a statement shifting position was plainly unlikely to succeed and was, in my judgment, rightly rejected by His Honour Judge Hamilton. That of course is one extreme. The other extreme would be to reject the development absolutely and treat the previous finding of fact as incapable of being revisited. There is, between these two extremes, an obvious middle way, and that is to conduct the disposal hearing in such a way as to adopt the process of preliminary hearing as the foundation, and then to make such adjustments as are necessary to reflect subsequent developments rigorously tested through the process of examination-in-chief and cross examination. His Honour Judge Hamilton clearly suspected that there was no greater value in the fifth statement than in the earlier statements, and in that suspicion he may be right. But no complete conclusion can be reached without affording the mother the opportunity of explaining herself in the witness-box and answering as best she can the local authority’s response, namely, that the fifth statement is contradicted by or is inconsistent with, the medical evidence.”

He added, para 16:

“There are, of course, both advantages and disadvantages of the practice that has developed for isolating the specific issues of fact for preliminary hearing. This case only illustrates one of the risks inherent in the process. I do not think that that risk can be eliminated. It amounts simply to one of the disadvantages to be set against other advantages. On balance the disadvantage is an acceptable and a manageable one providing that courts avoid either the extreme of plunging into a complete reopening of the issue of fact, or, on the other extreme, excluding the subsequent development from proper consideration in its context at the disposal hearing.”

I return to events in the present case.

Parker J identified the fundamental dispute between the parties as being “whether the boy’s retraction undermined the findings made in their totality and undermined the boy’s credibility in its totality.” Referring to what Hale J had said in Re B, she said that, in the context of this case, the District Judge needed to consider whether there was any new evidence or information casting doubt on the original findings. Having found the District Judge’s approach to be flawed, Parker J continued:

“It is necessary in my view that there should be a proper examination of whether, and if so, to what extent the boy’s retraction, and mother’s, impact upon the totality of the findings made by the judge. I stress the threads of evidence supporting these nine findings are very much woven together.”

Parker J referred to Charles J’s three stage approach in Birmingham (No 1) and said:

“I have decided that I will permit limited reconsideration / review of the challenge to the earlier findings. The second stage is the extent of the evidence for the review that I have directed. The judge will have to decide the extent she will need to investigate and the District Judge will be wholly in charge of hearing the review.”

Parker J concluded:

“I allow the appeal and cross-appeal. I delete the discharge of earlier hearings and restore them. I direct that the hearing will remain before DJ Aitken from 30^th September on the basis that the Judge must consider whether on the basis presented by the step-father and/or the mother the Judge’s findings of January 2013 should be deleted or varied having taken into account the retractions of the mother and the boy, the extent to which those retractions affect the assessment of credibility of the boy and with specific regard to his assertion that he has concocted evidence of injury.”

I read that as indicating that, although Parker J had laid down the broad parameters within which the further hearing should take place, it was for the District Judge to determine the fine detail of that hearing before coming to her ultimate conclusions on the facts.

At the outset of her judgment the District Judge said that “it is important for a judge who is being asked to reconsider findings in the light of new evidence to keep an open mind.” She continued, “I must carefully consider the new evidence from ZZ.” She gave herself a ‘Lucas’ direction: R v Lucas (Ruth) [1981] QB 720. She then proceeded to rehearse in considerable detail the course of the hearing and the evidence she had heard, before turning to consider the law.

The District Judge directed herself that the test in respect of reconsidering the findings is set out in Birmingham (No 1). Having set out the key passages, including what Charles J had said in Birmingham (No 1) paras 55, 62 (where it will be recalled he had referred to “a high test”), she continued:

“The first and second stages identified by Charles J have already been dealt with, firstly at the hearing on 13^th August 2013 and then by the subsequent appeal to Parker J. I must deal with the third stage in that the findings of the court are to be reconsidered because ZZ has retracted the allegations which he made against his stepfather, and his mother has retracted her allegations of domestic violence. The court must weigh up both of these retractions against the other evidence and there must be strong and cogent evidence that the original findings of the court are unsafe in order that the high test referred to above before findings can be set aside is passed.”

She said:

“I have looked again at all the evidence before me at the fact finding hearing and considered whether the new evidence from ZZ and ZN casts doubt on my previous findings.”

After a further analysis of the evidence the District Judge said this:

“The evidence from ZZ, the mother, R and the maternal grandparents cannot be relied upon in respect of discharging findings … I find that their evidence was tailored to fit in with ZZ’s account … I … have reached the conclusion that when their evidence to the court is analysed it suggests some form of agreement between the adults in this family to overturn the fact finding judgment, so that the four younger children could return to their mother and SA”.

She expressed her conclusions as follows (it is para 63 of her judgment):

“In conclusion I do not find the retractions by either ZZ or ZN to be credible. It follows that, having considered the authorities on reconsideration of findings referred to above, I do not find that the high test has been passed and I do not discharge any of the findings made following the fact finding hearing.”

I need not repeat what I have already said by way of summary of the remainder of the District Judge’s judgment.

Baker J was concerned about the approach that had been adopted by the District Judge, though recognising that it followed the approach outlined by Parker J and had indeed been put forward by counsel. Having referred to Birmingham (No 1) and Re B, Baker J said this:

“Those decisions concerned cases in which the court was being asked to review findings made some years earlier in previous proceedings. The present case, on the other hand, involved a rehearing of allegations arising in the same proceedings which had not yet concluded. In those circumstances it seems to me questionable whether the three-stage test proposed by Charles J in the Birmingham case is a correct approach. If a rehearing has been ordered within the same proceedings, it seems to me that it is a complete free hearing. Of course, some of the evidence given before may be taken as read. It will not normally be necessary for every single aspect of the evidence at the first hearing to be revisited at the rehearing, but where there is new evidence – in this case alleged retractions of previous allegations – the court must look at all relevant material again in the light of the new evidence.”

He added:

“A recent example of this is a case in which, as it happens, I have been involved (see my decision in Re L & M [2013] EWHC 1569 (Fam), in which I conducted a rehearing following a successful appeal to the Court of Appeal).”

The reference is to Re M (Fact-Finding Hearing: Injuries to Skull) [2012] EWCA Civ 1710, [2013] 2 FLR 322.

Baker J continued:

“The danger of the approach adopted by District Judge Aitken is that it shifts the burden of proof on to the respondent. That is indeed what I fear may have happened, as set out in the passage I have quoted from paragraph 63 of her judgment. Accordingly, applying the law relating to applications for permission to appeal, I conclude that it is arguable that District Judge Aitken made an error of law in her approach to the process of reviewing the hearing. I therefore give the father permission to appeal.”

He directed that the appeal was to be listed before me:

“This case highlights a number of issues concerning the approach in cases where further evidence emerges in the course of family proceedings, in particular between a fact finding hearing and the final welfare hearing. I consider this question should be considered by the President of the Family Division. Accordingly, with his permission, I direct that this appeal be listed before him”.

Before turning to consider the grounds of appeal I note that although all three judges (District Judge Aitken, Parker J and Baker J) were referred to Birmingham (No 1), none of them was referred to Birmingham (No 2). The significance of the latter case was first appreciated by Mr Roger McCarthy QC – he had appeared in the Birmingham litigation – whose first involvement in the present case was before me.

In common with McFarlane J in Birmingham (No 2), I agree with Charles J’s formulation in Birmingham (No 1) of the three stage approach.

The same three stage approach applies, in my judgment, whether the issue arises before the same judge or a different judge, whether in the same or different proceedings, and whether in relation to the same or different children. I do not, with all respect to Baker J’s tentative comment, think that different approaches are called for in different forensic contexts. The attempt to create such a forensic taxonomy would, I fear, be productive merely of satellite litigation. Of course, the application of the general approach in any particular case will reflect the circumstances of that case.

So far as concerns the first stage I agree with what Hale J said in Re B, in particular in the passage I have set out above. I add this: one does not get beyond the first stage unless there is some real reason to believe that the earlier findings require revisiting. Mere speculation and hope are not enough. There must be solid grounds for challenge. But for my own part I would be disinclined to set the test any higher. I have misgivings about McFarlane J’s use in Birmingham (No 2), paras 42, 55, of the words I have emphasised in paragraphs 16-17 above. I suspect that in significant part they reflected the approach of Lord Nicholls of Birkenhead. Be that as it may, I think, with great respect to McFarlane J, that the nuance is wrong.

So far as concerns the second stage, the ambit of the review or rehearing, I doubt that one can sensibly be prescriptive. Much will turn on the forensic context and the circumstances of the particular case.

So far as concerns the third stage, the proper approach in my judgment, subject only to what I have said in paragraph 33 above, is that spelt out by McFarlane J in Birmingham (No 2) in the passages I have quoted. There is an evidential burden on those who seek to displace an earlier finding – in that sense they have to ‘make the running’ – but the legal burden of proof remains throughout where it was at the outset. The judge has to consider the fresh evidence alongside the earlier material before coming to a conclusion in the light of the totality of the material before the court. I think that Charles J’s phrase “a high test” is best avoided at this as at previous stages. I can well understand why, in the particular circumstances of Birmingham (No 1), where there were concurrent findings of two High Court judges and the Court of Appeal, Charles J used those words, but to elevate them to a test – a legal principle – is unwarranted, unnecessary and potentially misleading. Indeed, I think with respect to Charles J that reference to “a high test” at the third stage is simply wrong, essentially for the reasons given by McFarlane J in Birmingham (No 2) at paras 42(iii) and 55.

As amended in accordance with Baker J’s order, there are six grounds of appeal:

SA was represented before me by Ms Diana Murphy. His appeal was opposed by the local authority, represented by Mr Roger McCarthy QC, by ZZ and AZ, represented by Ms Annie Dixon, and by the children’s guardian for FA, ARA, KA and ASA, represented by Ms Shiva Ancliffe. I should record that in her skeleton argument Ms Dixon said that ZZ and AZ did not wish to be involved in any further court proceedings, that they opposed any appeal on the basis that there might be a re-hearing and that ZZ, having had District Judge Aitken’s judgment of 9 October 2013 explained to him, had not instructed her to challenge or go behind it.

I can take these three grounds together.

Ms Murphy submits, in short, that District Judge Aitken effectively reversed the burden of proof; that she adopted the wrong “high test”; that, rather than assessing the retractions in the light of all the evidence, she looked at them in isolation; and that, having concluded that they were not credible, wrongly held that the “high test” had not been passed. She submits that District Judge Aitken should have reviewed all the factual evidence, not just the retractions, to evaluate whether threshold was met, that she failed to do so and failed to evaluate the differing accounts she had from different witnesses.

There is, at the end of the day, quite a short answer to this. Read as a whole, as it must be, her judgment shows that District Judge Aitken did not reverse the burden of proof and that she correctly understood the task upon which she was embarked. She came to findings that are, in my judgment, unassailable.

The crucial part of her judgment is paragraph 63. I have already set it out but it bears repetition:

“In conclusion I do not find the retractions by either ZZ or ZN to be credible. It follows that, having considered the authorities on reconsideration of findings referred to above, I do not find that the high test has been passed and I do not discharge any of the findings made following the fact finding hearing.”

The first sentence contains a finding of fact which was plainly open to District Judge Aitken and against which, at the end of the day, Ms Murphy has not been able to mount any effective challenge. True it is, of course, given the law as I have set it out, that District Judge Aitken fell into error when in the second sentence she directed herself by reference to “the high test”. But this mis-direction is irrelevant. District Judge Aitken rejected the retraction evidence because she did not believe it.

As Mr McCarthy correctly submits, given the finding that the retractions were not credible, there was simply no factual or legal basis for discharging or varying the original findings. Moreover, given that there was, as he says, no basis for discharging those findings, it cannot matter what legal test District Judge Aitken was applying. The ‘new’ evidence did not satisfy any test, however formulated; put another way, given this finding, the original findings had to stand whatever test was adopted.

The simple reality is that District Judge Aitken, quite properly, gave the parties the opportunity to give further evidence, and allowed the other parties to test that evidence, before coming to a final conclusion on the totality of the evidence, having heard days of evidence from a number of family and non-family witnesses. She listened again to the ABE interviews, viewed the photographs and listened to the witnesses, as she said, “with an open mind”. In her judgment, District Judge Aitken reviewed the evidence fully and carefully, cross-referencing back to the earlier hearing where appropriate, and analysed the chronology of retractions together with the circumstances in which they were made. Her approach to the issues raised by the retractions was, as Mr McCarthy convincingly submits, careful, detailed and balanced. There is no apparent flaw in her approach. Her conclusions on the evidence are unassailable.

Ms Murphy submits that District Judge Aitken’s findings on threshold were not open to her on the totality of the evidence. With all respect to Ms Murphy, the argument is little short of hopeless. If District Judge Aitken was justified in coming to her findings of primary fact, and that is the assumption on which this ground of appeal falls to be decided, then it was clearly open to her to conclude as she did in relation to threshold.

Ms Murphy submits that threshold, if met, was only to “a relatively low level” and that the circumstances were not such as to justify the separation of the children from their parents, something to be contemplated only if “nothing else will do”. This again is little short of hopeless. District Judge Aitken directed herself appropriately and gave more than adequate reasons justifying her decision. There is, at the end of the day, no basis for any successful complaint that District Judge Aitken’s conclusion was wrong.

The complaint here is that SA’s counsel was “prevented” from asking questions in cross-examination of both the social worker and the guardian about the circumstances in which they had received ZZ’s retraction and as to its reliability. Ms Murphy also raises questions as to whether, as she puts it, ZZ was “properly represented at all” at the final hearing. It is right to record that Ms Murphy did not pursue either point with any great vigour.

As to the first point, I was not shown any transcript of the hearing, and without a transcript or detailed note of the relevant exchanges there are obvious difficulties in an appellate court engaging in a complaint of this character. As to the second point I observe only that this is not a point that ZZ has chosen to take.

For these reasons I have concluded that this appeal fails and must be dismissed.

I conclude with a comment about the use of separate fact-finding hearings in care cases.

The current approach is to discourage their use except in a relatively limited group of cases: see Re S, Cambridgeshire County Council v PS and others [2014] EWCA Civ 25, paras 26-32. Ryder LJ made clear, para 29, that a split hearing in a care case will usually be appropriate only in either “the most simple cases where there [is] 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” or “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.” He went on, “For almost all other cases, the procedure is inappropriate.” I agree. See also, to the same effect, Re S (Children, W & T) [2014] EWCA Civ 638, para 71.

I add only this. The benefit of hindsight is enormous, and the split hearing in the present case was directed at a time when such things were fashionable, but it might be thought that the present case stands as an object lesson in the perils of the split hearing. The risks to which Thorpe LI drew attention have in the present case borne spectacular fruit, to the disadvantage of everyone and the discredit of the system. Even if one leaves out of account the further delay generated by the appeal which is currently before me, the proceedings which had been launched in July 2012 were not concluded until October 2013. How much quicker and more smoothly everything would have gone if there had not been a split hearing.