In the matter of P-S (Children)

On 28 March 2017 His Honour Judge Tolson QC granted care orders to the London Borough of Camden in respect of two young boys who I shall refer to as P and S. P is aged 5 and S is aged 2. They have different fathers. Their mother lives with the father of S and is separated from the father of P. The effect of the orders is that the children are living with their respective paternal grandparents.

The children’s mother and the father of S would like the children to return to their care, but that is not the issue in the proceedings in this court. There are also contact issues in respect of the mother and the father of S which will need to be decided. The father of P is happy with the care arrangements that involve his son being cared for by his parents, within which he has a role to play.

At the conclusion of the hearing of this appeal, and because the court had heard extensive argument, it was decided to reserve the contact issues and remit them to the President of the Family Division, sitting alone. The court had been made aware that because of the outstanding issues on the merits and in the absence of legal representation, the children’s mother wanted an adjournment. The court deferred that application until the end of the hearing in order to ensure that it safeguarded the children’s interests by taking the most appropriate and expeditious course which has the consequence that the mother’s wish to make representations about the welfare issues that remain is preserved and she may be able to obtain legal or pro bono assistance for the adjourned hearing.

The issue in this appeal is whether the judge was wrong in the circumstances that occurred to decline to make special guardianship orders [SGOs] to the paternal grandparents. The appellant before this court is the children’s guardian, Mr. Jeff Boyd. The appeal is supported by the local authority. The paternal grandparents had never been made parties to the proceedings in the Family Court but during case management in this court they were joined and have taken an active part in the appeal supporting the guardian. The father of P wrote a letter to the court in support of his parents. The paternal grandparents of S concede that at the time of the final hearing in March 2017 their position was not that an immediate SGO should have been made but that an adjournment was necessary to facilitate the transfer of S to their care within which S would have been protected by an interim care order.

The father of S was represented and the children’s mother appeared in person. They opposed the appeal on the basis that they did not want any step to be taken which prejudiced the mother’s intended application, supported by the father of S, to discharge the care orders so as to have the care of the children returned to them.

The Association of Lawyers for Children applied for and were granted leave to intervene to assist the court in relation to the issues of law and principle. The court has been greatly assisted by the written materials and oral submissions of all parties.

The background circumstances can be taken quite shortly given that the issues in the appeal were issues of law and principle. The care proceedings began in January 2016 and were heard in the Family Drug and Alcohol Court (FDAC) in London. Unfortunately, the mother and the father of S relapsed and by the time of the final hearing had withdrawn from the FDAC process (i.e. from the process of rehabilitation in that court which is managed by a judge and monitored by a specialist team). The mother continued to argue that she could care for the children but the judge concluded that she was not a ‘viable option’ or ‘realistic option’ as a carer. There is no appeal against that conclusion or the findings of fact that support it which easily satisfy the threshold in section 31 of the Children Act 1989.

During the proceedings P was living with the maternal grandmother and S was living with a paternal aunt and her partner under interim care orders. By the time of the final hearing S’s carers were not in a position to continue to care for him. The maternal grandmother withdrew her application to care for P on the first day of the final hearing.

One of the reasons why the care proceedings were not completed within 26 weeks was that S’s aunt and her partner had initially been assessed and approved as S’s long term carers. When they changed their minds in November 2016 as a consequence of the alleged conduct of S’s parents, the local authority had to consider a new care plan for S and that led to the adjournment of what had been intended to be a final hearing on 28 November 2016.

Both sets of paternal grandparents were assessed by the local authority as potential special guardians in accordance with section 14A(8) of the 1989 Act. Both reports led to recommendations that the children be made the subject of SGOs to their respective paternal grandparents. P’s father agreed with the process and the recommendation for P. The local authority filed care plans based upon the recommendations. The children’s guardian supported the recommendations for placement under SGOs but raised some issues with the plans for contact and support, which were narrowed as the hearing progressed. The recommendations were opposed by the children’s mother and the father of S. There is no issue that the assessment reports were completed in accordance with the regulatory scheme and there was no challenge to their content during the hearing.

Neither the local authority nor the grandparents made an application for special guardianship with the consequence that the court was invited to make SGOs of its own motion. The grandparents were not represented before the court and had not been joined, although they were physically present and called into court by the judge for a part of one of the days of the hearing. The grandparents had no opportunity to have any legal advice about what was happening in the Family Court, something that has now been remedied as a consequence of this court’s directions for their joinder. They were surprised when the judge made full care orders.

The court has been told in additional evidence that both boys are doing well in the care of their grandparents. P’s father agrees. The local authority and the children’s guardian with the consent of P’s father continue to submit that SGOs are appropriate. Subject to appropriate funding and the answers to questions they had properly raised before the appeal was heard, the grandparents support the making of SGOs. They confirmed within the appeal that their questions had been answered to their satisfaction. The local authority was able to demonstrate to the court that it had come to agreements with the grandparents for appropriate special guardianship support plans comprising weekly financial assistance, support for contact and other long term support that the children will need for a period of three years.

There are three strands to the errors that all represented parties before this court identify in the family court’s decision: a) the lack of any adequate reasoning for making care orders rather than interim care orders or special guardianship orders, b) the reliance of the judge on informal guidance that was neither approved guidance nor peer reviewed research capable of being scrutinised or challenged by the parties and c) procedural unfairness. I shall take each in turn. The court is mindful of the fact that each of the represented parties before it (except S’s father) have taken the same position in respect of each issue and accordingly the court has tested with the interveners each of the propositions in respect of which they would otherwise have reached a consensus.

The propositions about which there is a large measure of agreement are as follows:

It is helpful to trace the judge’s reasoning by setting out how he came to his conclusion in his judgment. The following extracts are sufficient:

“1....It is not a case in which I must consider rival realistic options in terms of the children’s future placements. Instead, the main question for me to resolve is the appropriate legal order which should govern a placement with the children’s respective paternal grandparents……

7.…the local authority and the Guardian contend that the children’s placements should take place under special guardianship orders………During the trial it has largely been left to me to raise concerns as to whether special guardianship orders in favour of the two sets of grandparents would be premature…….

10.

In this case the children might be placed with the paternal grandparents under either a care order, a special guardianship order, or a child arrangements order. These are very different orders. A care order creates parental responsibility in the local authority which, under section 33(4) of the Act may be exercised by the local authority if they are “satisfied that it is necessary to do so in order to safeguard or promote the child’s welfare”…….

16.

Section 14A provides for those who may make an application for a special guardianship order…….the grandparents come within the definition of those who may apply for a special guardianship order.

17.

There is also a power for a court to make a special guardianship order of the court’s own motion. That power is found at section 14A(6)(b).

20.

……It suffices to say that during my time as designated family judge here at the Central Family Court I must have made upwards of 30 special guardianship orders. I have, however, yet to encounter an application for such an order. On every occasion I have been invited by the local authority, whether opposed by another party or unopposed, to make the order of my own motion. That is not just the default position, but it appears to be the universal practice amongst authorities who use this court centre. This is the largest family court centre in England…….My purely personal impression is that the practice has changed in recent times.

23.

Whilst I do not suggest that these children should be the subject of care orders for their minority, the real balance in the case is in my judgment between special guardianship orders now and care orders (although not interim orders). The care plan under such care orders would be that if all goes well, then applications for special guardianship orders should follow in due course. By the expression ‘in due course’ I mean ‘when the new placements are regarded as settled and working well for the children’. In this case that might perhaps be in about a year from now…….

24.

……both sets of grandparents have been assessed in accordance with the Statute and the accompanying Regulations. The assessments are positive……

25.

My first concern is, however, that neither child is currently living with the proposed special guardians. During the course of argument, I mentioned that, last year, a letter had been written to interested parties by Keehan J, the Family Division Liaison Judge for the Midlands Circuit. It discussed the use of special guardianship orders. The view promulgated by Keehan J, as a result of a meeting with the chairs of the Circuit’s Local Family Justice Boards, was that “a special guardianship order should not be made, absent compelling and cogent reasons, until the child has lived for an appreciable period with the prospective special guardians.” Such guidance is not, of course, binding upon me but in passing I observe, with some deference, that it appears to amount to sound common sense……

28.

All this leads me to believe that someone has to be in charge of a process which oversees not just the move of the children to a new home, and their settling in, but also the implementation and progression of a closely controlled contact regime in circumstances where it is unclear what the parents’ reaction will be to the children’s move and equally unclear as to how they will handle time with the children in the very different circumstances which would apply……

30.

The next matter which concerns me is the position of the grandparents – within these proceedings as well as towards the children. As I listened to the case being developed, I did so in the complete absence of the grandparents – of the proposed special guardians. They were not parties. They were not represented. They were not present. They were not intended to be witnesses. Had an application been made – properly sponsored by the local authority which after all is the prime mover in this change to the children’s lives – then the grandparents would have been parties, represented, present and witnesses……

31 ….I have had the conduct of this case since the IRH on 3 February 2017. I could then have (i) made the grandparents parties (although that would not necessarily have secured representation for them); (ii) asked them to file a statement; (iii) invited them to give evidence; (iv) encouraged a special guardianship application at that stage. I did not take any of these steps, nor was I invited to do so……In truth, however, with the exception of my concerns surrounding their lack of participation in the process, the grounds on which I propose to reject the local authority case for special guardianship orders would have remained whatever step had been taken at the IRH. I know a great deal about the grandparents. I am not making special guardianship orders, but it is not because I lack information about the proposed special guardians.

32.

I invited the grandparents into court before they spoke to the professionals (all of whom were of course advocating special guardianship) so that at least they could hear the guardian, the representatives and myself debating the issues as the guardian gave evidence. They spoke with professionals afterwards. The result of this exercise was that they confirmed their wish to be special guardians immediately and for the children not to be subject to care orders…….I remain concerned, however, as to the process here. I am not convinced that the grandparents have been sufficiently involved. It is stating the obvious to observe that the effect of making an application to a court is to involve the applicants closely in the process.

34.

A short-term care order meets many of the concerns expressed in the previous paragraphs…..It is common ground that the transfer of the children to the grandparents, which is happening as I write this judgment, will not be delayed for want of special guardianship orders, or by any further assessment process.

It is evident that the judge recognised that the only realistic placement options that he had were with the paternal grandparents. His concern was the viability of those placements: not because they were unassessed but because they were untested in the specific context of the possible interference with them by the children’s mother and the father of S. It was in that context that on the merits the judge wanted to be assured that the control and parental responsibility which vests in special guardians would be sufficient to manage the relationship with the parents. The alternative was control and parental responsibility being vested in the local authority through care orders. The problem to be solved was whether the relationships and capabilities of the grandparents were strong enough or needed to be supported and tested before SGOs were made.

The solution to the problem was in the choice of order: SGO, care order or interim care order and an adjournment. The route to the solution lay in an evaluation of the evidence including oral evidence from professional witnesses, the parents and the proposed carers i.e. the paternal grandparents. It is clear from the judgment and from a transcript of the judge’s discussions with the advocates during the hearing that the judge had the problem and the solutions in mind. What was missing was a route to the preferred solution. Having identified the problem and the range of solutions the judge did not go on to evaluate that evidence. That necessarily meant that the propositions advanced in the discussion and the conclusions reached in the judgment take the form of assumptions that were not reasoned and which are now challenged.

It is submitted that the judge might have felt impelled to come to the conclusion that he did because of the constraints of time i.e. the proceedings had been delayed beyond the statutory time limit. That the 26 week time limit set out in section 32(1)(a) of the 1989 Act had already been exceeded and excused in accordance with the provisions of the legislative scheme is plain. It is also plain that there is an inevitable tension between the adverse effects of delay and the time needed for more complex welfare determinations where, for example, a placement order or special guardianship assessment are needed. In addition, there are many examples of cases where, as in this case, the realistic options for the care of a child change during the proceedings. They changed twice for these children: at the end of the FDAC process and when their interim carers were no longer available. That inevitably has the tendency to lengthen proceedings.

The time constraints in assessing long term placements are such that the statutory timetable can only be met if proposed placements are identified at the start of the proceedings. Provided the realistic options for the children are identified right at the beginning of the local authority’s intervention, delay can be minimised but not always avoided. It is minimised by the local authority undertaking planning pre-proceedings, or as soon as the threshold circumstances are in their knowledge, which identifies all of the realistic options for a child’s care. That usually involves detailed discussions with the extended family leading to the assessment of potential carers as early as possible. The planning process avoids firefighting i.e. responding to events and instead promotes a purposeful, positive stance on the assessment of potential carers to avoid dealing with them in sequence.

This court has not been asked to undertake a detailed analysis of the local authority’s planning process. I am content to assume that in the absence of any adverse comment they followed good practice given that they tried a rehabilitative process that was recommended, had identified alternative family carers who had been assessed and when that failed they kept the children in their extended families by coming to agreements with the paternal grandparents who were unequivocally recommended in a special guardianship assessment process that was unchallenged. There is no suggestion in this case that there was either inadequate evidence about the statutory and regulatory requirements or that they had not been properly assessed.

An analysis of the case management orders that were made in the proceedings demonstrate that there was an allocated judge for the FDAC process and that had the process been successful, the parents of S could have been identified as the realistic option for the care of both children within the statutory timetable of 26 weeks. Furthermore, contingency planning and the assessment of relatives for the purpose of special guardianship orders was undertaken relatively early in the proceedings and, again, with sufficient time to consider the then proposed special guardians within 26 weeks or with only a minimal need to extend time.

When the FDAC process came to an end, there was undoubtedly a pause described as being necessary to enable the parties to re-consider all the options, given that the children’s relatives who were at that time their carers had decided to withdraw. That pause was only intended to last a month but despite the intention of the judge in the FDAC who directed the case be listed for an Issues Resolution Hearing in September 2016 that did not occur and no-one appears to have taken any steps to prevent delay or identify the remaining issues. Fortunately, the matter came before His Honour Judge Cryan (for the first time) at an Issues Resolution Hearing on 28 November 2016 when firm judicial case management was restored. It was Judge Cryan who put in place the remaining work that needed to be done to identify whether the paternal grandparents were realistic options for the care of P and S. It is regrettable that the matter was not listed before him for an IRH in September 2016 as directed by the judge in the FDAC.

It is too often the case that it can be said that there has been inadequate planning either pre-proceedings or sufficiently early in proceedings to identify all of the realistic options for a child and then to complete assessments within a timetable that is appropriate for the child. It is important to state that this court has no basis for coming to such a conclusion in this case. Sequential planning may be necessary when unforeseen events occur and it achieves nothing in this case to spend any more time speculating whether the paternal grandparents could have been assessed earlier. All that needs to be said is that the best planning for a child includes contingency planning and care plans should always identify the realistic options for the care of a child as the contingencies to a local authority’s preferred option. In setting this out I am only repeating the good practice and mechanisms that this court described in W v Neath Port Talbot & Ors [2013] EWCA Civ 1227, sub nom Re W (Care Proceedings: Functions of Court and Local Authority) [2014] 2 FLR 431.

The real issue in this case was not whether the proceedings could be completed within 26 weeks if special guardianship orders were to be made subsequent to an FDAC process or whether the planning process properly identified the paternal grandparents at the earliest stage given that the realistic options changed and narrowed over time. Although that was the tension alluded to by the parties during the hearing, the import of the submissions was that the judge must have had in his mind the need to determine proceedings expeditiously and that he gave too much prominence to the impact of delay in the evaluation of factors he needed to consider and, necessarily, not enough weight to other welfare factors.

It has to be said that his judgment does not betray that approach and the assessment reports were complete and the evidence was filed before the final hearing within a timetable that was delayed but extended by the court for good reason when the FDAC process came to an end. One can speculate whether delay was a factor in his decision not to evaluate the benefits and detriments of an interim care order, because that would have lengthened the proceedings, but given that the proceedings were already outside the 26 week window, I am not prepared to conclude that the judge fell into that trap.

Given the comprehensive way in which my Lord, the President, dealt with these issues in Re M-F (Children) [2014] EWCA Civ 991, [2015] 1 WLR 909 at [25] to[28] and in Re S (A Child) [2014] EWCC B44 (Fam) at [28] to [34], all that needs to be said is that the circumstances in which it might be appropriate to extend the timetable for a child beyond 26 weeks because it is ‘necessary to enable the court to resolve the proceedings justly’ have been fully reasoned in a decision of this court which provides for all the circumstances of this case. It is not necessary to go any further.

As I remarked at [16] and [17] it was the absence of any testing of the assumptions raised in discussion which created the problem with which this court now has to grapple. The judge was concerned about the relationship between the grandparents and each of the children in the context of continuing discord with the mother and the father of S. It is also right to note that it was not until the commencement of the appeal before this court that the special guardianship support plans were agreed between the local authority and the grandparents. The judge identified what were potentially adverse factors to balance against the positive factors in the special guardianship assessments which might lead to the conclusion that a trial placement of the children was required before vesting parental responsibility and control in the grandparents. That deserved more than a cursory analysis not least because the local authority and the children’s guardian had come to a clear and agreed contrary opinion on the basis of rigorous assessment material that apparently demonstrated that the positives outweighed the negatives.

In order to test the assumptions the judge had described in his discussion with the advocates, he could have heard evidence about them and from that drawn conclusions. The judge records in his judgment that he heard some oral evidence but it is plain from his judgment that such evidence as there was either did not touch on the issues that he was raising or was unhelpful. That may be unsurprising given that the local authority and the children’s guardian disagreed with the judge and were agreed among themselves and also that no advocate was pursuing the issues the judge wanted to pursue. In that circumstance, as inquisitorial tribunals know, there must be an enhanced caution in a judge not to ‘simply’ rely on his or her own pre-conceptions or opinions and to ensure that as provisional conclusions are formed in judgment they are adequately tested so that they are soundly based on evidential conclusions.

It would also have been legitimate, if properly reasoned, for the judge to conclude that he needed more evidence with the consequence that the time for the proceedings might need to have been extended. In order to come to either conclusion the judge needed to identify the risk that he sought to protect the children against and reason the options that were open to him on the evidence. He ought to have tested his own assumptions and the opinions of the professional witnesses in oral evidence and by hearing evidence from the paternal grandparents. He would have been assisted by evidence from the mother and the father of S but, as has sadly been the case more than once in these proceedings, they had absented themselves and the judge was left with a history from which only inferences could be drawn. Had the judge reasoned his concerns on the evidence he would have had a proper basis for conducting an evaluation of the benefits and detriments of each order that was available to him.

In that context, it is not surprising that the judge’s evaluation of the merits of each option and the available orders was incomplete. The judge agreed with the parties that a child arrangements order was not in the interests of either child and he was right to do so on the merits. No-one pursues that option before this court. That left SGOs, full care orders and interim care orders with an adjournment.

I agree with the paternal grandparents of S that if and in so far as the judge needed more time to ensure that the relationship of the grandparents with the child and the parents was such that it was in the interests of each child to make an SGO, that could, if reasoned, have been an appropriate basis upon which to adjourn the proceedings. If the judge had not in his own mind resolved what form of final order was appropriate it could have been an appropriate course having regard to the guidance given by this court in W v Neath Port Talbot (supra) and would not have cut across the principles described in In re S (Minors) (Care Order: Implementation of Care Plan), In re W (Minors) (Care Order: Adequacy of Care Plan) [2002] UKHL 10, [2002] 2 AC 291.

Given that the judge did not analyse the benefits and detriments of the remaining orders that were available to him, this court is left with various propositions that the judge appeared to rely upon to justify the full care orders that he made. It suffices to say that in the absence of any evidential basis for the same, none of the propositions are sufficient and I regret to have to conclude that the imposition of care orders was not adequately reasoned. For the avoidance of doubt, I shall deal with three propositions that appeared to influence the judge: the first is that a full care order can be short term, the second is that there is some equivalence between the way in which a court should consider the permanence of an SGO and the permanence of an adoption order and the third is that the court was able to rely on the informal guidance about which complaint is made.

The concept of a short term care order within which the placements could be tested was raised by the judge as a justification for making full care orders. Aside from the welfare merits of the orders, which were not adequately reasoned, the concept of a short term order is flawed. There is no mechanism for a care order to be discharged on the happening of a fixed event or otherwise to be limited in time. The exercise of parental responsibility by a local authority cannot be constrained once a full care order is made other than on public law principles of unlawfulness, unreasonableness and irrationality. The judge should have reflected on the fact that if the local authority did not in due course apply to discharge the care orders themselves it would have been incumbent on the proposed special guardians to do so and to satisfy the test for leave to make that application without the benefit of legal aid, given that in the circumstance of a disagreement with the local authority it would be highly unlikely that the special guardians would be in receipt of funding from them.

Furthermore, the judge did not follow the guidance given in W v Neath Port Talbot and obtain from the local authority section 31A care plans for each of the children setting out the plan that he wanted them to pursue, namely a trial of the proposed placements by the local authority. The lack of scrutiny by the court of the plans that are required was contrary to section 31 (3A)(a) of the 1989 Act.

It is trite law that whatever the degree of permanence a special guardianship order provides for a child, the order does not have the same characteristics as an adoption order. It is not intended to have. Special guardianship was introduced to provide permanence in the care of children who cannot return to their birth families but where adoption is not appropriate. The legislative framework is different. It does not extinguish the parental responsibility of a child’s parents or terminate their legal relationship with the child, it can be varied or discharged (but not by a parent without the court’s leave) and it comes to an end when the child reaches the age of 18. There is no direct equivalence with an adoption order and the protections around it are accordingly different. If the judge intended to suggest that the statutory schemes are the same or very similar then that was inappropriate.

In particular, adoption orders cannot be made unless a child has had his or her home with an adoptive applicant for a prescribed period and an adoption order may not be made unless the court is satisfied that sufficient opportunities to see the child with the applicants have been given (see section 42 of the Adoption and Children Act 2002). There are no similar residence requirements in the statutory scheme that governs special guardianship orders. The scheme for approval of potential adopters and their matching with children is also completely different from the assessment provisions relating to special guardians where the assessment is child specific.

The regulatory scheme, that is reg 21 and the Schedule to the Special Guardianship Regulations 2005, as amended, prescribes the elements that are to be assessed which include an applicant’s ‘current and past relationship’ with the child. The regulations were amended by the Government in 2016 to include the need for assessments to be more robust in assessing whether a person is capable of caring for a child into adulthood: Special Guardianship review: report on findings, Government consultation response, December 2015. The opportunity to include provision for a period of assessment of a child living with a proposed special guardian was not taken by the Government. It is neither a statutory nor a regulatory requirement.

There is, however, considerable informed debate about what is good practice in the way in which SGOs are applied for and used. The ALC relied upon a body of research including from other jurisdictions to illustrate their submissions about that debate. The domestic research on the issues includes that funded by the Nuffield Foundation: Harwin J, Alrouh B, Palmer M, Broadhurst K and Swift S (2017) A national study of the usage of supervision orders and special guardianship over time (2007-2016), Universities of Brunel and Lancaster, which describes the increasing role of SGOs and the changing trend in the ratio between placement orders and SGOs which began in 2012 and Harwin J, Alrouh B, Palmer M, Broadhurst K and Swift S (2017) Considering the case for parity in policy and practice between adoption and special guardianship: findings from a population wide study, Universities of Brunel and Lancaster. Research was also undertaken by Wade et al in 2014: Wade J, Sinclair I, Stuttard L and Simmonds J (November 2014) Investigating Special Guardianship: experiences, challenges and outcomes, Social Policy Research Unit, University of York / BAAF which identifies that one in seven children were found not to have been living with their eventual carer before an SGO was granted. There are also ‘research in practice’ papers issued by the Department for Education in 2015 and the Association of Directors of Children’s Services in 2015 and a learned commentary on policy and practice published by Coram BAAF in 2016. It is plain that there are real policy and practice issues that arise about the way in which SGOs are used.

The ALC also drew attention to the difficulties proposed special guardians face who do not have the support, including the financial support of the local authority. They have to give three months’ notice of their intention to make an application to the local authority, pay significant fees to the court and try and obtain legal advice and representation, which will be means and merits tested for public funding. Unless the local authority consents, they have to obtain the leave of the court to make the application where one of the factors where a child is looked after by the local authority is the authority’s plans for the child’s future which will include consideration of whether any departure from the local authority’s plan might disrupt the child’s life to such an extent that the child would be harmed by it.

None of the materials to which I have referred was before the judge. Necessarily, he did not use them to justify taking a position about whether a child should live with a proposed special guardian before an SGO is made. It is not for this court to come to a conclusion on the merits or otherwise of that research given that this court has not had the opportunity to scrutinise and test the same. What the research does is to demonstrate the lacuna which the judge was identifying and trying to fill. There is a real need for authoritative guidance to sit alongside the statutory materials which would address what is evidence based, peer reviewed research, what is the reliable data about the outcomes that different practices achieve and the good practice that an analysis of those outcomes suggests. Such guidance would have been invaluable in this case and, we are told, in other cases where similar decisions need to be made.

In coming to the judgment that he did, the judge was influenced by informal guidance given by a High Court Judge in his role as a leadership judge. That is not the same as authoritative guidance or a practice direction for the reasons that follow. As it happens, the leadership judge is not the Family Division Liaison Judge for the Central Family Court but this court was told that the informal guidance that was referred to has become something of a commonplace in the Family Court. Its origins are commendable, i.e. training from academics and practitioners in the field which helps describe the range of expert material that is available to the court. I can assume for the purposes of this judgment that the expert opinion on which it is based is evidence based, peer reviewed and is held by a reasonable body of professionals in the expert discipline concerned although I emphasise that those essential methodological protections have not been demonstrated to this court.

There are two relevant letters from Keehan J. The first is dated 19 May 2016 and was apparently intended for family judges on the Midland Circuit. Its preamble recorded that “The following list of issues and points of practice have been seen and approved by all DFJs [designated family judges] and the Chairs of the Circuits’ (sic) LFJBs [local family justice boards]” and the advice that was given was that an SGO should not be made unless: “absent cogent reasons to the contrary, the child has been placed with the proposed SGO applicants/parties for a considerable period”.

The second letter is undated but was apparently written after 19 July 2016 in the same terms although a caveat is introduced that “it all depends on the particular circumstances of each case”. The caveat related to whether a court should make a care order and leave it to the local authority and the proposed special guardians to decide when the time is right to place a child and thereafter make an application for an SGO or keep the proceedings going by making interim care orders until the court is satisfied that the time is right for an application to be assented to. The second letter also identified a need to consider how more cases could be planned and managed to meet the statutory timetable. This court was told that other leadership judges elsewhere have written in similar terms although the only other example that this court is aware of makes it clear that the views expressed “are not intended to cut across the provisions of primary or secondary legislation”. If I may say so, that was both wise and appropriately cautious given the terms of the statutory scheme.

The recipients of the letter are not clear but it is understood from the submissions of the ALC that their members who are children’s lawyers have received a copy. In written submissions they alerted the court to a practice which is developing of reliance being placed on the letter to influence courts to take a particular course, for example, to encourage a court to make a full care order rather than to seek outstanding evidence about the merits in particular where a child has not lived with a prospective special guardian for any or any appreciable period of time. Although none of the parties in this case relied on the letter, it was cited by the judge in his judgment as part of the discussion, with the consequence that there is a perception that the contents of it were put in issue without any opportunity for challenge.

It is convenient to set out shortly the methods by which guidance is given which can be relied upon by the Family Court in an individual case. There is a summary by the President in Re B (A Child) [2017] EWCA Civ 1579, with which I respectfully agree. It is, of course, one of the functions of the Court of Appeal to give binding guidance (see, for example Piglowska v Piglowski [1999] 2 FLR 763 at 784H-785D). The High Court also gives persuasive guidance which if not given by the President will often record that the President has read and agreed with the guidance given. In each case, the court has had the opportunity to scrutinise and the parties have had the opportunity to test the propositions that are expounded.

Practice Directions may also be a source of invaluable guidance. They do not have the force of statute, are not scrutinised or approved by Parliament and cannot change the law but they are used to describe good practice. By convention, they go through an elaborate process of scrutiny by Rules Committees although there is no power in those committees to prescribe their content. They are the responsibility of the delegated judicial office holder and the Minister concerned. The power to issue Practice Directions includes that of the Lord Chief Justice with the agreement of the Lord Chancellor to give designated directions under section 13 and schedule 2 of the Constitutional Reform Act 2005 (which is delegated by the Lord Chief Justice to the President as Head of Family Justice) and the power in the President under section 81 of the Courts Act 2003. Under section 81(4) of the 2003 Act the agreement of the Lord Chancellor is not needed if the guidance concerns the interpretation of the law or the making of judicial decisions.

For many years the family courts have also had the benefit of non-statutory practice guidance issued by the President or with the President’s agreement by bodies such as the Family Justice Council, the Children Act Advisory Committee and the President’s Inter-Disciplinary Committee. Typically, guidance of this kind reflects a transparent process of inter-disciplinary working supported by evidence based research, a report of a working party or council that is scrutinised by the body concerned and then the adoption of recommendations by the judiciary, professional bodies and practitioners and/or Government. Guidance of this kind can always be challenged in court.

The complaint that is made about the informal guidance that was identified in this case is that it fell into none of these categories. It did not identify the research or basis upon which the guidance was given, it had not been the object of scrutiny or consultation in any environment where those responsible were accountable for that process, it was not transparently issued with an acknowledgement of responsibility, that is with the intention that its contents should be relied upon and it did not have a status that permitted a party to challenge its contents if a court or party in a particular case sought to rely on it.

The problem thus identified is not new. It was described in ‘Judicial Proposals for the Modernisation of Family Justice’ in 2012. The recommendations of that report were accepted by the then Lord Chief Justice and Government. They included the collation of “Frameworks of Good Practice’ to identify the research and authoritative guidance that exists. I would respectfully urge the President to consider whether further steps need to be taken to bring together existing research and guidance about special guardianship into a framework or to encourage agencies and practitioners within the family justice system to work together to promulgate good practice in a more coherent way. I would also urge the President to recommend to leadership judges who have good practice initiatives that they liaise with him before publication to ensure that the problem that has arisen in this case is avoided in future.

The mischief that is identified is not that guidance is needed or given. It is that it is relied upon without scrutiny or, where appropriate, the opportunity for challenge. If the opinion that arises from evidence based expert material is agreed by the parties in a case, for example because it is referred to by a witness who relies on it and it can be produced if necessary, then it may be that the court will consider it and, if appropriate, adopt it in its evaluation. That process is sometimes facilitated by the good practice being published as part of academic studies, recommendations of the Family Justice Council or other learned bodies of standing. What is important is that the process of reliance on the evidence is agreed. If the opinion is disputed, then if it is relevant and necessary, expert evidence on the point in question can be directed by the court.

The informal guidance that was relied upon in this case had none of those protections. It did not identify the evidence based research upon which it relied nor was it scrutinised by any of the methods I have described. The opinion it expresses might be right but it is not an appropriate vehicle for an opinion of the kind expressed to be relied upon in court in an individual case. For these reasons, it was understandable but inappropriate for the judge to have relied on the letters identified. That is not to suggest that the use of research and references to research by witnesses or judges is to be disapproved or discouraged. Quite the contrary, it has been the practice in family proceedings since at least the days of Lord Justice Thorpe’s inter-disciplinary committee for good practice to be identified so that it can be used. The key to its use is the identification of the materials upon which the guidance is based so that the court can decide whether there is an issue that arises and, if so, whether expert evidence is necessary.

The statutory scheme that applies to special guardianship orders has previously been considered by this court in Re H (A Child) (Analysis of realistic options and SGOs) [2015] EWCA Civ 406, [2016] 1 FLR 286 CA. Given that the judge cited that authority it was clearly in his contemplation. The paternal grandparents were able to make applications for special guardianship orders because sections 14A(3) and 14(5)(c) when read together with section 10(5)(c)(ii) of the 1989 Act provide that they are entitled to do so if they have the consent of the local authority where the child is in the care of that local authority. That would have been so in this case had anyone directed their minds to the situation.

In the absence of the local authority’s consent, the grandparents would have been able to apply for leave to make an application under section 10(9) of the 1989 Act where the factors to be considered by the court are specified. In a case where the local authority does not consent the leave application is an important protection for the child and the child’s parents. It is not a rubber stamp. Where leave is granted, an appropriate balance is struck between the applicants, the child, the child’s parents and the local authority. It is relatively common to find that local authorities who give consent to an application being made, that is who support the application on the merits, will help fund the applicant by providing representation. That happened in this case when the matter came on appeal and after a case management indication to that effect was given by this court. It ought to have happened earlier.

The residual power in the court to consider making a special guardianship order of its own motion in section 14A(6)(b) of the Act should not be the normal or default process because it avoids the protections that I have just referred to. That is not to say that circumstances will not arise where that residual process is in the interests of the child and the court is able to have regard to the protections in sections 14 and 10 in its decision making, but it should not be the normal process. Not only does it tend to avoid the protections in the statutory scheme but it tends to avoid good planning by the local authority and the court which will include identifying the status of the prospective special guardians, how they will achieve effective access to justice and such case management directions as will provide fairness to all parties by notice of the proceedings, the disclosure of evidence and the ability to take advice. In so far as the judge indicates that the Central Family Court and local authorities appearing in it had yet to come to terms with the need to follow this court’s guidance in Re H, that time has now arrived.

It is difficult to understand why steps were not taken on 3 February 2017 to consider effective access to justice for the grandparents. It was clear from the recordings in the order made that the key issues in the case included whether each of the children should “reside in a family placement under an SGO”. It may be that the apparent consensus between the local authority, the children’s guardian and the grandparents on the ultimate order they all sought obscured the immediate procedural issue before the court. Given the procedural unfairness that undoubtedly was the consequence, I have no hesitation in coming to the conclusion that it was wrong not to have made appropriate provision for the grandparents to obtain effective access to justice at the final hearing. To leave them on the sidelines without party status, without documents and without advice and without any mechanism being identified for the parents of S to cross examine them on their proposals was unfair in more than one respect. From the children’s perspective, it meant that part of their case was assumed to be incomplete when it could have been tested.

The solution would have been either to direct that an application for an SGO be made so that case management directions on that application relating to party status, disclosure and time for advice could be made or for case management directions to be made that otherwise secured the same procedural protections. At the Issue Resolution Hearing on 3 February 2017 the facts were such that the grandparents were a realistic option for the care of the children. The direction for an SGO assessment which is a mandatory pre-condition to the making of an order had been made and the reports were complete. I am accordingly at a loss to understand why directions were neither asked for nor made that dealt with how that application would be considered by the court. The consequence is stark even from the judge’s own judgment. The grandparents did not know what was happening, did not have the evidence upon which the court was making a decision, were unable to take advice and in the event, in my judgment, did not have effective access to justice. That was not in the interests of the children. The procedure was accordingly unfair.

For all these reasons, I would allow the appeal, set aside the care orders that were made and substitute special guardianship orders in favour of the paternal grandparents for each child. There is now no issue that an SGO is the only appropriate order to make in all the circumstances and the uncontradicted assessments of the grandparents and the subsequent evidence of the success of the placements more than adequately provide the basis for the same.

I agree with the judgment of Sir Ernest Ryder and with the supplementary comments by the President of the Family Division.

I also agree. By way of supplement to the judgment of Sir Ernest Ryder, with whose reasoning and analysis I agree, I add a few observations.

Published research and practical experience demonstrate the increasing use in recent years of SGOs in the context of care proceedings, sometimes alone, sometimes in conjunction with a supervision order. Practical experience and much anecdotal material have identified two important issues giving rise to much concern: first, the not infrequent examples of cases in which a SGO is proposed for the placement of a child with a relative with whom the child has never previously lived and whose relationship with the child may be tenuous or non-existent; secondly, worries that the assessments relied on by the court in deciding whether or not to make a SGO are not always as rigorous as might be thought appropriate. In terms of process, these concerns raise serious questions as to how they are best addressed having regard to the imperative to conclude care cases within 26 weeks as required by section 32(1)(a)(ii) of the Children Act 1989.

My Lord has referred to the approval by this court in In re M-F (Children) (Care Proceedings: Extension of Time Limit) [2014] EWCA Civ 991, [2015] 1 WLR 909, of what I had previously said in In re S (A Child) (Interim Care Order: Residential Assessment) Note [2015] 1 WLR 925, same case Re S (Parenting Assessment) [2014] 2 FLR 575. It may be helpful to set out some of what I said in In re S.

Referring to the 26-week rule, I said this (In re S, para 24):

“Section 32(1)(a)(ii) does not describe some mere aspiration or target, nor does it prescribe an average. It defines, subject only to the qualification in section 32(5) and compliance with the requirements of sections 32(6)(7), a mandatory limit which applies to all cases.”

Turning to consider section 32(5), I set out (paras 26-27) the general principle that, if the court does not have the kind of evidence it needs if it to be properly equipped to decide the issues before it, then an adjournment must be directed, even if this takes the case over 26 weeks. For the outcome is not to be determined by rigorous adherence to an inflexible timetable if justice is thereby potentially denied. I went on (para 29) to “wholeheartedly endorse” the words of Pauffley J in In re L (A Child) (Family Proceedings: Practice and Procedure) (Practice Note) [2014] EWHC 279 (Fam), [2014] 1 WLR 2795, [2014] 1 FLR 1384, para 40: “Justice must never be sacrificed on the altar of speed.”

I continued (paras 30-34):

“30

So despite the imperative demand of section 32(1)(a)(ii) of the Children Act 1989, there can be exceptions. But … it is vital to recall the equally imperative language of sections 32(5) and 32(7). An extension beyond 26 weeks is to be permitted only if it is “necessary to enable the court to resolve the proceedings justly” … Moreover, extensions are “not to be granted routinely” and require “specific justification”.

31

In what circumstances may the qualification in section 32(5) apply?

32

This is not the occasion for any elaborate discussion of a question which, in the final analysis, can be determined only on a case by case basis. But some preliminary and necessarily tentative observations are appropriate.

33

There will, as it seems to me, be three different forensic contexts in which an extension of the 26-week time limit in accordance with section 32(5) may be “necessary”:

(i)

The first is where the case can be identified from the outset, or at least very early on, as one which it may not be possible to resolve justly within 26 weeks. Experience will no doubt identify the kind of cases that may fall within this category. Four examples which readily spring to mind (no doubt others will emerge) are (a) very heavy cases involving the most complex medical evidence where a separate fact finding hearing is directed … (b) FDAC type cases …, (c) cases with an international element where investigations or assessments have to be carried out abroad and (d) cases where the parent’s disabilities require recourse to special assessments or measures …

(ii)

The second is where, despite appropriately robust and vigorous judicial case management, something unexpectedly emerges to change the nature of the proceedings too late in the day to enable the case to be concluded justly within 26 weeks. Examples which come to mind are: (a) cases proceeding on allegations of neglect or emotional harm where allegations of sexual abuse subsequently surface, (b) cases which are unexpectedly “derailed” because of the death, serious illness or imprisonment of the proposed carer, and (c) cases where a realistic alternative family carer emerges late in the day.

(iii)

The third is where litigation failure on the part of one or more of the parties makes it impossible to complete the case justly within 26 weeks …

34

I repeat, because the point is so important, that in no case can an extension beyond 26 weeks be authorised unless it is “necessary” to enable the court to resolve the proceedings “justly”. Only the imperative demands of justice – fair process – or of the child’s welfare will suffice.”

I draw attention to the reference in para (i)(b) to what I referred to as “FDAC type cases” and to the reference in para (ii)(c) to what I described as “cases where a realistic alternative family carer emerges late in the day.” Having described what I meant by FDAC type cases, I continued (para 38):

“Viewed from a judicial perspective a vital component of the FDAC approach has to be a robust and realistic appraisal at the outset of what is possible within the child’s timescale and an equally robust and realistic ongoing appraisal throughout of whether what is needed is indeed being achieved (or not) within the child’s timescale. These appraisals must be evidence based, with a solid foundation, not driven by sentiment or a hope that “something may turn up”. Typically three questions will have to be addressed. First, is there some solid, evidence based, reason to believe that the parent is committed to making the necessary changes? If so, secondly, is there some solid, evidence based, reason to believe that the parent will be able to maintain that commitment? If so, thirdly, is there some solid, evidence based, reason to believe that the parent will be able to make the necessary changes within the child’s timescale?”

Returning to care cases where the making of a SGO is in contemplation, an acute problem arises when, as not infrequently happens, the prospective special guardian is identified late in the day, perhaps before the placement has been sufficiently assessed and tested. What is the court to do? The answer, in my judgment, is in an application of the principles to be found in the passages from In re S set out above.

The first question is whether the proposed special guardian is a ‘runner’, to adopt the language of McFarlane LJ in In re W (A Child) (Adoption: Grandparents’ Competing Claim) [2016] EWCA Civ 793, [2017] 1 WLR 889, [2017] 2 FLR 31, para 70. This appraisal, to adopt the language of In re S, must be “evidence based, with a solid foundation, not driven by sentiment or … hope.” But the appraisal, assessment or evaluation (call it what you will) need not necessarily be too lengthy or too searching at this stage; what is sometimes referred to as a viability assessment or something similar may well suffice. If the proposed special guardian is ruled out at this stage, then so be it. If not, the judge will need to consider carefully what further steps need to be taken, in all the circumstances of the particular case, before the court can be satisfied that the proposed SGO should indeed be made.

What further assessment, addressing which issues, is necessary to enable the judge to come to a properly informed conclusion? How long will the necessary assessment take – something on which the professional opinion of the proposed assessor is likely to be of crucial importance? If the child has never lived with, or has only a tenuous relationship with, the proposed special guardian, what steps need to be taken and over what period to test the proposed placement? These are some of the questions the judge may need to have answered; no doubt there will be others.

If the answer to these questions demonstrates that the process cannot be completed justly, fairly and in a manner compatible with the child’s welfare within 26 weeks, then time must be extended. There can be – there must be – no question of abbreviating what is necessary in terms of fair process, and necessary to achieve the proper evaluation and furthering of the child’s welfare, by concern about the possible impact of such necessary delay upon the court’s performance statistics. In relation to SGOs, as elsewhere, justice must never be sacrificed upon the altar of speed.

These are issues which Keehan J sought to address in the two documents to which my Lord has referred. I agree with my Lord that there is a real need for authoritative guidance to sit alongside the statutory materials and with his characterisation of the way in which such guidance should be prepared. However, and not least given all the research which the ALC in particular has brought to our attention, including that to which my Lord has referred, this is not an exercise which we are properly equipped to undertake. Nor, with respect, was Keehan J, even though, as his second letter makes clear, he had the benefit of the no doubt invaluable input of all the Designated Family Judges and Chairs of all the Local Family Justice Boards on the Midland Circuit.

Given the nature of the relevant kinds of questions, as I have described them above, which the court will need to address, and the way in which, as described by my Lord, the task needs to proceed, the appropriate body to undertake the investigation of what form any guidance should take is the multi-disciplinary Family Justice Council. I therefore propose to invite the Family Justice Council to undertake this task and to make available to it all the research and other relevant materials which were put before us.