Re U (A Child)

This is the judgment of the Court to which all members of the court have contributed. It concerns a little girl, U, born on the 11^th July 2001.

The case returned to the Court of Appeal on the 8^th December for further consideration of the refusal of the Court of Appeal (The President, Thorpe LJ and Mantell LJ) to grant permission to appeal on the 4^th March 2004 (see the judgment of the court of the 14^th May 2004, reported [2004] 3 WLR 753. The further application is made subject to the principles in Taylor v Lawrence [2003] QB 528, [2002] EWCA Civ 90 and the provisions of CPR 52.17.

We gave our decision to refuse the renewed application to the parties on the 14^th December2004 and provided a short Press release. We now give our reasons for our decision.

CPR 52.17 is headed “Reopening of final appeals”. 52.17(1) provides:

“The Court of Appeal or the High Court will not reopen a final determination of any appeal unless –

(a)

it is necessary to do so in order to avoid real injustice;

(b)

the circumstances are exceptional and make it appropriate to reopen the appeal; and

(c)

there is no alternative effective remedy.”

By paragraph (2) “appeal” includes an application for permission to appeal. There is no doubt that this rule is specifically intended to reflect the reasoning in the judgment of this court in Taylor v Lawrence. While this authority is very well known in the legal profession, it seems to us especially important to explain its effect in the context of our decision in this case. There is a pressing public interest, which has to be understood, in confining the circumstances in which the court will reopen an appeal that has already been finally determined to cases of the most exceptional kind; and this is so across all areas of the law, including family law. Accordingly we make no apology for addressing the authority of Taylor v Lawrence in some detail.

In that case judgment had been given against the defendants in a county court action. They appealed in part on the ground of an appearance of bias by the judge. It emerged that the judge and his wife had engaged the claimants’ solicitors to amend their wills the night before judgment was given in the action. The appeal was dismissed in January 2001. Later, the defendants learnt that the judge had not paid for the services provided by the solicitors during the trial. The defendants applied for permission to reopen their appeal. The application was heard and determined by a constitution of five judges of this court (Lord Woolf CJ, Lord Phillips of Worth Matravers MR, Ward, Brooke and Chadwick LJJ). As is made clear in the judgment of the court delivered by the Lord Chief Justice, a court of five was exceptionally constituted in order to resolve the important issue of jurisdiction that arose in the case, namely whether there is any power at all to reopen an appeal which has been finally determined.

It was submitted by Mr Corner acting as advocate to the court that the Court of Appeal enjoys no such power. He cited much learning, not least the decision in Flower v Lloyd 6 Ch D 297, in which Sir George Jessel MR concluded at 300- 301 that if the Court of Appeal “has once determined an appeal, it has no further jurisdiction”. There were some dicta which went the other way. The court noted more than once that a judgment obtained by fraud could be set aside by order of the court in a fresh action brought for the purpose, and such recourse could no doubt be had as readily in a case where the judgment impugned was that of the Court of Appeal as where it had been given at first instance. But this, of course, does not in truth involve the reopening of concluded proceedings in the sense in which that is sought to be done here, and was attempted in Taylor v Lawrence. It is one thing to re-litigate an issue where the vehicle for doing so is the very same set of proceedings in which the issue had earlier been concluded. It is quite another to bring a fresh action to impugn an earlier judgment.

As the court pointed out at 540, the availability of a claim to set aside a judgment obtained by fraud does not mean there are no other situations where serious injustice may occur if there is no power to reopen an appeal. And so the court proceeded to consider, and ultimately to accept (though in guarded terms), the submission by counsel for the defendants to the effect that the Court of Appeal indeed possessed the jurisdiction to reopen previously concluded decisions of its own.

The argument was put three ways. First it was said that the relevant jurisdiction was to be found in the County Court Rules 1981, as transposed to the Court of Appeal by force of s.15(3) of the Supreme Court Act 1981 (which we will not cite). Secondly, it was said that there existed a jurisdiction analogous to that discussed in Wood v Gahlings, The Times, 29 November 1996. Lastly, another analogy was put forward, with that class of cases in which, in special and exceptional instances, the Court of Appeal has exercised a power to reopen its judgments in order to oversee and regulate the implementation and enforcement of its own orders.

The first ground was rejected in short order and we need say no more about it. Wood v Gahlings was a case where there was a further appeal after a retrial. At the further appeal it was submitted that a fraud had been practised on this court hearing the appeal after the first trial. It was held that the accusation of fraud was not established, but observations were made obiter by Lord Woolf MR (as he then was) and Aldous LJ “suggesting that there was power in this court in the appropriate circumstances to intervene if actual fraud could be established”. The court in Taylor v Lawrence observed:

“Fraud has always been treated as an exceptional case… If, however, it is arguable that the Court of Appeal is able to reopen a decision where it has been obtained by fraud, this opens the door to the argument that there is jurisdiction to reopen an appeal in other exceptional cases.”

What was in contemplation here was not the bringing of a fresh action to set aside a judgment based on fraud, but rather the actual reopening, and thus rehearing, of the appeal in which the fraud was said to have been practised.

As regards the third argument raised to support a jurisdiction to reopen a concluded appeal, the court in Taylor v Lawrence drew attention to two family cases, In re C (A Child) and In re J (A Child), both unreported.In the first of these Dame Elizabeth Butler-Sloss P drew attention to the fact that in Hague Convention cases the Court of Appeal “inevitably has a continuing jurisdiction for the purpose of implementation; as it has in other forms of civil litigation, but particularly in family cases”. However, with respect it is to be noted that this form of “continuing jurisdiction” is in principle no more controversial than the court’s everyday grant of a liberty to apply when in handing down a final judgment it perceives that there may be difficulties, not then and there identifiable, in carrying the court’s order into effect; and the court in In re C was not saying anything at variance with this.

In the second case, however (In re J), the court made observations which touch more closely the true question of jurisdiction raised in Taylor v Lawrence. Dame Elizabeth Butler-Sloss P referred to “the possibility of a residual power in the court with the final decision-making process [to set aside an earlier final order]… if… there is some factor outside the decision in the case itself… It would have… to be a factor that flawed the decision, and a factor which was outside the ambit of the decision itself”. Robert Walker LJ (as he then was) referred to “a very limited residual class of cases in which in quite extraordinary circumstances a decision might be revisited.” And he recalled the example given by Lord Donaldson of Lymington MR in which by mistake a Lord Justice determined an appeal by reference to the wrong set of case papers. May LJ envisaged, but did not affirm, the possibility that the Court of Appeal enjoyed a jurisdiction similar to that exercised by the House of Lords in Ex p. Pinochet (No 2) [2000] 1 AC 119.

The court in Taylor v Lawrence then entered into some discussion of the Pinochet case, emphasising that the reasoning which allowed an earlier order of the House of Lords to be revisited depended on the status of the House as a final appellate court. That was a premise not automatically applicable to the Court of Appeal; but “… in some cases no appeal will lie to the House of Lords and in others an appeal to the House of Lords will not be appropriate. Thus for practical purposes in many cases this court is the final court of appeal. The difference between this court and the House of Lords is therefore one of degree…” Then a little later:

“It is the reality of the situation which means that we cannot, as Sir George Jessel MR did in Flower v Lloyd…, take refuge in the fact that there is an alternative remedy. If there is no effective right of appeal to the House of Lords and this court is the only court which can provide a remedy then in our judgment there can arise the ‘exceptional circumstances’ to which Russell LJ referred in In re Barrell Enterprises…”

Then it was stated that the Court of Appeal, albeit entirely a creature of statute, must possess implicit powers “to control its own procedure so as to prevent it being used to achieve injustice”. Passages from the speeches of Lord Diplock in Bremer Vulkan Schiffbau und Maschinenfabrik [1981] AC 909and of Lord Morris of Borth-y-Gest in Connelly v DPP [1964] AC 1254are cited. These observations follow, appearing towards the end of the section of the judgment in Taylor v Lawrence dealing with jurisdiction:

“54… The residual jurisdiction which we are satisfied is vested in a court of appeal to avoid real injustice in exceptional circumstances is linked to a discretion which enables the court to confine the use of that jurisdiction to the cases in which it is appropriate for it to be exercised. There is a tension between a court having a residual jurisdiction of the type to which we are here referring and the need to have finality in litigation. The ability to reopen proceedings after the ordinary appeal process has been concluded can also create injustice. There therefore needs to be a procedure which will ensure that proceedings will only be reopened when there is a real requirement for that to happen.

55.

One situation where this can occur is a situation where it is alleged, as here, that a decision is invalid because the court which made it was biased. If bias is established, there has been a breach of natural justice. The need to maintain confidence in the administration of justice makes it imperative that there should be a remedy. The need for an effective remedy in such a case may justify this court in taking the exceptional course of reopening proceedings which it has already heard and determined. What will be of the greatest importance is that it should be clearly established that a significant injustice has probably occurred and that there is no alternative effective remedy. The effect of reopening the appeal on others and the extent to which the complaining party is the author of his own misfortune will also be important considerations…”

It is clear that whenever the residual jurisdiction established by the judgment in Taylor v Lawrence is sought to be invoked, the court must be satisfied that the case falls within the exceptional category there described before it will accede to the application and reopen the case. One may without levity ask the question, how exceptional is exceptional? The language used by the court (and in CPR 52(17)(1)) is necessarily general: apart from the descriptive phrase “exceptional circumstances”, the requirements are that the probability of a significant (CPR – “real”) injustice must be clearly established, and that there be no effective alternative remedy.

It seems to us to be of the first importance to distinguish the kind of case in which the residual jurisdiction might properly be invoked from one where all that is said is that there exists fresh evidence which could have a substantial effect on the outcome of the case. The principles upon which this court, on a first appeal brought in the ordinary way, will admit new evidence and decide the case in light of it (or remit the matter for a fresh trial) are well known. They are encapsulated in the decision in Ladd v Marshall [1954] 1 WLR 1489 (which is referred to in Taylor v Lawrence) with some little modification since the CPR came into force into which it is unnecessary to go. There are essentially three requirements: (a) the evidence could not have been obtained with reasonable diligence for use at the trial; (b) the evidence would probably have had an important (though not necessarily decisive) influence on the result of the case; (c) the evidence must be credible (though it need not be incontrovertible).

These requirements are of an altogether different and less demanding order from what must be shown to invoke the residual jurisdiction to reopen a concluded appeal. The Ladd v Marshall rules may promote the admission of fresh evidence where there is no more than a possibility that an injustice, in the shape of a factually erroneous result at the trial, has been perpetrated. The possibility, it is true, must be more than merely fanciful. But the Taylor v Lawrence jurisdiction can in our judgment only be properly invoked where it is demonstrated that the integrity of the earlier litigation process, whether at trial or at the first appeal, has been critically undermined. We think this language appropriate because the jurisdiction is by no means solely concerned with the case where the earlier process has or may have produced a wrong result (which must be the whole scope of a fresh evidence case), but rather, at least primarily, with special circumstances where the process itself has been corrupted. The instances variously discussed in Taylor v Lawrence or in other learning there cited are instructive. Fraud (where relied on to reopen a concluded appeal rather than found a fresh cause of action – Wood v Gahlings); bias; the eccentric case where the judge had read the wrong papers; the vice in all these cases is not, or not necessarily, that the decision was factually incorrect but that it was arrived at by a corrupted process. Such instances are so far from the norm that they will inevitably be exceptional. And it is the corruption of justice that as a matter of policy is most likely to validate an exceptional recourse; a recourse which relegates the high importance of finality in litigation to second place.

By contrast a fresh evidence case may disclose nothing to suggest that the process of justice has been corrupted. It may just be that fresh evidence, not previously available or at least not reasonably obtainable for use at the trial, has come along which had it been admitted at trial might have had a significant impact on the result. Such instances are, broadly, the stuff of first-time-round appeals based on new evidence. They are, categorically, not the proper subject-matter of Taylor v Lawrence applications. The principle of finality yields so as to allow a first appeal on Ladd v Marshall grounds. But it will prevail so as to disallow a second appeal – a Taylor v Lawrence application – on such grounds.

Even so, is there no case in which fresh evidence might justify a Taylor v Lawrence application in the absence of some other factor which has corrupted the litigation process? Since, as the court’s judgment shows, the ultimate rationale of Taylor v Lawrence is the correction of injustice, it would surely be wrong to answer with an absolute negative. The examples we have mentioned – bias, and the others – cannot be said to constitute the only proper instances for the jurisdiction’s application. Nothing in the reasoning justifies so strait-jacketed an approach. Indeed, it does not justify a brightline rule by which, for any application to reopen an appeal to succeed, it must be shown that the litigation process has been corrupted, as we have put it. In our view the case where the process has been corrupted is the paradigm case: not necessarily the only case.

However if the discovery of fresh evidence is ever to justify reopening a concluded appeal, the case must at least have this in common with the instances of corrupted process: the injustice that would be perpetrated if the appeal is not reopened must be so grave as to overbear the pressing claims of finality in litigation – especially pressing where what is contemplated is a second appeal. Finality is itself a function of justice, and one of great importance. Here the observations of Lord Wilberforce in The Ampthill Peerage [1977] AC 547, set out in Taylor v Lawrence, particularly repay attention. Now, it is to be noted that the strictures to be found in the Ladd v Marshall principles, not least the requirement that the fresh evidence proffered could not have been obtained with reasonable diligence for use at the trial, are justified by the demands of finality; and Ladd v Marshall applies to a first appeal. To entertain a second appeal on fresh evidence grounds, greater strictures are required. It follows that, as we have already indicated, Ladd v Marshall grounds will in principle not suffice to justify a second appeal.

What will? In our judgment it must at least be shown, not merely that the fresh evidence demonstrates a real possibility that an erroneous result was arrived at in the earlier proceedings (first instance or appellate), but that there exists a powerful probability that such a result has in fact been perpetrated. That, in our view, is a necessary but by no means a sufficient condition for a successful application under CPR 52.17(1). It is to be remembered that apart from the requirement of no alternative remedy, “[t]he effect of reopening the appeal on others and the extent to which the complaining party is the author of his own misfortune will also be important considerations” (Taylor v Lawrence, 547). Earlier we stated that the Taylor v Lawrence jurisdiction can only be properly invoked where it is demonstrated that the integrity of the earlier litigation process, whether at trial or at the first appeal, has been critically undermined. That test will generally be met where the process has been corrupted. It may be met where it is shown that a wrong result was earlier arrived at. It will not be met where it is shown only that a wrong result may have been arrived at.

This was never a case in which the integrity of the earlier litigation process, whether at trial before Bracewell J or at the first appeal, had been critically undermined. As must be plain from our treatment of the new evidence itself, it is very far from being a case in which it is demonstrated that a wrong result was earlier arrived at. The application to reopen the earlier appeal process pursuant to CPR 52.17(1) was misconceived in principle. That said, given the acute importance of the case to the parties and the interests of the child, we propose to look at the detail of the material now put forward. It is by no means to be supposed that such a narrative judgment will be appropriate in every Taylor v Lawrence application.

The history of the case prior to the hearing of the permission to appeal in the Court of Appeal is set out in the previous judgment. Since the refusal of the permission to appeal, the decision as to the future of the child U was made by Kirkwood J on the 25^th March 2004. In his judgment, he dealt with events between the judgment of Bracewell J dated 7^th November 2002, and the hearing before him.

Kirkwood J found that the mother had failed to co-operate with the agencies or with the court after the decision was given by Bracewell J. He said

“It is a matter of regret and, I add, of sorrow that U’s mother has not felt able to lend her co-operation to following through with any of this work. She has not seen the need to participate in any sort of therapeutic programme whether to address her own personal problems or her relationship with her child. That is very serious. No court could reasonably arrange the placement of this child with this mother without having a very full understanding of her and very clear professional guidance that whatever went wrong has been addressed; that U would be physically and emotionally safe in her mother’s care. Indeed the mother surely recognises that. Her decision not to give any oral evidence to the court at this hearing (itself an unusual decision in such circumstances) was one taken deliberately and with the benefit of advice, and left the court without any opportunity of making an assessment of its own. Knowledge about the mother is not far advanced from the material that was before Mrs Justice Bracewell.”

This was sad since there had been at least a chance that the child might have been reunited with her mother after further investigations and assessments.

The maternal grandmother looked after U for a period but she did not accept that her daughter posed a risk to her granddaughter and she was considered unsuitable to be a carer. She agreed that long term she could not care for the child. She made no application to the court.

It was then agreed, in consultation with the paternal family, that the paternal grandmother should take over the care of U with their support. This agreement was based upon the clear understanding that the father and the rest of his family who lived in the home of the paternal grandparents accepted the findings of Bracewell J and would co-operate in excluding the mother from playing any role in the child’s life unless it was properly supervised.

Kirkwood J’s judgment recorded the way in which the family, particularly the mother and the father, deliberately misled the local authority as to their involvement in the care of the child and their breaches of clear provisions to ensure that the mother did not have unsupervised contact with U.

The father, in a statement of evidence dated 16^th June 2003 accepted the findings of Bracewell J that at that time the mother remained a risk to his daughter. The local authority with expert help engaged in a full and careful assessment of the paternal family. It was not completed until later than expected, that is to say in August 2003. The detailed assessment included a passage in which the writer formed the view that the father sounded genuine when he agreed with his family that he should terminate his relationship with his wife after she made it clear that she did not need and would not accept any therapy or help to deal with the issues nor co-operate with professionals. The wife throughout did not accept the end of their marriage. The father told the psychologist on three occasions in September 2003 that he had separated from his wife although he had subsequently seen her to try to stop her pestering him. His solicitor wrote to the mother’s solicitors on the 19^th September 2003 warning that if the mother continued to contact and harass him, he would not hesitate to seek injunctive relief against her and that U’s placement with the paternal family was at risk from her actions.

The local authority’s suspicions were, nonetheless, aroused and in accordance with its powers under the Regulation of Investigatory Powers Act 2000, as well as seeking guidance from the Court, observation of the family was undertaken. The parents were indeed seeing each other in clear breach of the clear preconditions set out in the agreement between the social workers and the paternal family including the father. The mother on the 8^th October was shown the results of the observations and on the same day the father disappeared with U, returning on the 10^th October after intervention from his family. U was immediately removed and placed with a foster family.

The father disclosed that he did not and never had accepted that his wife had harmed the child and he intended to clear her name. He further disclosed that he had remained in a relationship with his wife throughout the period. He had deceived and lied to his family, the professionals working on the case and the court. So had the mother. The deception was elaborate, including, as it did, the letter from his solicitor to the mother’s solicitor alleging harassment by the mother.

Kirkwood J made a care order on the basis of a care plan that U should be placed for adoption. She is now the subject of a freeing order and we were told by Miss Rowe QC for the local authority that active steps were being taken to place her with an adoptive family, subject of course to the outcome of the present application.

Taylor v. Lawrence applications are comparatively rare and there is little established practice directing their preparation, particularly if ordered in for a hearing on notice with appeal to follow if permission granted. Manifestly the further application required careful judicial management and Thorpe LJ assumed control of it on behalf of the Court of Appeal. Permission was granted to the mother to disclose the documents in the case to Professor D, who had not previously been involved in the case, in July and August 2004. Thorpe LJ directed an oral hearing on notice on the 28^th September. His order contained further directions to ensure orderly and timely preparation. At that stage the application was supported by a statement from the solicitor for the mother, Mrs Harman, dated 8^th September 2004, a report from Professor D dated 11^th July 2004 and a second report from him dated 30^th August 2004. The second report was stated to be a preliminary report prepared in haste and without the benefit of all relevant documents. Accordingly the first direction was that the applicant should file any further evidence by 8^th October, the second required the evidence in response by 5^th November (subsequently extended to 12^th November by agreement between the parties), and the third the evidence in reply by 19^th November. The order then provided for skeleton arguments to be filed by 26^th November and that any application for further directions was to be made to Thorpe LJ, in writing in the first instance.

The applicant elected not to file any further evidence in support of the application. That must seem surprising given the preliminary nature of Professor D’s second report. Accordingly the evidence in response from the respondent’s experts, Dr D, Dr R and Dr H, was directed to Professor D’s first and second reports. Dr R’s response was dated 2^nd November, Dr D’s 9^th November and, although Dr H’s report was delivered on the 11^th November, it was unfortunately not served on the applicant until the 17^th November.

Obviously at that stage an application by the applicant for an extension of say seven days to the deadline for filing evidence in reply would have been likely to succeed. No such application was received and no evidence in reply was filed.

In the meantime the parties had agreed amongst themselves that there should be a meeting of experts. That was revealed to the court by a letter from the Local Authority dated 13^th October (although dictated 7^th October), a letter from the applicant’s solicitors dated 18^th October and a reply from the Local Authority dated 21^st October. The letter of the 13^th October sought directions under four paragraphs. The first two paragraphs were agreed. The third paragraph was also agreed to the extent that it was accepted that in default of agreement as to the list of questions to be put to the experts at their meeting, then the parties were free to submit their own list of questions to be circulated to all the experts in advance.

It was the last direction that both in negotiation and implementation created difficulties. The Local Authority proposed the conventional direction that the meeting of experts should be minuted by the solicitor for the child, the costs of the exercise to be shared between the parties. The applicant’s solicitor strongly objected. She maintained that the Guardian had “formed a view about our client and the risk that she poses to the child”. Accordingly she asserted that the independence of the solicitor for the child was also compromised. Consequently she stated that she was seeking a retired High Court Judge to moderate the meeting. In response to this unjustified attack on the Guardian’s impartiality the Local Authority, perhaps unwisely, adopted a neutral stance. This attack on the guardian’s impartiality was renewed in a letter from the applicant’s solicitor dated 29^th October in which she wrote: -

“As you will have seen, the Court has indicated that the guardian should attend the expert meeting.

However, we would ask you to pass on to CAFCASS, who will be funding the cost of the guardian’s attendance, the fact that if our client succeeds in her appeal, she will be making application to the Court that a new guardian be appointed for any rehearing.

We would also wish for you to consider your position as the child’s solicitor.”

When the correspondence was submitted to the court Thorpe LJ directed: -

That direction seems not to have been much regarded by the parties for in the following week they agreed elaborate arrangements for the meeting of experts to be moderated by Sir Phillip Otton and transcribed by a stenographer. These developments were referred to the court and on the 2^nd November the President resolved the few issues in contention by directing that the Guardian should attend the meeting of experts and that the issue of costs should be dealt with at the hearing. It was also recorded that the court had no objection to the appointment of Sir Philip Otton, agreed by all parties, to chair the meeting.

In the same week the solicitors for the father sought by letter of the 25^th October directions that a further expert, Dr J, be authorised to read the papers and thereafter submit his report by 20^th November. By letter of the 29^th October the solicitor for the child supported that application. Before ruling the President wished to know whether Professor D supported the involvement of Dr J. Leading counsel then acting for the applicant reported to the court: “Professor D does not consider that Dr J needs to be instructed …our position on behalf of the mother, is that we do not support the application for leave for the father to instruct Dr J prior to the hearing fixed for 8^th December.” However the issue thus joined was resolved by a letter of 1^st November from the father’s solicitors withdrawing the application to instruct Dr J. These exchanges assumed considerable significance when, at the hearing, Mr Scott, who appeared for the father, sought an adjournment to enable an expert opinion to be obtained from Dr J.

We were told that on the 11^th November the BBC suggested to the mother or her advisors that Professor G, a consultant paediatric epidemiologist, might give expert evidence in support of the application. The mother’s solicitors made no application to the court but instructed Professor G and obtained from her a report, dated the 23^rd November that was in circulation prior to the experts meeting on the 25^th November. Her report was particularly contentious in that it attacked the validity of research published in articles in the medical press in 1996 and 1998, both co-authored by Dr D.

By much the same route the mother’s solicitor obtained a report from a Professor H, (suggested by a journalist on 1^st December). This report is less contentious since Professor H is a mathematician specialising in statistics. His report was dated 2^nd December 2004.

The drafting of questions for the experts’ meeting did not proceed happily and what emerged may well have contributed to the moderator’s approach. For he seems to have proceeded on an assumption that, since there was a profound disagreement between experts of obvious eminence, the application on the 8^th December would at least be enlarged to permit further scientific investigations and reports in preparation for a further hearing in the Spring. At the conclusion of the professional exchanges the moderator required each of the four participants to submit a final position statement on a single sheet of A4 by noon on 3^rd December. He also required responses to the reports of Professor G and Professor H to be filed by the experts for the first respondent by 3^rd December. Finally he indicated that a further report from Professor D to record his subsequent investigations and research was to be filed by 4^th February 2005. This last direction demonstrates the unwarranted assumption of a further hearing after the 8^th December. The court had fixed the 8^th December for the determination of the Taylor v. Lawrence application and any appeal that might result. That fixture recognised the child’s need for finality, particularly in the light of the order of Kirkwood J setting U on the road to adoption.

The moderator’s requirement was met by the First Respondent’s experts in that Dr D filed his position statement and responses to Professors G and H on the 2^nd December, Dr R filed his position statement on the 30^th November and Dr H filed his position statement on the 2^nd December. Professor D’s compliance was less orthodox in that on the 5^th December, only one clear day before the hearing, he filed a twelve page report which effectively became his final report in the application.

This detailed chronology of events demonstrates how vain was the court’s endeavour to control the timetable and to ensure that the papers were complete for the court’s preparation in the week preceding the hearing. Mr Jackson Q.C., first instructed on 24^th November on behalf of the mother, filed his principal skeleton on the 3^rd December, Miss Rowe Q.C. on the 1^st December and Miss Hodgson, on behalf of the guardian, on the 1^st December.

The court’s preparations were complicated by an application by the BBC for a raft of orders, necessitating arrangements for a preliminary sitting on the 7^th December. The parties filed their skeletons in response to that application, work which was then rendered superfluous by the BBC’s withdrawal on the afternoon of the 6^th December as a result of objection to their application by the mother. An additional skeleton from Mr Jackson was filed on the 7^th December, entirely excusable given that the documents upon which it was based, contemporaneous transcripts of conversations recorded by the ambulance teams who had taken U to hospital, had only been unearthed by his junior, Miss Wood, on the 5^th December. What is less excusable is that the application for the admission of the evidence of Professors G and H was only issued on the 26^th November and listed for hearing on the 7^th December.

We do not need to detail all the avoidable problems caused by the disregard of the court’s plain intention to control and manage the case, particularly on the part of the applicant’s solicitor and Professor D. It is perfectly plain from the correspondence that the First Respondent continued to await a final report from Professor D after the 8^th October and indeed well into November. Whilst we appreciate that the issues were fully ventilated at the experts’ meeting on the 25^th November, Professor D effectively delayed his final report until 5^th December. He was entitled to the last word by way of reply but not by the submission of his first fully considered report. Some prejudice, albeit less, was no doubt caused to the Local Authority’s experts, and particularly Dr D, by the filing of such a contentious report from Professor G at such a late stage in the timetable.

On the day before the hearing of this application we were asked to accept the additional evidence of Professor G and Professor H. Despite considerable reservations about the late presentation, the failure to seek the permission of the court and manner in which this additional evidence had been provided to the other parties and to the court, we took the view that the court should not refuse to consider the additional evidence on this renewed application. Neither the local authority nor the guardian objected to this course. We therefore took account of the evidence of Professor G and Professor H. We also accepted as additional evidence the transcripts of the conversations of the ambulance teams. As we set out above we refused the father’s application for an adjournment for the purpose of enabling an expert opinion to be obtained from Dr J.

Mr Peter Jackson, who now represents the mother, in his submissions to this Court set out a number of grounds upon which he attacked the basis of the judgment of Bracewell J (the 1^st judgment) and consequently the decision of this Court in the first application for permission to appeal (the 1^st application). The exceptional circumstances which, he submitted, entitled this Court to reopen the case were based partly on matters arising between the 1^st judgment and the 1^st appeal and partly the additional evidence obtained after the 1^st application. Since the 1^st judgment Mr Jackson submitted that there has been an evolving recognition of the need for caution in accepting allegations of child abuse. He pointed to the line of cases in the Court of Appeal (Criminal Division) in R v Clarke, R v Patel and R v Cannings. Given the light thrown on these cases by the Court of Appeal (Criminal Division) he asked us to look again at the 1^st judgment which he divided into the three aspects of medical evidence, stress on the mother and her credibility. He (i) criticised the absence of positive evidence of harm caused by the mother; (ii) challenged the judge’s findings on credibility; (iii) challenged the foundations upon which the previous medical evidence had been based and relied upon the additional medical evidence provided by Professor D to demonstrate its fallibility; (iv) relied upon alternative medical explanations for U’s condition.

We ruled that Mr Jackson was not entitled, on an application to reopen the 1^st application, to challenge matters which could properly have formed part of the submissions in the 1^st application and were not informed by subsequent events or additional evidence. He was not therefore able to pursue points (i) or (ii) above.

In addition to the reports and evidence of Professor D with which we deal below, Mr Jackson relied upon further information obtained in Bangladesh in support of the suggested genetic origin of U’s condition and the transcripts of the discussions between the ambulance teams who came for U and the mother to demonstrate that Bracewell J’s assessment that the mother was lying was flawed.

Mr Scott, for the father, supported the submissions of Mr Jackson and gave us a helpful insight into the meaning of gastro-oesophageal reflux (GOR) to elucidate his submission that this was the correct diagnosis of U’s condition.

The local authority opposed the application. Miss Rowe submitted that the late diagnosis of reflux was based solely upon the recent history given by the parents to Professor D which was inconsistent with the earlier accounts given by the mother to the medical professionals. She relied upon the findings of Bracewell J and Kirkwood J as to the lack of credibility of the mother and father. Apart from one recording of a 999 call the findings remained unchallenged and unchallengeable. The evidence of a genetic basis disclosed a muddled picture and the relevance of a genetic element remained as incomplete and problematical as it had been before the original Court of Appeal.

The guardian took a neutral position in the written submissions on the issue whether to reopen the permission to appeal. Miss Hodgson in helpful written and oral submissions, reminded us of the evidence of Professor P on the genetic issue and that the child remained in limbo during these prolonged proceedings.

Any appraisal of Professor D’s contribution must include all three of his reports. His first states in the introduction:

“On 30^th June 2004 I was approached by (the applicant’s solicitor) to ask if I would be willing to read the Court of Appeal judgment and prepare a short commentary in which I would analyse the judgment in medical (rather than legal) terms insofar as it relates to the case of U. My sole source of information about the care proceedings is the judgment of the Court of Appeal.”

The following are the themes developed by Professor D:

There can be no doubting that Professor D’s stance on the 11^th July was to question the qualification of the experts and the validity of their contributions both clinically and forensically. At this stage the theory of G.O.R. was raised in paragraphs 40 and 66 but assumed no prominence.

Professor D’s second report dated the 30^th August throughout stressed its preliminary character. In paragraph 7 he wrote: -

“I am concerned about the exceptional brevity and incomplete nature of this report, an inevitable consequence of the extreme urgency of the situation, namely the impending adoption of the child.”

Again the final section headed Preliminary Conclusions opens thus: -

“Still required (from me) is a detailed analysis of the four episodes that have been hitherto attributed to intentional suffocation, to look in greater depth at the history obtained at the time, the examination findings, and the results of the investigations that were performed. ”

Anyone giving due weight to those qualifications would surely have expected the completion of the forensic task, upon which the application solely depended, both within the Court’s timescale and to enable his professional colleagues, whose work he was indicting, to know with clarity what criticisms they had to meet.

That said it is relevant to identify the main themes of the second report: -

In his conclusions Professor D explained why “important preliminary points” contra indicated intentional suffocation. He then criticised his colleagues either for having failed to take a history or for having taken an inadequate history. In his final paragraph he offered his preliminary conclusion that there were three possible explanations for the four episodes of which G.O.R. was “in terms of the available medical information, the most probable.”

We have already explained the circumstances in which Professor D’s final report came to be submitted. In paragraph 4 he wrote: -

“Between submitting my report on 30^th August 2004 and the experts meeting on 25^th November 2004, I have undertaken additional work on the case and on the issues that arise. This work partly comprised reading additional papers and reports, but mainly consisted of studying various aspects of the medical literature.

However he continued to stress that even this report did not merit description as final. In paragraph 8 he said:

“…I have not yet had an opportunity to cross check my findings and my analysis with the new Court bundles and with the transcript of the entries in the medical records.”

In the next paragraph he wrote:

“In the light of this additional data it has been put to me that my analysis of the records needs to be revisited and completed, and I accept this.”

Furthermore the penultimate section of his report was headed “Further Work Needed” and over the course of the two following pages he identified the need for an agreed set of records, study of the relationship between G.O.R. and A.L.T.E., together with further examination of genetics.

Viewing these three reports together there emerges a distinct impression of a busy specialist who never quite finds the time to complete the task. Given that the mother’s application had been previously refused, with the consequence that she was reduced to such prospects as the decision in Taylor v. Lawrence afforded her, crucial to her success was a thorough, complete and convincing expert opinion. It might be said that Professor D’s contribution lacks any of those characteristics. However we must not overlook the fact that in writing his third report Professor D had been given the clearest indication from the moderator that the hearing on the 8^th December was likely to be but the preliminary to a final hearing in the Court of Appeal in 2005.

Whilst we appreciate that Professor D was entitled to restrict his contribution to matters purely medical, his elevation of the history which he had obtained from the parents as a proper foundation for his preliminary conclusion that G.O.R was the most probable explanation has to be questioned. Was it realistic given that the mother’s evidence had been rejected as untruthful at the trial, not just on details concerning the child’s admission, but across a wide front? However in fairness to Professor D we must note that in his final report he wrote: -

“I appreciate that aspects of the family’s behaviour have been a cause for concern, and I imagine that further information about this may be found in the papers relating to the disposal hearing, which have not been disclosed to me. However these are not medical matters and the significance of these aspects when it comes to considering the cause of the A.L.T.E can only be a matter for the court.”

What then of Professor D’s opinion that the most probable cause of the A.L.T.E. episodes was G.O.R.? Since that was not an argument raised during the course of the previous application for permission to appeal, it is necessary to return to the medical evidence before the trial judge.

Professor H in his report of the 22^nd May 2002 dealt with G.O.R. at paragraph 7. He stated: -

“As stated above, gastroeophageal reflux, from the stomach into the oesophagus and pharynx, has been suggested as a possible factor leading to airway obstruction in young infants. Gastroesophageal reflux is a commonly observed phenomenon in infants with many cases remaining asymptomatic and most improving spontaneously. It is clear that the presence of acid in the larynx can lead to airway obstruction and to so-called mixed apneas (a combination of obstruction and depression of the respiratory drive). However, most recent evidence refutes gastroesophageal reflux as a cause of apnoeas in infants presenting with apparent life threatening episodes and it appears likely that it is a chance association. I note that L was not investigated for gastroesophageal reflux. In my opinion, this is appropriate given the evidence, but even were this to be contemplated at this stage, its presence or absence would not be helpful in determining the cause of earlier events.”

Subsequently there was a meeting of the experts chaired by the guardian ad litem who dealt with a series of agreed questions. This was question 5: -

“Is G.O.R. a possible explanation of the reported/presenting symptoms in relation to these episodes? Would such symptoms be likely to be episodic and/or potentially self-resolving had they been caused by such means?”

There followed an exchange of views between the three experts on the relevance of G.O.R., its mechanism, and the debate in the scientific literature on the relationship between G.O.R. and apnoea. It is quite clear that none of the three answered the first question in the affirmative. This consensus led to Bracewell J recording amongst the conclusions of the experts: -

“The possible medical conditions considered, being G.O.R., Laryngeal Malacia and Trachea Malacia, are all much less likely than abuse to explain L’s pattern of illness.”

In reaching this opinion the experts naturally concentrated on the contemporaneous records, containing the statements of the parents about U’s health, together with the observations of the healthcare professionals. There is no doubt that the contemporaneous records do not support a diagnosis of G.O.R. The transcript of the meeting on the 25^th November 2004 contains this statement from Professor D: -

“I would not have diagnosed reflux on the basis of the records that Dr D has very carefully gone through – all the occasions when vomiting was or was not mentioned. On the basis of that information alone, I would not have made a diagnosis of reflux.”

A few pages earlier in the transcript Professor D stated that his diagnosis was “based totally on the history of the parents.”

Those concessions led to the obvious challenge which appears from this passage in the transcript which merits full citation: -

“THE CHAIRMAN: Do you want to come to it now and deal with it? Because I think we had been looking at the contemporaneous records, and Dr D seems to have applied himself to that. You may feel that as you have not had the opportunity to look at the records, you do not want to come into the discussion on that. I am in your hands really.

PROFESSOR D: I think I have already responded, in that Dr R put me on the spot and pointed out that his understanding of the contemporaneous …pointed out that there were contemporaneous records. I cannot remember how he worded it, but I said that I would not have diagnosed reflux on the basis of those written records. It does not include the health visitor ones necessarily, but I was really referring to the hospital ones. So I have answered your question?

DR R: Yes, I think you said it was a normal child, on the basis –

PROFESSOR D: I would not have diagnosed reflux on the basis of those records.

THE CHAIRMAN: Do you want to come back?

DR D: T, do you feel that the absence of reference to abnormal vomiting in the contemporaneous records in any way undermines what the parents told you in July?

PROFESSOR D: That is a very good question, but I think it really gets to the issue that, as you rightly pointed out in your excellent report, it is really going to be a matter for the court to decide on the reliability of the parents’ history. And I do not want to stray into that territory, because I think that is a matter for the judge rather than for me, as I understand it.

THE CHAIRMAN: Well, if I may say so, I think that is a perfectly proper stance to take, that you are not here to evaluate the account that was given to you. And that is a job for outside. I think that may be so.”

In her skeleton argument Miss Rowe included a table contrasting the stark difference between the parent’s account to Professor D in August 2004, covering L’s state of health from birth to age four and a half months, with the contemporaneous records. The records clearly show that, apart from a brief attack of gastro-enteritis in mid-August and another in mid-November, there is not only no record of L vomiting but specific entries of an absence of vomiting in mid-September and mid-October 2001. On the final occasion that U was admitted to hospital on the 13^th October 2001, she remained on the paediatric ward for about 21 days and during that period there were only two occasions of vomiting recorded.

Finally it must not be overlooked that a causative link between G.O.R. and A.L.T.E. is still not scientifically established. It is enough to quote a concession made by Professor D at the meeting of the 25^th November recorded at p.79 of the transcript. He said: -

“It does seem to me an important issue that has been raised, quite rightly, by Dr H as to whether there is a causal relationship between reflux and ALTEs. And he and others have questioned that. That is a scientific question that is at the heart of, if you like, my argument. Because if the science is there is no causal link, then it becomes irrelevant.

Now, I have already done some work on this, and it is a very complex area. And I believe more work needs to be done, because I think Dr H has quite rightly put his finger on a very important question, and it seems to me that the court needs to have an attempt of an answer to that question.”

The needs of this child cannot be postponed nor subordinated to future scientific research and debate.

An essential element of the new information provided to us is based upon the accounts of both parents to Professor D. This required their accounts of the state of health of U to be believable and acceptable. It is clear that their accounts to Professor D were not consistent with the contemporaneous medical evidence.

These are transcripts of conversations between the ambulance teams and the mother at the times of the four admissions of U to hospital. The records are said to cast doubt on two adverse findings of Bracewell J in her judgment. In relation to finding 1, on the 5^th August the report of the ambulance crew, timed 20.41, stated

“Severe respiratory distress …breathing problems…unconscious…no breathing.”

The mother in her oral evidence said that she had told the ambulance crew that the baby was not breathing. Without that recently acquired information the judge found, on the basis of the ambulance record, timed 20.48, which referred to ‘collic’ and other symptoms but not to ‘no breathing’, that the mother had given a different picture to the ambulance crew from that given to the hospital. With the new evidence, the adverse conclusion drawn by the judge as to the evidence of the mother on that point was not justified. The second finding is unclear and does not, in our view, support the challenge to the finding of the judge.

We accept, reading the ambulance team’s transcript, that Bracewell J was not justified in making her first adverse finding. But her assessment of the behaviour, demeanour and credibility of the mother gathered from hearing the entire first part of the case were so obvious and so extensive that a single incorrect finding on the basis of one incident cannot by itself undermine the findings of lack of credibility of the mother. Kirkwood J’s assessment of the lack of credibility of the father and continuing lack of credibility of the mother is a vital part of the reconsideration by this court of all the relevant factors.

Given that the parents were rejected as truthful witnesses and then found to have perpetrated a deception on the local authority, it would be particularly dangerous to prefer a belated account given in the circumstances that it was. It is thus apparent that the edifice of Professor D’s opinion is built on sand rather than rock.

The possibility of a genetic origin to U’s condition was considered prior to the hearing before this court. Professor P wrote a report, dated the 21^st February 2004 in which he considered that he would not be able to state that on the balance of probabilities there was a genetic cause and added that

“… the family history is unsubstantiated and needs to be treated with great caution.”

In our first judgment we accepted that advice. Journalists from the BBC became involved in this case and made further inquiries in Bangladesh. A journalist, believing that U had suffered from fits which necessitated admission to hospital, video-recorded a three year old cousin of the mother, called Tariq, suffering a fit in front of him. An explanation of the fit was given by Tariq’s mother to the effect that he suffered fits when he had a fever. These fits were diagnosed as febrile convulsions. His grandmother and other relatives described how other infants in the family suffered from unexplained fits. According to the journalist the family claimed more than 30 infant deaths in the family going back four generations.

This further information remained imprecise and incomplete and was, in our judgment, insufficient to cast doubt on the conclusion of the Court of Appeal at paragraph 83 of the first judgment of the Court.

She is an experienced paediatric epidemiologist. She commented in her evidence on two papers referred to and relied upon by Dr D in his original report dated 27^th February 2002 and at the experts’ first meeting on 6^th September 2002, both in the evidence considered by Bracewell J. Paper A was a report on a study by three paediatricians in association with the British Paediatric Association on the epidemiology of Munchausen Syndrome by Proxy, non-accidental poisoning and non-accidental suffocation. Paper B was a follow up of the children referred to in Paper A. Professor G was critical of the methodology of the study and the lack of detail and of verification. She reminded us that, while it was important to recognise that parents did harm their children, it was also important to recognise that there were many children with disorders for whom no cause could be found. She suggested that there was a temptation for some health professionals to make assumptions adverse to the family. In summary, in her view, the two Papers were no better than anecdotal reports and were not a scientifically valid approach to get at the truth of what might or might not have been happening.

In response Dr D explained that the two Papers had passed through the peer review process of the editorial committee of Archives of Disease in childhood before publication. He agreed that, in common with most scientific papers, they had deficiencies and the comments of Professor G were in some respect valid. But they had to be seen in the wider context of the available and practical methodologies in the study of child abuse. This was the only research they had. In his opinion the Papers had limitations which related in the main to the difficulties of prosecuting research in the area of child abuse. He did not consider that they misled but believed that they made a considerable contribution to the knowledge in this field.

We do not consider that this additional evidence from Professor G is of sufficient importance to justify by itself a review of the case nor does it add sufficient weight to the other evidence to which we have referred above.

Professor H wrote a paper in Paediatricand Perinatal Epidemiology 2004 on “Multiple sudden infant deaths – coincidence or beyond coincidence?” He cast doubt on the use of statistics in the Clarke trial. Although the article does not relate directly to the child U, the subject of Professor H’s paper does bear at least some relevance to the issues in this case. He was instructed by the solicitor for the mother to prepare a report on statistical aspects relating to multiple events and the evidence of Dr D and Dr R. Professor H’s report questioned the statistical evidence given by Dr D and Dr R and cited in the judgment of the court of the 14^th May 2004 and contended that, although inadvertent, the evidence relating to statistics was misleading. The basis of this contention is formed from Professor H’s statistical work on sudden infant deaths transferred onto the occurrences at issue in this case.

Professor H refers to evidence of Dr D at paragraph 69 of the court’s judgement of the 14^th May 2004 that “if you have two cot deaths, the relevant probability between that being natural causes or abuse is roughly 50:50” and contends that it is at odds with the statistics he has comprised from his work of multiple sudden infant deaths. Although perhaps compelling at first glance, Professor H does concede the obvious question when his report notes,

“How do cot death statistics relate to the present case? In the absence of relevant data it is impossible to say whether the relative probabilities for natural: abuse in cases of sudden infant deaths would transfer to similar ratios for episodes of the sort experienced by child U. But there is no reason to suspect that they would be vastly different.”

Professor H similarly questions an exchange cited at paragraph 53 of the judgment of this court of the 14^th May 2002 with Dr R concerning statistics. Dr R states that the number of cases of first presentation to hospital which lead to a second presentation is “much less than 1 in 20”. Although Professor H’s eventual contention is that the line of questioning regarding a third or fourth time is worrying in that it suggests that low re-admission rates are a measure of likelihood of abuse and might be misinterpreted, he does concede that as to the “1 in 20” figure, he has “no reason to take issue with this figure”.

We do not consider that in this case the issue of statistics was central to the decision nor that the evidence of Professor H is of sufficient weight to justify a reopening of the permission to appeal.

In conclusion we do not consider that sufficient doubt has been cast on the evidence of the three medical witnesses at the hearing before Bracewell J to justify a reopening of the medical issues. The serious lack of credibility of both parents casts such doubt on the account given by the parents to Professor D at a late stage, inconsistent as it is with contemporaneous medical records, that it does not begin to surmount the high hurdle required by CPR 52.17. For the reasons set out above, we were satisfied that the issues raised on the further application came nowhere close to the principles laid down in Taylor v Lawrence and that no significant injustice had occurred. There were therefore no grounds upon which permission to reopen the appeal should be granted.

We have set out above the steps taken by the solicitor for the mother to exclude the guardian from playing any part in the further experts’ meeting, even to the extent of indicating that she should not even attend the meeting. We should like to express our concern that the local authority should have remained relatively neutral on this point and that the guardian does not appear to have made a strong objection to this attack on her impartiality. We should like to make it clear that in the view of the court there were no grounds revealed in the papers which justified the removal of the guardian from her traditional role nor from playing any further part in the case had it gone further. Equally there would appear to be no grounds advanced for the suggestion in the letter of the 29^th October from Mrs Harman (see above) that the solicitor for the child should consider her position.

We are also concerned that the approach adopted by Mrs Harman might reflect that taken by other solicitors acting for parents in these difficult and fraught cases. We feel therefore that some comments should be made about the functions and role of the guardian of a child in a care application in order to remind the legal advisers as to the unsuitability of the course adopted in this case and with the hope that it may in future avoid a repetition of misunderstandings as to the guardian’s role.

In the Children Act 1989 the role of the guardian of the child is crucial. A guardian is appointed to represent the child unless the court is satisfied that it is not necessary to do so in order to safeguard the child’s interests see section 41(1). The Family Proceedings Rules 1991(FPR) and the Family Proceedings Courts (Matrimonial Proceedings etc) Rules 1991 set out in detail the statutory requirements of a guardian, usually appointed from CAFCASS. A guardian is appointed unless the court is satisfied that it is not necessary to do so to safeguard the child’s interests. Rules 11 and 11A (FPR), for example, provide for the powers and duties of a guardian which continue for such time as is specified in the appointment or until terminated by the court, see rule 10(9). The independence of the guardian is underlined by the exclusion of anyone who comes within the provisions of rule 10(7), such as a person employed by the applicant local authority. This independence on behalf of the child is an essential element in the hearing of a care application. The court is entitled to rely upon and does rely upon the investigations of and pays close attention to the recommendations of the guardian who looks at the case from an impartial and objective standpoint. The guardian has a duty to form a view of the application and, where appropriate, to recommend a course honestly believed by the guardian to be in the best interests of the child he/she represents, whether it be to advise the return of the child to the family or to advise the removal of the child permanently from the natural family.

A party is not entitled in a private law dispute to have the court welfare officer removed from the case (now the CAFCASS reporter) on the ground that the report is favourable to the other party. In the same way a party in a public law case is not entitled to attempt to dispense with the guardian or to sideline or marginalise the guardian on the ground that in an earlier part of the case, the guardian’s recommendation had been adverse to that party (see generally independence of court welfare officer and guardian in Re S (a Minor) (Guardian ad Litem: Welfare Officer) [1993] 1 FLR 110; [1992] 2 FCR 554). Other than in exceptional circumstances of specific bias or impropriety, the independence of the guardian may not be impugned as a result of the decision of the guardian to support the proposed care plan of the local authority.

In the present case, the guardian at an earlier stage had chaired the experts’ meeting which was a normal procedure in a difficult case. There was no good reason to exclude her or her solicitor from the neutral role of chairing the later experts’ meeting in which Professor D participated. To do so was to misunderstand and to diminish the role of the guardian. We are surprised that the guardian and the local authority did not demur more strongly at an expensive and unnecessary alternative to the use of the guardian or her solicitor as chairman. The cost of inviting Sir Philip Otton to chair the meeting and to make himself familiar with all the papers added extra cost to the public purse, one which neither the local authority nor the Legal Services Commission should have been asked to bear.