B (A Child)

This appeal concerns two matters: the extent of the obligation to cross-examine witnesses about accusations made against them, and the correct legal approach to determining responsibility for inflicted injuries where there are only two potential perpetrators. It raises no new point of law.

It is a mother’s appeal from a fact-finding decision given in the Family Court at Swindon by Recorder Miller on 5 July 2018. The case, which is ongoing, concerns married parents and their only child, a boy born in October 2017, to whom I shall refer as ‘Nathan’.

Nathan is a much-loved child, and no one would have expected him to have come to harm. However, in mid-November 2017 when he was only a few weeks old, he was admitted to hospital with an injured leg. Investigations revealed that he had fractures to both lower legs and six rib fractures. The local authority took care proceedings and Nathan was in due course allowed to go home to his parents under continuous grandparental supervision. However, in April 2018 the parents separated. At that point, Nathan remained in the supervised care of his mother; since the recorder’s decision he has been cared for by his father.

The medical evidence about the injuries was uncontested. The only people who could have caused them were the parents. The only task for the court at the threshold stage was to identify if possible which was responsible.

The matter came before the recorder in June, when he heard evidence and submissions over three days. In particular, he heard extensive evidence from the parents. He reserved judgment.

The position of the parties was this: the local authority submitted that the injuries were caused by one or other parent and that neither could be excluded; the mother stated that she had not caused the injuries and (although she saw no indication that he had done so) she accused the father of being responsible; the father stated that he had not caused the injuries and (because he had seen no indication that she had done so) he did not directly accuse the mother of being responsible, but accepted that this was the inevitable corollary of his denial; the Guardian was neutral.

The framework for the recorder’s decision was therefore a relatively straightforward one. There were a number of elements on which he could depend:

However unexpectedly, Nathan had sustained serious inflicted injuries.

The only candidates were the parents.

The nature of the injuries (involving forceful twisting, jerking and squeezing) did not allow for an accidental explanation, and there was anyhow no account of any accident.

There were three separate applications of force, occurring on one, two or three occasions within about 10 days of admission to hospital.

The injuries were not likely to be deliberate but to be the result of loss of control.

It was most unlikely that they were caused by more than one parent.

There were no obvious risk factors in respect of either parent and many positives in respect of each of them.

They worked as a team and were attuned to Nathan’s needs.

Both had the opportunity to cause the injuries, the mother more so than the father as she was the main caregiver and he had returned to work.

The house was a small one in which, if both parents were at home, it would be difficult for one parent to cause injuries on this scale without the other becoming aware.

One parent was not telling the truth.

There was no evidence of collusion.

The judgment, which is structured and lucid, summarises the background, the police and medical evidence, the oral evidence and the parties’ cases. The recorder succinctly summarised the applicable law, noting at [58] that the standard of proof is the balance of probabilities and at [60] that he should attempt to identify the perpetrator of an inflicted injury but should not strain to do so.

The recorder proceeded to his conclusions in this way:

Is it possible to identify who caused the fractures? There are only two possible perpetrators; mother and father. This is accepted by each of them and accords with the evidence I have heard and read. [64]

Criticism has been made by counsel on behalf of mother of the local authority’s threshold document for failing to particularise its case.… Whereas there are significant elements of case preparation by the local authority of this case which have not been of a good standard I am quite satisfied that the parents have been given sufficient particulars of the local authority’s case on the injuries in order that they can respond appropriately. There have been no surprises to them about the issues in this case. They have known about the injuries [Nathan] sustained and have had full access to all documents prepared to further the inquiry into the nature and cause of those injuries. I am satisfied there has been no infringement of their Article 6 rights in this regard. It is of course open to a local authority when it has considered the evidence at a fact finding hearing to modify the findings it seeks for example to remove an individual from the pool of possible perpetrators, but in this case the local authority has not considered it possible to do this. [65]

Both parents love Nathan and had shown appropriate emotion during the hearing; however, the fact remains that he sustained serious injuries in their care. [69-70]

The family home is a small house where noise from downstairs could be heard upstairs. It is difficult to conceive that noise out of the ordinary could go unnoticed by any adult present; even if asleep, they would be likely to awake. [72]

In my judgement, it is more likely that [Nathan] was injured when he was alone with one of his parents. I find this because I think it probable that if both were present, they would support each other, thereby sharing the caring tasks and relieving any pressures those may create. It is clear that the parents had organised themselves and apparently worked as a team, sharing tasks. I think it improbable that one would stand by and let the other struggle in their presence. It is more likely that if alone tensions could build and potentially be overwhelming without the support of the other. [74]

It is clear that mother had greater opportunity to inflict the fractures upon [Nathan]. However, mere opportunity does not necessarily equate with the probability she did so. I have to look at and assess all the evidence. [75]

I have concluded the injury to the leg did not occur whilst both parents were at home together later on 16 November and into 17 November. I do so for a number of reasons. In my judgement, had it occurred at this time [Nathan] would have screamed in pain to which both his parents would have responded and both would have been aware. Even if mother was asleep upstairs it is probable she would have woken and investigated immediately. It is also likely that had father caused it at this time, he would have been showing signs of agitation and frustration of which mother would have been aware.… [80]

In my judgement, the mother’s greater opportunity to cause these injuries is indeed significant. It is likely that in the week of 13 November, she was feeling emotional and struggling to care for [Nathan]. Father noticed this and took steps to help and support. This aspect of his evidence was not challenged. Father came across as straightforward and practical. He is organised and thoughtful. He was sensitive to his wife’s needs. She saw no signs in him of stress or agitation or frustration with [Nathan]. As I have indicated when they were together, they worked as a team. Father did not have the opportunity to cause these injuries unnoticed. Mother was upset that she couldn’t breastfeed [Nathan] as she would have wanted. I do not criticise her for feeling that way, it is natural. There were occasions when [Nathan] would not settle and she was alone with him no doubt in pain from her caesarean. Her mobility was restricted. There is evidence that on 14 November, for example, she’d had a bad day with [Nathan]. She accepted in evidence that it had not been an easy time for her but didn’t suggest that this was a problem. In my judgement, it is highly unlikely that both parents caused the injuries to [Nathan] and therefore I find that one of them did. In my judgement, I am able, without straining to do so, to identify mother as the perpetrator of each of [Nathan]’s injuries. It is improbable she did so deliberately and I am satisfied she did not. It is likely that at a time when she was alone with [Nathan] and had been for some time, stress levels built and she was overwhelmed by the situation and inflicted force on him causing these fractures. It is probable that [Nathan] was not settling and crying and this caused her to lose control. I find that she knows she used force and hurt him but was not immediately aware that fractures resulted. I am satisfied that she is overwhelmed by grief and guilt about this. It was not a deliberate, malicious act, but the actions of someone who lost control at a time when the stress of caring for a young baby overwhelmed her. By the time father came home she had probably been able to restore calm and regain her composure and therefore it would not have been necessarily apparent to father what had happened during his absence. I have already said I am satisfied she loves [Nathan] deeply, and my findings do not alter this. [83]

I am not able to say if all the injuries were caused at the same time, albeit with three separate applications of force however I can find that the father was not present when any of these injuries were caused and was probably at work. [84]

In my judgment, it is more likely than not that mother caused these injuries and that as the pool consists of two possible perpetrators, I find that she did so and that father did not. Therefore, he is removed from the pool with a finding that he did not cause these injuries to his son. [85]

On behalf of the mother, Mr Geekie QC and Ms Pope bring two grounds of appeal. Firstly, they argue that the finding against the mother was unfair because no other party was putting an explicit case that she should be identified as the sole perpetrator, and also because she had not been questioned about certain matters that the recorder relied upon in reaching his conclusion; as a result, she was deprived of the opportunity to answer the case against her. Secondly, they argue that the recorder applied the wrong legal test in that instead of considering each parent individually and asking himself whether the evidence established a probability that they were responsible, he instead impermissibly compared one parent against the other and convicted the mother because he in effect found her to be the likelier perpetrator.

During the course of argument, it was also submitted by Mr Geekie that certain aspects of the judge’s reasoning were unjustified by the evidence or even perverse. Those submissions went beyond the grounds of appeal as pleaded. Had there been any substance in them, I would in a matter of this importance, and where no procedural prejudice to other parties arose, have been inclined to allow an amendment of the grounds of appeal. However, Mr Geekie did not particularly press this and on examination the complaints came nowhere close to amounting to a potentially successful ‘reasons’ challenge. I therefore say no more about them.

In relation to the first ground of appeal, Mr Geekie criticised the way in which the local authority had expressed its case up to and during the trial, saying that it was not clear exactly how it was putting matters. Then, when the mother come to give evidence, her cross-examination did not include questions that directly accused her of responsibility for the injuries. There were also certain matters relating to her physical and mental state in the week before the injuries were discovered, about the acoustics in the home and about the likelihood of her being aware if the father had caused the injuries downstairs while she was asleep upstairs: Mr Geekie contends that the mother was not directly asked about these matters and that if she had been, she could have spoken about them. As it was, the mother was taken by surprise by the way the recorder reasoned his conclusion. Seeking to make good these points by a close analysis of the judgment, and in particular of paragraph 83 (set out above), Mr Geekie identified matters that he argued had either not been put to the mother or were speculative.

In relation to the second ground of appeal, Mr Geekie argues that, whatever the recorder may have said, the route by which he arrived at his conclusion was in effect by a comparison between one parent and the other, rather than by a separate appraisal of the evidence in respect of each of them and a proper application of the probability standard.

Both contentions are challenged by Mr Vine QC and Ms Gallagher on behalf of the local authority and by Ms Cook QC and Mr Ker-Reid on behalf of the father. On behalf of Nathan, Mr Jenkins and Ms Scammell filed helpful written submissions that, while maintaining the Guardian’s neutrality, offer no support for the mother’s arguments.

It is an elementary feature of a fair hearing that an adverse finding can only be made where the person in question knows of the allegation and the substance of the supporting evidence and has had a reasonable opportunity to respond. With effective case-management, the definition of the issues will make clear what findings are being sought and the opportunity to respond will arise in the course of the evidence, both written and oral.

The obligation to put relevant matters in cross-examination has been stated in a number of authorities from Browne v Dunn [1894] 6 R 67 (HL) to Re W (A child) [2016] EWCA Civ 1140, at [88-89]. Still more recently, the Privy Council considered the matter in Chen v Ng [2017] UKPC 27, where the Board’s judgment, given by Lords Neuberger and Mance, includes these passages:

In Chen, the process had been unfair because the two grounds given by the judge for disbelieving a party had not been put to him in evidence. Re W was a very different situation where completely unheralded findings were made against professional witnesses. Those findings were described by Mr Geekie, who also appeared in that case, as being outside the four corners of the case.

In assessing fairness, what is important is substance not form. The question of whether an adverse case has been sufficiently put to a witness is likely to be informed by the five factors set out in Chen at [55]. It is case-specific and rooted in the real world of litigation in which overall fairness can be achieved in a range of ways. In general a party is required to challenge in cross-examination the evidence of any witness of the opposing party if he wishes to submit to the court that the evidence should not be accepted on that point; if a party has decided not to cross-examine on a particular important point, he will be in difficulty in submitting the evidence should be rejected: see Phipson on Evidence 19^th ed. 12.12. However, the rule is not an absolute one, and there will be cases in which it will be pointless to put formal challenges to a witness who knows perfectly well that his or her evidence is disputed, and where the challenge could in reality go no further than “I put it to you that you are lying”. As my lord, Lord Justice Newey, put it in Howlett v Davies [2017] EWCA Civ 1696 at [39]:

The proper approach to cases where injury has undoubtedly been inflicted and where there are several possible perpetrators is clear and applies as much to those cases where there are only two possible candidates as to those where there are more. The court first considers whether there is sufficient evidence to identify a perpetrator on the balance of probabilities; if there is not, it goes on to consider in relation to each candidate whether there is a real possibility that they might have caused the injury and excludes those of which this cannot be said: North Yorkshire County Council v SA [2003] EWCA Civ 839, per Dame Elizabeth Butler-Sloss P at [26].

Even where there are only two possible perpetrators, there will be cases where a judge remains genuinely uncertain at the end of a fact-finding hearing and cannot identify the person responsible on the balance of probabilities. The court should not strain to identify a perpetrator in such circumstances: Re D (Care Proceedings: Preliminary Hearing) [2009] EWCA Civ 472 at [12].

In what Mr Geekie described as a simple binary case like the present one, the identification of one person as the perpetrator on the balance of probabilities carries the logical corollary that the second person must be excluded. However, the correct legal approach is to survey the evidence as a whole as it relates to each individual in order to arrive at a conclusion about whether the allegation has been made out in relation to one or other on a balance of probability. Evidentially, this will involve considering the individuals separately and together, and no doubt comparing the probabilities in respect of each of them. However, in the end the court must still ask itself the right question, which is not “who is the more likely?” but “does the evidence establish that this individual probably caused this injury?” In a case where there are more than two possible perpetrators, there are clear dangers in identifying an individual simply because they are the likeliest candidate, as this could lead to an identification on evidence that fell short of a probability. Although the danger does not arise in this form where there are only two possible perpetrators, the correct question is the same, if only to avoid the risk of an incorrect identification being made by a linear process of exclusion.

There is in my view nothing in either ground of appeal. This court will pay considerable respect to the findings of fact of trial judges and will only interfere if they are shown to be wrong or to have been reached by an invalid process. Here, the recorder reached a careful and considered decision. He did not misapply the law and his evaluation of the evidence led him to a reasoned conclusion that was well within the range of possibilities properly open to him.

This was a case in which there was no direct evidence of the assault(s) on Nathan: he cannot tell us and the only person who can is the person responsible. In such a case, the scope for conventional cross-examination is inevitably curtailed by the limits on the other parties’ knowledge. The court must arrive at a conclusion on the balance of probabilities by integrating the known facts with the inferences that can be drawn from the surrounding circumstances and with its assessment of the witnesses.

There was nothing unfair about the process that led to the finding of fact against the mother. These illustrations show that the finding was squarely ‘within the four corners of the case’:

On 24 November 2017, the mother took part in a voluntary police interview “about your suspected involvement in a criminal offence” [S78ii].

The case management order on 3 May 2018 recorded that the local authority was seeking a finding that the mother and/or the father were in the pool of perpetrators, and that a fact-finding hearing was to take place at which “in the event that one or other of the parents is excluded from the pool of perpetrators, the court may go on to consider whether any final order can be made” [C69-70].

On 11 June 2018, the local authority filed a parenting assessment [S275] that made recommendations on three alternative bases: that the mother was exonerated, that the father was exonerated and that neither was exonerated.

As for the parents, prior to the trial neither had made an explicit accusation against the other, but as these limited extracts from their evidence show, both told the court with varying degrees of enthusiasm that the other parent must be responsible:

In the parties’ final written submissions, both parents argued that they should be exonerated and that the other should be identified as having caused the injuries.

All this material overwhelmingly contradicts the mother’s case that no party was seeking a finding that she was responsible for the injuries. There is no possible way in which she could have considered herself immune from jeopardy while at the same time having a ‘free shot’ against the father. Her evidence ran to 80 pages of transcript, of which all but 13 pages consisted of cross-examination and questioning by the recorder. On the supposition that the father, like the local authority, had no knowledge of when and how the injuries were caused, the scope for putting blunt challenges in cross-examination was extremely limited. Instead, his experienced counsel, Mr Ker-Reid conducted a detailed cross-examination focusing on the surrounding circumstances in a way that was not attritional or hostile. However, the very exercise itself can only have been understood as being a means of furthering the father’s case that he was not responsible for the injuries and that by necessary implication the mother was. For its part, the local authority, faced with a bare denial had, as Mr Vine says, no other angle to pursue with the mother. The recorder also asked a considerable number of questions at the end of each parent’s evidence, plainly directed towards elucidating which one of them might have caused the injuries. During this hearing, Mr Geekie was unable to give examples of specific questions that should have been asked of the mother in cross-examination but were not.

The further complaint that the recorder based his conclusions on matters that the mother had not been asked about is again not borne out by a reading of the record. As Mr Vine demonstrates, matters relating to the mother’s health were dealt with in her own evidence and there was considerable debate about the acoustics in the house, as seen in the numerous references to the evidence about that subject in her skeleton argument on this appeal at [35-37]. As to the recorder’s reliance on the father’s demeanour within the home, the mother was not cross-examined about it because it was her case that he had never been other than calm.

In relation to the first ground of appeal, I therefore conclude that the mother was fully aware of the case against her and had a full opportunity to respond. In saying at [65] that there was no infringement of the mother’s right to a fair trial, the recorder was right. Her complaint is in reality a complaint that her evidence was not accepted. That could not be a basis for interfering with the assessment of the recorder, who had an unrivalled opportunity to assess the probabilities and who reached a conclusion that was clearly open to him on this evidence.

As to the second ground of appeal, the recorder was fully addressed on the law and well understood the task he had in hand. He reached his conclusion having thoroughly reviewed the evidence in respect of both parents. He did not exonerate the father by a process of exclusion or by reasoning that he was less likely to have caused the injuries than the mother, but because, as he said at [85], “it is more likely than not that mother caused these injuries”. Mr Geekie argues that in reality he reached this conclusion because he thought the mother the more likely perpetrator in circumstances where the evidence did not justify a finding on the balance of probability. My reading of the judgment does not support this submission, but rather shows that the recorder approached the matter in accordance with correct principle.

For these reasons, I have concluded that the recorder’s decision comfortably survives the examination it has received, and I would dismiss this appeal.

I agree.

I also agree.