England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM SWINDON COUNTY COURT
HER HONOUR JUDGE KATHARINE MARSHALL

		Royal Courts of Justice Strand, London, WC2A 2LL
		20 December 2012

Lord Justice Munby :

1. This is an appeal, pursuant to permission granted by the judge, from findings of fact made by Her Honour Judge Katharine Marshall in the Swindon County Court and set out in a reserved judgment handed down on 31 August 2012. The judge had been conducting a fact finding hearing as part of care proceedings brought by a local authority in relation to two children: L, a boy born on [a date in] 2008, and M, a girl born on [a date in] 2011. The fact finding hearing had occupied seven days between 6 and 19 June 2012.

The conundrum

2. It was not and is not disputed that when M was admitted to hospital on 3 August 2011, when she was just five months old, she had three (possibly four) fractured ribs, those fractures having occurred on at least two occasions, and fractures to her skull. What lifts this case out of the ordinary – indeed makes it unprecedented and thus unique in the experience of the very distinguished experts who gave evidence – is that although M had indubitably suffered complex bilateral skull fractures, described by one of the experts as "spectacular", she had not suffered the very serious brain injuries which would be expected to accompany fractures of such severity.
3. At the very beginning of her judgment the judge described the case as "most unusual and difficult." She explained why (paragraph 2):

"I should record at the outset that the skull fractures are spectacular, so complex and extensive that they have been described by the experts as beyond anything they have seen before in a child of M's age in their considerable collective experience. Her clinical presentation, in that she appeared to be reasonably well and suffering no apparent pain or neurological effects, was not just unusual in the circumstances, but inexplicable. This conundrum has perplexed the experts involved and goes to the heart of this case."

The expert evidence

4. The judge read and heard evidence from four very experienced experts: Dr Stephen Chapman, Consultant Paediatric Radiologist at Birmingham Children's Hospital (report dated 5 November 2011, letter dated 4 December 2011 and addendum report dated 28 January 2012), Mr Peter Richards, Consultant Paediatric Neurosurgeon at the John Radcliffe Hospital, Oxford (report dated 31 March 2012), Dr Katharine Halliday, Consultant Paediatric Radiologist at University Hospital Nottingham (report dated 12 April 2012) and Professor Nicholas Bishop, Professor of Paediatric Bone Disease at the University of Sheffield and Honorary Consultant Paediatrician at the Sheffield Children's NHS Trust (report dated 8 May 2012 and addendum report dated 11 May 2012). All four experts participated in an experts meeting, conducted by conference telephone and chaired by the guardian's solicitor, on 15 May 2012. The fact finding hearing began as I have said on 6 June 2012.
5. To put the expert evidence in context: The mother gave an account of an incident on 16 July 2011 when she squeezed M's chest to avoid dropping her as she fell out of the bath. The father gave an account of having dropped M on 2 August 2011 when she hit her head on a low table before she hit the floor.
6. M was x-rayed on 3 August 2011 and again on 5 August 2011. Dr Halliday and Dr Chapman were agreed that the older rib fractures, to the left 5^th and 6^th ribs, were likely to be some 2-4 weeks old when x-rayed and so to have occurred according to Dr Halliday (working back from the x-ray on 3 August 2011) between 6 and 20 July 2011 or, according to Dr Chapman (working back from the x-ray on 5 August 2011) between 8 and 22 July 2011. The more recent fracture, to the left 8^th rib, could be dated more precisely because of changes in its appearance between the two x-rays. Dr Halliday and Dr Chapman were firm in their view that this fracture could not have occurred less than four days before the second x-ray, in which case the father's account of what he said happened on 2 August 2011 could not be an explanation for any of the rib fractures.
7. In contrast to the position in relation to the rib fractures, the experts were all agreed that skull fractures cannot be aged from the appearance of the fracture lines.
8. Against this background, it is convenient to start with Mr Richards' report. For present purposes the key passages in his report are paragraphs 2.2, 2.4, 2.8 and 2.9. I set them out in full:

"2.2 The skull fracturing is extensive and bilateral. The exact mechanism cannot be determined. Such extensive fracturing is normally seen in high velocity impacts or in crash injuries. Both are normally associated with significant intracranial injury although occasionally in clinical practice patients present following high force injuries where significant brain injury could be expected and are found to have extensive fracturing bilaterally, (a term commonly referred to as eggshell skull) with no brain injury apparent. It is presumed in these circumstances that the skull has absorbed the impact and caused the extensive fracturing in circumstances where otherwise significant force would be transmitted to the brain.

2.4 An episode of potential injury was described the day before presentation to hospital. However, given the extensive skull fracturing it would be expected that had it been recent there would have been pain caused by the fracturing. This pain would have been apparent in the child being unsettled and crying and particularly if the head was handled. The amount of time a child remains in this state following a skull fracture is not known for certain and clinically is know to be variable. However, clinical experience from managing skull fractures, both caused by trauma and post-operatively, would suggest that pain persists, particularly if the head is handled for 48-72 hours after a skull fracture. I would therefore consider on this basis that the event described on 2^nd August 2012 was unlikely to be the cause of the skull fractures.

2.8 In the information available to me I can identify no event described of significant head injury which could explain the skull fracturing which is unexplained. M did not have the neuromuscular development to injure herself without a carer being aware. In the absence of a history of a significant head injury there can only be three possible explanations.

(i) M has an as yet undiagnosed bone fragility syndrome that has not been identified. Given the degree of skull fracturing she has, yet without clinical evidence of brain function or constitutional disturbance, and the lack of imaging features of brain injury this is a possibility and may explain extensive fracturing from an event which would normally not cause injury.

(ii) An event did occur which a carer was aware of but they did not have the mental capacity for some reason to appreciate it was capable of causing significant head injury.

(iii) An event did occur which a carer has not disclosed to medical or investigating authorities

What the explanation was cannot be determined on the basis of the medical features alone but should be for the Court to determine.

2.9 In summary, therefore, M presented with a skull fracture as well as rib fractures. There were no clinical features to suggest she had suffered a skull fracture and the absence of constitutional disturbance and pain would suggest that the fracture had occurred at least 72 hours beforehand. The persistence of scalp swelling with some blood in it would suggest that the fracture had occurred within ten days of the scan of 5^th August. I can identify no features of abnormal bone fragility but would suggest the opinion of an expert in metabolic bone disease be sought to exclude this possibility. If abnormal bone fragility is excluded, the fractures remain unexplained, but would have required an event with significant impact or crushing involved to cause such fractures."

9. It was following this report that Professor Bishop was asked to report. His opinion as set out in his first report was clear:

"Her previous x-rays have been put onto our x-ray viewing system and the extensive skull fracturing and rib fractures noted previously by Dr Chapman and Dr Halliday are clearly evident. The new x-rays of the spine show no evidence of vertebral deformity; the contour and size of the vertebrae look entirely normal in both the lateral and AP views and there is no abnormality of texture to any of the bones visualised.

My clinical opinion is that M has a normal skeleton. I do not think that either the skull fracturing or the rib fractures represent a manifestation of an underlying bone fragility problem. I think it is likely that the fractures were caused by non-accidental injury."

In his addendum report he said:

"My opinion remains unaltered; I think M has a normal skeleton and that she is likely to have suffered non-accidental injury."

10. I appreciate that "what is said at an experts' meeting is not strictly evidence in the case" (per McFarlane J in Oxfordshire County Council v DP, RS & BS [2005] EWHC 2156 (Fam), [2008] 2 FLR 1708, para 109), though it can of course be put in cross-examination or re-examination. But it is convenient, setting the scene for the evidence that was given by the experts at the trial, to see how this aspect of the matter developed at the experts' meeting on 15 May 2012.
11. The issue was first discussed before Dr Halliday joined the meeting. Mr Richards said:

"I can only speak about the head injuries, but there was a lump on the head, very extensive – I would call spectacular – skull fracturing, but no evidence of any injuries to the brain substance or bleeding inside the head."

Asked whether they all agreed that there were "extensive skull fractures, with minimal brain injury seen on the films?", both Mr Richards said Dr Chapman said "Yes". Professor Bishop said:

"I'd agree the extensive skull fractures. I defer to Dr Chapman and Mr Richards about the degree of brain injury. It's not my area of expertise."

A little later Mr Richards said this:

"Well, there must have been an impact against a hard surface or crushing of the head. What is unusual – the question of force is difficult, and what is unusual is to have such a degree of skull fracturing without any brain injury. But the fact there are fractures – they don't occur spontaneously; there must have been some form of impact or crushing of the head … If the head is crushed so much that you get such significant skull fracturing, you'd get some damage to the underlying brain as well. Provided that we're dealing with normal skull bone, and this is more Professor Bishop's field."

The following interchange then followed:

"NB I think from my perspective, there is no doubt in my mind that this child's skeleton is entirely normal. I've never seen fracturing like this even in a child with abnormal bones. And we see children who've got very abnormal bones indeed – we simply don't see this degree of skull fracturing. With normal ---

PR So, Professor Bishop, would you agree that this in my mind is spectacular fracturing?

NB I would say it's the most extensive skull fracture I have ever seen, irrespective of whether there's an underlying brain injury or not. So ---

PR Which is one of the difficulties for me, because yes, I've seen extensive fracturing like this before. The last time I remember seeing it was when someone walked into a rotating helicopter tail blade, and got that extensive fracturing, but did not survive very long … And that is unusual, the fact that there is very extensive fracturing, and yet no significant injury of radiology and no apparent clinical injury at all, no clinical disturbance whatsoever.

NB As I say, it's not accompanied by changes in other parts of the skeleton that would lead my to believe that there's any underlying bone disease. And having reviewed the child recently, only literally last week, the skeleton looks entirely normal to me, both clinically and radiologically."

Dr Chapman agreed:

"I think the mechanism's either going to be crushing or multiple impacts, and I don't really know what is the actual mechanism … it could be either [one massive impact to the top of the head, but possibly two impacts each side of the head] … We've all said this is an absolutely spectacular fracture, and to say that it could have occurred from one impact alone is again I think speculation, because I don't think any of us have that experience."

Professor Bishop said "I'd agree with that." Asked, "So it would be very difficult to speculate as to what it was that caused the injury?", Mr Richards said:

"Yes. And of course nobody has done research on living infants to fracture their skulls, to see what happens. The only research really that's been done on fractured skulls and what causes them is that rather unpleasant research of dropping dead babies before you do a post mortem on them, and seeing whether you fracture their skulls."

12. The discussion returned to this topic after Dr Halliday came on line. Mr Richards said "The degree of fracturing we all consider is spectacular, outside of our usual expertise", to which Dr Halliday said "Yes." Mr Richards continued, "it's odd that there's such extensive fracturing, not much scalp swelling, and a seemingly happy baby" to which Dr Halliday responded:

"And no brain injury, yes. Yes, it is odd."

13. The significance of this discussion for present purposes is perhaps two-fold. First, all the experts had addressed this issue during the experts' meeting, so it was not coming to them 'out of the blue' at the trial. Second, the experts had been left baffled at the end of the meeting, so it is reasonable to imagine that they were thinking about it in the three weeks that elapsed between the experts' meeting and the trial.
14. During the trial Mr Richards gave evidence on 7 June 2012, Dr Halliday and Dr Chapman (in that order) on 12 June 2012 and Professor Bishop on 13 June 2012.
15. Examined in chief by Mr John Ker-Reid on behalf of the guardian, and referring to what was found when M was x-rayed, Mr Richards said:

"I think they got quite a surprise because there was very extensive skull fracturing yet minimal scalp swelling because again you get scalp swelling with fractures in a normal skull. No underlying injury of the brain, so clinically it was a surprise. So the fact that there was no pain on presentation on 3 August would suggest either this is something very unusual and some very extensive fracturing has been caused in an unusual way that was not painful or the pain associated with it has come and gone or M's response to pain is very unusual. It was odd that what I attributed to be spectacular skull fracturing there was no pain or disturbance shown by M or reported by M from carers or people(?) in the time preceding presentation to hospital … I am particularly surprised by, given the degree of fracturing, the lack of pain, the lack of disturbance, the lack of swelling and the lack of underlying, damage to the brain. So there is something strange going on in this case."

A little later he added:

"most people that I have seen of all age groups, with that degree of fracturing, have either been unconscious in intensive care units or dead."

And then:

"to have that extent of skull fracturing from a normal skull, that requires a lot of force to cause that extent of fracturing and yet not a hint of any bleeding, bruising or damage to the underlying brain is very odd and unusual. As I say, most cases that you see with this degree of skull fracturing are either in an intensive care unit or, you know, not surviving."

16. Mr Richards was cross-examined by Mr Anthony Kirk QC on behalf of the mother:

"Mr Kirk … There is something rather unusual and something rather puzzling about this child's presentation and the fractures that she has sustained.

Mr Richards Yes. Taking all the aspects – if we look at the three aspects of the fractures, so the fractures themselves, her response to the fractures and the scalp swelling associated with these fractures, either her bones fracture from very minor trauma and that is why there is severe fracturing but very little scalp swelling, or her response to it in terms of pain and suffering is highly abnormal. The whole picture is, with that degree of scalp swelling I would have expected significant brain injury and if there was not significant brain injury and she was conscious I would have expected to be in great pain and disturbance and, given the degree of fracturing, with normal force required to cause that fracturing, I would have expected a lot of scalp swelling.

Mr Kirk Yes. Could these complex fractures have been caused, in M's case, by a relatively minor trauma?

Mr Richards I would not have expected it to have occurred from a minor trauma if her skull was completely normal but that is a question in my mind. Is her skull in some way abnormal so that she gets these fractures with trauma that would not normally cause these fractures?

Mr Kirk Was that addressed in your experts' meeting?

Mr Richards It was discussed. Professor Bishop who has looked at the metabolic aspects of bone(?) disease has not been able to identify anything. We have all commented that we have not seen fracturing like this except in very extreme circumstances. I did raise one circumstance where I have seen similar severe fracturing, forgetting the job that her father was doing, but it was an adult injured by a helicopter walking into a rotor blade, a rear rotor blade who had very extensive fracturing and died several hours afterwards which is the kind of injury that I have encountered if you see this kind of fracturing. So if this kind of fracturing with minimal response were to occur there must be something odd about the bones and Professor Bishop cannot find anything but he did say he has not seen fracturing like this either so why he cannot find any metabolic abnormality to explain it, I still wonder in my mind if there is some explanation we are not capable of finding as doctors.

Mr Kirk Thank you. And we cannot rule that out either, Mr Richards, can we?

Mr Richards Oh, absolutely. This is (inaudible) unprecedented in the experience of someone like myself who has been in neurosurgery for nearly thirty years, Steven Chapman who has been a radiologist for many, many years, Professor Bishop who has specialised in bone abnormalities and between us we have got a lot of years where we have seen lots of things but we have not seen this with such a minimum disturbance to a child."

17. Mr Richards was then cross-examined by Mr Frank Feehan QC on behalf of the father. I need not set it all out, for necessarily it traversed ground already covered by Mr Kirk:

"Mr Feehan … the problem is, as you see it, that there is no such brain injury. You have looked at the scans and you, and in fact nobody, sees anything with regard to brain injury in this case?

Mr Richards That is correct. And it is not just on the inside of the brain; the outside, her scalp, does not seem to have reacted in a way you would normally expect, this sort of high force injury that you would expect to cause that fracturing.

…

Mr Feehan In your experience, where someone has suffered an insult to the head of this sort on a normal skull and therefore has suffered brain injury in almost every case, how long do the symptoms of that brain injury last, if they survive?

Mr Richards Well, if they survive … Not many with this degree of injury, fracturing and damage do survive. I had a recent one which was a very great surprise because the initial scan gave us the impression she was not going to survive. She was run over by a Ford Fiesta in a supermarket car park. She got under the car, it drove over her head, the people screamed at the guy to stop so he reversed and did it again and she spent three or four days on a ventilator in intensive care. We at one stage thought she would not survive. She did survive and recovered. She was in hospital for about a month to six weeks before she could go home.

…

Mr Feehan Essentially, it all comes down to this, does it not, that given the clinical picture and the timing of this the notion of inflicted injury without other injuries such as massive amounts of other broken bones or brain injury just does not fit?

Mr Richards That is correct. It does not add up. To me as a simple(?) surgeon, it suggests there is something wrong with this young lady's skull. It is unduly fragile which is why I suggested that we needed a metabolic expert to look into this and he has not been able to identify it but it still does not mean to me that this is a normal skull and as we do in medicine, we sometimes show things to colleagues, not with the full story, not with any names. I showed the CT scans to one of my senior colleagues. There are four of us in the department and the one who is just below me in terms of age and I said to him "Look at this scan. This is a child with no brain injury and very little scalp swelling" and he immediately said "There is something wrong with the bones, isn't there?" … This brings to mind for me that a man of Professor Bishop's extreme skill has not been able to find anything …"

18. Mr Feehan returned to the point towards the end of his cross-examination:

"Mr Feehan … we are simply left with this then, that had this been a deliberately inflicted injury causing these fractures in a normal skull, five days before presentation to hospital, first of all your view is she may well not have survived and even if she had survived she would simply not have presented in the way that she did.

Mr Richards Yes, I think she would have ... again, it is odd that she does not have so much scalp swelling with these degree of fractures. Even if she had got away with the brain, there is still the soft tissue injury you would expect to be associated with it and you can have fractures all over the head, while you have just one small area of slight swelling, as that is all you see radiologically, why not over the whole of the head, unless the trauma that caused these fractures was less than you would normally expect in these fractures and that is why the response to it in the brain, the patient and the scalp is much less."

19. Dr Halliday was cross-examined by Mr Kirk. She agreed with what Mr Richards had said at the end of his cross-examination on this point by Mr Kirk (see paragraph 16 above). She went on:

"l think there are several things about this fracture which are extremely unusual. One is its severity. The other is the fact that it is not associated with any brain injury and the third thing which is unusual is that it has … the amount of soft tissue swelling on the CT scan is not very much for such a severe fracture and I think all of us are in agreement that we think these things are very unusual. And can't really explain them."

Asked by Mr Kirk whether this was "without parallel in terms of medical reporting and medical annals" she replied:

"Yes, yes, yes they are very unusual. Very unusual."

20. Dr Halliday was cross-examined by Mr Feehan. Referring to Mr Richards she said:

"l wouldn't say M's skull can't be normal as you know we doctors don't like saying things like that, we just say it is very, very unusual so I agree entirely with him that it is very, very unusual. Does it mean that M's skull cannot be normal? No I don't think it means that, I just think she has had a very unusual injury. There is some ... either ... I mean it could be that her skull is abnormal or it could be that she has ... there is some sort of quirk of mechanics about the way she has fallen or that she has had several blows or ... I think there are a number of explanations but they are extremely unusual."

21. Dr Chapman was cross-examined on the point by Mr Feehan:

"Mr Feehan Dr Richards said that his experience of a child suffering sufficient trauma to cause these skull fractures would mean that that child, again, if that child survived, that child would be in intensive care for upwards of two weeks and in hospital, in his experience, for at least a month, if not longer, and yet here we know there was a child who presented to hospital as happy, playful and alert. How does that fit with the kind of forces you describe as being necessary to cause these skull fractures?

Dr Chapman Well, if Mr Richards is right in that the child would be ill for many, many days, and clearly wasn't upset and ill at the time she was presented for medical attention on the 3^rd, then that would imply that the injury occurred nearer to 26^th July than it did to 3^rd August, and (inaudible) ...

Mr Feehan I'll ask you to pause there because it was clear from Mr Richards' evidence that the timeframe was simply insufficient to allow recovery. Even the wider timeframe was insufficient to allow recovery of a child to the state seen by 3rd August in this child. Now, let me say this, again, you are a radiologist and he is a clinician in that sense. He sees the children himself and has experience of how children respond to these sorts of injuries. Given that's his evidence, how does it fit with the level of trauma that you say is required?

Dr Chapman I'm not sure how to answer that."

Towards the end of cross-examination, Dr Chapman said this:

"Well, I mean, you could argue that (inaudible) that there are an awful lot of inexplicable things in this child."

And then at the very end:

"Mr Feehan And you, however, Dr Chapman, will be aware from your experience, and no doubt from other cases, that, in giving evidence, you have to at least acknowledge the real possibility that something has happened which simply you can't explain.

Dr Chapman Of course."

22. Cross-examined by Mr Kirk, Dr Chapman associated himself with what Mr Richards had said at the end of his cross-examination on this point by Mr Kirk.
23. Professor Bishop was cross-examined by Mr Kirk:

"Mr Kirk These fractures are, the skull fractures I'm talking about, are indeed highly unusual, if not, dare I go as far as to say, in the annals of medical science, unique in one so young?

Professor Bishop (inaudible) I certainly have not seen anything similar in my clinical practice, focusing on children with bone disease, in the last 16/17 years. That is not to say that it can't occur and hasn't occurred in other places and I have simply not been aware of publications that might record such fracturing. But talking with Dr Chapman and Mr Richards, who have also had a lot of experience with looking at x-rays and with traumatic injuries to the skull, they were very much of the same view, that these were exceptionally extensive fractures. And my understanding is that it was for that reason that people were concerned there might be an underlying metabolic bone disease that could give rise to such appearances and that was why I was asked to see M and examine her."

A little later there was this interchange:

"Mr Kirk … How can one say that this little girl, M, is not, in fact, exhibiting the signs of some new or perhaps hitherto unknown medical disorder in the light of the fractures and her presentation?

Professor Bishop Thank you. Yes. That, I suppose, is a Donald Rumsfeld question. We don't know what we don't know. She certainly ... there is a possibility that she could have a hitherto unknown and undiagnosed metabolic bone disorder. However, in my experience, children who have such a disorder usually continue to manifest problems associated with such a disorder throughout their life, albeit more frequently at some times in their lives than at other times. And my expectation, based on my clinical practice over the 15 years, would be that we would still be able to discern, at the age of a year, some evidence of underlying bone fragility problem at the age of a year in a child who suffered fractures earlier during infancy. And that's, as I say, based on our clinical experience on my understanding of the literature and many discussions over the years with friends and colleagues who are also involved in the field. Of course, it is still possible that at the end of all of that this is something new that nobody has ever come across before, but I would still expect, as I say, to see some other evidence of a persisting problem with the bones had the skeleton been so fragile that it had given rise both to the rib fractures and the skull fracturing that were observed originally.

Mr Kirk Thank you for that, Professor Bishop, but just by way of, as it were, come-back or a question, further question on that, if this is, and I use the words 'completely new and unchartered territory', one surely can't be so, as it were, positive in terms of your predictions and predictors, what would you say in answer to that?

Professor Bishop I think it's absolutely true that every year that goes by we find some new disorder or other that can give rise to a problem with your bones, but it's a question of balance of probabilities here and the number of children that we have seen over the years who present in the way that M presented and who've had bone fragility problems that give rise to such fractures and who then never exhibit any further features of bone fragility subsequently, that, in my experience, would be unique.

Mr Kirk And just, I think, finally this, if I may, you talk about M's presentation. Professor Bishop, I don't want to take you back through all of that because her Honour, the court and others have heard from other experts, I described her presentation earlier perhaps as being unique, but certainly the fact that this child remained alive with no massive swelling to the scalp, no significant brain injury and handled perfectly very well in hospital could themselves indeed be described as unique features; would you agree with me on that?

Professor Bishop I would agree in terms of it's surprising that M suffered no brain injury given the extent of her skull fracture. But, as I said before, that degree of skull fracturing in itself is very unusual, in fact, unique in my experience, and if ... I don't know what else to say, really, other than yes, it's a very unusual presentation. But, going back to what I said previously, if there was any specific problems with the bones that was likely to give rise to that degree of fracturing I would expect to see some evidence of consistence of that fracturing and further fracturing beyond the time when she first presented to hospital. That would be more in keeping with my clinical experience of how such disease tend to manifest themselves not just at the time of presentation but subsequently as well."

24. Cross-examined by Mr Feehan, Professsor Bishop said this:

"in my previous experience of similar matters it is surprising that M did not suffer any brain tissue injury commensurate with the degree of fracturing that she has."

He was asked by Mr Feehan, "how can you explain the extent of skull fractures, given the lack of brain injury?" He replied:

"Well, I defer to Mr Richards in terms of his experience of these matters, and as I say, it is not my area of expertise drawing conclusions from the extent of skull fracturing that's unlikely to be a brain injury. One might postulate if one was speculating here that M has been very fortunate in that the force to which her head was subjected result in fracturing which dissipated the force of the impact or whatever it was that caused the fracturing sufficiently that the brain was relatively protected and she is indeed very fortunate in that regard. But that is speculation on my part. I don't have any good answer for you as to why this degree of fracturing has not resulted in more extensive intracranial injury."

His response to the next question was:

"I will defer to Mr Richards in terms of his much more extensive experience of these matters, and I think I'd have to refer you back to my previous answer that if you've got energy applied to a bone which is sufficient to fracture it, effectively what you are doing when you fracture it is to dissipate the energy but the bone, if absorbing some of that energy and then it's breaking when the energy exceeds its capacity to absorb that ... sorry, when the energy of the force exceeds the bone's capacity to absorb it. Now, why in this particular circumstance bone has fractured in the way that it has done I can't answer and I can't begin to speculate on the amount of force that might have been required to cause such a fracture. What I can say is that M's skeleton, as it appeared to me on 8^th May, was normal, and there was nothing that I could find to give me a reason to suspect that her bones were other than normal."

He was asked by Mr Feehan "what else might be an explanation for what we see?" Professor Bishop's answer was "I think you are asking me to speculate beyond the limits of my expertise."

The expert evidence – summary

25. I have deliberately set out the expert evidence relating to the skull fractures at some length. It is important to see exactly what these very experienced and distinguished experts were and were not saying. But it is also important to be able to get a 'feel' for the case and to see – we can only read it on the printed page; the judge of course could watch and hear – not merely the care with which skilled counsel confronted each of them in turn with the central conundrum in the case but also the very evident difficulties they all had in coming to any answers.
26. That said, what the judge was presented with at the end of the day was a marked uniformity of expert opinion:

i) M had "spectacular" fractures of her skull.

ii) What I will call 'conventional' medical opinion, based on very extensive experience, would have expected such extensive fractures to be accompanied by: (a) extensive brain damage, (b) extensive swelling of the scalp (c) pain and (d) a patient presenting as either dead (or dying) or unconscious and requiring intensive care.

iii) In contrast here, M had no significant brain injury, only minimal scalp swelling and presented as apparently pain-free and reasonably well, "a seemingly happy baby".

iv) In the experience of all four experts M's case was unique: none of them had ever seen such a case and none was identified from the literature.

v) None of the experts was able to come up with any explanation that went beyond what each acknowledged was mere speculation.

In sum, M's skull fractures were inexplicable.

27. The language used by these very experienced experts is striking. A selective anthology includes, as we have seen, such phrases as: "cannot be determined", "remain unexplained", "something strange going on", "very odd and unusual", "it does not add up" (Mr Richards); "several things … extremely unusual", "can't really explain them", "without parallel", "very, very unusual", "extremely unusual" (Dr Halliday); "an awful lot of inexplicable things" (Dr Chapman); "very unusual, in fact, unique", "a very unusual presentation", "surprising", "I don't have any good answer", "I can't answer and I can't begin to speculate" (Professor Bishop); "unprecedented" (Mr Richards, agreed by Dr Halliday, Dr Chapman); "speculation" (Professor Bishop, Dr Chapman). The judge was entirely justified in characterising the case (paragraph 2) as "not just unusual … but inexplicable."
28. There is a further important point that emerges from the expert evidence. Mr Richards had envisaged as a possible explanation that M might have abnormal bone fragility – that, after all, was why Professor Bishop was asked for his opinion. But Professor Bishop was adamant that there was no such abnormality; M, in his opinion, has a normal skeleton. This left Mr Richards completely at a loss. Professor Bishop's evidence did not make the case easier; it made it more difficult, indeed inexplicable. As Mr Richards had said in his original report, and he never wavered, nor did any of the others gainsay him, "If abnormal bone fragility is excluded, the fractures remain unexplained." But equally, Professor Bishop was unable to come up with any explanation, assuming that is, as he believed, that M's skeleton was normal. The simple fact, given Professor Bishop's evidence, is that none of these very experienced experts had any explanation for what they were seeing. That is the central fact with which Judge Marshall had to grapple. Did she?
29. There is one other feature of the expert evidence that requires comment. The only material in the medical literature to which the judge was referred was research carried out on the cadavers of small children to determine the height from which they had to be dropped to produce fractures of the skull. This of its very nature did not assist the judge in grappling with the problem that confronted her, namely a very young child presenting with the "spectacular" fractures M had suffered. The mystery was not that M had survived – for Mr Richards was aware of such cases, at least in adults – but the absence of associated trauma. Now I have to say that I would find it very surprising if there is nothing to be found in the literature relating to spectacular skull fractures and only a little less surprising if there is nothing to be found in the literature relating to spectacular skull fractures where there is no associated trauma. Be that as it may, in a case where it was being asserted that something was, from a medical perspective, unprecedented and inexplicable, it would surely have assisted the judge either to be taken to such literature as there is, which might have provided a clue to what had happened, or to be told that an appropriately extensive search of the literature had produced nothing, in which case a finding could more confidently have been made that what had happened was inexplicable.

The judgment

30. Judge Marshall summarised the legal principles she had to apply largely by reference to the ten principles set out by Baker J in Re JS (A Minor) [2012] EWHC 1370 (Fam), paras 36-45. No one has criticised her for doing this and she was, in my judgment, well justified in adopting this approach. Baker J's analysis of the authorities is both helpful and, so far as it goes, accurate.
31. As part of that exercise Judge Marshall referred to what Hedley J said in Re R (Care Proceedings: Causation) [2011] EWHC 1715 (Fam), [2011] 2 FLR 1384, para 10. I propose to expand the quotation slightly. Hedley J referred to what Moses LJ had said in R v Henderson; Butler; Oyediran [2010] EWCA Crim 1269, [2011] 1 FLR 547, para 1: "There remains a temptation to believe that it is always possible to identify the cause of injury to a child." He continued:

"The temptation there described is ever present in family proceedings too and, in my judgment, should be as firmly resisted there as the courts are required to resist it in criminal law. In other words, there has to be factored into every case which concerns a disputed aetiology giving rise to significant harm a consideration as to whether the cause is unknown. That affects neither the burden nor the standard of proof. It is simply a factor to be taken into account in deciding whether the causation advanced by the one shouldering the burden of proof is established on the balance of probabilities."

He added this (para 19):

"I have been impressed over the years by the willingness of the best paediatricians and those who practise in the specialities of paediatric medicine to recognise how much we do not know about the growth patterns and what goes wrong in them, particularly in infants. Since they grow at a remarkable speed and cannot themselves give any clue as to what is happening inside them, and since research using control samples is self-evidently impossible in many areas, perhaps we should not be surprised. In my judgment, a conclusion of unknown aetiology in respect of an infant represents neither professional nor forensic failure. It simply recognises that we still have much to learn and it also recognises that it is dangerous and wrong to infer non-accidental injury merely from the absence of any other understood mechanism. Maybe it simply represents a general acknowledgement that we are fearfully and wonderfully made."

I respectfully agree with every word of what Hedley J said.

32. In addition Judge Marshall appropriately made reference to what Dame Elizabeth Butler-Sloss P said in In re U (A Child) (Department for Education and Skills intervening), In re B (A Child) (Department for Education and Skills intervening) [2004] EWCA Civ 567, [2005] Fam 134, para 23:

"… there is a broad measure of agreement as to some of the considerations emphasised by the judgment in R v Cannings that are of direct application in care proceedings. We adopt the following. (i) The cause of an injury or an episode that cannot be explained scientifically remains equivocal. (ii) Recurrence is not in itself probative. (iii) Particular caution is necessary in any case where the medical experts disagree, one opinion declining to exclude a reasonable possibility of natural cause. (iv) The Court must always be on guard against the over-dogmatic expert, the expert whose reputation or amour propre is at stake, or the expert who has developed a scientific prejudice. (v) The judge in care proceedings must never forget that today's medical certainty may be discarded by the next generation of experts or that scientific research will throw light into corners that are at present dark."

Mr Feehan understandably places particular reliance on (i) and (v).

33. Having dealt with the law, Judge Marshall turned to consider the background, first the general family background and then the events immediately before and after M's admission to hospital on 3 August 2011. She then turned to the medical evidence, commenting (paragraph 48) that:

"There is a large measure of agreement between the experts, but their evidence requires careful consideration as it is apparent that they all continue to find this a particularly perplexing and unusual case."

34. The judge dealt first with Professor Bishop's evidence, saying (paragraph 49):

"I note that Mr Richards remained perplexed when giving his evidence as to how M could have sustained these injuries, but apparently without accompanying brain disturbance or significant pain and swelling and wondered if in fact there might be some underlying disorder that the experts have not yet found. On this issue, in so far as it relates to fragile bones, I prefer to rely on the evidence of Professor Bishop, this being his particular field of expertise. The observations he makes and the explanations he gives for reaching his conclusions appear well reasoned and supported by extensive clinical experience. His conclusions are consistent with the radiological observations of Dr Halliday and Dr Chapman who found no signs of bone disorder and did not find it a likely explanation in this case."

In relation to this, three comments need to be made. First, Mr Richards was not merely "perplexed"; he was, as the judge went on to say, "baffled". Second, it was not just Mr Richards who was perplexed and baffled; so were all the experts, as the judge had in fact acknowledged in the previous paragraph. Third, the fact that, on the issue of underlying bone disorder, the judge preferred to rely on Professor Bishop's evidence, did not resolve the case or even make it any easier to resolve. In fact, as I have already pointed out, it made it more difficult.

35. The judge then considered the rib fractures before turning to consider the skull fractures. She summarised the evidence of Mr Richards, Dr Chapman and Dr Halliday. Of Mr Richards she said (paragraph 61):

"Mr Richards remained baffled by the lack of brain injury which he would expect to be present if M's skull has been crushed or compressed. It would require just the right amount of force to be applied to cause the bone to crack, but then to immediately stop. Rather than a crushing injury, as the main fracture appears to be from the top of the head spreading out, he suggested a blow to the top of the head might be a possible cause."

Again, I make the point that it was not just Mr Richards who remained baffled.

36. Judge Marshall then turned to an evaluation of the background evidence. Finally, and for present purposes this is the crucial part of her judgment, she set out her findings in relation to M's injuries. She said (paragraph 71):

"There seems to be some acceptance that the injuries are likely to have been caused by trauma; what I have to determine is whether the local authority has proved that these were caused non-accidentally, rather than accidentally as the parents have described. The evidence of the parents and the other witnesses about these matters requires to be assessed in terms of its credibility and reliability and considered together with the medical evidence."

37. She went on to say that "There are a number of puzzling factors in this case." So there were; but some matters, amongst them the most important, were not merely puzzling but inexplicable. The first group related to the parents' behaviour on 3 August 2011 and the following day. In the course of dealing with that, and specifically in relation to the account the father gave to the police on 4 August 2011 of the incident he alleged had happened on 2 August 2011, the judge said this (paragraphs 78-79):

"… this explanation does not fit easily with how M might reasonably be expected to have presented had she just sustained the skull fractures and cannot, if I accept the medical evidence, explain the 8^th rib fracture. I see no reason not to rely on the radiologists' opinion that the 8^th rib fracture must have occurred earlier than 2 August 2011, and likely several days before. This evidence is given by two very experienced radiologists no doubt well-used to dating fractures from observed periosteal reaction, and is based not only their considerable experience, but their knowledge of the relevant medical literature. I rely on Mr Richards' opinion to find that M's clinical presentation on 3 August is not consistent with her having suffered the extensive skull fractures found less than 24 hours earlier. I am satisfied that this would be the case whether or not M has bone fragility, and this leads me to reject Mr Richards' hypothesis that M may have some as yet undetected or unknown condition that affects her bones, which is in any event speculation outside his area of expertise.

The evidence of Professor Bishop and the radiologists supports a finding that bone fragility is not a feature in this case, and I rule this out as a possible explanation."

I repeat the point I have already made in paragraph 34 above.

38. The judge then considered how the parents' evidence more generally fitted the medical evidence in relation to the fractures. She then grappled with what she said was a conundrum that could be explained only by speculation (paragraphs 82-83):

"A finding that the skull and 8^th rib fracture occurred as early as 26 July would fit with the evidence that M appeared fine on 24 July when seen by the maternal grandmother, and it would be likely that she would be some way to recovering when seen 5 days later on 31 July 2011. It might also perhaps go some way to explaining M's clinical presentation as at 3 August 2011, so perplexing to the experts in the light of the fractures. Dr Halliday and Dr Chapman have both postulated that the swelling seen at that time might well have been resolving …

If indeed it is a real possibility that M suffered such serious injuries around 26 July, for which no medical treatment was sought, I still have to ask myself what was it that happened on 3 August that led to M being taken to the medical centre. This is, on the face of it, a conundrum and any attempt to explain it can only be speculation. However, it is not, in my view, so inexplicable as to exclude any other possibility than that the parents are telling the truth, particularly given the discrepancies and difficulties I have already identified in these accounts."

39. Judge Marshall set out her conclusions in three paragraphs that I should quote in full (paragraphs 84-86):

"This is not a case where I find the parents accounts so compelling or reliable that I am persuaded to accept what they say and reject well-founded medical evidence where it does not fit. I am not persuaded that this is a case where the court should find that there is something unexplained, beyond current medical knowledge. Father on his own account did not tell the truth at the beginning, he says from 'cowardice', and it is possible that he has continued to give an account that when it was first given was designed to cover up or minimise what really happened to M. Mother's evidence I have found to be truthful on a number of issues, but the very difficult circumstances she has had to contend with and her refusal to accept help previously when she might reasonably have expected to need it, coupled with her concerns about the involvement of social services suggests that she is likely to be less than open in admitting problems, be fearful of the consequences, and not give a fully honest account.

There are features of this case that are indeed perplexing, but I apply the legal principles that I set out at the beginning of this judgement in order to reach a decision. I am required to consider matters on the balance of probabilities. I have to balance the likelihood that these parents are telling the truth against the likelihood that they are not, and some other incident has occurred about which the court has not been told. Based on my analysis of the evidence as set out above, I conclude that the evidence is weighted in favour of a finding that it is more likely than not that the parents have not told this court the truth about what has happened to M.

How M's skull fractures were caused remains a mystery, but there must, of course, be an explanation, albeit at this stage unknown. The fact that the parents have not provided a truthful account supports a finding on the evidence, including the drawing of reasonable inference, that it is more likely than not that the injuries are non-accidental. A finding that M has suffered non-accidental injuries, including a rib fracture, post-dating the earlier rib fractures lends support to a finding on the balance of probabilities that the earlier rib injuries are also more likely than not non-accidental. I also factor in the known correlation between rib fractures and non-accidental injury in children of this age."

40. Finally, Judge Marshall turned to consider whether she could identify which of the parents was the perpetrator. She said she could not (paragraph 87):

"Having found M's injuries to be more likely than not non-accidental, I have also considered whether it is possible to identify who may be the perpetrator of those injuries. The local authority asks the court to find that both parents remain in the pool of potential perpetrators, although in cross examination, it was put to father that he had made up his account to protect mother. The evidence suggests that mother can be fiery, has a temper and has previously been violent towards father, causing him to have a black eye. She has also had difficulties in the past in coping. I do not find this sufficient to allow the court to conclude that, in temper, mother would harm her children. Previously when mother was having difficulty in coping, she would hand over the care of L to father, and when she was not able to do that in 2011, father being absent, there is no evidence the children came to harm in her sole care. The evidence suggests that at the time of these injuries, it was father who was having difficulty in his relationship with M, who had become so attached to mother that she did not like being fed by anyone else. Father had spent a significant period of time away from the family, and mother's evidence suggests that they were trying to improve the bond between M and father. Father is also said to be more 'laid-back' than mother and had coped previously when he had to take over care of L. After careful consideration, I find myself unable, on the balance of probabilities to determine that either mother or father is more likely than not the perpetrator and both must remain in the pool."

The appeal

41. The judge, as I have said, gave permission to appeal. The father's appellant's notice was filed on 7 September 2012. It contained four grounds of appeal and sought an order for a new trial. The mother's notice followed a little later. It adopted the father's grounds of appeal. The appeal came on before us on 12 November 2012.
42. The father was represented, as below, by Mr Frank Feehan QC and Ms Alexa Storey-Rea, who had filed a most helpful skeleton argument. The mother, although present, was not represented. Mr Anthony Kirk QC and Miss Shona Rogers, who had appeared for her below, had filed a skeleton argument saying that they adopted and endorsed the submissions contained in the father's grounds of appeal and skeleton argument and explaining that separate representation could not be justified in a case where the mother was publicly funded. In fact Miss Rogers was present, pro bono, with a watching brief and to support the mother. I am grateful to her for this, as I am sure the mother is also. The local authority was represented by Ms Frances Judd QC, who had not appeared below, and Ms Hayley Griffiths, who did. The children's guardian was, perfectly appropriately in the circumstances, neither present nor represented, but Mr John Ker-Reid, who had appeared for her below, filed a most helpful skeleton argument.

The parents' submissions

43. The parents' central submission, articulated and elaborated with great skill by Mr Feehan, is that Judge Marshall never really grappled with the fact that the experts found M's presentation to be inexplicable and that she failed adequately to address the parents' case that there was, on a proper consideration of all the medical evidence, good reason to conclude that M had some unknown and undiagnosed disorder of her bones. The possibility of such a disorder, Mr Feehan submits, simply could not be ruled out, not least in the light of Professor Bishop's reference to what he called the "Donald Rumsfeld question" and his acknowledgement that "there is a possibility that she could have a hitherto unknown and undiagnosed metabolic bone disorder" and that "every year that goes by we find some new disorder or other that can give rise to a problem with your bones." Putting the same point rather differently, the judge, Mr Feehan says, failed to heed what Dame Elizabeth Butler-Sloss P had said in In re U (A Child) (Department for Education and Skills intervening), In re B (A Child) (Department for Education and Skills intervening) [2004] EWCA Civ 567, [2005] Fam 134, para 23, and failed to give adequate weight to what Hedley J had said in Re R (Care Proceedings: Causation) [2011] EWHC 1715 (Fam), [2011] 2 FLR 1384, para 10.
44. More specifically, Mr Feehan says, Judge Marshall never explained how the various deficiencies in the parents' evidence could justify a finding against them when, as the judge herself recognised (paragraphs 2, 86), the skull fractures were "inexplicable", their causation "a mystery" and the explanation "unknown."

The local authority's submissions

45. Ms Judd, in her equally skilful submissions on behalf of the local authority drew attention to what Mr Richards had said in paragraph 2.8 of his report, where he had identified possible explanations, if some bone fragility syndrome was ruled out, involving events of which a carer would have been aware. She identified a number of significant discrepancies in and difficulties with the parents' evidence which, she submits, entitled Judge Marshall to reject it. She identified various features in the case which, she says, "are all hallmarks of non-accidental injury." Nothing that the parents had described could, she submits, properly explain either the rib fractures or the skull fractures.
46. Ms Judd submits that Judge Marshall was entitled to prefer Professor Bishop's evidence to that of Mr Richards, while remaining, as she did, says Ms Judd, acutely conscious of Mr Richards' evidence and the dilemma he had identified. The judge directed herself to what Dame Elizabeth Butler-Sloss P and Hedley J had said. What she said about causation being a "mystery" and the explanation "unknown" was merely a reference to the fact that the parents had not given a truthful account as to what had happened. The judge's inference that the injuries were non-accidental was, Ms Judd submits, a proper finding that was open to her on all the evidence before her, as were her findings as to when the skull fractures were most likely to have taken place. The judge, she says, carefully had regard to all the evidence in the case and was scrupulously fair in her approach. The judge appropriately gave herself a Lucas direction (see R v Lucas (Ruth) [1981] QB 720). In all the circumstances, Ms Judd submits, Judge Marshall was acting within the bounds of her discretion and making findings that were open to her on the evidence. This court, she says, cannot and should not interfere.
47. In support of these submissions, Ms Judd points out that Judge Marshall was steeped in this case in a way an appellate court never can be: see Re A (Fact-Finding: Disputed Findings) [2011] EWCA Civ 12, [2011] 1 FLR 1817, paras 36-38. And appropriately she reminded us of what Lord Hoffmann said in Piglowska v Piglowski [1999] 1 WLR 1360, 1372, and urged us to resist what Lord Hoffmann had identified as the temptation to resort to narrow textual analysis as a means of finding fault with the judgment under scrutiny.

The guardian's submissions

48. Mr Ker-Reid in his measured and thoughtful submissions suggests that:

"The judge was faced with a range of unknowns not determinable by medical science in a living infant."

M's presentation and the parents' lack of credibility, he says, made the issue of timing "very problematic, imprecise and unsafe" for the experts and the judge, but Judge Marshall was entitled, he says, to take the view that the parents' accounts were unreliable on "context, timing, effect and consequences." He acknowledges that on the issue of the skull fractures, and given the evidence, the judge was faced with a "very difficult decision" in the light of the authorities.

49. Mr Ker-Reid says that Judge Marshall's analysis was "sound and careful." She directed herself to the arguments and the relevant authorities. She balanced, he says, the "very difficult and probably unique elements of the case" with what he submits was the clearest and most reliable evidence that there was no bone abnormality and with the parents lack of credibility, honesty and reliability. There was, he submits, clear evidence of serial inflicted injury of the ribs. In conclusion, he submits that the judge's "sound analysis provides, on a difficult balance, the prevailing merit argued by the local authority for this judgment to remain undisturbed on appeal."

Discussion

50. I should at once pay tribute to the care and skill with which Judge Marshall went about the exceptionally difficult task facing her. Her judgment, if I may say so, is impressive and, in all respects save one, compelling. But for one crucial factor it would, I suspect, have been impregnable. If, in addition to the other factures, M had had only a simple skull fracture then the judge's findings would in all probability have been unassailable.
51. As it is, many of her findings cannot, I suspect, sensibly be challenged. Thus, even allowing for the fact that M's presentation was medically inexplicable, I think Mr Feehan would have faced a very uphill task if he had tried to argue (though he did not need to) that the judge was not entitled to conclude, as she did (paragraph 84), that:

"This is not a case where I find the parents accounts so compelling or reliable that I am persuaded to accept what they say and reject well-founded medical evidence where it does not fit."

That was not a finding driven by the medical frame of reference; it was, in large part, as the judgment explains, based on difficulties in the parents' own evidence. Similarly, it is difficult to challenge what the judge said a little later (paragraph 85):

"I conclude that the evidence is weighted in favour of a finding that it is more likely than not that the parents have not told this court the truth about what has happened to M."

52. But, of course, the case was not as simple as that.
53. For present purposes one can, I think, start with two propositions which I shall assume, even if not necessarily accepting: first, that the judge, for the reasons she gave, was entitled to rely on Professor Bishop's evidence; second, and again for the reasons she gave, that the judge was entitled to reject the parents' evidence. But where did this leave her? Facing the central conundrum that all the experts were baffled – none could explain M's presentation in the light of the skull fractures – and, as I have already pointed out, that her acceptance of Professor Bishop's evidence did not resolve the case, it made it more difficult.
54. In addressing that central conundrum Judge Marshall, in my judgment, fell into error. One can see the three facets of that error developing gradually through the key passages in her judgment. In the first place (see paragraphs 49, 61), by focusing on Mr Richards she tended to downplay the fact that, as she had previously recognised (paragraphs 2, 48), all the experts were baffled. Second (see paragraphs 49, 79), she failed to appreciate (or if she did appreciate it she failed to address) the fact that her acceptance of Professor Bishop's evidence did not resolve the case but made it more difficult. Third, and crucially (see paragraph 86), she failed to explain how she was able to move direct from acceptance that:

"How M's skull fractures were caused remains a mystery, but there must, of course, be an explanation, albeit at this stage unknown",

to a finding in the very next sentence that:

"The fact that the parents have not provided a truthful account supports a finding on the evidence, including the drawing of reasonable inference, that it is more likely than not that the injuries are non-accidental."

55. In relation to this, I should make clear that I cannot accept Ms Judd's reading of what the judge was saying here. After all, at the very beginning of her judgment (paragraph 2) Judge Marshall had described M's presentation as "not just unusual … but inexplicable", and in my judgment she was, for the reasons I have already given, entirely justified in characterising the case in this way. That was, after all, the effect of the expert evidence. She had, correctly, acknowledged (paragraph 48) that all the experts found it a "particularly perplexing and unusual case" and (paragraph 85) that there were features of the case "that are indeed perplexing." So if, when she was describing causation as a "mystery" and the explanation as "unknown" (paragraph 86), Judge Marshall was meaning no more than that the parents had provided no adequate explanation – and that is not how I read her words given what she had said elsewhere in her judgment – her analysis would be liable to the further criticism that she was at this point simply not engaging at all with the substance of the expert evidence.
56. There is another difficulty with Judge Marshall's judgment which in a sense encapsulates all the others. It is to be found in a passage (paragraph 84) that I have already quoted:

"This is not a case where I find the parents accounts so compelling or reliable that I am persuaded to accept what they say and reject well-founded medical evidence where it does not fit. I am not persuaded that this is a case where the court should find that there is something unexplained, beyond current medical knowledge. Father on his own account did not tell the truth … (emphasis added)."

As I have already explained I have no particular difficulty with the first sentence, nor for that matter with the third. But the second sentence, which I have emphasised, presents major difficulties. The judge's refusal to find that there was "something unexplained, beyond current medical knowledge" was not merely inconsistent with what she had said at the outset of her judgment (paragraph 2) and, as I read it, with what she said only two paragraphs further on (paragraph 86); perhaps more to the point it was inconsistent with the expert evidence. At the very least, it was a finding which, given the expert evidence, required much more explanation, a much more convincing process of reasoning, than is to be found anywhere in the judgment.

57. In these circumstances I conclude that, for all its many merits, this judgment cannot stand. Judge Marshall has simply not explained how, in the light of the expert evidence, she was able to arrive at her final conclusion. It may be that there was, in the light of the evidence, no way in which the judge could have arrived at her conclusion, though that is not, in my judgment, something we can safely and properly decide for ourselves. But certainly, and it is on this basis that, in my judgment, this appeal must be allowed, the judge has not explained how she arrived at her conclusion. That omission, going to the central issue in the case, is fatal and necessarily vitiates her findings.
58. Standing back from all the detail it might be said that given this collection of injuries – rib fractures caused on two occasions and very severe skull fractures – arising in circumstances where there is no suggestion that anyone other than the parents (or one of them) could have been responsible, and where the parents' accounts and explanations have appropriately been rejected, it was open to a judge to find as Judge Marshall did. So, I have little doubt, it would have been if the case had fallen to be determined in accordance with a conventional medical frame of reference. But it did not, for, to repeat the point, so far as the skull fractures were concerned M's presentation was inexplicable assessed by reference to conventional medical opinion and experience. And it must follow from the inability of conventional learning to explain the skull fractures that the judge's findings in relation to the rib fractures are also vulnerable. So there is, in my judgment, no scope for drawing a distinction and upholding the judge's findings in relation to the one whilst upsetting her judgment on the other.

Conclusions

59. It was for these reasons that, at the conclusion of the hearing on 12 November 2012, I agreed with my Lords that this appeal had to be allowed.
60. Mr Feehan and Ms Judd were at one in agreeing that it would not be appropriate for us to substitute our own findings and that if the appeal were to be allowed it could only be on the basis that the matter is re-tried. I agree. I do not, in the circumstances, think it would be appropriate to remit the case to Judge Marshall. The re-trial should, in my judgment, be before a judge of the Family Division, sitting either on circuit or in London as the Family Division Liaison Judge for the Western Circuit, Baker J, may determine.
61. Careful thought will need to be given to what directions are required for the re-trial. That is not a matter for us to resolve, but for the trial judge who will no doubt wish to have a case management or directions hearing. It may assist, however, if I make two points. First, in the light of what I have already said, an appropriately extensive search of the literature should surely be undertaken. Second, thought needs to be given as to how the expert evidence before Judge Marshall can be condensed or summarised for the re-trial so as to avoid taking the experts over the entire terrain again.
62. It will of course be for the judge who conducts the re-trial to determine the matter afresh in the light of all the evidence before him or her. It may be that a search of the literature will provide a clue to what had happened, so that the judge at the re-trial is not faced with something quite as inexplicable as Judge Marshall was. It may be that even an extensive search of the literature will produce nothing, in which case the judge may be faced with the same inexplicable mystery as confronted Judge Marshall. In that event, the judge will have to confront the same fundamental issue as Judge Marshall. Is it possible on the totality of the evidence to come to a finding despite the unexplained mystery? And if so, what process of reasoning can safely lead the judge to his finding? Or is the case one where, applying what Dame Elizabeth Butler-Sloss P and Hedley J said, no finding is possible?

Lord Justice Kitchin :

63. I agree.

Sir Stephen Sedley :

64. I also agree.