In October and November 2009, Lacie (L), then a two month old baby, sustained a number of serious injuries. Care proceedings were begun as a result. On 8 April 2011, at the end of an 18 day hearing to determine whether the injuries were non-accidental and if so who caused them, HHJ Cameron found that certain of the injuries were non-accidental and that they had been caused by L’s mother (M). M accepts the finding that there were non-accidental injuries but appeals against the finding that she was responsible for them.

It is necessary to set the scene by identifying the people who were central to L’s life at the time of her injury. The relationship between M and L’s father (F) came to an end in the early stages of M’s pregnancy with L but F had contact with L following her birth and since mid August 2011 L has been living with him. When she was about 6 months pregnant, M formed a relationship with RH. RH spent time with M, and L once she was born, and on occasions looked after L for short periods on his own, but it appears he was not living at M’s house. M’s mother (GM) and sister (JH) were also a regular feature of L’s life. Indeed the care of M’s older daughter (who is now 13) is shared between M and GM. M looks after her two boys, OA and LA, herself. They are aged 9 and 5 respectively. Living more or less full time in M’s house with her, L and the boys at the time of L’s injuries was ML. His status was that of friend/lodger.

The central parties in the care proceedings are the local authority, M, F and L’s guardian (GAL). A number of other people were joined as interveners for the fact finding hearing. They included RH, ML, GM and JH. Gradually, over the course of the hearing, it was possible for interveners to be excluded from the pool of possible perpetrators and the focus narrowed ultimately to a consideration only of RH, ML and M. ML acted in person during the hearing but the other parties and principal interveners were represented.

ML has not played any part in the appeal proceedings. The parties appearing before us were the appellant M, the local authority, F, RH and L by her guardian.

Judge Cameron’s judgment was reserved for several weeks and delivered orally over many hours. Transcribed, it is a long document and it refers comprehensively to the evidence that the judge received. In refusing permission to appeal, the judge said of it that it was “a very lengthy reserved judgment which took many days thoroughly to consider, draft, and reflect upon….”. It will be appreciated that a judgment (and hearing) of such length incorporates a great deal of material upon which I will draw only to the limited extent that is necessary to explain why, in my view, this appeal should be dismissed. References here to the judgment below will be identified simply as “§x”.

The first injury that L suffered, when she was around two weeks old, was bruising on her hand(s). There was certainly bruising on the palm of her right hand and the judge recorded (§50) that similar bruises were spoken of on the left hand too. It appears that there was hand bruising again on 28 October 2010 (see §79 and §117).

Dr Robinson gave paediatric evidence at the hearing. He said that bruising is extremely uncommon in a non-ambulatory baby and would involve blunt trauma, the child possibly being pinched, squeezed, hit or gripped. If the cause was accidental, it would be memorable. He found the hand bruising most unusual and very worrying in an infant of a few weeks of age.

No explanation for the marks was offered to the judge by M. She had earlier suggested that they were eczema, telling the hospital when L was taken there on 3 November 2009 with what turned out to be various fractures that L had had bruise-like areas on her hand which the GP thought was eczema. GM, who had noticed the marks and asked M about them, was also told by her that the doctor had said the marks were eczema. The evidence established that no GP had ever said this to M and that she had not even mentioned the marks to the doctors to whom she took L for other complaints. Judge Cameron attached significance to this and commented (§80) that this failure of M to mention the marks

The next injury in time may have been the transverse fracture of L’s left tibia. The timing of this injury is not exact. It was seen when L was given a skeletal x-ray on 4 November 2009 and was then healing. The doctors agreed during the hearing that it could be put at two to four weeks old. There appears to have been no explanation at all for this fracture.

L sustained what the judge described as “a deep cut” on the right side of her nose. Dr Robinson felt this to be a most unusual place and uncommon for an injury (§81), very worrying if it goes along with the injury to the hand and if there was no credible history of a memorable event.

When enlarged colour photographs of L were produced during the last week of the hearing, they showed, as the judge put it at §82, “possible further damage” (my emphasis) to the left nostril. The judgment seems to have proceeded at other points (§50 and §520) on the basis that there was in fact damage to the left nostril although we were told that the enlarged photographs were not shown to the doctors for them to give an opinion about this.

The nose injury probably occurred on 18 October 2009. L was in the care of RH when it happened; M was on the school run. When she returned, RH gave her an explanation of how the injury was caused which she apparently accepted as plausible and innocent, that is that he was feeding the baby with a bottle and she moved and he caught her on the side of her nose with his long fingernail. Although he told M about the incident, RH did not mention it to the police when they interviewed him or in his statements because, he said, he did not think it important at the time and he had too much to think about.

Judge Cameron treated RH’s explanation with caution. She appreciated that there were “all sorts of dynamics operating in the background here” as she said at §86 and that there were “untruths circulating around that nasal injury”. However she concluded (§87), on the balance of probabilities, that it was “an accidental unforced injury for which RH accurately took immediate and proper responsibility and that it was not any sort of episode of loss of control or frustration on his part.”

When she was presented at hospital on 3 November 2009 (the circumstances of which I shall set out later), L had a number of recent fractures: a displaced oblique spiral fracture of the mid-shaft of the right femur, distal metaphyseal fractures of the right tibia, a fracture of the metacarpal of the middle finger of the right hand, and greenstick or buckle fractures of the right distal radius and ulna. No healing reaction was visible on the x-rays of these on 3 and 4 November 2009. The doctors inferred from this that the fractures were then no more than 7 days old. However the timing was narrowed down in relation to the femur when it was x-rayed again on 9 November and there was still no healing reaction. As this meant that it must have been caused within the 7 days preceding the x-ray, it followed that it could be dated to 2 or 3 November and more specifically to the 24 hour period commencing with the early evening of 2 November. The judge specifically recorded (§134) that Dr Sprigg (a consultant paediatric radiologist) did not exclude that that fracture occurred at bath time on 3 November. She also recorded (§116 which can be found set out at paragraph 19 below) that Dr Robinson’s analysis was that it occurred at 6 p.m. on 3 November.

The first doctor to see L at hospital, at 7.48 p.m. on 3 November, reported swelling and tenderness. When examined at 1.45 a.m. on 4 November 2009, L was observed to have a swollen and tender right thigh with two bruises visible.

There was expert medical evidence about a number of aspects of the injuries. I did not detect from the judgment or from argument during the appeal that there was any significant difference between the doctors.

Dr Robinson’s evidence was that a rapid gripping or twisting force around the bone or knee joint was the typical causative mechanism for a spiral fracture such as that of the femur and he thought it also the likely causation for the bruises and swelling observed. Such a fracture would require significant force and be acutely painful, the pain being excruciating when the limb was manipulated causing bone and nerve to rub together with the result that the baby would cry as never heard before (§104). He said that L would have screamed in agony following the injury. In older children there may be severe pain for 30 to 60 minutes; a young baby would have remained in pain and fractious for many hours (§106). His evidence also contemplated that the infant may be in less pain when the limb is partly immobilised or supported (see §107 and 111). Swelling usually occurs within minutes of injury and certainly within an hour or two. It can develop and then gradually subside and disappear over 48 to 72 hours (§107).

Dr Robinson did not think it possible for the metaphyseal fracture to have been caused in the same act of force as the femur but it could have been done within minutes in one episode of abuse. The hand/wrist injury would also have required a separate action but could have been within the “same time frame of abuse” (§114).

The judge set out Dr Robinson’s analysis and conclusions as follows:

At §118, the judge turned to the evidence of Dr Sprigg about the fractures. She recorded that he “favoured three separate events” although in theory there could have been a minimum of two. I think the first of the three events in contemplation is the fracture of the left tibia which Dr Sprigg emphasised was older and could not have happened on 2 or 3 November as the femur fracture did. It follows that Dr Sprigg was contemplating two events which caused the most recent fractures.

Dr Sprigg also dealt with the pain that the injured child would suffer. The judge recorded his evidence at §119 as follows:

The judge went on to record more of Dr Sprigg’s evidence about pain as follows:

What the judge found about the events of 2 and 3 November and the personalities/reliability of those involved in them can only really be appreciated properly by reading the judgment as a whole. However, I do need to pick out some matters to serve as a foundation for what I say later.

The judge accepted the evidence of GM (§293) and JH (§233). She described GM as “a scrupulously fair and caring grandmother” (§503),

She found that ML’s recall and his evidence were flawed by imprecision and poor memory (§394). She found that he knew more than he had been prepared to divulge and his truthfulness could not be relied on (§548). She accepted the submission that he was “a strange admixture of a manipulated fool and a calculated deceiver” (§410). She did not consider that there was any real likelihood that he had hurt the baby and no one has argued that she was wrong about this.

She set down a number of her impressions of RH including that she found him “a dull, quiet and rather boring man” who “did not appear to operate on a short fuse” (§420) and she said that she did “not find that RH is an inherently violent man” (§442 and see also §477). She was critical of his approach to his evidence in some respects (see e.g. §449 where she said he had been “cavalier and unhelpful in the extreme in his witness statement”). He also lied to the court about his drug taking during his relationship with M (§424) and the judge commented that that damaged his credibility (§426).

As for M, the judge found that she “can very quickly become very angry and reactive” (see §304 and e.g. §307, §323, §377). She found that her credibility had been undermined (§312). She felt that at times during the evidence M was “quite a consummate actress” (§342), and that she made up fantasies to make her look important and independent in front of her sister (§351). As the judge said, her picture of M and her view as to whether there was a real possibility or likelihood that she had injured L was made up from “numerous strands and hallmarks in her own evidence and in other witnesses’ evidence about her” drawn together (§294).

Faced with evidence which was undermined by a significant number of lies, the judge gave herself an appropriate Lucas direction at the commencement of her judgment (§41) and there is every reason to suppose that she kept that in mind in considering each of the witnesses to whom it was relevant.

The judge described the weekend of Saturday 31 October/Sunday 1 November as a “febrile weekend” (§514). It commenced with a spat between M and RH on the Saturday evening when there was a Halloween party. The evidence was (e.g. §273, §411, §434) that RH was rude to GM on the Saturday evening, calling her a crap mother in what the judge called a “boorish out of the blue outburst” (§325). GM’s evidence (§276) was that when she visited on the Sunday evening, M met her at the front door in a really upset state; she had puffy eyes and had been crying and said that RH had been aggressive and violent to her the night before and started an argument. She told GM that he had drunk two 2-litre bottles of vodka and taken drugs.

L was with F for contact until around midday on Sunday 1 November. At some point after she was returned, RH left and M was at home with L and the boys for the rest of that day and overnight that night and on the night of Monday 2 November. She thought that ML stayed at the house overnight on both nights (§358). She saw GM early on the Sunday evening and met her sister at the noodle bar on the Monday evening with L. RH was not present at all during Monday although M and he were in contact by text, Facebook and telephone that as well as after he left on the Sunday.

There was evidence that L was unsettled on Sunday and Monday, 1 and 2 November 2009. The judge referred to this at various points in the judgment, for example at §315 where she referred to M’s account in her police interview that L was grizzling and did not settle until 10 or 11 p.m. on either day. When M and RH spoke on the telephone on the Monday evening, RH remarked to M that he could hear L screaming in the background. M seems to have given an account suggesting that this had gone on for some time. RH’s evidence was that M said she had had no sleep on the Sunday night. RH said he had never heard L scream like that before; it was similar to the really bad cry that he heard from her at the hospital. He said to M that if it continued, she must make an appointment at the doctor’s. He had never known the child to be up that late at night before (see §448 and surrounding paragraphs dealing with RH’s evidence; §316 dealing with M’s evidence). The judge commented, presumably referring to the state of affairs that Monday evening, that “[t]he Court views that as a possibly crucial event” (§448).

M took L to GM’s on Tuesday morning, 3 November, and GM looked after her whilst M took the boys to school. GM’s evidence (§280) was that L arrived whingey because M had not given her a bottle; M said she had been in a rush and L had only taken part of her milk (§363). GM first prop-fed her and then held her properly on her lap. She said there was no indication of discomfort although she did not change the baby’s nappy. M, GM, JH and L went shopping in town. The judge recorded GM’s evidence about this period as being that “the baby was her normal whingey and grizzly self for two or three hours that morning while they were shopping, and so on.”

M spent two hours or so in the car after pushing L around town for a little while because L was grizzly and then fell asleep and she did not want to disturb her after her unsettled night the night before (§364). After that, she picked up RH at around lunch-time, dropped off JH and went home. Her evidence was that the baby started to cry again when taken out of the car and she felt she just needed feeding. The baby was prop-fed whilst the shopping was put away. At some point she was moved from the car seat where she had spent much of the day to the baby bouncer. M and RH watched the television. The evidence was that L began to cry during the afternoon before M left at 2.50 p.m. to fetch the boys from school, leaving RH to look after L. She was out until 4 p.m.. RH’s evidence (§453) was that L was unsettled and crying. He patted her on his shoulder and gave her a bottle; she threw up all over him. He telephoned M to see how long she would be. There were at least this call and another one between the two of them whilst M was out. JH was standing next to M and she heard L screaming down the telephone, sounding very unsettled, when RH had first rung to say she had been sick (§262). The telephone records showed that the first call between M and RH was at 3.08 p.m. (§454).

RH kept the baby in his arms until M and JH arrived back (§456). As recorded by the judge (§263 et seq; and see §370 for M’s account), JH’s evidence was that “[s]he had felt that L had been grumpy all day on the Tuesday, although this was not unusual for L”. Just before she left M’s at about 4.30 p.m., JH heard L screaming as OA bounced her up and down on his lap. The child was distressed and very unsettled, making “a dreadful, gut-wrenching, heart-breaking sound”. M was in the kitchen, RH in the kitchen or upstairs. It was all a bit of a mad rush because the boys were due to attend a birthday party later in the afternoon. JH took L from OA. L fell asleep within 5 minutes while being held and supported.

The timings of events during the later afternoon and early evening are difficult to pin down. M had to get the boys to their party by 4.30 p.m. and took L with her, then returned home. She was there until she went to pick up the boys. There is uncertainty as to precisely when she did that but it was clearly around 6.30 p.m..

ML arrived at the house some time between 4 p.m. and 5 p.m. (§416). M reported that L had been a bit whingey for most of the day. He cuddled her whilst M cooked dinner for him, RH and herself. L was a little bit whingey but she did not appear to be in any pain. Every time he moved or went to put her down, she cried. She fell asleep in his arms but when put down on the settee, woke up and started to cry again. It seems that ML kept L on his lap asleep whilst M went with RH to fetch the boys back from the party (see, for example §§417 and 466).

A relative of the boys’ father visited in the late afternoon, possibly about 5 p.m. to use the computer. He conveyed a picture of an unsettled baby (§231) but he was not clear about events and unable to help the judge (§229).

After returning home with the boys from the party, M agreed to give L a bath. This seems to have been RH’s idea. ML took the baby up to M. M’s account was that when she removed L’s nappy for that purpose, she noted her leg was funny/floppy/not right. RH said that he heard L upstairs after ML had come downstairs (§468) and he viewed her cry as quite a loud and significant cry. M came running downstairs with L saying there was something wrong with her leg. L was “proper screaming, red in the face” (§469 and surrounding paragraphs). She yelled as RH touched her leg; it was weird and he said that they should go to hospital. He heard L scream at the hospital when they moved her leg for the x-ray which was a very high pitched horrible scream but the scream at the house was not the same (§471).

There was evidence about telephone calls made by M that evening at around the time of the bath/journey to hospital. The timings of the calls are not entirely clear. JH recalled that M had telephoned her at 6.42 p.m. to say that there was something wrong with L’s leg (§266). It was put to her that there might have been a second conversation about the leg but she was emphatic that she only remembered one conversation between 6.40 p.m. and 7.00 p.m. from M. The judge recorded at §268 that JH

GM gave evidence that M telephoned her between 6 and 7 p.m., certainly after 6 p.m., sounding “anxious distressed – very distressed – upset, really different, but not panicking” (§281). M said to her, “Mum I think something’s happened to L; I think she’s dislocated her leg in the baby bouncer”. GM told the court that she told M not to speak to her but to get the baby up to the hospital (§282).

M gave an account that she had phoned her mother about having the boys after she discovered that something was not right with the leg (§371) and denied that she telephoned JH at about 6.45 p.m. to say that there was something wrong with L’s leg.

GM joined M, L and RH at the hospital. She held L for three hours. When L was put on a bed to be examined or when GM stirred, L made a noise that GM said she would never forget, a continual, horrendous screaming (§283). She had never heard such a scream from L before (§284). The poorly leg was noticed to be flopping and bouncing as GM walked along the corridor with L and L’s ankles were taped together by a nurse to reduce the movement.

At 8 or 8.30 a.m. on 4 November, by which time L had spent a night in hospital being investigated for the symptoms in her leg and was already in traction, M telephoned F and said that she “might have to take L to hospital as she may have a broken leg” (§215). Not surprisingly, the judge found that that “was a rather strange and, frankly, inaccurate way to put it”.

The judge made certain findings as she went along in her judgment but the main section setting out her conclusions is to be found at the end of it. She arrived at her decision by “[d]rawing all the multifarious strands here together” (§502) as was entirely proper. Her approach to the evidence is perhaps encapsulated at §533 where she said that:

She accepted that the nose injury was innocently caused by RH (§520). She said of RH:

She said there was no evidence that RH had taken cocaine during his relationship with M other than on one occasion before L was born and that the court could not

The judge had heard evidence from a former long term partner of RH’s, EM, about his disposition and behaviour whilst they were together and this plainly formed part of the material upon which she based her conclusions. RH spent four weeks in custody as a result of having assaulted EM. His evidence was that he had not hit her but pushed her (§423). He said that it was at his request that EM had called the police and told them he had hit her. They did this because he wanted to go to prison to assist him in keeping off drugs. EM gave evidence to Judge Cameron who found her “a confident, impressive, truthful witness who gave compelling evidence in an honest and really straightforward manner” (§479) and considered that “[a]long with JH and GM, she gave the most genuine and credible account” (§500). EM described RH as a brilliant father who was not a big drinker but had smoked cannabis which made him chilled, sleepy and relaxed. She said that he was coming off cannabis in 2007 at the time of the push and the police involvement, the circumstances of which she confirmed were essentially as RH described (§483). EM told the judge that she knew RH was not capable of hurting L although she had called him “a hot-headed person, with a vicious tongue sometimes” in an email to M with whom she had become friendly (§490).

The judge concluded (e.g. §502) that the common denominator for all the injuries except the nose injuries was the mother, being careful to check that she had the opportunity to cause the injuries (§521).

She said (§502):

She viewed what happened as (§510)

The existence of injury early on was clearly influential in her thinking in this respect as she said (§513):

The judge gave real consideration to the notion that the child had received some, or all, of the most recent catalogue of injuries before the afternoon of Tuesday 3 November. She said that “the indicators arise during the febrile weekend commencing on the Halloween evening” when L was said to be unsettled (§514). She said:

She also set out an alternative scenario as follows:

She said (§517) that she was unable precisely to time the material event, given the lies told by the mother, supported by ML (and see also §522). She did, however, set out how the child was at various times in order to examine whether it indicated when she had been injured.

She considered the possibility that L had been the subject of “untoward rough handling or dropping or yanking that Sunday afternoon [1 November]” causing M to feel distressed and guilty and therefore to present to GM that evening as “unjustifiably upset” (§522). She set out that the baby noticeably did cry or scream when moved from one surface to another at various times on the Monday and Tuesday and was only comfortable when held and supported (§522). She set out a version of the events of 3 November at §518, emphasising that L was in the car seat most of the day “therefore the thigh supported”, crying when removed from the car at lunch time and screaming abnormally when moved to the bouncer prior to M leaving at 2.50 “perhaps because the thigh is not so supported in the bouncer”. She said (§518):

An important passage in relation to timing is §539 where the judge said:

At §540, she again stressed that she could not pinpoint exactly when the major fracture occurred but, she said, “those occasions are the possibilities”.

The judge listed a number of other matters in the conclusion section of the judgment which we can infer she considered lent support for her conclusion that it was M who was responsible. They included:

M’s “pat and far-fetched explanation” to GM between 6.30 and 7.30 on 3 November that L had dislocated her leg on the baby bouncer, which the judge considered was “very likely a swift fabrication concocted between her and ML when they both saw upstairs that the leg looked worryingly floppy” (§523).

The telephone call to JH at 6.42 p.m. whilst M was waiting for the boys to come out of the party or when they got back after the short car journey (§524).

That L seemed to be nervy, fractious and whingey at home but different elsewhere when handled by loving confident competent people (§525) and that L’s needs did not appear to have been the main focus of M’s loving attention which was instead directed to RH (§526, 527, 530).

M not being keen to bath L on 3 November (§528).

M not responding to L’s scream when she was bounced by OA on the afternoon of 3 November whereas JH did (§528 and 530).

M not suspecting or seeking to blame RH until the actual hearing and continuing in a relationship with him for 10 months, which the judge found was because she knew that she alone was responsible for hurting L (§541 and §350).

Further “strands leading inexorably to M damaging her child” were highlighted at §537 which I need not set out here.

For the most part, all parties other than M supported the judge’s analysis of the evidence and findings of fact. The one exception to that is in relation to the femur fracture which both the local authority and F submitted the judge may have thought occurred earlier than in reality it must have done. The local authority’s position derived from the radiological evidence which (if precise) narrowed the timing of the injury to a 24 hour period commencing with the early evening of Monday 2 November making it unlikely that the femur fracture was caused on the Sunday evening as the judge may have thought. The local authority submitted that whilst the judge may have overlooked that feature of the radiological evidence, it had no significant impact on her overall findings. F’s position was based on the symptoms that would be expected in association with a femur fracture. If this court took the view that the judge had concluded that the femur fracture was likely to have been caused before Tuesday 3 November, then F accepted that that was against the weight of the evidence. However, he too submitted that any such error was not fatal to the integrity of the judgment as a whole and in particular to the judge’s conclusion that M was solely responsible for L’s injuries.

It does appear to be highly unlikely that L had suffered the femur fracture by the time GM saw her on the morning of 3 November given that the evidence leads one to expect that L’s discomfort from that particular injury would have been such that GM, handling L as she did, would inevitably have noticed it and remarked upon it. Accordingly, in so far as the judge thought that there could have been one over-reaction by M on the Monday evening or in the early morning of Tuesday causing all the more recent injuries (§516), it is probably correct to categorise that as against the weight of the evidence. If the judge had gone so far as to make a finding that all the injuries (or at least the femur fracture) were present before GM saw L on Tuesday, such an error may have been capable of undermining her findings overall because it would have pointed her towards M by virtue of the fact that the injuries would have been inflicted when RH was largely not around. However, not surprisingly given the state of the evidence, the judge said a number of times that she was unable to be precise about the timing of the recent injuries and she made no specific findings about it. Her consideration of who was responsible for them was by no means tied to a particular time frame and she set out herself (§517) the alternative that the damage was done during rough handling at bathtime on the Tuesday evening whilst at §539 she appeared to be considering a third possibility, namely that the more recent injuries were inflicted in two stages, the possible timing of which included late Monday/early Tuesday and/or at bath time on Tuesday. No doubt the two stages that she had in mind were the infliction first of the injuries except for the femur fracture (or some of them) leading to the child being out of sorts on the morning of 3 November and secondly of the more acutely painful femur fracture. The medical and other evidence permitted these permutations and I note also that M’s counsel conceded in his skeleton that it was proper for the judge to say at §138 that GM and JH would not necessarily have noticed the less painful tibial injury on the morning of 3 November. I therefore accept the submissions of the local authority and F that even if one of the possibilities as to timing that the judge entertained was not in fact feasible, that did not undermine her conclusions.

Counsel for M very sensibly confined his submissions to what he described as the six better points that he wished to make in support of M’s appeal. They were all designed to establish that RH was a more likely candidate than M to have caused L’s injuries.

The first point that he made concerned the timing of the November injuries and built on the improbability of L having sustained the femur fracture prior to parting company with GM at the end of the morning of 3 November. Counsel submitted that the evidence was such that the judge should have found that RH injured the baby in the period during which he looked after her whilst M was away collecting the boys from school that afternoon of 3 November. He sought to focus attention in particular on the 18 minutes between M leaving the house and the first telephone call between her and RH, during which RH reported that L had been sick. It was during that telephone call that JH, who was standing beside M, could hear L screaming in a way she had not heard before. Counsel relied also on RH’s account of the baby’s eyes rolling during this period (§113 and §460) which Dr Robinson thought a most unusual sign which showed the baby was very distressed. He also pointed out that it was when M and JH returned with the boys that JH heard the dreadful scream from L when she was bounced on OA’s lap (§263). In addition, he relied on RH having said in his December 2010 statement that he only had sole care of L for 30 minutes at most. He submitted, as his second main appeal point, that the judge should have classed this as a deliberate lie by RH told in order to protect himself by reducing his opportunity to cause L’s injuries and therefore indicative of guilt.

The judge’s findings about this period of time presented an obstacle to counsel’s arguments. The judge did not ignore the points counsel made to us but dealt with them in the light of her own assessment of the witnesses, as she was required to do. She was well aware of RH’s lack of candour and the cavalier and unhelpful approach he took in his witness statement when he asserted he had had care of L for no more than 30 minutes. However, at §456 she gave her view that RH’s account of what he did whilst waiting for M to come back from school “certainly does not sound anything like the aftermath of a man who has lost his temper with a young baby, or dropped her” and she said that she did not find that RH had “the resourcefulness or motive to lie about all of this”. She asked pertinently why, if RH had lost control and hurt the baby, he would have wanted M to come home as quickly as possible (§459), acknowledging that panic and fear were possible explanations but setting out her view that RH was frightened at the ferocity of the crying and at his inability to settle L, coupled with disliking the vomit.

I thought that a further possible difficulty with the submission that the femur fracture had occurred at this time was M’s own evidence about the period as recorded by the judge (§348). It was to the effect that L was already screaming when she left the house and also crying when she came back so, as the judge recorded it, “there was nothing weird there”. Indeed, M apparently never felt suspicious that something had happened to L in RH’s care until, having exhausted the possibility of osteogenesis imperfecta as the culprit, she finally blamed him during the hearing (§340). I do appreciate, however, that counsel’s argument was that the reason for M’s failure to spot anything when she returned with the boys from school was that the initial period of 30 to 60 minutes of uncontrollable crying from the injury would have been over. The medical evidence (§109 and 119) was that if L was swaddled, comforted and not moved the pain would lessen and she may fall asleep; in addition, Calpol would reduce the pain. Counsel also submitted that when L was held comfortably with her leg well supported she was alright, as was seen between bath time that evening and the pain being aggravated by procedures at the hospital.

Counsel’s third point related to the nose injury undoubtedly caused to L by RH. He argued that it was wrong of the judge to find that RH’s accidental explanation for this was adequate. In fact, he said, the judge should have realised that the injury on the other nostril was not explained in this way and should have taken more account of Dr Robinson’s evidence that the nose was a most unusual place and uncommon for an injury and of the Maguire study about the infrequency of accidental injury in non-mobile babies. If the judge had found, as it was submitted that she should have done, that this was a non-accidental injury caused by RH, then that would have indicated that the other injuries were also caused by him, given the unlikelihood of there being two different perpetrators.

The state of affairs with regard to the possible injury on the other side of the nose was not very satisfactory. The thought that there was further injury only emerged during the last week of the trial when enlarged photographs were produced, too late for the doctors to see them. It is not even clear that the judge accepted there was an injury on the other side because, as I said above at paragraph 11, at one point in the judgment she said that the photographs indicated “possible” damage to the other nostril (§82). Nevertheless, there can be no doubt that she had the evidence of it in mind when evaluating RH’s explanation. She also had Dr Robinson’s evidence in mind (see §81). It did not bind her to reach a conclusion that the nose injury was non-accidental. Appropriately, she took into account the untruths circulating about the injury and treated RH’s explanation with caution but ultimately accepted it. It has not been demonstrated that this was a conclusion that was not open to her on the evidence.

Counsel’s fourth point concerned the judge’s acceptance of EM as an impressive and honest witness whose evidence contributed to the picture that the judge built up of RH as peace-loving and unlikely to be cruel or violent to the baby. Counsel submitted that EM was not honest. She was party to the events of 2007 when RH was prosecuted for assault and even on her own account she made the situation out to the police to be more serious so as to enable RH to be treated for his cannabis use. The judge knew this; she set it out at §485. Counsel suggested that EM also lied to the court when she denied that RH had taken cocaine once their son was born. He invited our attention to the transcript of RH’s evidence in cross-examination which was that he had a cocaine addiction which counsel put to him was “at a time when you had met [EM] in a relationship lasting seven years” to which he replied “Yes”. What the judge recorded EM as saying was that once [their son] was born he never took cocaine again to her knowledge (§487). It will be evident that a comparison of these two pieces of evidence does not demonstrate that EM was necessarily lying. It was also submitted that it was significant that EM had denied there had been any incidents of domestic abuse, either physical or verbal, save for the incident at the end of 2007 whereas RH had told the social worker that their son had witnessed a lot of domestic violence (by which, RH said in court, he meant “a lot of arguing” (G8)). Given the flexibility of the description “domestic violence” these days, it is difficult to draw from this feature anything fatal to the judge’s acceptance of EM as a truthful witness either.

The fifth point advanced on behalf of M concerned RH’s demonstrated untruthfulness in connection with his use of drugs. His evidence was that he had not used cannabis whilst with M and witnesses were cross-examined on that basis. It was only some way into the case that he admitted that that was a lie although he continued to maintain that his cannabis use was less than others said it was. The judge had RH’s position firmly in mind throughout her judgment, for example referring at §14 to his “frank lies about his taking of drugs”, at §253 to the fact that he was clearly lying about being abstinent from drugs for two years, and reverting to the topic again in her conclusions at §538. Counsel complained that this recognition of RH’s lies was not sufficient in that it did not take into account that RH continued to lie about the extent of his drug use and also that the judge failed to recognise the significance of his drug taking. He pointed out that RH had told a social worker in the course of a viability assessment at the end of 2007/beginning of 2008 that cannabis affected his moods and that he had snapped and begun smashing up the house at the time of the purported punch to EM which counsel submitted was of significance in indicating propensity to cause injuries to L. Counsel also reminded us that there had been other examples of unpleasant behaviour on the part of RH such as his approach to GM on Halloween night.

RH’s unpleasant behaviour was recorded in the judgment just as were his lies about drug use and there is nothing to suggest that either was ignored. Even if the judgment itself were to leave any doubt as to the attention paid by the judge to these features (which it does not), one has only to look at the transcript which shows that she herself asked RH a considerable number of very searching questions on diverse aspects of his behaviour, including in relation to his drug taking and his lies about it (see G126 to G139).

Counsel’s final point on M’s behalf was that the judge came to a premature conclusion about who was responsible before considering the arguments about it. This was based upon the judge’s §350 in which she said that from matters such as M’s lack of suspicion on RH and her absence of soul searching, it could be concluded, on the balance of probabilities, that M knew she herself had injured L. It was submitted that not only was this premature, it was also unfounded because M’s approach was explicable by her conviction that osteogenesis imperfecta was to blame.

I am not persuaded by this submission. Whilst a judgment inevitably has to be set out in a linear fashion, the thought processes that underlie it are not linear. The judge would not have begun upon a long judgment such as this one, given at the conclusion of a lengthy hearing, without a very considerable amount of preparation, preparation which is evident in the detail included in the final product and to which the judge referred explicitly when refusing permission to appeal. It was only relatively shortly before §350 (at §294) that she herself observed that her picture of M had been built up from numerous strands and hallmarks as is indeed apparent from reading her judgment as a whole. As well as the matters she identified expressly in §§349 and 350, all those strands no doubt also properly supported her conclusion about the reason for M’s lack of suspicions about others. I do not accept that the mere position of §350 within the judgment tells us anything significant about the path of the judge’s reasoning or gives rise to any suspicion that it was flawed. What matters is the integrity of the reasoning process in relation to her conclusion and as a whole and that is not undermined in this case by this feature of the geography of the judgment.

Having considered all the points made on behalf of M, both separately and together, I have not been persuaded that the judge fell into error in the conclusions that she drew from the evidence before her or that she failed to explain them sufficiently. There was, in fact, no shortage of evidence upon which the judge was entitled to conclude that M was responsible for L’s injuries. Relevant features were identified in the body of the judgment and in the judge’s conclusions, such as M’s personality and circumstances, the way in which she focussed her attention on RH rather than L, her lying and obfuscating, the “febrile weekend” commencing on Halloween when M was in charge and the child had been heard crying, M’s explanation to GM that L had dislocated her leg in the baby bouncer, the fact that it was not M who was really disturbed by L’s crying on 3 November but JH, the matters set out in §537, and so on.

Counsel for M rightly acknowledged the difficulty in appealing successfully against a finding of fact made by a judge who had heard evidence for many days. There are bound to be matters to which reference cannot be made in the judgment, either because they are hard to describe (such as a sense gained from the way in which a witness gives evidence) or because it is impossible to set out every last detail in a judgment of manageable size. Furthermore, it is not at all uncommon for loose ends to remain at the conclusion of such a hearing and for there to be unresolved contradictions in the evidence. One obvious reason for this is that inevitably not everyone has told the judge the truth.

Judge Cameron reflected the difficulties in this case when she remarked (§17) that “notwithstanding that wealth of evidence and exploration, this remains a difficult detective story, ‘a conundrum’, as one Counsel rightly put it”. It is the judge who is in a position to assemble the jigsaw of facts and impressions in order to arrive at findings on the balance of probability. As it has not been demonstrated that Judge Cameron fell into error in so doing here, I would dismiss the appeal.

I agree.