RSS Feed

Tag Archives: balance of probabilities

Mostyn J gets dissed by Court of Appeal despite not being the Judge in the case being appealed


Re A Children 2018

Long-time readers will have been enjoying the regular frank exchange of views and pleasantries between Mostyn J and the Court of Appeal, but this is a new one.  The Court of Appeal in this case overturned a Judge who had been following Mostyn J’s guidance in a High Court case and therefore had the opportunity to say that Mostyn J was wrong as a sideswipe.


Did they resist this?

Reader, they did not.


  1. In A County Council v M & F, upon which the judge relied, Mostyn J having set out passages from Re B (and Baroness Hale’s confirmation of Re B found in Re S-B [2010] 1 All ER 705, SC,) went on:
    1. “16. Thus the law sets a simple probability standard of 51/49, but the more serious or improbable the allegation the greater the need, generally speaking, for evidential “cogency”. In AA v NA and Others [2010] 2 FLR 1173, FD, I attempted to summarise these principles at para 24:

17. Thus, it is clear that in all civil proceedings P cannot be set higher than a scintilla above 0.5. The various judicial statements that a more serious charge requires more clear evidence is not an elevation of P > 0.5. The requirement of evidential clarity is quite distinct from an elevation of the probability standard. Were it otherwise, and, say, an allegation of rape or murder of a child made in civil proceedings required P to be set at > 0.6 then one could end up in the position where a court considered that P in such a case was, say 0.51 but still had to find that it did not happen; when, as a matter of probability, is was more likely that not that it did. This would be absurd and perverse. P must always be set at > 0.5 in civil proceedings, but subject to the proviso that the more serious the allegation so the evidence must be clearer.”

  1. With the greatest respect to the erudition of Mostyn J’s arithmetical approach to the application of the ‘simple balance of probabilities’, I do not agree that it represents the appropriate approach, and it seems to me that this passage had, in part, led the judge to decide that, in order to determine whether the local authority had discharged the burden of proof to the necessary standard, he had to adopt the same approach. As a consequence, the judge mistakenly attached a percentage to each of the possibilities and thereafter, added together the percentages which he attributed to an innocent explanation and before concluding that, only if the resulting sum was 49% or less, could the court make a finding of inflicted injury


Perhaps envisaging a ‘says who?’ response to their very polite (if you are not a lawyer) ground and pound of Mostyn J, the Court of Appeal pre-empt this


  1. In A County Council v M & F Mostyn J had drawn on the shipping case of The Popi M ( Rhesa Shipping Co.S.A. v Edmunds, Rhesa Shipping Co.SA v Fenton Insurance Co Ltd) [1985] 1 WLR 948 HL,(Popi M) as an example of ” the burden of proof coming to the rescue”[18]. Lord Brandon, in his celebrated passage in Popi M, in declining to apply the dictum of Sherlock Holmes to the effect that “when you have eliminated the impossible, whatever remains, however improbable, must be the truth” said:
    1. “The first reason is one which I have already sought to emphasise as being of great importance, namely, that the judge is not bound always to make a finding one way or the other with regard to the facts averred by the parties. He has open to him the third alternative of saying that the party on whom the burden of proof lies in relation to any averment made by him has failed to discharge that burden. No judge likes to decide cases on burden of proof if he can legitimately avoid having to do so. There are cases, however, in which, owing to the unsatisfactory state of the evidence or otherwise, deciding on the burden of proof is the only just course for him to take.

The second reason is that the dictum can only apply when all relevant facts are known, so that all possible explanations, except a single extremely improbable one, can properly be eliminated

  1. Recently (and after A County Council v M&F), in Nulty Deceased v Milton Keynes Borough Council [2013] EWCA Civ 15, [2013] 1 WLR 1183 Lord Justice Toulson (as he then was) considered the use of an arithmetical approach to the standard of proof. Having first considered Popi M he went on:
    1. “33. Lord Brandon concluded, at 957, that the judge ought to have found simply that the ship owners’ case was not proved.

34. A case based on circumstantial evidence depends for its cogency on the combination of relevant circumstances and the likelihood or unlikelihood of coincidence. A party advancing it argues that the circumstances can only or most probably be accounted for by the explanation which it suggests. Consideration of such a case necessarily involves looking at the whole picture, including what gaps there are in the evidence, whether the individual factors relied upon are in themselves properly established, what factors may point away from the suggested explanation and what other explanation might fit the circumstances. As Lord Mance observed in Datec Electronics Holdings Limited v UPS limited [2007] UKHL 23, [2007] 1 WLR 1325, at 48 and 50, there is an inherent risk that a systematic consideration of the possibilities could become a process of elimination “leading to no more than a conclusion regarding the least unlikely cause of loss”, which was the fault identified in The Popi M. So at the end of any such systematic analysis, the court has to stand back and ask itself the ultimate question whether it is satisfied that the suggested explanation is more likely than not to be true. The elimination of other possibilities as more implausible may well lead to that conclusion, but that will be a conclusion of fact: there is no rule of law that it must do so. I do not read any of the statements in any of the other authorities to which we were referred as intending to suggest otherwise.

35. The civil “balance of probability” test means no less and no more than that the court must be satisfied on rational and objective grounds that the case for believing that the suggested means of causation occurred is stronger than the case for not so believing. In the USA the usual formulation of this standard is a “preponderance of the evidence”. In the British Commonwealth the generally favoured term is a “balance of probability”. They mean the same. Sometimes the “balance of probability” standard is expressed mathematically as “50 + % probability”, but this can carry with it a danger of pseudo-mathematics, as the argument in this case demonstrated. When judging whether a case for believing that an event was caused in a particular way is stronger than the case for not so believing, the process is not scientific (although it may obviously include evaluation of scientific evidence) and to express the probability of some event having happened in percentage terms is illusory.

36. Mr Rigney submitted that balance of probability means a probability greater than 50%. If there is a closed list of possibilities, and if one possibility is more likely than the other, by definition that has a greater probability than 50%. If there is a closed list of more than two possibilities, the court should ascribe a probability factor to them individually in order to determine whether one had a probability figure greater than 50%.

37. I would reject that approach. It is not only over-formulaic but it is intrinsically unsound. The chances of something happening in the future may be expressed in terms of percentage. Epidemiological evidence may enable doctors to say that on average smokers increase their risk of lung cancer by X%. But you cannot properly say that there is a 25 per cent chance that something has happened: Hotson v East Berkshire Health Authority [1987] AC 750. Either it has or it has not. In deciding a question of past fact the court will, of course, give the answer which it believes is more likely to be (more probably) the right answer than the wrong answer, but it arrives at its conclusion by considering on an overall assessment of the evidence (i.e. on a preponderance of the evidence) whether the case for believing that the suggested event happened is more compelling than the case for not reaching that belief (which is not necessarily the same as believing positively that it did not happen)”.

  1. I accept that there may occasionally be cases where, at the conclusion of the evidence and submissions, the court will ultimately say that the local authority has not discharged the burden of proof to the requisite standard and thus decline to make the findings. That this is the case goes hand in hand with the well-established law that suspicion, or even strong suspicion, is not enough to discharge the burden of proof. The court must look at each possibility, both individually and together, factoring in all the evidence available including the medical evidence before deciding whether the “fact in issue more probably occurred than not” (Re B: Lord Hoffman).
  2. In my judgment what one draws from Popi M and Nulty Deceased is that:
  3. i) Judges will decide a case on the burden of proof alone only when driven to it and where no other course is open to him given the unsatisfactory state of the evidence.

ii) Consideration of such a case necessarily involves looking at the whole picture, including what gaps there are in the evidence, whether the individual factors relied upon are in themselves properly established, what factors may point away from the suggested explanation and what other explanation might fit the circumstances.

iii) The court arrives at its conclusion by considering whether on an overall assessment of the evidence (i.e. on a preponderance of the evidence) the case for believing that the suggested event happened is more compelling than the case for not reaching that belief (which is not necessarily the same as believing positively that it did not happen) and not by reference to percentage possibilities or probabilities.

  1. In my judgment the judge fell into error, not only by the use of a “pseudo- mathematical” approach to the burden of proof, but in any event, he allowed the ‘burden of proof to come to [his] rescue’ prematurely.


I’m sure that Mostyn J is delighted by the dismissal of his P>0.5 formulation as ‘pseudo-mathematical’


The case they were talking about is one I wrote about here


but for my part, the more troubling one, where the mathematics (or pseudo-mathematics) applied to the balance of probabilities directly affect the outcome is here  (three years later, building on Re M and F  and building on the Popi shipping law case but overlooking the Nulty civil negligence about a fire and electrical engineering  law case)


Anyway, the soup and nuts of both of them is that Mostyn J looked at a variety of explanations, malign and benign for incident X and then ascribed percentages to them, and saying whilst the malign explanation might be more likely than not than any individual benign explanation, he was instead totalling up the chance he had ascribed to each of the benign explanations and deciding that he could not say that the chance of malign explanation was higher than all of the possible benign explanations added together.  So what he was doing was saying  ‘There are 3 explanations. I think that the most likely of those three is that mother did this.  But if I ascribe percentage possibilities to each option, I might still decide that the two alternative explanations add up to more than 50%, so I’m not able to say that mother did this’


Anyway, the Court of Appeal say that the Court should not get into such esoteric exercises and simply say that on the balance of probabilities what do they say is the more likely than not explanation for event X.  Which is good news for anyone who doesn’t want to take a course in probability theory.


This case is desperately sad, even by care proceedings standards  – a ten year old girl is found dead. The police assume accidental strangulation by falling off a bunk and getting trapped in decorative netting. Poppi Worthington style errors are made in the investigation, and then evidence comes to light suggesting that the ten year old had been sexually assaulted (there is talk of DNA being present in intimate areas) and concerns then arise that the ten year old either hung herself intentionally or was killed  (deliberately or unintentionally as part of choking).  That obviously had massive implications for the other five children of the family.

At final hearing, the Judge concluded that the evidence that the girl was sexually assaulted was made out, but he could not say who perpetrated the assault  (there’s some odd wording about why the LA were refused their request to call the police officer who analysed the DNA samples) and whether it might be member of extended family or an intruder.  The Judge found that despite some conflicting expert evidence about causation of the death  (the medical research is that accidental strangulation happens rarely and to much much younger children) he was not able to make a finding that the malign explanation outweighed each of the possible benign explanations. Threshold was not met, the other five children went home.

The Court of Appeal concluded that


  1. In my judgment the judge fell into error, not only by the use of a “pseudo- mathematical” approach to the burden of proof, but in any event, he allowed the ‘burden of proof to come to [his] rescue’ prematurely.
  2. In my judgment the judge had failed to look at the whole picture. Not only did he fail to marry up the fact that S sustained two sets of injuries (one of which was fatal) but the judge, faced with the incontrovertible evidence in relation to the genital injuries, carried out no analysis of the available evidence in order to see whether an accident (for example) was a likely cause. Whilst in other circumstances I might have identified, or highlighted by way of example, certain evidence which I believe merited consideration by the judge, given my view that the appeal must be allowed and the matter remitted for rehearing, it would not be appropriate for me to comment further.
  3. Only if, having carried out such a comprehensive review of the evidence, a judge remains unable to make findings of fact as to causation, can he or she be thrown onto the burden of proof as the determinative element.
  4. In my judgment, in this most difficult of cases and in the most trying of circumstances, the judge failed to carry out such an analysis before relying on the burden of proof. This, when coupled with the erroneous conclusions of the judge in respect of the genital injuries and his failure to give those injuries any weight when considering whether S died as a consequence of an inflicted injury, must, in my judgment, lead to the appeal being allowed and the order set aside.
  5. I have considered with a deal of anxiety whether the case should be remitted given the lapse of time and that the family are reunited. I have however come to the unequivocal conclusion that it must. If S was killed other than by accident or suicide, it happened in that household and no one has any idea how or in what circumstances it came about. This is not a case, tragic and serious though that would be, where a child may have been shaken in an understandable momentary loss of self-control by an exhausted parent. This was a 10 year old child, and if it was the case that her death was caused by some unknown person strangling her with a ligature, the risk and child protection issues in respect of her surviving sister and brothers cannot be over stated. Traumatic though a fresh trial would be, it cannot be viewed as other than a proportionate outcome if, as they say is their intention, the local authority pursues the case.


That’s obviously a dreadful state of affairs either way.  Either something awful and malicious happened to this ten year old, in which case children were wrongly returned to the care of the parents  OR it didn’t, and having secured the return of their five surviving children having been under awful suspicion the parents have to go through it all again.  That’s unbearable however it turns out.