image of 4 Paper Buildings logoCoram ChambersFamily Law Week Email SubscriptionGarden CourtDNA Legal1 Garden CourtHarcourt Chamberssite by Zehuti

Making black and white from a palette of shades of grey: the binary nature of fact finding hearings after AA v NA & Ors

Byron James considers the implications of Mostyn J's approach to fact finding.

Byron James, barrister, 14 Gray's Inn Square

Byron James, barrister, 14 Gray's Inn Square

Underpinning most children practitioners' contentment with their daily life is the comforting knowledge that they will never have to sit in a maths class again; in fact numbers are often only relevant in so far as how one divides seven, and even then, precision is a guideline rather than a rule. What to make then of the maths lesson given by Mostyn J's during AA v NA & Ors [2010] EWHC 1282? This not being just any maths lesson, but an advanced one; dealing with the binomial theory of probability and its application to fact finding hearings.

Fact finding hearings have their troubles, of that there can be no doubt. In an attempt to stem the flow of unnecessary split hearings being listed, the President in May this year proffered a gentle reminder to courts and practitioners that separate fact finding hearings are not the norm. What of the procedures to counter the flow: Scott schedules exist as embarrassed creatures, skulking around, wishing they were thinner; statements allege, allege, allege, with more heat than light – and corroborating evidence? Well, that passed into extinction a long time ago. So often, one comes to a fact finding involving allegations that bear little relevance to the child and, even if they do, can hardly ever be proved save for the court 'preferring the evidence of one party over the other'; this last expression being a depressingly repeated mantra by the judiciary that rarely inspires with confidence in the correct outcome. The stigma attached to someone who is held to have been violent is significant: it is a warning sign to those close, both presently and prospectively, and also to professionals; the more freely it is attached without genuine basis, the more diluted it becomes.

The lack of corroborating evidence is, however, both understandable and frustrating. Like rape, domestic violence is rarely done in public; the lack of any ostensible indicators of what has happened does not necessarily suggest that it did not happen. Can it be fair, therefore, to penalise someone for making an allegation when they are unable to adduce evidence of an event not witnessed by others? Conversely, however, there is a difficulty for a man countering such allegations: angry that he is there, accused of such things, perhaps having been denied contact for some time, against a woman who comes with nothing but her tears and shaky voice to support her case. Is it really enough to rely upon a perception of those two, and their oral evidence, to get to the truth: one appears angry; the other a victim. On that alone, the result seems obvious, especially when giving the benefit of the doubt to wanting to ensure the child's safety.

The burden of proof deployed at the actual fact finding hearing is therefore all important. It should feature in every opening and closing of a case and was set out by Mostyn J in AA v NA & Ors [2010] EWHC 1282. The real blame for the maths lesson lies with Lord Hoffman, who started it in Re B (Minors) [2008] UKHL 35 at para 2:

"If a legal rule requires a fact to be proved (a "fact in issue"), a judge or jury must decide whether or not it happened. There is no room for a finding that it might have happened. The law operates a binary system in which the only values are 0 and 1. The fact either happened or it did not. If the tribunal is left in doubt, the doubt is resolved by a rule that one party or the other carries the burden of proof. If the party who bears the burden of proof fails to discharge it, a value of 0 is returned and the fact is treated as not having happened. If he does discharge it, a value of 1 is returned and the fact is treated as having happened."

The Judge went onto explain that this was referable to the binomial theory of probability, the term for an event that can be broken down into two probabilities (expressed as 'p' and 'q') where the two terms are complimentary, so: q = 1 ? p.  To explain this, he moved onto coin tossing:

"If you have a double headed penny then the probability (P) of tossing a head is absolute certainty. This is expressed as P = 1. The probability of not tossing a head is zero. This is expressed as P = 0 or Q = 1. For a normal penny the probability is tossing a head is P = 0.5, and the probability of not tossing a head is Q = 0.5. The probability of rolling a six on a die is one in 6: P = 0.1666 and Q = 0.8333."

The comparison of judicial determination with tossing a coin is perhaps unfortunate. Instead of a comparable, or even directly referable, link in the method of determination, rather, it was ground laying for the mathematical interpretation of the civil standard of proof:

"In civil proceedings the standard of proof is stated to be 'on the balance of probabilities'. Expressed mathematically this is P > 0.5. If a court were to find on the evidence that P = 0.5 (i.e. that the occurrence of the event was as likely as not) then the standard would not be met, as the respondent to the allegation that is sought to be proved is always entitled to the benefit of the doubt.

"Thus, it is clear that in all civil proceedings P cannot be set higher than a scintilla above 0.5."

The applicability of mathematical principles here could be thought an attempt to find some form of certainty in an area where subjectivity, perception and assumption usually reign. It is of course important to get the starting point right. If a judge misdirects herself on what is required to find something as fact that is problematic. Of course, the extent to which a mathematical principle is ever relevant to something that is more art than science is clearly questionable.  The important point here on directing a judge during a fact finding is the difference between 'as likely as not' and 'more likely than not'. This was the error found by the Mostyn J of the direction at first instance:

"The burden of proof is upon the party who makes the allegations. It is not reversed, that is it is not for the other party to establish the allegations are not made out. It is open to the court to make the following findings:

1. That an allegation is true on the balance of probability
2. That an allegation is false."

With the 'correct' version being:

"The burden of proof is upon the party who makes the allegations. It is not reversed, that is it is not for the other party to establish the allegations are not made out. It is open to the court to make the following findings on the balance of probability:

1. That an allegation is true
2. That an allegation is false."

The error was (as the judge would approve) in mathematical form: the district judge's formula provided for P>0.5 ¬ P=0; whereas the formula should simply be P>0.5 ¬ P?0.5.

Going forward then, we have the nomenclature of the binary system of fact finding. Either an allegation is proved or it is not; a finding that an event is as likely as not to have happened is not the same as a finding that an event is more likely than not to have happened. In fact, it goes further: a judgment should purport only to have two values in respect of an allegation, a positive or a negative, requiring the court to say whether something did or did not happen. Consider again, set out above, the lack of evidence often involved in these hearings: is the Court being asked to make black and white from a palette of only shades of grey? When discussing the merit of the 80 allegations made by the father against the mother, Mostyn J stated:

"Many of these were wholly trivial, unkind and unnecessary and were designed to destabilise her. They were all found to be false, by which the District Judge surely meant that they had not been proved to the requisite standard. For none of these was P > 0.5. It might be said that there is no difference between setting them aside and leaving the findings intact as in each case the charges simply remain unproven. But given the binary system of fact-finding explained by Lord Hoffmann the effect of the judgment is to return for each of these allegations a value of zero, and a finding that they did not happen."

In a finding then that a particular allegation is 'not proved', Hoffman LJ/Mostyn J's binary system of fact finding should surely lead the court further, as in the last line of the above quote, to state not simply that the allegation was not proved but rather that 'it did not happen'. This creates the odd scenario where if a court has found that something was as likely to have happened as not, it must consequently find that it did not happen. A tautological nightmare. However, consider further the relevance of the fact finding hearing itself to the proceedings generally. Why does it exist if it is not to consider the parties differently after it? In the bigger picture of the litigation, the court must find that either one parent is violent and/or one parent is a liar, with the latter having deprived or changed contact as a result of that lie, a form of harm toward the child. Should practitioners now seek to make more of an allegation not proved, to use the binary nature of fact finding hearings to insist upon the active 'did not happen' rather than the passive 'was not proved' and, consequently, draw suitably negative inferences in respect of the parent making (judicially found) false allegations?

Clearly, like a world cup hosted in Africa, there is massive scope here for unfairness and exploitation of the unfortunate.  Either fact finding hearings are determinative of something useful or they are not; if they are not, they should not be undertaken. The supposed useful purpose of them is to provide findings of fact in relation to incidents that directly harm the child, including domestic violence by one parent on another, although of course, there are various forms of harm. If the binary form of fact finding is adopted, or is indeed a correct statement of the status quo, then surely it should be more than just the extreme cases of fantasy in the Scott schedule that should be examined. If a court finds that an allegation did not happen, should it not have a duty to examine the circumstances in which the allegation has affected, for example, contact and the harm caused by a resulting deprivation of a parental relationship? Should there be a finding that the alleging parent has caused the child harm when the allegation is found not to have happened? Can this be just? Can one really ignore the evidential difficulties, the shades of grey reality determining fine line decision making, to pretend that such things are capable of correct judicial stamping of 'did' or 'did not' happen? 

The reality is that a finding of 'not proved' rather than 'did not happen' is actually required in fact finding hearings. Otherwise, why should the binary nature only be applied to the fact finding, why not also to the wrong doing: whatever happens, one or both parties will come out badly. The possibility for unfairness in such circumstances is really quite great.

Maths then sends a shiver down the spine for reasons other than the classroom: the mental scars left from trying to understand the 'Fourier series' still remain, but Mostyn J's maths lesson gives even greater cause for concern now.  

14 Gray's Inn Square