What do we mean by proving something ‘on the balance of probabilities’ ?


How and why do we have two different standards of proof in civil and criminal proceedings?

‘The balance of probabilities’ is the standard of proof used in all civil court proceedings, so includes care proceedings.

The other standard of proof we use is the one in criminal cases which is ‘beyond a reasonable doubt’ or that the jury must be ‘sure’ a crime was committed, which is a higher standard due to the very serious consequences that can follow a criminal conviction, such as loss of liberty.

Its often difficult for non lawyers to understand that there are two separate systems which work on two different ‘standards of proof’ and many parents struggle to accept that the lower civil standard of proof is used to make findings about parents which are often just as serious as a decision to send someone to prison. For example, the family court gets to decide whether a child is removed permanently from the parents’ care. This struggle becomes even more acute around the issue of removing children at risk of future emotional harm – i.e. for something that hasn’t happened yet, but the court thinks will.

I agree that if the State takes your child away, that is a very serious and significant interference in the family life of both parent and child. But if children are left in dangerous situations, that also puts them at serious risk of being hurt, either emotionally or physically. Therefore, it is a deliberate decision to use balance of probabilities in family cases because we could not always prove children were at risk using such a high standard of proof, or it may take much longer to find and gather together the evidence to have a court hearing, leaving the children at risk of harm while this is done.

Baker J discussed the issue of the burden and standard of proof in 2013:

In English law, the House of Lords has now concluded definitively that in order to determine whether an event has happened it has to be proved by the person making the allegation on the simple balance of probabilities. Where the law establishes a threshold based on likelihood, for example that a child is likely to suffer significant harm as a result of the care he or she would be likely to receive not being what it would be reasonable for a parent to give, the House of Lords has also concluded that such a likelihood, meaning a real possibility, can only be established on the basis of established facts proved on a balance of probabilities.

There are those who considered that to require the proof of past harm was a misreading of the intention of Parliament, and that a system devoted to child protection should not imposed such a high hurdle. It was argued, and in some quarters is still argued, that since we would not insist on proof before protecting our own children from risk, we should adopt the same cautious approach when protecting other, more vulnerable children. The House of Lords has of course firmly rejected that approach, which of course would at one extreme involve removing children from their parents on the basis of mere suspicion.

However, it is clear that certainly in ‘finely balanced cases’ that a finding can often be very difficult for parents to accept, particularly as any finding then becomes ‘the truth’ unless and until it is challenged. Our current system of fact finding permits only two outcomes (the ‘binary system’) – either it did or it didn’t happen.

Application of the standard of proof in care proceedings .

In care proceedings, the Judge has to be satisfied that the evidence to show that your child is suffering or is at risk of suffering significant harm has been proved on the balance of probabilities. This phrase has been explained to mean ‘more likely than not, or ‘ ‘more than 50/50’.

For a while, the courts did approach the standard of proof for serious allegations in  family cases as being similar to the standard in criminal cases, as it was felt that such serious allegations with such serious consequences required a high level of proof.

The courts however from 2004 onwards decided to move away from this approach and confirmed it by a decision in the House of Lords in 2008  (The House of Lords is now called the Supreme Court).

Baronness Hale said at paragraph 69 of her judgment:

There are some proceedings, though civil in form, whose nature is such that it is appropriate to apply the criminal standard of proof. Divorce proceedings in the olden days of the matrimonial “offence” may have been another example (see Bater v Bater [1951] P 35). But care proceedings are not of that nature. They are not there to punish or to deter anyone. The consequences of breaking a care order are not penal. Care proceedings are there to protect a child from harm. The consequences for the child of getting it wrong are equally serious either way.

Baronness Hale stated ‘loud and clear’ that the standard of proof in care proceedings is the simple balance of probabilities, neither more nor less.

Neither the seriousness of the allegation nor the seriousness of the consequences should make any difference to the standard of proof to be applied in determining the facts. The inherent probabilities are simply something to be taken into account, where relevant, in deciding where the truth lies. […] It may be unlikely that any person looking after a baby would take him by the wrist and swing him against the wall, causing multiple fractures and other injuries. But once the evidence is clear that that is indeed what has happened to the child, it ceases to be improbable. Some-one looking after the child at the relevant time must have done it. The inherent improbability of the event has no relevance to deciding who that was. The simple balance of probabilities test should be applied.

The dangers of ‘pseudo maths’ to determine the balance of probabilities

The court examined this in the tragic case of A (A child) 2018.  In November 2016 S, aged only 10 years, was found dead in her bedroom. Initially it was thought this was a tragic accident – that she had become entangled in decorative netting around her bed and injuries to her neck caused her death. There were problems at the outset in the police gathering evidence at the scene.

However it was then discovered she had injuries to her genitals, probably inflicted about 12 hours before she died and the view of the experts was that this made it more likely that the neck injuries were deliberately caused, probably in some sexually motivated homicide.

However at the first hearing the LA failed to persuade the Judge to make a finding on the balance of probabilities that both sets of injuries were deliberately inflicted. The LA appealed successfully to the Court of Appeal who criticised the first instance Judge for failing to look at the wide canvass of the evidence and to consider neck and genital injuries together, and also for attempting to apply ‘pseudo maths’ to the analysis of the balance of probabilities, in particular by identifying percentages for each possibility and adding them up. As the Judge could not reach ’51’ by this method, he concluded the LA had not met the standard of proof.

The Judge relied upon the decision of Mostyn J in A County Council v M & F [2011] EWHC 1804 (Fam), [2012] 2 FLR 939 as support for his contention that, where there are rival hypotheses, the judge is not bound to make a finding and that as a consequence the burden of proof is the only course to take. The judge went further saying at para 98 of his judgment that:

“98. I find that this is one of those unusual cases where the burden of proof comes to the judge’s rescue. Aggregating, as I must, the probability of suicide together with the probability of accident, I find that the aggregate of these two is more than 50 per cent. Doing the best that I can, I find that the possibility of suicide is about 10 per cent, and the possibility of accident and a perpetrated act are about 45 per cent each. It would be wrong for anyone to regard these figures as in any way accurate, for of course they are not. They persuade me, however, that the local authority has not discharged the burden of proof which is upon it. I am not satisfied, on the balance of probabilities, that this was a perpetrated act, albeit that I recognise that it is one of three possibilities. On the basis, however, that I do not discard the least probable and then allow a competition between the other two options, but that I should look at each of the alternative possibilities and aggregate them together, I am quite satisfied that the burden of proof in this case is not discharged. Accordingly, I do not find that the local authority’s case is proved in respect of any of the contested issues. By application of the binary principle, it is the finding of this court that neither the father, the mother, M or N are responsible for a sexual assault on L and nor are they responsible for her death. Accordingly, threshold is not met.”

The Court of Appeal firmly rejected this approach at para 51 of their judgment and emphasised that the starting point remains the test articulated in Re B (Minors) 2008 2 WLR 1 HL, discussed above:

“70. My Lords, for that reason I would go further and announce loud and clear that the standard of proof in finding the facts necessary to establish the threshold under section 31(2) or the welfare considerations in section 1 of the 1989 Act is the simple balance of probabilities, neither more nor less. Neither the seriousness of the allegation nor the seriousness of the consequences should make any difference to the standard of proof to be applied in determining the facts. The inherent probabilities are simply something to be taken into account, where relevant, in deciding where the truth lies.


It is the binary nature of our fact finding system that makes the ‘balance of probabilities’ sometimes an uneasy place to be. If the court finds something happened then that is ‘the truth’ that all must abide by for all future dealings with the family. If the court finds something didn’t happen then it simply never did. I am not alone in feeling uneasy about the consequences of this from findings made on the lower standard of proof, particularly when the fact finding exercise is offered as ‘the child’s right’ to know ‘the truth’ about what happened – See comment from the Justice Gap and  The Transparency Project in ‘further reading’ below.

However, what is clear is that the Judge must meticulously examine a broad canvass of evidence and not be swayed by any ‘pseudo mathematics’.  while I can sympathise with a Judge who attempts to impose some clarity on the analytical process by affording a ‘value’ to the options of suicide, accident, or deliberate killing, this is now clearly identified and rejected as unlawful.


Further Reading

  • There is an interesting article about the importance of ‘beyond reasonable doubt’ by BarristerBlogger;
  • Here is a useful  article by Simon Goddard which discusses in more detail the standard of proof generally, and with particular reference to cases involving suspected non – accidental injury.
  • We discuss how to get the best evidence to make the right decisions for children in our post ‘Achieving Best Evidence in Children Act cases’.
  • There is concern that the ‘balance of probabilities’ standard is structurally unfair – can a ‘fact’ really be found on 51% certainty? See this article from The Justice Gap, commenting on the tragic case of Poppi Worthington.
  • You can’t handle the truth – The Transparency Project, raises unease that more weight is being put on the balance of probabilities than it can reasonably carry.
  • Lucy Reed discusses on Pink Tape her unease that barristers facing a disciplinary charge benefit from ‘beyond reasonable doubt’ but parents will lose their children on the lesser civil standard.




36 thoughts on “What do we mean by proving something ‘on the balance of probabilities’ ?

  1. C

    “So, in care proceedings, the Judge has to be satisfied that the evidence to show that your child is suffering or is at risk of suffering significant harm has been proved on the balance of probabilities. This phrase has been explained to mean ‘more likely than not, or ‘ ‘more than 50/50′.”

    Does this same ‘balance of probabilities’ hold true for when Social Services are transforming a contact into a child protection referral? Do they have any duty to verify the truth/source of information received before they create records for the children?


    1. Sarah Phillimore Post author

      This is an interesting example of how finely balanced some of the arguments can be about imposing a duty of care on professionals who are investigating child abuse. House of Lords in 2005. Might not be entirely on point but thought it worth a read if you haven’t already.


      1. C

        Many thanks for the links. I hadn’t read them, but have now.

        I find these judgments obscene. How expensive must it have been for the aggrieved families? The terrible implication is that CSC can do almost whatever they like, and behave as poorly as they wish – as in our case, for instance, sending misinformation to the Ombudsman, and maintaining misleading, demonstrable falsehoods throughout their complaint reports – and even though we can absolutely prove them falsehoods, we have no hope of bringing them to book, because their behaving like the Stasi is considered to be ‘in the best interests of the child’. How depressing! CSC’s general level of ineptness makes their unaccountability similar to handing the nuclear launch button to the village idiot. I had thought that John Hemming was slightly unhinged, but I am now coming round to his opinion. How can you continue to function in the midst of such a hopeless fiasco?

        I return to where I began, readily acknowledging the legitimate grievances of these particular appellants, against whom no suspicions whatever remain, sufferers from a presumed want of professional skill and care on the part of the doctors treating their children. It is they, I acknowledge, who are paying the price of the law’s denial of a duty of care. But it is a price they pay in the interests of children generally. The well-being of innumerable children up and down the land depends crucially upon doctors and social workers concerned with their safety being subjected by the law to but a single duty: that of safeguarding the child’s own welfare. It is that imperative which in my judgment must determine the outcome of these appeals. For these reasons, together with those given by my noble and learned friends Lord Nicholls of Birkenhead and Lord Rodger of Earlsferry, I would dismiss them.

        Thus the wrongly accused must be martyred to protect the general. Remind me, did we stop ducking potential witches? With these judgments, it appears not.

        Thank you for you effort in sending me these judgments. in proving to me that there is no justice, you may at least have saved some of my children’s inheritance.

  2. Sarah Phillimore

    I am not a social worker, so can’t reliably comment but I would HOPE that no professional takes serious action without being confident about the information that prompts their action.

    But I know when a situation is perceived as ‘urgent’ there may not be time for considered checks about sources etc.

    As with everything, it’s trying to get a balance between need for speed but need to act proportionally. I accept that sometimes that balance isn’t maintained.

  3. C

    Thanks for the reply. The local authority that I have issues with today sent me an update on a recent case MXA v London Borough IHJ/14/0272. sent to discourage me from pursuing them I guess – but the Update they sent me is an ‘advertorial’ from the solicitors who won the case, and I can’t find any other information that would help me to assess how similar/disimilar our case is from this.
    Except that I would say that a police report of sexual abuse, plus a previous allegation might trump a request for advice from a nursery due to nappy rash. But I don’t know what to believe any more ; this stuff is surreal.

    Where might I find info. on this case and similar cases please?

  4. C

    Hmm. I didn’t know what ‘ex tempore’ judgment meant.
    This from wikipedia: An ex tempore judgment, being off the cuff, does not entail the same preparation as a reserved decision. Consequently, it will not be thought out to the same degree.

    …But if not thought through, does it still retain the same validity?

    To imply that CSC can maintain records brimming with misinformation would seem a dodgy decision… surely there must be some threshold of proof/balance of probabilities?

    1. Sarah Phillimore Post author

      Ex temp ore does not mean it isn’t thought through at all, just not thought through to the same degree as a written judgement which may be prepared over several days or even weeks.

  5. Julie

    Hi – I remember seeing a ref to this case recently (MXA v (1) HARROW LONDON BOROUGH COUNCIL (2) WEST BERKSHIRE COUNCIL (3) TAUNTON DEAN BOROUGH COUNCIL (4) WOKINGHAM BOROUGH COUNCIL (2014) heard on 4th June) and have now found a report which I’m afraid I can’t pass on for copyright reasons but if there is a specific question I will try to summarise relevant bits.

    1. Sarah Phillimore Post author

      Thanks Julie, do you know if it is going to be reported anywhere?

  6. C

    Hello Julie,

    ..Because a report by Weightmans of the MXA. v. case was sent to us by the Local Authority, I was trying to asses the similarities/differences between that case and ours. Without knowing more about the detail of the case, it is hard to know what precise questions to ask. I can send you details of our complaint, if you wish, that you could then measure against the MXA case, but would need to do that confidentially. Perhaps Sarah might supply you with my email address if that would be acceptable to you.

    Many thanks.

    1. Julie

      I can look at the case again and post a summary here of the basic facts & law that were at issue. Hope that will help.

      1. Sarah Phillimore Post author

        Thanks Julie that would be really helpful – I think we also need to expand the post about making a complaint to look at the issue of negligence/duty of care of social worker towards parent.

  7. C

    Thank you. Our concerns are to do with what we see as an unjustified referral, that came about when a nursery called to ask for advice on what transpired to be nappy rash. the contact was recorded very inaccurately, and as a result, was escalated towards a s.47. I say towards, because CSC never got round to holding a strategy discussion and the police claim that they never intended to take any action on the information fax.ed to them. But nearly three years later, all the inaccuracies remain on our children’s files, and these will be maintained/ disseminated by a Social Services who we consider less trustworthy than the Stasi, since they have been duplicitous throughout our complains process, have manufactured falsehoods in their reports, and then sent those on to the LGO… I have to stop, I believe I am starting to sound a little crazy – but then, I wouldn’t believe it if it hadn’t happened to us.

  8. Julie

    Here is a short summary of the abstract I have read on the case in question. If this is an unreported case, it may not be binding on other courts but it may give an indication of the purposes for which a public body can legally process and store data.

    MXA v Harrow LBC & three other LAs
    High Court; 4th June 2014; Mr Justice Bean
    A stepfather claimed breach of data protection principles and breach of his human rights. His claims were struck out (he did not attend the hearing and was presumably unrepresented.)
    The family had moved in and out of four different LA areas. A child protection referral had been made to Harrow. There was a police report setting out a number of relevant recent and older convictions, warnings and allegations. The family moved to another LA area and Harrow passed on the information. The family moved again and the information was passed on again. There was a new referral and an investigation began; he signed consent forms to the sharing of information; care proceedings began. He alleged that the LAs had together disseminated damaging information about him maliciously and not checked if it was true.
    The court decided that the LAs were undertaking child protection investigations, had held the info only as long as necessary (Data protection principle), had passed it on when necessary to protect the child’s welfare, and had not disseminated it beyond that. The LAs had taken reasonable steps to check the accuracy of the info. Processing it was necessary to enable them to comply with statutory duties. There was no evidence of malice or recklessness by the LAs.
    Two cases were cited (which I think are reported on bailii): Johnson v Medical Defence Union Ltd [2007] EWCA Civ 262 and JD v West Berkshire Community Health NHS Trust [2005] UKHL 23

    1. Sarah Phillimore

      Thanks Julie that is really helpful, I will aim to get a post up about these issues within the next week or so.

  9. C

    Thanks Julie, and everyone for info.

    The Data Protection Act does not require us to share data. Rather, it sets out a limited set of circumstances in which it is permitted. One of these is with consent. A second is where it is necessary “for the exercise of any functions conferred on any person by or under any enactment”. This second one is widely relied on, but it requires what is termed a “statutory gateway”.

    One such statutory gateway is section 47 of the Children Act 1989. In the context of a section 47 enquiry, we can share relevant information without consent (subject, of course, to complying with the relevant statutory guidance).

    That being the framework, the judgment in our case begins to make sense. There was no section 47 enquiry. There was no consent. The data gathering was unlawful.

    I take this to mean that in our case the data sharing was unlawful. This seems to me to be compounded by the fact that the CSC complaint – which went through three stages – did not discover this. Instead the I.O. did not upheld our complaint that the parents were not informed, saying that: ‘In my view the telephone conversation made it clear that the parents did not know about the decision to contact…’. This seems bewildering to us, but elsewhere the Adjudicator stated: It is not a matter of whether a complaint is factually correct.. The issue is whether the question should have been asked. In my opinion it was not an omission to fail to ask it. . He was actually talking about asking whether there was any risk of harm – but I suppose the same principle applies in his view.

    Anyway, thus encouraged – I have invested some money in getting proper advice from a barrister…

  10. Sarah Phillimore Post author

    Let us know what happens, I think this is an interesting issue and probably of relevance to many.

  11. C

    Update: Because you asked to be informed of progress.
    It is exceedingly difficult to find a direct access barrister with the necessary skillset to cover both professional negligence/misfeasance and child protection law. Perhaps a useful addition to your site might be some legal resources. I’m going round in circles with professional negligence experts saying it is about family, and family barristers saying it is about professional negligence…

    I’ve built a password protected site with summaries, chronologies, report documents etc. all linked and navigable so that someone doesn’t have to plough through mountains of paper – but the security settings created by I.T. departments in chambers prevent quite a few from being able to access.

    Still – I keep discovering fresh sets to badger.

  12. Sarah Phillimore Post author

    Sorry to hear you are having problems re direct access. We have added a link about legal advice if you check the links and resources tab – there is one firm of solicitors in the north west called Farleys who say they specialise in these kind of claims. They might be able to point you in the direction of someone who feels able to take it on.

  13. C

    Thanks for the Farleys link.
    I am favouring direct access if possible as I feel it might be more efficient – if I can find the right barrister who bridges family, DPA and professional negligence. Considering the behavioural level of Social Services, one would expect them to be out there in droves…

    1. Sarah Phillimore Post author

      I feel that I am only just starting out in my appreciation of the issues of professional negligence and I haven’t conducted any cases in that area yet, but if you get stuck I would be happy to review the papers and provide an advice pro bono; I am direct access trained. But my experience is pretty limited at the moment. If you think that would be helpful, please do get in touch. Its an interesting area and I would like to explore it further.

      1. C

        Dear Sarah,
        That would be fantastic. Really fantastic. I have built a website containing all the documents, with summaries of issues, chronologies, the contact/referral etc. – all with navigable menus and interactive click-button links to the relevant parts of each document. All in the interests of barristers being able to access quickly and easily.
        Not everything is on there… but enough to make one barrister question my sanity. I figured if I have to pursue this mess I might as well polish my javascript web development skills at the same time…
        I’ve made the website username and password protected – but you, as the site owner should be able to see my email address. If you email me directly, I’ll send you the relevant links – and all my details. Oh, thank you – if you are able to explore this. It will be your summertime reading.
        Once I’ve finished with all the legal presentations, I reckon all the shameless bungling on show has some potential for a sit-com.
        If you can’t see my details, then I could contact you via chambers…?

        1. Sarah Phillimore Post author

          Yes, I have got your email I will send you one from my Chambers’ address now.

  14. Pingback: As if People Were Humans – Humanistic psychology in the Family Courts | Child Protection Resource

  15. Pingback: The limits of family court influence over adult disputes about children Part II | The Transparency Project

  16. Angelo Granda

    A Parent’s View

    It would seem ,if one takes into account all the court argument on the subject, that lawyers and especially partially qualified trainee solicitors may well have some amount of difficulty when pondering the standard of proof required in the civil family courts. It is also easy to imagine that the lower Magistrates ( Family Proceedings) fail to understand the issues fully either.Yet it is that Court which carries out the initial fact-finding in most cases : the Magistrates decision if they issue an ICO early on tends to pre-determine the outcome in the Family Court itself later on.It certainly has a great linfluence.
    One does,of course,expect highly trained barristers to be au fait with the standards in a civil court and with those pertaining in criminal courts and the difference.

    To what extent to mere parents understand the difference?

    In a criminal court , the jury hears all the evidence and if it is established that a party has given false evidence,failed to produce evidence,neglected to obey guidelines etc.,the jury can refuse to consider any of that party’s evidence credible and SAFE.Their unlawful aactions will be unacceptable!
    To be continued.

  17. Angelo Granda


    In the inferior civil courts there is a lower standard of proof.A Judge has to make a decision at his own discretion which relies on his or her consideration of ALL the evidence entered into a factual matrix produced to the Court.It is generally accepted that the evidence comprises of a mixture of facts,half-facts,truth and lies ,allegations,speculation and professional opi ions,intelligence reports etc.It is known for certain that the decision made cannot be a totally fair one based on facts alone but it is the duty of the Judge to make one anyway.The decision can never be entirely unbiased as with a twelve -person jury .
    Any decision is made on the balance of probabilities and the balance may be very tight perhaps down to a 51/49 split.
    Therefore, the decision may rely on the smallest snippet of evidence; not any of the evidence can be ignored or discounted,all goes on to the scales of justice and must be counted equally to the rest
    According to the latest addition to the post above,one cannot calculate the value of any one piece of evidence mathematically,every one must be considered a part of a whole .
    Thus,should any one small part of the evidence be false and untrue ,should any portion of freely available evidence be withheld then the decision made cannot be a totally fair one; it must contain an element of error and be unsafe to a degree.
    I reckon that whilst such standards are acceptable in normal civil courts ( even secret ones) ,whilst they may even be acceptable in Family Courts it can only be acceptable if limits be placed on the severity of sanctions imposed.The permanent liquidation of families and/or permanence plans should never be ordered by a civil court.It is disproportionate and out of line with article 8, surely? An inferior court cannot exceed the powers of the higher court with higher standards.

    All comments welcome.

  18. Angelo Granda

    It would seem ,Sarah,please correct me if I misunderstand it , that the latest judgment rebuts the elephants and mice theory; how ironic that it should be due to a local authority appeal given that they profit so much from discounting small mistakes in most cases.
    Obviously,the team who successfully brought the appeal did not have to rely on the legal funding commission or it would not have been brought.They tend to be brilliant when acting for the authorities.

    If the elephants cannot be evaluated mathematically neither can the mice so the argument and findings can work both ways.

    But will it? That will depend on whether parents have access to justice and proportionality,I suppose, as always!

    Can anyone think of an article of the ECHR convention which is not contravened by this child a protection system?
    That is a serious question.
    Parents and their lawyers might find better solutions if they think in reverse.Instead of analysing precedent with regard to successful appeals by other parents which are few and far between,examine winning appeals by LA’s.
    This LA won because the appeal court decided that notwithstanding the fact-finding had relied on some false representations (major ones) ,the apparent failure of father to cooperate and work together with sws overrode it.
    So there you are,lawyers!
    All you have to do is demonstrate to Court that the LA’s will not follow frameworks and cooperate by working together and supporting parents.
    You will find the evidence you need right at the outset of cases usually when they start to flout guidelines.

  19. Angelo Granda

    I repeat,Sarah,my question is serious.
    Can anyone think of an article of the ECHR which is not contravened by this judicial system?

    1. Sarah Phillimore Post author

      The Children Act has been challenged in the ECtHR. The ECtHR upheld the legislation as complaint with the ECHR. So I am afraid the Judges in the ECtHR itself do not think your question is serious.

      1. Angelo Granda

        The Children Act isn’t the main problem.I am no lawyer just a plain and ordinary parent but I would uphold the legislation too were I presiding at the ECHR.
        My question was about our judicial system and access to a fair and proportionate hearing etc.
        What do you think? The Act seems okay.

        1. Sarah Phillimore Post author

          I think if matters continue down their present road – with high vacancy rates in judicial posts and local court buildings being shut, then there is a real risk that individuals will have to wait so long for their cases to be heard, that they will not in any real sense have access to a fair hearing. Justice delayed, is justice denied.

Comments are closed.