Should the standard of proof for child sexual abuse be lowered so that more of those accused can be tried and convicted in the criminal courts? This seemed to be the argument being made by Anne Longfield, children’s commissioner for England, on the Today programme this morning. (interview starts at 1.12.20 into the programme)
Longfield was responding to the family court’s fact-finding as to whether Poppi Worthington’s father had sexually abused her. On the balance of probabilities, the court found that he had. He has never been charged with or convicted of any criminal offence. The gist of Longfield’s argument seemed to be that because the vast majority of child sexual abuse cases never reach court because evidence doesn’t meet the required criminal standard of proof, society needs to rethink what we are willing to accept as sufficient evidence in a criminal trial for this particular charge.
Challenged by presenter Sarah Montague about whether she is recommending that [we drop*] a fundamental tenet of our justice system – that nobody is convicted of a crime unless proved beyond reasonable doubt to have committed it – Longfield seems confused about what she’s asking for. But if you analyse the implications, I think it’s pretty clear.
“I don’t think it’s as definite as saying move from one [standard of proof] to the other… ” she says. “We’ve got two stark examples here of different levels of proof, and the one we’ve got at the moment [in criminal trials] clearly isn’t working for those children who need justice from the abuse they’ve experienced… we need to better decide what does constitute good evidence.”
Sarah Montague points out that there are other crimes in which very few cases that are tried result in conviction: does Longfield want to apply this lower standard of proof to those other tricky-to-charge and hard-to-convict crimes too?
According to the children’s commissioner, however, CSA is “a very particular crime”. This seems to imply that the dynamics of familial sexual abuse and the difficulties in evidencing it mean that different – lesser? currently inadmissible? – forms of evidence should be accepted in criminal courts. It seems to me that she’s saying that we should as a society countenance and discuss the possibility that people could be convicted of a particularly vile crime on a less rigorous standard of proof than if you were charged with, say, fraud.
If Longfield really believes that evidence that is currently not viewed as strong enough for a jury to be sure should be given more weight, then she needs to come out and say that it is therefore acceptable that people will be convicted of criminal acts without us being able to be sure they’ve done something terrible – and some will very likely be convicted who are innocent, with all the horrendous consequences for themselves and their families that entails.
I mean, I struggle badly enough with the 51% balance of probabilities business when it comes to family courts anyway; clearly that’s to do with keeping children safe, but it has nevertheless created some appallingly destructive results for parents and children, as various published judgements are now demonstrating. So what Longfield is suggesting – without coming out and fully acknowledging and owning the very serious consequences for families, including of course children, when a conviction is wrong – properly shocks me.
Louise Tickle
[Ed’s Note : Last week The Today Programme also ran some pieces about the Poppi Worthington case, which appears to be what has prompted the discussion about the standard of proof in family cases and generally regarding sexual abuse. There is a piece at 0810 on 20 Jan, where Dr Hannah Quirke (Senior lecturer in Criminal Law and Justice at the University of Manchester) and Sir Mark Potter (former judge of the Court of Appeal and President of the Family Division) discuss Poppi’s case and some of the legal issues surrounding the proving of abuse of this sort – see here. We’ve written about the transparency issues arising from Poppi’s case here and given a brief timeline, but will try and write about that discussion when we are able. Lucy Reed]
* correction 16.49 25/01/16
If I were facing trial on a charge of sexual assault upon a child I would damn well want to be found guilty beyond a reasonable doubt. This is an insane suggestion.
It’s not the standard of proof that’s the problem here, it is a police investigation that couldn’t even aspire to being abysmal.
Sarah Montague’s questions were perfectly valid but I don’t think Anne Longfield had been accurately briefed about them. She was talking about the difficulty in making a finding based on evidence from young people who have been abused (a very serious problem). That is not what happened in this case, where it appears that the decision not to prosecute was because of procedural mistakes in gathering physcial evidence. I haven’t seen any suggestion that there was witness evidence that fell below the standard needed. If there was, this was unfortunately not captured in this brief interview. It was also unhelpful to hear constant references to a ‘burden’ of proof, which is irrelevant – it is the standard of proof which varies between ‘beyond reasonable doubt’ (criminal) and ‘on balance of probabilites’ (civil). The burden of proof is only where that burden lies. This hugely problematic area was subject to intense debate after the Cannings appeal, and if it is being revisited now, it would be sensible to look at those cases as well, not just one in isolation.
To better understand Anne Longfield’s comments, read the Children’s Commissioner’s Report of November 2015 on CSA, particularly chapter 8.2 where she asserts that there were 425,000 hidden CSA cases over a 2 year period.
That equates to a trebling of Crown Court cases listed for trial if all were pursued. The trouble is that the Report’s methodology is questionable. It does not say (and there are no footnotes showing a breakdown) whether all cases reported by police fell into the CSA category or were merely neglect matters. Nor which cases have been double-counted e.g. EPO by police then an Interim Care Order by the Council within 24 hours for the same child.
Given that the Commissioner is responsible for the protection of children’s interests in institutions, it is surprising that the Report includes no analysis of the danger of lowering the standard of proof for children on remand for sex offences.
Additionally, the Report recommends that social workers conduct ABE interviews and that intermediaries are present. Nowhere is the issue of the cost of training social workers to conduct interviews which do not produce inadmissible evidence addressed. Nor is the cost of intermediaries (frequently at a cost of £250 via agencies) factored in.
It is a shame that not a single lawyer was on the Panel producing an £86,000 Report dealing with key issues in family and criminal evidence.
Hello Alistair, I haven’t read the whole report but the government’s statistics show that the number of children involved in all care proceedings each year is an average of 28,000, which falls some way short of the 200,000 plus that is suggested (425,000 over 2 years). Only a smallish proportion of those cases will involve CSA, even fewer will involve a trial of the issues like in the Poppi Worthington case. So there are not 425,000 “hidden CSA cases” in the sense of cases dealt with in private family proceedings that don’t make it to the criminal courts – my reading of that part of the report is that the estimate is of the number of victims of CSA, and they are hidden because a vast number of them are victims that are not known about, so the question of EPO / ICO etc doesn’t really arise. I don’t think the report is saying that all 425,000 are subject to child protection processes / known children.
the report you are looking at is here I think. And the Govt stats I was referring to are here.
Hope I have not misunderstood the thrust of your comment.
I think Alistair is arguing that he has doubts about the figures that the report quotes about known cases (from the police and the care system). Therefore, if the statistical model used to arrive at unknown cases is based on inaccurate ‘known’ figures, then the overall conclusions (on prevalence) have to be questioned.