Diluting Defendants’ Rights: Corroboration

The law until 1994

Until 1994, the criminal law of England and Wales required that there should always be ‘corroboration warning’ in sexual offence cases. A conviction where no warning had been given or was given improperly would most likely be overturned on appeal.

A jury could convict a defendant in such a case where there was no corroboration, as long as the appropriate warning had been given.

The Court of Appeal in DPP v Kilbourne (1973) AC 729 said that ‘corroboration’…

by itself means no more than evidence tending to confirm other evidence. [E]vidence which is (a) admissible and (b) relevant to the evidence requiring corroboration and, if believed, confirming it in the required particulars, is capable of being corroboration of that evidence and, when believed, is in fact such corroboration.

Lord Salmon, R v Henry and Manning (1969) 53 CrAppR 150 set out why, in his view, a corroboration warning was appropriate.

What the judge has to do is to use clear and simple language that will without any doubt convey to the jury that in cases of alleged sexual offences it is really dangerous to convict on the evidence of the woman or girl alone. This is dangerous because human experience has shown that in these courts girls and women do sometimes tell an entirely false story which is very easy to fabricate, but extremely difficult to refute. Such stories are fabricated for all sorts of reasons… and sometimes for no reason at all. The judge should… tell the jury that… they
have to look at the particular facts of the particular case and if, having given full weight to the warning, they come to the conclusion that in the particular case the woman or girl without any real doubt is speaking the truth, then… they are entitled to convict.

Lord Salmon’s focus on “girls and women” does seem perniciously old-fashioned, and indefensible in today’s terms [1].

However, this highly experienced senior judge also makes a good point. It is one which needs to be made today, even though it is almost blasphemous in today’s victim-obsessed culture.

An allegation of a sexual offence remains today “easy to fabricate” and “difficult to refute”, and it is a matter of fact [2] – whatever proponents of the easier sexual offence convictions may argue – that such fabrications do take place. Any evaluation of the law as it stood, and any comparison with the law as it is today, has to take that into account.

The level of ‘similar fact’ evidence [3] which could be used to corroborate an allegation was strictly controlled. The similarity between the alleged ‘similar fact’ evidence and the offence charged had to go beyond merely suggesting that the accused was the kind of person to commit the offence charged, and had to be so striking as to point strongly to the conclusion that the defendant was in fact guilty of the offence charged.

So, until 1994, a conviction in a sexual offence case was technically possible without corroborative evidence, as long as the appropriate warning had been given. But as a matter of practice proceedings were not generally taken unless there was corroborative evidence available. Otherwise, the view was taken that the case failed the “likelihood of a conviction” test.

The law from 1994

In 1991 the Law Commission recommended that the corroboration rules should be scrapped. As their report notes [4], the Criminal Bar Association argued that a corroboration warning in these cases was an “an essential safeguard for the defendant”. However, the Law Commission believed that “the present general obligations of a trial judge with regard to doubtful or unreliable evidence” would adequately preserve a defendant’s position. Thus, a rule mandating a warning was to be replaced with a generalized hope that judges would ‘play fair’ with defendants.

So the mandatory requirement for a corroboration warning in trials of sexual offences was scrapped by the Criminal Justice and Public Order Act 1994.


The scrapping of this protective mechanism – a provision which made it less likely that an accused would be convicted on shaky complainant evidence – no doubt led to a small increase in rightful convictions. Unfortunately, it also made it much more likely that innocent accused were convicted of these serious offences.

Quite apart from the (unknown and unknowable) statistics there, that’s not the real point. It is the cumulative effect of a number of legal and practice changes, all weakening defendants’ rights, which put us in the place we are today.

Many of those cumulative elements will be discussed in later posts. Prime among them, though, are the police’s and CPS’s dogmatic assumption that complainants never lie and are never mistaken, and the CPS’s craven willingness to prosecute even weak sexual offense cases, leaving it to the jury to pick up responsibility for making the hard decision, if it can. Together, these politically motivated stances cost the taxpayer money and, even more importantly, damage lives unnecessarily.

In the end, it’s a question of balancing justice. No one wants sex offenders to evade conviction. But, even more, no one should want innocent men and women convicted of a sexual offence. If it were just a matter of some minor changes in the law, that would be one thing. But it’s not. Changes in corroboration rules. A dilution of when ‘bad character’ can be brought in. Weakening of ‘similar fact evidence’ provisions. A blurring between evidence of criminality and more general ‘bad behavior’. Changes to hearsay rules. Making Legal Aid harder to get. And, above all, the politics of prosecuting nearly everything.

There’s a pattern here. The change is all in one direction: to make convictions more likely and acquittals (even where there should be an acquittal) harder to obtain.

Part of the problem is that there is little likelihood of going back. Once a rule (even one needing finessing, such as the corroboration warning) is abrogated, it’s likely gone for good.

What we’re seeing over the last twenty or so years is a wholesale surrender to a form of political correctness which says that complainants in sexual offence cases rarely lie, and that defendants are usually guilty. And that’s not true, right, or just.


  1. As the 1991 Law Commission report on Corroboration of Evidence in Criminal Trials pointed out, “In recent years, it appears that a general form of direction is often used which does not distinguish between male and female complainants”.  ↩
  2. See the evidence accumulated in their newsletters by www.safari-uk.org.  ↩
  3. There will be a post on similar fact evidence coming soon.  ↩
  4. Paragraph 3.4 of the report.  ↩

Defendant anonymity: a provocative suggestion…

Of course, there is another solution to the impasse over defendant anonymity until conviction in sexual offence cases. It’s an idea which those who argue for a free and open press and a transparent justice system might well appreciate.

It’s this. Repeal the law providing complainants in sexual offence cases with anonymity. Indeed, make it possible that their names and general location will be available to the public.

In this way, we can flip the arguments that those such as Emily Thornberry put forward on their head.

She has argued that publicizing the name of a suspect in these cases may well enable the public to make further complaint against the same person. Equally, for all we know, a given complainant might be – might well be – a serial complainant. Thus, publicizing their name and where they live might allow others who have been the subject of previous (false) complaints to have their memory jogged and come forward, etc, etc.

Likewise, anonymity may not be (according to her argument, at least) in the complainant’s best interests, because, if it is alleged against that person that they have lied, exculpatory witness will not know to come forward to support them either. And, arguably, if it is likely that the identity of a complainant is known, there is less risk that people will fabricate allegations, as indeed they sometimes do.

Meanwhile, Sir Cliff Richard has canceled his involvement which would have given him the freedom of the town in Portugal where he sometimes lives, and he has now been interviewed under caution, surrendering voluntarily and strenuously denying the truth of the allegations.

For the moment, those watching the story unfold are divided between those who will now (on the basis of no evidence whatsoever) perceive him “warily” because, after all, child abuse is (apparently!!) “the new “normal” in the land of fame and fortune”, and those such as Charlie Brooker, who points out that “Cliff hasn’t been charged or found guilty of anything, except on the internet, where he’s already serving concurrent 140-character sentences”.

Perhaps the justified brouhaha over the way the police and the BBC handled all of this will lead to a watershed moment where justice is offered to suspects as well as complainants.

In any event, let’s have it one way or the other. Anonymity for complainants and suspects in these cases. Or open season for all of them. Except that the latter isn’t going to happen any time soon. And probably not the former either.

Defendant anonymity: an emerging debate?

A debate seems to be emerging around the search of Sir Cliff Richard’s Sunningdale home and the tip-off to the BBC. It is about time that this debate takes place. But it is a pity that the way it arises should be so prejudicial to a person who has not yet even been interviewed by police about a criminal complaint made about him.

The Guardian reports that South Yorkshire Police have formally complained about the BBC’s coverage of the search of Richard’s property.

The force was contacted some weeks ago by a BBC reporter who made it clear he knew of the existence of an investigation. It was clear he [was] in a position to publish it.

The force was reluctant to co-operate but felt that to do otherwise would risk losing any potential evidence, so in the interests of the investigation it was agreed that the reporter would be notified of the date of the house search in return for delaying publication of any of the facts.

Contrary to media reports, this decision was not taken in order to maximise publicity, it was taken to preserve any potential evidence.

South Yorkshire [P]olice considers it disappointing that the BBC was slow to acknowledge that the force was not the source of the leak.

A letter of complaint has been sent to the director-general of the BBC making it clear that the broadcaster appears to have contravened its editorial guidelines.

Additionally, Geoffrey Robertson QC has weighed into the fray in an Independent article. His claim that “orchestrating massive publicity for the raid on his house” seems, in the light of South Yorkshire Police’s complaint, possibly wide of the mark. However, he does make various telling points:-

1. The search was made in broad daylight when the police knew Richard was away from the premises.

2. Police codes provide that “the officer in charge of the search shall first try to communicate with the occupier” before entering the premises, and this aspect of the code was not complied with in this case.

3. The law requires that, for a search warrant to be granted, a justice of the peace should be satisfied that ”there is material on the premises both relevant and of substantial value“ to the investigation of a criminal offence. Robertson wonders how this element of the law can possibly be satisfied for the search of premises in 2014 in relation to an offence which is alleged to have been committed in another part of the country nearly 30 years ago.

4. The warrant should only be issued if it is “not practicable to communicate” with the owner of the premises, and there could be no possible difficulty in finding a way to communicate with Sir Cliff Richard, of all people.

5. Police codes provide that powers of search should “be used fairly, responsibly, with respect for occupiers of premises being searched”. Far from this being the case, the search was conducted in the full glare of media publicity.

6. The police should be required to get search warrants from circuit judges, rather than from non-legally-trained lay justices of the peace.

Now, there are counter-arguments to at least two of Robertson’s points, though at least one of them does not really hold water.

In respect of point 2, paragraph 6.4 of the Police and Criminal Evidence Act 1984 Code B states that communication with the occupier should be attempted ”unless the search premises are unoccupied“, and indeed they were unoccupied. On the other hand, the police will have known that they were unoccupied, and may have even waited until that was the case before initiating the search.

Point 3 above: it is invariable practice that a search is conducted in historical abuse cases, and papers, computers, cameras, disks taken away for analysis. Whether this is a form of fishing for evidence to bolster a case, or a reasonable extension of a police investigation into other areas of inquiry, is a matter of opinion.

Whatever view one takes here, it is arguable that, under the current law, police should have waited until Richard was at the premises, visited him there, arrested him as necessary, and conducted a search under their general powers of search after arrest under Section 32 of the Police and Criminal Evidence Act 1984. Any arrest and search should have taken place discreetly, probably after hours of darkness, and without notice to or discussion with the media, whatever their state of prior knowledge.

But Robertson makes a further claim, about the effect of high-profile searches and prosecutorial delay on a suspect’s life:-

If the outrageous treatment of Paul Gambuccini and Jimmy Tarbuck is any guide, Cliff Richard will remain in a cruel limbo for 18 months or so until the police and the CPS decide whether to charge him. This has been one of the most intolerable features of other high-profile arrests for ”historic” offences, namely the inability of police and prosecutors to deliver Magna Carta’s truly historic promise that justice will not be delayed.

The CPS has taken up to 2 years to tell journalists like Patrick Foster that they will not be prosecuted, after unnecessary dawn raids, and publicity every time they are bailed. This lack of care for their liberty is amoral, because it subjects them to drawn-out psychological cruelty. If the CPS cannot decide whether to prosecute 3 months after receiving the police file, it should not prosecute at all.

Robertson makes two other claims, first that the search clearly broke the privacy provisions of article 8 of the European Convention of Human Rights; and, second, that most European countries have time limits which prevent prosecutions of most sexual offences after a lapse of 10 years. After 25 years, fair trial becomes very difficult, as memories dim, alibi witnesses die and records disappear. We will assess these arguments in later posts.

For the moment though, let’s congratulate Robertson for standing up for the rights of the suspect in historical abuse investigations. Suspects in such cases are human beings too, with human rights under the law. It appears that the criminal justice system in England and Wales has decided that these rights no longer really count for anything when stacked up against their pressing need to be seen to be doing something, anything, to satisfy the public that they are doing all they can to stem the alleged tide of child abuse, which is any event a myth born of a generalised public fear and anxiety.

If we are not yet back in Salem, we are getting pretty close to being there.


Defendant anonymity: an update

The 14 August 2014 search of Sir Cliff Richard’s Berkshire home after an allegation of sexual molestation of a minor in 1985 and the attendant press coverage highlight the pressing need for defendant anonymity in alleged sexual offence cases, as we claimed in two recent posts.

Various aspects of this event demand inquiry.

Cliff Richard, on holiday in Portugal, was given no notice of the search. But the press probably were, since they were able to send reporters and helicopters to cover the story as it happened. South Yorkshire police have said that news of their search of Richard’s property had been leaked to the BBC, but that they were not to blame.

Cliff Richard certainly believes that the press were forewarned, and issued this statement:-

For many months I have been aware of allegations against me of historic impropriety, which have been circulating online. The allegations are completely false. Up until now, I have chosen not to dignify the false allegations with a response, as it would just give them more oxygen. However, the police attended my apartment in Berkshire today without notice, except, it would appear, to the press. I am not presently in the UK but it goes without saying that I will cooperate fully should the police wish to speak to me.

Why were the media forewarned of this search?

And so the trial by media commences, before the entertainer has even been questioned by police. There are already news articles (too many to cite) about Richard’s bachelor lifestyle, his Christian faith, his homes in various parts of the world. There are aerial shots of the Berkshire home which was the subject of the raid, and long lens photos of police officers donning latex gloves to comb through the property.

Richard will no doubt be interviewed by police. But, even if he is not charged, or if he is charged, tried and acquitted, the die is already cast. The merest suggestion of this kind of sexual impropriety stains a person’s life, and Richard will need to live under this shadow for some time to come.

Presciently, we suggested in our second post on defendant anonymity that the identity of people such as Nigel Evans, charged and then acquitted of sexual offences, should not be available for media reporting unless and until a conviction ensues. And now Evans has argued on television – rightly, in our view – that there should be defendant anonymity in these cases.

Whatever the final outcome for Cliff Richard and others, the point holds true. A person is innocent until proven guilty, and there should be no trial by media. We are all for freedom of the press. But today the press has license to smear and to ruin people’s lives. There urgently needs to be a ban on publicising the names of those against whom sexual allegations are made.

Defendant anonymity (part 2)

[Go here for the first part of this extended post…]

A summary of the debate (continued)

Since 1981 it has been the law in the Republic of Ireland that defendants in rape [1] trials cannot be identified pre-conviction. There is a provision [2] – which deals with Thornberry’s claim about the availability of defence witnesses – for the defence to apply to the court to have defendant anonymity lifted in order to induce potential witnesses to come forward.

There are those who argue, post-Clifford, that this provision is not in the public interest. But it has been on the Irish statute book for 33 years. Should England and Wales have a similar provision, but for a rather wider range of sexual offences?


As we pointed out in our last post, in 2013 fully 32% of those charged in England and Wales for rape, child sex abuse, indecent assault and other types of sexual offending were not convicted or cautioned.

10,365 sexual offences were proceeded against in 2013. 5,659 people were found guilty. 1,349 people were cautioned. This means that 3,357 people (just over 32%) were not convicted and retain, in law, their innocence. Nearly 3,500 people, in one year alone, risked being publicly identified with the slur and stigma of sexual offending and carrying that burden into the rest of their lives.

Note that this is despite the relaxation of the law since the 1990s in terms of corroboration and hearsay. The conviction rate in 2013 for this kind of offence was truly lamentable. It confirms our belief that the Crown Prosecution Service does not weed out enough cases before the charging stage.

Once a sexual complaint is made, and as long as the complainant sticks to his or her complaint, the likelihood is that the matter will proceed to charge and to trial. If there are internal inconsistencies in the complainant’s evidence, some attempt will be made by CPS and police to close up those holes. But the inconsistencies – unless they make it crystal clear that the complainant was fabricating and that no offence could have taken place – will not usually lead to a decision not to prosecute. This is because the CPS prefers to have such matters tested by the court, rather than take such decisions itself. It is politically more expedient, in the current climate, to be criticized for obtaining too few convictions than to be criticized for being ‘soft’ on sexual offending.

To see how the CPS’s sometimes cavalier attitude towards charging and proceeding to trial can work out on the ground, let’s look at the prosecution of Nigel Evans, former deputy speaker of the House of Commons. This was a high-profile case. A prosecution was mounted even though complainants said in court that they did not consider themselves victims of any criminal offence, nor had they wanted to complain to police. Some of the witnesses “suggested they felt pressured by police into appearing as alleged victims in the high-profile trial.”

There was another complainant who alleged rape against Evans, thought the facts were strongly disputed around the issue of consent. The prosecution aim here was clearly to use the weaker allegations to bolster the rather stronger one, a technique known as “bundling”.

Nigel Evans was acquitted on all counts. He is having difficulty recouping his £130,000 legal defence costs. He is likely to carry the effect of these failed proceedings publicly for many years. Personal details of his personal life have been under the public gaze as a result.

So the question ought to be, not “why should there be pre-conviction defendant anonymity in the these cases?”, but “why should there not be?”. The government, CPS and police cannot have it both ways. If they insist for reasons of political expediency on charging people a third of whom they know will probably be acquitted, then the case for defendant anonymity is overwhelming.

Michelle Healey argues at Halbury’s Law Exchange that:-

… those accused of sexual offences ought not to be named unless they are charged. This would offer a person who is investigated but then exonerated the chance to move forward with their life without negative publicity. It would also ensure that if the CPS were making a charging decision they would look only at the offence charged and assess the weight of the evidence to see whether there is a reasonable prospect of success, rather than looking at numerous allegations and allowing potentially weaker allegations to be charged in order to shore up a conviction on potentially stronger cases… Once charged then it is right and proper that a suspect be named, as they would be in any other criminal case.

Healey is working along the right lines, but she does not go far enough. If defendant anonymity is lifted post-charge and pre-conviction, untold damage can still be done where there is an eventual acquittal, as there was in 32% of cases in 2013. Defendant anonymity in sexual offence cases should be in place from the moment of original complaint until the moment of conviction.

  1. This provision doesn’t extend to other kinds of sexual offence.  ↩
  2. Section 8 (3) of the Criminal Law (Rape) Act 1981.  ↩

Defendant anonymity (part 1)

In England and Wales, the complainant in a sexual offence prosecution is conferred life-long anonymity. Nothing can be published which is likely to lead members of the public to identify them. It would be contempt of court to do so.

Should the defendant in such cases be granted anonymity also, at least unless and until convicted?

There has been vigorous debate over the last few years about this. Over two posts, we will describe some of the arguments made for and against, and then contribute some thoughts of our own.

A summary of the debate

A lively discussion took place in November 2013 between philosopher A C Grayling and Emily Thornberry, Labour MP and Shadow Attorney General.

A C Grayling’s explicit “one fundamental issue at stake” [1] is “the unacceptable risk of besmirching innocent people by parading them in public as being under suspicion for abhorrent crimes”. Even if a defendant in this kind of case is acquitted, stigma and suspicion are likely to remain. Conferring anonymity on a defendant until a conviction takes place honours the presumption of innocence and avoids the jeopardy of causing serious injury to that person:-

… publicising the name and alleged crimes of a “suspect”—a person under suspicion; a person under a cloud therefore—is already to begin punishing that person, and if she or he is innocent, that punishment is unjust [2]

For A C Grayling, all arguments which try to counter this fundamental issue are slippery slope arguments.

Thornberry disagrees. A C Grayling’s argument is based on “a patrician hunch that the public is too stupid to grasp the meaning of the presumption of innocence or of an acquittal by a jury of one’s peers” [3]. We do not have a police force that keeps the public in the dark [4]. The aim has always been that the public works with the police to solve cases. Naming defendants can jog the memory of potential witnesses, both for the prosecution and for the defence [5]. Anonymity would not be in the defendant’s best interests, because exculpatory witness will not know to come forward [6].

Thornberry continues:-

By offering [suspected sex offenders] a layer of protection not provided to other suspects, we would fan a pernicious myth that there is something inherently unreliable, unstable and vindictive about rape complainants, setting them apart from other types of victim.


Recent research by the… Director of Public Prosecutions… has discredited the idea that false allegations are a particular problem with rape cases… Introducing anonymity for rape suspects would set the clock back on efforts to bring a more enlightened approach to these cases.

But that’s not the point. No one is claiming that the system is rife with false allegations, though there are probably more than the DPP’s statistic would lead us to believe. In any event, there are doubtless more cases where a complainant’s evidence is unreliable for a whole range of possible reasons [7]. The better test would be to inquire as to how many convictions and how many acquittals there have been for sexual offences over a given period.

In May 2014 the media reported that “conviction rates in England and Wales for rape, child sex abuse, indecent assault and other types of sexual offending fell from 61 per cent to 55 per cent in 2013”.

Fully 32% of those charged who were not convicted or cautioned. This suggests (1) that too many complaints proceed to prosecution (see more on this in our next post); and (2) that A C Grayling’s “one fundamental issue” is more telling than Thornberry wishes to admit.

Barrister Blogger is supportive of defendant anonymity. He points out that it was in fact the law, in rape cases, in England and Wales between 1976 and 1988 – it’s been done before. Yes, he admits, there is a logical inconsistency in granting defendant anonymity in sexual offence cases but not in other serious cases, such as murder. But:-

The fact is that the law already recognises the unique difficulties of sexual cases by providing blanket anonymity to complainants, before, during and after any trial. Similar considerations, including a justified fear of a crazily disproportionate reaction to an untested allegation, suggest that defendants in these cases should at least be given a similar protection until they are convicted.

What’s more, he says, anonymity would lead to fairer trials, especially in celebrity cases, and it would “avoid the pre-trial public relations battle”.

Legal academic Richard Glover is not convinced. He takes the point about potential damage to innocent defendants’ lives and reputations, and “can certainly sympathise with” their predicament. However, he feels that “the important principle of open justice” militates against defendant anonymity.

That’s probably the strongest point against defendant anonymity. But ultimately it’s not compelling. Open justice is a hallmark of a democratic society. Yet the principle has already been breached for complainants. And what price “open justice” where the outcome of an acquittal is that an innocent person justifiably fears unwarranted and undeserved social stigma?

In our next post we consider the position in the Republic of Ireland, where there is defendant anonymity in rape cases, and we set out our own conclusions.


  1. for which he is roundly criticized by Thornberry.  ↩
  2. Barrister Blogger and law academic Richard Glover both put forward versions of this argument, one to support it, the other to say it is “misleading”.  ↩
  3. This is an ad hominem argument, and so probably should be discounted. No one is suggesting people are stupid. But the emotional potency surrounding sexual offences is so great that, despite logic, stigma is likely to last, despite acquittal. As in: “Wasn’t
    s/he the person who was accused of…?”
  4. Which way does Thornberry want it? We are being “kept in the dark” about the identity of sexual offence complainants, and she and most people believe that that is right and appropriate. Rather than using emotive language, it would be better to concentrate on why, in her view, there should be no parity here between complainant and unconvicted defendant.  ↩
  5. A specious argument. If publicity does help to bring forward any further complaints, why not gazette all the names of defendants charged with sexual offences? Because that would smack of witch hunt and persecution. Thornberry’s “emboldening more victims” arguments carries little weight.  ↩
  6. This is not how it works. A defendant and his or her lawyers can and will find their own witnesses.  ↩
  7. though it is almost blasphemous to say so in the current climate…  ↩