The Jeremy Bamber files: three crucial questions about the White House Farm massacre

In 1986, Bamber was convicted of the murders of five members of his family. He has always protested his innocence. This year, with the help of leading experts, we have re-examined some of the most compelling evidence

Warning: this piece contains crime-scene photographs

The White House Farm massacre is one of Britain’s most notorious and horrifying crimes. On 7 August 1985, five members of the Bamber family were shot dead at their farmhouse in Essex, England. This Wednesday marked exactly 40 years since Jeremy Bamber was convicted of the murders of his adoptive parents, his sister and her six-year-old twin boys. He was initially jailed for a minimum of 25 years, later upgraded to a whole-life sentence.

If Bamber did kill his family, it is right that he remains in prison. But there have always been questions about the safety of his convictions, and these have grown recently. My late reporting partner, Eric Allison, formerly the prisons correspondent for the Guardian, spent years investigating the case and we began reporting on these questions together. If Bamber isn’t guilty, the idea of an innocent man spending 41 years and counting in prison, wrongfully convicted of killing his loved ones, is horrifying.

Bamber’s case has been submitted three times to the Criminal Cases Review Commission (CCRC), the body responsible for investigating alleged miscarriages of justice. If the CCRC believes there is a “real possibility” that the verdict will be quashed or the sentence changed, it then refers cases back to the court of appeal. Bamber’s case was referred once by the CCRC to the court of appeal in 2001, and the panel of judges concluded: “The more we examined the detail of the case, the more likely we thought it to be that the jury were right.” In 2012, the CCRC refused to refer Bamber’s case back to the court of appeal. In 2021, he made a new submission; the CCRC has rejected the four of 10 points it has so far analysed.

In recent years, the CCRC has been widely criticised, not only for the time it takes to investigate alleged wrongful convictions, but for failing to investigate them, full stop. The Bamber case is a prime example. The CCRC has taken five years to examine those four points, and there has seemingly been no attempt to investigate the issues that Bamber insists make his conviction unsafe or prove his innocence.

In fact, at times, the CCRC seems to have gone out of its way to do the opposite – to challenge convincing evidence by returning to trial evidence that has subsequently been refuted or proved unreliable.

In 2024, the investigative journalist Heidi Blake wrote a brilliant piece for the New Yorker, challenging the safety of Bamber’s conviction. This was subsequently turned into the podcast Blood Relatives. She interviewed a former police officer, Nicholas Milbank, who told her that he was monitoring the open phone line at White House Farm and confirmed that at 6.09am – at which point Bamber had been standing outside the farmhouse with the police for two hours – he received a 999 call from inside the farmhouse. If there was still somebody alive in the house, this would obviously have cleared Bamber.

Instead of interviewing Milbank itself, the CCRC allowed Essex police (whom Milbank still worked for) to interview him, and the force reported back to the commission that he had never, to his knowledge, spoken to the New Yorker. The CCRC chose to believe Essex police rather than the New Yorker, one of the world’s most respected publications. Essex police also informed the CCRC that Milbank had died soon after his denial. The New Yorker podcast came out a few months later, and it emerged that Milbank had indeed spoken to Blake – there was audio of their interview – and had confirmed the 999 call from inside the house.

When Dame Vera Baird KC became interim chair of the CCRC last year, she said of the commission: “They don’t communicate with applicants, are reluctant to challenge the court of appeal, look for reasons not to refer rather than to refer and are quite often incompetent.”

In our series The Bamber Files, we have attempted to do some of the work the CCRC should have been doing all along. Forensic scientists have been commissioned to examine key issues and we have returned to original police logs, witness statements and court transcripts to expose inconsistencies that appear to make Bamber’s conviction unsafe.

This is a summary of what has been discovered in the Bamber files, and why it is important for the CCRC to reconsider these points, or consider them for the first time.

At Bamber’s trial, the jury was presented with a binary choice – either Bamber killed five members of his family, or Sheila Caffell killed her parents and her children before turning the gun on herself.

The use of a silencer was central to Bamber being found guilty. The silencer in question was found by Bamber’s relatives in a downstairs cupboard at the farmhouse that had already been checked by the police. From the start, therefore, its provenance was in question.

The 1986 trial turned on the silencer: the prosecution stated that, with the silencer attached to the rifle – and therefore adding to the gun’s length – Sheila Caffell’s arms were too short for her to have shot herself. They also asked how Caffell could have shot herself twice, in the throat and the neck, with the silencer attached, then walked downstairs, replaced the silencer in the cupboard and returned upstairs to die. This clearly didn’t make sense.

There was another significant factor that explained why the silencer implicated Bamber. The prosecution alleged that it contained blood flakes belonging to Caffell. In fact, it was known at the time that this blood only matched her blood grouping, and this grouping could have belonged to either her or her uncle Robert Boutflour, a vital prosecution witness who regularly used the guns for hunting.

In addition, the finding for one of the five components that were analysed in the blood group testing was identical to that found in the blood of some animals, such as rabbits, which were regularly shot on the farm. The jury was never told this.

During the jury’s deliberations, they sent a question to the judge, Mr Justice Drake, asking if the blood in the silencer definitely belonged to Caffell. Yes, he replied, adding: “On the evidence of the silencer alone you may find Mr Bamber guilty.” Drake, who died in 2014, arguably misled the jury. About 20 minutes later, the jury returned, finding Bamber guilty of the five murders on a 10-2 majority.

In 2012, Bamber’s team commissioned the American forensic expert Dr David Fowler, formerly the chief medical examiner of the US state of Maryland, to examine the bullet wound injuries to Caffell. His report concluded that she was shot without a silencer. Two equally eminent American pathologists peer-reviewed Fowler’s findings and agreed with him. After brief consideration, the CCRC rejected Fowler’s findings in 2012.

Astonishingly, it wasn’t until nine years later, in 2021, that Bamber could resubmit this evidence to the CCRC with a detailed critique of its rejection – and not until last year that the commission finished analysing it. Even more astonishingly, the evidence was rejected on a technicality.

In its provisional statement of reasons, published in March 2025, the CCRC said that it rejected Fowler’s assertion that no silencer had been used because this did not align with the trial testimony. It claimed that Fowler characterised the wounds to Caffell as resulting from “firm contact” with the skin, whereas the prosecution’s firearms witness Malcolm Fletcher had said the lower wound may have been the result of a shot between one and three inches away.

If the CCRC had reservations about Fowler’s work, surely it should have commissioned its own forensic experts to examine Caffell’s wounds and see if its findings pointed to the use or otherwise of a silencer? It didn’t. In its conclusion on the silencer, the CCRC stated that there was “no real possibility that evidence from Dr Fowler might lead the court of appeal to quash any of Mr Bamber’s convictions”.

Last year, the Guardian commissioned Prof Jason Payne-James, a specialist in forensic and legal medicine, to examine the crime-scene photographs, including Caffell’s injuries. He concluded that the bullet injuries were caused by a contact or close-range shooting from an Anschütz 525 rifle (the gun in question) without a silencer attached.

Payne-James, the president of the European Council of Legal & Forensic Medicine, specialises in the medical recording and interpretation of injuries. He told the Guardian: “The pattern imprint on the skin is not large enough to suggest that a silencer was used, either at very close range or in contact with her body. These are close-range bullet holes, and the nature of the moderator or silencer is such that you’d expect some form of pattern imprint equivalent to the diameter of the silencer if it was used in contact or at very close range.”

The use of the silencer was one of the main factors in convicting Bamber, as the jury had been told that it would have been impossible for Caffell to have killed herself with it attached to the rifle.

We believe the conclusion reached by Payne-James – that the shots to Caffell’s neck and throat were “in contact or at very close range” – overrides any objection that the CCRC had to Fowler’s earlier findings, and should ensure a referral back to the court of appeal.

Last year, the Guardian was handed the crime-scene photographs by Bamber. At his trial, only a few photos were available to him and the jury. At his 2002 appeal, another batch was made available. But it was only in 2011 that he received the complete batch of 406 photographs.

Essex police had finally given the negatives to the CCRC, and Bamber asked the miscarriage of justice body to pay to have the negatives turned into hi-res photos. The CCRC refused. In the end, Bamber paid more than £2,000 for the process, with money raised by supporters.

If the CCRC had wished to investigate Bamber’s case seriously, the crime-scene photographs were a gift. They provided the evidence from which Fowler and Payne-James concluded that Sheila Caffell was not shot with a gun that had a silencer attached.

The crime-scene photographs also contained a number of images from the master bedroom that suggested there had been a struggle between June Bamber and Sheila Caffell. We asked the leading forensic biologist Sue Carney-Gannon to analyse the bloodstain patterns. While she concluded that she couldn’t tell whether there had been a struggle, she made far more significant findings.

Carney-Gannon wrote in her report: “In my opinion, if Sheila Caffell took her own life using the rifle, as the defence for Jeremy Bamber assert, then this explains the distribution of blood on her nightdress and right arm.

“In my opinion, these findings are accounted for by her having taken her own life with the rifle; blood from the wound under her chin running down the upper right front of her clothing and splashing on to her right forearm, then veering to her right side as her body and arms fell.”

At trial, there was an obscure argument about whether Bamber had called Essex police at 3.26am or 3.36am. It was impossible for the jury to know the relevance of this at the time. The counsel for the defence, Geoffrey Rivlin, preferred the earlier time, because he thought it looked bad that Bamber said he had taken a call from his father at around 3.15am and waited roughly 20 minutes before calling the police. Rivlin thought 10 minutes would seem more acceptable to a jury.

In fact, as Bamber told the Guardian, in the intervening 20 minutes, he had been trying to call his father back after the line went dead and was asking advice from his then girlfriend, Julie Mugford. Most importantly, he claims, he had no idea that the call from his father was the precursor to five of his family being murdered. As far as he was concerned, he says, the rifle his sister had grabbed was empty and he believed his father would just take it back from her.

But the difference between 3.26am and 3.36am may be far more significant than it was ever thought to be at trial. By comparing police logs, witness statements and court transcripts only disclosed to Bamber 25 years after the trial, the evidence suggests there were two calls to the police about the incident, rather than the one call that Essex police and the CCRC insist on.

Rather than the officer PC Michael West making a 10-minute error in recording the time he received the call from Jeremy Bamber, our new evidence suggests there was an earlier call made by Nevill Bamber at 3.26am, followed by one from Jeremy Bamber at 3.36am.

The CCRC insists the 10-minute discrepancy in the two call logs is simply a mistake and Bamber actually called at 3.26am. If this is the case, then the prosecution chronology makes sense. After the 3.26am call, officers were radioed at 3.30am, while investigating local burglaries on a nearby industrial estate, and returned straight to the local station to collect their police car and left at 3.35am. However, statements from officers who attended the incident say that they spoke to West at 3.37am, when he told them he had Bamber on the phone.

At the trial, the jury was only shown one log from a civilian phone officer, Malcolm Bonnett, who said he took a call from West about the incident and sent out the cars. The time at the top of his log says 3.26am. West gave evidence that he had filled in a “C1” log about the incident, stating that the time of the call was 3.36am. But the jury was never shown this log. Mysteriously, it had disappeared. A quarter of a century later, the log was disclosed to Bamber.

When we compared Bonnett’s and West’s logs, the details were very different – they didn’t seem to suggest one call, but two calls from two people. Bonnett’s log refers to a “daughter” gone berserk; West’s refers to the caller’s sister. Bonnett’s log has White House Farm as the address at the top; West’s has the address of Bamber’s cottage. Bonnett’s log says Caffell is 26; West’s says she is 27 (she had actually just turned 28). Bonnett’s log doesn’t mention her mental health; West’s does. Bonnett’s log doesn’t have Nevill Bamber’s age; West’s states he is 62 (Bamber had mistakenly added an extra year to his father’s age). And, crucially, Bonnett’s log is recorded at 3.26am, West’s at 3.36am.

The obvious conclusion is that two calls were made to the police – one by Nevill Bamber, the other by Jeremy Bamber. There could be no stronger alibi for Bamber than an earlier call from his father saying his daughter had “gone berserk” and grabbed the gun. Essex police has always maintained that Bamber’s version of events, in which Nevill called him to say Sheila was attacking the family, is a fiction he created to provide himself with an alibi.

Bamber’s campaign team has argued for 15 years that there were two calls. But we discovered that a 1988 internal Essex police inquiry into the policing failures in the Bamber investigation, known as the Dickinson inquiry, makes no reference to a 3.26am call from Bamber, and also accepted the 3.36am timings that West had stated in his log.

Analysis of West’s cross-examination at the trial also revealed one crucial sentence. When asked by Rivlin whether Bamber was told that police cars would be sent to attend the scene, West stated that the cars had already been sent by the time he spoke to Bamber. How could that be possible if there was only one call?

If Bamber did call at 3.36am, how can police have been radioed at 3.30am, and how can the first car have left at 3.35am, as officers who attended the scene write in their statements? The only logical conclusion is that Bamber called at 3.36am, after an earlier call was made, presumably – given the precise contents of Bonnett’s call log – by Nevill.

The CCRC does not appear to have examined these crucial documents. In its provisional statement of reasoning, it simply says that the 3.26am/3.36am argument was aired at the trial and doesn’t need to be revisited. On a basic level, this is true. It was discussed in court. On another level, it is nonsense, because the jury had no idea that West’s C1 form existed or that it was so different from Bonnett’s log. If Rivlin and the defence team had been given access to this, they would surely have spotted the inconsistencies; the question of two calls would have arisen and Bamber may not have been convicted.

Bamber has already waited five years for the CCRC to finish looking at his latest submission and there is no sign that the commission is anywhere near the end of the process. There are still six out of 10 points for them to consider. Meanwhile, Bamber, now 65, having spent more than four decades in prison, continues to serve out his whole-life sentence, with no prospect of release unless his conviction is overturned.

Bamber’s case is one of the most controversial in modern British criminal history. It would cause the establishment huge embarrassment, and call into question the efficacy and fairness of our legal system, if it were found that Bamber had been wrongfully convicted. For Essex police, the CCRC and the court of appeal, the stakes are extremely high. That is exactly why the evidence uncovered in the Bamber Files needs to be reconsidered.

A spokesperson for the Jeremy Bamber Innocence Campaign says: “The jury that convicted Jeremy heard none of the evidence, most of which has been uncovered since his trial, that points to his innocence. Now, the Guardian’s investigations have been another major body blow to the prosecution’s case. The obvious question is: why did it take a newspaper to do this work, rather than the CCRC?

“After the searing criticism of its performance in a number of recent cases, and a change of leadership imposed upon it by the government, the CCRC has recently shown signs of improvement in some areas of its work. In the past it has let Jeremy down badly. Now it must show it can address these previous failings and refer his case back to the court of appeal for them to rule on.”

A spokesperson for Essex police said: “In August 1985, the lives of five people, including two children, were needlessly, tragically and callously cut short when they were murdered in their own home by Jeremy Bamber. In the years that followed, this case has been the subject of several appeals and reviews by the court of appeal and the Criminal Cases Review Commission – all of these processes have never found anything other than Bamber is the person responsible for killing his adoptive parents Nevill and June, sister Sheila Caffell and her two sons Nicholas and Daniel.”

A CCRC spokesperson said: “Mr Bamber has made three applications to the CCRC. The first application led to a referral to the court of appeal in 2001. His conviction was subsequently upheld. The second application to the CCRC in 2004 did not result in a referral. A third application received in 2021 is now under review. A provisional decision on part of this latest application was issued to his legal representatives, and they responded to us in January this year. Work now continues to consider additional matters which were raised in Mr Bamber’s application.”