The Home Office has admitted that a review of how the Southport killer slipped through the net of its counterterrorism programme contained a key inaccuracy — a further sign of the shambolic handling of the case.
Its “rapid learning review” of Prevent, the counterterrorism programme, claimed that Axel Rudakubana was on the police’s radar six months before officers actually came into contact with him.
The rapid learning review was commissioned soon after Rudakubana’s deadly attack on a Taylor Swift-themed dance class on July 29 last year to examine how and why his referrals to the Prevent scheme were not followed up properly. Rudakubana, 18, is serving a life sentence with a minimum term of 52 years for the murders of Bebe King, six, Elsie Dot Stancombe, seven, and Alice da Silva Aguiar, nine.
The review concluded that officers had “prematurely” dismissed the threat he posed on each of the three occasions he was flagged to the programme between December 2019 and April 2021.
• Southport families say inquiry must focus on online safety
It also found that the police might have missed chances to stop the Southport attacker because of spelling mistakes. Prevent officers misspelt Rudakubana’s name on his second and third referrals, which may have led to the premature closure of his case because officers were not able to see his previous referrals on the system.
Yet the review of what went wrong itself contains inaccuracies that pose further questions about Prevent’s record-keeping.
It states that Rudakubana first came to the attention of the police in April 2019, when he would have been 12, after phoning the National Crime Agency to report bullying and admitting he had taken a knife into school. It then describes how the NCA contacted local police officers, who visited Rudakubana at home.
That account conflicts with Lancashire police’s version of events: they said their first interaction with Rudakubana came in October 2019, when he was 13, after he contacted Childline to make similar remarks.
After being asked to explain the discrepancy, the Home Office admitted its account was incorrect and blamed it on the speed at which the Prevent review was carried out. The review relied on Prevent case management records. Rudakubana was never in direct contact with the NCA, which only received a report from Childline about the teenager after that service became concerned by the comments he had made in October 2019.
The NCA then passed its concerns to Lancashire police, who carried out a home visit and made a vulnerable child referral to a collection of agencies referred to as Mash (multi-agency safeguarding hub). The Prevent review stated that no Mash referral “could be identified” two months later in December 2019.
The NCA confirmed that the Home Office review had inaccurately stated that the agency had been in contact directly with Rudakubana, saying: “The NCA became aware of information about Axel Rudakubana in October 2019. The information was shared with policing and Rudakubana was visited by local officers.”
Lord Carlile of Berriew, the UK’s former independent reviewer of terrorism, said it was “inexcusable” for public services such as Prevent not to get the basics right.
“Accuracy is absolutely essential,” he said. “Both at the time of dealing with the person concerned, and in reporting what happened to a wider public.”
Chris Philp, the shadow home secretary, concurred, saying that it was “important that a wider review of Prevent in these cases covers the accuracy of record-keeping and recording”.
He added: “As we have seen, these cases can be a matter of life and death so there is no scope for mistakes.”
The Home Office is preparing to publish its rapid learning review on Wednesday of the murder of Sir David Amess, whose killer, Ali Harbi Ali, was also flagged to Prevent before the attack.
It is expected to reveal similar failures to those in the Rudakubana case, including record-keeping errors that led to a breakdown in communication between Prevent officers and the police, and to conclude that inquiries into Ali were dropped prematurely.
There were signs of his possible radicalisation years before he killed Amess, the review will say, according to sources, including his sudden change of appearance and comments about the student loan system and modern music breaching Islamic law.
Learning reviews are carried out when an individual with previous involvement with Prevent carries out a terrorist attack or serious violence.
It is highly unusual to publish the reviews but ministers have taken the decision to do so in the interests of transparency.
However, Neil Basu, the former head of counterterrorism, said the errors highlighted why it was counterproductive to publish rapid learning reviews, which are usually internal documents, in full.
He said that while the Home Office should have published the review’s conclusions, the decision to share the review in full was “clearly made to avoid allegations of a cover-up, because that has become the conspiracy theory”.
He added: “I don’t believe mistakes in a rapid report are a cover-up; they may be genuine errors, which is why it is not good to publish a rapid report.
“But [regarding] the general conclusions of the report — that the systems weren’t robust enough; the system needs to be improved — it’s really important to do a rapid review for those reasons, to start examining the system. You shouldn’t wait for a full public inquiry in order to make changes to a system that clearly has deficiencies.”
The Home Office said the Southport review was conducted swiftly to identify what went wrong as soon as possible and relied on Prevent case management records.
It said a wider review being carried out by Lord Anderson of Ipswich, the interim Prevent commissioner, will contain more thorough recommendations to avoid a repeat of the failures in the Southport case.
