Recommendations

Recommendations

¶54. Many of the issues identified above are the subject of recommendations later in this report in respect of individual agencies or sectors. But by their very nature, fundamental problems about the relationship between different agencies cannot be resolved by action from those agencies alone. The ‘multi-agency’ approach in AR’s case completely failed to assess and react to the evident high level of danger that he posed. That failure means that detailed consideration should be given to restructuring fundamentally the way that our society handles those who pose a real risk of killing or causing serious harm to others (including ‘violence fixated’ individuals).

¶55. I therefore make the following three related recommendations.

¶56. First, I consider that there should be a single, dedicated agency or structure with responsibility for monitoring and co-ordinating any necessary information sharing or gathering and any intervention, providing direction, and holding the relevant agencies to account. The form will be a matter for consideration, but it could be through a mechanism for the appointment of a pre-existing agency as the ‘lead agency’; the expansion or creation of a dedicated multi-agency structure (for example MAPPA or akin to MAPPA); or the creation of a new agency. As I return to in Chapter 8: Prevent and Counter Terrorism Policing, it is right to note that, on any view, Prevent’s current statutory remit does not extend to all forms of risk of harm to others. A question that will need to be considered is whether this new mechanism, structure or agency would be separate from, or incorporated into, Prevent.

¶57. Among others, Ms Louise Anderson, Director of Children’s Social Care for LCC, gave considered and helpful evidence about the various multi-agency frameworks that exist, as well as on where responsibility for identifying and assessing risk lies.⁴⁴⁴⁴ T/S: 31/114/13-118/4, T/S: 31/119/20-122/12. In her view, FCAMHS was the principal agency with the training to be able to assess whether a child is a risk to another person. I have addressed FCAMHS in Chapter 10: AR’s healthcare, but I am sure that FCAMHS cannot be the entirety of the answer as to how to identify and address risk to others, nor are they in any way currently resourced or commissioned to do this. Ms Anderson’s evidence served to emphasise the significant flaws in the present system.

¶58. Further, as Ms Anderson explained and as I address in detail in Chapter 9: Social care, the multi-agency arrangements for child protection under Working Together to Safeguard Children, and the Early Help provision offered by LCC’s Child and Family Wellbeing Service, are heavily focused on preventing and removing risks to a child, rather than addressing risks from the child to others.⁴⁵⁴⁵ T/S: 31/77/6-16. Although Children’s Social Care ought to be involved in managing those latter risks, I do not consider that the Working Together to Safeguard Children framework has sufficient flexibility to be the primary means of addressing risks to others.

¶59. The other frameworks to which Ms Anderson drew my attention were MAPPA and MASP. But as set out above, I do not consider that these were appropriate frameworks for identifying or managing the risk posed by AR.

¶60. Ultimately, the evidence I heard amply demonstrated the pressing need for identification of a single agency or structure, and most likely an identified individual within that agency or structure, with responsibility for ensuring proper coordination between other involved agencies. This should include responsibility for the implementation of any plan to address the risks posed by individuals such as AR.

Recommendation 1: Phase 2 should consider what single agency or structure should be appointed or established to record, monitor and co-ordinate interventions for children and young people who present a high risk of serious harm. This must be matched with cultural change so that agencies are prepared to own and manage risk appropriately not just refer it on to others.

¶61. Second, and relatedly, this case has demonstrated the urgent need for a method of assessing risk which ‘talks’ across social care, mental health, education, policing and the criminal justice system. At present there is a lack of even a shared risk form, in which, amongst other things, the basic history is set out accurately, clearly and concisely. There was a wide range of exercises purporting to be risk assessments carried out, by different individuals at different stages, applying different frameworks and with varying degrees of rigour. Dr Tina Irani, the expert forensic child psychiatrist instructed by the Inquiry, made clear that what is required is a structured risk assessment tool to help with thinking about risk over time: neither purely actuarial nor ‘gut feeling’ risk assessments are adequate in this context.⁴⁶⁴⁶ T/S: 26/17/22-21/19. Ms Stephanie Roberts-Bibby, of the Youth Justice Board, was clear that the existing structured risk assessment tool for youth offending was both difficult to work with and in need of updating, with the development of a new tool needing collaborative work across a range of agencies:⁴⁷⁴⁷ T/S: 32/32/15-35/16. I agree, but such a tool needs to be used not only for those children or young people who are in contact with the youth justice system, but any child or young person who poses a serious risk of harm to others.

¶62. Ultimately, there needs to be a clear mechanism established, potentially with a grounding in statute, to identify and assess whether a young person poses a risk of causing serious harm to the public. This assessment should be the result of multi-agency involvement. The agency responsible for this work needs to be identified, and the level and detail of the assessment agreed. That agency should undertake or co-ordinate this work and ensure that it is completed. Furthermore, the responsible agency should ensure the appropriate sharing of information, to address the persistent issues over information sharing. Dr Anthony Molyneux of Alder Hey CAMHS advocated a “unified system” which would enable public sector agencies to access each other’s information.⁴⁸⁴⁸ T/S: 24/88/2-89/18. That is likely to be unachievable for a variety of principled and practical reasons, but a single risk assessment tool, under the co-ordination of a single responsible agency, would be a proportionate way of reaching the same position. Once a risk has been appropriately identified and assessed, where necessary there should be interventions in place to address it.

¶63. The absence of a common risk assessment mechanism means that there is a significant gap in the ability of the public services to provide protection for the public from those who pose a real threat of serious harm.

Recommendation 2: Phase 2 should consider the development of a shared multi-agency risk-assessment tool that is clear, accessible and suitable for use across public sector services.

¶64. Third, as I have already indicated, Merseyside Police’s post-attack investigation and this Inquiry have revealed the degrading, violent and misogynistic material that AR was accessing from home. He was able to access that material and use his home internet to purchase an arsenal of weapons notwithstanding the concerns strongly raised by The Acorns School as early as November 2019 about AR’s internet use.

¶65. Important steps have been taken in the United Kingdom via the Online Safety Act 2023 and strengthened age verification measures around online knife purchase. Other countries have taken, or are actively considering taking, a variety of other measures regarding children’s safety online and related mental health concerns. The agencies involved with AR – and his parents – could and should have done more within their existing powers to influence and supervise his online behaviour. However, I consider that there is a pressing question about whether further statutory powers to monitor or restrict internet use are required. Any such powers should in my view be tightly focussed on children/young people who have already been responsible for significant violence and show signs of a fascination with violence being fostered online.

¶66. I am, of course, alive to the issues of liberty and freedom of expression that characterised the careful balance that had to be struck in the Online Safety Act. However, echoing remarks made by ACC Winstanley at the end of his evidence, currently children, “at the click of a button […] can see the most horrific and horrendous incidents and they can go online and purchase things which […] I can see no legitimate purpose for, in crossbows, in machetes”. In 2019, AR had armed himself with knives on a significant number of occasions, and carried out a premeditated serious assault, confessing to an intention to use the knife on his intended target if the hockey stick did not inflict the damage he wanted. When AR’s parents were unable or unwilling to set boundaries at home, many observers would think it inexplicable that agencies were not empowered in such a case to enforce an appropriate level of control or supervision of his online behaviour in the years that followed, while AR was instead able to view graphically violent material and buy weapons online. Insight into what AR was doing online would have revealed the true level of his dangerousness.

¶67. Both the implications of such further powers (even if the threshold for their use is a high one) and an appropriate mechanism to deliver them were far beyond the scope of what could be considered in Phase 1 of this Inquiry. They require further and wider consideration in Phase 2 of this Inquiry.

Recommendation 3: Phase 2 should consider whether there should be a further ability to restrict or monitor access to the internet on the part of children and young people, if a significant threshold is passed concerning the risk they pose to others.

¶68. I will wish to examine how such an ability might, where necessary, be supported by a court-imposed order, including how any such order could relate to individuals within the Prevent programme.