I welcome our witnesses to the third evidence session in our inquiry on new forms of extremism. We have had some really interesting evidence so far and we are looking forward to hearing from both of our witnesses today. Could you introduce yourselves, please?

I am Leo Ratledge. I am the co-director of the Child Rights International Network.

I am Dr Jane Horton. I am an educator and a researcher at Liverpool University.

Thank you very much. We have a series of questions, all helping us with the background that we need for this inquiry. We will start with Jo White.

We need to be cautious in assuming that the increase is as significant as it might appear, particularly looking at the 2024-25 statistics. I certainly agree with the Bingham report, which I understand you will be hearing about later today. That says that practitioners are seeing more and more young people with complex and multiple needs coming forward. On the increase, the statistics from 2024-25 show that there was a surge of referrals in the first quarter in light of the Rudakubana inquiry and the interim guidance for practitioners on the referrals of violence with certain individuals. There was also the instruction from the Prevent referrals forum. They make that look like more of a significant increase than is necessarily the case. Prior to that, the referral figures had been quite steady from 2015 to 2023. It is interesting when we look at the high referral rates in the education sector since 2015 and question why they are such high rates. Teachers are under a huge amount of pressure to monitor and report young people in the context in schools when they are extremely busy and there is lots of high pressure to do safeguarding, which creates a great amount of anxiety among the teaching profession. Prevent has now become very embedded in all the safeguarding systems and procedures and it is in the key directives in teacher standards, in Ofsted and QTS. It is a huge priority for teachers in all sectors of the education system. Day to day they have to log their concerns on safeguarding software, and they are expected to log any little observation that they make in the classroom, often on to a system called CPOMS. There is a tendency, as Lord Anderson observed, of adopting a “better safe than sorry” approach. From a teaching perspective, that creates great anxiety. They want to do the right thing. They don’t want to miss anything and are frightened of missing anything because the consequences could be so great. My research showed that teachers are in a heightened state of alert because there is the Prevent training and the incidents in the wider world. This makes them hypervigilant to the indicators of radicalisation, again the fear of missing something. They are in a state of high alert. I think all these factors lead us to understand the large numbers of referrals that are coming from the education sector, particularly bearing in mind that only 7% to 9% of those are then adopted as Channel cases. It is a very low percentage. It was 17% in 2025 but we have said that the figures there are a little bit different because of the change in the methodology.

I agree with what Dr Horton said. We have seen increases in certain areas, but I think we should be careful about characterising that as an increase in extremism across the board relating to children. There are some trends that we can look at. We have had the most recent Prevent data for a year, for example, which showed a very sharp spike in right-wing extremism for the referrals. There are certainly trends within the data where we can see that certain forms of extremism are becoming more prevalent, but I think we must be cautious about looking at that as an increase, per se, and rather look at the phenomena within that. I know that the Committee has heard some evidence about com networks, for example, which is a specific form that grooming and exploitation can take with children. Elements of that are a new phenomenon that we should be aware of, but it is not necessarily the case that extremism among children is rising in general.

Certainly. This is not the centre for our work, particularly not for counter-terrorism measures and how they impact children, but we have seen it with sexual exploitation and abuse of children. These are quite closely-knit networks that operate across social media but also across private messaging services. They are in some ways communities that are taking place in an online space. These kinds of communities existed before but the way that they use technology and operate through technology has changed, so there is an underlying phenomenon that is consistent.

The whole issue of vulnerability is very complicated, and it is a term that is hotly contested in academia. In the Shawcross review, that language was somewhat changed to consider more susceptibility. One of the issues that has come to the fore recently has been about children with complex and multiple needs, and for the first time in the Prevent referral data we have seen mental health and neurodiversity recorded. I noted that 34% of the referral figures have one mental health or neurodiverse condition, which is an over-representation when we look at the wider population where about 1% of the population would be classified as having a neurodiversity. We must be extremely careful to not assume either correlation or causation with that data. This year Ofsted unfortunately released some training materials through a freedom of information request that came from Rights & Security International. In the training materials it said that children and young people with autism were at greater risk of being susceptible to extremism because they are more likely to develop special interests and use the internet to find friends. Some very important research by Alice Siberry from Vulnerability & Policing Futures says that counter-terrorism practitioners often believe that autistic traits, such as fixation or socialisation, might make people more vulnerable to radicalisation but this is surface level knowledge, and it overlooks the wider systemic factors and the way algorithms operate to draw people in. It ignores the lack of provision that is available for young people who are neurodiverse. There is a real risk of those vulnerabilities being misinterpreted as extremist intent, which I think calls for greater training for practitioners in neurodiversity, not just in policing but in education and those sorts of supporting services.

Dr Horton, I want to test if I misunderstood what you said. Are you saying that because teachers and educators are very strongly encouraged to be on the lookout for Prevent indicators, that is driving the increase in the statistics? If I am right in saying that, how does it compare to periods when extremism was very high up the public agenda just after 9/11 and just after 7/7? How is it different now than then, if you see what I am driving at?

Yes, I think that events in the wider world cause this hypervigilance. I conducted my research in the wake of the Manchester Arena bombing. That had somewhat passed but there was still the sense of being alert to what was going on in the city, what do I need to look out for and a state of anxiety. That was identified in Lord Anderson’s report in the wake of the Southport attack. I think that certainly puts teachers under a huge amount of pressure to make a referral.

Is it higher now than it was in the aftermath of 9/11 and 7/7? I know the statistics will not be comparable because Prevent was not around then but how do the statistics compare?

It is very hard to gauge it because the recording only came about in 2015, so I don’t think there was anything collected that would give that indication. There was no formal mechanism for recording it anyway, but prior to 2015, concerns referred through schools were just in the hundreds rather than the figures that we have today.

Picking up on what Chris just said, events promote hypervigilance. Do they also, though, promote more people into extremism?

Young people? That is not particularly my field. I don’t know if Leo would like to say anything in relation to that.

I think the short answer is that we don’t really know. There is a quite complicated series of drivers for what an individual is attracted to. We occasionally see copycat claims. That is certainly a feature, but I can’t say whether it is a statistically relevant one.

One of my colleagues will ask in detail about the internet but I want to ask a preliminary question on com networks. Some of the evidence we have had has said that com networks should be treated as extremism. Others have said, “No, don’t treat it as extremism, it is different”. What is your view on that? Secondly, do you know what percentage of the referrals are com-related?

The data is not sufficiently broken down for us to identify what percentage of the referrals are com related. Com networks are not necessarily tied to counter-terrorism methods. We would not expect all of them to be showing up in Prevent referrals, although there are several categories under which the Prevent referrals are organised that could overlap. It is not easy to identify that from the public data. It may be that the Home Office has more, but it is not in the public domain.

Dr Horton, do you want to add something?

No, it is not my area of speciality. Thank you.

Hello. I am Peter Prinsley. I am the MP for Bury St Edmunds and Stowmarket. I am interested to know how you think about this at a constituency-based level. What am I to say to a constituent on the doorstep if I am asked “What are you doing about all the terrorism we have got in our schools?” Do you have any information about what the efficacy of a Prevent programme is in a constituency like mine, for instance?

The data is not broken down so granularly. Regionally, yes, and regional Prevent practitioners would advise on what the risks are for a particular area. I am not sure it would come down to that level of detail, but I can find out if you are interested in that.

I have the sense that people think that there is a whole process in place here that has been very successful in preventing all this stuff. Are you able to say how many episodes of terrorism have in fact been prevented? Is there any way of knowing that?

There are several sources that you can look at to try to answer part of that question. In general, Prevent has not been good at publishing data about the effectiveness of its programmes. Some limited information has been published. The Lord Anderson report looked a little bit at some of the disengagement programmes that have been conducted through prisons, for example. This is information that we would very much like to have but don’t have. One of the criticisms that has particularly come from civil society about the Prevent programme is that the scrutiny has not taken place in a public way that allows scrutiny of the effectiveness of individual programming.

For a programme like Prevent to have public support, it would need to demonstrate its effectiveness rather than we sit here supposing that. Being able to somehow come up with a figure to say, “We have prevented 90% of all possible—”

I agree entirely.

That is missing.

It is a missing piece of information.

Yes, absolutely.

Could you ever do that? Could you ever prove that what didn’t happen was due to Prevent?

I imagine if you called the security services to give testimony, they would be able to point to individual attempts that they have blocked at various stages. I understand that that would not be in the public domain, but they will certainly have that information. On individual programmes, you could certainly design evaluation processes for how it influenced certain characteristics that might leave someone prone to engaging with extremism, counter-terrorism and so on. It would be possible, and I assume that there have been some evaluations that have not been published.

I think that the Intelligence and Security Committee would look at that and have access to more of that data because of the nature of the work it does. It is not in public.

But the public are interested in it.

Yes, absolutely.

To follow up on some of the stuff that has been mentioned already, do you think there is sufficient publicly available data for people outside of government agencies to understand what is happening in communities? Do you think the data is robust and comprehensive enough for us to be able to tackle extremism, either to prevent it or to intervene?

Speaking from the practitioners’ perspective and the teaching point of view, they have said to me that when they are making a referral they often feel as though it disappears into a black hole, “I don’t know what happens”. They have said they don’t know the consequences of the referral that they make, and that makes it very difficult to learn from their experiences and what they have done. With issues of confidentiality, it is hugely difficult to share that information, but I think anything that gives a better quality of information and makes it publicly accessible, including to professionals like teachers, is a good thing and is much needed.

As an educator, do you feel that you have a sufficient understanding of where radicalisation is happening and where the real danger points are, so that teachers are given the tools to prevent? I don’t mean Prevent, with a capital “P”; I mean to prevent.

A local authority Prevent officer would probably cascade local profiles of risk through to a safeguarding lead in a school. Whether that filters down to individual teachers would depend on whether it was felt that it was necessary.

The data is improving but there is still not enough in the public domain to understand the effectiveness and the extent to which phenomena are changing. A lot of teachers—and I am very interested in how this shows up in your research, Dr Horton—would say that the training has been quite limited and it is quite common, certainly unless a teacher is a safeguarding lead, for them to take the training once and then not to keep up to date with changing evidence. I think that is a significant feature and I don’t think that is the teachers’ fault at all. It is partly a product of the range of safeguarding obligations that fall on them with different mandates that seriously complicates the issue for them.

You don’t think it is because there are gaps in the data? You think it is that the data collection is not being communicated and shared for maybe issues of national security or whatever?

I think it is a combination of both.

I will turn now to identifying and intervening on extremism behaviour, particularly with young people. What do you think are the most effective strategies for identifying signs of extremist behaviour or radicalisation? I suppose that is on the spectrum of early intervention but also moving up towards the Prevent indicators that would trigger a formal reporting process, just identifying the signs of extremist behaviour.

It is an incredibly complicated thing to do. There are multiple sources of information that provide checklists of signs of radicalisation. They are contested. They are often signs that can equally be interpreted as normal teenage behaviour, which causes confusion. The starting point— the vulnerability assessment framework as it was and the Prevent assessment framework as it is now—is a very difficult tool to navigate. If we are thinking about how to improve the quality of the referrals and where to intervene, it is almost like we need to go back a step and build trust between the teacher and the pupil so that those relationships can be nurtured in the school environment, so that a teacher knows the pupil and can genuinely identify when something is wrong. For me, that is the starting point that needs to be at the heart of any changes that are introduced.

Do you think there are sufficient ways of triggering that process or teachers being able to identify concerns without going down the adversarial official route?

Teachers do an awful lot of that in schools. There is an awful lot of what you might call pre-referrals that never make it out into the wider system but are handled within the school. I saw some excellent examples of that in my research, conducted by teachers at a pupil referral unit, who are extremely skilled at safeguarding, working with the most troubled learners with multiple and complex needs. They can have that kind of dialogue with the pupils, answer difficult questions and prevent a referral going forward to Prevent.

I agree very much with what Jane said. The point about taking a step back from immediately trying to identify extremism is incredibly helpful. If there is an increased focus on the features that make children vulnerable to a variety of negative outcomes, we can in some ways ensure that they get the service at an earlier stage, whether that is special educational needs, which are in short supply in some schools, or about the huge backlog in child and adolescent mental healthcare. I was alarmed by the latest data produced by the Children’s Commissioner for England with 40,000 children waiting more than two years for a mental health referral. I think all these features can leave children quite vulnerable to other negative things happening afterwards. By taking it right back to basic safeguarding, we can short circuit a lot of the negative things that come afterwards.

On moving on the route of safeguarding to counter extremism, are there identifiable patterns on how young people move from what you said was typical teenage behaviour or curiosity into something more sinister and dangerous?

The basis of assessing radicalisation, which lies at the heart of Prevent, is still quite contested from an academic perspective. The idea that lesser behaviours necessarily graduate to more serious is not supported, necessarily, by the evidence that we have. Trying to look at the stages at which that occurs is a very difficult analytical exercise. I am sorry I don’t have a simple answer on that, but the answer is that it is very difficult.

I agree wholeheartedly with that. That tipping point has not been established in research. There are multiple pathways to radicalisation. It is not a single pathway. A lot of the current research that we see—I am sure you have seen as well—is moving away from the emphasis on personalities and the individual to looking at much wider drivers of radicalisation. Research conducted by the University of Liverpool, consistent with a lot of other research that has been conducted, looks at social exclusion and polarisation as drivers of radicalisation. Evidence submitted by Professor Abbas echoes that. Practitioners say that lack of resources and Government disinvestment in communities are important contextual factors that contribute to young people gravitating towards extremist ideologies. The causal link and the point of where they cross over from extremist sentiments and thinking to wanting to commit an act of violence is just not known.

The only thing I will add is that in many ways if children are engaged with other safeguarding mechanisms at an earlier stage, they will have more contact with professionals who will be able to see when something is changing the child. Good relationships with professionals who work with children are essential. It is the only way that we can really see when something has changed and a new response is necessary.

I want to move on to children’s access to the internet and unfettered access to social media. You mentioned that some algorithms are particularly risky for those with certain psychological problems. Do you think that what is now a complete free-for-all in children having access to the internet has increased the referrals to Prevent and actual acts of extremism? Do you think it is a contributing factor?

Certainly, changes in technology have altered the phenomena that we see. It is very difficult to draw a bright line between what is happening in online space and what is happening offline. Most children don’t necessarily see a neat divide between a digital world and a real world anymore. Their relationships are spread across both and the way they communicate and seek information is across both. We have seen negative impacts from algorithms that can feed children’s access to harmful information. We have seen moves on some of the large platforms away from moderating content, which makes it more likely that children will see damaging and extreme content. This is undeniable. A causal effect is more difficult to establish and, if so, how that operates but we certainly see it in the way that children are experiencing this kind of information.

How do you think children can be best supported to navigate the internet safely? Do you think the Online Safety Act—given that may be too left field—is sufficient in protecting children from extremism?

There are some excellent resources out there that help teachers support young people to navigate the internet safely. The suggestion that is made in the curriculum review to focus more on life skills will help enormously, if time can be given to support young people to do that and navigate these platforms safely. One of the things that has happened is that these are presented as a negative space, but young people are using those spaces for a reason, and they are offering very pro-social opportunities to engage with like-minded individuals. It is recognising that and obviously helping them to navigate it safely. Organisations like Educate Against Hate, the Association for Citizenship Teaching have got superb tools that they provide to the education sector to help them do that, and for parents as well.

I would say that, because the phenomenon touches every part of their lives, there is not a single intervention that can respond to it. Effective regulations are a key part of that. The UK is choosing to do this through the Online Safety Act, which is at a partial point of implementation—I would say at this point there are several ways that Ofcom will be developing new provisions on that, so we are part way through that reform. The other side is education of children. The access through schools is the strongest opportunity to provide education on a nationwide level that we get, and we do need some improvements in that area. There are some excellent examples that Jane points to, but we need to be looking at how children can engage critically with what they see online, how they respond, how they know how to respond when something goes wrong, who they reach out to, and how they deal with complaints processes. The incorporation of complaints processes through regulation is mutually reinforcing when it comes to education. You also need to look at education for parents. It is a much more difficult target but a lot of the support that children will be looking to receive when things go wrong will come through parents, caregivers and families. Finally, other professionals who are engaged with children. So, those four elements are important. Because things are moving so fast, all that needs to be implemented with a significant investment in evaluation—what is working, what is not—and an iterative process of improving that.

I appreciate your point about causation, but you have both acknowledged that children have more access to extremist content than ever before. Do you think there is a case to limit children’s access to the online world and to social media in particular?

I would argue that children and young people are often one step ahead of us, so any attempts to limit or curtail their freedom online will be met with a way around it; a way of navigating that limitation.

I would agree that we need to be quite cautious about assuming that limits, in particular legal limits, will be effective on that front. There is a place for regulation of social media. Through GDPR we have a digital age of 13 for controlling data, which impacts at what stage children can consent to use social media.

That is quite young.

It is quite young, but one thing I would say with regards to that is that if we set an age too high, we are at a risk of creating a cliff edge whereby children are not supported to learn to use tools in a safe way. Some of the proposals around raising that to 18 would carry some very significant risks once children are out of that supportive environment for how they learn to use these tools.

Can I just push a little bit more on the Online Safety Act? Mr Ratledge, you were very diplomatic in terms of saying the implementation is ongoing. There are some strong views in this place that Ofcom is not perhaps going fast enough or robustly enough in implementing them. On the flip side, we have a Parliament discussing a petition of half a million signatures in a week or two’s time asking for the Online Safety Act to be repealed all together. Can you be clear in your view of do you think the Online Safety Act, if implemented robustly, could or will reduce risks to extremism for young people?

It is very difficult to say. When the Online Safety Act was passed it kicked a lot of questions down the road for the regulator, and now the regulator is trying to implement them, with partial success. In all honestly, I cannot say that it has been a whole success, but the reason that progress has been so slow is that what has been asked of them is an incredibly difficult task. As an organisation, we do not have a view about whether a repeal would be a good idea, but the challenges that face Ofcom now would exist were Parliament to try to enact a new Online Safety Act tomorrow. It is more a matter of where those decisions are made rather than how they are made.

Whatever you think of specificity of the Online Safety Act—you talked with my colleague about social media bans in terms of the Online Safety Act’s principle around putting proactive responsibilities on online content platforms of taking responsibility for the way that their algorithms and the impact that their content has. Is that overall approach the right one and one that is suitable and is important in terms of combatting extremism behaviours?

Yes, that is a good approach—I would agree with that. That, coupled with strong educational support and skills developed among young people, will be a powerful way of helping to build resilience in young people.

You have both mentioned the link between mental health and the significant waiting times for mental health services and Prevent referrals, and this is something we are going to ask of subsequent panel, but, as you mentioned it, do you think that in some cases people are being referred to Prevent as a way of shortcutting waiting times, which I think the Shawcross review of Prevent suggested was the case?

We have seen anecdotal cases of this. It is not possible to identify at what scale it is happening through the data that is available to us, but we are familiar with examples of exactly that happening.

I would agree with that. I have had that said to me in conversations, anecdotally, and it is understandable because people in safeguarding, people in education, want the best for young people. They are not doing that to misuse public money or to circumvent systems. They are doing it because they are trying to get the best outcome for young people. If that means going through Prevent to fast-track support, that will be done.

When I was a child, it was a simple matter that we were simply told not to talk to strangers. That was how our parents kept us safe. Do you think that we ought to be thinking about how we mandate within primary schools some sort of online safety education, so that all the children are told—everybody is instructed—about the dangers of the internet and the sort of things that are safe and not safe for them? If we were to mandate that in primary schools, we would obviously need to give the teachers the tools to do this, but it seems to me that at the heart of this is telling our children how to behave. We were told how to behave, but then we did not have an internet that was so dangerous. Rather than telling the internet companies how to behaviour, should we be spending a little bit more time educating our children? Is there any way in which that would be a practical suggestion?

I would agree. More education on this—as I said, with the curriculum review introducing more life skills—if time and space is given to those changes, they could add great effect and they could help children to develop the skills and the resilience that they need to navigate a complicated, online world that was not around when we were children. An awful lot of good work goes on in primary schools. When children move on to secondary school there is obviously good work there as well, however when children get their own free will and they develop their own interests, then telling them what to do becomes less powerful—

Which is why we need to do it very early.

Yes.

We need to encourage good internet habits at a very early age for these people.

The introduction of citizenship into primary schools would hopefully address that.

On that point, I have heard one defence of the Australian model: it is a nice binary, it is simple—as with do not talk to strangers—and teaching nuance at an early age is a fool’s errand and you need that simple rule that gives parents something to say, “You cannot do it”. Are you resistant to those ideas? I might be as well for that matter.

Difficult question. We must not assume that all children come from a background where the parents would have their best interests at heart. Obviously, that would be the ideal, but that is not necessarily always the case, and they may not have that influence at all.

I would be sceptical of the bright line Australian model. I imagine we will see a great deal of evidence about how it is implemented and what that looks like. It strikes me that it is impractical to try to prevent children using technology that will be accessible whether there is a ban in Australia or not. Whether children are breaking the law to access these services, they will continue to do so, and we must accept that as a fact of reality. I agree very much with what you said about having education at the earlier stages. We cannot treat simple regulations as a way of shutting down the realities of children’s experiences. The stranger example keeps coming up, but for me this is one of those that tells us that we need to have nuance because children are not most likely to experience abuse at the hand of a stranger: they are most likely to experience it from people in their family, people that they trust. If anything, it pushes us more towards needing to have a nuanced, critical look at how we respond.

In his review, Lord Anderson recommended that Prevent be connected to other safeguarding and violence prevention strategies. Dr Horton, you have already mentioned that that is happening in the informal pre-Prevent stage. Is there more to be done there?

The Bingham report suggests moving Prevent into a separate sphere so that it just deals with individuals who are likely to progress onto terrorism and that other violence fascinated individuals, for example, would be dealt with in a broader safeguarding. So, rather than broadening the scope of Prevent, I would say narrow the scope of Prevent but improve wider safeguarding for the other concerns that we are seeing so much of in the education system and in wider society.

Does it require a strategic approach where you work out which of these strategies are being used so they are not counterproductive?

In our view—and this aligns with some of the recommendations that came out of Lord Anderson’s report—we need a more radical overhaul of safeguarding in general, in that if the concern of, say, a teacher in the education system, is that, “I can see that this child has particular needs and vulnerabilities, but I do not necessarily know what they are”, that that can go to a safeguarding team that has the relevant expertise and can decide what the next steps are, without that being a Prevent referral that we know will usually not be the final outcome for a referral, because 90% of those referrals end up going somewhere else or being discontinued. What this would do would be focus more on how the different services need to work together without skewing that towards a counter-terrorism model. Although it does fall within broader safeguarding frameworks, in many ways Prevent departs from safeguarding in quite important ways. It is not a model that is fundamentally based on assessments of the welfare and best interests of a child, but it is skewed by the intervention of policing. It is, in part, an investigatory process as well as a safeguarding process. If the starting point is a safeguarding referral—and it may be that through that information collecting, gathering, discussion process, it becomes clear that there is a CT need, a referral could take place from there. The benefit of that would be that it would massively reduce the number of false positives—as is their terms in the literature—children being brought into a counter-terrorism process that is inappropriate and that they find stigmatising, and that then makes them reluctant to engage with other services in the future because of the breakdown in trust.

Four-fifths of referrals do not make it to a Channel discussion, and five-sixths do not get a Channel intervention. Is there a formal process as to what happens with that vast majority?

There are a number of points where a child could be redirected to other services. In lots of cases, there will be a number of cases that are first triaged by counter-terrorism policing where there is no counter-terrorism engagement, so the process stops there. It may be that in schools or other settings that is taken up within other safeguarding models—that is not clear from the data. There are a number of other points throughout the process that, again, could lead to other referrals. It might be that part way down the line a mental health need is established as necessary and that a mental health referral takes place. It is in those kinds of cases that we might expect to see a shortcutting of mental health waiting lists, for example. It could be that it reaches a Channel panel and there is no consent to continue with an intervention. Again, that would end the process because all interventions are consent based within the Prevent programme.

One of the things that we need to remember is that if all this time children are being viewed through a lens of suspicion, that could be hugely damaging. Whether the referral gets progressed or not, even for a referral to be thrown out at the very early stages, that child has still been officially recorded. To be scrutinised like that—to be put on that pathway to a referral—could be hugely damaging, not just for them but for their family as well. I have seen some very good examples that I can mention in pupil referral units, for example, where that has been offset and diverted away from to protect that child. I would argue that there is a lot of that going on in education, that teachers are doing their best to protect children from a negative referral.

So, the large percentage of false positives on referrals is a real danger, is it not?

Yes.

What can we do to stop that? Do educational professionals, teachers, have the skills to work out who they should be referring?

There is a lot of evidence that suggests teachers say they are confident with the skills that they have. They gain confidence from the training. When we drill down into it, the confidence is that they know how to make a referral; they know how to follow that procedure. Something that came out of my own research is that what they were less confident about was what indicators they should be looking out for: what are the worrying signs of concern, when to make that referral. It is that more detailed understanding that is perhaps not as evident. Something that was mentioned is that with training, if it is online training, it is often given at the end of a busy school year, it is something that teachers are expected to rush through. So even though Prevent is a hugely serious concern, they are expected to rush through it, get it done to a deadline and to then show the management that they have completed the training and ticked that off. It is about time, opportunities for teachers to have the chance to do this better—they want to do it better—rather than it being rushed at the end of the year. I would argue it needs more time.

So, they know how to do a referral, but not why, a lot of the times?

Yes.

Given that there will always be the issues with time for training and teachers have a huge number of responsibilities, and being risk-averse of not making a mistake, is there something we can do with the referral system so that it does not stigmatise in the way you are telling me that it does?

One year on after the Shawcross review, I understand that the right was given to retain the data for six years—I can check on that—but it was questioned whether that should be the case. Even if a child’s referral is not progressed through the system, it can stay on their record and teachers are aware of that, which makes them very wary—it can make them wary about referring into a system that would do that.

There is an element that in so far as these referrals are taking place within a counter-terrorism framework, it is very difficult to remove the stigma that comes with that. That can particularly be at its most damaging when children feel that they have been wrongly referred in part because of a misunderstanding of an element of their identity, culture or religion. A merging of those senses of what is very important to them as a person within a counter-terrorism framework entered appropriately is always going to be stigmatising.

Is there an argument for not having counter-terrorism for that stage—one recognises a child that is vulnerable and not bringing in this Prevent approach at that point—or is that naive?

Yes.

That is the view?

In my view, yes.

Yes.

If the starting point was a safeguarding assessment based on vulnerabilities and needs of a child, that was separate from a counter-terrorism process, and that if counter-terrorism concerns became clear through that. The stage would happen first that was child focused and that would likely reduce the stigma of the experience.

What percentage of children who do go to the channelling point, engage with it? Do we have figures on that?

We do. I do not have them in front of me, but I would be happy to write to you if that would be useful.

In broad terms, do most engage?

It varies a little by age group. There are a sizeable number who decline to be involved.

Do you know what factors influence whether the child engages?

So far as I am aware, that would not be available as part of the public data.

We heard from previous evidence sessions with the previous Home Secretary and Lord Anderson about the benefits of Channel—how children who were going down the wrong path were supported through that. For some children it ends up being a diversion onto a more positive outcome. You have raised concerns about the negative consequences of being referred to Prevent. I can understand if someone has misunderstood the signals that they thought a child was saying something and they weren’t, but assuming they are correctly interpreting and are right to have safeguarding concerns, how can it be negative for a child to be referred to Prevent?

What I would say specifically about Channel is that it takes place at the very end of the process and to take part in a Channel programme, the person involved must consent. The group of people who have gone through Channel are people who want to and, therefore, you are more likely to see positive examples come out of that. What did not feature in Lord Anderson’s report are the false positives—they were not addressed within the report at all. The terms of reference make sense for why that was the case, because it started as an analysis of two cases that had gone wrong with catastrophic consequences.

Can you walk me through a false positive? Forgive me if I am being dense here, but does false positive mean that someone who had absolutely no extremist inclinations or expressions at all was mis-identified by the system? Or does it mean that someone who does have concerning behaviours but does not meet the threshold that would require an intervention?

It would be somebody who had been referred when there was not a counter-terrorism related issue at the core of their needs.

Around the stigma that we do identify with these children who are referred to Prevent, presumably during the era when Islamic extremism was the top public policy terrorism concern, one would imagine that it would be young men from Muslim backgrounds who felt particularly stigmatised by referrals. Does your research find that that is still ongoing? Do you also find that there is increasing stigma around other kinds of ideologies now—far right, misogynist? How do these stigmas affect these children?

Obviously the data has changed and the demographics of the concerns that are being referred through has changed over the years, but that legacy of the anti-Muslim bias is a long legacy, and it is very hard to shake off. It will take a while for that to be felt. If we look at the figures from this year, 19% of the Asian minority population referred in Prevent figures, although they only make up 9% of the wider population. So, there is still that over referral. This is the first time we have had Home Office ethnicity data provided, and it quite starkly stands out as being disappointing. It will take a long time for the damaging legacy to be moved away from.

Are we certain it is over referral? What is the male and female divide, for example?

Overwhelmingly male.

That does not necessarily mean that we are over referring men or male children, it just means they may be exhibiting more concerning signs. We do not know if that is over or under referral from a demographic group.

Far right referrals are more likely to be accepted by Channel than Islamis referrals, which would suggest that there is either—it could suggest any number of things—but it may be that there is a different threshold in terms of when professionals are referring; it could be about who is likely to take up offerings from Channel. We do not have that from the data. In terms of cases adopted, we do see differences between different groups.

Is there not a fundamental difference between counter-terrorism and extremism? Do you think that our interventions are too focused on extremism or looked at through the lens of counter-terrorism? In the absence of an extremist strategy from the Government, is there a case or an argument for widening out and recognising the impact of extremism on children and wider society outwith counter-terrorism?

I would argue, yes, and it goes back to what we were saying before that to go further downstream through education to tackle the drivers against—

You think there should be a counter extremism strategy that is led by Government?

We have some hesitation around grouping of different phenomena as extremism. I would be cautious about treating all of them as if they operate in the same way. What we really need is more granularity. It has been one of the criticisms of the expansion of the Prevent policy to incorporate other forms of extremism that would not be counter-terrorism is that it does not necessarily tailor the response to the threat. I would be cautious about grouping different phenomena without deeper analysis.

I want to pick up on something that I thought you said—I was not quite sure. Did you say a minute ago that the likelihood of being referred for the danger of far-right activity was greater than the likelihood of being referred for the danger of Islamic?

Yes.

Do you have an explanation for that?

The most recent data has shown a very rapid spike. There have been more than twice as many referrals for the far-right in the last 12-month period. That is a sharp increase. I could speculate about why that is, but we do not have—

Do you think that there is some sort of intrinsic reluctance to make a referral where the fear is that of Islamic radicalisation for fear of being seen to be prejudiced?

I do not believe that to be the case. I know that Dr Horton’s research was focused on how teachers experience the referral process, so she may be better placed to speak about it than me.

In Lord Anderson’s report, he said that the referrals from Muslim majority schools were rare, and from Islamic faith schools were practically unknown. He interpreted that as reflecting suspicions among Prevent from the Muslim population. My interpretation would be that it reflects a better knowledge of mainstream Islamic cultural practices. It is not that those practitioners are suspicious of the system, it is that they understand what a concerning indicator would be. That is an argument for building those things into training for all teachers, to understand what the difference is between mainstream religiosity and cultural behaviours and extremist ones.

The question that I asked has a particular political resonance, as you probably understand. I can well understand that referrals from Muslim and Islamic schools would be less, but is Islamic groups within mainstream schools compared to, say, referrals for far-right activities? Would you see any contrast there?

I would argue that most teachers are professional enough to override those concerns. Safeguarding training is very robust. Equal opportunities training is very robust. They would not be concerned with those.

We have reached the point where we are supposed to be letting you go. If you are okay to stay for another five to 10 minutes, we only have a couple more questions.

Just on that point, surely the issue is not just referrals, the issue is those that get through channelling. Did you not also tell us that a smaller percentage of those referred on Islamist grounds are channelled than those on right-wing grounds—did you tell us that earlier?

Yes, right-wing cases are more likely to be adopted as Channel cases.

Is it not contrary to what was being suggested that there is in fact a greater likelihood of referring people for Islamism when there is no reason to than for right wing? Is that what the evidence shows?

It would only be consistent with the data; it would not be the only explanation.

My next question is on youth diversion orders, which is proposed are going to come in. What is your view on youth diversion orders at this stage?

One of the criticisms of how diversions operate within a counter-terrorism process has been that there is relatively little diversion compared to what would exist for other offences. In principle, it is a very helpful addition. What I would add to that is that there will be a number of safeguards that will need to be in place to make sure that it is an effective measure. The first is that it will be very important that a child has legal advice through the process, in part so that they understand what is happening to them, but also so that any restrictions that are placed through the diversion order are properly tailored. It will be very important that it is not treated as the only tool to be used in this setting, and that it does not sideline other measures of informal diversion that might be possible. At earlier stages, we will need some very clear criteria for how they are used, guidance in terms of how they are implemented, and I would be very keen to see research and evaluation within a year or two after they are implemented. We are tentatively hopeful that they bring something new to the space.

You said, “legal advice”. You would have a solicitor there?

Yes, depending on the setting that took place, it could be a solicitor, yes.

I can say this as a lawyer myself; lawyers do not always help the process when we are talking with something like that. Why is ‘advice’ not sufficient? Why would it have to be ‘legal advice’ that might become overly legalised—the whole process?

The diversion orders are designed to be commissioned within a court setting. A child navigating that, in whether to comply with a diversion order, would have legal implications for them, so it is crucial that they understand what they are when they are choosing how they want to engage.

We have heard previously, in other decisions, about the lack of evidence base around extremism and policy approaches. What do you think the priorities should be for future research in this area, considering the current gaps that there are? What could the Home Office and Government do to support that?

There is a necessity for more research into this area, particularly with young people themselves. One of the problems that we face is that universities are institutionally very risk-averse and getting approval for research is incredibly difficult. I had to change the direction of my research. I wanted to do that directly with young people and was strongly persuaded against that; it became too challenging to do that. It is essential that more research is conducted.

I would agree with that. I would also like to see more research about children’s experiences of the digital world and how that intersects with this issue. There is relatively little research that deals with the way that different groups of children experience online services. For example, we have seen very little addressing how children with disabilities experience the same kind of technology but in different ways, so there is a wealth of research to be done there. I would like to see more evaluations published publicly about Prevent and the efficacy of interventions. That would be an enormously helpful contribution.

What are the barriers to carrying out that research? You mentioned the research ethics of universities. Do you think a clear steer from the Government about intent or priority for that work would be helpful?

It would be, yes.

Thank you very much. We have only overrun by five minutes. I do apologise for keeping you longer but, as you can see, we have a lot of questions, and we are very keen to learn about this. Thank you, you have been incredibly informative and helpful. If there is anything else—I know you said you would write to us—but if anything strikes you that you think it would be useful for the Committee to know about after you have finished here, please do get in touch. We are still happy to receive further comments. I will draw this panel to an end. Witnesses: Rt Hon Professor John Denham PC, Kenny Bowie and Councillor Sara Conway.

Welcome to our second panel of the afternoon, some familiar faces for me on this panel. Perhaps the witnesses could introduce themselves and we will then go into questions.

I am John Denham. I am a former Communities Secretary and one-time chair of the Home Affairs Select Committee. I am a professorial fellow at Southampton University, and I was a member of the Independent Commission on Counter-Terrorism Law that reported a couple of weeks ago.

I am not quite as grand as that. I am Kenny Bowie, and I am the director of Strategy and Metropolitan Police Oversight at the mayor’s Office for Policing and Crime in London.

I am less grand than that. I am Councillor Sara Conway. I am a councillor in Barnet. I am here for the Local Government Association where I am co-chair of the Special Interest Group on Countering Extremism for members.

We are very keen to understand about what your priorities are at a local level. How much is tackling extremism a priority at a local level?

It varies. There is no specific statutory duty. There is legislation that cuts across it. For the LGA, it is a mixed picture across the country. We were talking outside of London—obviously that is the area I know best—and in terms of the local area that I am part of in Barnet; we have been funding that post for two years. The Government funding was pulled a couple of years ago. We think it really matters. As the LGA, there has been a big gap since 2020-21 when the community co-ordinators got pulled, and that is why the LGA has made the commitment to carry on doing some of the work that is there. It really matters for local government. We are the convenors of place; we know our communities best—MPs too, obviously—but councillors and councils, and it is an important role, particularly at the present time. There is a need to be doing things locally, but also nationally as well. There needs to be proper funding—you would expect local government to be saying that—for mental health and all the different components, and to track cohesion nationally as well as being able to do that locally. I am sure we will get into it more, but in terms of the online world, there needs to be a national pushback on disinformation and co-production with young people.

Do you want to add anything?

I would say that, at a London-wide level, it is a priority for the mayor, following the attacks in 2017, as well as the foiled ones. We did a huge city-wide consultation on all of this to see what we should do, as a result of that. That came back saying that we should set up a countering violent extremism team and that we should set up a new fund for civil society organisations to be able to bid into, not to compete with Prevent, but to complement it and work with the central government and local authorities. Both of those things were accepted by the mayor, and we have gone on over the years from strength to strength with the Shared Endeavour Fund. There has been £6 million invested in it over the last six years, of which roughly 25% has come from match funding, which is a good model. It does remain a priority in London for the mayor.

On the evidence that we receive nationally, it is the enormous range of issues that are now facing local authorities, the police and other agencies. The desire of groups like Isis to stimulate terrorist attacks has not gone away. We have the mixed, unclear, uncertain motives via this fixation. On the broader sense, with the asylum hotel disturbances—the disturbances after Southport—people are dealing with an enormous range of issues, generally much less well-resourced than they feel they need to be to deal with them effectively, so the agenda is growing, not shrinking.

In terms of how you balance priorities—because you have a lot of priorities—where would you say tackling extremism sits in terms of the priorities that you have to focus on? I appreciate you do have a lot of things to focus on.

Genuinely, it is right up there. We have a team of three that is dedicated to it. We invest £875,000 of money in it every single year. We chair, through the deputy mayor, the London CONTEST Board where we bring together both central government through, what was previously the Prevent director but now will be the countering terrorism director in the Home Office following some structural changes there, along with others. We do see it as a being a big priority. For example, we bring in experts to talk. We have David Anderson coming in to talk about his next report—or his current report—in a couple of weeks, to talk about violence-fascinated individuals, how we deal with that in London when London is the centre of 60% of the pursue threat across the country and it has so many iconic venues. We cannot afford for it not to be seen as a top priority in that way. National CT policing headquarters obviously sits within the Metropolitan police, and we have paid the ESB for a large chunk of investment to allow the coalition with some of our other operational partners there. We do take that role very seriously, so high importance.

I am Peter Prinsley. I am the MP for Bury St Edmunds. Thank you very much for coming this afternoon. This is about local approaches to tackling extremism. You have answered the question about how much of a priority this is given. Do you have sufficient resources at your disposal? If you were to be given additional resources, what would you do with them?

For the LGA generally, mental health is a key issue for young people. If it is all right, I want to share the UK Youth Endowment Fund that has come out this month. I presume you have seen it, but it really struck me. That talked about that in the past year, more than four in five 13 to 17-year-olds have seen conversations online about hurting, either physically or emotionally, specific groups, and over a third said they had taken part in those conversations, either to support or to challenge; and that 70% of 13 to 17-year-olds had seen real world violence or online. When you start to think about that, and what that means locally—I had a conversation with our Prevent officer yesterday, before coming to this—and you were saying this as well—but the level of referrals in quarter two of this year is unprecedented. I am sure that is after Southport as well, but there is all this other stuff going on too. The numbers I have is that in the first six months of 2025-26, just in one local authority—that is in Barnet—969 education professionals, LA front staff, commission partners received workshops, training, and so on. Also working with education, safeguarding leads, foster carers, representatives from adult and health. We feel that we have pretty good best practice, but across the country, it is patchy and needs proper resourcing. There used to be 35 community co-ordinators that were funded by central government back in 2020, and that was vital because, as we all know, these relationships and understanding develop with trust. You need those relationships in place, which local government is very good at doing, but needs to be able to resource properly, otherwise we are all just firefighting and we cannot put the proper structures and processes around this.

That leads to the next bit of the question, which is all about how a local authority goes about co-ordinating the activities of the schools, youth services, health and community organisations and so on? From what you have just said, it sounds like something that was there has been dismantled, am I correct?

It is patchy.

First, we did not do a scientific survey on the inquiry, but we have a lot of evidence of areas that used to have youth services that could be deployed as part of this work, where those services no longer exist. Secondly, you have areas where they are perhaps slightly better resourced, and they have managed to protect some of this work by raiding other budgets, because they used to have funding for community cohesion. That has gone, they have taken it somewhere else. Thirdly, people pile in behind Prevent, because it is the only service there. Prevent is meant to be there for a very specific problem of tackling terrorism. We had evidence from people saying they refer people to Prevent because it is the only way you can get accelerated support from CAMHS services for mental health issues. We are referring people that we know are not going to be terrorists, but they are possibly a danger to themselves and other people. It is distorting it in all those three different ways.

That leads on to the final bit of the question, which is about how we build trust in communities where there is probably widespread scepticism about whether any of this counter-extremism stuff is effective? How do we sell this to the sceptical community?

I do not know. We did not find anywhere that thought the communities would be sceptical. What we heard time and time again was we need a properly directed community cohesion strategy. So, 20 years ago, in April 2005, this Committee called for a national social cohesion strategy dealing with these issues. There has not been one in 20 years. To the extent there are cohesion strategies, it is because local people have done it. We talked about the Shared Endeavour Fund in London, things of that sort. People running local authorities, people doing this work, think that unless you build trusted relationships across groups in the community, so you have resilience, confidence, the ability to discuss difficult issues, you are always dealing with problems downstream, and you need to invest upstream to stop that happening.

We can supply the LGA case studies of best practice and what is being done in different areas across the country. That is a strength of a national local government association.

In another world, I know a little bit about how interfaith things are happening, and that the strength of interfaith activities is extremely patchy. There are some places where it works very well; many where it is non-existent.

You talk about how you build trust, one of the things we try to do when we are funding civil society organisations is using some of those organisations as a way into those conversations. It is a very powerful way, because they are seen as being perhaps more neutral than others are. We talk about interfaith here—we use people like Maccabi GB and others like that, who can go in and have those conversations—those Jewish/Muslim conversations. That is a different way into that conversation, rather than, say, a police officer or a politician standing up and trying to have it.

What happens nationally has a local ramification. When we went to West Yorkshire, we were told that when the previous Government decided to stop funding the interfaith network, that was seen locally by groups that had been working together as a signal that interfaith work was no longer valued. The amounts of money involved are probably tiny, but that symbolic thing that “this work is no longer valued” was picked up in local communities and people thought, “Have we been wasting our time?” I am putting words into people’s mouths, but there was a sense that it was not valued like it had been.

Mr Bowie, these various interventions, how do you assess the success of them?

The first thing is that when we put out the call for grantees to apply, we have a published theory of change that is developed based on best practice, not just in this country but across the globe. That is done with a partner. It used to be the Institute for Strategic Dialogue, now it is the Science of PCVE. Equally, we then run evaluations on each of these. We have reached over 180,000 Londoners across the first five calls, and hopefully we will reach another 50,000 with call six. On average, we have had 10,000 responses, which gives us a really good way of looking at this. We do assess against the four criteria, and I am happy to send you that. It is all published. Equally, we know that it is very difficult to do these things in a longitudinal way. We are confident that what we are assessing is the best quality evaluation that you can do of this kind of work. It has been commended by both the EU and the UN in terms of it being best practice. We are very proud of the work that we have done in that space.

How does the Prevent threshold levels compare with those accessing safeguarding pathways: mental health support or other safeguarding pathways?

The pictures that we got were that you have thresholds for Prevent that are designed around terrorism and the likelihood to possibly progress to commit a terrorist act because of defined ideology. We were told that there is no clearly defined threshold for somebody who is fixated with violence. One of the problems you then have with the system is: what do you do with that person? One place may say they will put them into the programme, because even though it is not designed for them, it is the closest we will get to expertise. Another place might take the decision not to refer them in because they do not meet the threshold. When we have argued in the report for what we call ‘the big front door’—a multi-agency approach to which young people can be referred without needing to meet a terrorism threshold—we think that is the right approach. You get everybody in, and then you find the right agency to take the lead. Those who are likely to go on—or could go on—and get involved in terrorist violence, would go through the Prevent Channel programme. For other people, it might be to do with substance abuse, mental health, neurodivergence issues, all those other factors. You get them to the most appropriate place. Rather than try to align other thresholds with Prevent, we would be better to have a broader approach for getting people into the system. Once they are in the system, they can be sent to the most appropriate place. There are bits of that practice—and they may well be in London—across the country, but it is not consistent.

I asked about that locally, and the officer who chairs our Prevent panel explained to me that that is what they do already. It is almost like a place for filtering, that there is bespoke matching to the needs of each individual, their vulnerabilities and the specific issues that they are presenting with. That naturally comes together. I took soundings from the police after Southport and everything earlier this year in the findings and was advised that we are following very good practice locally. All Channel members attend the Channel Hydra training, with the simulation and methodology, and that seems to work well in practice. Also, the thresholds are obviously different. It is quite difficult to compare the different matrixes and so on, and the Prevent ones are set by SO15. Where it works, the approach that has been outlined in the report is sort of what is happening anyway. That may be a way of looking at how you structure that across the country: share best practice and fund that effectively.

If you look in London, already more than one in 20 of the referrals into Prevent are where there is no fixed ideology, but it might be about violence fascinated individuals, so it is already happening. Probably not on the scale that you would want if there was national guidance or an approach on it, but it is the same people who would be sitting round having a discussion under Channel. If it was a terrorist ideologist, we would be discussing a violence fascinated individual. In a way, what Professor Denham is talking about there about the single front door—and that is why we are so keen to talk to David Anderson next month—it does, in principle, sound like the right approach.

The issue of the previous panel and yourself, and that we have seen from the evidence that we have had, that Prevent is very clearly being used to try to shortcut waiting times in mental health services, what would you do about that? What policy changes would you suggest ensuring that Prevent is not being used as a fast-track?

We would all vote money.

To some extent—and we say this in the report—you cannot take away what is there now until you have built something better. If Prevent is the only focal point around which things happen, you must continue with it. Building a better multi-agency approach does undoubtedly mean saying, “How do we make sure that child and adolescent mental services are in place?” That some other services are in place. It will take some investment over time. We were not convinced that the money that is currently being spent in the system is necessarily being spent as efficiently as it could be, because of all these different pressures on the system. It is very dependent on good practice at local levels to sort things out. Over time, you would need to build up some of the other multi-agency services that would not necessarily be there now.

I would also advise having more of a conversation with the LGA, because that has been the stopgap with the service and funding that disappeared, to try to provide some sort of support for local government. The bit that I am involved with meets quarterly and different councillors from across the country join that to talk about best practice, the issues, and so on. There is something there, but it needs a lot more work.

Do you get the sense that, because of the amount of referrals going in, the very fact of processing and assessing those referrals is using up valuable NHS resource, as part of multi-agency panels, for example, that would be best used providing treatment rather than sifting Prevent referrals?

The point that I am making is that you have this significant emergent problem of the people who do not fit the terrorist category, for whom Prevent is designed. They are a danger. They are not terrorists, but they are a danger to themselves and the wider community, and that can have a horrific cost. We have already accepted that we invest in counter-terrorism work because the cost of terrorism can be very high. This is about keeping the public safe and extending the services to a group of people who are vulnerable and a danger to the wider community is common sense, because part of the job is to the keep the public safe.

I would not want to say that, and we can give case studies of what is working, what is not, and maybe give some more information on that. We are very lucky in London, because it is almost like a partnership. The shared endeavour work happens in a lot of our local schools, and it means that Prevent can work with them and refer different people, and sometimes target different schools for certain workshops, which is important. We are talking about funding and what do you do. Quite a lot of that is the world in which young people are living. I cited the online stuff, unless there is those points of contact in the local community, the trusted family circles, youth providers, whatever space that is—faith, sometimes—as well as a national mechanism to push back on some of this stuff. I was at a session with young Muslim girls at the weekend around hate crime and VAWG, the combination between the two, listening to them a lot more. We were there to talk through and signpost a bit and one of them said, “TikTok is our news”. She was 14 years old, hijab-wearing and the world has changed so it is unsurprising that Prevent is not struggling but is not quite fitting what is there. It is what all those other sources are around there that need to work, and local government is a key part of that but so are these other things.

It is quite hard to find something that does not exist. We certainly talk to some places where, as Prevent had been scaled back and as cohesion funding had been scaled back, activities had been reduced. In fairness we must wait for the outcome of the Southport inquiry, but I think it is reasonable to say that there are places where the relationships do not work as well as they could do and where greater guidance is necessary. For all the places that we met who have gone to every effort to keep it going, I would be surprised if there are not a similar number of places who have just felt unable to invest extra money into the system.

Just following on the same theme, you have been kind enough to touch on the challenges of those associated when you are making assessments of no clear ideology. Can you expand a little bit on those specific challenges when the assessments have been made by Channel Panels?

You would know in London. What we were told in Manchester was, for example, they had to bring in people with expertise in violence reduction to the Channel Panels who had not necessarily been there before. They have had to bring in people with mental health expertise into the Channel Panels. Maybe you have always had that in London but certainly in Manchester they said because the numbers had shot up—and I think they have shot up particularly since Southport where there was a safety first sense of getting people into the system somehow—they had had to bring new expertise into the panels to deal with those cases but I am sure you know more about London.

I have down here that in Barnet—which is obviously the bit that I know best—where Prevent has never worked in isolation, very often involves a multitude of services in support of both individuals and their families so that there are referrals sometimes into early help, vulnerable adolescents, Rise Mutual that maybe are better fitted than the Prevent Channel for it. You would want to see that happening everywhere. I think with all the work you are doing that the multi-agency approach and how that works best is when everyone is sitting around the table together.

It is quite common in London and in London 39% of our referrals for the last published figures are either where there is no clear ideology or there is no ideology identified but they are susceptible to radicalisation. This has become something over the last four or five years that has become much more common for them. What is new and where we have tried to help with the mayor’s Office is around violence-fascinated individuals. We have had three roundtables that have brought various people from local authorities, violence reduction units—to the point you have made there—and the police, to try to talk about what is the right way forward in that and how you build the system around that. Some of those people may have psychopathic tendencies and they go from a spectrum of people who just potentially have a genuinely morbid fascination right up to others who genuinely want to do a school shooter level attack, so what are the interventions that you can put in and what is the evidence base around that. That is not straightforward and those are the sort of conversations we are starting to have at a London-wide level.

Maybe Councillor Conway, this is one for you to start with, but from a local authority perspective how do you make sure that the Channel Panels are truly accountable, representative and reflect the communities that they are there to serve?

Again, I will look at my notes on this, but it is difficult because we cannot talk about cases and things like that, but certainly I chair the Safer Communities Partnership Board. We have quarterly updates at that. You were saying about the London situation. The situation is such generally that at our meeting last week we had a specific focus on hate crime on Prevent, on the work that we have been doing on cohesion because we were one of the areas targeted back in the summer of 2024, and we co-produced a big online resource with local communities to push back on that narrative. I said at that meeting that, on a very difficult day, I watch it because it shows all different videos from across our communities produced by them that really push back on some of the narrative that is out there of the division and the levels of violence and everything falling apart. That balance is very important too, but in terms of scrutiny it is very much there. I am looking locally at if we have more of that, so we had a session recently where the police leading on it briefed our cabinet and CMT as well. It felt very necessary for reassurance as well after Southport that we were keeping proper track on it. That probably also needs to evolve as time goes on. We were talking about this outside. It will be different in every area, so there are similarities, but it needs to work locally too.

Do you have anything else to add on that, Professor Denham, just to do with the accountability and making sure that the Panels are truly representative?

I do not think I have anything to add particularly on the things that Councillor Conway has talked about because I do not have that level of expertise.

Can you expand on the relationship between local authorities and the Home Office when trying to ensure that we are delivering a Prevent programme where the allocation of assessment is done as appropriately as possible? Is that relationship and interaction between local authorities and the Home Office as good as it can be to ensure the assessment is working as well as it can?

I do not want to speak about it at a national level but at a London level we are lucky in that the local authorities have their own Prevent-style meeting, and they come to the London CONTEST Board. We have the Home Office Director who attends that and we have questioned the then Director—not the new one—about a Prevent risk is calculated and how it is funded. We would certainly argue that London should probably get more money than it does. However, you would expect me to say that. In London I think it does work quite well. There is always room for improvement in terms of how that co-ordination and work could go but London, especially when you speak to CT policing about this, we are seen as an area of best practice in terms of how we join that up from top to bottom. I cannot speak for the rest of the panel.

Locally and nationally there are the annual checks that are done and everything else. The Channel Hydra training was spoken well of in terms of the simulation and the models that it does. If it is possible to come back on your previous question, it is something that we have done locally, and I think the Belong Network—that I have given evidence with at different things—has case studies of what local government is trying to do in this space around diversity everywhere. We have set up different roundtables with different minority majority local communities, our biggest diverse communities. We have one happening this evening. That has been to systemise the relationship into local government, so that with our officers we are able to build trust, take temperature checks and to have those conversations. I just want to come back on that because of what you were saying. Of course, money matters but it is also what you do with what you have locally as well.

The picture we picked up was one where Home Office and local are not always in step with each other. If you take Dovetail—which is an evolution of Prevent, which was much more towards the multi-agency approach, which ran for about four or five years as a pilot approach—a lot of the people we talked to who had been involved in that thought that it was moving towards a better model. There were some issues about it and trust and the relationship between counterterrorism, policing and the local authority partners but, rather than perhaps resolve that, things moved back away from that model. I would have said that most of the people we talked to in local authorities want to move one way or another in a multi-agency direction and would like the support of the Home Office in doing that. To pick up on Councillor Conway’s point, I do not think anybody wants the Home Office just to tell people what to go and do because how you organise services at local level will have to reflect what is on the ground. So, some places will have multi-agency safeguarding hubs, they will have violence against women and girls action groups, they will have violence reduction units and other people will have some of those and not the other ones and some of them will have none of them in some areas but the basic principles of design I think can be set out and I think that the Home Office should work with local government perhaps more than it does at the moment.

Really you have anticipated my question. Lord Anderson recommended that Prevent be brought into a broader safeguarding process. What we are talking about is identifying vulnerable children and diverting them to the best solution and what I think you are telling me is Prevent is being used somehow to do that in a not utterly appropriate form but that is how you are using it but the way presumably is to have Prevent as a second tier as one of the routes one goes to and redivert children to wherever is more appropriate. Presumably one could then remove some of the stigma that currently affects those diverted to Prevent at the outset.

I would agree with all those things. I think Lord Anderson’s report and ours are very much in the same area of proposal and, yes, you clearly need within the system to have terrorist-related capacity because some of the people will be at risk of becoming involved in terrorism and that requires specialist responses but if you have a broader way in for young people who are at risk you do not have to label somebody as a terrorist potentially to get them into the system but you can divert them to the most appropriate place after that. Then we suggest in the report, we did not say that the Prevent duties should be scrapped but we basically said once this is seen as a broader safeguarding route then what is the justification of having a separate Prevent referral requirement? What you are doing is talking about teachers doing what they are trained to do anyway which is identify safeguarding risk and send people to the appropriate place. Then you have that 90% of young people who get referred into Prevent to whom nothing is done now, you take that away from them and the potentially damaging effect on community relations of having somebody’s child labelled as a risk when they have problems that need to be sorted out.

Nationally it is roughly 54% or 55% of Prevent referrals are under-18s. It is slightly lower than that in London at I think around 44%. Equally, you see an increasing number of the disrupted plots feature children as a whole and there must be a better way of getting into children before they get to that stage and avoiding either the stigma, if you want to call it that, of going through Channel, Prevent or whatever but equally I think the Youth Diversion Orders going through Parliament. We need to wait and see once they come in how they get utilised and exactly how they are going to operate in practice but that is a potentially good way of taking children down a non-criminal route and diverting them away from some of the extremism or terrorist stuff, which I think is quite a promising development now.

I agree with that but, whatever the system, however you do it for local government it needs to be properly funded. For something like the Youth Diversion Order for it to really have any traction it needs to be attractive to young people. It is not unlike some of the work that goes on around exploitation. This is a different form of exploitation which you touched on, and I would say certainly London’s Violence Reduction Unit has some very good models of that, but they are well-resourced. We have run a couple of programmes locally with that additional funding and they work very well because again you can make it bespoke for your own local needs, which is very important, so you get the regional framework and that view across an area as well. I worry slightly that if you are just looking at one local authority and no one stops at a border, a boundary of a local authority, so it is that wider piece as well that is very important regionally.

Vast sums of money are being wasted now via Prevent. If four-fifths are not getting through to the Channel process that means police are spending time on an inappropriate referral, which is a huge waste of resource. Presumably if it was the other way around that filtering would not happen at that point, would it?

It is difficult to say it is an inappropriate referral when it is a voluntary scheme so people may be referred in certain—

I am not saying it in a critical way.

I take the wider point. I think if you are looking at this there should be an opportunity for people to engage with the programme if they are willing to, which can take them away potentially from the path that they are on. I take the point that there may be a better way. There is always a better way of doing stuff. It is just how you find it.

Obviously some of those young people who do not go into Channel have problems, but because they do not go into Channel their problems are not being dealt with. The crucial thing is to make sure those issues are dealt with appropriately.

So for the LGA we can certainly send some case studies. Some are anonymised and some are not, but we do have that sort of resource and we can follow up with that.

I have already touched on the Shared Endeavour Fund and have already committed to Mr Kohler to send through all the reviews that we have of that and if anybody would like to come out to see some of the projects and programmes that we have underway we are very proud of them and are very happy to help facilitate all of that for you, whether that is some of the stuff that we take against terrorism online or whether it is stuff working with some of London’s premier league football teams we are very happy to help.

We can also pass on details of people we talked to in Manchester and west Yorkshire who seem to have good records of doing work around radicalisation and community cohesion with some very good evidence of how that got them through the last two very difficult years, for example, so we can pass those on.

For us there is a guidance and a toolkit that the Government have funded that the Belong Network have put together that I think is due to appear soon. Certainly, we were signing off some stuff for it recently so that work is happening, yes.

We publish all our findings. We have a public Theory of Change that the organisations can look at, which is based on best practice. For example, at the launch of Call 6 we brought all the organisations together responding to the recommendation of the previous review about building capacity and we did teach them about how to maximise their presentational skills so they could have maximum impact with the young people that they are working with so that is exactly the space that we operate in.

We certainly had very good evidence from the special interest group of the LGA, which I forget the title of.

SIGCE, Special Interest Group on Countering Extremism. I hope that is right.

That if I have that right is a network of local authority practitioners who are leading the front-line work on this, and we mentioned them in the report as a potential source of spreading good practice around the country as things develop.

Yes, I think as I answered before and as Kenny was saying they are interesting but there needs to be the proper resourcing. It needs to be something that young people would choose to do and would work. I think co-production is very important. With our exploitation strategy that is now called Keeping Young People Safe we have produced the whole thing with young people at the heart of it; we have done the same with our violence against women and girls’ stuff. All these things need to be everybody together and local government is very well placed to be that convener of place.

To my knowledge most of the available evidence would suggest where possible to avoid criminalising children. Now, obviously there will be certain children who are involved in terrorist activity or extremist activity who will need to be prosecuted, who will need to go to prison but I think in many cases it would be far better to go down a non-criminal route and try to address that young person’s needs as Professor Denham was saying and take them away from that. That must be better for the child, and it must be less costly to the state in the longer term. I agree about it being properly resourced, but I think the evidence would suggest that is the right approach.

We could not go into these in detail because they do not exist or they did not when we did our report. The observation I would make and I would not want to take the Committee too far away from the main topic is they arise because the definition of terrorism is so broad that you have young people being involved in activities that could be legally classed as terrorist adjacent and you therefore must invent something that is not a legal process to get them out of a problem. If you did not have such a tight definition of terrorism, you might have a broader approach. In practical terms, the interesting thing from me is that if we move towards this multi-agency diversion route where would Youth Diversion Orders sit within that? We will need a range of activities for young people who need to be assisted. It would be interesting to see whether they turn out to be necessary or whether they will just be part of a pattern of provision that is available.

That goes without saying. Mental health is very important. We convened a meeting recently at the London level with the LGA’s Thrive London and the community’s team for working with one of our communities who felt that young people were having to self-silence and as well as seeing a lot of trauma with the footage from the Middle East and everything else. If I had a wish list I think that keeping mental health as part of this at every step of the way would be important and also creating safe spaces and the tools and techniques, some of which is part of the Shared Endeavour Fund, to disagree, to push back on disinformation, to be able to have all those tools and all those skills that some of our young people do not get anymore. It is always in these briefing papers and, it is true, there is a generation that grew up with the impacts of covid and not being at school and who went to university in a different way. There is that holistic approach that is needed.

I would take it a step further back and say that I think one of the most important things is how you build among younger people psycho-resilience and critical thinking skills because so much of what they will see is the TikTok news or the misinformation and disinformation online or it will be stuff where they potentially feel targeted themselves about what they see online and being able to build that psycho-resilience or social inclusion, call it what you like, I think those skills are quite fundamental to building the resilience to the hatred, extremism, terrorism and making sure that you do not get to a stage where you have to have intervention in the first place.

Your caveat about not any money is making me think. Even if there was not money—I hope there would be but even if there is not money—for the Government to set out what a social cohesion strategy should look like and what the expectations are would be enormously useful. There are places that are doing it very well; there are places that would like that support. There are people who would like that expectation just so that we have the sense that we are all working in the same direction because that would include issues such as resilience among young people. It would include how schools and other places learn to handle difficult discussions. One of the things you worry about the online world is if the teachers and others do not feel capable of allowing discussions to take place in the classroom it does not mean the young people will not go and try to find information about the same issues somewhere much more dangerous. This was an issue 20 years ago and it has moved online now but it is the same issue. The other area just to back Councillor Conway, mental health is the service issue that came up most often in our discussions. It is not the only missing element, CAMHS is struggling everywhere, but whether there is any way of better integrating the mental health support that is needed for this group of young people that would be enormously helpful to lots of people.

Excuse my ignorance but can you give me in broad terms what a social cohesion strategy would look like?

It is all the things you can do to enable people from different backgrounds and communities to meet, to understand together, to learn how to discuss difficult issues together, to work out what it is in common they want to achieve themselves, their families, their children so you do not have breakdowns of information. If you do that several things follow. You will reduce the overall level of extremism because extremism is likely to grow where there is misunderstanding and lack of knowledge and prejudice. Secondly, you will get more resilience because when something does happen, and no one is suggesting these strategies will prevent anything bad ever happening because that is not the real world, it will be resilient because most people will respond collectively because they know each other and they do not want it to happen. Thirdly, you will find more trust in the sorts of things we have been talking about this afternoon because people will be willing to refer members of their family or their friends to, as it were, the authorities or the system if they are genuinely convinced that this is a system for helping people and sorting out problems, not something that is going to stigmatise or get them into the criminal justice system unnecessarily. It is a myriad of different activities. One of the things we suggested in the report was bringing back the comprehensive community cohesion survey that operated briefly when Prevent was first launched, which enabled local authorities to gather a lot of data about interactions, confidence and trust in neighbourhoods, how people thought about their area, their understanding of different groups because at least if that were done we would have a baseline of information and every local authority would have the same baseline of information. It is one of those areas—and I think you mentioned earlier that you had some of that data here in London, but it is not uniform, having a baseline of understanding the extent to which people feel they know, trust and can rely on other communities and other groups and then looking at the interventions that bring people together.

This is the essence of a liberal democracy, of course.

It is important to a liberal democracy, yes. The ability to discuss, Mr Kohler, I think is crucial. One of the things that we are at great risk of now is losing our nerve to have difficult discussions with our neighbours about issues on which we disagree. This is not samosas and cucumber sandwich events. This is creating the space where you can have discussions of difficult issues that are safe. That is a crucial part of the way our society needs to operate and there are places that do that and before this session Councillor Conway and I were talking about some of their local activities, but it is not consistent and not uniform in its approach and it is very dangerous. A lot of what we have been talking about today, not all of it, is about the problems that come downstream if you have not done that first.

Thank you very much. We have come in on time so that is unusual on this inquiry, but you have been very good at answering our questions succinctly and giving us all the information. We very much appreciate your time, and I will bring this to a conclusion now.