Schools are rightly expected to notice when something is wrong. We expect teachers to identify changes in a child’s behaviour, recognise emerging safeguarding concerns, notice deteriorating attendance, respond to mental health difficulties and identify barriers that might indicate additional needs. We have spent years, quite rightly, encouraging schools to intervene earlier rather than waiting for problems to escalate. But what happens when a school identifies the problem and there is nowhere for the child to go? That question was brought into sharp focus by the children’s commissioner’s recent findings on referrals to Prevent. Her analysis found that at least 65 per cent of referrals involving children made by schools and colleges were returned to education settings when the referral was closed. In 93 per cent of these cases, the child had not progressed to a Prevent panel. The commissioner described professionals seeking help for children but encountering “closed doors and dead ends”. The findings relate specifically to Prevent, but the phrase will feel uncomfortably familiar to many school leaders. Because increasingly, schools are becoming the service that cannot say no. The problem is rarely that other professionals do not care about children. Nor is it that thresholds exist without reason. Specialist services have finite capacity and inevitably have to make decisions about where that capacity is most urgently needed. But, viewed from inside a school, the consequences can be difficult. A child may not meet the threshold for specialist mental health support or social care intervention but will still be struggling. Referrals may close. The child’s needs do not. And tomorrow morning, that child will usually walk back through the school gates. Eligibility criteria This creates a group of children who are not necessarily without need, but whose needs do not fit neatly within the eligibility criteria of the services designed to support them. Look at the extent to which schools have adapted their staffing structures. Schools employ family support workers, attendance officers, counsellors, pastoral teams, mental health practitioners, behaviour specialists and increasingly sophisticated inclusion teams. Much of this work is exceptional and some of it represents precisely the kind of early intervention we should want schools to provide. It would therefore be far too simplistic to argue that schools should retreat behind a boundary marked education and declare everything beyond teaching and learning somebody else’s responsibility. Children do not experience their lives in professional silos. A child experiencing anxiety does not leave it at the classroom door. Schools should not ignore those realities. But there is a difference between recognising that education is inseparable from children’s wider lives and allowing schools to become the default destination whenever another part of the system cannot respond. There is a paradox here. Schools have become extraordinarily resourceful at responding to unmet need. They do so because the alternative is often to leave a child unsupported. But the better schools become at compensating for gaps elsewhere, the easier it becomes for those gaps to form part of the normal architecture of children’s services. What schools once did because nobody else could do so begins to look like something schools were always supposed to do. That should concern us. Not because pastoral support, inclusion or early intervention sit outside the purpose of education, but because every additional responsibility draws on finite capacity. There is also an important question about expertise. Schools can provide enormously valuable support, but teachers and pastoral staff cannot simply substitute for every specialist professional whose service a child cannot access. Early intervention There is a wider policy contradiction here. Across education, health and children’s services, we talk constantly about the importance of early intervention. The principle is compelling: identify need early, respond quickly and prevent difficulties from becoming more complex. But early identification is only useful if something happens next. There is little benefit in schools becoming increasingly sophisticated at identifying emerging need if the wider system can respond only once that need has become sufficiently acute. This is why the children’s commissioner’s call for clearer routes to appropriate support matters beyond Prevent. None of this requires another debate about whether schools should focus solely on academic education or take responsibility for the whole child. That is a false choice. Effective schools have always understood that children learn best when they are safe, supported, attending and able to participate. Schools will therefore always have an important role in identifying need, providing early help and coordinating support around children and families. But partnership only works when there are partners on the other side. If schools are expected to identify need earlier, there must be accessible services to which that need can be directed. If specialist thresholds necessarily remain high, there needs to be sufficient support beneath them. And where responsibility genuinely belongs across education, health and social care, accountability for providing that support needs to be shared too. Perhaps the real question is not whether schools are doing too much. It is whether we have gradually constructed a system which depends on them doing so. For children, organisational boundaries are largely irrelevant. But our willingness to help children should not become the reason those boundaries, and the responsibilities that sit behind them, cease to matter. A genuinely child-centred system should not depend on schools becoming increasingly good at filling the spaces left by everyone else.