Sure Start is a useful starting point for understanding a much wider shift in British social policy. On its own terms, it was an early-years programme built around a plausible and humane idea: that poverty and disadvantage are reproduced partly through early childhood environments, and that joined-up local services might improve children’s life chances before disadvantage becomes hardened. It drew intellectual support from the wider early-intervention tradition, especially American evidence from programmes such as Perry Preschool, Abecedarian, the Chicago Child-Parent Centers, and Head Start. The underlying theory was that early disadvantage has cumulative effects, so intervention before school could produce benefits across health, education, behaviour, and later social outcomes.
That was not an irrational theory. Early childhood clearly matters. Parenting clearly matters. Children do not arrive at school as blank slates, and family environment can shape vocabulary, attachment, self-regulation, health, attendance, behaviour, and later achievement. Nor was Sure Start simply a fantasy imported from American social science. Later UK evidence, especially from administrative-data studies, suggests that Sure Start produced measurable benefits in areas such as hospitalisations, school outcomes, absences, special educational needs, and some youth outcomes. The strongest conclusion is not that any early-years programme automatically works, but that well-resourced, accessible, integrated early-years services in disadvantaged communities can produce real medium-term benefits.
But Sure Start also reveals the limits of the early-intervention worldview. The evidence did not mechanically imply this particular policy design. It was evidence interpreted through a progressive theory of disadvantage. The problem was understood less as parental failure, family disorder, weak norms, or lack of discipline, and more as unmet need, poor access to services, and insufficient support. Sure Start was therefore designed to support and reshape family environments indirectly rather than confront parental failure directly. That does not make it wrong. It does mean the debate should not be framed as evidence versus prejudice. It was evidence plus values: values about what parents owe children, what the state may demand of parents, how far public policy should go in correcting family environments, and how much coercion a liberal state is prepared to use.
This distinction matters because a plausible causal story is not enough. Given almost any social problem, it is possible to invent several plausible solutions. If children arrive at school with poor language, poor behaviour, poor health, or poor self-regulation, one can argue for parenting classes, welfare conditionality, family support workers, nursery expansion, compulsory home visits, stricter school discipline, child protection intervention, cash transfers, marriage support, neighbourhood policing, or cultural campaigns around parental responsibility. The serious evidence question is not whether one favoured intervention is consistent with a plausible theory. It is whether that specific intervention beats rival interventions. Sure Start clears the weaker test: it addressed real early-years mechanisms and appears to have produced measurable benefits. It does not fully clear the stronger test: that it was demonstrably the best or most cost-effective way to deal with poor parenting and early disadvantage compared with harder-edged, more targeted, or more coercive alternatives.
This is where Sure Start becomes more than an early-years policy. It becomes an example of a broader governing style. Modern Britain has increasingly converted failures of family formation, school authority, social norms, community life, poverty management, digital childhood, and parenting into demand for professional services. Once that conversion happens, the answer to almost every crisis becomes more staff, more funding, more programmes, more guidance, more support, and more institutional responsibility. The state does not necessarily solve the original problem. It absorbs the problem into a service system.
The moral thread running through this system is care. The assumption is that social failure is, at root, a deficit of care, compassion, recognition, support, or access. Since one can never be infinitely caring, the system is always guilty, always underfunded, always in crisis, and always entitled to expand. This produces a politics in which the existence of suffering becomes evidence that the service system must grow, while the failure of the service system becomes evidence that it was never large enough. A more serious model would still care about suffering children, distressed adolescents, neglected families, and vulnerable adults. But it would put care inside limits. It would distinguish illness from distress, need from demand, support from indulgence, and compassion from the permanent removal of responsibility.
The problem is not compassion itself. The problem is compassion when it becomes an open-ended governing principle. If every difficulty is redescribed as unmet need, then every institution becomes morally liable for providing care. Parents are not failing; they are unsupported. Pupils are not misbehaving; they are dysregulated. Students are not struggling with ordinary unhappiness; they are experiencing a mental-health crisis. Citizens are not making bad choices; they are being failed by systems. There is truth in some of this language. There are real traumas, real disabilities, real injustices, and real institutional failures. But when the framework expands too far, it weakens the concepts of duty, agency, discipline, hierarchy, and consequence. It also overwhelms the very services it creates.
This is one reason Britain can have both decades of early intervention and a serious child and adolescent mental-health crisis. Sure Start may have improved some outcomes for exposed cohorts relative to what would otherwise have happened, but it was never large enough, sustained enough, or broad enough to prevent a later national surge in distress. Its evidence is about marginal causal improvement, not social transformation. A programme can work in a narrow statistical sense while the wider society still deteriorates. If family life, digital childhood, school authority, peer culture, housing, welfare dependency, community breakdown, and therapeutic expectations are all moving in the wrong direction, then a service intervention may reduce harm without reversing the larger trend.
The same logic helps explain the rise of lived experience as a political tool. Lived experience gives emotional authority to policy claims. It can expose failures that aggregate data miss, and it can force institutions to notice people who were previously ignored. But it is also vulnerable to selection bias, availability bias, and advocacy packaging. Campaigning NGOs understand this very well. A vivid personal story can do what a statistical table cannot: it can create urgency, guilt, identification, and pressure. In that sense, lived experience often functions like a behavioural nudge. It bypasses abstract reasoning by making a problem emotionally immediate. The serious version of lived experience treats it as qualitative evidence: a way to generate hypotheses, reveal mechanisms, and identify blind spots. The unserious version treats it as a trump card over population-level data, counterfactual testing, and cost-effectiveness.
This produces a familiar pipeline. Compassion supplies the moral premise. Lived experience supplies the testimonial authority. NGOs convert testimony into campaigning pressure. Professionals convert campaigning pressure into service categories. The state converts service categories into programmes, guidance, funding streams, legal duties, and institutional obligations. None of this requires conspiracy. It is a system of aligned incentives. Campaigners gain causes. Professionals gain authority. Institutions gain responsibilities and budgets. Politicians gain moral language and addressable constituencies. The people who are meant to be helped may gain services, but they also become the raw material of a permanent politics of need.
This care-based politics has also been shaped by changes in political representation. Greater female representation did not mechanically create the caring state, but it helped legitimate and expand the political authority of care. As women became more present in politics, public administration, health, education, NGOs, and the professions, issues organised around vulnerability, emotional harm, safeguarding, childcare, abuse, wellbeing, and support gained status. That corrected real older neglect and brutality. The older order often ignored suffering, excused abuse, overvalued hierarchy, and underprovided for children and vulnerable people. But the correction created its own imbalance. The language of care became increasingly difficult to oppose, while limits, discipline, punishment, authority, parental responsibility, and institutional gatekeeping became harder to defend publicly.
There is also a political-market logic. As women became a larger independent electoral constituency, parties, especially parties of the left, had incentives to frame more social problems in the language of care, protection, services, and compassion. This does not mean that female voters all think alike, or that care politics belongs only to women. It means that democratic politics learned to address women as a distinct political market partly through issues of vulnerability, family, public services, childcare, safety, welfare, and emotional harm. The same logic later operated through gay men and then wider LGBT constituencies, though with recognition rather than care as the central currency. Both forms of politics begin with genuine injustice. Both can produce legitimate reforms. Both can also generate self-expanding systems in which vulnerability becomes a permanent resource for parties, NGOs, professionals, and institutions.
Education is one of the clearest places where this governing style has expanded. Schools and universities have increasingly become welfare-facing institutions. This began for legitimate reasons. Children do bring real family problems, trauma, poverty, disability, and mental-health difficulties into school. Students do arrive at university with vulnerabilities that cannot simply be ignored. But the cumulative effect has been to subordinate academic formation, discipline, and standards to wellbeing, inclusion, safeguarding, and risk management. In schools, the teacher becomes part educator, part social worker, part behaviour therapist, and part safeguarding officer. In universities, the lecturer becomes part academic, part service provider, part emotional-support agent, and part risk-managed representative of institutional care.
This shift connects directly to the expansion of higher education. The failure was not simply that too many people went to university. It was that mass higher education expanded the status form of university faster than it expanded hard intellectual and technical capacity. A country does not become more capable merely by increasing the number of degree holders. It becomes more capable when more people master difficult, useful, reality-constrained knowledge. Mathematics is the clean example because expansion in that field would have meant an increase in genuine national capability. The same could be said of engineering, physics, computer science, medicine, and other demanding disciplines. Instead, much of the expansion occurred in softer fields, including parts of the social sciences and applied professional disciplines, where students are often trained to interpret society through structural harm, unmet need, identity, inequality, and institutional failure.
That does not make those subjects worthless. Some of their knowledge is useful, and some of the injustices they describe are real. But the expansion produced a large credentialed class whose status and employment depend on interpreting society as a network of harms requiring expert intervention. This class then enters schools, universities, charities, local government, HR departments, public health bodies, equality offices, safeguarding teams, policy units, and NGOs. It is trained to name problems, classify vulnerability, process claims, design interventions, and justify institutional expansion. It is also often weak at checking its own premises against hard counterfactual evidence. The result is a class that sees society through unmet need and structural harm while working in institutions whose budgets and authority grow when unmet need and structural harm are discovered.
That is the feedback loop. The more problems are found, the more the system justifies itself. The more the system fails to resolve them, the more it argues that it was never resourced enough. The language is moralised as compassion, so scepticism can be framed as cruelty rather than as a demand for evidence, limits, and accountability. The system therefore becomes difficult to challenge. To question whether a service works is to appear indifferent to suffering. To ask whether parents should be held more responsible is to appear punitive. To ask whether some distress is ordinary rather than clinical is to appear dismissive. To ask whether schools should prioritise authority and knowledge over wellbeing is to appear harsh. To ask whether universities should stop treating adults as vulnerable clients is to appear uncaring.
Sure Start remains a revealing case because it contains both sides of the argument. The pro-Sure Start case was not foolish. Early years matter, and there is evidence that integrated early-years support can improve outcomes. But the programme also expressed a wider tendency to translate social and familial failure into service demand. It treated poor parenting and early disadvantage primarily as problems of support rather than responsibility. That may sometimes be the right approach. It cannot be the only approach. A society that responds to every failure with more care, more recognition, and more professional support eventually loses the ability to say that some people must be expected to behave better, that some institutions must enforce standards, and that not every form of distress should become a claim on the state.
The central issue, then, is not whether Britain should care less. It is whether care should remain bounded by evidence, responsibility, authority, and limits. A serious social policy would still protect children from neglect, support families in difficulty, fund useful early-years provision, and listen to people failed by institutions. But it would also ask harder questions. Does this intervention outperform the alternatives? Does it change behaviour or merely process demand? Does it strengthen families or substitute for them? Does it build capability or dependency? Does it solve the upstream problem or professionalise its consequences? Does it help the genuinely vulnerable, or does it expand the category of vulnerability until ordinary life itself becomes a service claim?
Sure Start therefore belongs to a much larger story. It began as a hopeful attempt to improve the life chances of disadvantaged children. It now also stands as an example of how modern Britain governs through care: identifying vulnerability, collecting testimony, expanding professional categories, funding services, and treating overload as proof of moral urgency. The challenge is to preserve the genuine insight that early environments matter while rejecting the weaker assumption that every social problem is best solved by more services. Without that distinction, compassion becomes not a virtue within policy but a machine for endless institutional expansion.