The Problem of Prevention

Another piece inspired by the Leeds Peace Conference

At the Leeds Peace Conference organised by Inspire North, His Honour Judge Guy Kearl KC, the Recorder of Leeds and the Senior Circuit Judge at the Leeds Combined Court Centre described two programmes designed to deter young people from knife crime. Having described two excellent programmes that were obviously doing good work he asked a question. Having told a story of a young man who saved a life because he had learned how to apply a tourniquet on one of the programmes he asked “How do you provide evidence that these prevention programmes are working?”. A life saved - yet still a demand for evidence.

Prevention is hard to prove in systems organised around measurable outputs because prevention concerns events that do not happen. How do you evidence a stabbing that never occurred? Most evidence frameworks are better at measuring interventions on visible problems than they are at recognising and evidencing the conditions that prevent problems emerging in the first place. The NHS can count admissions, the police can count arrests, courts can count prosecutions and our prisons can count inmates. But how do we count trust, dignity, belonging, courage, moral formation and the restoration of hope? Because these are the ground from which non-violence and peace can emerge. It seems that our systems become biased toward treating damage rather than cultivating the conditions for flourishing. A hospital A&E unit overflowing with knife injuries and other trauma generates lots of data, urgency, targets, dashboards, funding streams (over £200bn a year now) and political visibility. A quiet summer evening in which nothing much happens generates nothing measurable at all, though that harmless evening may represent the greatest achievement.

The Tyranny of the Counterfactual

Programmes like the Leeds Crown Court initiatives operate within complex human systems where causation is distributed, relational and emergent rather than direct and linear. They are part of a causal web in our city. The demand for “proof” fails to recognise this and assumes a more linear mechanistic model of causation where Intervention A causes Outcome B. But reality is fundamentally processual and relational rather than mechanistic. Our lives unfold through countless interactions and histories. In this frame, prevention programmes are not “causes” in a linear sense; but are influences in a living social ecology. This makes them difficult to isolate and prove experimentally.

The Politics of Evidence

There is a political dimension to this too. “Evidence-based practice” sounds so rational and obviously fair. We are urged to ‘look at the evidence’ and ‘follow the science’. But evidence regimes are never neutral because they privilege some forms of knowing over others. They privilege quantitative over qualitative, measurable over experiential, short-term over long-term, visible outcomes over subtle cultural shifts, and certainty over wisdom. As a result relational and preventative work is disadvantaged. The programmes most likely to transform young lives may produce the weakest managerial evidence, while programmes with marginal human impact can produce strong metrics. The A&E unit that treats dozens of stabbings has done nothing to reduce knife crime. Our metrics tend to capture what is easy to count, not necessarily what matters most.

When prevention programmes must continually prove themselves numerically, they are often pressured to evolve around visible metrics rather than deep relational work. The danger is that storytelling is replaced by dashboards, and building trust takes a back seat to performance indicators. Short term impact matters more than long term shifts and the results of a managerial audit of performance and impact is more highly valued than the act of moral imagination that originally designed the prevention strategies.

The Tourniquet Story Matters

The tourniquet story Judge Kearl shared matters because it avoids any abstraction. A young person saved a life, because of a programme. We have always transmitted moral knowledge narratively because our stories reveal meaning, context and transformation in ways that statistics alone cannot. Our real expertise emerges through context-rich practical wisdom rather than the identification of abstract universal models or performance dashboards. A culture obsessed with scalable evidence and ‘rolling out what works’ loses contact with our lived realities. One saved life does not “prove” the programme works in a positivist sense. But it offers a tremendous clue.

The Ethics of Care Perspective

From an ethics of care perspective, we can ask very different questions. Instead of “Can you prove this intervention causes reduced knife crime?”

we might ask:

• Are these young people being encountered with dignity?

• Are relationships of trust being formed?

• Are adults showing up consistently?

• Are spaces of belonging being created?

• Are conditions for violence being discouraged?

• Are young people becoming more capable of caring for themselves and others?

• Are communities becoming more relationally resilient?

Such care is not sentimentality but a political and moral practice that maintains, and repairs our world so that we can live in it as well as possible. Programmes like the Leeds Crown Court initiative are not just “interventions” but forms of such social repair.

The Fear Beneath the Demand for Proof

The demand for certainty often arises from institutional fear. Public bodies fear accusations of wasting money, media attack and accusations of naivety. Metrics can counter such accusations and offer a sense of certainty that this works. But in complex human life, certainty is illusory. This does not mean abandoning evidence, but becoming more nuanced about what evidence can and do. And, what it can’t. When discerning the effectiveness of a programme in repairing and maintaining our world there are a whole raft of things that matter:

• Quantitative evidence

• Qualitative evidence

• Narrative

• Professional judgement

• Lived experience

• Ethical intuition, and

• Community wisdom (genuine common sense?).

But our evaluation systems often recognise only the first item on this list as legitimate.

Toward a Different Understanding of Evidence

I think we need a broader understanding of evidence itself. Not evidence as mechanistic proof, but evidence as signs of attentiveness and accumulated practical wisdom, as patterns of human flourishing observed carefully over time. Evidence as relational discernment.

This would require more epistemological humility, recognising that the most important dimensions of human life may not always be fully measurable, and that perhaps communities know more than our spreadsheets and dashboards can capture. Can our institutions learn to act on forms of knowing that cannot be reduced entirely to managerial proof?

If they cannot, we may continue funding the measurable aftermath of violence while underfunding the immeasurable and complex web of conditions that can prevent it.