Before Harm Becomes Irreversible

If we act early because of what might happen to a child, should we also look early at what our own actions might do?

There is a logic at the heart of modern child protection that is difficult to disagree with.

We should not always wait for harm to happen.

If there are serious reasons to believe that a child may be in danger, waiting for certainty may mean waiting too long. By the time every doubt has disappeared, the very thing we hoped to prevent may already have happened.

That is why risk matters.

It asks us to look forward.

Not only at what has happened, but at what might happen.

There is wisdom in that.

But recently, while thinking about Britain’s apology for historic forced adoption, another question occurred to me.

It is a simple one.

If we believe in acting early because harm may occur in the future, should we apply some of that same caution to the actions of the system itself?

This is not an argument against child protection.

Nor is it an argument that intervention and non-intervention carry equal risks in every case.

They do not.

There are children who need protection urgently. There are circumstances in which removing a child is necessary. Any serious discussion of child protection has to begin by accepting that.

But intervention also creates a future.

That future is harder to talk about.

When a child is separated from a parent, something happens immediately: the child is removed from one situation.

But many other things begin at the same moment.

A relationship changes.

Contact may become supervised, reduced or eventually disappear.

A child may begin to attach to another family.

Brothers and sisters may grow up apart.

Grandparents may move from being part of everyday life to names in a file.

A temporary arrangement may, over time, become very difficult to reverse.

None of this proves that the original intervention was wrong.

Sometimes these losses may be the painful price of keeping a child safe.

But they are still consequences.

And if consequences matter before intervention, perhaps they should matter after it too.

This is where I find the question difficult to put aside.

Child protection asks professionals to imagine the future of a family.

What might happen if nothing changes?

What might happen if this child remains here?

Could today’s concern become tomorrow’s harm?

These are serious questions.

But there is another possible future sitting beside them.

What happens if we intervene?

What if the assessment is partly wrong?

What if the family could have changed with the right help?

What if a relative who seems unsuitable today could become a safe carer with support?

What if a relationship we allow to disappear turns out, years later, to have mattered deeply to the child?

What if a decision made to protect one part of a child’s life unintentionally removes something else the child will later need?

We cannot know these things with certainty.

But that is precisely the point.

Risk assessment exists because the future is uncertain.

So perhaps uncertainty should make us curious in both directions.

There is a phrase we often hear in public life:

better safe than sorry.

It sounds sensible.

But in decisions that can permanently alter human relationships, “safe” and “sorry” are not always sitting neatly on opposite sides.

Sometimes an intervention prevents terrible harm.

Sometimes an intervention also carries losses of its own.

The difficult work is not choosing between “protection” and “harm” as though one belongs entirely to one side.

It is trying to understand where harm may lie in both directions, and which risks can still be reduced.

That may require a slightly different habit of mind.

Before asking only:

What might happen if we do not act?

perhaps we should also ask:

What might happen because we do?

And then one more question:

If we later discover that we were wrong, what will still be possible to repair?

That last question matters because not all decisions age in the same way.

Some can be changed.

Some can be reviewed.

Some mistakes can be corrected.

Others become harder to undo with every passing month.

A child settles.

Relationships weaken.

Time passes.

Life reorganises itself around the decision.

Eventually, even if our understanding changes, reality may no longer be able to follow.

This is one reason the recent historical apology is so thought-provoking.

It reminds us that institutions can sincerely believe they are acting properly and still be judged differently by history.

That does not mean today’s institutions are repeating yesterday’s mistakes.

It means something more ordinary, and perhaps more important:

no generation has the advantage of knowing in advance exactly how history will understand it.

We act with the knowledge we have.

We use the evidence available to us.

We make judgments.

And sometimes we are wrong.

The question, then, is not how to create a system that never makes mistakes.

There will never be such a system.

The better question may be whether we can design decisions so that uncertainty itself receives some respect.

Can we preserve relationships where it is safe to preserve them?

Can we keep genuinely temporary decisions temporary long enough for uncertainty to clear?

Can we give support early enough that we learn what a family can do with help, rather than only what it cannot do without it?

Can we distinguish carefully between what we know and what we fear may happen?

Can we revisit predictions when new evidence appears?

And when the consequences of a decision may be permanent, can the level of scrutiny rise with the level of irreversibility?

None of these questions makes protecting children less important.

Perhaps they make protection more complete.

Because a child’s future is not only a future in which something dangerous might happen.

It is also a future containing identity, relationships, memory and belonging.

Those things can be harmed too.

And this brings me back to the question that started this observation.

Modern child protection has developed a powerful ability to look ahead.

It tries to see harm before harm arrives.

That is an important achievement.

But perhaps the next step in institutional maturity is to turn that same capacity, occasionally, towards itself.

Not because the institution is the danger.

Not because intervention is inherently wrong.

But because powerful decisions also have futures.

And when those futures may become difficult or impossible to reverse, they deserve to be imagined before they arrive.

We should not have to wait thirty years for hindsight to ask questions that can still be asked today.

Perhaps one of the lessons of history is not simply that we should learn from past mistakes.

Perhaps it is that we should become better at recognising the possibility of our own mistakes before their consequences become history.

That seems worth asking now.

While the future is still the future.