Danny Stone MBE is the Chief Executive of the Antisemitism Policy Trust.

Throughout the period of Jeremy Corbyn’s leadership of the Labour party, one of the most heinous and upsetting elements for me was the constant reference to allegations and experiences of anti-Jewish racism as a smear.

As I pointed out repeatedly at that time, and still do, our British approach to hate crime, learned so painfully through the Macpherson process which followed the racist murder of Stephen Lawrence, is to treat those reporting incidents with proper care and to investigate with due diligence.

The definition of hate is victim-led when it comes to reporting, and rightly so.

Of course, the criminal law also rightly sets a high threshold for illegal hate, so that our freedoms are not impinged, and of course not every report will be a crime. Moreover, there will always be those that seek to make mischief of some sort.

However, this approach, of victim-led hate reporting has had significant and important positive impacts for police, and communities, in diagnosing harm, extremism, and failing integration or community-cohesion efforts.

The police message to affected communities is clear: that they want to encourage victims to come forward. We all know that recording rules can be complex, but they should not be a concern to a victim at their time of distress. The police want us to tell them when we experience serious harm and let them decide which category it falls into.

Whether that ends up as a hate crime, or a non-crime hate incident, the impact on our sense of safety can be exactly the same. Our collective learning from many inquiries into horrific crimes is that there were signs of escalating risk that could, if properly assessed and reacted to, have in some cases prevented death.

Take the tragic case of Fiona Pilkington, who killed herself and her severely disabled daughter Francecca Hardwick some 15 years ago, following a decade of torment by local youths. The 33 reported incidents to the police were never considered as a series, as they targeted different members of Pilkington’s family and at different locations, but the accumulative effect on the family brought her to a situation to take unthinkable action to remove her daughter from such a hostile environment.

Many of those incidents should have been recorded as non-crime hate incidents and the Independent Police Complaints Commission report found that the failure to consider the overall harm to the family prevented the interventions that could have protected them and prevented the tragedy.

Despite this, more recently non-crime hate incidents have come in for criticism.

Harry Miller was interviewed by Humberside Police over allegedly transphobic tweets. He wasn’t charged and then took the police to court to challenge their right to retain a record of the incident. He won on the specific claim against Humberside, but the court ruled that in general the police could continue to retain data relating to non-crime hate incidents.

From here, the debate has turned largely to questions of free speech and the impact of residual data handling, though this has also become twinned with interconnected discussion of transphobia and women’s rights. Much of the debate of late has centered on this issue, but discussion can become febrile and even extreme.

We should not lose sight of other fractures in society and the role played in recording them. Holocaust denial and antisemitic hate speech will often be sub-criminal, but society recognises the harm in the denial of historical facts surrounding this unique atrocity and others. We have many examples of individuals who have been influenced by hate and misinformation and have been motivated to acts of extreme violence.

At the heart of the debate is the need to balance often conflicting rights to free speech and protection from harm. What our agencies and our society need is a system that can recognise and seek to balance these rights, and recognise when conflict has the potential to escalate to harm.

A backbench peer sought to amend the Police, Crime, Sentencing and Courts Bill, currently making its way through Parliament, to regulate non-crime hate incidents. Though that amendment was withdrawn, it gathered widespread support from across the House; so much so that the Government took the matter away and intends to produce its own Code of Practice to dictate the recording of these non-crime incidents.

That Code of Practice may well turn out to be good but for me, the action has been taken without enough of a case having been made for these incidents being captured in the first place. As the Bill makes its way for Royal Assent, it is important that it is.

In Britain, organisations working to record incidents of antisemitic, homophobic, and anti-Muslim attacks are broadly considered to report data that is authoritative, robust and reliable. In some cases, there are data sharing agreements with the police to anonymise and exchange incident details.

If a new law, or indeed code, is drawn up in such a way that it prevents local police and non-police bodies from retaining non-crime hate incident data then that could be devastating, not just for those organisations involved but for our national ability to understand, interrogate and take action to address many forms of harm.

We must remain victim-led in the reporting of hate, not to do so would be an abdication of our duty to learn the hard lessons from a racist past.