The Presumption is Going. But Without Criminal Justice Reform, Family Courts Will Keep Making Decisions in a Vacuum

The repeal of the presumption of parental contact is being presented as a watershed moment in child protection. And it is. For years, survivors and children have lived with the reality that contact was treated as the default, even in cases where abuse was alleged, minimised or reframed as “conflict”.

This reform is a significant and welcome step forward. It reflects years of survivor advocacy and a growing recognition of the risks children face in abusive contexts. But legal reform on paper does not automatically translate into safer outcomes in practice.

Removing a presumption in family law, however, does not solve a deeper problem.

Family courts do not operate in isolation. While they make independent welfare decisions on the balance of probabilities, they rely heavily on information generated elsewhere: police investigations, CPS charging decisions, risk assessments and any criminal findings. When coercive and controlling behaviour is not prosecuted, is undercharged or results in minimal sentences, family courts are often left assessing complex patterns of risk without clear or consistent criminal justice signals.

That vacuum is where unsafe decisions can still be made, even with the best judicial intentions.

The Case That Illustrates the Gap

There are publicly discussed campaign cases, including that of Skye Nicholls, where a perpetrator was convicted of coercive and controlling behaviour and the sentencing judge explicitly linked the conduct to the victim’s eventual suicide, only for the sentence to be reduced on appeal and the assessed risk downgraded.

The precise legal details of individual cases may vary across reports and appeals, but the principle they illustrate is deeply concerning: a victim dies, coercive control is recognised, yet the final criminal justice signal about risk becomes weaker, not stronger.

A victim dies. The perpetrator is convicted of coercive control. The sentencing outcome is relatively short. The risk level is later reassessed as lower than originally determined.

What does that leave the family court with?

A conviction that does not clearly communicate lethality risk. A sentence that appears inconsistent with the severity of harm. A downgraded risk classification despite the most serious possible outcome.

This is not simply about punishment. It is about signalling. Criminal justice outcomes are one of the primary ways risk is communicated across systems. When that signal is weak, inconsistent or ambiguous, other courts are forced to make safeguarding decisions without a stable evidential anchor.

A Legislative Win That Does Not Solve the Structural Problem

The removal of the presumption of contact is an important corrective. It acknowledges what survivors have long known: that contact is not automatically in a child’s best interests where abuse is present.

But the reform assumes something crucial. It assumes that the criminal justice system is reliably identifying, prosecuting and evidencing coercive control in the first place.

That assumption does not hold.

Conviction rates for coercive and controlling behaviour remain low relative to prevalence. Where prosecutions do occur, sentences vary significantly. Risk assessments can shift between stages of the process. Appeals can alter the narrative that other safeguarding bodies rely upon.

The result is that family courts are still asked to assess patterns of psychological harm without the benefit of consistent criminal findings.

Why Coercive Control Is So Hard to Prosecute

Coercive control is not an incident-based crime. It is a pattern-based one.

The evidential model of criminal justice, however, is still heavily oriented around discrete, provable events: an assault, a threat, a single identifiable act. Coercive control rarely presents like that. It is cumulative. It is subtle. It often looks trivial in isolation and devastating only in aggregate.

Perpetrators who use coercive control frequently exploit this structure. They may present as calm, credible and socially adept to professionals, while privately engaging in behaviour that steadily erodes the victim’s autonomy and sense of reality. The public persona and the private conduct do not align and the discrepancy itself can make evidencing the behaviour harder.

This creates a fundamental mismatch. The law requires proof beyond reasonable doubt of a course of conduct. The harm is experienced as a gradual psychological diminishment. The incidents that make up that course of conduct can appear minor when viewed individually, even though the overall pattern is anything but.

Passive Systemic Blindness vs Active Perpetrator Exploitation

It is important to distinguish between two dynamics that operate simultaneously.

The first is passive systemic blindness. Different agencies record different fragments of information. Police logs, GP records, school concerns and housing reports may all capture aspects of risk, but no single system always sees the whole pattern. Reviews of serious harm and domestic homicide cases have repeatedly found failures in information sharing and pattern recognition across agencies.

The second is active perpetrator exploitation. Some perpetrators deliberately behave in ways that appear reasonable or benign to outsiders while intensifying psychological control in private. They understand that systems look for visible incidents. They keep their behaviour just below obvious criminal thresholds, creating a trail of concerning but individually explainable events.

The law is therefore trying to prove a pattern in a landscape where the data itself is fragmented and where some perpetrators are consciously shaping how they appear within each silo.

The Eight Stages and the Escalation Problem

The research on domestic homicide timelines shows that coercive control is not a late-stage development. It is often present from the very beginning. What may initially appear as intense affection, rapid commitment or “love-bombing” can be the early grooming phase, drawing the victim into the perpetrator’s sphere of influence. This is followed by boundary testing, increasing surveillance and gradual isolation.

By the time professionals encounter the case, the pattern may already be deeply embedded, even if no single incident appears serious in isolation. The timeline is therefore not about when coercive control begins, but how it escalates and becomes more dangerous over time.

If criminal justice responses do not reliably capture these early stages of coercive control, family courts may only see risk once escalation has already occurred. By then, the opportunity for early safeguarding intervention has often already been lost.

Inconsistent Risk Signals and Their Consequences

Risk assessment is not merely a descriptive exercise. It is communicative. A high-risk classification tells other agencies that the behaviour presents serious ongoing danger. A reduced classification or a minimal sentence, can unintentionally signal that the conduct was less serious than it truly was.

This matters enormously in post-separation contexts. Research and practice experience consistently show that risk can increase after separation, precisely when perpetrators lose direct control. At that stage, they may engage with multiple systems simultaneously: family court, police, child contact arrangements and sometimes housing or social services.

If those systems receive inconsistent signals about risk, the perpetrator’s behaviour can appear less concerning than the victim’s lived experience would suggest. The fragmentation does not just obscure harm. It can actively enable continued control across institutional settings.

The Child Protection Register and the Limits of Data Integration

Proposals for child abuse or child protection registers and broader data-sharing reforms, are often framed as the solution to fragmentation. A single identifier linking information across agencies could, in theory, allow patterns of concern to be recognised more easily.

But data integration cannot compensate for weak underlying criminal outcomes.

If coercive control is not investigated, not charged or results in minimal sentencing, then the data being integrated will still underrepresent the true level of risk. We will simply be connecting incomplete signals more efficiently. Integration is valuable, but it does not replace the need for consistent recognition and prosecution of pattern-based abuse.

The Vacuum Facing Family Courts

When criminal justice outcomes are inconsistent, family courts are left to fill the gap. Judges must evaluate allegations of coercive control without always having the benefit of clear criminal findings and often with limited evidential tools compared to the criminal courts.

This is an unenviable position. Family courts are tasked with making forward-looking risk decisions for children, sometimes in the absence of definitive criminal determinations about the perpetrator’s behaviour. The repeal of the presumption of contact acknowledges that risk must be taken seriously, but it does not provide the robust criminal justice evidence that would make those decisions more straightforward and safer.

What Needs to Change

If the repeal of the presumption of contact is to achieve its intended safeguarding impact, it must be accompanied by reform within the criminal justice response to coercive control.

That includes:

• A more consistent prosecutorial approach to pattern-based abuse

• Judicial training that recognises the cumulative and potentially lethal nature of coercive control

• Sentencing practices that clearly communicate seriousness and risk

• Investigative models that focus on courses of conduct rather than isolated incidents

Without these changes, family courts will continue to operate with incomplete risk pictures. They will be asked to assess complex patterns of psychological harm without the consistent backing of criminal convictions that accurately reflect those patterns.

The Core Problem

Coercive control is often described as “murder in slow motion”. That phrase is powerful, but legally it presents a challenge. How do you prove something that is subtle, cumulative and often deliberately concealed behind a socially acceptable public persona?

The answer cannot be that the harm is simply too complex to evidence. Survivors document. They preserve messages. They obtain witness accounts. They report patterns over time. Yet even then, the translation of that material into criminal charges is not guaranteed.

This does not mean the system is indifferent. It means the system is still adapting to a form of harm that does not fit neatly within traditional evidential models.

Conclusion

The repeal of the presumption of contact is a necessary and welcome reform. It recognises that the safety of children and survivors must not be subordinated to assumptions about parental involvement.

But legislative change in family law cannot, on its own, solve the deeper evidential and prosecutorial challenges of coercive and controlling behaviour. Until criminal justice responses consistently recognise, charge and sentence pattern-based abuse in a way that accurately communicates risk, family courts will continue to make decisions in an evidential vacuum.

The presumption may be going. The vacuum remains.

These issues rarely arise from lack of care. They arise from blind spots in how systems interpret risk, patterns and evidence. My work at Safe Haven Education focuses on helping organisations and professionals recognise those blind spots before they lead to unsafe outcomes.