This is not simply a new criminal offence

The most interesting thing about what is happening in Canada is not one statute.

It is the direction of travel.

Canadian law is increasingly recognising coercive control in three distinct but connected ways.

That matters because coercive control has always been difficult for legal systems built around identifying discrete events.

A punch is an event.

A threat is an event.

A particular abusive message is an event.

Coercive control is often something different.

It may consist of hundreds of individually explainable acts which, when viewed together, establish an architecture of domination: surveillance, isolation, intimidation, financial restriction, manipulation of children, threats, humiliation, regulation of everyday life, interference with relationships, litigation pressure and the systematic erosion of another person’s autonomy.

Look at only one incident and the abuse may appear insignificant.

Look at the pattern and the case may look completely different.

Canada has recognised coercive control in family law since 2021

Canada did not suddenly discover coercive control in 2026.

Important amendments to the federal Divorce Act came into force on 1 March 2021.

The Act defines “family violence” broadly. Crucially, the definition is not restricted to behaviour which constitutes a criminal offence.

It includes violent or threatening behaviour, conduct causing fear for someone’s safety and a pattern of coercive and controlling behaviour.

It can include:

• physical abuse;
• sexual abuse;
• threats to kill or cause bodily harm;
• harassment and stalking;
• psychological abuse;
• financial abuse;
• failure to provide necessities;
• threats against animals or property; and
• harm to animals or damage to property.

The definition is particularly important for children because it expressly includes a child’s direct or indirect exposure to family violence.

Children do not need to be physically assaulted to be affected by abuse

This is an area where both Canadian law and the law of England and Wales have moved significantly.

A child can be harmed by domestic abuse even where the abusive parent never lays a hand on the child.

Children can live inside the consequences of coercive control.

They may watch one parent modify everything they say and do to avoid another parent’s reaction.

They may understand which subjects cannot safely be discussed.

They may become messengers.

They may be questioned after spending time with the other parent.

Their telephone contact may be monitored.

Their relationships with grandparents, siblings or the other side of their family may become contested territory.

They may learn that affection itself carries consequences.

They may therefore experience the abusive system even when they are not its original target.

Canada’s Divorce Act expressly requires courts examining family violence to consider whether a child was directly or indirectly exposed to it and the resulting physical, emotional or psychological harm or risk of harm.

This is not a peripheral issue. It is part of the best-interests assessment.

Canada asks courts to identify the pattern

Section 16 of the Canadian Divorce Act requires courts dealing with family violence to consider a number of factors, including:

• the nature, seriousness and frequency of the family violence;
• when it occurred;
• whether there is a pattern of coercive and controlling behaviour;
• whether the violence was directed at the child;
• whether the child was directly or indirectly exposed to it;
• physical, emotional and psychological harm or risk of harm;
• whether the safety of the child or another family member has been compromised;
• whether the behaviour caused fear for safety; and
• what steps, if any, the person responsible has taken to prevent further violence and improve their parenting.

Justice Canada’s explanatory material goes further.

It recognises that coercive and controlling family violence is cumulative.

It also recognises something practitioners dealing with post-separation abuse will immediately understand: children and parenting arrangements themselves may be used as instruments of continued control.

Justice Canada gives examples including refusing to comply with parenting orders or threatening the former partner with the loss of parenting time.

The critical question is often not “What happened on Tuesday?” It is “What system of behaviour existed before, during and after Tuesday — and what did that system do to the adults and children living inside it?”

Then came Ahluwalia v Ahluwalia

On 15 May 2026, the Supreme Court of Canada delivered a major judgment in Ahluwalia v Ahluwalia.

The case concerned whether Canadian law should recognise a civil claim capable of addressing the distinctive harm caused by intimate partner violence.

The Court concluded that existing torts did not fully capture that harm and recognised a new tort of intimate partner violence.

This is important because traditional civil claims may identify particular wrongs — such as assault or intentional infliction of emotional distress — while still missing what coercive control does as a whole.

The Supreme Court described intimate partner violence in terms of coercive and controlling conduct capable of undermining a person’s dignity, autonomy and equality.

It is not limited to physical violence.

Relevant behaviour may include:

• isolation;
• humiliation;
• surveillance;
• financial control;
• sexual coercion;
• intimidation; and
• other behaviour which, understood contextually, amounts to coercive control.

The significance is difficult to overstate.

The law is recognising that there can be a distinct injury in stripping another person of meaningful autonomy over time.

That injury cannot always be understood by separating the relationship into a spreadsheet of incidents.

The new Canadian criminal offence

Canada then took another major step.

The Protecting Victims Act, formerly Bill C-16, received Royal Assent on 18 June 2026.

Among a wider package of criminal justice reforms, the legislation creates a specific offence aimed at a pattern of coercive or controlling conduct towards an intimate partner.

The legislation is intended to capture combinations or repeated instances of conduct including violence, sexual coercion and other controlling behaviour which would reasonably be expected to cause an intimate partner to believe that their physical or psychological safety is threatened.

The statutory scheme identifies areas of a victim’s life which coercive behaviour may attempt to control, including:

• social interactions;
• communications;
• childcare decisions;
• employment;
• education;
• finances;
• property;
• physical appearance;
• access to healthcare or medication;
• expression of opinions; and
• participation in linguistic or cultural communities.

Threats of suicide or self-harm used as part of a coercive pattern may also fall within the framework where the statutory requirements are satisfied.

The new offence carries a maximum sentence of 10 years’ imprisonment on indictment.

But there is an important caveat: it is not yet operational

The reason given for the delayed commencement is particularly telling.

Canada wants time for training and implementation.

That is a lesson in itself.

Coercive control legislation can fail if police, lawyers, social workers, prosecutors, judges and safeguarding professionals continue to search only for incidents.

You cannot legislate for patterns of abuse and then train professionals to look exclusively for episodes.

Nor can a legal system sensibly say that coercive control is cumulative and contextual while forcing victims to reduce years of behaviour to a small number of decontextualised allegations.

Canada has also connected coercive control with femicide

The 2026 reforms go further again.

Canada’s new legislation changes the treatment of certain murders occurring in circumstances involving coercive control, hate, sexual violence or exploitation and expressly uses the term femicide where the victim is female.

That is significant because coercive control is not merely being treated as unpleasant relationship behaviour.

It is being recognised within a risk continuum.

This reflects something domestic-abuse specialists have argued for years: coercive control is relevant not only because of the psychological damage it causes but because patterns of domination, stalking, threats, possessiveness and escalating post-separation behaviour can also be important indicators of serious risk.

Separation does not necessarily end coercive control.

In some cases, it changes its form.

So how does England and Wales compare?

England and Wales should not be presented as legally backward on coercive control.

We already have substantial legislation.

Section 76 of the Serious Crime Act 2015 created the offence of controlling or coercive behaviour.

The Domestic Abuse Act 2021 subsequently broadened the framework, including removing the previous requirement that perpetrator and victim be living together for the coercive-control offence to apply.

That was crucial for recognising post-separation abuse.

The Domestic Abuse Act also provides a broad statutory definition of domestic abuse encompassing:

• physical or sexual abuse;
• violent or threatening behaviour;
• controlling or coercive behaviour;
• economic abuse; and
• psychological, emotional or other abuse.

Importantly, section 3 recognises children as victims of domestic abuse where they see, hear or experience the effects of abuse between connected adults and are related to them.

So England and Wales already possess many of the building blocks.

The difficulty is often not whether the law recognises coercive control.

The difficulty is how reliably the family justice system identifies, evidences and responds to it.

PD12J is stronger than many litigants realise

Practice Direction 12J governs child-arrangements proceedings where domestic abuse is alleged, admitted or there is other reason to believe that domestic abuse has occurred.

It expressly recognises controlling and coercive behaviour.

It requires courts to consider domestic abuse at every stage of proceedings and, where relevant, identify factual and welfare issues and ensure that child arrangements do not expose a child or parent to further harm.

Where abuse is established, the court may need to consider its effect on:

• the child;
• the child’s relationship with each parent;
• the parent with whom the child lives;
• future arrangements;
• the likelihood of future abusive behaviour; and
• each parent’s capacity to understand the effect of the abuse.

PD12J also directs attention to whether a parent is genuinely promoting the child’s welfare or using the proceedings to continue domestic abuse against the other parent.

That provision should be read carefully.

Because post-separation coercive control can migrate directly into family proceedings.

The family court itself can become part of the machinery of control

Litigation is not inherently abusive.

Parents must be able to bring genuine welfare disputes before a court.

But proceedings can be weaponised.

That may happen through repeated applications, unnecessary communications, relentless allegations, financially exhausting procedural activity, demands for information, manipulation of handovers, monitoring through children or continued attempts to exercise control over the other parent’s daily life.

Parliament expressly recognised this risk when reforming section 91(14) of the Children Act 1989.

PD12J and Practice Direction 12Q acknowledge that further applications may themselves form part of a pattern of controlling or coercive behaviour.

The court can therefore restrict further Children Act applications without permission where the statutory criteria are met.

That is a safeguarding power, not simply a case-management convenience.

Re H-N: stop looking at abuse through a keyhole

The Court of Appeal’s decision in Re H-N and Others (Children) (Domestic Abuse: Finding of Fact Hearings) [2021] EWCA Civ 448 remains central to understanding coercive control in private children proceedings.

One of its most important implications is that the court must not lose the overall evidential picture by treating allegations as completely disconnected events.

Coercive control is, by definition, about pattern and context.

A traditional schedule of allegations may sometimes assist in identifying disputed facts.

But it can also become a poor analytical tool where the allegation is fundamentally that numerous behaviours worked together to establish domination, fear, dependency or control.

This concern has now been reinforced procedurally.

Amendments to PD12J taking effect from 5 January 2026 expressly moved away from an overly rigid focus on “Scott Schedules” and emphasised a holistic assessment of domestic-abuse allegations, including patterns of coercive and controlling behaviour.

The real comparison: Canada has made the pattern unusually explicit

England and Wales recognise coercive control.

Canada recognises coercive control.

The difference is therefore more subtle than “Canada has it and we do not”.

The interesting feature of the Canadian approach is the degree to which the same underlying concept is becoming visible across multiple legal domains.

What England and Wales should learn from Canada

The lesson is not necessarily that Parliament needs another definition.

We have definitions.

We have statutory guidance.

We have PD12J.

We have appellate authority.

We have specific recognition of post-separation coercive control.

The harder question is whether the system consistently translates those principles into decision-making.

1. Pattern analysis needs to become routine

Courts should ask not merely whether individual allegations are proved but whether the evidence as a whole establishes a pattern and, if it does, what that pattern means for parenting capacity, risk and the child’s welfare.

2. Post-separation conduct matters

Separation should not create an artificial evidential cut-off.

Coercive control may continue through finances, digital surveillance, children, correspondence, handovers, schools, healthcare decisions or litigation.

3. Behaviour involving children needs to be analysed in context

A dispute about a child’s telephone, contact arrangements, passport, clothing, school information or relationships with relatives may be an ordinary parenting disagreement.

Or it may form part of a wider pattern.

Context decides which.

4. Professionals need pattern-recognition training

Canada’s decision to delay commencement of its new criminal offence while training is developed deserves attention.

Legislation cannot recognise coercive control effectively if professional practice repeatedly atomises it.

5. The child’s experience must not disappear behind the parental dispute

Family proceedings often speak about domestic abuse as though it concerns the adults and child arrangements as though they concern the child.

In reality, those questions may be inseparable.

A parent who controls, intimidates or terrorises the other parent may affect that parent’s capacity to care freely, the emotional environment in which the child develops and the child’s own understanding of relationships, loyalty, safety and autonomy.

6. Absence of prosecution is not absence of abuse

This point is fundamental.

Family courts work to a different evidential standard and perform a different function from criminal courts.

A person does not need to have been prosecuted or convicted before domestic abuse can be relevant to a welfare determination.

Canada makes this particularly explicit in its definition of family violence.

England and Wales should be equally clear in practice.

For litigants in person: evidence the pattern, not just the worst day

People experiencing coercive control often make one of two understandable mistakes when preparing family-court evidence.

Some describe everything.

The evidence becomes hundreds of pages long and the central pattern disappears.

Others describe only the most serious incidents.

The court then sees several apparently disconnected events and never understands the controlling system surrounding them.

The answer is not necessarily more evidence.

It is better organised evidence.

The objective is not to attach a sinister interpretation to every disagreement.

It is to make the evidence capable of answering the correct legal question.

Is this ordinary conflict?

Is it a handful of isolated incidents?

Or is there evidence of a sustained pattern in which one person has attempted to subordinate, frighten, isolate, monitor or control another?

Conflict and coercive control are not the same thing

This distinction matters enormously in private children proceedings.

“High conflict” can become a dangerously imprecise label.

Two adults arguing with each other is not necessarily the same phenomenon as one adult establishing a system of coercive control over another.

Calling both situations “conflict” risks flattening the power dynamics.

It may also create false symmetry.

A victim’s resistance, anxiety, protective behaviour or refusal to cooperate with unsafe arrangements should not automatically be treated as equivalent to the behaviour which produced that response.

Context matters.

Pattern matters.

Power matters.

And the child’s lived experience of that dynamic matters.

The deeper significance of Ahluwalia: autonomy itself can be harmed

Perhaps the most intellectually important aspect of the Canadian Supreme Court’s 2026 decision is its recognition that coercive control is capable of producing a distinctive injury to autonomy and equality.

That is useful far beyond damages litigation.

It helps explain why domestic abuse cannot always be measured by visible injury.

Imagine living for years while another person determines who you speak to, what money you can access, what you wear, where you go, when you sleep, what opinions you can safely express, how you parent, who you can contact and what consequences follow if you disobey.

There may be no single incident capable of explaining that experience.

The harm lies partly in the accumulated destruction of freedom.

That is precisely why coercive control requires contextual legal analysis.

And this matters profoundly in family court

Parenting requires more than an ability to feed a child, provide accommodation and get them to school.

It involves recognising that a child is an autonomous developing person with relationships, feelings, loyalties and emotional needs of their own.

Where a parent has engaged in coercive control, the family court may need to ask difficult welfare questions.

Can that parent distinguish the child’s needs from their own need for control?

Can they support the child’s relationship with the other parent?

Can they tolerate the child holding affection for people they dislike?

Can they respect appropriate privacy and boundaries?

Can they make joint decisions without using parental responsibility as leverage?

Do they understand the impact of the abuse?

Has the behaviour genuinely changed?

These are welfare questions, not punishment.

What I hope happens next in England and Wales

I do not think the answer is simply to keep passing legislation and assume the problem has therefore been solved.

The statutory architecture in England and Wales is already considerably stronger than many people realise.

What we need is consistent implementation.

I would like to see:

• stronger and continuing judicial education on coercive control and post-separation abuse;
• greater consistency between PD12J on paper and its application in individual cases;
• proper use of holistic evidence rather than mechanically reducing abuse to isolated incidents;
• clear analysis of the distinction between conflict and coercive control;
• greater recognition of children as victims of the abusive environment, not merely witnesses to adult behaviour;
• careful examination of the use of children, communication and litigation as potential mechanisms of post-separation control;
• better coordination between criminal justice, family justice, safeguarding and domestic-abuse services;
• greater emphasis on behaviour, impact and risk rather than stereotypes about how a “real victim” ought to present;
• proper examination of whether a perpetrator demonstrates insight and sustained behavioural change; and
• family-court decisions which explain transparently how domestic-abuse findings have affected the welfare analysis.

Most of that does not require inventing a new legal concept.

It requires using the concepts we already have properly.

Canada is worth watching

Canada has not solved domestic abuse.

No legislation can make that claim.

Nor should comparative legal analysis become a competition in which one jurisdiction is declared “better” than another.

England and Wales have themselves been internationally significant in criminalising coercive and controlling behaviour.

But Canada’s developing framework deserves serious attention.

Its federal family law tells courts explicitly to look for coercive and controlling patterns.

Its Supreme Court has recognised that intimate partner violence can cause a distinctive injury to dignity, autonomy and equality which traditional incident-based torts do not fully capture.

Its Parliament has now enacted a specific criminal coercive-control offence with a maximum penalty of ten years on indictment.

And, tellingly, Canada is delaying implementation of that offence to allow time for training.

There is a message in all of that.

Coercive control is not merely a category of behaviour. It is a way of understanding abuse.

Once the law accepts that, procedure, evidence gathering, professional training and judicial analysis need to follow.

That is where the next stage of reform in England and Wales should concentrate.