Can Abusers Really Change? What the Research Shows

Can abusers really change? It's one of the most searched questions in this entire area, and most of the answers you'll find online are either a flat no or a hopeful maybe. Neither is accurate, so let's look at the research.

If you're asking it, you're probably asking about a specific person. Maybe you've left and part of you is still waiting to hear they've become someone different. Maybe you're still in it, and they've promised things will change, and there's been a good few weeks that make you want to believe them. Or maybe you're co-parenting and you need to know whether the person you're dealing with is ever going to be different to deal with.

The research does have an answer to this, and it's clear enough to be genuinely useful. Change is possible. It's also rare, it depends on conditions that are almost never present in an ordinary relationship, and there's one particular kind of change that the evidence says is much harder to achieve than the rest.

What the Research on Perpetrator Programmes Shows

The largest UK study of this question is Project Mirabal, a multi-site evaluation of domestic violence perpetrator programmes in England and Wales, led by Liz Kelly at London Metropolitan University and Nicole Westmarland at Durham University and published in 2015.

What made it unusually valuable is that it didn't just count incidents. It measured success across six different dimensions, including something the researchers called space for action, meaning the degree to which partners got their freedom and agency back.

On physical and sexual violence, the findings were striking. The researchers reported that these forms of abuse weren't just reduced but ended for the majority of women whose partners attended the programmes. Some of the specific figures are stark: the proportion of women reporting sexual coercion, and the proportion reporting a weapon being used against them, both fell from around three in ten before the programme to zero a year after it started.

That's a genuinely important finding. Structured intervention can stop serious physical and sexual violence for a lot of people.

However, the same research found something that matters enormously if the relationship you're thinking about was primarily controlling rather than violent. The controlling behaviours, the excuse-making, and the patterns that sat underneath the violence continued for a significant proportion of men even after they'd been through a programme. Physical violence proved considerably easier to stop than coercive control.

Why Coercive Control Is Harder to Change Than Violence

That finding makes complete sense once you understand where each kind of behaviour comes from.

Physical violence is a behaviour, and behaviours can be targeted directly. You can learn to recognise escalation, remove yourself, and choose not to act. A great deal of intervention work is built around exactly that, and it works.

Coercive control isn't a single behaviour. It's the expression of a belief system - the entitlement schema, the conviction that one person's needs and comfort take precedence over the other's, and that managing the other person is a reasonable way to secure them. You can stop hitting someone without ever touching that belief. So while the belief stays intact, it simply finds other ways to express itself.

This is why someone can genuinely stop being physically violent and the relationship can still feel exactly as suffocating as it did before. The most visible symptom has gone, but the thing producing it is still there. Kind of like treating a fever without touching the infection.

It also explains a pattern people describe constantly: the partner who changed, and yet somehow nothing really changed at all. They stopped shouting and other obvious, overt things. However, the rules, the monitoring, the atmosphere and the sense of walking on eggshells all carried on regardless.

What Real Change Actually Requires

The research and the practice literature are fairly consistent about the conditions change depends on, and they're demanding.

It requires structured, external intervention. Belief systems don't get rebuilt through good intentions inside the environment that formed them. Something outside the relationship has to challenge the thinking directly and hold the person to it.

It requires sustained engagement over a long period of time. Not a few weeks of effort after a crisis, but months of work, with the old patterns tested repeatedly under real conditions.

It also requires the person to actually engage with the work, rather than simply attending it. There's a significant difference between someone who goes to a programme because they've been told to and someone who has started to see their own thinking as the problem, and the outcomes tend to reflect that difference.

The Mirabal researchers observed that one of the things programmes achieved was shifting how men understood their own behaviour - moving from talking about stand-alone incidents of violence towards beginning to recognise ongoing coercive control. That shift in understanding is where real change starts. It's also the point at which a lot of people stall.

Why It Almost Never Happens Inside the Relationship

This is the really important part for you to know.

Almost none of the conditions above can be supplied by a partner. You can't be the external challenge, because you're inside the system being challenged. You can't provide structured intervention, because you're the person the thinking is organised around. You can't make someone engage with their own belief system, no matter how clearly you describe it or how long you're prepared to wait.

That's why the improvements that do appear inside relationships so often follow a recognisable pattern. They come after a crisis, after you pull away, or after you say you're leaving. They last as long as the pressure lasts. Then once things feel secure again, they fade away. That isn't transformation, it's stabilisation - the relationship being returned to functional. This is a very different outcome from the person truly changing.

It also means that the waiting has a cost, and it's usually borne entirely by you. Every month spent hoping for a change the conditions don't support is a month spent inside the same pattern, adjusting yourself around it.

The Honest Answer

So can abusers change? Some do. It's possible, and the evidence shows it happening, particularly where physical violence is concerned and particularly within properly structured programmes.

However, the kind of change most people are hoping for - the belief system underneath coercive control shifting - is the hardest kind to achieve, it happens least often, and it almost never happens in response to a partner's patience, love or effort.

Because of this, wanting to know if they could change may not be the most useful place to start. The focus is better spent on what you do while you're waiting to find out, and whether waiting is a decision you'd make if you could see the evidence clearly.

Where to go next...

That decision is exactly what Module 4 of The Relationship Reality Check is built around. It gives you a research-grounded answer to whether change is realistic in your specific situation, explains why the pull back keeps coming even when the evidence says otherwise, and takes you through the Decision Clarity Grid™ so you can make a choice from clarity rather than hope, fear or exhaustion. The cohort starts on Monday 28 September, and enrolment closes on Sunday 27th.

👉 [Enrol in The Relationship Reality Check here] 👈

If you read this after the deadline, you can join the waitlist for the next cohort here:

👉 [Join the Waitlist] 👈

This week's YouTube video goes further into this from inside perpetrator intervention work:

👉 [Watch the YouTube video here] 👈

For the connected mechanisms, Why Some People Can Never Take Accountability: The Psychology of the Entitlement Schema covers the belief system that makes coercive control so persistent, and Why They Never Took Responsibility for Anything explains the thinking habits that sit on top of it.

If you're not on the Clarity Drop list, it's one email every Tuesday breaking down a single psychological concept from that week's topic, with a practical tool you can actually use and first access to everything new from THRC.

[Join The Clarity Drop weekly email here]