Rethinking Victimology

When we talk about victimology in violence against women and girls (VAWG), the conversation usually starts with a demographic profile: age, relationship to perpetrator, prior victimisation, economic position. This is necessary groundwork. But it’s not the whole picture. A critical criminology approach asks a second, harder question underneath it: who gets recognised as a victim in the first place, and who doesn’t, and what happens to the women who fall outside that recognition once they come into contact with the very institutions meant to protect them.

The standard profile

Most victimology frameworks in this field cover:

• Demographics: age, race, class, disability, migration or immigration status, sexuality
• Relationship to perpetrator: partner, family member, acquaintance, colleague, or institutional actor (police, employer, the state itself)
• Victimisation history: prior abuse, repeat victimisation, how the harm escalated over time
• Structural position: economic dependency, housing precarity, language barriers, immigration status tied to an abuser
• Institutional contact: police reports, health service visits, refuge use, and the points at which these contacts failed to intervene

This profile is meant to give a factual map of who is affected and how. What it doesn’t explain is why some of these women show up clearly in the data and official record, and others barely show up at all.

One gap sits in that second category. “Relationship to perpetrator” tends to assume the bond is either romantic, biological, or casual, leaving no space for close, chosen relationships that are neither: found family, long term flatmates who have become carers, faith community ties, or friendships that function as a primary support structure rather than a casual acquaintance.

In England and Wales, the Domestic Abuse Act 2021 defines abuse based on being “personally connected”, which covers intimate partners, former partners, and relatives. A woman abused by someone she considers family in every practical sense, but who isn’t legally family and was never her partner, may not be recognised as experiencing domestic abuse at all under that statutory definition. She’d be routed through generic violence or harassment frameworks instead, which don’t carry the same specialist response, risk assessment tools, or refuge eligibility.

The ideal victim problem

Nils Christie’s concept of the “ideal victim” is useful here. Christie argued that society extends full victim status, belief, sympathy, protection, mainly to people who fit a narrow template: weak, blameless, doing nothing to provoke harm, harmed by a stranger they couldn’t reasonably have avoided. He developed this in a civil society model, where visibility was mediated by a small number of accountable gatekeepers: police, courts, the press.

That model no longer holds. Society now operates closer to Foucault’s panopticon, where everyone is potentially visible to everyone else, and where people begin regulating their own conduct in anticipation of being watched and judged, whether or not anyone is actually watching at that moment. For a victim, this means credibility is no longer built through one encounter with a police officer or one court appearance; it’s built and unravelled continuously, in public, often by an audience with no accountability and no stake in the outcome.

Almost no real case of VAWG fits Christie’s original template cleanly. Most abuse happens within relationships, not from strangers. Many women have complicated histories: substance use, prior contact with police as suspects rather than victims, sex work, immigration status that makes reporting dangerous rather than protective. Some fight back. None of that makes the violence they experience less real. But it changes how they’re treated, both by institutions and by the wider public watching. A woman who is white, sober, married, and composed is far more likely to be believed than a woman who is not, and in a panopticon society, that composure now has to be performed not just for police, but for anyone who might be watching. This gap isn’t incidental, it’s structural, and it now operates through self-surveillance as much as through institutional judgement.

This matters beyond visibility. A zemiological view of harm, one that looks past what is formally criminalised to what actually damages people, is needed here, because much of what happens in this panopticon plays out entirely outside any criminal process. A woman whose case doesn’t fit the template can face coordinated disbelief, harassment, and doxxing from a crowd with no formal role in her case at all. False narratives spread this way don’t just fail to protect real victims, they sometimes wrongly brand someone else as a perpetrator, generating a different kind of harm again. None of it requires a court, a police force, or even a named institution to inflict damage; a platform’s incentive structure and an anonymous crowd are enough.

Flipping the question: why do police forces blame victims?

Most accounts of victim-blaming treat it as an individual failing: an officer with poor training, a bad attitude, a lack of empathy. That framing is comforting because it implies the fix is simple: better training, better guidance, a new policy document. The evidence from inquiry after inquiry into UK policing (Casey, the Angiolini Inquiry, HMICFRS thematic reviews of VAWG) points somewhere less comfortable: victim-blaming is not a glitch in police culture, it is a load-bearing part of how the institution manages its own workload, risk, and legitimacy.

Michael Lipsky’s concept of street level bureaucracy explains why. Lipsky argued that frontline workers operating under chronic resource pressure, teachers, social workers, police officers, inevitably develop informal coping mechanisms, rationing, categorising, discretion, that become the de facto policy on the ground, regardless of what official guidance says. Victim-blaming, seen this way, isn’t a departure from how the institution is meant to work, it’s an adaptation that lets an overstretched system keep functioning.

A few mechanisms do a lot of that work:

• Credibility as a resource-rationing device. Forces are stretched, and every report has to be triaged. Doubting a victim’s account, quietly or otherwise, is one of the fastest ways to close a file with no further action, without appearing to have failed to act. Blame becomes a form of gatekeeping dressed up as scepticism.
• Risk assessment tools that reward passivity. Frameworks such as DASH ask about fear, compliance, and vulnerability in ways that favour women who present as frightened and cooperative, and the tool itself is overdue for review and revision to remain fit for purpose. A woman who is angry, who has fought back, or who continues contact with her abuser for practical reasons often scores as lower risk, not because she is safer, but because she doesn’t perform the expected script. This becomes especially damaging where family courts pressure a woman into maintaining contact with her abuser under threat of losing custody of her children. That pressure is itself a form of abuse.
• Institutional self-protection, increasingly under its own surveillance. Where an officer or force has already missed warning signs, disputing a victim’s later account can retroactively justify earlier inaction. This has sharpened rather than softened as officers themselves become more visible, through body-worn video, complaints processes, performance metrics, and public scrutiny. The result can be a defensible record built for the institution’s protection, rather than a genuine account built for the victim’s.
• Occupational culture and case-hardening. Robert Reiner’s account of police “canteen culture”, marked by suspicion, cynicism, and in-group solidarity, helps explain why officers who spend years in public protection roles often become desensitised to repeat callers and complex cases, especially where the same address generates multiple reports. That desensitisation gets read back onto the victim as a character judgement, “she keeps going back”, rather than understood as an entirely predictable pattern of coercive control.

Seen this way, victim-blaming isn’t a failure of individual judgement so much as a rational adaptation to an under-resourced system that needs to produce the appearance of manageable caseloads. It is a construct, and each of the mechanisms above benefits the institution: a closed file, a lower risk score, a defensible record, a manageable caseload. None of them is free. The cost is paid by the women whose accounts are quietly discounted to produce that appearance of order. That doesn’t excuse the practice. It does explain why training alone rarely shifts it.

Secondary victimisation and institutional betrayal

This is where the concept of secondary victimisation earns its place in the profile. First named by Symonds in 1980, secondary victimisation is the harm caused not by the original perpetrator but by the response of institutions afterwards: being disbelieved, having a report downgraded, being asked to repeat a disclosure multiple times to different officers, having a case closed with no explanation. For many women, this stage causes as much psychological harm as the original assault, and it frequently deters them from ever reporting again. It is, in effect, the practical outcome of not being recognised as an ideal victim in the first place: the woman whose account doesn’t fit the expected template is the one most likely to be doubted, and doubt is where secondary victimisation begins.

Where secondary victimisation is often treated as an isolated bad experience, a poor interview, an unsympathetic officer, Jennifer Freyd’s concept of institutional betrayal describes something more sustained: harm caused by an institution’s failure to protect someone who depended on it and trusted it, compounding the original trauma rather than simply failing to address it. This is the better framework for what happens over time. A woman labelled as “difficult” or “non-compliant” in one service carries that label into the next. A reputation follows her between police, social work, and family court, so that each new professional she meets has already read a case file shaped by the last one’s disbelief. At this stage, she is no longer just failing to be recognised as a victim, she is being actively reconstructed by the system as a source of the problem, and the institutions she turned to for protection become a second source of harm in their own right.

Questions worth asking

Bringing these strands together, a fuller set of victimology questions might include:

About the person:

• What is their relationship to the person who harmed them, including any close chosen bonds that fall outside legal or biological categories?
• What structural factors (economic dependency, immigration status, disability) limited their options?
• What was their history of prior contact with services, and how did those contacts go?

About the system around them:

• Who is treated as a legitimate victim, and by what institutional logic?
• How do race, class, and immigration status intersect to shape whether someone is believed?
• What role do police, courts, and media play in constructing or denying victim status?
• What role did public or online reaction play in shaping how this case was treated by formal institutions?
• Where in the process did secondary victimisation occur, and what institutional pressure produced it?
• Has repeated contact with services produced institutional betrayal, a label or reputation that now shapes how new professionals treat her?
• Whose account is missing from the data, who doesn’t report, and why not?

Food for thought

Christie was writing in 1986. Symonds in 1980. Lipsky’s account of street level bureaucracy is from the same era. The theoretical foundations this field still leans on were built for a world without smartphones, without algorithmic feeds, without a case going viral before an officer has even opened a file. That doesn’t make them wrong, the mechanisms they describe are still operating, but it does mean victimology has some catching up to do.

The harder question isn’t whether any of this happens; the evidence is clear that it does, repeatedly, and to women who are already carrying the most. The harder question is what institutional response would look like if it didn’t require a woman to perform credibility at every single stage, to police, to a risk assessment tool, to a courtroom, to an anonymous crowd online, each with its own version of the ideal victim she’s expected to resemble. Nobody has fully answered that yet. It’s worth continuing to ask.