State Crime: Definition, Types, Offenders and Victims

Last Updated on September 26, 2026 by Karl Thompson

State crime is crime committed by, or on behalf of, governments in pursuit of state policy. It is the most harmful category of crime there is, State crime is crime committed by, or on behalf of, governments in pursuit of state policy. It is the most harmful category of crime there is, and also the hardest to prosecute — because the state is the body that makes the law in the first place. This post covers:

1. What state crime is, and why it is difficult to define
2. Why state crime matters more than any other kind
3. McLaughlin’s four types of state crime
4. The typical offenders of state crime
5. The typical victims, and the four types of victim
6. Whether state crime is criminal or deviant
7. Genocide, torture, corruption and war crimes, with examples
8. Institutional racism as state crime, and the Casey Review
9. Why ordinary people take part in state crimes
10. Evaluations and contemporary evidence

State crime is taught towards the end of the crime and deviance module, alongside globalisation and crime and green crime. It comes up on Paper 3 (7192/3).

What is State Crime?

Penny Green and Tony Ward (2004) define state crime as:

“State organisational deviance involving the violation of human rights.”

Many A-level textbooks give their definition in a looser form — “illegal or deviant activities perpetrated by the state, or with the complicity of state agencies” — and often date it 2005. Either wording will be understood by an examiner, but the book is State Crime: Governments, Violence and Corruption (Pluto Press, 2004).

The exact wording matters more than it might look, because Green and Ward chose it deliberately. Notice that their definition says deviance rather than crime, and human rights rather than law.

This is because the state makes the law. A government can therefore only be shown to be criminal on the rare occasions when it denounces itself for breaking its own rules — which is not something governments often do. Worse, a state can simply legalise what it wants to do. If a government passes a law permitting indefinite detention without trial, then indefinite detention without trial is no longer illegal in that country, and a definition of state crime based on national law has nothing left to say about it.

Defining state crime by the violation of human rights escapes that trap. Human rights standards are external to any single government, so they can be used to judge a state that has rewritten its own laws.

This is the single most important analytical point in the topic. In eight years of marking Paper 3 I saw very few students make it, and the ones who did were almost always in the top band. If you can explain why state crime has to be defined by human rights rather than by law, you are doing the thing examiners are looking for.

Why State Crime Matters

Scale. States control large territories and large populations, so when they turn to crime the number of victims is on a different order to anything an individual offender can manage. Green and Ward cite mid-range estimates of around 169 million people killed by governments between 1900 and 1987, excluding deaths in war. No amount of street crime approaches that.

Concealment. The state also controls the institutions that would ordinarily investigate, record and prosecute crime — the police, the courts, the statistical agencies. A state committing crimes is therefore very well placed to hide them, which is why state crime barely appears in official statistics.

The state is supposed to be the solution. We give governments a monopoly on legitimate force precisely so they can protect us. When the state is the offender, the institution that is supposed to provide protection is the source of the harm, and there is nowhere obvious to appeal.

Types of State Crime: McLaughlin’s Four Categories

Eugene McLaughlin (2001) identifies four categories of state crime:

• Crimes by security and police forces — genocide, torture, disappearances, imprisonment without trial
• Political crimes — corruption, censorship, electoral fraud
• Economic crimes — violations of health and safety laws, and of other regulations governing working conditions
• Social and cultural crimes — institutional racism

A useful way to revise this is to find a real example for each category. “The state does bad things” will get you nowhere; the category and the case will.

Who are the Typical Offenders of State Crime?

The typical offender is not a typical criminal, and that is precisely what makes this topic interesting.

The offenders are people in positions of authority: political leaders, senior civil servants, military commanders, police officers, intelligence officials. They are overwhelmingly older, better educated and wealthier than the people who fill the prisons for ordinary crime, which turns the usual sociological picture of the criminal on its head.

They also tend to act on behalf of an organisation rather than for personal gain, which is where Green and Ward’s phrase “state organisational deviance” earns its keep. A soldier committing an atrocity, or an official signing off a policy of detention without trial, is generally acting within a chain of command and in pursuit of what they understand to be state policy. This makes state crime closer to corporate crime than to street crime.

Most are never prosecuted. Because the offender is the state or an agent of it, prosecution generally requires either the regime to fall or an international body to intervene. Both are rare and both take years.

And most do not see themselves as criminals at all. This is the finding that makes the topic sociological rather than psychological. Studies of imprisoned atrocity offenders in Bosnia describe prison staff seeing them as ordinary people transformed into perpetrators by extraordinary circumstances, rather than as monsters. For much more on this, see my post on atrocity crimes in criminology, which goes into the research in depth.

Who are the Victims of State Crime?

State crime victims are the group most often left out of the topic altogether, which is odd given that they are its whole point. If you have been set a question on the typical victims of state crime, this section is the one you want.

The victims are usually the powerless, and usually members of groups the state has already marginalised — ethnic minorities, religious minorities, political opponents, indigenous peoples, the poor, and populations of countries subject to intervention by more powerful states.

Victims are also frequently defined out of the category before anything happens to them. Before a state harms a group on a large scale, it typically dehumanises them in official language and in the media. That process makes the harm easier to carry out and easier for the population to accept. It is a straightforward application of labelling theory — except that here it is the state doing the labelling.

They often appear in no statistics at all. They may not be recorded, they may not survive to report, and the agency that would record them is the offender.

And they are frequently in other countries. Where a wealthy state commits crimes abroad — through war, intervention or the actions of its corporations — the victims have no vote and no access to the courts of the country responsible.

Types of victims of state crime

It helps to separate them out, because a question asking about types of victim wants more than a list of groups.

Direct victims are those harmed by the act itself — the people killed, tortured, imprisoned without trial, or displaced. These are the victims of genocide, of torture, of detention without charge.

Indirect victims are harmed by the consequences. Families of the disappeared, communities whose land or water is poisoned by a state-permitted industry, populations left without healthcare or education because public money was diverted. The harm is real and the causal chain is longer, which is exactly why it is easier for a state to deny.

Secondary victims are the citizens of the offending state itself. They pay for it, their institutions are corrupted by it, and they are lied to about it. They rarely think of themselves as victims at all.

Denied victims are the category that matters most analytically: people harmed by the state who are never granted the victim label in the first place. A state that does not recognise a harm has done has no victims to count, and this is where state crime and critical victimology meet. Critical victimologists make exactly this argument about the state’s power to withhold the label — I have covered it in more depth in my post on victimology.

HOWEVER, it is worth noting that state crime also has secondary victims closer to home: the citizens of the offending state, who pay for it, whose institutions are corrupted by it, and who are lied to about it.

Is State Crime Criminal or Deviant?

This is a question students are often set, and the short answer is: usually deviant, sometimes criminal, and the gap between the two is the point.

The case for deviant. Much state crime breaks no national law, because the state writes the laws. Green and Ward define it as deviance for exactly this reason. Under a definition based on national law, a great deal of the most harmful state activity is simply not criminal.

The case for criminal. Since 1945 a body of international law has developed which does make some state actions criminal regardless of what national law says — the Nuremberg trials, the UN Genocide Convention (1948), the Universal Declaration of Human Rights (1948), and the International Criminal Court, established in 2002, which prosecutes genocide, crimes against humanity, war crimes and the crime of aggression.

The problem with the criminal answer. International courts depend on states cooperating with them, and the states most likely to commit state crimes are the least likely to cooperate. Several of the world’s most powerful countries have not ratified the ICC’s founding statute. So the law exists, but enforcement is selective — which is a criticism Marxists make of the criminal justice system generally, operating here at global scale.

A good answer says both. State crime is deviant by definition and criminal in principle, but only occasionally criminal in practice — and whether it is prosecuted depends far more on the power of the state involved than on the severity of what was done.

Genocide

The UN Convention on the Prevention and Punishment of the Crime of Genocide (1948), Article II, defines genocide as acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group. The acts listed are:

• Killing members of the group
• Causing serious bodily or mental harm to members of the group
• Deliberately inflicting conditions of life calculated to bring about the group’s physical destruction
• Imposing measures intended to prevent births within the group
• Forcibly transferring children of the group to another group

Note that killing is only one of the five. A state can commit genocide without a single execution — which surprises most students, and is worth remembering precisely because it surprises them.

Cambodia, 1975–79

Under the Khmer Rouge, most scholarly estimates place the death toll between 1.5 and 3 million people — roughly a fifth to a quarter of Cambodia’s population.

A precise figure is impossible, and the reason is itself a methodological point worth making: the regime destroyed the institutions capable of documenting what it did. The scale of a state crime is hardest to measure precisely where the state crime is worst — which is a good line to have ready for a methods question about official statistics.

Rwanda, 1994

In around a hundred days, an estimated 800,000 people were killed, the great majority of them Tutsi. This is usually given as approximately 70% of the Tutsi population and around 20% of Rwanda’s total population.

The Rwandan genocide illustrates the dehumanisation point above with unusual clarity: state-controlled radio referred to Tutsis as inyenzi, cockroaches, in the months beforehand.

Torture and Detention Without Trial

Torture is prohibited absolutely under international law — there are no permitted exceptions, even in war — and it remains widespread.

Guantanamo Bay is the example most often used because it involves a wealthy democratic state rather than a distant regime. Around 780 people have been held there since 2002, most without charge or trial. Fifteen remained as of January 2026, some of them cleared for release years earlier and still detained. The camp entered its twenty-fifth year in January 2026.

This is a strong evaluation point against the assumption that state crime is something that happens elsewhere.

Corruption

Political corruption takes several forms: diverting public money into private accounts, awarding government contracts in return for bribes, and electoral fraud.

Transparency International’s Corruption Perceptions Index scores countries from 0 (highly corrupt) to 100 (very clean). In the 2025 index, published in February 2026, the global average was 42 — the lowest in over a decade.

There is a broad association between low scores, poverty and armed conflict, and many of the worst state crimes are committed during civil wars. Paul Collier argues that the problem begins with how political office is understood: in some countries it is treated not as public service but as a means of extracting money for oneself and one’s backers.

HOWEVER, corruption is not confined to poor countries. Russia scored 22 out of 100 in 2025, ranking 157th of 182 countries, and China scored 43, barely above the global average, despite being the world’s second largest economy. Economic development does not by itself produce clean government, which suggests that political institutions and accountability matter more than national income.

War Crimes and State Collapse

Treating a population badly enough breeds the conditions for civil war, and civil war brings its own category of state crime with it.

Child soldiers. The recruitment and use of children in armed forces is a war crime under the Rome Statute, and is a recurring feature of conflicts in which state authority has broken down.

Rape as a weapon of war. Sexual violence used systematically against a civilian population is a crime against humanity, not an incidental consequence of conflict. The distinction matters: systematic use implies planning and command responsibility, which is what makes it prosecutable.

In extreme cases the state loses control of its territory altogether and armed groups take over the functions of government — as has happened at different points in Afghanistan, Somalia, Syria and Iraq. This produces an awkward question for the definition of state crime: if a group controls territory and a population but is not a recognised government, are its crimes state crimes? Green and Ward’s definition, which turns on organisational deviance and human rights rather than on formal statehood, handles this better than a definition based on national law could.

Institutional Racism as State Crime

Institutional racism is racism expressed through the practices of institutions, as distinct from racism by individuals — McLaughlin’s fourth category.

Following the Macpherson Inquiry (1999) into the Metropolitan Police’s failure to investigate the murder of Stephen Lawrence, the Met was found to be institutionally racist.

The Casey Review (March 2023) found the problem unresolved. Baroness Louise Casey’s independent review — commissioned after the murder of Sarah Everard by a serving officer — found the Met to be institutionally racist, misogynistic and homophobic, both in how officers and staff are treated internally and in how communities are policed. It concluded that the force was failing women and children and was unable to police itself.

A useful evaluation point: Commissioner Sir Mark Rowley accepted the substance of Casey’s findings but publicly resisted the word “institutional”. That refusal is sociologically interesting in its own right — it is a state agency contesting the label rather than the evidence, which is exactly what labelling theorists would predict a powerful institution to do.

Why do Ordinary People Take Part in State Crimes?

This is the question that turns a descriptive answer into an analytical one, and criminologists working on atrocity crimes have developed a useful way of approaching it: look at three levels at once.

Macro — the level of the state and society. State crimes happen in particular structural conditions: economic crisis, war or civil conflict, an authoritarian government, and an official ideology that identifies a group as a threat. These conditions are not created by the individuals who carry the crimes out.

Meso — the level of the organisation. Most perpetrators are acting within an institution — an army unit, a police force, a bureaucracy — with a chain of command, strong internal loyalty and clear pressure to conform. Responsibility gets distributed across the organisation until nobody feels they hold much of it. This is why organisational structure matters more than individual pathology.

Micro — the level of the individual. Individual perpetrators are usually not distinguishable from anyone else beforehand. What changes is the situation, and the process by which victims are dehumanised until harming them feels permissible.

The conclusion drawn from this research is uncomfortable and worth stating plainly: state crimes are generally committed by ordinary people placed in extraordinary circumstances, not by people who were criminal to begin with.

If you want the full academic treatment of this, including the research on imprisoned perpetrators and the debate about punishment, I have written it up separately in atrocity crimes in criminology. That post is written at degree level.

State Crime and Human Rights

Herman and Julia Schwendinger (1970), in ‘Defenders of Order or Guardians of Human Rights?’, argued that criminologists should define crime in terms of the violation of basic human rights rather than in terms of the law. On this view, any state which denies its people those rights is committing a crime, whatever its own statute book says.

This is a transgressive position — it crosses the boundary of traditional criminology by studying harm rather than crime — and it is the same move green criminologists make when they study environmental harm that no law prohibits.

Contemporary Evidence on State Crime (2026 update)

Below I set out three current cases, each of which can be used as evidence in an exam answer.

The Casey Review and institutional racism

See the section above for the detail. The point to carry into an answer is that this is a 2023 finding about a British institution, not a historical one about somewhere else — which makes it much harder for anyone to argue that state crime is something that happens elsewhere.

Guantanamo Bay is still open

Around 780 people have been held at Guantanamo since 2002, most without charge or trial. Fifteen remained as of January 2026, some cleared for release years earlier and still there. The camp entered its twenty-fifth year in January 2026.

Twenty-five years is longer than most of the students reading this have been alive, which is worth saying out loud when the topic comes up.

The 2025 Corruption Perceptions Index

Transparency International’s 2025 index, published in February 2026, put the global average at 42 — the lowest in over a decade. Russia scored 22 and ranked 157th of 182; China scored 43. Neither is a poor country, which is the finding that matters for evaluation.

Evaluating the Study of State Crime

In favour

The concept draws attention to the most harmful crimes there are, which mainstream criminology had largely ignored. It exposes the fact that the law serves the powerful, since states define their own harmful acts as legal — a point that supports the Marxist analysis of crime. And the Casey Review shows the concept has purchase on a British institution, not only on distant regimes.

Criticisms

It is usual in A-level sociology to criticise one position using others.

Functionalists would argue that laws express the collective morality of a society, and that state agents policing and punishing crime reinforce the boundaries of acceptable behaviour. They would add that citizens generally come to less harm with a stable state than without one — nearly all wealthy nations have large public sectors providing health, education and infrastructure, and the alternative to a functioning state is usually not freedom but a failed state.

Defining crime by human rights rather than law means relying on contested moral standards. Whose human rights, defined by whom? Different societies genuinely disagree about which rights are basic, and a definition that a state can reject is a weak basis for prosecution.

Interactionists would point out that “state crime” is itself a label, applied by some groups to others and applied very selectively. Western states are quick to identify state crime in countries they oppose and slow to see it in allies — little is said about Saudi Arabia’s treatment of women, or restrictions on freedom of speech in China.

It is extremely difficult to research. States control the evidence, and the worst state crimes destroy the records that would measure them. Most of what sociologists know comes from post-hoc inquiries, defectors and journalism, all of which have obvious limitations.

Crime and Deviance Revision Bundle

If you like this sort of thing, then you might like my crime and deviance revision bundle — [LINK NEEDED: Sellfy URL for the crime and deviance bundle] — which contains the following:

[IMAGE: retain existing Crime-Deviance-A-Level-Revision.png cover, linked to the Sellfy product]

1. 32 pages of revision notes covering the entire A-level sociology crime and deviance specification
2. Seven colour mind maps covering sociological perspectives on crime and deviance
3. 12 exam practice questions including short answer, 10-mark and essay exemplars

Written specifically for the AQA A-level sociology specification.

Sources used to write this post

• Green, P. and Ward, T. (2004) State Crime: Governments, Violence and Corruption. London: Pluto Press
• McLaughlin, E. (2001) ‘Political violence, terrorism and states of fear’, in Muncie, J. and McLaughlin, E. (eds) The Problem of Crime, 2nd edn. London: SAGE
• Schwendinger, H. and Schwendinger, J. (1970) ‘Defenders of Order or Guardians of Human Rights?’
• UN Convention on the Prevention and Punishment of the Crime of Genocide (1948), Article II
• The Baroness Casey Review: final report (March 2023)
• Transparency International Corruption Perceptions Index 2025
• The Macpherson Inquiry (1999) into the matters arising from the death of Stephen Lawrence

Signposting and Related Posts

This post has been written primarily for students studying the crime and deviance topic as part of the AQA’s A-Level Sociology course, where it appears on Paper 3 (7192/3).

For worked cases you should read my post on examples of state crimes, which covers recent and current cases in detail and is the natural next step after this post.

State crime is usually taught alongside globalisation and crime, which explains how criminal activity crosses borders, and green crime, which overlaps with state crime wherever a government permits or conceals environmental damage.

Students wanting to take this further should read atrocity crimes in criminology, which covers the academic research on genocide, war crimes and crimes against humanity at degree level.

The victims of state crime are treated more fully here than on the victimology post, which covers positivist and critical victimology, the hierarchy of victimisation and Tombs and Whyte on workplace deaths. The two topics share an argument — the state’s power to decide who counts as a victim — so read them together.

The theoretical background is in the Marxist perspective on crime and labelling theory, both of which are applied above.

Please click here to return to the homepage – ReviseSociology.com