Nothing to Hide:

The case against the ban on extreme pornography

By Nick Cowen




  • The ban on possession of ‘extreme pornography’ was introduced in 2009 and extended to 2015. The law, as drafted, bans depictions of some sex acts that can be conducted safely and consensually between adults, with a specific risk of pros- ecution posed to LGBT minorities.
  • The Crown Prosecution Service reports more than a thousand offences prosecuted each year, implying significant enforcement costs that could be deployed effectively elsewhere.
  • A significant minority of the British population enjoy sexually aggressive fantasy scenarios but do not pose a specific risk of committing violent or sexual offences.
  • Access to pornography has increased dramatically in recent years, yet social harms imputed to pornography (especially violence against women) have reduced moderately but significantly.
  • While some survey evidence claims a correlation between individual use of pornography and sexual aggression, econometric evidence suggests this is not a causal relationship and that, if anything, increased access to pornography can reduce measurable social harms.
  • The ban itself represents a potential risk to political integrity. Like the ban on homosexuality in much of the 20th century, prohibitions on private sexual conduct can be used to silence, blackmail and corrupt individuals in positions of authority and responsibility.
  • There are better policies for reducing violence against women in the dimensions of criminal justice, education and economic reform.
  • The prevailing free speech doctrine in the United States shows that it is realistically possible to tackle damaging forms of expression while also having strong protections for sexual expression both of real consenting activity and fictional performances.


FOREWORD by Myles Jackman, Solicitor

Four words: canary in the coalmine. I have a confession. I appropriated the phrase “Pornography is the Canary in the Coalmine of Free Speech” from Nick Cowen. Fortunately he was gracious enough to deny it. Yet I cannot deny that Nick’s intuitive feel for the liberal civil-liberties cause has fundamentally informed my presentation of the sexual freedom of expression case. As the mechanisms used against pornography are often tests deployed to evaluate the relative ease with which more widespread state censorship can be delivered.

As members of the sexual liberties campaign Backlash, Nick and I have often argued that not only is the law a blunt tool for enforcing presumed moral values; the state and our policy-makers are entirely divorced from current social values and the technological change driving them. Simon Walsh’s acquittal for possessing supposedly extreme pornography illustrates the criminal justice system’s inherent in- comprehension of LGBTQ lifestyle activities. The sympathy displayed by the vast majority of heterosexual society to Walsh’s Newsnight appearance was sufficient for the Crown Prosecution Service to revise their Guidelines.
Whilst this was clearly a victory in itself, I agree with Nick that the revised Guide- lines are little more than a cosmetic change and that the entire area of obscenity law needs to be redrafted, starting with the extreme pornography law.

One of the most disheartening examples of this problem is presented by the popular mobile messaging service WhatsApp that allows multiple-member group messaging. I have represented a number of defendants who have received unsolicited images as part of a WhatsApp group. They had their phones seized by Police for unrelated offences, which were dropped later; then charged with possessing extreme pornographic images as a kind of “consolation prize” offence.

It is extremely hard to disprove criminal intent in possession cases without expert forensic IT examination and accurate legal advice. Last year two defendants were so manifestly failed by the legal system that they ended up representing themselves at their own sentencing hearing, rather than continuing to pay for (inaccurate) legal advice. The sentencing Judge even acknowledged that their culpability was at vanishing point, but failed to recognise that this was a miscarriage of justice in itself.
Whilst they received a conditional discharge (the Court’s equivalent of a Police Caution) this will be recorded against them and may prohibit their ability to travel to the US under American “moral turpitude” provisions. Thus the receipt of un- wanted and unrequested images led to two professionals having criminal records; being shamed publicly in the press; and a continued impact upon their futures.

Nick’s work is exceptionally valuable in making the argument for law reform in this confused and confusing area, which is in desperate need of review.


"This passion to discover the real sentiments of others is naturally so strong, that it often degenerates into a troublesome and impertinent curiosity to pry into those secrets of our neighbours which they have very justifiable reasons for concealing"
Adam Smith, The Theory of Moral Sentiments

In August 2012, Simon Walsh, a prominent barrister, City of London alderman, and former aide to mayor Boris Johnson, was prosecuted for possession of ‘extreme pornography’. His alleged crime was possession of photographs depicting ‘fisting’ and ‘urethral sounding’ taken at a private all-male sex party where Walsh was a participant. The prosecution claimed that the acts depicted were extreme because they could cause serious harm. The jury heard expert evidence from a surgeon that the acts, which are relatively commonly practiced within the LGBT community, could be conducted safely. It took the jury just a few minutes of deliberation to reject all charges.

Despite the ‘not guilty’ verdict, the trial came at great personal cost to Walsh. Intimate details of his sex life were exposed to judgement in a very public forum. Moreover, the Crown Prosecution Service has continued to argue that the grounds for prosecution were sound and that the images were ‘extreme’, leaving depictions of these practices open to further prosecutions. This suggests a particular legal vulnerability for gay men and other sexual minorities. This is a perverse result for a law that was originally intended to address violence against women.

In this report, I present a case for reform of this prohibition, and also a general defence of a liberal policy approach to sexually explicit expression. It is based on research conducted at Worcester College, University of Oxford and the Department of Political Economy, King’s College London. Part of this research is published in the American Journal of Political Science.[1]



A Labour government banned the possession of what it termed ‘extreme pornogra- phy’ under Section 63 of the Criminal Justice and Immigration Act 2008 (but only initiated in 2009). The Conservative-led Coalition government expanded the law under Section 37 of the Criminal Justice and Courts Act 2015. The result is a law that bans possession of still or moving images which meet the following criteria.[2] They must be:

  • Intended for sexual arousal
  • Realistic
  • Grossly Obscene

In addition, they must depict at least one of the following:

  • Acts of non-consensual penetration
  • An act which threatens a person’s life
  • An act likely to cause serious harm to the breasts, genitals or anus • Bestiality
  • Necrophilia



Table 1 displays the best data available on how often the extreme porn ban is used. The Crown Prosecution Service reports more than a thousand prosecutions each year for extreme pornography.[3] The Ministry of Justice does not give a specific category for the offence. It is listed under ‘other possession of obscene materials’. The MOJ suggests there were a little fewer than 200 cases each year from 2011 to 2014, including cautions and court prosecutions.[4] The reason for the discrepancy with the CPS figures is probably because the MOJ only reports the ‘index’ offence, the most serious offence when multiple charges are laid against an individual. This mean that in the majority of cases, possession of extreme pornography is an additional allegation made alongside a more serious offence.

Table 1: Crown Prosecution Service and Ministry of Justice data on prosecutions

Table 1: Crown Prosecution Service and Ministry of Justice data on prosecutions

Using either measures, the law is being used much more frequently than the Government anticipated. We know comparatively little about the circumstances of most cases because those prosecuted are usually keen to plead guilty and accept a sanction with as little public exposure as possible. Prosecution statistics indicate that many cases involve depictions of bestiality. While bestiality raises legitimate concerns with animal cruelty, many images may amount to harmless (if poor taste) jokes. For example, one failed prosecution in Wales involved possession of an image of a man having sex while wearing a tiger costume.[5]

As with a great deal of public expenditure, it is hard to make accurate estimates of the unit cost of pursuing each offence. The Ministry of Justice estimates that the cost of bringing about each extreme pornography case is around £10,000, not including police costs.[6] If each prosecution charge were a case on its own, that would imply costs of more than £15 million in 2014 alone. In practice, that is probably an overestimate as a large number of those offences are likely to involve several offences of a similar kind. If we take the approximate 200 figure offered by the Ministry of Justice as more likely indicative of the number of unique cases (the actual correspondence is still far from perfect but the scale is at least plausible), then criminal justice system costs are around £2 million per annum.

These costs are relatively trivial compared to total expenditure in the criminal justice system. However, it is worth considering the opportunity cost of pursuing individuals for pornography possession with resources and expertise that could otherwise be used to support victims of actual violence, and pursue people who have committed contact sexual offences such as rape and sexual assault. This is even more pertinent when we consider the range of people potentially targeted by extreme pornography legislation.



Systematic surveys of the prevalence pornography consumption in the United Kingdom is relatively scarce. One of the more credible is the somewhat dated but helpful British Sexual Fantasy Research Project 2007. Based on a representative sample survey of 19,000 adults in the United Kingdom, it found that: 86% of men and 56% of women had viewed pornography.[7] 29% fantasise about playing a dominant or “aggressive” role during sex; 33% fantasise about playing a submissive or “passive” role during sex; 4% fantasise about being “violent” towards someone else; 6% fantasise about violence being vested on themselves by another person.[8]

This means that around 2.2 million men and women have violent sexual fantasies of some kind, and nearly a third of all British adults fantasise about sexual domination and submission. These statistics indicate that the number of men and women interested in fantasy pornographic depictions of non-consensual sexual encounters is likely to be quite high. A central, perhaps conservative, estimate might be around 930,000 men and 640,000 women. There is no evidential link to suggest that any of these individuals pose a specific risk of committing sexual offences.

Crucially, fantasy scenarios involving coercion are shared by both men and women. Both men and women fantasise about aggressive sex in both dominant and submissive roles. The argument in favour of criminalising extreme pornography has been characterised as a means of protecting women and supporting women’s interests and standing in society. Unfortunately, these claims discount the potential impact of criminalisation on female viewers of pornography.

It can be psychologically and personally destructive for an individual of any gender to have their private consensual fantasies exposed for public scrutiny. As a result, one would hope that any law that is bound to bring about some of these consequences would have strong evidence that it would effectively address social harms. I shall now argue that evidence is lacking.



Around the world, censorship proponents associate pornography with all manner of social harms, including higher divorce rates, substance addiction and even devil worship. In the United Kingdom, public justification for restrictions tend to focus on presumed effects on violence against women, as well as women’s status in society. While scientific evidence is divided on the impact of pornography, I argue that the weight of evidence suggests that allowing consenting adults to access pornography either has no causal effect in terms of sexual violence and other social harms, or has a small positive effect (i.e. reducing observed violence against women).

In trying to justify the notion that pornography causes violence, censorship proponents face a steep evidential hurdle. Access to pornography is temporally reverse correlated (rather than correlated) with measures of violence against women. Most recently, the staggering rise in internet access has happened alongside a marked reduction in women’s reported experience of violent crime.
To illustrate the trend in the United Kingdom, Figure 1 shows the increase in the proportion of households with internet access alongside some key measures of women’s victimisation. This is the percentage of women who report experiencing partner abuse (non-sexual), any sexual assault (including attempts) and stalking (including in later years, harassment online) in a given year. Data sources are the Crime Survey for England and Wales and the Office for National Statistics.[9,10] Note, for the sake of space, household internet access is not plotted against the same vertical scale axis (data points are labelled). This crime survey data is collected independently of police reports, and is one of the most systematic attempts to generate a representative picture of the population’s experience of crime and social disorder. While true representativeness is impossible, this formal survey approach comes closest to approaching it.

Figure 1: Household internet access (%) (right axis) compared to violence against women measures (left axis)

Figure 1: Household internet access (%) (right axis) compared to violence against women measures (left axis)

This means that women had a one in twenty estimated annual risk of suffering at least one form of sexual assault in 2006 and just over a one in fifty chance in 2014. In noting this reduction, I am not suggesting that violence against women has be- come a marginal social problem. It is a severe social problem and perhaps the most urgent problem in criminal justice policy. It is due to its importance that it is critical that we understand what the real causal contributions are to gender-based violence.

This apparent inverse relationship between access to pornography and sexual violence is a common finding across the world, especially in the United States.[11] Milton Diamond, emeritus professor at the University of Hawaii, comments in one systematic survey of the academic literature on pornography: ‘objections to erotic materials are often made on the basis of supposed actual, social or moral harm to women. No such cause and effect has been demonstrated with any negative con- sequence. It is relevant to mention here that a temporal correlation between pornography and any effect is a necessary condition before one can rationally entertain the idea that there is a positive statistical correlation between pornography and any negative effect.’[12] He proposes, ‘if anything, there is an inverse causal relationship between an increase in pornography and sex crimes’.[13]

In response, evidence supporting censorship of pornography tends to use different research designs. These studies produce interesting results from a social science perspective but I suggest they are less relevant for evaluating the likely impact of public policies that restrict access to pornography. Professors Catherine Itzin, Anne Taket and Liz Kelly conducted a study for the Home Office designed to offer the evidence-based case for prohibiting extreme pornography.[14] It combined five 8 meta-analyses of studies researching aggression after exposure to pornography in a laboratory; experimental and non-experimental studies of the relationship between exposure to pornography and the acceptance of rape myths; studies analysing the reactions of criminal sex offenders to pornography; one of published research on the general effects of pornography and one on the relationship between consuming pornography and attitudes towards violence. In addition, the authors included 32 other primary studies that were not included in previous meta-analyses. The sheer size of the research considered here was a major selling point for the research. For example, a total of 3786 male students were trumpeted to have been examined over the course of the meta-analyses and included studies.

Given this represented the rationale for criminalisation, it is remarkable how mild the summary conclusions are. They claim there were ‘some harmful effects from extreme pornography on some who access it’ and that ‘Men who are predisposed to aggression, or have a history of sexual and other aggression were more susceptible to the influence of extreme pornographic material.’[15] Attwood and Smith criticise the assessment for failing to justify the assumptions that ‘aggressive behaviour’ and ‘rape myth acceptance’ can be measured objectively and that laboratory studies can relate to real world social effects.[16] With respect to the experimental research that features in the meta-analysis, Kendall notes several likely difficulties: ‘attitudes towards rape – or at best, physiological arousal – can be measured, not rapes... Moreover, in actual market consumption, pornography is disproportionately consumed by people who are specifically seeking sexual release, not randomly assigned to typically unaroused people as in a laboratory.’[17]

In other words, what experiments in psychology miss out by design is the fact that people who access pornography in society, rather than a laboratory, are autonomous actors with their own motivations. They are likely to attach different meanings to the media they choose to access than to similar media that they are randomly assigned to watch by an experimenter. Unless we assume a simplistic mechanical model of human psychology, how people interpret and reflect on particular media is going to be crucial for how they subsequently behave. Relying on experimental studies of this kind is rather like preferring to test the effects of smoking by getting students to smoke in a laboratory and then fill out a form on whether they ‘feel’ like they are more likely to get lung disease, while ignoring evidence of how many people have actually contracted lung disease and how this relates to smoking in the real world.



In recognition of some these concerns, a more careful recent meta-analysis excluded experimental studies and included only analyses based on surveys of representative populations (often high school or university students).18 These results offer much more compelling evidence at least of a correlation between individual use of pornography and sexual aggression. The weakness is that they have no direct evidence of causation. We cannot know from these studies whether people with a propensity to access pornography are likely to become sexually aggressive or whether sexually aggressive people also choose to access pornography.

These correlations ought to be reconcilable with the aggregate social observations discussed above, but this has not been attempted by proponents of censorship. When it came to examining correlational studies, the Home Office study’s criteria for inclusion stated: ‘If measures are carried out only at the level of a group and the data cannot be related to individuals, then the study is to be excluded from further consideration’.[19] The justification for this criteria were as follows: ‘These studies were excluded in view of the inability of such designs to allow any detailed exploration of hypotheses in relation to causality, and the danger of ecological fallacy – the bias that may occur because an association observed between variables on an aggregate level does not necessarily represent an association that exists at an individual level.’[20] These are reasonable methodological concerns but considering the studies admitted to the assessment also have methodological weaknesses, this represents somewhat of a double standard.

The criteria had the effect of excluding all studies that attempted to research the effects of the availability of pornography on whole populations through the use of information like national crime data. This is relevant because, of course, criminal prohibitions and other forms of regulation are applied to whole populations, not a subset. It was partly considerations of survey data of this sort that persuaded the Williams Committee in 1979, which produced a previous report for the Home Office, to take a comparatively sceptical view of censorship. A key finding in that instance was that, like today, there was no obvious relationship between sex crime and the increase in the availability of pornography in the United Kingdom (in fact sex crime as a proportion of serious crime was declining at the time of this increase in availability).[21] The decision to ignore aggregate analyses in the Home Office assessment led to the explicit exclusion of two well-known studies by Berl Kutchinsky.[22] One of these looked at four European states where ‘pictorial pornography including violent/dominant varieties (in the form of picture magazines, and films/videos used at home or shown in arcades or cinemas) has developed from extreme 10 scarcity to relative abundance.’[23]

The results: ‘in none of the countries did rape increase more than nonsexual violent crimes. This finding in itself would seem sufficient to discard the hypothesis that pornography causes rape. In fact, in three countries, Denmark, Sweden and West Germany, rape increased less than nonsexual assault, and in West Germany rape did not increase at all.’[24]
The Home Office assessment also excluded a study by Milton Diamond and Ayako Uchiyama that used a similar analysis of sex crime data in Japan to conclude: ‘It is certainly clear from our data and analysis that a massive increase in available pornography in Japan has been correlated with a dramatic decrease in sexual crimes.’[25] The outcome of this study is particularly interesting when considering extreme pornography, due to the popularity of extraordinarily violent themes in Japanese pornography.

There remain plenty of difficulties with drawing conclusions from aggregate data. One problem is extrapolating what the real level of sexual abuse is from mere police reports, which are often the only kind of data available for specific districts. Besides widespread under-reporting of sex crimes, they are also recorded differently from one state or district to another, and the recording definitions sometimes change over time. Another difficulty is to control for other factors that might render a statistical correlation either spurious or indirect, or hide a correlation from view. Changes in crime rates over time might reflect more general institutional and cultural changes, while correlations within geographical regions could reflect un- observable features of the population rather than actual links between pornography and violence.

These challenges, though not yet decisively overcome, have been tackled using econometric techniques designed to isolate causality. Winai Wongsurawat uses an instrumental variable (the availability of private mail boxes) to try to control for some of the unobserved background factors of cross-sectional data. He found that in areas where private mail box services are more available, individuals were more likely to buy pornography, probably because it is easier to avoid detection and, therefore, embarrassment. He used data from the 1990s before widespread use of the Internet may have rendered some of these issues far less relevant. He found that when this instrumental measure was applied to cross-sectional data comparing rates of rape victimisation and the use of pornography, the apparent positive correlation turned negative. In other words, when the increase in the use of pornography was associated with ease of access (rather than other background social factors), it 11 was associated with fewer sex crimes rather than more.

Wongsurawat concludes: ‘These findings potentially reconcile two contradicting lines of research on the social effects of pornography; one that links legalisation of adult material to a decline in crime in Europe and one that points to a positive correlation between crime and sales of sex magazines within the United State. My results strongly suggest that unobservable population characteristics severely bias upwards... estimated harmful effects of such magazines and hence wrongly attribute social ills arising from unobservable population features to sales of pornography.’[26]
It would be imprudent to claim that these analyses conclusively show that the greater availability of pornography plays a causal role in reducing sex crimes. It is possible that an unmeasurable intervening factor, for example more socially liberal institutions, leads simultaneously to greater use of and tolerance for pornography, and to more respect for women’s rights. However, what we can say with greater confidence is that these studies represent a body of evidence (discounted almost out of hand by the government) that the use of pornography does not make women more likely to be victims of sexual violence. More than anything else, if those who favour prohibition wish to maintain that pornography is associated with sexual violence, they need some account of why this connection seems to have so little effect on aggregate social outcomes, or even has the opposite effect to that expected.

Since the Internet is associated with a dramatic broadening of the types of pornography available (including extreme pornography as defined in British law), and this dramatic increase in their availability has not led to a general increase in sex crimes, we might infer that the correlation between the consumption of extreme pornography and violence against women is limited, non-existent or even negative (reducing crime).


As a result of this comparative lack of observable evidence, scholarly proponents of censorship do not rely very much on arguments that pornography causes direct specifiable harm. McGlynn, Rackley and Ward have proposed instead a concept of ‘cultural harm’.[27] On this account, the pornography ban itself cannot necessarily be expected to tackle an observable social problem, but it might help challenge the symbolic and discursive frameworks through which women are seen and categorised. By passing the ban, the state sends a message that it is unacceptable to treat women as sex objects.

Such a justification is difficult to evaluate empirically. However, it is possible to inquire whether this use of legislation as a form of signalling is likely to be effective on its own terms. Successfully sending ‘a message’ requires that an audience receives
it and understands it. If the intended audience is the public, then it is unlikely that passing a law is an effective means of sending a message regarding women’s status. Relatively few people are aware when a new criminal law has been passed. Those
that become aware can easily misunderstand its content or moral intentions. This is particularly likely when the law itself is sufficiently complex that legal professionals can easily misunderstand its intentions.

There are also perverse interpretations of the message. For example, it could signal that the state regards pornography as sufficiently dangerous that it can be blamed for real world violence, rather than the actual perpetrators. Some sex offenders argue that they are addicted to pornography in an attempt to mitigate their own responsibility for violent acts, perhaps to seek a lighter penalty. The scientific basis for the existence of ‘porn addiction’ is flimsy to say the least.[28] Nevertheless, if the government officially asserts that pornography is dangerous, then that strengthens those sorts of defences.

Alternatively, as seems to have happened in Walsh’s case, the received message could simply be that the state approves of punishing people with minority sexual interests. The result is that in pursuit of an abstract goal of challenging symbolic representations of women, other people must be singled out on the basis of their private sexual interests, almost as scapegoats, to be punished for social harms for which they are not plausibly individually responsible.




While the relatively minor costs of enforcement were discussed above, a deeper concern is the potential impact that prohibitions on private conduct can have on political institutions and wider public accountability. Some historical context can help here. A key rationale for decriminalising homosexuality in 1967 was not so much widespread moral acceptance of homosexuality. Instead, the prohibition had become widely understood as a ‘blackmailer’s charter’. The ban on homosexual acts had not caused people to stop engaging in such acts, but it had exposed many otherwise law-abiding citizens to being branded criminals. Extorting money, or favours, from homosexuals in return for not revealing their sexual orientation was commonplace.[29]

This had a particular impact on political accountability as it meant that private homosexuals with official public roles could, if discovered, be manipulated by opportunistic actors as well as foreign agencies. Persecuting sexual minorities as criminals was not just an outrageous evil, it was a self-inflicted wound on the body politic. In banning a range of material that represents relatively popular sexual fantasy material (including material enjoyed by LGBT individuals), governments risk reintroducing this sort of scenario and making blackmail over private sexuality or embarrassing personal conduct a common problem once again.

A couple of recent events can illustrate the surprising interaction between extreme porn accusations and wider public disputes. Conservative MP Andrew Mitchell, then Chief Whip, got into a now infamous altercation with some police officers stationed at Downing Street. During this now infamous ‘plebgate’ fallout, several police officers involved were arrested for possession of extreme pornography.[30] These allegations were irrelevant to their dispute with the Government but came to light at just the moment that they found themselves in conflict with a member of the Cabinet. In a similar case, several teaching staff who were under scrutiny during the so-called ‘Trojan Horse’ schools scandal in Birmingham were arrested on the unrelated issue of possessing extreme pornography.[31]

These events are subject to reasonable alternative interpretations. However, it is plausible to suggest that these allegations damaged the credibility of parties who were already involved in very public, acrimonious disputes. Accidental or incidental possession of extreme pornography by otherwise law-abiding citizens, including people in public roles, is quite common. As I have argued, such possession is usually not dangerous in itself. This means there is no particular reason for police to search actively for individuals committing a crime of this kind. However, if an individual comes under public scrutiny for other reasons, their possession of extreme pornography (in fact, even the allegation) can be used to silence them. Although it is difficult to quantify the extent or the risk, anti-pornography laws (alongside other offences designed to deter private vice, such as drug prohibition) can harm free political debate and accountability in public institutions. In essence, the law is an invitation to corruption.



I have argued that the case for censorship on the grounds of protecting women from violence is empirically unsound, and that the ‘symbolic’ case for censorship is an unconvincing way of the state signalling a commitment to improving women’s status. This raises the question what alternative ways there are for reducing violence against women. I think there are a number of approaches. We should not dismiss criminal justice responses entirely. There is a significant body of research showing that police detection and policing resources play a significant role in reducing crime.[32] Violence against women is no exception to this general find- 14 ing so convicting more perpetrators of sexual assault and other forms of abuse is one effective approach. Increased punishments, essentially more and longer prison sentences is more contested. On balance, increasing punishment seems to be crime reductive, but it can also have a criminogenic impact depending on how prisons are organised.[33, 34]

Criminalisation is a blunt tool, but a liberal approach to pornography does not rule out all kinds of criminalisation. Besides images of child abuse which have been criminalised for decades, there is a strong case for the criminalisation of so-called ‘revenge pornography’, the non-consensual sharing of intimate images, as has now happened under section 33 of the Criminal Justice and Courts Act.[35, 36] When used in a threatening or harassing way, these specific uses of pornography are a form of violence and so legitimately prohibited and deterred with the prospect of meaningful penalties. In addition, what we know about the impact of sex education on future experience of sexual violence is suggestive and positive. Access to sex education in schools is associated with reductions in harmful sexual activity, including, critically, instances of nonvolitional sexual encounters.[37]

Perhaps the most important dimension of all is likely to be economic. Sexual violence and related forms of abuse are disproportionately perpetrated against people in poverty. Recent research found an increase in violence against women around the time of the great recession, suggesting that both economic deprivation and negative shocks to income make women more likely to be victims of violent crime and men to be perpetrators.[38] In fact, it could be that the improvement in women’s absolute and relative economic status in recent years was a decisive factor explaining the modest reductions in violence against women that has been observed.

There are a great many relationships between deprivation and violence. I am going to suggest just one straightforward one here. The ability to control where and how one lives is expensive. Women who lack material resources of their own face desperate trade-offs when considering the prospect of leaving, or escaping, an abusive partner, associate or an otherwise insecure living environment.[39] For example, they may not be able to afford a physically secure dwelling for themselves and their children (if they have children). They face possibilities like moving to a different city at great immediate cost and risk in the hope of deterring further attacks from former partners, while at the same time losing their existing social and economic support. If they have nowhere secure to go, they may have no choice but to remain with an abusive partner. Having more personal economic resources does not eliminate these problems (no one is rendered immune to things like domestic violence or sexual assault by wealth) but it is an important tool for ameliorating and ending abusive situations when they arise.

Besides providing specific support and shelter for women in crisis, the role of economic factors suggests that more general economic reforms could be crucial for ending the widespread phenomenon of violence against women. Tackling Britain’s chronic shortage of housing, which has so many related social costs already, could be important for giving women in economically precarious situations real alternatives to staying in unsafe environments. Similarly, replacing our current precarious system of needs-assessed welfare with a more universal system of support, such as a basic income, could help women, and families, facing sudden shocks to their income avoid the sort of home and community situations where they can become a target of abuse.



In this report, I have argued that there is little to commend the ban on extreme pornography in the United Kingdom, and several reasons to be concerned about it in theory and practice. I do not see it as fulfilling, even badly, any pressing public policy goal. It could be abolished with no real negative impact, and a significant positive impact in terms of better securing the rights of sexual minorities. If abolition is politically impossible, then the law could be productively amended so that images created through consensual activities between adults are not subject to prosecution.

Some proponents of censorship suggest that an ‘absolutist’ approach to free ex- pression, especially sexually explicit expression, is simply impossible and unreasonable.[40] No regime, on their account, has ever allowed unlimited free speech, especially when people widely affirm some kinds of speech as harmful or offensive. I think this is true on some level. Trying to defend an abstract idea of absolute frees peech is conceptually unsound. I think that a doctrine of free speech is instead 16 better considered in terms of procedural constraints on how states may regulate.

This works in a fashion in the United States, where a great deal of state regulation of expression is permitted in practice but, due to the First Amendment, that regulation is required to follow a norm of non-content discrimination.[41] This gives states and the federal government sufficient power to prohibit the possession of images of real abuse, as well as the non-consensual sharing of intimate images. Yet the content produced using consensual events and fictional performances are constitutionally protected. There are a great many ways in which the US justice system fails to live up to the standards of a free society, but the relative strength of the First Amendment shows that a consistent protection of free expression is not just part of a liberal utopia but a realistic scenario worth striving for in the United Kingdom.



  • Attwood, F. & Smith, C. (2010) Extreme Concern: Regulating ‘Dangerous Pictures’ in the United Kingdom. Journal of Law and Society. 37 (1), 171–188.
  • Attwood, F. & Walters, C. (2013) Fifty Shades and the law: Regulating sex and sex media in the UK. Sexualities. [Online] 16 (8), 974–979.
  • Cowen, N. (2015) Millian Liberalism and Extreme Pornography: Millian liberalism and extreme pornography. American Journal of Political Science. [Online] 60 (4), 509-520.
  • Diamond, M. (2009) Pornography, public acceptance and sex related crime: A review. International Journal of Law and Psychiatry. [Online] 32 (5), 304–314.
  • Dymock, A. (2012) Comment: Prosecutors, extreme porn and 50 Shades of Grey. 8 August. [online]. Available from: of-grey.
  • Dymock, A. (2013) ‘Criminalising Extreme Pornography: Five Years On’ – Dymock on The question of regulating pornography: is criminalisation the answer? Inherently Human. 22 May. [Online]. Available from: https://inherentlyhuman.wordpress. com/2013/05/22/criminalising-extreme-pornography-five-years-on-dymock-on-the-question-of-regulating-pornography-is-criminalisation-the-answer/.
  • Ferguson, C. J. & Hartley, R. D. (2009) The pleasure is momentary...the expense damnable? Aggression and Violent Behavior. [Online] 14 (5), 323–329.
  • Itzin, C. et al. (2007) The evidence of harm to adults relating to exposure to extreme pornographic material: a rapid evidence assessment (REA).
  • Joyal, C. C. et al. (2015) What Exactly Is an Unusual Sexual Fantasy?: Sexual Fantasies in the General Population. The Journal of Sexual Medicine. [Online] 12 (2), 328–340.
  • McGlynn, C. & Rackley, E. (2009) Criminalising Extreme Pornography: A Lost Opportunity. Criminal Law Review. 4245–260.
  • McGlynn, C. & Ward, I. (2009) Pornography, Pragmatism, and Proscription. Jour- nal of Law and Society. 36 (3), 327–351.
  • Murray, A. D. (2009) The reclassification of extreme pornographic images. Modern Law Review. 72 (1), 73–90.
  • Rackley, E. & McGlynn, C. (2013) Prosecuting the possession of extreme pornography: a misunderstood and misused law. Criminal law review. (5), 400–405.
  • Strossen, N. (2000) Defending Pornography: Free Speech, Sex, and the Fight for Women’s Rights. 2nd edition. New York: New York University Press.
  • Strossen, N. (2007) ‘Pornography’, in John Meadowcroft (ed.) Prohibitions. London: Institute of Economic Affairs. pp. 139–163.


Foot Notes

[1] Nick Cowen, “Millian Liberalism and Extreme Pornography,” American Journal of Political Science, December 2015, 60 (4), 509-520n/a–n/a,doi:10.1111/ajps.12238

[2] John Woodhouse, “Extreme Pornography,” Briefing paper (London: House of Commons Library, January 7, 2016), 7.

[3] “CPS Violence against Women and Girls Crime Report 2014-2015” (London: Crown Prosecution Service, June 2015).

[4] See section 86.4: file/428937/outcomes-by-offence-tables.xlsx

[5] Laurence Dodds, “‘Tiger Porn’ Bus Driver Fights for Change in Law,” Telegraph, October 28, 2014, Online edition,

[6] “Criminal Justice and Courts Bill,” Impact Assessment (London: Ministry of Justice, June 18, 2014), 8,

[7] Brett Kahr, Sex and the Psyche: The Truth about Our Most Secret Fantasies (London: Penguin, 2008), 88.

[8] Ibid., 588–590.



[11] Christopher J. Ferguson and Richard D. Hartley, “The Pleasure Is Momentary...the Expense Damnable?,” Aggression and Violent Behavior 14, no. 5 (September 2009): 323–29, doi:10.1016/j.avb.2009.04.008.

[12] Milton Diamond, “Pornography, Public Acceptance and Sex Related Crime: A Review,” International Journal of Law and Psychiatry 32, no. 5 (September 2009): 312, doi:10.1016/j.ijlp.2009.06.004.

[13] Ibid.

[14] Catherine Itzin, Ann Taket, and Liz Kelly, “The Evidence of Harm to Adults Relating to Exposure to Extreme Pornographic Material: A Rapid Evidence Assessment (REA)” (Ministry of Justice, September 2007),

[15] Ibid., 1.

[16] Feona Attwood and Clarissa Smith, “Extreme Concern: Regulating ‘Dangerous Pictures’ in the United
Kingdom,” Journal of Law and Society 37, no. 1 (March 2010): 175.

[17] Todd D. Kendall, “Pornography, Rape, and the Internet” (The John E. Walker Department of Economics, July 2007), pdf.

[18] Paul J. Wright, Robert S. Tokunaga, and Ashley Kraus, “A Meta-Analysis of Pornography Consumption and Actual Acts of Sexual Aggression in General Population Studies: Pornography and Sexual Aggression,” Journal of Communication, December 2015, n/a–n/a, doi:10.1111/jcom.12201.

[19] Itzin, Taket, and Kelly, “The Evidence of Harm to Adults Relating to Exposure to Extreme Pornographic Material: A Rapid Evidence Assessment (REA),” 37.

[20] Ibid., 4.

[21] Bernard Williams, Report of the Committee on Obscenity and Film Censorship, vol. 7772, Command
Papers (London: The Stationary Office, 1979), 79.

[22] Berl Kutchinsky, “The Effect of Easy Availability of Pornography on the Incidence of Sex Crimes: The Danish Experience,” Journal of Social Issues 29 (1973): 163–81; Berl Kutchinsky, “Pornography and Rape – Theory and Practice: Evidence from Crime Data in 4 Countries Where Pornography Is Easily Available,” International Journal of Law and Psychiatry 14 (1991): 47–64.

[23] Kutchinsky, “Pornography and Rape – Theory and Practice: Evidence from Crime Data in 4 Countries Where Pornography Is Easily Available,” 61.

[24] Kutchinsky, “Pornography and Rape – Theory and Practice: Evidence from Crime Data in 4 Countries Where Pornography Is Easily Available.”

[25] Carol Harrington and Tainui Neilson, Review of Research on Sexual Violence in Audio-Visual Media Commissioned (Office of Film and Literature Classification, 2009),

[26] W. Wongsurawat, “Pornography and Social Ills: Evidence from the Early 1990s,” Journal of Applied Economics 9, no. 1 (2006): 206.

[27] Clare McGlynn and Ian Ward, “Pornography, Pragmatism, and Proscription,” Journal of Law and Society 36, no. 3 (2009): 327–51; Clare McGlynn and Erika Rackley, “Justifying the Criminalisation of Extreme Pornography,” Inter Alia 1/2, no. 7 (2010): 8–11.

[28] David Ley, Nicole Prause, and Peter Finn, “The Emperor Has No Clothes: A Review of the ‘Pornography Addiction’ Model,” Current Sexual Health Reports 6, no. 2 (June 2014): 94–105, doi:10.1007/s11930-014-0016-8.

[29] Christie Davies, The Strange Death of Moral Britain (New Brunswick, N.J.: Transaction Publishers, 2004), 93.

[30] Lewis Smith, “No 10 Police Officers Arrested on Suspicion of Sending Each Other ‘Extreme Pornographic Images,’” Independent, February 8, 2014, sec. Crime, news/uk/crime/police-officers-arrested-on-suspicion-of-sending-each-other-extreme-pornographic- images-9116209.html.

[31] “Birmingham Teaching Staff Held in ‘Extreme Porn’ Inquiry,” BBC News, March 13, 2014, sec. Birmingham & Black Country,

[32] Lu Han, Siddhartha Bandyopadhyay, and Samrat Bhattacharya, “Determinants of Violent and Property Crimes in England and Wales: A Panel Data Analysis,” Applied Economics 45, no. 34 (December 2013): 4820–30, doi:10.1080/00036846.2013.806782; Robert Witt, Alan Clarke, and Nigel Fielding, “Crime and Economic Activity: A Panel Data Approach,” British Journal of Criminology 39, no. 3 (Summer 1999): 391–400; Andrew Von Hirsch et al., Criminal Deterrence and Sentencing Severity (Oxford: Hart, 1999).

[33] William Spelman, “Prisons and Crime, Backwards in High Heels,” Journal of Quantitative Criminology 29, no. 4 (April 13, 2013): 643–74, doi:10.1007/s10940-013-9193-2; William Spelman, “Specifying the Relationship Between Crime and Prisons,” Journal of Quantitative Criminology 24, no. 2 (February 29, 2008): 149–78, doi:10.1007/s10940-008-9042-x.

[34] Daniel S. Nagin, “Deterrence: A Review of the Evidence by a Criminologist for Economists,” Annual Review of Economics 5, no. 1 (January 2013): 83–105, doi:10.1146/annurev-economics-072412-131310; F. T. Cullen, C. L. Jonson, and D. S. Nagin, “Prisons Do Not Reduce Recidivism: The High Cost of Ignoring Science,” The Prison Journal 91, no. 3 Suppl (July 19, 2011): 48S – 65S, doi:10.1177/0032885511415224.

[35] Nadine Strossen, “Pornography,” in Prohibitions, ed. John Meadowcroft (London: Institute of Economic Affairs, 2007), 139–63.

[36] Criminal Justice and Courts Act, 2015, sec. 33, section/37/enacted.

[37] Wendy Macdowall et al., “Associations between Source of Information about Sex and Sexual Health Outcomes in Britain: Findings from the Third National Survey of Sexual Attitudes and Lifestyles (Natsal-3),” BMJ Open 5, no. 3 (March 5, 2015): e007837–e007837, doi:10.1136/ bmjopen-2015-007837.

[38] Sylvia Walby, Jude Towers, and Brian Francis, “Is Violent Crime Increasing or Decreasing? A New Methodology to Measure Repeat Attacks Making Visible the Significance of Gender and Domestic Relations,” British Journal of Criminology, December 31, 2015, azv131, doi:10.1093/bjc/azv131.

[39] Charlene K. Baker et al., “Domestic Violence, Housing Instability, and Homelessness: A Review of Housing Policies and Program Practices for Meeting the Needs of Survivors,” Aggression and Violent Behavior 15, no. 6 (November 2010): 430–39, doi:10.1016/j.avb.2010.07.005.

[40] McGlynn and Ward, “Pornography, Pragmatism, and Proscription.”

[41] James Weinstein, “An Overview of American Free Speech Doctrine,” in Extreme Speech and Democracy, ed. Ivan Hare and James Weinstein (Oxford ; New York: Oxford University Press, 2009), 220–32.