Ch 12. Sex

Even now when women feature in crime stories they are typically depicted as victims not perpetrators. And their demise is especially newsworthy if, like Miss Darrow, they are young and decorous and ideally middle class.

A classic example which made headlines around the world was the murder of British student Meredith Kercher who had her throat cut while studying Italian in Perugia in 2007. The case attracted special fascination because one of the accused was another attractive female, her American flatmate Amanda Knox.




Around the world, with remarkable consistency, women are about a fifth as likely as men to be arrested, and usually for less serious offences.

In England and Wales females make up only about 21 per cent of all those arrested for indictable offences, most of which are handling stolen property or drug offences. Only 10 per cent of adult women sentenced to prison are convicted of offences involving violence, and women’s involvement in crime tends to be fairly consistent across industrial countries. In the US, females account for just over 20 per cent of arrests and around 16 per cent of all convictions. Self-report surveys, in which people are asked anonymously to confess their bad behaviour, show a similar pattern: the fairer sex really does seem to be fairer, admitting to only 14 per cent of violent offences. (Sources: Ministry of Justice, London, and US Dept of Justice criminal offender statistics.)




Why, all things considered, do women suffer far less deliberate injury than men with only a third of the risk of being murdered?

According to the British Crime Survey, recorded crime rates and other official studies, compared to men, women are almost half as likely to suffer violence and make up less than a third of murder victims, but tend to worry more about crime and have rather more confidence in the criminal justice system.


And why, if equality is a goal, do women get much shorter sentences than men for similar offences?

Women are much less likely to receive custodial sentences than men, except for first offences of violence where the imprisonment rate is similar, and women generally receive shorter sentences for a comparable offence. Only about 5 per cent of the UK prison population is female. Even in the US, where the proportion of women jailed has increased significantly, the figure is only 7 per cent. Part of the leniency is because of women’s role as parents. Mothers are much more likely to gain sympathy from a court than is the case with fathers. For example, there was national outrage when a serial burglar was spared prison so he could look after his five children. []




Or, paradoxically, if they cross a Rubicon of sex crime, why are they reviled much more than men?

Crimes against children always cause anger and revulsion but more so with women who apparently are thought to have betrayed their sex. The child killer Myra Hindley remained an icon of evil for almost forty years until and beyond her death in 2002, and even a child killer’s girlfriend, Maxine Carr, became a national hate figure. (Myra Hindley became infatuated by a bank robber, Ian Brady, and between 1963 and 1965 the two of them abducted and tortured five children for sexual gratification. Despite evidence of repentance some years later, there were frequent death threats against Hindley and she remained in prison until her death at the age of sixty. Maxine Carr was jailed for perverting the course of justice after she gave a false alibi to protect her boyfriend Ian Huntley who, in her absence in August 2002, had murdered two ten-year-old schoolgirls.)




When researchers failed to find much ugliness, excess hair or lumpiness in fallen women, Sigmund Freud proposed that when women turned from innate passivity to crime it was because they were envious of penises.

Of course Freud was big on penises, first elaborating his theory in 1908. He thought boys were sexually attracted to their mothers (the Oedipus complex) and were kept in line by fears that their father would castrate them. Hence they developed a conscience or superego. Girls could scarcely fear castration but fancied their fathers and so were generally compliant to keep in dad’s good books. Normal girls eventually overcame penis envy by giving birth, but some women started offending in order to emulate men. See Sigmund Freud, Some Psychical Consequences of the Anatomical Distinction Between the Sexes, 1925, or The Standard Edition of the Complete Psychological Works of Sigmund Freud, ed. James Strachey, Vintage Classics, London, 2001.




Freda Adler, the first notable feminist on the criminology scene, rejected the thought that females are the weaker sex, innately programmed for nurturing, biologically predisposed to being good. In Sisters in Crime: The Rise of the New Female Criminal, she seemed determined to prove that women are intrinsically as nasty as men.

Freda Adler, Sisters in Crime: The Rise of the New Female Criminal, McGraw Hill, 1976. ‘In the same way that women are demanding equal opportunity in fields of legitimate endeavor, a similar number of determined women are forcing their way into the world of major crimes’ (p. 13).




A liberal American lawyer, together with a feminist sociologist, trawled through Old Bailey records from the eighteenth and nineteenth centuries. They reported that for the first 100 years or so almost half the defendants were female, three or four times the proportion we see now.

Malcolm Feeley and Deborah Little, The Vanishing Female: The Decline of Women in the Criminal Process, 1687–1912, Law and Society Review 25, 1991, 719.




Certainly a lot of women went to jail. In 1900 almost a fifth of all prisoners were female, compared to one-twentieth today.

Jonathan Woodcock, Sentencing provisions and recent sentencing trends, July 2012.




By 2005 a survey of 30,000 primary and secondary school pupils in England and Wales revealed a third of girls indulged in vandalism, just the same as boys, and stole almost as much as well (with 25 per cent of girls admitting to theft compared to 29 per cent of boys), though boys were still twice as likely to truant and were four times more likely to carry a knife.

Derrick Armstrong, Jean Hine, Sue Hacking, Remos Armaos, Roy Jones, Nicolai Klessinger, Alan France, Children, risk and crime: the On Track Youth Lifestyles Surveys, Home Office Research Study 278, 2005.




…according to one of the authors:

The good news, and perhaps unexpected, is that the 2005 youngsters we studied have less problematic behaviour than those in the 1985 cohort. The bad news, however, is that twenty years ago boys drugged, drank, smoked, truanted, stole, vandalised and fought more than girls. Today it is very different. Girls now significantly smoke and binge-drink more than boys. They truant, steal and fight at similar rates to boys but engage in under-aged sex earlier than boys.

The surveys were part of a programme to reduce truancy, delinquency and other problem behaviours, and are detailed in Rcd Williams & Colin Prichard, Breaking the Cycle of Educational Deprivation, Open University Press, ISBN: 0335219187. The sample sizes were small, kids may simply have been more honest in admitting bad behaviour in the later questionnaire, and the findings might only reflect behaviours on the English south coast. The quotation is from Professor Colin Pritchard of Bournemouth University’s Institute of Health and Community Studies, cited by PA News, 18 May 2006 and confirmed through correspondence with the author.




Arrest rates corroborated the trend. In the US the FBI reported a significant rise in bank raids led by women1 and in 2007, for the first time since the war, more women were detained for violence in England and Wales than for shoplifting.2

1According to the FBI, 6.2 per cent of bank heists in 2008 were committed by women, a 25 per cent increase in five years in what appeared to be a consistently rising trend. (Source: CNN, 20 February 2009.)

2 Arrests for Recorded Crime (Notifiable Offences) and the Operation of Certain Police Powers under PACE England and Wales 2006/07: Statistical bulletin, Ministry of Justice, London, 30 July 2008, p. 2.




Women still commit only one murder for every nine perpetrated by men1 and they remain conspicuously absent from most big-league crime outrages. Only one in sixty of America’s mass shootings like Columbine, Virginia Tec or Newton was committed by a woman.2

1The best long-term figures come from the US, where there are aggregated homicide data from 1980 through 2008. The offending rate for males (15.1 per 100,000) was almost nine times higher than the rate for females (1.7 per 100,000). Source: Alexia Cooper and Erica Smith, ‘Homicide Trends in the United States, 1980–2008’, US Bureau of Justice, 16 November 2011, NCJ 236018[] An earlier study of almost a quarter of a million US homicides between 1976 and 1987 found women perpetrated 14.7 per cent of homicides. Source: A. L. Kellermann and J. A. Mercy, ‘Men, women, and murder: gender-specific differences in rates of fatal violence and victimization’, The Journal of Trauma, 1992, Vol. 33, No. 1, pp. 1–5 [].

Women are also less likely than men to engage actively in genocide. It has long been known that women cannot be relied upon to hold a distinctive moral line in upholding liberal values; females are as capable as males of prejudice, religious zealotry and political extremism. For example, at times Hitler had more support from women than from men (notably in July 1932 when some 6.5m German women voted Nazi) []. An analysis of archives in central and Eastern Europe has also revealed that tens of thousands of women took part in casual acts of sadism and murder (See Hitler’s Furies, German Women in the Nazi Killing Fields, Wendy Lower, Chatto & Windus, 2013). But of about 55,000 guards who served in Nazi extermination camps only 3,700 were women. Women also played a leading role in the Rwandan genocide of 1994, with an African Rights Report claiming, “The extent to which women were involved in the killings is unprecedented anywhere in the world” []; but even there women more often acted as “cheerleaders”, providing psychological and physical motivation to killers who were mostly male.

2  In 2006 in California Jennifer Sanmarco shot a neighbour then drove to the mail sorting office where she used to work and shot dead six other people before committing suicide. The UK has not experienced a similar lone female mass killer since Mary Ann Cotton was executed in 1873 for murdering her mother, three husbands, a lover, eight children and several others.




As crime rates tumbled girls and women began to lead the downward trend. Arrest rates for women began to drop twice as fast as those for men,1 and supervision orders for girls dropped much faster than those for boys.2

1Women and the criminal justice system, Ministry of Justice, London, 2012. Between 2006/07 and 2010/11 there was an 8 per cent reduction in the number of arrests by police forces in England and Wales (from 1,482,156 to 1,360,451). There was a 13 per cent decrease for females and a 7 per cent decrease for males. []

2Youth Justice Statistics 2010/11 England and Wales, Youth Justice Board, London, 2012, p. 18. The number of supervised girls aged ten to fourteen was 6,121 in 2010/11, a reduction of 39 per cent from 2009/10 while the number of boys aged ten to fourteen was 16,076 in 2010/11, a reduction of 25 per cent. Among fifteen- to seventeen-year-olds the drop was 24 per cent for girls and 14 per cent for boys. []




In this Darwinian view, women steal less for the same reason they tend to earn less: they are as competitive as men but with different priorities.1 And it does seem to be true that while girls may not have superior morality they do have different tactics.2

1Anne Campbell, A mind of her own: The evolutionary psychology of women. Oxford: Oxford University Press, 2nd ed., 2012.

2From an evolutionary perspective there is no reason why men and women would have different morals, but instead would have different ways of reaching their objectives. However, this explanation does not satisfy the politics of sex. The internet is angry with sometimes facile exchanges between (mostly American) feminists and men’s groups about whether men or women are basically nastier. Both sides are highly selective but the misogynists’ arguments are often highly misleading. For example, they claim that many Nazi concentration guards were women, but in fact only 3,600 were out of 55,000. They point out that women commit more child murders than men, but this is usually in the first year of infancy because of post-natal depression; and that children are three times more likely to be fatally abused in mother-only families than in father-only households, but the killer is often a male. (As with lions and some other mammals there can be primitive evolutionary advantages for infanticide. At any rate infanticide by step-parents is 100 times more common than by biological parents – sometimes called the Cinderella effect. See Martin Daly and Margo Wilson, The Truth About Cinderella, Weidenfeld and Nicolson, 1998.)




In 1989 the Canadian Journal of Behavioural Science published the results of a survey that was celebrated as a classic exposé of the problem of ‘battered wives’.

L. W. Kennedy and D. G. Dutton, The incidence of wife assault in Alberta, Canadian Journal of Behavioural Science, 21, 1989, pp. 40–53.




While 10.8 per cent of the men surveyed had pushed, grabbed or thrown objects at their spouses, 12.4 per cent of women had done so too; and though 2.5 per cent of men used serious violence, so did 4.7 per cent of women. Marilyn Kwong, who exposed the expurgation, examined eight other studies too, and found the pattern was universal.

Marilyn J. Kwong and Kim Bartholomew, Simon Fraser University, Donald G. Dutton, University of British Columbia, Gender differences in patterns of relationship violence in Alberta, Canadian Journal of Behavioural Science, 1999, 31:3, 150–60.




Feminism did a vital job putting domestic violence on the agenda… but the very success of feminism and its flattery by mainstream authority caused the politics of dissent to become institutionalised. The politics soon came to define the problem.

A UK review of the literature on domestic violence notes that feminist beliefs about patriarchy have come to dominate policy regardless of evidence and to the exclusion of contra-evidence. This also applies to academic research where, “Social scientists are often motivated by an ideologically based view of their subject that makes it hard for them to think beyond a narrow range of acceptable theoretically stances. This is particularly apparent in research on violence between intimate partners.” Louise Dixon, John Archer and Nicholas Graham-Kevan, “Perpetrator programmes for partner violence: are they based on ideology or evidence?” Legal and Criminological Psychology, 2012. Vol 17, pp 196-215.




The politics soon came to define the problem, and for decades professional interventions assumed that men were always the aggressors and if women were violent it must have been in self-defence.

The best-known approach was called the Duluth model after a town in Minnesota where an especially vicious domestic homicide in 1980 gave rise to a project to rescue battered women and prosecute their male partners. The Duluth approach also involved raising women’s consciousness to assume that men in general use physical and sexual violence to assert control.




One researcher complains that we have been bamboozled by ‘nothing less than the rejection and devaluing of the scientific method, in favour of politically acceptable interpretations of discursive material’.1 Despite all this, officials continue to use exaggerated claims, and many agencies, including the police, help propagate the propaganda.2

1 John Archer, ‘The trouble with “doing boy”’, The Psychologist, 2004, Vol. 17, No. 3, p. 132. The author’s own survey of the literature reveals scientific howlers, overgeneralisation, overreliance on estimates, painting-in detail or airbrushing it out.

2 Criminologist Professor Betsy Stanko , who ran the Economic and Social Research Council’s domestic violence programme, was influential in promoting the women-as-victims agenda, and was taken on as adviser to the Cabinet Office and then to the Metropolitan Police, who have been keen to be seen to be politically correct on the issue.




Many such reports also ignored the thought that women can be violent to men.

It took many years before there was an academic backlash to the feminist agenda, such as a review of studies called ‘Riding the Donkey Backwards: Men as the Unacceptable Victims of Marital Violence’, by neurologist Malcolm George (Journal of Men’s Studies, Vol. 3, No. 2, November 1994, pp. 137–59).

Since then, when studies have been ‘gender-inclusive’, avoiding the preconception that violence is typically male on female, they tend to show “that women are as likely to be physically aggressive towards their partner as men are, if not more so.*” In fact, at least in Western nations, there appears to be a contrasting pattern of sex differences: men are more aggressive to men than women are to women, whereas women are as aggressive to their male partners as men are to their female partners. This may be because boys are taught from an early age that it is wrong to hit girls whereas several studies have demonstrated there is much more social acceptance of girls showing aggression to boys. Women can also be as controlling as men. When 1,100 British students were given questionnaires about their own aggression, women were significantly more physically and verbally aggressive to their partners than men were. And in contrast to the feminist ‘male control theory’, males and females reported that their partners used controlling behaviour (sometimes called ‘intimate terrorism’) at a similar rate. (Elizabeth Bates, Nicola Graham-Kevan and John Archer, ‘Testing Predictions from the Male Control Theory,’ Aggressive Behavior, 2014, Vol 40, *p43.)[]




Erin Pizzey, the feminist who founded one of the world’s first women’s refuges, has been trying ever since to set the record straight:

I will never forget one woman, who was staying in my refuge, telling me, in chilling tones, ‘knives are a great leveller’. That is the reality of domestic violence. It is far less clear-cut than the ideologues like to pretend, with their neat division between female victims and male oppressors. The truth is that much of the violence takes place in squalid, tortured relationships, often involving drink and drugs, where both partners are guilty of verbal and physical assault. In the refuge I opened in 1971, for example, of the first 100 women through the door, 62 admitted that they had also perpetrated violence against their partners.

Erin Pizzey, Daily Mail, 29 July 2008. The Goldhawk Road Women’s Liberation Movement Centre in west London became the world’s first women’s refuge when a woman escaping an abusive husband was allowed to stay there overnight. I knew Erin Pizzey slightly (she was married to a BBC reporter who was a colleague of mine on Man Alive) and she was always driven by compassion more than ideology and was well aware that domestic violence was part of family dynamics, not a conspiracy of men to keep women in subservience. She was equally concerned with child victims (she had been abused by her own parents) and is, in short, precisely the sort of woman who hated the biased and anti-male stereotypes which flourished in the wake of her domestic violence initiative.[]




As early as the 1980s, large-scale US research began to endorse Erin Pizzey’s view: women often initiated violence and could give as bad as they got.

Stets and Straus, in Don Sutton, ed., Rethinking Domestic Violence, University of BC Press, Vancouver, 2006, ISBN: 9780774810159.




Although in many cases neither men nor women reported injury or emotional effects, about one in ten in both genders had suffered bleeding or broken bones and 3 per cent or men and 2 per cent of women had attempted suicide.

A major report on domestic violence in England and Wales, using data between 2012 and 2015, concluded that males accounted for more than a third of all domestic violence victims (650,000 men and 1.2m women, according to the Crime Survey of England and Wales) and were the victims of more than a quarter of domestic murder (117 males and 315 females). (Statistical Bulletin, ‘Domestic abuse in England and Wales: year ending March 2016,’ ONS, 9 December 2016 [].

An example unusually caught over a long period on video formed the basis of a 2024 TV documentary, My Wife, My Attacker, The Secret Footage (Channel 4, 18 March 2024). A Yorkshire woman who worked at the Ministry of  Justice had blamed her husband for abuse over twenty years but recordings from a baby-minder nannycam proved she was consistently the aggressor, belittling, threatening and so injuring her partner that she was jailed for four years for coercive and violent behavious. (The Times, 15 March 2025).

While domestic violence as a whole appears to be declining, the proportion of male victims has been fairly consistent over many decades. A British Crime Survey in 1995 found around a quarter of those assaulted were men, but this was thought to be an underestimate since it was largely based on face-to-face interviews – whereas more recent data are obtained from self-completion questionnaires in which people tend to divulge more. Thus, a BCS self-report survey sampled 10,800 people in 1999 and found men made up a third of those injured and a quarter of those who suffered repeated violence. (Catriona Mirrlees-Black, Domestic Violence: Findings from a new British Crime Survey self-completion questionnaire, Home Office Study 191, London, 1999.) Another big survey in 2001 put the proportion of male victims at 40 per cent, with injured males making up 43.5 per cent of assaults classed as severe; and three years later another and still bigger survey (of 22,500 people) found much the same. (Supplement to 2004/5 BCS, see Dewar Research, Government statistics on domestic violence, August 2006. []) However, when all the data from 1995 to 2010 were put together it looked like about a third of all domestic violence victims are male and they suffer about a quarter of all assaults. (Government Statistics on Domestic Violence: Estimated prevalence of domestic violence, England and Wales 1995 – 2010/11, Dewar Research, 2012. [])




Not that most men would confide how they were injured.

Gender stereotyping works both ways, and assumptions about manliness are so deeply rooted that even people who regard themselves as thoughtful and open-minded tend to deride male victimisation as absurd or at least vanishingly rare. The idea is so counterintuitive that even hard data are often rejected as implausible. And while there are thousands of academic papers and whole academic journals devoted to Violence Against Women, few scholars bother to look at aggression towards men.

Even if persuaded that it does happen people tend to blame the victims. For example, when 168 American adults were shown matched vignettes of male and female intimate violence, they ranked male victims as much less deserving of sympathy than female ones, and less worthy of medical or psychological support. Just as importantly, they disparaged male victims: they rated them less successful, less powerful and less socially desirable than female victims. Little wonder that researchers conclude, ‘preconceived notions based on gender roles impede the identification of men as victims’. (Thomas A Paul and Kathleen Hart, ‘Third-party perceptions of intimate partner violence victimization in men,’ Psychology of Men & Masculinities, 2022 [].)

This third-party stereotyping is shared by male victims themselves, who fear they will be regarded as wimps or be suspected of being bullies who provoked retaliation. Interviews with 26 British men who self-identified as victims found a marked reluctance to seek help, and some encountered contempt and suspicion when they did so. ‘First and foremost, the importance of maintaining a sense of masculinity consistently underpinned the participants’ narratives.’ (Hogan, K. F., Clarke, V., & Ward, T., 2021, ‘Men’s experiences of help‐seeking for perpetrated intimate partner violence: A qualitative exploration’, Counselling & Psychotherapy Research, 21(4), 934–945.

It is hardly surprising that male victims are undercounted and overlooked. As Ken Gregory, a fit and powerfully-built man said after his wife tipped boiling water over his head in the worst of several unprovoked attacks, ‘As a man who is a bit older and who isn’t exactly small, there is a perception that you can’t be a victim of domestic violence. But it should be the same message that they put out for women – don’t be frightened, you don’t have to put up with it.’ (The Times, 13 March 2015).



The researchers point out that because women are usually physically less powerful they tended to come off worse but ‘naively’ believed that if they hit their partner he would not hit back. There may be something to this assumption since, at least in Western nations,  boys are taught from an early age that it is wrong to hit girls whereas several studies have demonstrated there is much more social acceptance of girls showing aggression to boys.

The Dunedin study shows, along with other studies, that women’s overall rates of partner violence perpetration are similar to those of men. This is not an isolated finding. Many studies have found that substantial numbers of women self-report abusive behaviours toward male partners, and epidemiological studies show that although males are more likely than females to engage in almost every type of violence, the single exception is family violence.

Terrie Moffitt, Avshalom Caspi, Michael Rutter and Phil Silva, Sex Differences in Antisocial Behaviour, Cambridge University Press, 2000.

An interesting example was provided by Britain’s shadow minister for preventing domestic violence who was herself arrested once for assaulting her husband. According to a newspaper report she acknowledged: “I’m not proud of what happened and I accept I was in the wrong.” (Sarah Champion MP, quoted in The Times, 24 Sept 2016 []).




Nor should we forget the extent of emotional bullying,1 where the wounds don’t show, or the effect on children,2 with the demonstrable likelihood that they will grow up to be violent themselves.3

1Maureen Outlaw, ‘No one type of intimate partner abuse: exploring physical and non-physical abuse among intimate partners’, Journal of Family Violence, Vol. 24, No. 4, May 2009, pp. 263–72. [doi: 10.1007/s10896-009-9228-5]

2When New Zealanders were studied from childhood through to their mid-twenties, almost half had been hit with an object by their parents, and 6 per cent described extreme physical punishment such as being beaten or knocked unconscious. These severe events were meted out equally to boys and girls, generally by the father or stepfather but sometimes by the mother, and usually out of rage rather than as a calibrated reprimand. Police became involved only when alerted by doctors or social workers when the children were very young.

3 John Hamel and Tonia Nicholls, ed., Family Interventions in Domestic Violence, Springer, NY, 2006, SBN10: 082610245X. This is one of the few books on domestic violence which is grounded in good science, which challenges glib paradigms, and which offers sound advice for tackling the problem.




Claire Turner, who founded a British support group after her female partner tried to strangle her, said:

You end up thinking that society will not think it serious enough because it was another woman who perpetrated the abuse. I did not report it. I really believed that women were great and incapable of being anything but nice to each other. But you come to realise that anybody in society has the potential to behave badly.

Quoted by The Times, London, 9 June 2007 following a conference organised by Greater Manchester Police.




Until at least the 1970s and ’80s the institutions of the state were steeped in prejudice. There was tactless condescension from judges1 and, as a seminal TV documentary showed with shocking candour in the 1980s, police officers sometimes treated rape complaints with crass insensitivity.2

1In 1982 an English judge caused outrage when he fined a rapist (rather than imprison him) because of what he called the victim’s ‘contributory negligence’. She had been hitchhiking. Nowadays nobody would say such a thing – but almost everyone would think it (or at least almost everyone would think it unwise for a girl to climb into a car with a stranger).

2 Film-makers Roger Graef and Charles Stewart spent a year filming work at a police station in Reading, Berkshire, and Episode 3 of their nine-part BBC documentary, Police (1982), showed a woman who complained of rape being bullied and taunted by three male officers on the presumption that because she had a history of mental illness she had probably made the story up. The screening provoked outrage and led to a wholesale review of how sex offences were dealt with by British police.





There was a widely held assumption that victims had probably been asking for it or at least had rashly encouraged it. Conviction rates were said to be a risible 10 per cent.

The 10 per cent figure is an approximation across the 1980s and across England, Wales, Scotland and Northern Ireland. But the calculation is based on interpretations provided by the Home Office Sex Offenders Review Team then led by Betty Moxon who overruled a minority report – also written by a woman – which challenged the maths. []




Several police forces set up dedicated sex crime facilities, with officers selected and trained for sensitivity

Police sensitivity to sex crimes was, and probably remains, patchy. A survey by the Association of Chief Police Officers in 2004 found that fewer than ten out of the forty-three police forces in England and Wales had dedicated rape investigation teams, and eight forces were not even using early evidence kits, which allow samples to be taken at the scene.




When that failed to raise conviction rates, in 2007 England’s Solicitor General announced packages of targets and ‘guidelines’ for judges and juries which would shift the presumption of innocence towards a presumption of guilt.

These were government proposals announced in December 2007, and several years later, with the conviction rate remaining broadly static, Alison Saunders head of the CPS, launched another high profile action plan to get more guilty verdicts. But some lawyers have expressed grave reservations and have advised that courts are not always the best way to resolve issues of sexual violence. For example, in June 2014 the attorney general, Dominic Grieve QC, warned that enthusiasm for justice could backfire. He pointed out that a policy to prosecute more cases would require the CPS to go to court with lower standards of evidence than normal, which juries would be likely to reject. In other words, a policy of boosting prosecutions would likely lead to a lower level of convictions. Such concerns are rejected, often stridently, by feminists, with Liz Kelly of the End Violence Against Women Coalition accusing Mr Grieve of denying rape victims equal access to justice (The Times, 3 July 2014, p5).




So far as we can tell, roughly 4 per cent of women in England & Wales are raped at some point in their lives, some repeatedly, and about 0.6 per cent of women (and 0.1 per cent of men) are victims of rapes and other serious sexual assaults each year. Yet, despite the fact that reporting rates have soared, fewer than 20 per cent go to the police.

A review of how rape cases are handled accepted as best guess that 11 per cent of those who have been raped tell the police about it (The Stern Review, Government Equalities Office, 2010, p. 32) but more recent evidence suggests reporting rates are higher. English and Welsh Crime Survey data from self-completion questionnaires aggregated between 2009 and 20012 estimated that around 20 per cent of women in England and Wales are victims of some sort of sexual offence or attempt at some point in their lives and nearly 4 per cent are raped. On average 2.5 per cent of women aged sixteen to fifty-nine and 0.4 per cent of men had experienced some form of unwanted sexual advance in the previous twelve months, ranging from indecent exposure through unwanted touching to rape. About 0.5 per cent of females and 0.1 per cent of males had been sexually penetrated, suggesting there were around 75,000 female and 12,000 male victims of rape; of whom around 90 per cent knew the perpetrator. Fewer than 20 per cent of rape victims said they had told the police. This corresponds closely to police figures which recorded about 15,000 rapes on females and 1,300 male rapes.[]




Officers have sometimes pressured women to abandon complaints – if they have no crime it improves their detection rates1 – but there is no evidence that police fail to prioritise sex offences in general. In fact the detection rates for sex crimes are comparable with many other crimes including robbery, burglary and fraud.2

1The Independent Police Complaints Commission described a ‘sorry chapter’ for the Met’s Sapphire team which investigated rapes in Southwark in south London. Complainants had been ‘close questioned’, resulting in ‘widespread’ retractions so ‘fewer crimes were recorded and therefore targets were easier to reach’. The Times, 27 February 2013.

2 Sanction detections for rape average at 25 per cent, about the same as for fraud and as opposed to 17 per cent for burglary and 20 per cent for robbery. By contrast homicide detection rates are usually between 80 and 95 per cent (e.g. 83 per cent of homicides resulted in a conviction in 2009/10 and 95 per cent in 2010/11). Source: Paul Taylor and Rupert Chaplin, Crimes detected in England and Wales 2010/11, Home Office Statistical Bulletin, July 2011, pp. 15–16. []

The determination to press rape cases led police to name suspects in the hope that other victims would come forward, which was a far remove from traditional police reticence along the lines that, ‘a man is helping police with their inquiries’. In fact the Leveson inquiry into press intrusion had recommended that suspects should not be named unless charged. But publicising arrests became routine in Operation Yewtree, a huge investigation triggered by widespread abuse of girls by DJ Jimmy Savile. This naming and inevitable shaming of suspects led to complaints of ‘abuses of policing’ similar to some of the worst intrusions by the press: ‘Need we slide into a situation where police have a free pass to humiliate and inconvenience citizens on anybody’s say-so?’ (Libby Purves, The Times, 29 April 2013, p25). ACPO at first proposed a blanket ban on naming suspects to get consistency (and to end the ‘bizarre parlour game’ where journalists sought to persuade press officers to leak the names) but some reporters bridled at what The Press Gazette described as ‘secret arrests’; and which a former DPP, Lord Macdonald, called, ‘frankly sinister’. In one case (that of TV presenter Stuart Hall accused of sex abuse) Lancashire detectives said that most of the thirteen complainants became known to them because of the publicity. ACPO subsequently modified its guidance to allow naming where publicity could provoke new witnesses to come forward, but only when a suspect had been charged.




For rape specifically the conviction rate is around 33 per cent with a further 23 per cent of those accused found guilty of lesser charges such as sexual assault.

The Ministry of Justice announced in April 2013 that 63 per cent of rape trials result in a conviction.  Albeit the conviction is not always for rape. Analysis of 2011 court suggests 33 per cent of men charged with rape were found guilty as charged and another 23 per cent were convicted of lesser charges, usually sexual assault. (Tables provided to author by Ministry of Justice, January 2013. Also ‘Sexual offences in England and Wales’, 2011, Ministry of Justice, 2012.) The proportion of convictions appears to have risen: an earlier more detailed study concluded 46 per cent of rapes result in a guilty verdict. (Liz Kelly, Jo Lovett and Linda Regan, A gap or chasm? Attrition in reported rape cases, Home Office Research Study 293, February 2005, London. [])




In November 2005 Amnesty International published poll findings which suggested one in three people believes that women who are drunk or who are flirtatious are at least partly responsible if they are raped, and a quarter feel the same about rape victims who dress provocatively.

ICM survey of 1,095 adults, Amnesty International []




Yet for any other crime, compensation can be reduced according to ‘the conduct of the applicant before, during or after the incident’.

The Criminal Injuries Compensation Scheme 2012, Grounds for withholding or reducing an award, para 25, p. 12, Ministry of Justice, 2012.

Incidentally, nobody seems to have noticed the irony that on the very same day the Foreign Office blamed excessive drinking for a rise in arrests of Britons overseas, an announcement which also made front-page news but which was received as a welcome warning.

According to the Guardian editorial, ‘notions of provocation have no place in sexual violence’ and hark back to the days when ‘misogynistic moralising clouded views about where responsibility for rape belongs’ (12 August 2008). According to the Daily Mail the big story that day was ‘Shame of binge drinks Britain abroad.’




Not even in the licentious days of the Charles II Restoration in the seventeenth century was it acceptable for women to dress as provocatively as they have done in Western culture since the 1960s.

Some people regard it as anathema to mention clothing in the context of sexual assault as though provocative clothing might be considered as an excuse for rape, or at least as a mitigating factor. This is a very modern taboo – the importance of dressing modestly would have been unremarkable in generations past. And it is understandable given the history of blaming women for their own misfortune. But it should never be unacceptable to consider potential dangers, especially when the crime of rape can be so damaging.

As it happens there is little evidence either way on whether alluring clothing does increase vulnerability to assault, and if there is an effect it is not conspicuous. Sexual assaults do not seem more likely in beach resorts, where swimwear is the norm, than in city centres. Nor is modest clothing a defence against rape in cultures where sexual assault is much more commonplace than in the UK. Moreover, there is no obvious correlation between the fashions for naked flesh and attacks by strangers, which is the sort of assaults scanty outfits might be presumed to trigger. Thus, as the paragraph continues:

Yet so-called stranger rapes, the sort most often reported by the newspapers, make up a small proportion of rapes that women divulge through surveys.

Even so many feminists themselves are concerned that provocative clothing is, literally, provocative. Indeed, there is often loud antagonism to skimpily-clothed models in newspapers, where partial nudity is held to sexually objectify women. There is an apparent inconsistency with that view and the loud defence of scantily-clothed woman in the streets, which for some reason is held not to sexually objectify the individual.




Getting legless isn’t new – think of Hogarth’s classic satire, Gin Lane

The gin craze was itself an example of a social problem caused by changing situations – in this case legislation from 1689 which encouraged domestic distillation of gin while heavily taxing imported wines and brandies. Gin takes its name from Dutch juniper or jenever which was used to flavour the distillate. The Gin Act of 1751 (the year of Hogarth’s cartoon) helped bring the epidemic to a close.




As we saw in Chapter 9, half of all women who have had penetrative sex unwillingly do not think they were raped and this proportion rises strongly when the assault involves a boyfriend, or if the woman is drunk or high on drugs: they led him on, they went too far, it wasn’t forcible, they didn’t make themselves clear…

See for example: A. S. Kahn, J. Jackson, C. Kully, K. Badger, J. Halvorsen, Psychology of Women Quarterly, Volume 27, Number 3, September 2003, pp. 233–42, Blackwell Publishing. This questioned eighty-nine women, only thirty-three of whom classed their experience as rape. In this, as in other studies, women were more likely to regard non-compliant sex as rape when the assailant was not their boyfriend, if they experienced more forceful assaults, or when they found it especially upsetting.

The re-stating of this by a Sunday newspaper on the launch of this book was accompanied by furious protests, not least from Jo Wood of Merseyside’s Rape and Sexual Abuse Centre who was quoted as saying, ‘These comments are ill-advised, uneducated and simply wrong.’ This was followed by a storm of insults on Twitter and on feminist blogs, and abusive articles in the Sun, Mirror, The Guardian and The Sunday Times. The Mail on Sunday which has confected the row offered me a right of reply, but then buried it beneath a larger article, again purporting to accuse me of insensitivity to rape victims. However, Jo Wood subsequently said she had been misled by the Mail on Sunday. ‘Having read the full version I am satisfied there is no intention to criticise victims of rape and that the comments made, when read in context, actually strengthen the arguments for sexual violence crimes to be treated with the empathy and respect that victims demand.’ A number of rape victims wrote privately endorsing this chapter: ‘It is a great shame we can’t talk seriously about the very real problem of rape because any comment that isn’t baying for the blood of all offenders everywhere is screamed down. The vitriol makes sensible discussion impossible.’




It is known as DFSA, or drug-facilitated sexual assault. But had she really been drugged? What about thousands of other women who made similar assumptions?

Research on university students in the UK and US during the heightened fears of spiking found a majority believed drug-adulterated drink to be a bigger threat than drink-driving, being mugged or burgled, despite never having experienced it themselves. Many claimed to know someone else to whom it had happened and routinely protected their drinks, often taking with them to the toilet. The authors suggest that the spiking narrative brought traditional feminine qualities – including policing one’s physical boundaries and watchfulness for predatory males – to bear on women’s new exposure to public drinking. See: Adam Burgess, Sarah Moore and Pamela Donovan, ‘Embodying Uncertainty? Understanding Heightened Risk Perception of Drink “Spiking”,’ British Journal of Criminology, Vol. 49(6), 2009.




That same year the Forensic Science Service published a study into DFSA – the largest of its kind.

Journal of Clinical Forensic Medicine, August 2005; 12 (4): 175–86.




In 98 per cent of cases there was no evidence of drugs other than self-administered alcohol, cannabis or cocaine.

The study rejected the ‘widespread classic “date rape” scenario in which a sedative/hypnotic tablet, capsule or powder is covertly introduced into a victim’s drink’. In fact it found virtually no traces of drugs other than those the victims had admitted to taking themselves. Of the 1,000 samples only twenty-one cases were found where depressant substances could not be explained by voluntary use, plus nine others where the women said they had been openly offered a stupefying drug or forced to take one.




This is in line with other studies in both the UK1 and the US, one of which was trenchant in its conclusions: ‘Detailed examination of the testing results does not support the contention that any single drug, apart from alcohol, can be particularly identified as a “date rape” drug.’2

1A study in Cardiff in 2005 checked seventy-five allegations by hospital admissions that their drinks had been spiked. Most of the plaintiffs simply turned out to be drunk, two-thirds of them well over the legal drink-drive limit. A few urine tests (19 per cent) showed signs of illegal drugs, but these might have been self-administered, and none was Rohypnol or GBH. Nor is it likely that many of the patient’s drinks had been spiked with extra alcohol; the research included breathalysing 900 late-night revellers, and found binge drinking is routine, suggesting drinkers underestimated their consumption. Hywel Hughes, Rachael Peters, Gareth Davies, and Keith Griffiths, A study of patients presenting to an emergency department having had a ‘spiked drink’, Emergency Medical Journal, Vol. 24, 2007, pp. 89–91

2 Journal of Clinical Forensic Medicine, 8 (4), of Clinical Forensic Medicine, 8 (4), December 2001, pp. 197–205. The study analysed urine samples from 3,303 individuals who claimed to have been ‘date raped’ with the help of drugs. ‘Detailed examination of the testing results does not support the contention that any single drug, apart from alcohol, can be particularly identified as a “date rape” drug. Rather, the alleged sexual assaults may often take place against a background of licit or recreational alcohol or drug use, where alcohol and other drugs are frequently taken together.’




The media were so credulous that several TV soaps as well BBC news and newspapers warned about the growing menace, creating a household name for an obscure pre-anaesthetic sedative called Rohypnol.

Rohypnol (technically flunitrazepam) is a benzodiazepine like Valium used mostly for pre-anaesthesia or short-term sleep disorders. It has not been prescribed by the NHS since 1992 and is illegal in the US but small quantities circulate illegally, mostly from South America, for sale to clubbers who like the sense of feeling drunk but without a hangover. The street version is known as ‘roofies’, ‘roach’ or ‘rope’. There are other drugs which can cause amnesia, like GHB (gammahydroxy-butyrate) and which are more rapidly eliminated from the body, though GHB is evident in hair for several weeks.




In the one proven case of drinks spiking in Britain, in which a doorman was sent to prison for tipping the anaesthetic GHB into a woman’s drink, the victim collapsed and had to be rushed to hospital.

Michael Wright, a pub doorman, was sentenced to five years at Newcastle Crown Court in 2005 for tipping GHB into a woman’s drink. []




… the media repeatedly reinforce the impression that Rohypnol is almost undetectable. Not so: some of the studies were on samples taken within hours and in any case some of the metabolites of hypnotic drugs can be detected for days or even weeks.

Benzodiazepines or their metabolites (substances created when the drug is metabolised by the body) can be detected in blood for six to forty-eight hours, in urine for three to seven days, and in hair for up to ninety days. Rohypnol is best detected within thirty-six hours of administration, but it yields at least nine different metabolites which can be identified by screening.




And no doubt women who say their drinks were spiked mostly believe it.

Adam Burgess, Pamela Donovan and Sarah Moore, ‘Embodying Uncertainty? Understanding Heightened Perception of Drink “Spiking”,’ British Journal of Criminology, 2009 [doi:10.1093/bjc/azp04].




Prostitution might or might not be the world’s oldest profession – as Ronald Reagan said, politics bear a striking resemblance. But even in these days of sexual liberation it continues to be secretive and frequently reviled, as was homosexuality a generation back – with clients as reluctant to admit to paying for sex as prostitutes often are to providing it.

Attitudes to prostitution are complex. The impression of a sleazy, enslaved and hollow-eyed streetwalker jostles with images of a warm-hearted and good-natured tart; and many concubines have gone on to become successful and admired (Eva Peron springs to mind) while others became iconic images (almost every priceless nude in every great art gallery around the world depicts a prostitute). And while polite society still pretends prostitution is a deviant activity, the extent of the business speaks for itself. Nobody knows how many UK prostitutes there are (a wild guess at 80,000 made by health worker Hilary Kinnell in the 1990s is widely quoted as though it were gospel, even by the Home Office, which – as referenced later in this chapter – Ms Kinnell describes as ‘bizarre’) but for millions of men it is as natural to pay for a sexual encounter as it is to pay for a drink or a thrill at a theme park. For some it has the added excitement of danger as, week after week, tabloid papers testify: all those footballers and A-list celebrities who get caught in flagrante, in one case with a streetwalker in Los Angeles. And think back to the to the time when one of England’s most senior law officials, the Director of Public Prosecutions, was caught kerb-crawling in central London. Here was a man who knew precisely what the law was, who was aware that he could pay for sex legally by hiring a call girl to a hotel, or by looking through the small ads and visiting a private flat; but like a moth to the flame he gravitated to the one adventure that would make his heart race. We should never forget that one of the attractions of behaving badly is adrenalin.




Small working-class brothels were more or less endured for many years

Paying the Price, Home Office, London, July 2004, pp. 91–3.




Street prostitutes are twelve times more likely to be murdered than other women.

C. Gabrielle Salfati, ‘Prostitute Homicide: An Overview of the Literature and Comparison to Sexual and Non-Sexual Female Victim Homicide’, in David Canter, Maria Ioannou, and Donna Youngs (eds) Safer Sex in the City: The Experience and Management of Street Prostitution, pp. 51–68. The Psychology, Crime and Law Series. Aldershot: Ashgate, 2009.




… the traditional reaction to a dead prostitute – ‘and then the silly girl went and got herself murdered’ as one magistrate famously remarked

‘And then the silly girl went and got herself murdered’ is a quote ascribed to a West Midlands magistrate in 1990 describing the fate of Gail Henderson who was strangled while selling sex on the streets of Wolverhampton in 1990.

Or we could copy New Zealand, Australia and some European countries which have licensed bordellos so that girls can work together and be regulated.

Prostitution in most forms is legal and/or regulated in Germany, Greece, the Netherlands, Portugal, Spain, Switzerland and Turkey and in most of Austria. It is tolerated in Belgium, while Denmark, Finland, France, Italy and Sweden have laws resembling those in the UK.




Street girls are not your average ‘escort’ or ‘masseuse’, nor the student who finds a lucrative way of paying her way through college, the hard-working female who prefers having sex to sitting at a factory bench, or the self-employed businesswoman who has a call-girl venture rather than going into hairdressing.

Feminists are divided over prostitution. Some want it ‘decriminalised’ while others hold a powerful, and ironically patronising, view that all sex workers are ‘victims’ rather than intelligent free agents who are choosing their career paths like anyone else. Although prostitutes have written of their shame and self-abasement others have described it as ‘empowering’. The condescending view of prostitutes in popular culture – and from anti-prostitution feminists – echoes century’s old moralising attitudes to women and are usually underpinned by wild assumptions and extrapolations rather than scholarly research. The mainstream media rarely dares to tackle this taboo. A television producer colleague of mine once featured half a dozen prostitutes in a documentary, five women and one man; but the film appalled her head of department, who demanded a re-edit, because, while two found their trade distasteful, four of the six said they enjoyed their work or that it was better than alternative careers. Systematic research suggests these six were pretty representative. A survey of off-street sex workers in Cardiff found the women were attracted to, ‘the benefits of self-employment, flexible working hours, job satisfaction and providing a public service,’ and some were, ‘drawn to the work because of the desire for sexual pleasure: I get to have sex with nice young men; You get to have fantastic sex, good friendships with clients and a little bit of extra money… .’ (Tracey Sagar and Debbie Jones, ‘Off street sex workers and victim-oriented policymaking at the local level,’ Crime Prevention and Community Safety, Vol 16, No4, 2014.)

The same team had similar findings from university students involved in the sex industry. A self-selection survey conducted by the Student Sex Work project and led by Swansea University prompted 6,750 respondents of whom roughly 5% claimed to supplement their incomes by stripping, phone sex erotic dancing, naked butler services or prostitution. “Good money and flexible hours were the positive aspects that were noted most, but also sexual pleasure was important for almost half of the respondents.” The biggest negatives were the secrecy involved and concerns about staying safe. Tracey Sagar, Debbie Jones, Katrien Symons and Jo Bowring, The Student Sex Work Project, Research Summary, Swansea University, March 2015, p25. []

In general stigma is the greatest problem reported by prostitutes which makes it hard for many to be open about how they earned their money. In 2015 a study of British sex workers (196 women, 28 men and 12 transgender people) found job satisfaction was high with two thirds saying their job was ‘fun’ and many describing their career as ‘flexible’ and ‘rewarding’. Almost 40% of respondents had a university degree. But the challenge of explaining what they had been up to was one reason almost a quarter felt it hard to change profession. (Results published in March 2015 by the charity National Ugly Mugs based on research by Teela Sanders et al of Leeds University funded by the Wellcome Trust.)

A larger survey of almost 500 prostitutes, male and female, found the females were generally more positive about their career choice than the males. Most were motivated by a ‘more luxurious lifestyle’, but two-thirds enjoyed ‘meeting people’, three-quarters liked the ‘independence’ and ‘choosing their own working hours’, and half said they ‘enjoyed the sex’. Around a quarter had felt threatened at some point in their careers but, in contrast to popular assumptions, three-quarters felt respected by their clients (a sentiment they reciprocated) and said sex work had ‘improved their self-confidence’. Less than one per cent reported that their clients usually took control of their encounters and only three per cent wanted to give up the work in the next few months. (Suzanne Jenkins, Beyond gender: an examination of exploitation in sex work, 2009. [])




Not surprisingly, headlines about the return of the white slave trade caused alarm, especially when government spokesmen were widely quoted saying that of 80,000 working girls in Britain, ‘the majority are working under control from traffickers, pimps or brothel owners’.

Daily Telegraph, 19 November 2008.




‘The government figures are completely fabricated,’ said one sex worker, ‘they just make them up.’ That wasn’t far off the truth. Because of stigmatisation and the furtive nature of the industry, the only number available, 80,000, had been extrapolated from a small survey compiled twenty years previously by a health outreach worker tackling HIV and AIDS in Birmingham.

Hilary Kinnell, ‘Prostitutes and Their Clients in Birmingham: Action research to measure and reduce risks of HIV’, The AIDS Letter, Royal Society of Medicine 19, 1990. Hilary Kinnell went on to specialise in outreach work with prostitutes and became as knowledgeable as anyone about the subject, publishing a book prompted by the Ipswich murders, Violence and Sex Work in Britain, Willan, London, 2008.




As for the alarming idea that most of the women were trapped by violent pimps – a figure of 80 per cent was cited by a former minister

A former Home Office minister, Fiona Mactaggart, told BBC Radio 4’s Today in Parliament (21 November 2008) that ‘something like 80 per cent of women in prostitution are controlled by their drug dealer, their pimp, or their trafficker’. Many other wild claims were made including one mostly attributed Labour MP Denis McShane that there were 25,000 victims of trafficking for sexual exploitation.




Nonetheless the Met set up a dedicated Human Exploitation and Organised Crime Unit called SCD9 and police around the country set up Operation Acumen to investigate the crisis and try to tackle it. Their estimate was that 30,000 women were working as escorts or from flats and brothels, of whom almost 12,000 women were vulnerable to violence or debt-bond and of whom 2,600 were definitely trafficked.

Operation Acumen, Setting the Record, The trafficking of migrant women in the England and Wales off-street prostitution sector, ACPO, 2010, p. 21.[].

See also The UK Threat Assessment of Organised Crime 2008/9, Serious and Organised Crime Agency, London, p. 37, which found some cases where sex migrants were ‘unaware of the extent of the control and exploitation they will suffer’.




There were more proactive raids in the run-up to the Olympics but the predicted surge in victims failed to materialise – just as it had in South Africa during the 2010 World Cup, when not one of 40,000 expected sex slaves was encountered.

Andrew Boff, London Assembly Member, Silence on Violence: Improving the Safety of Women; The policing of off-street sex work and sex trafficking in London, a report to the Mayor of London, 2012. [] This must rank as one of the most useful and thoughtful surveys ever presented by a local politician and contains references, explanations and analysis on London police raids on brothels and their results.




In fact the most reliable figure we have for people brought to Britain on false pretences and exploited for sexual purposes is approximately 250 a year. This is half the number of people ‘conclusively’ recognised as ‘victims of trafficking’ by the national reporting scheme coordinated by the police. The other 250 or so recognised victims are thought to be exploited for labour or domestic servitude.

The National Referral Mechanism compiled figures from across the UK which were analysed and published quarterly by the Serious Organised Crime Agency. Just over 2,000 ‘potential’ victims were referred in 2011 and on average about a quarter of all cases receive a ‘positive conclusive decision’. Of these 500 about half are thought to be for sexual exploitation though the proportions vary over time. Intelligence Assessment, UK Human Trafficking Centre: A baseline assessment on the nature and scale of human trafficking in 2011, SOCA, 2012. See also quarterly National Referral Mechanism Statistics. []

For further reading see a well-researched essay on how the trafficking scare ‘infantilises’ women: Dr Brooke Magnanti, The Sex Myth: why everything we’re told is wrong, Weidenfeld & Nicolson, 2012. As Dr Magnanti observes, so many beliefs about women in sex work, held with ‘almost religious fervour’, are easily disproven.




It’s worth hearing at length her exasperation at the scare:

Firstly, there was a critical mass of female parliamentarians eager to be seen to be doing something for women, and who used trafficking rhetoric and inflated trafficking figures which exploited migration fears … Secondly, these were conveyed by a news media dependent on ‘client journalism’ and news agencies producing ‘churnalism’ from government press releases. Third, there were significant vested interests of politicized senior police officers who, using pseudo-scare tactics lobbied for more power and pressure groups influenced by USA prohibition research who supported the rise of ‘spin’ as an integral political tool. It was easy to spin material on sex work to a public who have little experience of or access to research material on sex work.

Linda Cusick, Hilary Kinnell, Belinda Brooks-Gordon and Rosie Campbell, ‘Wild guesses and conflated meanings? Estimating the size of the sex worker population in Britain’, UK Network of Sex Work Projects, Critical Social Policy, Vol. 29(4), 2009, pp. 703–19. []