Rape rates

Official rape rates are socially constructed and often misleading (Concha Reference Concha2023). One study of crime in major cities of the world (Zimring Reference Zimring2012, p. 45) relied on government statistics to conclude that rape rates per 100,000 population in New York (10.6), London (30.7), and Sydney (51.4) were 5–25 times higher than the rape rate in Tokyo (1.8). Similarly, a cross-national study found that Japan had the third lowest official rape rate among fifty-four countries, behind only Armenia and Azerbaijan (Concha Reference Concha2023, p. 44). According to government data compiled by the United Nations, Japan has a per capita rape rate of 1.1 per 100,000 while Sweden, England, and the United States have rates that are 30–50 times higher (Ito Reference Ito2021, p. 139). These figures echo claims by the Japanese government, but three serious problems plague them. First, different societies define rape differently. Second, rape rates can be expressed with different indicators, from the number of offenses known to or recorded by police, to the number of arrests, or prosecutions, or convictions, or incarcerations. To compare rape rates that use different indicators is to compare apples and bananas. Third and most fundamentally, government statistics reflect only a portion of all crime because they do not include unreported offenses, which are common in the case of rape but not equally common across societies (Archer and Gartner Reference Archer and Gartner1984, p. 29). For reasons explained below, the under-reporting rate for rape in Japan may well be higher than the under-reporting rate in many other democracies.

It is difficult to make sound comparisons of rape rates across jurisdictions, but when efforts are made to compare like with like, as was done in a recent study of rape rates in thirty European nations, many of the differences that appear in official figures decline or disappear (Swedish National Council for Crime Prevention 2020). Victimization surveys also facilitate cross-national comparison by addressing the problem of underreporting in official crime statistics. In these surveys, people are asked whether and when they have been victimized by crime of various kinds. This method of measuring crime rates is far from perfect because memory fails and people may be motivated to minimize or misstate their experiences. However, victimization surveys are widely considered a useful tool for addressing the problem of under-reporting. Evidence from the International Crime Victim Survey shows that Japan ranks among the safest societies in the world for the crimes of robbery, burglary, assault, and the aggregate category of “victimization by any common crime.” However, for “sexual offenses” Japan ranks far less well (Johnson Reference Johnson2018, p. 34). This finding is consistent with what is known from sources which show that more than 70 percent of Japanese women have been sexually molested (Makino Reference Makino2019, p. 81). Few of these victims report it to the police, often because they fear being embarrassed or because they believe criminal justice officials will not believe them or will not care. None of the unreported cases is included in official statistics.

Victimization surveys have also been conducted by the Research and Training Institute of Japan's Ministry of Justice. The most salient question asks about victimization in a “sex case” (sei teki jiken) within the last five years. The percentage of respondents answering “yes” has declined in each survey, from a high of 2.7 percent in 2000 to 2.5 percent in 2004, 2.0 percent in 2008, 1.3 percent in 2012, and 1.0 percent in 2019 (Research and Training Institute 2020, p. 9). By this measure, the five-year sexual victimization rate declined more than 60 percent in two decades (over the same period of time Japan's aggregate crime rate declined 70 percent). In the victimization survey for 2019, the percentage of respondents saying they had been sexually victimized in the previous year was 0.4 percent (Research and Training Institute 2020, p. 7), which means more than 500,000 persons were victimized by a sex crime that year. This is approximately fifty times more sex offenses than are recorded by Japanese police in the same period (Makino Reference Makino2019, p. 20). As for the crime of rape more narrowly, a victimization survey by Japan's Cabinet Office in 2017 found that 7.8 percent of women (one in thirteen) and 1.5 percent of men (one in sixty seven) said they had been victimized by “forcible sexual intercourse and the like” at least once in their lifetime (Ito Reference Ito2019, p. 54). A similar survey conducted in 2022 found that these figures had declined, to 6.9 percent for women and 1.0 percent for men (Osawa Reference Osawa2023, p. 30). Still, by these measures approximately five million Japanese women have been raped at least once, as have 700,000 men, and many of these victims were young at the time of the crime. An internet survey by NHK in 2022 obtained responses from 38,383 victims of sexual violence. Three-quarters of the victims were under the age of twenty at the time of the first offense, and the average age at first victimization was just fifteen years. Similarly, a national survey of youth ages 16–24 carried out by the Cabinet Office in 2022 found that one-quarter of these young people had been sexually victimized, many of them repeatedly. The effects – depression, PTSD, suicidal ideation, and more – are often serious and long-lasting (Osawa Reference Osawa2023, pp. 122–30).

Rape law and recent reforms

Rape is prohibited by Article 177 of Japan's Penal Code, but the legal definition of this crime has long been problematic, and there is also a large gap between law-on-the-books and law-in-action. At the end of the day – after laws are passed and legal principles pronounced – it is law-in-action that matters most. As students of law and society often observe, “Whoever hath an absolute authority to interpret [and enforce] any written or spoken laws, it is he who is truly the Law-giver to all intents and purposes, and not the person who first wrote or spoke them” (Hoadly Reference Hoadly1717).

Japan's law of rape was revised in 2017 and then again in 2023. The changes were driven partly by transnational social forces, including a revolution in criminal-sex law that is rooted in “the intensified global celebration of free-standing personhood” (Frank, Camp, and Boutcher Reference Frank, Camp and Boutcher2010) and foreign opinion that is critical of Japan's legal regime. Domestic forces have also pushed for reform. Most notably, Japanese media have highlighted how the country lags other nations in its treatment of sex crime and domestic violence, and Japanese civil society has demanded greater protection and support for people who are vulnerable to sexual victimization. Revisions to the Penal Code were also spurred by lamentations that Japan's statutory punishments for rape were more lenient than the punishments for robbery and by several surprising acquittals in rape trials that made reform seem urgent. All of these domestic forces of change were rooted in the voices, anger, and activism of people who have been harmed by sex crime and by people who are dismayed by the impunity that sex offenders frequently enjoy.

The 2017 reform made four main changes to Japan's Penal Code. First, the definition of rape was broadened, from a law that said only forcible male–female genital intercourse could be rape, to one which forbids forcible genital, anal, or oral intercourse between male and female or male and male – but even after this reform coercive female–female sex was not considered “rape.” Second, the minimum punishment for rape was raised from three years imprisonment to five years. Third, the requirement of a victim's complaint (kokuso) for the criminal prosecution of rape was dropped, which makes it possible for prosecutors to charge rape even when the victim does not complain. Finally, new crimes were created to treat sexual conduct as criminal when it is done by a legal guardian or by some other adult in a position of authority.

The changes in 2017 were the first sex crime reforms in Japan's Penal Code since 1907, but the revised law remained narrow in scope and problematic in content in two main ways. First, to convict someone of rape, prosecutors still had to prove that “violence or intimidation” was involved, and according to interpretations by Japan's Supreme Court, the violence or intimidation had to be such that it was “remarkably difficult” for the victim to resist (Ito Reference Ito2019, p. 27). In this way, the 2017 reform fell short of international standards by failing to focus on the victim's lack of consent (Goto Reference Goto2018). Second, to convict someone of quasi-rape, which is the sexual assault of a person who is unconscious or otherwise unable to resist, prosecutors still had to prove the victim was “incapable of resistance.” With this requirement still in place, the revised Penal Code stated that if a person had sex with a victim who was intoxicated or who was in a mental state that made it difficult to refuse a sexual advance, the act should be considered quasi-rape, not rape. In other words, as a matter of Japanese law, sex with a person who could not consent may be “partly” or “almost” rape but not “real rape” (Estrich Reference Estrich1988; Johnson Reference Johnson2022, p. 168).

Thus, until the Penal Code was reformed for a second time in 2023, Japan had no consent-based standard for rape. In its absence, prosecutors and judges had to discern whether violence and intimidation were “remarkably difficult” for the victim to resist. This force-based regime contradicted what is known about sex offenders (few of whom use weapons or outright violence) and their victims (only 20 percent of whom say they resisted) (Osawa Reference Osawa2023, p. 143). Japan's revised law also stood in sharp contrast to the United States, where “the modern trend is to focus on non-consent and to downplay the element of force” (West Reference West2011, p. 113), and it differed from the law of rape in other nations as well. Attorney Ito Kazuko (Reference Ito2019, pp. 107–26) compared Japan to seven other countries (Canada, Germany, England & Wales, Sweden, Finland, South Korea, and Taiwan) and concluded that criminal law in Japan is the least sensitive to issues of consent. The most similar society was South Korea, where prosecutors have to prove that force or threat of force was present when the sex act was committed, but where the #MeToo movement is significantly stronger than in Japan, which has “led to accusations of sexual abuse against an array of prominent [Korean] men,” some of whom were convicted of a sex offense and sentenced to lengthy terms of imprisonment (Yoon Reference Yoon2023). Another sign of South Korea's more aggressive approach to prosecuting rape is the conviction rate in sex crime cases, which is around 80 percent there compared to 99 percent in Japan. Prosecutors in South Korea are charging rape even when there is a significant chance of acquittal, but this seldom happens in Japan.

Sexual desire is often perceived as “unfeminine” in Japan.Footnote ⁴ In the shadow of this social norm, “no” might mean “yes,” and “only physical displays of refusal suffice to usurp a man's belief in a woman's agreement to sex” (Kemp Reference Kemp2020). These attitudes are not exclusive to men, for many women believe it is their duty to ward off male advances (Makino Reference Makino2019). Dubious folk beliefs and “rape myths” also shape perceptions of when sex is rape. One holds that a man should not be prosecuted for a sex crime if the woman's back was not touching a wall when he first made contact, because she could have (and should have) been able to retreat. Another holds that a crime cannot occur if the man places a handkerchief under the woman, because her cooperation must have been necessary for him to lay it down (West Reference West2011, p. 113). In these ways, fallacious reasoning about consent is common in Japan (Mukai and Watamura Reference Mukai and Watamura2022), as illustrated by a survey conducted by NHK in 2017 (Ito Reference Ito2019, p. 94). The prompt was “Things That Lead You to Think the Other Person Consents to Sex,” and the levels of agreement are in parentheses:

• – Eating together, just the two of you (11 percent).

• – Wearing revealing clothing (23 percent).

• – Getting in a car, just the two of you (25 percent).

• – Drinking together, just the two of you (27 percent).

• – Being drunk (35 percent).

There are also cultural pressures for victims of sex crime not to report their victimization (Dussich Reference Dussich2001). Many feel shame, embarrassment, and concern about their own reputation and that of their family (Yamawaki Reference Yamawaki2009). When victims do tell family or friends, the response often minimizes or excuses the offense or blames the victim for inviting or encouraging the offender's behavior (Osawa Reference Osawa2023, p. 151). For these reasons, a victim's decision not to report can reflect a reasoned cost–benefit analysis about what good, and what harm, will come from reporting their experience. Rape myths operate here too, by positing that rape was the victim's fault, that women “asked for it” by the way they dressed or drank, that real rape can only be committed by a stranger, and that if the victim did not physically resist then it could not have been rape (Yamawaki and Tschanz Reference Yamawaki and Tschanz2005). One study found that while many Japanese college students say they believe rape is a serious problem in their society, they also doubt rape claims significantly more often than college students in the United States (Stephens et al. Reference Stephens, Kamimura, Yamawaki, Bhattacharya, Mo, Birkholz, Makomenaw and Olson2016).

The tendency not to believe victims reflects a “credibility complex” – a cluster of patriarchal norms that shape judgments about rape by pressuring people to discount the credibility of (mostly female) accusers and to inflate the credibility of the (mostly male) accused (Tuerkheimer Reference Tuerkheimer2021, p. 3). Culture and law are the main drivers of this credibility complex, and they interact to produce large “credibility discounts” for many of the claims made by victims, about what the truth is, whether the other party's conduct is blameworthy, and whether the incident is really worthy of concern (Tuerkheimer Reference Tuerkheimer2021, p. 10). Since the vast majority of victims of rape are women and girls, this tendency to view victims as “untrustworthy threats to the good names of good men” can be called “himpathy” (Brodsky Reference Brodsky2021, p. 6).

In 2019, not-guilty verdicts in four rape trials led to protests in Japan and ridicule abroad over judicial decisions which concluded that sex is not criminal if the victim did not resist “enough.” In a case in Fukuoka the court found the male defendant not guilty of quasi-rape (incapacitated rape) after having sex with a woman who drank until she lost consciousness. The court said the woman was unable to resist, but it concluded that because she had not said anything the man mistakenly believed she had consented. In a lay judge trial in Shizuoka, the male defendant was accused of forcing a woman he had never met to give him oral sex and of injuring her in the process. The mixed panel of three professional judges and six citizens found him “not guilty” because the woman did not sufficiently resist and because the man said he had not deliberately assaulted her. In another case in Shizuoka, a 14-year-old girl testified that she was repeatedly raped by her father over two years period, but the court concluded that her statements were unreliable and acquitted him of that charge, though the court did find the father guilty of possessing child pornography and fined him 100,000 yen ($1,000). Finally, in the Okazaki branch court of Nagoya, three judges found that a father who had been having sex with his 19-year-old daughter since she was in the second year of junior high had sexually abused her, but they acquitted him of rape. In the court's view, it may have been difficult for the daughter to resist, but “it could not be concluded that she was so afraid she could not refuse.” Here again, the victim did not resist “enough.”Footnote ⁵

The retrograde decisions in these rape trials reveal how patriarchal values are deeply encoded in Japanese law and culture. The judicial decisions also provoked a backlash. Ultimately, three of the four acquittals were reversed after prosecutors appealed, and public dismay about impunity for sex offenders helped mobilize a social movement that led to the second round of Penal Code reform. This time, the age of sexual consent was raised from thirteen to sixteen. The statute of limitations for criminal prosecution was extended from ten years to fifteen years for rape and from fifteen years to twenty for indecent assault resulting in injury. The sexual assault of an unconscious person was finally recognized as real rape. Most importantly, the crime of rape was redefined from “forcible sexual intercourse” (kyosei seiko) to “non-consensual sexual intercourse” (fudoi seikozai). Following this revision in 2023, Japan no longer requires violence or coercion for rape to be proven (Keiho 2023). Sexual intercourse without consent is now considered rape, and prosecutors can prove rape by showing that it was “difficult for the victim to form, express, or fulfill the intention not to consent.” The new law lists eight ways in which this may happen, such as when a victim has a mental or physical disorder or is under the influence of alcohol or drugs, or when an offender's social or economic status causes the victim to fear the consequences of saying “no.” This complicated approach to consent is hardly a model of clarity (Asahi Shimbun/Asia and Japan Watch 2022a), but it is a significant improvement over the force-based law that prevailed until 2023.

Time will tell whether the legal reforms of 2023 result in more justice for victims and less impunity for sex offenders. Research about law and society is replete with accounts of large gaps between what the law says and what it does in actual practice (Calavita Reference Calavita2016, pp. 109–34), and police and prosecutors in Japan have coopted, deflected, and defeated many previous attempts to control their discretion (Johnson and Vanoverbeke Reference Johnson and Vanoverbeke2020; Suo Reference Suo2015). Still, events in other countries suggest there may be room for optimism. In 2018, Sweden changed its legal definition of rape to “sex without consent,” and prosecutors no longer had to prove the use or threat of violence or coercion. In the next two years, the country's conviction rate for rape rose 75 percent (Batha Reference Batha2020). More generally, a review of rape law revisions in forty countries from 1945 to 2005 concluded that legal reform is “strongly associated with elevated police reporting,” though the strength of this association depends on the domestic context (Frank, Hardinge, and Wosick-Correa Reference Frank, Hardinge and Wosick-Correa2009).

Victim reporting and police recording

Three contingencies shape a rape victim's decision to go to the police. First, a victim must name the experience as problematic (“I was raped”), then the victim must blame someone for it (“he did it”), and then the victim must make a claim about it (“he should be punished for hurting me”). There is nothing automatic or inevitable about this “naming, blaming, and claiming” process (Felstiner, Abel, and Sarat Reference Felstiner, Abel and Sarat1980–81). Many rapes are never perceived by the victim or are only perceived long after the fact (Spring-Voice 2020), and the lack of good sex education in Japanese families and schools helps explain this failure to “name” (Kuwahara Reference Kuwahara2019).Footnote ⁶ Moreover, many of the acts that victims do deem “rape” are not transformed into legal cases because there is no blaming and claiming. In these ways, labeling a sex act as “rape” depends on the actions of victims and police, and social and legal norms strongly shape their decisions (Concha Reference Concha2023).

Japan has low levels of reporting for rape and other sex crimes. The numbers vary from study to study, but they all point in the same direction: the vast majority of victims never report it to police. In 2018 the National Police Agency's “White Paper on Crime Victims” stated that only 3.7 percent of victims of sex crime contacted the police (Hanzai Higaisha Hakusho 2018), and a 2020 survey by Japan's Cabinet Office (2020) found that only 5.6 percent of victims did so. In several other accounts, less than 5 percent of incidents of sexual violence were reported to police (Asahi Shimbun/Asia and Japan Watch 2019; Human Rights Watch 2018; Kemp Reference Kemp2020; Oppenheim Reference Oppenheim2019). NHK's survey in 2022 found that 10 percent of the victims of sexual violence reported to police – but police did not record (juri) 29 percent of those cases, and another 34 percent were never sent (soken) to prosecutors (Osawa Reference Osawa2023, pp. 153–55). Thus, the NHK data indicate that prosecutors received a little more than one-third (37 percent) of all reported cases. Overall, the evidence from Japan suggests that only 5–10 percent of victims of sex crime report it to police. By this estimate, for each sexual offense that is reported, 10–20 are not.Footnote ⁷

The evidence also shows that rape by a stranger in Japan is much more likely to be reported to police than rape by a friend, acquaintance, or family member. In a Cabinet Office survey in 2020, half of the victims who said they reported to police were raped by a stranger even though only 12 percent or so of all rapes in Japan are committed by strangers (Johnson Reference Johnson2022, p. 164).Footnote ⁸ In Japan as in many nations, the vast majority of rapes are between people who know each other, and these “acquaintance rapes” are especially unlikely to be reported.Footnote ⁹ A large share of acquaintance rapes also involve victims who are minors or young adults. For many of them, it is emotionally and practically difficult to make a formal complaint about an older person who has influence over their life.

The propensity of victims to report rape is shaped by various social forces, including the “rape myths” (explained above) which lead many victims to conclude that it is in their own interest not to go to the police (Osawa Reference Osawa2023, p. 151; Research and Training Institute 2020, p. 179). The level of public trust in police also influences victim reporting. Some studies have argued that Japanese police enjoy great public trust (Ames Reference Ames1981; Bayley Reference Bayley1991), and police in Japan say they solve the vast majority of sex crimes. In one recent year police said they solved 97 percent of rape cases and 91 percent of forcible indecency crimes (White Paper on Crime 2021). These “clearance rates” for sex offenses are much higher than the clearance rates in many European nations (Swedish National Council for Crime Prevention 2020), but they are partly a function of how few victims report to police and how few of the reported cases are actually recorded. Police often use this “unfounding” decision to erase cases in which they are convinced a crime occurred but believe the likelihood of arrest and prosecution is low. A survey by Spring-Voice (2020, p. 25) found the number of rape cases recorded by police to be half the number of cases reported to them by victims, while NHK's survey found the proportion recorded by police and sent to prosecutors to be closer to one-third (Osawa Reference Osawa2023, pp. 153–55). By these measures, police deem unreliable 50 percent or more of reported sex crimes, often by concluding that the victim's account is not sufficiently credible.Footnote ¹⁰ Most of these decisions are discretionary, and police often pressure victims to acquiesce to their judgment (Ito Reference Ito2017; Makino Reference Makino2019). The low levels of victim reporting and police recording also keep Japan's clearance rate (cases solved/cases known) artificially high by suppressing the number of rapes in the denominator. Moreover, public trust in Japanese police is actually quite low (Cao and Stack Reference Cao and Stack2005; Tsushima and Hamai Reference Tsushima and Hamai2015), and this shapes victims' perceptions of what good and harm will come from filing a rape complaint. One common harm is an investigation process that re-victimizes victims of sex crime by making them reenact their experience. This problem would be mitigated if Japanese police were better trained and less patriarchal, but only 8 percent of police in Japan are women. In the United States, where 11 percent of police are women, research shows that female police respond more effectively to violence against women than their male colleagues do (Miller and Armstrong Reference Miller and Armstrong2018, p. 92).

In sum, Japanese police are not reliable allies of the victims of rape or other sex crimes, even when the victim is a police officer (Makino Reference Makino2019, pp. 234–47). In many cases police do not even deploy “coercive investigation” (kyosei sosa) techniques, such as arrest and detention, and their interrogations of criminal suspects are often shorter and less intense than for other serious offenses. Confessions have long been the cornerstone of Japanese criminal justice, and interrogation is the main means of making suspects talk (Johnson Reference Johnson2002). But in rape cases, when male suspects are given the benefit of the doubt and female victims are given a credibility discount, several pernicious consequences follow. Truth is not clarified. Rape statistics are distorted and a false sense of security is encouraged. The needs of victims are not met, and many of them are re-victimized. Sex offenders are not held accountable – and they may be emboldened to offend again.Footnote ¹¹ Public safety is sacrificed. And justice is not done. In this context, and when naming-blaming-and-claiming might result in blowback from family, friends, and co-workers, a victim's decision to “lump it” often feels like the least bad of a bunch of bad options.

Prosecution

We have seen that much caseload attrition occurs when victims of rape decide not to report and police refuse to record and investigate. What about prosecution?

Prosecutors in Japan tend to be cautious about charging cases, and this is the main reason for a conviction rate that exceeds 99 percent (Johnson Reference Johnson2002, pp. 214–42). Their cautious charging policy brings significant benefits, including a more careful approach to using the criminal sanction than occurs in other nations. But it also produces problems. Most notably, many victims feel betrayed when their offender is not charged, and there is impunity for many serious offenders. These problems are serious in sex crime cases.

Most rapes occur in places where a third party is not present, and prosecutors often regard the scene of the crime as a “black box” which conceals evidence about whether the woman was willing (Ito Reference Ito2017). Concerns about victim credibility also shape prosecutorial decision-making. Some of these challenges could be overcome through the effective use of physical evidence, but a recent review found that rape kits were available in less than one-third of Japan's forty-seven prefectures (Kemp Reference Kemp2020), and Japanese law enforcement relies heavily on confessions for proof (Johnson Reference Johnson2002, p. 243). The result is that prosecutors charge a little more than one-third of the rape cases recorded by police, while their charge rate for forcible indecency is slightly higher (Ibusuki, Reference Ibusukiforthcoming).

Two other facts help explain why many sex crimes are not prosecuted in Japan. First, some victims withdraw their complaint before a case can be charged because of financial payments (jidan) they receive from the accused. In 2014, 27 percent of all formal rape complaints were withdrawn in this way, as were nearly 57 percent of forcible indecency complaints (Ibusuki, Reference Ibusukiforthcoming). These settlements are not well studied, but their frequency suggests that money can buy leniency for sex offenders, both at the charging stage and at sentencing. In a gang rape case involving eight co-offenders, compensation of 1.5 million yen (about $12,500) reduced the prison sentence to three years for each of the six offenders who paid this amount, compared with a four-year sentence imposed on two offenders who did not or could not pay (Johnson Reference Johnson2002, pp. 201–10). From the prosecutor's point of view, this kind of payment is a way for the offender to demonstrate remorse to the victim and deference to the decision-makers who hold his fate in their hands.

A second hallmark of the non-prosecution of sex crime is the claim commonly made by prosecutors that there is “insufficient evidence” (kengi fujubun) to charge and convict. And for them the main issue is often consent. In recent years, “insufficient evidence” was the reason given for not charging 44–52 percent of all uncharged rape and forcible indecency cases, which is more than double the rate of “insufficient evidence” dispositions that prosecutors reached in other Penal Code cases that did not get charged (Ibusuki, Reference Ibusukiforthcoming). At the same time, prosecutors are less likely to “suspend prosecution” (kiso yuyo) in sex crime cases than for other serious crimes. To “suspend prosecution” is to declare that although there is sufficient evidence to convict, charges will not be filed for practical or policy reasons, as when an indictment will not foster rehabilitation, promote public safety, or serve the interests of justice. In effect, a suspended prosecution is a discretionary decision that not charging an offender is better than seeking punishment.

Japan's approach to prosecuting rape has become substantially more cautious in recent years, with the charge rate for rape cases declining from 65.8 percent in 2005 to 32.7 percent in 2017 (Ito Reference Ito2019, p. 59). In other words, prosecutors in 2017 were half as likely to charge a rape case as they were 12 years earlier (charge rates for other crimes have also declined substantially). Prosecutors' caution can also be seen in a ratio that juxtaposes the two main reasons for non-prosecution: “suspended prosecution” versus “insufficient evidence.” For all Penal Code offenses (theft, burglary, robbery, and the like) this ratio is ~3.5–1, while for rape it is ~1–5, and for forcible indecency ~1–4. These large differences – claims of “insufficient evidence” are 14–18 times more common in sex crime cases than in other cases – reflect three truths about prosecutor culture. First, questions of consent were important to prosecutors even before a consent-based definition of rape was legislated in 2023. Second, prosecutors do not want to acknowledge that many of their decisions to withhold charges in sex crime cases are discretionary (this is why they avoid the “suspended prosecution” category). Third, the purported lack of evidence in sex crime cases often occurs because prosecutors and police have not seriously investigated (Ito Reference Ito2017, Reference Ito2019; Makino Reference Makino2019). Rape myths and the credibility complex drive these problematic practices by encouraging prosecutors and police to doubt accusers and protect abusers.

Judges influence prosecutorial decision-making through the decisions they make about trial procedures, admissible evidence, guilt, and sentencing (Burns Reference Burns2005). And it is judges who define the meaning of “beyond a reasonable doubt,” thereby establishing the threshold of proof required for conviction, and it is judges who, by acquitting defendants, prompt prosecutors to shrink into a protective crouch when they are deciding whether to charge. As one prosecutor observed, “Losing a case is a career killer in the eyes of many of my colleagues” and acquittals can produce “a chilling effect on what gets prosecuted” (quoted in Yamamoto and Adelstein Reference Yamamoto and Adelstein2019). It is also judges who ruled that the “violence or intimidation” in a rape case must be sufficient to make it “remarkably difficult” for the victim to resist. Until the Penal Code reform of 2023, prosecutors who ignored this judicial imperative did so at their peril.

Japanese prosecutors could be called “cautious” or “timid” or “sexist” or “patriarchal” in their approach to charging sex crimes. Whatever the characterization, one result of their charging policy is a very high conviction rate. From 2015 to 2019 the charge rate for rape cases received from police varied between 33 and 39 percent, and the conviction rate for rape in this five-year period was 99.2 percent (just 13 acquittals in 1,577 charged cases). As these numbers show, prosecutors in Japan charge rape only if a conviction is all but certain. There is nothing necessary or inevitable about this approach to prosecution. It could be different, and many people in Japan want it to be different because they believe the present practice sacrifices too many victims, too much deterrence, and too much justice (Johnson Reference Johnson2022).

Japan's high conviction rate would decline if prosecutors adopted a more aggressive charging policy, but such a change could also produce more wrongful convictions. In the United States and United Kingdom it is estimated that 2–8 percent of rape complaints are false (Avalos Reference Avalos2016; Miller and Armstrong Reference Miller and Armstrong2018, p. 38). The rate of false complaints in Japan is unknown, but similar issues occur there too. So we need to ask: what balance should be struck between crime control values and due process values in the prosecution of sex crime, and what is an acceptable risk of wrongful conviction? These questions are explored in Suo Masayuki's hit film, “Sore demo Boku wa Yatte Nai” (“Even So I Didn't Do It”), which was released in 2007. Based on a true story, the film depicts an apparently innocent man who was charged with groping a woman on a crowded train. In the film as in real life, the man was convicted and sentenced to prison. Suo's film has done much to raise awareness of problems in Japanese criminal justice, but when I interviewed him in 2017 he said that while he is proud of this work, he also wonders whether it has harmed the cause of gender justice in Japan by focusing on the serious but uncommon problem of wrongful accusations. Similar concerns have been raised about well-intentioned efforts by Japanese lawyers to prevent miscarriages of justice in sexual molestation cases (Chikan Enzai Higaisha Nettowaaku 2002). Some Japanese companies even sell insurance that will pay for legal representation if a person is arrested for groping. One product's premium is 590 yen per month (Chikan Enzai Herupu Koru tsuki Bengoshi Hiyo Hoken 2022). Even critics of the weak enforcement of anti-groping laws in Japan acknowledge that wrongful accusations are a serious problem, but they also fear that the movement to prevent wrongful convictions may be reinforcing “a culture of rape” (Makino Reference Makino2019, pp. 161–208).

Punishment

Research shows that deterring crime depends on the severity, speed, and certainty of punishment. Of these three factors, certainty is the most important (Kleiman Reference Kleiman2009, p. 3). There is no magic number, but some studies suggest that “crime increases in places that have lower than a 30 percent arrest rate, and it decreases in places with arrest rates higher than 30 percent” (Van Rooij and Fine Reference Van Rooij and Fine2021, p. 32). As we have seen, this threshold for effective deterrence is far higher than the arrest rate for rape in Japan. We have also seen that Japan has a 99 percent conviction rate for rape when prosecutors charge, yet 98–99 percent of rape offenses do not lead to conviction.Footnote ¹² When people who have raped are charged and convicted, how are they punished?

According to a White Paper on Crime which combined the crimes of rape (kyosei seiko) and forcible indecency (kyosei waisetsu) in its summary statistics, 1,184 of these sex offenders were convicted in 2020, and more than half (55.3 percent) received a suspended sentence (shikko yuyo), which means they were not incarcerated after conviction (Hanzai Hakusho 2021, p. 41). The remaining convicts (44.7 percent) were sentenced to prison (jikkei). By comparison, only 25.2 percent of convicted robbers (goto) received a suspended sentence, as did 48.2 percent of convicted thieves (setto). Thus, people convicted of rape or forcible indecency were treated much more leniently than people convicted of robbery (another serious crime involving violence or intimidation), and they were even treated a little more leniently than people convicted of theft, which is not a violent crime. Among the 45 percent of convicted sex offenders who were sentenced to prison, the most common sentencing ranges were 3–5 years (14.0 percent) and 2–3 years (11.2 percent). Only one in seven was sentenced to a term of imprisonment exceeding five years, and less than one in fifty was sentenced to ten years or more.

How do Japan's sentencing patterns for sex crime compare with sentencing patterns in other countries? Apples-to-apples data are hard to find, but two comparisons can be made. First, consider the average time served in prison for rape in some developed democracies for 1980–1999: USA (60 months), Australia (51 months), Scotland (36 months), England and Wales (34 months), Switzerland (25 months), Netherlands (16 months), and Sweden (15 months). In this study, the European average is 30 months and the American average is double that (Latzer Reference Latzer2022, p. 113). Both of these figures are significantly longer than the average sentence for convicted sex offenders in Japan (25 months) even though the figures for other countries are “time served,” not sentence imposed. The sentencing averages for Europe and America also refer to a period of time when sexual assaults were punished more leniently than they are today.

The second comparison comes from a recent study that calculated the average sentence length for sexual assault convictions in several countries, including the United States (12.2 years), England and Wales (5.2 years), Belgium (4.4 years), France (3.9 years), and Finland (2.9 years). In this research, the only country with a lower average sentence length than Japan (2.1 years) was the Netherlands (0.9 years), and in some nations (including the United States and Australia) the average time served for sexual assault was greater than the average sentence imposed in Japan (Kazemian Reference Kazemian2022, p. 19).

The data are far from perfect, but even after great caseload attrition – a process that should be selecting the “worst of the worst” cases for criminal punishment – sentencing for convicted sex offenders in Japan appears to be very lenient (Ito Reference Ito2019). The Japanese public seems aware of this fact even while they doubt the claims made by many victims of sex crime. A survey in 2019 found that 86 percent of adults believed “Japan is too forgiving of sex offenders,” and there was little difference between men and women respondents, with more than 80 percent of both saying Japan should be tougher on sex crime (Baseel Reference Baseel2019).Footnote ¹³

One other telling fact helps reveal how patriarchal norms shape perceptions of sex crime in Japanese law and society. To make criminal justice more democratically responsive to citizens' sensibilities, Japan established a lay judge system (saibanin seido) in 2009. This new trial system is the most consequential reform in Japanese criminal justice since the Occupation (Johnson and Vanoverbeke Reference Johnson and Vanoverbeke2020). Now, six ordinary citizens sit with three professional judges to decide guilt and impose sentence in some serious cases (about 3 percent of criminal cases are tried by lay judge panels). But strikingly, four serious sex crimes – rape, quasi-rape, forcible indecency, and quasi-forcible indecency – remain ineligible for trial by lay judges. Indeed, when the Lay Judge Law was being debated and drafted, these crimes were not even seriously considered for inclusion in the list of eligible offenses. Instead, only “rape at the scene of a robbery,” “rape causing death or injury,” and “forcible indecency causing death or injury” can be adjudicated by lay judges. In these cases, lay judges have been challenging the going rates for the sentencing of sex offenders that prevailed under the system of adjudication by professional judges, often by nudging punishment towards more severity, and often against significant judicial resistance (Johnson and Vanoverbeke Reference Johnson and Vanoverbeke2020). However, the 124 sex crime cases tried by lay judge panels in 2020 was less than 10 percent of all rape and forcible indecency cases (n = 1,308) adjudicated in that year. The vast majority of Japan's most serious sex crimes have been deemed insufficiently important to warrant lay participation in adjudication.