Gender and crime in EuroAmerican research

By the twentieth century, social science methods were increasingly popular across the globe. Japan, Argentina, and India had growing numbers of trained social scientists studying crime (Anderson Reference Anderson2004; Nakatani Reference Nakatani, Becker and Wetzell2006; Salvatore Reference Salvatore, Becker and Wetzell2006); and some in Japan and the Soviet Union even focused explicitly on women (Marran Reference Marran2007; Kowalsky Reference Kowalsky2009). Likewise, Zhou Shuzhao and research team Liu and Xu developed their own transnational translations of social science methodologies to analyze women’s crime in early twentieth-century China. In her study, Zhou (Reference Zhou1932) cited individual researchers Sidney D. Gamble, William Healy, and William I. Thomas, as well as collaborative works by Mabel Ruth Fernald, Mary Holmes Steven Hayes, and Almena Dawley; and married couple and research team Sheldon and Eleanor Glueck. Liu and Xu’s study cited Henry H. Goddard, Cesare Lombroso, Maurice Parmelee, and Edwin H. Sutherland. It is not always obvious how or why they cited these particular thinkers and elided others. However, Zhou, Liu, and Xu’s citational web reflected their familiarity with contemporary global social science. It also situated them broadly within a transnational social science community and specifically within the group of global researchers who focused on women’s crime. The following section examines the specific question of women’s crime in works by European and U.S. scholars that Zhou, Liu and Xu cited with particular attention to Fernald et al., whose women-focused study Zhou specifically praised. It argues that Zhou, Liu and Xu strategically translated – and sometimes refuted – an assortment of methods, statistics, and case studies to craft their novel argument that Chinese women’s crime was fundamentally a socioeconomic issue.

Many scholars (Horn Reference Horn2015; Hahn Rafter et al. Reference Hahn Rafter, Posick and Rocque2016) trace the scientific study of crime to the discipline of criminal anthropology primarily developed by Italian criminologist Cesare Lombroso, whose theory of the “born criminal” was referenced by both Zhou, Liu and Xu. ^{Footnote 1} Lombroso was one of the first to examine crime through scientific methods, and though he acknowledged some social and cultural factors (Scheider Reference Scheider2010), he largely gained international acclaim for his biological crime theories, especially in the U.S. (G. Lombroso and Lombroso Reference Lombroso and Lombroso1911: xix, 5; Hahn Rafter et al. Reference Hahn Rafter, Posick and Rocque2016: 70). On women specifically, Lombroso’s La Donna Delinquente: La Prostituta e la Donna Normale (1893; hereafter Criminal Woman) argued that most women were less prone to crime than men, but pathologized those who committed crimes as masculine and outside of “normal” womanhood. Most women, he claimed, were menstruating at the time they committed their crimes. In deviant women, such as prostitutes, he wrote that menstruation was “always abnormally prolonged, repeated, or abundant” (C. Lombroso and Ferrero Reference Lombroso and Ferrero2003 [1893]: 160). His central argument was that biological, physiological, and mental characteristics in women influenced criminal behavior. By the early twentieth century, many social scientists read this not only as biologism, but as arguing for women’s inferiority, and promptly began to denounce his work (Brown Reference Brown, Gelsthorpe and Morse1990; Woods Reference Woods, Panfil and Peterson2014: 141–43). Instead, research on the impact of “social learning experience” (Jones Reference Jones1986: 12–13) in the Chicago school of sociology became increasingly influential (Bulmer Reference Bulmer1986), and social explanations for crime rose to prominence.

This shift, however, did not represent a distinct break from social scientists’ fixation on biology (Rose and Abi-Rached Reference Rose and Abi-Rached2013; Horn Reference Horn2015), particularly for the study of women’s crime. William Healy, William I. Thomas, and Maurice Parmelee were such examples: All three were cited in Zhou, Liu, and Xu’s studies and overwhelmingly focused on women’s biology and physiology. Criminologist and child psychologist William Healy – affiliated with Harvard and the University of Chicago during his long career – included many case studies of women that focused on their particular sex characteristics, especially menstruation and pregnancy in The Individual Delinquent (Healy Reference Healy1915). For criminal women, Healy argued, “there is excessive tendency towards misconduct and irrational behavior at menstrual periods” as well as when women were premenstrual or pregnant (Healy Reference Healy1915: 627–34). Like Healy, William I. Thomas – also affiliated with Chicago – was suspicious of women and their sexuality. His work The Unadjusted Girl (Thomas Reference Thomas1923) examined three thousand summaries of the behavior of deviant girls to argue that antisocial behavior developed in girlhood: Girls, Thomas wrote, have “an impulse to get amusement, adventure, pretty clothes, favorable notice, distinction [and] freedom…” (Thomas Reference Thomas1923: 109). Sex was a tool to secure these desires, one that poor girls were especially likely to use for their amusement and survival (Thomas Reference Thomas1923: 151). Maurice Parmelee – trained at Columbia University – argued in his 1918 book Criminology that women were hysterical, physically weak, passive, and surpassed men in their “deceitfulness, lying, hypocrisy, malicious gossip, backbiting, slander, nagging…and a weaker sense of social solidarity and justice” (Parmelee Reference Parmelee1918: 246). Popular opinion suggested that women were underrepresented in the overall prison population because they were more moral than men, Parmelee wrote, but he disagreed; he argued that women were normally confined to the domestic sphere and thus most of their crimes simply went unnoticed (Parmelee Reference Parmelee1918: 231–48). Healy, Thomas, and Parmelee reflected a trend in the study of women’s crime: Though all three believed that social factors shaped crime, for women, they continued to focus on biological, physiological, and psychological factors.

In contrast, women researchers Mabel Ruth Fernald, Mary Holmes Steven Hayes, and Almena Dawley presented a study of original statistical data on women inmates that challenged such biological theories. All were early pioneers of psychology in social work and reformatories and shared similar academic backgrounds to Healy, Thomas, and Parmelee: Fernald obtained her doctorate at the University of Chicago in 1910 studying under psychologist James Rowland Angell; Hayes also received her Ph.D. in Psychology there in 1910; and Dawley her M.A. in 1915 (Springfield Republican 1907; Willrich Reference Willrich1998:67–68n10; Koppes Reference Koppes1997: 502; Walsh and Savickas Reference Walsh and Savickas2005: 28). All three women also lived and worked in Bedford Hills, New York; Fernald worked as a psychologist at the Bedford Hills Reformatory, while Hayes and Dawley held positions at the Rockefeller-funded Bureau of Social Hygiene (New York Times 1914; New York Times 1962; Koppes Reference Koppes1997: 502), where William I. Thomas also led several projects on crime (Thomas Reference Thomas1928). Through these local connections, they undertook a massive study of women inmates across the state of New York: A Study of Women Delinquents in New York State (1920). They worked under the guidance of Katharine Bement Davis, a notable reformer of the Progressive era who headed a number of women’s institutions and conducted research on women’s criminal psychology and sexuality (Freedman Reference Freedman1984: 116; Fitzpatrick Reference Fitzpatrick1994: 92–94; Deegan Reference Deegan2003; Garton Reference Garton2010: 85–86). In their study, Fernald et al. (Reference Fernald, Hayes and Dawley1920) developed a moderate analysis of women’s crime that was “neither biological nor psychological nor economic” (Freedman Reference Freedman1984: 120) largely based on statistics and case studies.

Fernald et al. (Reference Fernald, Hayes and Dawley1920) collected data from 1915 to 1917 on the backgrounds and experiences of 587 women incarcerated in New York state reformatories, prisons, penitentiaries, workhouses, and religious institutions, as well as a group of women on probation. Although they claimed that it was not their goal to “defend any specific thesis or to combat established ideas regarding the characteristics” of their subjects, many of their core claims centered on refuting the claims of prominent criminologists, including Cesare Lombroso, William Healy, and Sheldon Glueck (Fernald et al. Reference Fernald, Hayes and Dawley1920: 5). In particular, Fernald et al. were deeply critical of Lombroso, whose claims “made no pretense of being scientific” and whose methods were little but “ipse dixit” and “crude comparison[s]” (ibid.: 246). The question of whether or not criminal women constituted a different “type,” a framework championed by Lombroso, also motivated their study. Fernald et al. rejected the notion of a Lombrosian “criminal type,” writing that “any search for a well-defined type of individual, appearing as the delinquent woman [italics in the original], will probably be fruitless” (ibid.: 528). ^{Footnote 2} However, Fernald et al. were still convinced that criminal women – and especially sex offenders or those with “sex irregularities” in their past – possessed “a somewhat inferior mentality” (ibid.: 525), drawing a slight but notable correlation between mental inferiority and crime.

Fernald et al. also conducted interviews to ascertain details such as each subject’s family life and upbringing, including moral standards, parental supervision, and economic status of the home; their race, age, religion, native language, and mental capacity; their marital and immigration status; their education, work, health, substance (alcohol and tobacco) habits, and sex history, including a pelvic exam and pap smear. They also gave IQ tests (Binet, Stanford-Binet, Woolley) and tests on areas of memory, problem solving, association, math, spelling, writing, reading, history, geography, and practical knowledge; several of which were created by or based on studies by Healy (ibid.: 43–46). In order to collect this information, they designed a regimented and strict, as well as deeply invasive, methodology: The subjects were held in solitary confinement for 2 weeks and allowed no visitors, then given a 10-day series of tests and interviews. This “quarantine period” was designed to avoid subjects coaching one other on test questions. It also had the benefit of making their subjects more compliant to testing “in preference to the tedium of sitting idle in her room” (ibid.: 40).

Fernald et al.’s attitudes towards their subjects were frequently dismissive and disdainful. They described the women participants as uncooperative, bemoaning their “grouchiness” and “obstinacy” when asked to undergo testing (ibid.: 40). Comparing them to disgruntled guests who disliked changes at their “favorite hotel,” they argued that their subjects were simply resentful of any kind of “innovation,” such as their scientific study (ibid.: 23). Some subjects, they noted, would either withhold information or give false information due to their resentful attitude, particularly older women, whom the researchers deemed “cannily” clever for refusing to participate because they deduced that the less that was known about them, the better (ibid.: 47, 206). In rare moments such as these, the researchers seemed to acknowledge their own subjects’ awareness of the intense scrutiny and pressures of their study.

Many of Fernald et al.’s conclusions rejected the biological claims of their male colleagues who studied crime, but also explanations of crime as wholly social or environmental. They found no evidence that criminals had “physical stigmata” on the body, as criminal anthropologists claimed (ibid.: 305). Although they believed that their subjects possessed low intelligence and mental capacity generally, they argued that there was “an extensive amount of overlapping” with the general population, and thus, lower intelligence was not synonymous with criminality. They took a strong stance against eugenicist trends, arguing that attempts to solve “the problem of mental deficiency” would not fix crime rates (ibid.: 433–44). However, they argued that most of their subjects were not “driven into wrong-doing by want” (ibid.: 361); rather, that poor and irregular work skills, training, and personal history, as well as sparse educational opportunities and achievements, were due to a combination of both low intelligence and poor socioeconomic conditions.

Although Fernald et al. rejected purely biological or socioeconomic explanations for the origins of women’s crime, they also reinforced racist scientific theories, particularly against Black women, and sexist notions about women generally. Based on their intelligence testing, the researchers argued that the Black women of their study were “appreciably lower mentally than the white [women]” (ibid.: 422). Specific examples of romantic or sexual relationships between their white women subjects and non-white men – and especially Black men – were cited as further proof of immorality or propensity for crime (ibid.: 220, 321n12, 411). In general, the researchers went to great lengths to note the intellectual superiority of white women and the immorality of white women who had relationships with non-white men. Although they made adjustments for issues such as nativity (immigration history) and non-native English skills, they nonetheless maintained a racial status quo.

Fernald et al. also argued that sexual deviance was central to understanding women’s crime. Although not all of the subjects of their study were arrested and convicted of such crimes, the researchers claimed that nearly all (84%) of their subjects were “sex offenders,” or had sexual encounters outside of marriage, often for money. They concluded that “the sex problem is one of the most prevalent ones in a study of delinquent women” (ibid.: 388), a popular notion amongst Progressive reformers who attempted to correlate female sexuality, prostitution, and crime (Lunbeck Reference Lunbeck1987: 523–4, 541–2n18). Both consensual sex and rape, at any age, were considered sexual encounters. For those who engaged in intermittent or sustained sex work to support themselves, Fernald et al. argued that, on the one hand, such women were intelligent and manipulative, making a “deliberate,” “easier,” and more “lucrative” choice of prostitution rather a “serious attempt” at “honest work” (Fernald et al. Reference Fernald, Hayes and Dawley1920: 391, 400–1). On the other hand, and paradoxically, the researchers suggested that the sex offenders in their study had a lower mental capacity than others who had never engaged in any transgressive sex (ibid.: 523). Fernald et al. directly connected intelligence with the ability to obtain and keep employment, suggesting that sex offender women lacked intelligence. Although Fernald et al. were also sympathetic to women “whose offense against the law is made possible only by the participation of men” who were not held accountable under the law (ibid.: 13), they also argued that most of their subjects likely grew up in households with “low moral standards” (ibid.: 244). Fernald et al. attempted to reframe the issue of engaging in sex work as an issue of mental capacity, but still entertained judgments of women’s morality.

Besides “sex offenders,” Fernald et al. also argued that some women in general were simply too intelligent to be arrested or convicted (ibid.: 28–29). Surprisingly, given their own highly educated backgrounds, they also suggested that more educated women may be more easily able to conceal their crimes (ibid.: 272). It was unclear why: Was it because education made women more deviously clever? Or simply because it may have given them more socioeconomic resources? Many professional women of this era put “the interests of their professions ahead of…gender interests” (Lunbeck Reference Lunbeck1996: 6); in this case, though Fernald et al. argued that the longstanding stereotype that women were generally less intelligent than men was untrue, they still suggested a correlation between intelligent or educated women and deviance.

Overall, Fernald et al. presented their statistical analysis and data with few concrete conclusions other than the argument that women’s delinquency was a result of two factors: poverty and a slightly lower mentality, especially for “sex offenders” (Fernald et al. Reference Fernald, Hayes and Dawley1920: 525). They also wholly rejected that women were generally of lesser intelligence than men and that lower intelligence was synonymous with criminal propensity; however, Fernald et al. also still argued that sex crimes were a defining category for women; that most women convicted of crimes had a slightly lower mentality; that white women were superior to Black women; and that those women that evaded the law were clever, manipulative, or perhaps even aided by their education to do so. Although the researchers were careful to only state the results of their data and avoid generalizing to all women, they nonetheless reified deep-seated notions of criminalized women as either somewhat stupid and hyper-sexual or cleverly devious, as well as upholding a white supremacist racial hierarchy. These ideas about deviance continued to reinforce a type of womanhood founded on “respectable” wage labor that was unattainable for most, and especially for Black working women (Hicks Reference Hicks2003). Their arguments were strikingly similar to those in the studies of Healy, Thomas, and Parmelee, though they took great pains to reject the notion that all women were inherently or latently criminal biologically by building an argument based on statistical data and case studies.

In Republican China, Zhou Shuzhao in particular found Fernald et al.’s study compelling. She cited Fernald’s individual studies on women criminals in New York, as well as Fernald et al.’s (Reference Fernald, Hayes and Dawley1920) study, to examine the impact of age on criminal behavior. Although Fernald found that chances of criminal behavior that resulted in incarceration decreased with age, Zhou argued that this may not be the case for poor women in Republican China. Instead, she theorized that crime might be the only avenue for older poor women whose mental and physical state was deteriorating (S. Zhou Reference Zhou1932: 51–52). Despite these disagreements, Zhou was clear about how important Fernald’s methods in particular were to her own work; indeed, she saw herself as following in the footsteps of figures such as Healy, Glueck, and Fernald, writing,

Liu and Xu had no such great praise in their study; however, they did frequently cite data from U.S. social scientists and agree with their critique of modern society’s ills. They quoted, for example, Parmelee to summarize the main takeaway from their study: “It’s no wonder that Parmelee, professor of criminology in the United States, said ‘Crime and poverty are the two greatest evils in society’” (Q. Liu and Xu Reference Liu and Xu1932: 88). On the one hand, these examples demonstrate how both Zhou Shuzhao and research team Liu and Xu understood themselves to be in conversation with, and deeply influenced by, U.S. social scientists. On the other hand – and like other Chinese intellectuals adapting ideas from the West – Zhou, Liu and Xu also “did not blindly follow” (H. Liu Reference Liu2007: 99) their U.S. models, but translated such transnational theories to develop their own unique frameworks for Chinese women’s crime.

Gender and crime in Republican China

The introduction of sociology as a discipline in China began during the Late Qing period at the end of the nineteenth century, when prominent statesmen concluded that sociology could be a powerful “tool of statecraft” for national reform (H. F. Chen Reference Chen2018: 67): Yan Fu introduced the concept of evolutionary sociology to China, including translating Herbert Spencer’s Study of Sociology (1873); Kang Youwei used his expansive knowledge of ancient Chinese academic texts to examine sociology issues and may have also taught about sociology at his academy, Wanmu Caotang, in Guangzhou (S. Liu Reference Liu2020: 35, 46, 74–75, 133). The following generation of scholars, such as Liang Qichao and Liang Shuming, built upon this introduction to examine issues of education, governance, and culture through sociology (S. Liu Reference Liu2020: 149–59, 191). In Qing China, sociology was understood as a means by which to strengthen the Chinese state and people.

By the beginning of the twentieth century, however, sociology as an academic discipline was still generally only taught at Christian colleges by missionaries but was growing in popularity (Dirlik Reference Dirlik, Dirlik, Li and Yen2012: 3; H. F. Chen Reference Chen2018: 12). The mid-1920s to mid-1930s brought a deluge of thousands of social surveys in both rural and urban spaces (Li Reference Li, Dirlik, Li and Yen2012: 69; H. F. Chen Reference Chen2018: 16). Heavily influenced by the Chicago school, which encouraged focusing on social surveys and field work in urban areas, newly trained sociologists set out to collect massive amounts of data, statistics, and case study vignettes (Dong Reference Dong2003: 211–12). The late 1930s, though violently marked by Japanese invasion and occupation, also heralded a new relationship between the state and the discipline of sociology. Although Nationalist leaders were suspicious of sociology and conflated it was socialism, the field was increasingly supported by the regime: Newly nationalized universities founded sociology departments; academic sociologists trained with social workers; and sociology grew popular as a university major, particularly due to the establishment of the Princeton Center in Beijing in 1906. By the 1910s, Princeton alumnus Sidney D. Gamble began offering a course on the social conditions of the capital city and later published his 1921 study, Peking: A Social Survey (Gamble Reference Gamble1921; Dong Reference Dong2003: 211–12; H. F. Chen Reference Chen2018: 18–22). This model of missionary-founded colleges with strong ties to elite universities exemplified the growing trend of sociology programs in Republican China.

A number of important figures in Chinese sociology, most of whom studied abroad in the U.S. or Europe, emerged during this formative period. Sociologist Sun Benwen was trained at Columbia University, New York University, and the University of Chicago, where he worked with sociologists Robert E. Park and William I. Thomas. Sun advocated for the institutionalization of academic sociology under the Nationalists, despite its association with socialism, and was widely known as “the founder of Chinese sociology” (Li Reference Li, Dirlik, Li and Yen2012: 68; H. F. Chen Reference Chen2018: 21–22). Chen Da – who studied at Reed College and Columbia University – researched issues of labor, migration, identity, and eugenics (Candela Reference Candela2015). Pan Guangdan, though most famous for his work on eugenics to improve the Chinese race (Sihn Reference Sihn2010; Chung Reference Chung2018), was educated at Dartmouth and Columbia University, becoming Professor of Sociology at Tsinghua University in 1934 (Rocha Reference Rocha2012: 3). Lei Jieqiong, China’s most famous woman sociologist, studied at the University of Southern California and from 1931 taught sociology at Yanjing, focusing on women and the family in her long academic and political career (H. Liu Reference Liu2005: 179; Dirlik Reference Dirlik, Dirlik, Li and Yen2012: 4; Yue Reference Yue, Dirlik, Li and Yen2012). ^{Footnote 4} Most famous for his work in anthropology, Fei Xiaotong studied sociology at Yanjing under sociologists Wu Wenzao and Robert E. Park (Arkush Reference Arkush1981: 28–31). With their direct connections to U.S. universities and scholars, as well as their own research and social surveys, this generation of Chinese scholars represented the strong development of sociology as an academic discipline and subject of inquiry in Republican China.

Although Western scholars writing on Chinese society during this period tended to view crime in China as culturally situated and individual, historians (Dong Reference Dong2003: 228) have shown that this new cohort of Chinese sociologists passionately argued that crime was fundamentally a social problem. The educated elite advocating for social reform in Late Qing and early Republican China also embodied a “crisis-ridden habitus” in which the neo-Confucian foundations of society, politics, culture, and intellectual life were dramatically transformed under global capitalism (Candela Reference Candela2015: 363–64). Social surveys in particular held a special place as a means of making “sense of the nation” in its nascent and emerging form (Lam Reference Lam2011: 2). Thus, on the question of women’s crime, the potential for social science knowledge and extensive surveys to answer China-specific questions was especially powerful: Who were China’s women criminals and how did they become criminals? What was their relationship to national weakness and the so-called “woman problem”? What could be done about the supposedly increasing social problem of women’s crime and who should do it? Social science promised objective data, understandable models, and perhaps, even possible solutions.

A number of male Chinese penologists, criminologists, and sociologists addressed, at least in passing, the question of women’s crime. One such example was the prominent sociologist Yan Jingyue, Zhou Shuzhao’s advisor. Born in Zhejiang, Yan studied sociology not only at Yanjing, but also at the University of Chicago, where he received his Ph.D. in 1934. As a young professor, he taught at Yanjing and Dongwu Universities, and also traveled the world participating in international penology conferences (Dikötter Reference Dikötter2002: 199–201). Yan’s exceptional familiarity with U.S. social science no doubt shaped his view that crime was fundamentally a social problem that social reform could rectify. He was a prolific researcher and author, publishing not only in China but also in English-language academic journals on sociology (Yan Reference Yan1931, Reference Yan1934). His 1928 study, “Sociological Analysis of Crime in Peking,” published by the Department of Sociology and Social Work at Yanjing and reprinted in the academic journal Sociological World, cited numerous English-language studies by prominent sociologists and criminologists, including Henry M. Boies, W.A. Bonger (Dutch), Harold S. Bucklin, Charles A. Ellwood, Sidney D. Gamble, John Lewis Gillin, Henry H. Goddard, and Maurice Parmelee (Yan Reference Yan1928: 43–45). Yan’s study focused primarily on male inmates, but also compared some data on men’s and women’s crime. Much like his counterparts in the U.S., Yan concluded that women’s crimes and women criminals were fundamentally different from the crimes of men, primarily due to social and physiological differences.

Yan argued, like Parmelee, that there were consistently far fewer incarcerated women than men in Republican China because women stayed within the home and encountered far less “evil coercion” than men did in their roles outside the home. This not only meant that they relied on men for their livelihood, but also that opportunities to commit a crime were “naturally less.” Yan did not offer an explicit connection to the patriarchal traditions that might foster these conditions, though he did note that the treatment of men and women in Chinese society created a “cowering attitude” (weisuo) in women and an “indomitable behavior” (yiwangwuqian de xingwei) in men (Yan Reference Yan1928: 8–9). Yan also did not use the lens of class to explore how poor women were still often forced to work outside of the home to make ends meet regardless of their gender. ^{Footnote 5} Instead, he focused only on how “society” (shehui) shaped and controlled both genders. This analysis aligned with Yan’s theory that the parameters of crime were always evolving with and defined by society (Dong Reference Dong2003: 222). For women in early twentieth-century China, these social parameters of crime were both a benefit and a curse: they were less easily exposed to or lured into crime due to their social roles, but also, could not independently support themselves.

In addition to the restrictions of Chinese society, however, Yan also argued that women’s bodies kept them from physically committing violent acts, including rape, because of their “physiology” (shengli) and “constitution” (tizhi) (Yan Reference Yan1928: 8). Yan’s definition of rape was unclear; however, he suggested that women were not physically strong enough to overpower and assault men. He provided data from 1920 to 1926, for example, in which there were no recorded instances of Beijing inmate women committing violent robberies. While men’s crimes were associated with violence, Yan described women’s crimes as “complicated” (fuza) and “concealed” (ancang). ^{Footnote 6} As a result of the hidden and surreptitious nature of women’s crimes, Yan argued that there was often a lack of “real or knowable evidence” and women’s crimes were thus more difficult to prove. To link this to his social theory of crime, Yan argued that the paucity of cases of women involved in violent crimes was evidence that there were “pros and cons” to the treatment of women in society. His conclusion suggested that a possible “pro” was that society produced women subjects who were less violent and more docile than men (Yan Reference Yan1928: 8–9). Yan’s suspicions of women as socially and physiologically different from men were sufficient enough for him to portray them as having a potentially higher, but unknown, propensity for crime.

These suspicions may explain why Yan was particularly critical of what he deemed were legal “advantages” (pianyi) women possessed over men; namely, that they were more likely to receive shorter custodial sentences, suspended sentences, or be granted parole (Yan Reference Yan1928: 8–9). He cited a publication from the Counselor’s Office of the Ministry of Justice that outlined the special treatment of married women charged with misdemeanors, which encouraged releasing women on bail “in accordance with the old laws and regulations” and changing sentences of imprisonment for such petty crimes to a fine:

Yan took this as evidence of women’s “advantages” in the courtroom (Tran Reference Tran2015). As historians (Kuo Reference Kuo2012; Tran Reference Tran2015) have shown, the developing Republican civil code of the late 1920s did potentially grant women increasing legal agency, particularly if they carried the status of either wife or concubine. The regulations from the Ministry of Justice he cited used two terms that specifically referred to married women: Funv – a common term that typically refers to older, married women – and guinv, a less common term which can refer to one who has been married out to her husband’s family. ^{Footnote 7} Yan, however, did not explore the fact that such “advantages” were predicated upon women’s reproductive labor and rearing of children. Women’s status as wives and mothers was the most important element in their sentencing; they may have been more likely to escape imprisonment than men, but only under the condition that they fulfilled their prescribed gender roles.

Yan also noted a 1913 judicial opinion on a case of “consenting to abduction” (heyou) in which the man, but not the woman, was punished. Although there was no specific law which outlined an exception for women, Yan believed that such cases were often the result of plans men and women made together to elope, but that women “always say that they ‘were seduced’ and can escape the law” (Yan Reference Yan1928: 8–9). Although Yan’s explanation acknowledged the autonomy of women in this example, he also elided a more complete explanation of the power differential between men and women during the Republican era and instead simply concluded that women were much more likely to be handed down a sentence of “not guilty” by the courts.

Yan’s assertion that women’s crimes were linked to their unique social status and physiology fit within the broader academic discourse on women’s bodies and crime in Republican China. Books written in the fields of sociology, physiology, and sexology generally argued that women with intense sexual desire were more physiologically inclined toward prostitution; that menstruating, pregnant, or menopausal women were more prone to crime; and that women, who were deemed less evolutionarily developed than men, were more likely to commit crimes such as abortion and infanticide (Dikötter Reference Dikötter1995: 50–51, Reference Dikötter2002: 188–89; Kiely Reference Kiely, Zelin and So2013: 145). Historian Jan Kiely argues that these studies portrayed criminalized women as “either hapless and ignorant or deeply evil,” a framing likely influenced by long standing historical discourses of female yin (Kiely Reference Kiely, Zelin and So2013: 151–52), or the feminine aspect of the yinyang “binary principle” which shaped the material world into gendered hierarchies (Furth Reference Furth1999: 21). This “gender system that stressed female weakness” (Furth Reference Furth1986: 44) may have informed new Republican-era scientific theories of the inferiority of women’s physiology, especially their reproductive and sexual biology. A younger generation of women scholars who focused primarily on women’s crime, however, including Yan’s own student Zhou Shuzhao, utilized the new language of social science differently.

Zhou, Liu and Xu on women’s crime

Not only was the field of sociology in Republican China dominated by male figures such as Yan Jingyue, but studies of crime also largely focused on male criminals. There were, however, two comprehensive studies on women inmates in the early 1930s, both of which were conducted by young Chinese women researchers. The most prominent and influential of them was Zhou Shuzhao, who studied under Yan at Yanjing as a graduate student and also worked with Robert E. Park when he was a visiting lecturer for a semester in 1932 (Ransmeier Reference Ransmeier2017: 275). Zhou published a survey of one hundred women prisoners in Beijing in 1932, as well as several other articles on the problem of women’s crime (S. Zhou Reference Zhou1934a) and statistical methodology (S. Zhou Reference Zhou1934b) in the 1930s. In the 1940s, it appears as though she may have temporarily relocated to Guizhou, where she penned some articles on local education and culture (S. Zhou Reference Zhou1942b, Reference Zhou1934a), before settling in Taiwan after WWII (Ransmeier Reference Ransmeier2017: 308). ^{Footnote 8} Although we know few details of her personal life, we know that Zhou came from an elite and prominent family. Her great-grandfather, Zhou Fu, served as governor-general of Jiangsu, Jiangxi, Guangdong, and Guangxi, and also drafted anti-drug trafficking laws as a Qing dynasty official in North China (Y. Zhou Reference Zhou2013: 2; Ransmeier Reference Ransmeier2017: 273). Her younger cousin and well-known Harvard-educated historian, Zhou Yiliang, described her as a writer who introduced him to new Chinese and British literature and attempted to give him a fashionable perm (Y. Zhou Reference Zhou2013: 4–5, 17). ^{Footnote 9} By all accounts, she seemed a modern young woman interested in foreign ideas and trends, like many urban youth of her generation.

We know far less about the research team Liu Qingyu and Xu Huifang. However, we do know that the pair were graduate students at Shanghai Baptist College (Kiely Reference Kiely, Zelin and So2013: 144–45), an institution founded by U.S. missionaries in 1906. The college became well known for its Department of Sociology founded in 1913, which was considered to “be a central part” of the college mission (Payne and Liu Reference Payne and Liu1932: 209). Compared to other missionary colleges, “the dominance of the Westerners was even more striking” at Shanghai Baptist and there were many issues with retaining Chinese faculty due to poor “salaries, status, and working conditions” (Lutz Reference Lutz1971: 197, 195). The sociology department itself was no exception: Daniel H. Kulp – Columbia University sociologist and author of Country Life in South China – served as chair, and Brown University professor of political science and sociology Harold S. Bucklin taught there from 1924 to 1925, when he published his study. A Social Survey of Sung-Ka-Hong, China (Rawlinson Reference Rawlinson1919; Bucklin Reference Bucklin1924; Kulp Reference Kulp1925; Lutz Reference Lutz1971: 110–11; Arkush Reference Arkush1981: 25; Mitchell Reference Mitchell1993: 334–35). Liu and Xu may have also had access to the Yangtzepoo Social Center – a laboratory for students “to obtain practical experiences and first-hand observation in social investigation” – and the International Relations Library, a collection of over 4,000 books from 40 different countries (Payne and Liu Reference Payne and Liu1932: 210–11). In 1920, Shanghai Baptist was also the first missionary college in China to become co-educational (Foreign Mission Board 1920; Ray Reference Ray1924: 11; Payne and Liu Reference Payne and Liu1932: 208). By 1931, when Liu and Xu were likely enrolled, there were 170 women students (Payne and Liu Reference Payne and Liu1932: 208). Shanghai Baptist was thus an exceptionally international space and a national center for sociology research with growing support for women scholars.

Although Liu and Xu’s Reference Liu and Xu1932 study was less cited and influential than Zhou’s, it was also a product in Western-run and funded missionary colleges. For sociology departments in particular, there was “little alteration” of U.S.-based course content and the curriculum thus likely looked the same as that of their counterparts at U.S. universities (Lutz Reference Lutz1971: 189). Their large-scale social surveys of Chinese women’s crime in the Republican period – the only ones of their kind – were thus produced in the international and “bicultural” environments of Yanjing (Rosenbaum Reference Rosenbaum2015) and Shanghai Baptist, institutions with direct and deep ties to Brown, Chicago, Columbia, Harvard, Princeton, and Yale, among other elite U.S. universities (West Reference West2007, Reference West1976; Gjedssø Bertelsen and Thybo Møller Reference Gjedssø Bertelsen and Møller2010; Fan Reference Fan2014). It is perhaps of little surprise, then, that Zhou, Liu, and Xu were extremely well-read in contemporary English language, U.S.-based sociological research and made connections to them in their own studies. Furthermore, there were similarities across their methodologies, conclusions, and female authorship. All of these factors merit closer examination and comparison to fully understand the study of women’s crime in 1930s China.

As previously shown, Zhou and research team Liu and Xu both crafted a comprehensive citational web that demonstrated their knowledge, especially U.S. researchers with connections to the Chicago school. The influence of these transnational works in their own research is clear in three important ways: (1) methodology (extensive interviews, statistical analysis), (2) primary emphasis on nonbiological factors, and (3) advocacy of institutional reform and improvement. Crucially, however, Zhou and research team Liu and Xu also broke ranks with both their academic peers and advisors in China and U.S. Chicago-school affiliated researchers: They presented women’s crime as solely determined by social factors, not biological, physiological, or psychological factors tied to gender. The biased views against women and their biology, found to varying degrees in both Yan and U.S. researchers’ studies, were nowhere to be found in the writings of Zhou, Liu, and Xu.

The methodological approaches of their studies fit what historian Hon Fai Chen called the “social survey movement” from 1927 to 1935 in which over nine thousand social science surveys were conducted by Chinese sociologists; 68 reports on some of these surveys were published in the journal Sociological World by the Yanjing Sociology department, including Zhou Shuzhao’s (H. F. Chen Reference Chen2018: 16–17). In contrast, Liu and Xu’s study was published in The Continent Magazine (Dalu zazhi), a Shanghai general interest magazine focused on social and cultural topics which only ran from 1932 to 1934 (“Dalu Zazhi [The Continent Magazine]” n.d.). Both Zhou and research team Liu and Xu interviewed hundreds of incarcerated women and compiled statistical data on every possible aspect of their lives (name, age, native place, bad habits, marital status, children, occupation, education, religion, family member backgrounds, income, living situation, social lives), as well as every detail of their arrest and incarceration (criminal charge, location of crime, length of sentence, date of conviction/incarceration) (Q. Liu and Xu Reference Liu and Xu1932: 73; S. Zhou Reference Zhou1932: 35). Zhou found that 82% of the women inmates in her survey were thirty years or older (S. Zhou Reference Zhou1932: 49); and in Shanghai 66% of Liu and Xu’s subjects ranged from 30 to 74 years old (Q. Liu and Xu Reference Liu and Xu1932: 75). Both groups were significantly older than the average male offenders in 1930s Republican prisons, most of whom were in their twenties and thirties (Kiely Reference Kiely2014: 87). Women inmates in China were also nearly all married; surveys of U.S. women inmates in 1910, in contrast, showed that the majority of women inmates were unmarried (Q. Wang Reference Wang1993: 15–16). In Zhou’s survey, 98% of women were or had been married (S. Zhou Reference Zhou1932: 53); Liu and Xu similarly found 96% (Q. Liu and Xu Reference Liu and Xu1932: 82). Both studies also noted that the vast majority of women were illiterate, with no formal education or vocational training whatsoever (Q. Liu and Xu Reference Liu and Xu1932: 85; S. Zhou Reference Zhou1932: 55–57). Finally, there were also notably high rates of unemployment: 44% for Zhou’s Beijing inmates (S. Zhou Reference Zhou1932: 57) and 30% for those in Shanghai (Q. Liu and Xu Reference Liu and Xu1932: 80). The average profile that emerged from their studies was an older, married or formerly married, illiterate woman who had few opportunities for either education or stable work outside the home.

Beyond the personal information and background of inmates, determining and classifying the types of crimes their women subjects committed were of particular interest for Zhou, Liu and Xu. Of her one hundred interviewees, Zhou found that the majority of women were charged with seduction (37%) and women were seven times more likely to commit forcible abduction crimes than men (S. Zhou Reference Zhou1932: 38). Both “seduction” and “harming the family” were part of a new category of criminal charges labeled “Offenses Against the Family” introduced during the legal reforms of the Republican period, particularly the 1935 Criminal Code. These laws were typically used to charge men who eloped with women under 20 years old by “seducing” them into a romantic relationship or having intercourse with “young, desperate women,” who were then convinced or coerced to be sold themselves (Ransmeier Reference Ransmeier2017: 288, 365n20). Republican lawmakers hoped “to support women’s social autonomy and protect their sexual choices” through such legislation while also maintaining and protecting the institution of the family (Z. Ma Reference Ma, Chen and Zelin2015b: 201). Some families, however, used these laws to control their daughters when they were unhappy with their choice of partner (Z. Ma Reference Ma, Chen and Zelin2015b: 192–93, 197–98). Zhou’s research showed the many complex layers of human trafficking: New Republican laws designed to protect women often had unintended consequences that could also harm them.

The next most common crimes for Beijing women in Zhou’s study were opium (10%) and harming the family (8%) (S. Zhou Reference Zhou1932: 41–43). In Liu and Xu’s study, the highest percentage of women were charged with “economic crimes” (78%), and the top three overall charges were related to opium (25%), abduction (18%), and kidnapping for ransom (8%) (Q. Liu and Xu Reference Liu and Xu1932: 76–77). One possibility for this grouping could be that these cases of seduction and abduction were considered trafficking for marriage or prostitution, as opposed to crimes like kidnapping for ransom. Abduction (guaipian) could indicate abduction for illicit sexual purposes, such as prostitution, which was a major issue during the Republican period; scholar Wang Juan, for example, found that 76% of sex workers in Beijing in 1936 had been abducted (Wang Reference Wang2006: 50–51). Liu and Xu also classified consensual seduction (heyou) and forcible abduction (lueyou) as both economic and sexual, a choice which revealed the economic underpinnings of such crimes (Q. Liu and Xu Reference Liu and Xu1932: 76). A “mature woman’s crime,” the trafficking of children and young women supplied a high demand for “servants, wives, slaves, concubines, child brides, prostitutes, apprentices, and adopted children” during the Late Qing and Republican periods (Ransmeier Reference Ransmeier2017: 306). Kidnapping in general was a major concern during the 1930s. In Shanghai, organizations such as the Anti-Kidnapping Society fought against the kidnapping of women and children for such purposes, especially sex work (Hershatter Reference Hershatter1997: 181–202). Age comparisons by criminal charges supported the theory that trafficking was a “mature woman’s crime”: 72% of Zhou’s interviewees who had been charged with seduction were over forty years old (S. Zhou Reference Zhou1932: 52). Older women, it seemed, had a particular position or situation that made them more likely to be charged with such crimes.

In general, such surprising data led to large questions about the causes of women’s crime: What created these trends, such as the prevalence of trafficking and drug-related crimes? What was specific, if anything, about women or their lived experiences and its relationship to crime? Zhou Shuzhao in particular seemed to have deep sympathies for the incarcerated and once even called their victims “stupider than” their own mothers for falling for their tricks (Ransmeier Reference Ransmeier2017: 278, 283). In a later 1934 essay published in Eastern Miscellany, “Beijing Women’s Crimes and the Woman Problem,” Zhou revisited the question of the origins of women’s crime. She dismissed ideas associated with Lombroso, writing: “What percentage of the so-called ‘natural criminals’ and ‘criminal type’ are there? I dare say it is almost none” (S. Zhou Reference Zhou1934a: 6). ^{Footnote 10} Although Zhou did not directly cite any of Cesare Lombroso’s works – including Criminal Woman – this was most likely her translation of Lombroso’s influential idea of the “born criminal” (tiansheng fanren). Liu and Xu also did not cite Lombroso’s Criminal Woman, but did cite and summarily dismiss the conclusions of his seminal work, Criminal Man; namely, they argued that IQ studies in China had found no notable differences in intelligence between the general and incarcerated populations (Q. Liu and Xu Reference Liu and Xu1932: 77–79). Liu and Xu also agreed with Zhou’s assertion that there were “almost none” who fit the hereditary criminal type, writing that there was no scientific evidence for such and it was “rarely seen” (Q. Liu and Xu Reference Liu and Xu1932: 71). By rejecting the notion that all incarcerated people, regardless of gender, were not genetically inferior to the non-incarcerated, Zhou, Liu, and Xu also bolstered their argument that incarcerated women were not biologically or mentally inferior to the general population or men.

Instead, Zhou, Liu, and Xu focused on systemic social factors. Zhou argued that criminals were the oppressed of society, and that poor women criminals – who lacked education, paths to divorce, or a strong familial support base – were the most oppressed. Zhou concluded that the biggest factor in this lack of access and subsequent turn to crime was the rampant poverty in the capital city. Roughly 29% of Zhou’s interviewees had no income, and 45% had an unstable income (S. Zhou Reference Zhou1932: 58). Their household heads, most of whom were men (70%), had also received no formal education (59%) and had no daily income (39%) (S. Zhou Reference Zhou1932: 65–67). Zhou cited research by Sidney D. Gamble, who noted that a 1914 census classified 12% of the Beijing population as “poor” or “very poor.” Within this group, nearly 45% were women, despite a severe gender imbalance within the overall population of Beijing in which women were roughly 37% of the general population. Previous estimates actually suggested much higher levels of poverty – potentially upwards of three-fourths of the capital population (Gamble Reference Gamble1921: 270). Zhou also cited other studies by Gamble and his research collaborator T’ien-p’ei Meng, as well as a study on handicraft workers in Beijing by L.K. Tao (Tao Menghe), and a study on subsistence levels by Fang Fu An (Fang Reference Fang1929; Tao Reference Tao1929). All of these studies combined to paint a picture of a struggling and impoverished underclass of which no small number were vulnerable women.

Liu and Xu’s findings for Shanghai were similar to those of Zhou; they primarily found that most incarcerated women in Shanghai were poor and economically exploited. The “evils” of urban spaces, which other scholars (Kiely Reference Kiely, Zelin and So2013) have noted were a concern for a number of Republican-era sociologists at this time, were especially acute in class-stratified Shanghai:

To fix these issues, Liu and Xu advocated for social reforms across education, labor, entertainment, police, hygiene, medicine, and much more. For women in particular, they stressed not only supporting women’s movements, improving women’s social status, and equal opportunity for both genders, but also the need for breaking negative psychosocial factors, such “improving women’s self-esteem and breaking the degrading mentality of ‘valuing men, demeaning women’ (zhong nan qing nv)” (Q. Liu and Xu Reference Liu and Xu1932: 92). Zhou too agreed that women’s situations gave them little else to resort to beyond crime. She stressed the inevitability of crime if women’s basic needs were not met, as well as for the importance of social science as an assessment tool and crime as a social issue:

The conclusion that women would simply have no need for crime if they were educated and given socioeconomic stability neatly summed up the central arguments of Zhou, Liu and Xu’s work on women’s crime. Rather than biological, physiological, or psychological defects, they argued that the origins of their crimes were purely a result of social and economic systems. This set them apart from their male colleagues in Republican China and the U.S., who stressed social factors but clung to the notion that there were inherent aspects to women’s biology, physiology, or psychology that led them to crime. Likewise, though Fernald et al. rejected arguments of women’s inferiority, Zhou, Liu and Xu disagreed with any arguments about the potential “lower mentality” of incarcerated women.