Abstract

The use of longitudinal data from the criminal records of a sample of 6,042 female prisoners in nineteenth- and twentieth-century Victoria reveals limitations in the traditional method of examining criminality within specific offense categories. Investigations devoted exclusively to particular categories of women’s offenses potentially obscures the extent to which women resorted to multiple forms of offending. Such versatile activity challenges conceptions of women as predominantly petty offenders by suggesting that some women were arrested for minor offenses because of their engagement in more serious crimes and their participation in criminal sub-cultures.

Most of the criminal offending by women in common-law jurisdictions during the nineteenth and early twentieth century fell into three main categories—property, personal, and public-order. As Williams comments in respect to Victorian England, “Whilst crimes of theft most often saw women convicted of felonies and sent to convict prisons and violent crimes stole newspaper headlines, women who drank excessively or sold themselves on the streets (often both) probably constituted the largest single group of female offenders in Victorian England.” Many historical studies of female offenders consequently approach the study of the different crimes committed by women in isolation from each other, largely treating the women involved in each category as belonging to a different group. They seldom deal with the potential overlap between these groups in depth.1

In Australia, scholarly work about women’s offending has been even more piecemeal; most of the analyses concentrate on women’s involvement in specific crimes. The only monograph offering a more overarching narrative of female offending, Judith Allen’s Sex and Secrets: Crimes involving Australian Women since 1880 (New York, 1990), for the most part discusses offenses connected to sexuality and reproduction so as to explore how the policing of such “gendered” crimes reinforced traditional power hierarchies. Other work devoted to female crime in Australia also tends to concentrate on the incidence and social discourse surrounding such “feminine” offenses as abortion, infanticide, baby farming, and prostitution. Only a few studies examine the policing of women for public-order offenses, particularly vagrancy and drunkenness, which represent the majority of the charges historically brought against women. Even less attention has been paid to female theft, despite the fact that property offenses dominate women’s felony indictments (due, in part, to the dearth of historical studies about property crime in Australia, beyond the ongoing popular and academic interest in “bushranging” crimes). Rarer still, both inside and outside Australia, are historical works that explore the interaction between these categories of offense as a reflection of the versatility in women’s offending patterns throughout their criminal careers.2

Only in recent years has a limited body of literature—in historical criminology and the history of criminal justice—about offending throughout the entire life course emerged, encouraged by the increasing digitization of historical records. The most thorough historical analysis of criminal careers to date—a study of persistent offenders in Crewe, England, between the 1880s and 1940s—concentrated largely on male offenders. The few female offenders included in the sample confirmed the conclusion, already advanced in historical research, that female recidivism was predominantly characterized by the “repeat prosecution of prostitutes and destitute women” for public-order offenses. Other works dealing with longitudinal female offending also tend to focus on the petty offending that comprised the bulk of women’s criminal careers. However, in a study of the criminal careers of thirty-three female prisoners interviewed as part of the 1887 Royal Commission into the Queensland prison system, Piper pointed out that despite their vast number of public-order offenses, the majority of these women were also versatile offenders who amassed a number of convictions for theft and violence.3

In contrast, a considerable corpus within the field of criminology traces patterns in crimes committed, relating to both offense specialization and versatility. Yet, many of the contemporary studies of specialization and versatility tend to rely exclusively or predominantly on male samples. A prominent exception is a broader study of criminal offending within the 1953 birth cohort in England by Francis, Soothill, and Fligelstone, which found that most of the women who committed crimes were versatile offenders. The bulk of the criminological literature further suggests that offending versatility is linked to more frequent offending, although Brame, Paternoster, and Bushway found that versatility and frequency operated independently of each other. On this basis, in line with Piper’s findings, the high levels of recidivism evident among nineteenth-century women in Australian and other English-speaking jurisdictions may well indicate that a significant proportion of these repeat offenders were also versatile offenders.4

This apparent scant specialization among female offenders could offer an important corrective to the understanding of women as predominantly petty offenders (even if the bulk of their crimes fell into this category). It could also cast new meaning on the policing of public-order offenses, suggesting that it was prompted less by concerns about order and morality than by the desire to control and prevent more serious forms of offending. Hence, this article examines patterns of offending throughout the criminal careers of women imprisoned in Victoria between 1860 and 1920 to assess the extent and significance of versatile offending. Ultimately, this analysis suggests that the historical study of crime—particularly as committed by women—benefits from moving beyond broad (and sometimes hollow) categories toward a more holistic view of criminal careers and the place of specific offenses within them.

Methodology

The main source informing this analysis is the Central Register of Female Prisoners, a series of records created by Victoria’s penal department to register the names, personal details, and convictions of women taken into custody in Victorian prisons. Upon a woman’s first entry to prison, the record created for her included a registration number to track her thenceforward through the system, even subsequent returns under different aliases. The format of the record-keeping system remained consistent for several decades, documenting such details as birthplace, year of birth, religion, occupation, literacy, marital status, the date of conviction, the offense, the court involved, and the sentence, as well as notes about appearance, personal background, and behavior in prison. Subsequent returns to prison saw further offenses added to the same page; information about other convictions meriting a fine or a short stay in police lockup, rather than imprisonment, often entered the log as well.5

Between 2014 and 2015, volunteers at the Public Records Office of Victoria created a spreadsheet of the names, registration numbers, and basic biographical details for all of the women listed in the Central Register. The authors of this article, with the help of a research assistant, later oversaw the transcription of the additional information contained in the prison records into an spss database. The result of this work is a sample of 6,042 individual women who first entered the central prison system between 1860 and 1920 (note, however, that records pertaining to prisoners first entering during 1871 are unavailable). The sample does not include women incarcerated in Victoria’s prisons during the study period whose first imprisonment came prior to 1860, since their records exist in earlier registers. Victoria’s record-keeping practices also meant that the offending histories of the women were not limited to this sixty-year period. Three women in the sample had convictions prior to 1860 for which they had not gone to prison; the earliest date for a woman’s first known conviction of this kind was September 20, 1854. Furthermore, 124 women continued offending after 1920; the latest date for a woman’s last-known conviction was December 23, 1947.

The richness of the register’s information makes it an invaluable source for understanding women’s offending in the context of their life courses, although the nature of information-collection practices makes it less than perfect. The information often came from the women themselves, except for conviction histories, which the prison staff seems to have validated with reference to official police files, sometimes even noting that a prisoner admitted to convictions that were not listed on their police record. Prison staff were occasionally careless, however, when recording convictions that had not resulted in imprisonment, noting, for example, that an offender had “previous convictions for minor offenses” without giving details. Hence, recidivism rates are probably underestimated to some extent. Information was missing here and there from the register, usually due to damage, but since the effect was random, it had little influence on the significance of the patterns found therein.

For the purposes of this article, two findings are important: (1) the first crime for which a woman was incarcerated and (2) whether its perpetrator specialized thereafter in violent crime, theft, or public-order offenses, or instead committed a mixed range of offenses during her criminal career. Chi-square analyses are used to examine correlations between particular types of offending and specific characteristics, such as the overall number of convictions, length of criminal career, age at onset of offending, and location of offending. The association of specific types of offense with offense switching and levels of recidivism are also examined. Such an investigation allows an analysis of the data regarding women’s profiles and the pathways that led them to crime, as well as of the relationship between serious and petty crime.

Type and Frequency of Offense

As with female offending elsewhere, the bulk of women’s criminal activity during this period in Victoria consisted of offenses against public order (see Table 1). Crimes of poverty—that is, vagrancy, begging, or lacking lawful means of support—represented the most common category for which women were first imprisoned, accounting for 2,074, or 34.3 percent, of female prisoners’ initial imprisonment. Disorderly, indecent, or riotous behavior was the reason for the first imprisonment of 1,220 first-time female prisoners (20.2 percent). Smaller numbers of women were also first imprisoned for such public-order offenses as drunkenness (311 or 5.1 percent), prostitution offenses (169 or 2.8 percent), obscene or abusive language (139 or 2.3 percent), and occupying or keeping a house frequented by thieves, rogues, or suspected persons (71 or 1.2 percent).

Table 1

Offense Types, First Entry to Prison, and Proportions of Versatile Offenders

offense categoryoffense typenumber of women first imprisoned for offenseversatile offenders (%)
Theft Larceny offenses 822 27.9 
Receiving stolen goods 116 24.1 
Pickpocketing 105 51.4 
Fraud offenses 98 13.3 
Robbery 34 58.8 
Burglary 22 36.4 
Stock offenses 13 53.8 
Violence Threatening life or harm 263 31.9 
Assault, wounding, injury, or attempted murder 176 37.5 
Murder or manslaughter 80 8.8 
Public order Vagrancy, begging, or lacking lawful means of support 2,074 18.7 
Disorderly, indecent, or riotous conduct 1,220 16.3 
Drunkenness 311 27.3 
Prostitution 169 29 
Obscene, indent, or abusive language 139 28.8 
Consorting with or keeping a house frequented by thieves, rogues, or suspected persons 71 32.4 
Other Offenses against justice or courts 43 37.2 
Offenses involving care of children 36 16.7 
Damaging property 40 42.5 
Concealment of birth 35 2.9 
Arson 33 12.1 
Intent to commit or aiding a felony 22 13.6 
Suicide 19 52.6 
Illegally selling liquor 17 11.8 
Abortion 11 
Bigamy 11.1 
Miscellaneous offenses 44 31.8 
Unknown 20 15 
offense categoryoffense typenumber of women first imprisoned for offenseversatile offenders (%)
Theft Larceny offenses 822 27.9 
Receiving stolen goods 116 24.1 
Pickpocketing 105 51.4 
Fraud offenses 98 13.3 
Robbery 34 58.8 
Burglary 22 36.4 
Stock offenses 13 53.8 
Violence Threatening life or harm 263 31.9 
Assault, wounding, injury, or attempted murder 176 37.5 
Murder or manslaughter 80 8.8 
Public order Vagrancy, begging, or lacking lawful means of support 2,074 18.7 
Disorderly, indecent, or riotous conduct 1,220 16.3 
Drunkenness 311 27.3 
Prostitution 169 29 
Obscene, indent, or abusive language 139 28.8 
Consorting with or keeping a house frequented by thieves, rogues, or suspected persons 71 32.4 
Other Offenses against justice or courts 43 37.2 
Offenses involving care of children 36 16.7 
Damaging property 40 42.5 
Concealment of birth 35 2.9 
Arson 33 12.1 
Intent to commit or aiding a felony 22 13.6 
Suicide 19 52.6 
Illegally selling liquor 17 11.8 
Abortion 11 
Bigamy 11.1 
Miscellaneous offenses 44 31.8 
Unknown 20 15 

Women who first entered the prison system following the commission of what might be considered serious crimes were in the minority. Only a small percentage of women (cumulatively around 2.5 percent) were first imprisoned for crimes generally regarded as serious but not involving theft or violence, such as justice offenses (43), damaging property (40), arson (33), aiding a felony (22), abortion (11), or bigamy (9). A sizable proportion (20.1 percent) first entered the prison system because of theft convictions; slightly less than 10 percent were first imprisoned for violent crimes. Most of the convictions for theft and violence were minor. Around 67 percent of the 1,210 women who first entered the prison system for theft were charged with larceny. Receiving stolen goods, pickpocketing, and fraud offenses were the next-most-popular categories. Only 22 women were imprisoned for burglary; 34 did time for robbery. Likewise, only 80 women were first imprisoned for murder or manslaughter; 176 women were convicted of assault, wounding, or causing injury; and 263 were charged with making violent threats. As the number of women entering the prison system declined (73 percent of the sample were first imprisoned before 1890), the proportion of women entering as a result of violent offenses fell (from 9.9 percent before 1890 to 5.1 percent afterward). This trend is the inverse of the contemporary one that links rising numbers of female prisoners to higher proportions of violent crime.6

The petty nature of the offenses that brought women into first contact with the prison system is further indicated by the court level at which women were tried—86.6 percent convicted at summary jurisdiction and 13.2 percent indicted before the General Sessions or Supreme Court. It was also reflected in the sentences passed against them: 74.4 percent of women received terms of six months or less, whereas only 2 percent received a term of more than two years’ duration. However, even these short sentences might be regarded as severe given the trivial nature of most women’s offenses; the crime of using obscene language potentially attracted a three-month prison term. This heavy-handed assignment of prison sentences has been cited as the primary reason why women constituted a higher proportion of the overall prison population in the nineteenth than in the twentieth century. As Finnane argues, where imprisonment is a common penalty for minor offenses, the consequence is likely to be a proportionately larger female prison population. By the 1890s, even officials often noted the futility of imprisoning large numbers of women for offenses of poverty or drunkenness, arguing that such practices were corruptive and simply led to women cycling in and out of prison on short sentences. During the latter part of the sample period, a profusion of refuges, charity homes, and inebriate asylums were introduced to divert women away from the prison system, probably contributing to the declining numbers of imprisoned women (see Figure 1).7

Fig. 1

The Number of Women First Entering Prison in Victoria by Type of Offense Committed throughout Their Criminal Careers, 1860–1918

Fig. 1

The Number of Women First Entering Prison in Victoria by Type of Offense Committed throughout Their Criminal Careers, 1860–1918

The number of women first entering the prison system peaked in 1872—likely due to a general crackdown on public disorderliness that also caused a similar increase in male crime rates, as well as the particularly rapid growth in Victoria’s female population between the 1861 and 1871 census. The number of women imprisoned declined thereafter, particularly toward the turn of the century. Yet while the overall number of women imprisoned in Victoria decreased during the 1890s and early 1900s, those individuals who went to prison tended to return more often. The majority of women in the sample were recidivists; single-conviction offenders comprised just 42.7 percent of the sample. Repeat offending by female criminals rose significantly over time—the mean of 2.6 convictions for those who first entered the prison system in the 1860s rising to 5.7 convictions for those first imprisoned in the 1890s and then to 7.6 convictions among women first imprisoned during the 1910s. Most repeat offenders were low-level recidivists—37.4 percent of the sample amassing two-to-five convictions. Mid-level recidivists (six-to-nine convictions) and chronic recidivists (ten or more convictions) comprised around 10 percent each of the remaining women.8

Even when taking recidivist activity into account, the vast bulk of women’s criminal histories still consisted of minor offenses. Only 7.2 percent of women ever received a sentence longer than twelve months; more than 80 percent of the sample were only ever tried in the summary courts. The proportion of women who engaged in theft or violent offending rises, however, when convictions throughout women’s criminal careers are considered. Public-order cases still account for the most significant share of women’s offending (involving 73.9 percent of the prisoners), though the entirety of women’s records shows a slight increase in the proportion of women who committed theft, violent assault, etc. At some point, 28.8 percent of women were convicted of theft, 12.9 percent of violent offenses, and 11.2 percent of various other crimes.

Findings about the extent of female offenders’ involvement in serious crime are not as dramatic as those of Piper for a much smaller Queensland-based sample (thirty-three), in which more than 50 percent of the women, during offending careers spanning from the 1870s to 1910s, had a conviction for theft and more than 40 percent a conviction for violence. Apart from differences in jurisdiction and sample size, the higher levels of versatile offending among these Queensland women may be attributable to particularities in their profiles and backgrounds. Most of the women in that sample were identified as recidivists who committed serious crimes within a pattern of chronic offending that derived from their membership in a prostitution sub-culture in the urban capital of Brisbane. The Victoria sample, in contrast, included higher numbers of regional offenders (that is, offenders outside the capital of Melbourne) and lower numbers of chronic offenders. But, as we demonstrate in the next section, those who conformed to the offender profile that dominated the Queensland sample were significantly more likely to be versatile offenders who interspersed incidents of serious crime with the usual records of public-order offenses.9

Versatile Offenders

Empirical criminological research consistently indicates that the majority of criminals commit diverse offenses throughout their lives, although they may go through short-term periods of specialization. However, the proportion of versatile offenders in our sample is smaller than that found in contemporary studies; the 1953 study of Francis, Soothill, and Fligelstone reported that around 66 percent of female offenders fell into the versatile category. In contrast, just 22.8 percent of Victoria’s female prisoners were versatile offenders; 52.7 percent were never convicted of anything more serious than public-order disturbances; 14.3 percent were convicted only of theft; 6 percent were convicted only of violent crimes; and 4 percent were convicted of other types of misdeed (see Table 2).10

Table 2

Criminal Career Characteristics by Different Types of Offender

variablecharacteristicnumber of womenversatile offenders (%)public order offenders (%)theft offenders (%)violent offenders (%)other offenders (%)
All offenders 6,036 22.8 52.7 14.2 6.1 4.3 
Number of convictionsa 1 conviction 2,572 58 21.8 11.4 8.8 
2–5 convictions 2,256 31.5 52.4 11.5 3.2 1.4 
6–9 convictions 623 49.6 45.7 4.3 0.2 0.2 
10 or more convictions 585 61 37.4 1.5 
Length of criminal careerb 3 years 4,360 11.6 55.8 18.6 8.1 5.8 
3–10 years 962 45.7 49.9 3.1 0.2 
10 years 714 60.2 36.8 2.1 0.4 0.4 
Age of onset of offendingc =18 years 589 23.1 59.1 14.6 1.2 
19–30 years 2,395 25.8 49.2 15.7 5.2 
30 years 3,010 20.4 54.1 13 7.7 4.9 
Location of offendingd Urban 4,249 23.3 52.4 14.1 6.5 3.7 
Regional 1,504 17 54.4 16 5.9 6.7 
Both 272 47.8 46 6.3 
Level of courte Summary court 5,063 20.3 62.6 9.8 2.3 
Higher court 602 6.1 52.8 17.6 23.4 
Both 359 86.4 11.7 1.7 0.3 
Number of prison infractionsf 0 prison infractions 4,357 17.7 54.7 15.1 7.5 
1–3 prison infractions 1,314 31.5 50.5 12.7 2.5 2.8 
4 or more prison infractions 365 52.6 35.9 8.8 2.2 0.5 
Number of aliasesg No aliases 4,301 13.2 56 17.1 8.2 5.4 
1 alias 982 39.9 47.1 10.1 0.9 1.9 
2–3 aliases 578 53.3 42.7 3.3 0.5 0.2 
4 or more aliases 175 62.3 34.9 2.9 
variablecharacteristicnumber of womenversatile offenders (%)public order offenders (%)theft offenders (%)violent offenders (%)other offenders (%)
All offenders 6,036 22.8 52.7 14.2 6.1 4.3 
Number of convictionsa 1 conviction 2,572 58 21.8 11.4 8.8 
2–5 convictions 2,256 31.5 52.4 11.5 3.2 1.4 
6–9 convictions 623 49.6 45.7 4.3 0.2 0.2 
10 or more convictions 585 61 37.4 1.5 
Length of criminal careerb 3 years 4,360 11.6 55.8 18.6 8.1 5.8 
3–10 years 962 45.7 49.9 3.1 0.2 
10 years 714 60.2 36.8 2.1 0.4 0.4 
Age of onset of offendingc =18 years 589 23.1 59.1 14.6 1.2 
19–30 years 2,395 25.8 49.2 15.7 5.2 
30 years 3,010 20.4 54.1 13 7.7 4.9 
Location of offendingd Urban 4,249 23.3 52.4 14.1 6.5 3.7 
Regional 1,504 17 54.4 16 5.9 6.7 
Both 272 47.8 46 6.3 
Level of courte Summary court 5,063 20.3 62.6 9.8 2.3 
Higher court 602 6.1 52.8 17.6 23.4 
Both 359 86.4 11.7 1.7 0.3 
Number of prison infractionsf 0 prison infractions 4,357 17.7 54.7 15.1 7.5 
1–3 prison infractions 1,314 31.5 50.5 12.7 2.5 2.8 
4 or more prison infractions 365 52.6 35.9 8.8 2.2 0.5 
Number of aliasesg No aliases 4,301 13.2 56 17.1 8.2 5.4 
1 alias 982 39.9 47.1 10.1 0.9 1.9 
2–3 aliases 578 53.3 42.7 3.3 0.5 0.2 
4 or more aliases 175 62.3 34.9 2.9 
a

χ2 (12) = 1942.7, p .001.

b

χ2 (8) = 1354.6, p .001.

c

χ2 (8) = 84.6, p .001.

d

χ2 (8) = 169.3, p .001.

e

χ2 (8) = 2754.4, p .001.

f

χ2 (12) = 347.3, p .001.

g

χ2 (12) = 968.2, p .001.

note 0 cells have an expected count less than 5. The “all offenders” variable includes every offender for whom conviction histories are known.

The sample also showed considerable fluctuations in the proportions of versatile offenders over time, suggesting that such patterns are considerably influenced by socio-historic context. Whereas only 8.2 percent of women entering Victoria’s prison system in 1864 would eventually commit different types of offense, this proportion reached a high of 77.6 percent in 1917. Although the percentage of versatile offenders rose and fell between different years, the overall trend in the sample was growth across time; from 1903 onward, the proportion of versatile offenders regularly matched or outstripped that of public-order-only offenders (see Figure 1).

This pattern of growth is likely a reflection of the strong association between versatile offending and recidivism, given that women who first entered the prison system in the early twentieth century were also prone to accumulating higher numbers of convictions than their nineteenth-century counterparts. Recidivism was a necessary condition of versatile offending; the exclusionary nature of the categories meant that a woman’s subsequent crimes had to cross categories for her to be a versatile offender. However, the sample also shows that the higher was the level of recidivism, the greater was the proportion of versatile offenders: 31.5 percent of low-level recidivists were versatile offenders, as opposed to 49.6 percent of mid-level and 61 percent of chronic recidivists (see Table 2). For instance, Elizabeth Turnbull, the most convicted woman in the study, had 188 convictions between 1910 and 1947, mostly for drunkenness and disorderly behavior; however, she also had several convictions for theft, though none for violence. The relationship between versatile offending and the overall length of women’s criminal careers was also clearly positive—unsurprisingly, since those who committed more offenses also offended for longer periods (see Table 2). Ellen Green, the woman with the longest criminal career—spanning fifty-two years from her first known conviction on May 25, 1860, to her last one on December 30, 1912—achieved a prolific fifty-eight convictions during this period, including one for assault, eight for larceny or pickpocketing, and forty-nine for various public-order offenses.

Versatile offenders did not comprise the bulk of the sample, but they did comprise a considerable proportion of the particularly troublesome offenders, as indicated by their cross-section of characteristics. Versatile offenders were over-represented among those who committed disciplinary infractions while imprisoned (see Table 2). Sarah Copas—convicted of eleven public-order offenses and four theft offenses between 1869 and 1891—held the record for the most infractions, thirty-six. Only 17.7 percent of those who maintained spotless prison records were versatile offenders, but more than 50 percent of the women who accumulated four or more infractions were versatile offenders. Frequent returns to prison likely increased the opportunities to misbehave, but a greater incidence of antisocial activity also intimates a reluctance to follow rules, whether in the outside world or inside prison. Versatile offending was also associated with a high number of alias identities in the records; 62.3 percent of the 175 women with four or more aliases committed different kinds of crime (see Table 2).

Criminologists have suggested that those who initiate offending at an early age and then persist in it show greater diversity in their offenses throughout their criminal careers. The distribution in age at first offense is different for the women in the sample than for modern female offenders; around 50 percent of the sample suffered their first conviction after the age of thirty, far later than in contemporary studies, although such studies typically show females as more likely than males to start offending later in life. However, versatile offenders were over-represented in the sample among those convicted at a younger age (see Table 2). In addition to a greater likelihood of being early-onset offenders, versatile offenders also tended to come from an urban environment. Women who offended in more than one category were under-represented among those convicted in regional courts (defined, for the purposes of this study, as courts outside Melbourne and its suburbs). They were, however, considerably over-represented among the 272 women convicted in both regional and urban courts (see Table 2)—47.8 percent of such prisoners—providing support for research linking offender mobility with greater diversity in and frequency of offending. Other risk factors associated with persistent offending for women, such as histories of substance abuse, homelessness, and sex work, likewise find expression in the significant proportions of versatile offenders among women convicted of drunkenness, vagrancy, and prostitution-related offenses (see Table 3).11

Table 3

Criminal Career Characteristics by Different Types of Offender

offense convictednumber of womenversatile offenders (%)mean number of total offensesmean number of public order offensesmean number of theft offensesmean number of violent offensesmean number of other offenses
Simple larceny 1,068 59.6 6.82 3.82 2.35 0.11 0.13 
Receiving 186 46.3 4.5 1.65 2.23 0.09 0.12 
Pickpocketing 177 73.4 7.46 4.66 2.05 0.11 0.21 
Burglary 108 83.3 13.64 9.82 1.94 0.52 0.66 
Robbery 59 71.2 4.61 2.2 1.73 0.22 0.1 
False pretences 54 25.9 5.41 1.44 3.81 0.04 0.04 
Homicide 94 18.1 1.78 0.5 0.09 1.03 0.05 
Nonfatal violence 456 62.3 7.45 4.88 0.55 1.34 0.18 
Arson 39 30.8 2.33 0.95 0.23 0.03 1.13 
Perjury 43 55.8 2.16 1.28 0.19 1.4 
Vagrancy 2,984 29.8 5.9 4.88 0.41 0.11 0.1 
Drunkenness 975 43 10.88 9.04 0.59 0.2 0.2 
Disorderly behavior 798 41.5 8.33 6.84 0.53 0.21 0.13 
Language 734 54.8 11.96 9.71 0.6 0.39 0.28 
Prostitution 424 43.9 8.71 7.19 0.64 0.11 0.29 
Breaking windows 86 95.3 17.7 14.72 0.8 0.47 0.98 
offense convictednumber of womenversatile offenders (%)mean number of total offensesmean number of public order offensesmean number of theft offensesmean number of violent offensesmean number of other offenses
Simple larceny 1,068 59.6 6.82 3.82 2.35 0.11 0.13 
Receiving 186 46.3 4.5 1.65 2.23 0.09 0.12 
Pickpocketing 177 73.4 7.46 4.66 2.05 0.11 0.21 
Burglary 108 83.3 13.64 9.82 1.94 0.52 0.66 
Robbery 59 71.2 4.61 2.2 1.73 0.22 0.1 
False pretences 54 25.9 5.41 1.44 3.81 0.04 0.04 
Homicide 94 18.1 1.78 0.5 0.09 1.03 0.05 
Nonfatal violence 456 62.3 7.45 4.88 0.55 1.34 0.18 
Arson 39 30.8 2.33 0.95 0.23 0.03 1.13 
Perjury 43 55.8 2.16 1.28 0.19 1.4 
Vagrancy 2,984 29.8 5.9 4.88 0.41 0.11 0.1 
Drunkenness 975 43 10.88 9.04 0.59 0.2 0.2 
Disorderly behavior 798 41.5 8.33 6.84 0.53 0.21 0.13 
Language 734 54.8 11.96 9.71 0.6 0.39 0.28 
Prostitution 424 43.9 8.71 7.19 0.64 0.11 0.29 
Breaking windows 86 95.3 17.7 14.72 0.8 0.47 0.98 

Yet, although we can definitely characterize the versatile female offenders in our sample as a more urban, early-onset, prolific, persistent, and disruptive subset of the female prison population, they were not uniform in their various crimes. The vast majority of the 1,376 versatile offenders were found guilty of public-order offenses—46.1 percent of public-order and theft offenses, 16.8 percent public-order and violent offenses, and 14.4 percent public-order and various other offenses. The next largest groups contained those who committed three distinct types of crime—5.9 percent public-order, theft, and violent offenses and 5.6 percent public-order, theft, and other offenses. Smaller proportions occupied the remaining offense-category combinations. Only 1.5 percent (twenty women) committed crimes in all four offense groups.

A closer look at specific offenses reveals even greater diversity in offending patterns, particularly regarding degrees of specialization and levels of recidivism (see Table 3). Although 50.8 percent of the women convicted of theft were versatile offenders, a breakdown of women within that category shows that the proportion comprised by versatile offenders ranged from 83.3 percent of the 108 women convicted of burglary to 25.9 percent of the fifty-four women convicted of false pretences, the most common type of fraud. The lower level of recidivism among women convicted of false pretences might be responsible for this lower level of versatility. However, although the mean number of total convictions for women convicted of false pretences (5.41) was lower than that for those convicted of some theft offenses, it was also higher than that for those convicted of others, like robbery (4.6), that had a large proportion of versatile offenders (71.2 percent). Moreover, the fact that women convicted of obtaining money or goods by false pretences had the highest mean average of theft convictions (3.8) suggests that the issue was not recidivism but a more specialized offending pattern.

In contrast, burglary, in addition to having the highest proportion of versatile offenders, also had the highest mean number of total convictions (13.64) for any theft offense, the primary cause being an elevated mean number of convictions for public-order offenses (9.82). Public-order offenses dominated the overall conviction records of women convicted of pickpocketing and larceny as well. The occasional thefts in such women’s records were probably connected to their more general participation in deviant sub-cultures and activities; scholars of prostitution note that the sex trade historically provided abundant opportunities for women to rob men in brothels and hotels, as well as on the streets.12

Unlike the case of theft, in which the most serious offenses (burglary, robbery, and pickpocketing) showed the highest proportion of versatile offenders, the kinds of violence most associated with mixed offending were the less serious ones. Of the 456 women convicted of nonfatal violence, 62.3 percent were versatile offenders. The average number of convictions among these women was 7.45, higher than that found in any theft offense except burglary. Contemporary findings tend to confirm that violent offenders commit more offenses than nonviolent ones. As in theft, however, most of the convictions that violent offenders accumulated appear to have been for public-order offenses; the mean average of public-order offenses among women convicted of nonfatal violence is 4.88.

The combination of recidivism and versatility was far lower among the ninety-four women convicted of homicide; their mean average of 1.78 total convictions suggests that most of these homicides were single convictions. Other historical studies usually construct homicides by women less as instances within a pattern of criminal activity than as aberrant episodes prompted by extreme circumstances—often a response to domestic violence or the stigma attached to unmarried motherhood. Contemporary literature likewise suggests that murders committed by women are generally one-time events linked to family relationships and/or bouts of mental illness. The few versatile offenders convicted of homicide in our sample were mostly convicted of one or two public-order offenses, such as drunkenness or vagrancy, following their incarceration, possibly indicating a difficulty in adjusting to life after an extended sentence. Only three women appear to have led a life of crime and disorder prior to their homicide convictions; Louisa Davis, the most persistent of the three, had twenty-six convictions between 1890 and 1911, one of which was for stabbing a female neighbor to death in 1891.13

Public-order offenses generally saw lower proportions of versatile offenders than did convictions for nonfatal violence or most kinds of theft. The 2,984 women convicted of vagrancy or insufficient means of support had the lowest proportion of versatile offenders of any public-order offense, 29.8 percent, as well as the lowest mean average of total convictions, 5.9—surprising results given the many women subjected to this charge, and the scholarly contention that officials often resorted to a charge of vagrancy to curtail prostitution in Victoria. If the vast majority of women convicted of vagrancy were prostitutes, the scrutiny that such activity received would likely have resulted in a higher recidivism rate than the one on record. Among the 423 women convicted specifically of prostitution-related offenses like soliciting, however, the average number of convictions was 8.71, the product of not only an elevated number of public-order but also theft offenses. Hence, the majority of vagrant women appear to have come from the ranks of the poverty-stricken rather than from a specific sub-group of prostitutes, or at least not from prostitutes who were chronically involved in crime or the delinquent street sub-culture associated with the trade. Charges of drunkenness, disorderly conduct, and obscene language rather than vagrancy were more likely the primary means of policing prostitution, since these categories showed far higher rates of recidivism and far higher proportions of versatile offenders. Women convicted of obscene language had a particularly high proportion of versatile and highly recidivist offenders, and were more likely than other offenders to amass convictions for violence.14

A miscellaneous form of property damage that deserves special attention is breaking windows or panes of glass; it has the strongest association with versatile offending (95.3 percent of the eighty-six women) and the highest average recidivism rate (17.7 convictions). An unusually high proportion of those who committed this offense also had convictions for theft and violent crime. In most instances, it was not linked, as might be expected, to attempted theft or burglary. Rather, it appears that breaking windows was a common form of vengeance among women from criminal sub-cultures. For example, as John Castieau, Melbourne’s Gaol Superintendent, noted in July 1870: “Joe Thompson the betting man the Proprietor of the Continental Cafe came up this afternoon to see a woman who in revenge for not being allowed inside the Cafe took up a stone smashed one of the large plate glass windows of the establishment; it seems that females are not allowed inside the cafe but that the waitresses who are dressed in fancy costumes are not any better than they should be consequently the outsiders have a down upon them their place of business which of course is frequented by most of the loose fish of Melbourne. Thompson is afraid the woman I spoke about will smash his windows again if she is released from prison he visited her in the hope of inducing her to go to New Zealand where she has friends.”15

Freeman, author of Lights and Shadows of Melbourne Life, likewise recorded that “smashing windows” was one of the favorite amusements of “larrikins” (youthful delinquents). The high correlation of this offense with serious and persistent offending is ironic, and potentially significant, given that one of the main arguments for policing minor or public-order offenses, as enshrined in the “broken windows” theory, was that such offenses produced criminal environments (discussed below).16

Broken Windows: Zero Tolerance or Labeling?

The tendency of historians to focus on specific types of offense has largely obscured the problems surrounding a significant minority of female prisoners, who comprised a majority of serious and recidivist offenders. The practice of analyzing crime and criminality within particular categories derives from the nineteenth century, when governments began annual compilations of crime statistics. Historians later adopted the classifications used in these reports, despite arguments that other coding would be preferable. Yet late nineteenth-century commentators were far from unaware of offenders’ propensity to mix minor and serious criminal acts, even drawing attention to particular subtypes—prostitutes robbing customers, vagrants breaking and entering, or drunks given to fits of violence. In their work on female offenders, Lombroso and Ferrero speculated that female “born criminals” were more prone to versatility than men, specializing “in not just one but several types of crime.” For instance, among prostitutes, such “mild crimes like theft, blackmail and stabbing” were supposed to be a “frequent occurrence.”17

Two criminological theories appear especially relevant to explaining the data in this article—the broken-windows theory (and its corresponding response, zero-tolerance policing) and labeling or social-exclusion theory. Wilson and Kelling maintained that the effect of unpoliced petty crime on a neighborhood can result in an escalation to more serious interpersonal infractions. In their view, physical incivility (like drawing graffiti, littering, or breaking windows) and behavioral incivility (like public urination, street prostitution, and rowdiness) created “no-go zones” for law-abiding members of society and hot spots for crime. The suggested response, especially popular in the United Kingdom and the United States since the 1980s, was a zero-tolerance policy, which involved a strict policing of petty crime and, to a lesser extent, neighborhood-rejuvenation programs. Australia did not embrace this style of policing to the same extent; offenses that once met with zero tolerance there—such as public drunkenness, obscene language, and disorderly conduct—have generally attracted more lenience since the 1970s. Unlike the zero-tolerance perspective, labeling and social exclusion theories maintain that criminals and non-criminals do not differ in any meaningful way. According to these theories, society’s response to deviants—especially as reflected in the media and the criminal-justice system but also in the larger community—is ultimately what determines whether miscreants will reform or become more deeply ensconced in crime.18

Did awareness of the potential for minor offenses to develop into more serious crime influence policing practices? Was the hard line taken against public-order offenses in Victoria during the late nineteenth and early twentieth century an anticipation of the “broken windows” theory? The increase in the rate of female imprisonment during the 1870s has generally been attributed to Melbourne’s crackdown on the rampant prostitution and “loose” conduct that arose during the mining boom of the 1850s and 1860s. Yet these concerns were also linked to more general fears about a growing “criminal class” in Melbourne guilty of practices much worse than moral peccadilloes. Efforts by authorities to “clean up” the city and make it more cosmopolitan and inviting to visitors and residents alike pre-dated the “broken windows” approach by almost a century. At any rate, according to McConville, the areas targeted by police shifted across the study period, from the central business district of Melbourne in the early years to its inner-city suburbs by the early 1900s. Zero-tolerance policing attempts appear to have been successful on the surface; women entering the system during the 1870s were likely to remain low-level recidivists convicted of public-order offenses only. However, there is no clear evidence that such women would have become problem offenders otherwise or that imprisonment was the determining factor in their desistance.19

Women who first entered prison because of violence or theft were the ones most likely to be over-represented among those who became versatile offenders (see Table 1). Rather than using public-order offenses to deter women from an escalation to more serious crimes, police might have used them to detain women who were serious offenders but difficult to prosecute accordingly. Nineteenth-century juries in Victoria—as in most jurisdictions then and now—were far less inclined to convict women than men. Instances appear in which police charged women with public-order offenses while also initiating more serious prosecutions against them, likely as backup should the felony charges fail. The knowledge that women with histories of theft and violence might continue committing such offenses with impunity may have encouraged police to harass these women after release. But re-arresting them more frequently and segregating them from the support of family and friends may instead have instigated these women to partake in more crime. Within the context of labeling theory, Tannenbaum refers to this cycle as the “dramatization of evil,” wherein the community seeking to prevent crime or morally suspect behavior actually creates an environment conducive to more of it occurring.20

The labeling and monitoring of offenders by the police during the late nineteenth century was facilitated in large part by the prison registers; apart from written descriptions of offenders, these files included photographs that helped officers to recognize former prisoners as they patroled their beat. Offenders commonly complained that this level of scrutiny impeded their ability to resume a normal life, all but guaranteeing their return to prison. Janet Dibben, one of the women in the sample, wrote a poem about her experiences in Melbourne gaol that alluded to the practice of ambitious policemen apprehending known offenders to increase their arrest quotas: “When you go out it’s now beware, / The bobbies are watching you everywhere; / It’s when you go out, and when you come in / They want a stripe, and that is the thing.” Dibben also wrote that it was not only police that singled former female prisoners out; family and friends also rejected them for their fall from respectability.21

Some contemporary studies show that the social exclusion of women due to criminal involvement, which exceeds that of men, may well contribute significantly to recidivism. Many female offenders feel impelled to conform to the label of social outcast thrust upon them, indulging in disorderly behavior that makes them easy prey for attentive police. Likewise, in Victoria, as the number of women imprisoned declined, those who were incarcerated might have become even more of an outcast group, thus accounting for the rising proportion of recidivist and versatile offenders toward the end of the period under study (see Figure 1). In that case, the best explanation for the over-representation of violence and theft offenders among versatile offenders may well be, broadly, the labeling or social-exclusion theories.22

A significant minority of all female prisoners mixed minor or public-order offending with the more serious crimes of theft or violence; such women comprised the majority of the most troublesome and recidivist of offenders. The results presented herein also show the considerable variation of longitudinal offending behavior evident within the broad categories of property, personal, and public-order offenses. This variation limits the usefulness of such groupings as a means to discuss offender “types.” The average woman convicted of vagrancy appears to have been on a vastly different criminal career trajectory from the one convicted of obscene language—not to mention the different outlooks for women convicted of murder compared to assault, or even burglary compared to false pretences.

These findings indicate that the traditional approach of examining crime and criminality predominantly through separate offense categories obscures important patterns, and that versatile offending warrants greater attention than it has currently received in criminal-justice history. The marriage of longitudinal criminological research issues to traditional historical methodologies, such as microhistory, offers fresh possibilities for exploring past criminal behavior. In particular, more longitudinal studies of the criminal careers of both historical and contemporary offenders are necessary to explain changes in the proportions of versatile offenders over time, and the relationship of such patterns to the incidence of crime committed by women—or men, for that matter—in society overall. Such research may have important implications for policing and sentencing strategies adopted to limit serious and persistent offending.

Notes

1 

Lucy Williams, Wayward Women: Female Offending in Victorian England (Barnsley, South Yorkshire, 2016), 117.

2 

For abortion, see Barbara Baird, “‘The Incompetent, Barbarous Old Lady Round the Corner’: The Image of the Backyard Abortionist in Pro-abortion Politics,” Hecate, XXII (1996), 7–26; Allen, “The Trials of Abortion in Late Nineteenth and Early Twentieth Century Australia,” Australian Cultural History, XII (1993), 87–99; Lyn Finch and Jon Stratton, “The Australian Working Class and the Practice of Abortion 1880–1939,” Journal of Australian Studies, XII (1988), 45–64; for infanticide, Kathy Laster, “Infanticide: A Litmus Test for Feminist Criminological Theory,” Australia and New Zealand Journal of Criminology, XXII (1989), 151–166; Allen, “Octavius Beale Reconsidered: Infanticide, Babyfarming and Abortion in NSW 1880–1939,” in Sydney Labour History Group (ed.), What Rough Beast?: The State and Social Order in Australian History (Sydney 1982), 111–129; for baby farming, Annie Cossins, The Baby Farmers: A Chilling Tale of Missing Babies, Shameful Secrets and Murder in 19th Century Australia (Crows Nest, 2013); Jude E. S. McCulloch, “Baby-Farming and Benevolence in Brisbane, 1885–1915,” Hecate, XXXVI (201), 42–56; for prostitution, Rae Frances, Selling Sex: A Hidden History of Prostitution (Sydney 2007); Kay Daniels, So Much Hard Work, (Sydney, 1984); Chris McConville, “The Location of Melbourne’s Prostitutes, 1870–1920,” Australian Historical Studies, XIX (1980), 86–97; for public-order offenses and women, Leigh S. L. Straw, Drunks, Pests and Harlots: Criminal Women in Perth and Fremantle, 1900–1939 (Knightsbridge, 2013); Piper, “‘A Growing Vice’”: The Truth about Brisbane Girls and Drunkenness in the Early Twentieth Century,” Journal of Australian Studies, XXXIV (2010), 485–497; idem, “All the Waters of Lethe: An Experience of Female Alcoholism in Federation Queensland,” Queensland Review, XVIII (2011), 85–97; Susanne E. Davies, “Vagrancy and the Victorians: The Social Construction of the Vagrant in Melbourne, 1880–1907,” unpub. Ph.D. diss. (Univ. of Melbourne, 1990). “Bushranging,” one of the few types of property crime to receive historically significant attention in Australia, is the prevalent nineteenth-century practice of robbing travelers or other targets (such as banks or homesteads) in rural areas and then retreating into the Australian wilderness or bush. The popular romantic view of bushranging as a form of “social banditry” has encouraged considerable scholarship about such thefts, predominantly perpetrated by men.

3 

Helen Johnston, Barry Godfrey, David Cox, and Jo Turner, “Reconstructing Prison Lives: Criminal Lives in the Digital Age,” Prison Service Journal, CCX (2013), 4–9; Godfrey, Cox, and Stephen Farrall, Criminal Lives: Family Life, Employment, and Offending (New York, 2007); idem, Serious Offenders: A Historical Study of Habitual Criminals (New York, 2010), 120; Helen Johnston, Godfrey, and Turner, “‘I Am Afraid She Is Perfectly Responsible For Her Actions and Is Simply Wicked’”: Reconstructing the Criminal Career of Julia Hyland,” in Anne-Marie Kilday and David Nash (eds.), Law, Crime and Deviance since 1700: Micro-studies in the History of Crime (London, 2017), 209–226; Marian H. A. C. Weevers, Margo Koster, and Catrien C. J. H. Bijleveld, “Swept Up from the Streets or Nowhere Else to Go? The Journeys of Dutch Female Beggars and Vagrants to the Oegstgeest State Labor institution in the Late Nineteenth Century,” Journal of Social History, XLVI (2012), 1–14; Weevers and Bijleveld, “Mad, Bad, or Sad? Dutch Female Beggars and Vagabonds Sent from the State Labor Institution to the State Mental Asylum at the Turn of the 19th Century,” Women and Criminal Justice, XXI (2014), 176–192; Piper, “‘I Go Out Worse Every Time’: Connections and Corruption in a Female Prison,” History Australia, IX (2012), 135.

4 

Marc Leblanc and Rolf Loeber, “Developmental Criminology Updated,” in Michael Tonry (ed.), Crime and Justice (Chicago, 1998), 115–198; Terrie E. Moffitt, “Adolescence-Limited and Life-Course Persistent Antisocial Behavior: A Developmental Taxonomy,” Psychological Review, C (1993), 674–701; Loeber and LeBlanc, “Toward a Developmental Criminology,” in Michael Tonry and Norval Morris (eds.), Crime and Justice (Chicago, 1990), 375–473; Robert Brame, Raymond Paternoster, and Shawn D. Bushway, “Criminal Offending Frequency and Offense Switching,” Journal of Contemporary Criminal Justice, XX (2004), 201–214; Brian Francis, Keith Soothill, and Rachel Fligelstone, “Identifying Patterns and Pathways of Offending Behaviour: A New Approach to Typologies of Crime,” European Journal of Criminology, I (2004), 48–87.

5 

People who went to “prison” on sentence had the full details of their convictions entered into this series. People given a sentence of twenty-four hours to a week in the police “lockup” attached to the station were not entered into the register. If they later incurred proper prison sentences, however, they would usually have these lockup convictions entered into their record under “previous history.”

6 

Felipe Estrada, Olof Bäckman, and Anders Nilsson, “The Darker Side of Equality? The Declining Gender Gap in Crime: Historical Trends and an Enhanced Analysis of Staggered Birth Cohorts,” British Journal of Criminology, LVI (2016), 1272–1290.

7 

Police Offences Statute 1865 (Victoria), s26; Mark Finnane, Punishment in Australian Society (Melbourne, 1997), 88; Shurlee Swain, “The Poor People of Melbourne,” in Graeme Davison, David Dunstan, and McConville (eds.), The Outcasts of Melbourne (North Sydney, 1985).

8 

Satyanshu K. Mukherjee, Evelyn N. Jacobsen, and John R. Walker, Source Book of Australian Criminal and Social Statistics 1804–1988 (Canberra, 1989), 286; Dean Wilson, The Beat: Policing a Victorian City (Melbourne, 2006)), 47–69, 190. Victoria’s female population increased by around 36% between 1861 and 1871, compared to just a 19% between 1871 and 1881.

9 

Piper, “‘I Go Out Worse Every Time.’”

10 

Jean Marie McGloin, Christopher Sullivan, and Alex Piquero, “Aggregating to Versatility? Transitions among Offender Types in the Short Term,” British Journal of Criminology, XLIX (2008), 243–264; Francis, Soothill, and Fligelstone, “Identifying Patterns.”

11 

LeBlanc and Loeber, “Developmental Criminology Updated”; Loeber and LeBlanc, “Toward a Developmental Criminology”; Moffitt, “Adolescence-Limited and Life-Course Persistent Antisocial Behavior”; Jeffrey M Ackerman and D. Kim Rossmo, “How Far to Travel? A Multilevel Analysis of the Residence-to-Crime Distance,” Journal of Quantitative Criminology, XXXI (2015), 237–262; Per-Olof H. Wikstrom, Urban Crime, Criminals, and Victims: The Swedish Experience in an Anglo-American Comparative Perspective (New York, 1991); Sally S. Simpson, Jennifer L. Yahner, and Laura Dugan, “Understanding Women’s Pathways to Jail: Analysing the Lives of Incarcerated Women,” Australian and New Zealand Journal of Criminology, XLI (2008), 84–108. For more information about the risk factors historically associated with female imprisonment and recidivism, see also Piper and Nagy, “Risk Factors and Pathways to Imprisonment among Incarcerated Women in Victoria, 1860–1920” (forthcoming).

12 

Piper, “‘Us Girls Don't Put One Another Away’: Relations among Melbourne's Prostitute Pickpockets, 1860-1920,” Women's History Review (forthcoming 2017), available now at doi 10:1080/09612025.2017.1321613.

13 

Loeber et al., “Constancy and Change in the Prevalence and Frequency of Offending When Based on Longitudinal Self-reports or Official Records: Comparisons by Gender, Race, and Crime Type,” Journal of Developmental and Life-Course Criminology, I (2015), 150–168; Piquero, “Frequency, Specialization and Violence in Offending Careers,” Journal of Research in Crime and Delinquency, XXXVII (2000), 392–418; Kathy Laster, “Arbitrary Chivalry: Women and Capital Punishment in Victoria, 1842–1967,” in David Philips and Susanne Davies (eds.), A Nation of Rogues: Crime, Law and Punishment in Colonial Australia (Melbourne, 1994), 166–186; Adam Louis Mahoney and Thanos Karatzias, “Violent Female Offending: An Exploration of Repeat and One-Time Offending,” International Journal of Forensic Mental Health, XI (2012), 191–202.

14 

Frances, Selling Sex.

15 

John B. Castieau, The Diaries of John Buckley Castieau (1855–1884), MS 2218, National Library of Australia; John Freeman, Lights and Shadows of Melbourne Life (London, 1888).

16 

George L. Kelling and James Q. Wilson, “Broken Windows: The Police and Neighbourhood Safety,” The Atlantic (March 1982), available at https://www.theatlantic.com/magazine/archive/1982/03/broken-windows/304465/.

17 

Finnane and Piper, “The Prosecution Project: Understanding the Changing Criminal Trial through Digital Tools,” Law and History Review, XXXIV (2016), 873–891; Cesare Lombroso and Guglielmo Ferrero, Criminal Woman, the Prostitute and the Normal Woman (Durham, 2004), 182–221.

18 

Although other criminological theories that aim to explain criminality from an individual’s perspective are applicatble (for example, general-strain, psychological, feminist, postmodern, cultural-criminology, Marxist theories, etc.), theories that work on a situational and social level can be more fruitful in a quantitative, longitudinal project like this one with a large cohort. Labeling theories have evolved into social-exclusion theories that are more nuanced; the original labeling theories could not explain initial criminal behavior, certain forms of property crime, nor labels in their useful form (for example, re-integrative shaming—labeling people to reform them). Kelling and Wilson, “Broken Windows”; Howard Becker, Outsiders: Studies in the Sociology of Deviance (London, 1963); Peter N. Grabosky, “Zero Tolerance Policing,” Trends and Issues in Crime and Criminal Justice No.102 (Canberra, 1999).

19 

McConville, “Location of Melbourne Prostitutes”; idem, “From ‘Criminal Class’ to ‘Underworld,’” in Graeme Davison, Dunstan, and idem (eds.), The Outcasts of Melbourne: Essays in Social History (Sydney, 1985), 69–90.

20 

Frank Tannenbaum, Crime and Community (Boston, 1938).

21 

Janet Dibben, Songs and Recitations; Written and Composed from Experience by a Lady That Has Travelled (St. Kilda, 1904), 8–10.

22 

Felipe Estrada and Anders Nilsson, “Does It Cost More to Be a Female Offender? A Life-Course Study of Childhood Circumstances, Crime, Drug Abuse, and Living Conditions,” Feminist Criminology, VII (2012), 196–219.