Desistance and Persistence: Women and Habitual Violence in Late Victorian and Edwardian England

Abstract

Historicized research on female habitual criminality provides insight into the interactions and struggles of marginalized women. Female recidivists, forming a large proportion of habitual offenders, prompted fears among Victorian and Edwardian contemporaries over the efficacy of the justice system. Recent research indicates that recidivism was a gendered process, women facing greater marginalization and stigmatization before, during and after offending than men, yet there remains little discussion focusing specifically on violent offending. This article examines violent female recidivism in the late Victorian and Edwardian period, focusing on women’s entries, desistance and persistence in violence and at how reoffending impacted their post-offending lives.

Keywords:

• Criminal justice system
• female offending
• habitual offending
• recidivism
• violent crime

INTRODUCTION

This article focuses on the lives and experiences of women who were violent habitual offenders in late Victorian and Edwardian England. During this period, crime was regarded a distinct, prominent threat which challenged contemporary ideals of social order, morality, self-restraint, and progress. Despite evidence of declining prosecution rates, the Victorian and Edwardian period was characterized by bursts of intense fear over crime and disorder; these so-called “moral panics” were highly sensationalized by the press, with notable examples including the widespread reportage on the violent street robberies and Jack the Ripper murders of the late-nineteenth century (Davis, 1980; Walkowitz, 1982). Habitual offending was also a major concern among Victorian contemporaries, an anxiety intensified by the fact that criminal statistics offered evidence of a rise in reoffending rates. In response to this emerging threat, the government passed a series of legislation aimed at penalizing habitual offenders (Habitual Criminals Act, 1869; Prevention of Crimes Act, 1871).

Crime was targeted not only through legislation but also by reforming the criminal justice system. At the entry point of law enforcement, the police were radically expanded both in officer numbers and in jurisdiction during the Victorian and Edwardian period; it also became compulsory for each county to establish a constabulary police force (County Police Act, 1839; County and Borough Police Act, 1856). The criminal courts, meanwhile, were a hierarchical system composed of three tribunals, with the Assizes and the Quarter Sessions at the top and middle levels handling serious, indictable offenses. The magistrates’ courts were the lowest tribunal and predominantly handled summary offenses, although the expansion of magisterial powers throughout the nineteenth century resulted in these courts becoming an increasingly important mechanism of the justice system (Bentley, 1998). The Victorian and Edwardian period also saw the introduction of new measures for handling juvenile offenders, including the development of reformatory and industrial schools in the 1850s and later the establishment of separate juvenile courts (Children Act, 1908). Moving beyond the courts, the Edwardian period introduced the foundations of the present-day probation service (Probation of Offenders Act, 1907).

Women then – as today – constituted only a small minority of those apprehended, prosecuted and/or convicted of criminal offenses, a phenomenon which scholars have attributed to a range of social, cultural and psychobiological factors (Arnot & Usborne, 1999; Emsley, 2005; Feeley & Little, 1991; Heidensohn & Silvestri, 2012; Heijden, 2016; Zedner, 1991). This gap was especially evident among offenses against the person, with violence being regarded an “overwhelmingly male enterprise” across “practically every historical setting” (Spierenburg, 1998); indeed, studies suggest that women accounted for just 5–20% of those charged with violent crimes across the modern period in England (Beattie, 1986; Eisner, 2003; Turner, 2009). However, women appear to have constituted a larger proportion of habitual offenders, at times even outnumbering male recidivists; Lucia Zedner’s research (1991) on Victorian England suggests that there were more female habitual offenders with over ten convictions than there were males. As such, female habitual offenders were regarded by late-nineteenth and early-twentieth century contemporaries to be especially problematic and irredeemable, with one contemporary newspaper stating that there was “some truth in the common belief that women who have once adopted a criminal life are less likely to be reclaimed from it than men.”

This article contributes to discussions on female habitual offending by examining the experiences of women prosecuted for multiple violent crimes at late Victorian and Edwardian magistrates’ courts. Drawing on historic newspapers, court records and census returns, it explores cases which occurred across two South West localities – Cheltenham and Exeter – during the years 1880–1910. Three lines of enquiry are followed: whether the backgrounds of female habitual offenders differed from those of non-reoffenders; which factors influenced desistance from (or persistence in) habitual violence; and, finally, whether female habitual offenders faced different life-course trajectories than non-reoffending women. Ultimately, it demonstrates the ways in which habitually violent women experienced difficulties in forming and maintaining social connections during their pre- and post-offending lives.

LITERATURE REVIEW

Influenced by the ‘life-course' approach adopted by criminologists, historians of crime have paid greater attention to recidivism over the last two decades. Lucy Frost and Hamish Maxwell-Stewart (2001), for example, drew on archival materials to examine recidivism and criminal trajectories within the context of convict transportation. Subsequent research by Godfrey et al. (2007, 2010) offers insight into the lives of habitual criminals in nineteenth- and twentieth-century England, demonstrating that offenders with stronger social connections or greater community integration had better chances of desisting from crime. Historians have also focused on the experiences of so-called “juvenile delinquents” (habitual youth offenders), with Godfrey et al. (2017) uncovering the life-course histories of youths who were institutionalized at reformatories, industrial schools and training ships in nineteenth- and twentieth-century England and ultimately emphasizing the impact of social care and skills training on children’s prospects of reforming.

A handful of studies have drawn specific attention to the issue of female habitual offending. Focusing largely on women’s experiences in the nineteenth and twentieth centuries, discussions extend across a range of jurisdictions including the UK, wider Europe, north America, and Australia (Boritch, 2005; Fingard, 1984; Piper & Nagy, 2017; Turner, 2009, 2011; Turner & Johnston, 2015; Vikström, 2011; Weevers & Bijleveld, 2014). A key focus within these studies is the extent to which contemporary discussions of the “criminal class”– regarded by nineteenth-century contemporaries as an amoral, unrespectable and distinct group of offenders – bore resemblance to reality. By examining the backgrounds of reoffenders, studies have generally disproved the existence of a criminal class yet indicated that habitual criminals fell within a distinct category of offenders often positioned in the lowest societal strata. For example, in a study of recidivism in late nineteenth-century Ontario, Helen Boritch (2005) argues that there was not a “dangerous or professional criminal class” but a “criminalized group consisting primarily of those at the bottom of the social hierarchy.”

Another focal point in historical studies is the controls – both informal social pressures and the formal criminal justice system – which influenced male and female desistance from crime. Evidence across jurisdictions indicates that family formation was less influential over female reoffenders than their male counterparts, many women having been married with children at the onset of offending. Evidence from Victorian and Edwardian England instead suggests that factors such as obtaining employment, securing stable accommodation, and having strong support networks were more strongly associated with female desistance (Godfrey et al., 2007; Turner, 2009; Turner & Johnston, 2015). In addition, there has been some research on the post-offending experiences of female recidivists; Lotta Vikström (2011) demonstrates that, for young female recidivists prosecuted in nineteenth-century Sweden, being stigmatized as a “criminal” had important repercussions on women’s lives after criminal desistance, including their life expectancies, relationships prospects and living circumstances.

Historical and criminological research on female habitual offending has tended to examine the full spectrum of women’s criminality rather than focusing specifically on habitual offending within one category of crime, thereby resulting in a lack of research on the nature of women’s habitual violence. Offering an exception, one criminological review of patterns in habitual violence has argued that women are not only “outstripped” by men as violent reoffenders but are also more likely to desist from violence (Weiner, 1989). However, there remains a dearth of research on violent female reoffenders, leaving questions regarding the onset, cessation, and aftermath of habitual violence by women in the past largely unanswered.

SOURCES AND METHODS

This article offers a primary study on violent female recidivism in late Victorian and Edwardian England. However, it was necessary to impose certain boundaries on the scope of this research. First, the research period 1880–1910 was selected not only for framing the final phase of the late Victorian and Edwardian period but also because it allowed the examination of a complete series of court records (available from 1880). Second, the Cheltenham and Exeter courts were chosen for two reasons: first, the excellent survival of registers and minute books for each court; second, the lack of research which focuses on female crime in the southwest of England, with only a handful of exceptions being available (Di Méo, 2021; King, 2006; Warner et al., 2008; Williams, 2000). Habitual offenders – also termed throughout as “reoffenders” and “recidivists” – are defined as those who were convicted of three or more violent crimes. Finally, the study focuses predominantly on acts of assault, then the most common form of violent crime, although more severe cases such as aggravated assault and GBH have been included. No acts of homicide by women feature in this study since no such cases were identified during the research.

The study was informed by a range of archival, literary and governmental primary sources. Archival court records for Cheltenham and Exeter magistrates’ courts, including registers and minute books, provided details such as the names of perpetrators and victims, offense classifications, verdicts, sentences, and victim statements. Insight was also offered through local and regional newspapers available online via the British Newspaper Archive: the primary newspapers used for the study include Western Times, Cheltenham Chronicle, Gloucester Citizen, Gloucestershire Echo, Cheltenham Mercury, Trewman’s Exeter Flying Post, and Exeter and Plymouth Gazette. These newspaper sources not only reported trial proceedings across the magistrates’ courts, Quarter Sessions and the Assizes but also provided rich descriptions of court statements made by defendants, victims, witnesses, magistrates, and judges. Finally, official census returns, available online via Ancestry, were used to obtain sociodemographic information both before and after offending, including women’s addresses, occupations, ages, birthplaces, family members, and other living arrangements.

These sources were used to create two datasets of violent female offending in Cheltenham and Exeter during the years 1880–1910: the first related only to habitual offenders (those convicted of three or more violent crimes); the second contained a random sample of fifty non-reoffending women (those convicted of only one violent crime), a control group used for contextualizing the experiences of recidivists. Data collection processes began by using court records to identify every female assault prosecution in Cheltenham and Exeter during the research period. Repeat offenders were then identified from this case list by checking overlapping information; simply sharing a name is not enough to confirm that two defendants represent the same woman. To be considered a “match,” defendants first needed to share the same/a similar name or to have used a known pseudonym or married name; they also needed to share at least two of the following sociodemographic characteristics: address; marital status; occupation; age; or names of family members. This process enabled the creation of two datasets for known female recidivists and non-recidivists, both being subsequently expanded when newspapers and censuses were used to obtain additional data on women’s backgrounds, experiences, and offending trajectories. Newspapers were also consulted to confirm whether any recidivists faced additional charges across regional magistrates’ courts, Quarter Sessions and/or the Assizes. Together, these data made it possible to recreate life-course histories for each habitual female offender and thus to follow women through the onset, duration and termination of their offending careers.

REOFFENDING PATTERNS AT CHELTENHAM AND EXETER

During the years 1880–1909, there were 1,507 assault prosecutions against women at the Cheltenham and Exeter magistrates’ courts. Approximately 1,040 individual women appeared as defendants in these cases. Only a small proportion – just 47 women (4.5%) – were categorized as recidivists. Despite their minority, reoffenders were responsible for a significant portion of women’s assaults, acting as defendants in 258 of the 1,507 (17.1%) prosecutions at Cheltenham and Exeter. These figures are lower than those identified in other historical studies of female recidivism: in a study of female petty offending in late Victorian Staffordshire, Jo Turner (2009) found that recidivists were responsible for almost half of all prosecutions, whereas Suki Haider (2013) has demonstrated that roughly 60% of female petty offenders prosecuted in late nineteenth-century Dundee were reconvicted. This discrepancy may be due to those studies having focused on a fuller range of crimes rather than solely on violent offenses, thus suggesting that female reoffending was less prevalent among crimes against the person than other types of crime.

Offending patterns among violent female reoffenders varied. The average gap between assault prosecutions was almost three years, while the average number of prosecutions and convictions was five and four respectively. A recidivist in Cheltenham had the highest number of prosecutions, being brought before the magistrates nineteen times for assault (but convicted only six times), whereas an Exeter woman convicted of assault seven times received the highest number of convictions. The average length of women’s offending careers at both Cheltenham and Exeter was eleven years, but the range was large; the longest offending record in this study spanned twenty-seven years, whereas the shortest offending career belonged to a woman whose four prosecutions all occurred in the same year.

Alongside acts of assault, most female recidivists were summoned for other offenses. A substantial proportion (70.2%) were prosecuted for other petty crimes, a high figure compared with the 20.8% of non-recidivists similarly prosecuted. Typically, these charges were public order offenses such as drinking, obscene language, wilful damage, and obstruction. Some women had considerable offending records: for example, one Cheltenham reoffender received six charges for assault, seven for drinking, four for obscene language, two for fighting, and one for causing an obstruction in the street. In addition, several recidivists were indicted for more serious offenses at either the Quarter Sessions or the Assizes: one recidivist and her husband were prosecuted at the Assizes in 1890 for malicious wounding, while another reoffender and her partner were prosecuted at the same court for perjury in 1906. As such, although violent recidivists constituted only a minority of women charged with assaults, they could pursue particularly lengthy offending careers and sometimes transgressed from committing acts of petty, non-indictable crimes to more serious offenses that were triable by a judge and jury.

SOCIOECONOMIC BACKGROUNDS AT THE ONSET OF OFFENDING

Criminologists have indicated that female recidivists are more likely than non-recidivists to face economic deprivation, unemployment, poor housing, substance abuse, and victimization prior to offending (Giordano et al., 2002; Gunnison, 2014; Huebner et al., 2010; Sommers et al., 1994; Uggen & Kruttschnitt, 1998). Similarly, historians have highlighted the marginalization of habitual offenders: in a study of female petty crime in Victorian Dundee, Haider (2013) denies the existence of a “criminal class” yet stresses that reoffenders “very often shared characteristics particular to the most marginalized sections of the working class” since they were “unskilled and inconsistently employed [and] were impoverished and…routinely abused alcohol”; similarly, Alana Piper (2012) argues that nineteenth-century Australian female recidivists constituted a “social underclass”.

There were some notable similarities in the backgrounds of the recidivists and non-recidivists in this study. First, women in both groups were of like ages at the onset of offending. The average age when first prosecuted for assault was 33 for recidivists and 34 for non-recidivists, although a large portion of all defendants were aged 16–30. The oldest of the recidivists and non-recidivists were 71 and 76 respectively when convicted of their first (recorded) offense, both therefore appearing to have had a particularly late onset. Meanwhile, the youngest defendant in the study was a 15-year-old non-reoffender who received magisterial lenience with the minimum fine of one shilling due to being “a very young woman” whose parents were “not very well off.” Similar results are presented in other historical studies: almost half of the recidivist women in a study of nineteenth-century Australia were under 30 before their first incarceration (Piper & Nagy, 2017), although research on Victorian and Edwardian Staffordshire suggests that most female reoffenders were aged 22–40 at their first prosecution (Turner, 2009). However, research suggests that female recidivism may be triggered at an earlier age in recent times than it was in the past, with contemporary studies indicating that modern-day women often begin offending in their late teens (Flood-Page et al., 2000; Graham & Bowling, 1995; McIvor et al., 2004).

Most recidivists (51.2%) and non-recidivists (62%) were local women born in either Cheltenham or Exeter. An additional third of recidivist and non-recidivist women were born elsewhere in the counties of Gloucestershire or Devon, often in towns or villages which neighbored Cheltenham or Exeter. Reoffenders were the most likely to be born outside of the county, although the proportion of women in this category was low (7.3% of recidivists compared to 2% of non-recidivists). An additional two women – one reoffender and one non-reoffender – were born outside of England, both of whom were Irish. As such, few women were “outsiders” to the areas in which they were prosecuted, most recidivists and non-recidivists having been “local” women born in Cheltenham, Exeter or surrounding areas.

Yet there were discrepancies in living circumstances. Although most defendants had lived in poorer areas of their towns, recidivist women faced less favorable residential settings. Non-recidivists more frequently lived in private family houses at the time of their first prosecution (80.4% of non-recidivists compared to 58.1% of recidivists). Reoffenders, meanwhile, were the most likely to live in shared lodgings: almost a third of these women – compared to 13% of non-reoffenders – lived in houses shared with other families or individuals, a situation which could entail renting apartments in a crowded house occupied by upwards of thirty people. They were also over twice as likely as non-recidivists to live in lodging houses. Temporary lodgings were more common among single reoffenders but also applied to women with families: one recidivist, for example, had been living in a lodging house with her unemployed husband and their six children prior to committing her first assault.

Disparities in marital conditions are also evident. Habitual offenders were the more likely to be unmarried at the onset of offending: 21.3% of recidivists compared to 8.6% of non-recidivists were unmarried when first appearing in courts. Among the married women, recidivists appear to have experienced – or at least reported – domestic abuse more frequently than non-recidivists, since 26.3% of married recidivists compared to 9.8% of non-recidivists prosecuted their husbands for assault during the research period. Many married reoffenders had long records of victimization and prosecuted their husbands several times for abuse. A Cheltenham recidivist summoned her husband six times for assault, first prosecuting him in 1880 when appearing in court with a bandaged head and testifying that he had “beat[en] her and kicked her several times” and “broke[n] all the things in the house.” After numerous subsequent assault prosecutions against him in the 1890s, the woman had separated from her husband by 1901, when she was supporting herself and their two daughters through her earnings as a hawker. Recidivist women also more frequently committed violence with their husbands: during the research period, 17.1% of recidivists compared to 11.8% of non-recidivists were prosecuted for acts of assault alongside their husbands. A recidivist woman and her spouse were convicted together for assaults on three occasions: reportage on the first case detailed how the “lively pair” had assaulted the woman’s arresting officer, while accounts of their final conviction described them as a “rough-looking couple” who had drunkenly assaulted their neighbor. Another recidivist was also convicted three times for committing assaults with her husband, one case having occurred when he “took her part” in a street fight.

Approximately 70% of both recidivists and non-recidivists had had children by the onset of offending. This suggests a greater prevalence of illegitimate children among recidivists, since they were more likely to have had children out of wedlock. Records of assaults by recidivists who were single mothers suggest the struggles of raising children with little financial assistance. When prosecuted for assault in 1901, a Cheltenham woman hoped that the Bench would be lenient since she was a mother of eight with “no one to support her.” Another recidivist brought three affiliation charges – then the legal mechanism for a woman to summon the father of an illegitimate child for failing to pay child support – against a former partner, stating in one hearing that he “ought to give her something” since she “could not afford them [the children].” Similarly, in a study of nineteenth-century Staffordshire, Turner (2009) argues that female habitual offenders were sometimes “abandoned” women whose partners had broken marriage promises and left them in pregnancy or with children to support.

One of the most apparent differences in the backgrounds of recidivists and non-recidivists was their apparent relationships with alcohol. Evidence of alcohol (ab)use was more prevalent among reoffenders; 25.5% of the recidivists in this study compared to just 2.1% of the non-recidivists were either alleged alcoholics or received multiple drinking charges. Some women had long histories of alcoholism, resulting in their stigmatization by police, magistrates, and the press. A Cheltenham recidivist was prosecuted seven times for drunkenness over the years 1882–1900 alongside numerous assault convictions. One report stated that she had been before the Bench “a great number of times” for drunkenness; at another hearing, the magistrates warned that “if she came there again she would be dealt with under the Habitual Drunkards Act.” Convicted thirteen times for drunkenness, another recidivist was described as a woman of “drunken habits and bad character”; a subsequent court appearance led the magistrates to lament that she had “received help from the ladies of the city” but had “spent her money in drink [and] left her children without food or clothes.” Taken together, then, the pre-offending histories of habitually violent women indicate that they may have experienced not only a greater degree of socioeconomic marginalization but also a greater reliance on alcohol than non-reoffending women.

DESISTANCE FROM VIOLENT CRIME: SOCIAL FACTORS

Most female habitual offenders in this study – 41 of the total 47 women in the recidivist cohort – eventually desisted from crime. This section of the article explores some of the social factors which may have played a role in influencing desistance among women, focusing in particular on whether women reached the stage referred to by criminologists as “primary desistance” (i.e. the cessation of offending, whereas secondary desistance refers to the adoption of a non-offending identity and tertiary desistance as a changed sense of belonging [Maruna & Farrall, 2004; McNeill, 2016]). Discussion will focus on whether either the “maturational” or “social bonds” theories—as outlined below—are applicable to the experiences of Victorian recidivists. Factors related to age, family life, residency and employment will be examined, although numerous influential controls identified within modern criminological studies have been excluded due to the limits associated with historical data, including the role played by mental health, individual agency and/or the completion of education (then largely inaccessible and unused by working-class girls due to the ineffectualness of the Elementary Education Act [1880]).

Explanations for desistance from crime began with the “maturational” theory put forward by Glueck and Glueck (1950), in which it was argued that, since offenders eventually “grow out” of crime with age and typically reduce or terminate their offending by the early 20s (the “age-crime curve”), desistance occurs as a natural result of the aging process. Criminologists now broadly agree that age is insufficient alone in fully explaining desistance, although it is held that maturation coincides with important “turning points” in the life course which can influence desistance. This has given rise to the “social bonds” theory, whereby these “turning points” – including obtaining employment, becoming a parent, and entering a serious relationship – allow offenders to build social capital and responsibility and thereby tie them to informal systems of social control (Barry, 2006; Carlsson, 2012; Sampson & Laub, 1993; 2003; Sommers et al., 1994). However, while these controls are most strongly associated with certain stages of the aging process, criminologists maintain that they remain important across the life trajectory. It has also been argued that individual choice, agency, and decision-making processes are integral to the process of desistance (Farrall & Bowling, 1999; Maruna, 2001). Taken together, it is typically held that “familiar themes of maturation, transitions, changed lifestyles, and relationships” all play a role in influencing criminal desistance (McIvor et al., 2004).

Criminological research on women’s desistance has grown in recent decades (Rodermond et al., 2016). These studies highlight a range of factors which can impact on women’s propensity to desist, although from the available evidence – most research focusing on the US, UK, continental Europe and Oceania – there appear to be some cultural and social differences in the effects of such factors. Evidence on the effects of marriage, for example, remain inconsistent: while numerous US-based studies suggest that marriage can promote a reduction and/or termination in offending, other studies, including further evidence from the US and the Netherlands, have found no link between marriage and desistance (Gunnison, 2001; Huebner et al., 2010; Zoutewelle-Terovan et al., 2014). Impacts may differ according to women’s sociodemographic backgrounds; one US study, for example, found that marriage was less strongly associated with desistance among African American women (Leverentz, 2014). The onset of motherhood has also received notable attention: although studies across Scotland, the US and Australia support the notion that motherhood can reduce female offending, evidence from the Netherlands and New Zealand indicates that childrearing can cause stress and thereby exacerbate women’s offending (Barry, 2006; Huebner et al., 2010; McIvor et al., 2004, 2009; Taylor, 2008; Uggen & Kruttschnitt, 1998; Zoutewelle-Terovan et al., 2014). Sociocultural differences may play a role here: one US-based study demonstrates that motherhood is more likely to have a positive impact on women from “advantaged” backgrounds, whereas another indicates that becoming a parent is more closely linked with desistance among white women than minority ethnic women (Giordano et al., 2011; Katz, 2000). More consensus has been reached regarding the impact of employment on women’s desistance, with studies across a range of jurisdictions finding that securing work is linked with women’s reduction and/or cessation of crime (Carlsson, 2012; Gunnison, 2001; Healy, 2012; Huebner et al., 2010; McIvor et al., 2004; Taylor, 2008; Verbruggen et al., 2012). Some exceptions are apparent: Monica Barry’s study (2006) on youth recidivism in Scotland, for example, stressed that women showed a propensity to desist even without employment, whereas Spencer De Li and Doris MacKenzie’s US-based study (2003) highlighted that employment can provide criminal opportunities for women and thereby encourage further crime. The generally view, however, is that employment can be effective for women, especially if they have regular working patterns and stable contracts.

How far, then, do “maturational” and “social bonds” theories apply to the experiences of female recidivists in the Victorian and Edwardian period? Evidence from Cheltenham and Exeter suggests that the “aging process” alone cannot account for women’s desistance. In the first place, the recidivists in this study were not particularly young at the onset of offending, the average age of recidivist offenders being 33 years when first convicted of assault. Second, their ages at desistance were (naturally) higher, with the average age at last recorded conviction being 43 years old. Criminological explanations that recidivists tend to “grow out” of their offending activities during their early 20s are therefore not applicable within this historical context.

It is possible that socioeconomic changes played a greater role in influencing desistance among these recidivist women. As aforementioned, one such factor attributed to desistance is the process of securing employment, an effect which may be triggered not only by changes in financial security, physical surroundings and maturity but also by a reduction in free time available for committing crime. However, obtaining employment appears to have had little impact on violent female reoffenders at Cheltenham and Exeter, since only a handful of women – just five – gained employment during their offending careers and all continued to offend. One Exeter woman had been unemployed and living with her husband during her first assault conviction, yet her subsequent employment as a laundress did little to encourage desistance: by that point, she was separated from her husband with seven children to support – a circumstance which possibly contributed to her criminality – and her continued offending resulted in two prison sentences for assault in 1901 and 1906.

If a change in occupational status failed to have an impact, moving to a new community was less influential still. Indeed, relocating to other neighborhoods may have exacerbated women’s offending; women typically moved to and from some of the poorest districts of their towns, where many were repeatedly involved in neighborly rows over territory, space, and children. Twenty-five of the forty-seven recidivists in the study moved house after the onset of offending, the majority of whom (twenty-two in total) reoffended. One Cheltenham woman moved six times during her offending career (1882–1894), committing at least one assault at each address, whereas an Exeter recidivist moved between the same two districts four times and reoffended each time. The remaining three women who relocated but desisted from crime may have followed different trajectories because they were moved from their communities into institutions. One woman had been summoned for several assaults in the years 1881–1894, during which time she had been living with her husband, a railway worker, and their three children. She was moved to a so-called “lunatic asylum” shortly after her last offense, at which she remained a patient for over a decade. She committed no further offenses either during or after her time at the asylum; her confinement likely resulted in her isolation from society and thus not only decreased her opportunities to offend but also lessened the external influence to do so.

Family formation, meanwhile, had a mixed impact on desistance from crime. Existing criminological studies suggest that, although motherhood can give women a sense of responsibility and impress notions of desistance, it can also exacerbate women’s offending by increasing financial and emotional stress (Giordano et al., 2002; Huebner et al., 2010; Katz, 2000; McIvor et al., 2004; Taylor, 2008; Zoutewelle-Terovan et al., 2014). Becoming a parent does not appear to have influenced desistance among violent female reoffenders in Cheltenham and Exeter: the four women who gave birth to their first child after the onset of offending continued to commit violent crime. The offending history of an Exeter woman offers insight into this issue. She made her first court appearances in 1880 and 1881 for assaulting neighbors, at which time she was unmarried without children. By 1885, she had given birth to two illegitimate children and was prosecuted for neglecting them: expressing that she was “very sorry for what had happened,” she asked the magistrates to “overlook” her conduct since she “had no work and could not maintain her children.” Subsequently becoming unemployed and moving in with her mother, she was convicted a further four times for child abuse and continuously testified to her inability to feed or clothe the children.

Marriage may have had a more positive influence over female recidivists. Although most of the recidivists had been married prior to offending, four of the five recidivists who became married after the onset of offending did not reoffend after marriage. The woman who persisted after marriage may have continued offending due to marrying a violent husband with whom she committed several assaults, a phenomenon referred to by criminologists as “negative assortative mating” whereby coupling with a violent partner can encourage violent behavior in the other party (Becker, 1981). For those who desisted, marriage may have offered a support network and financial stability. Consider, for example, the case of a Cheltenham recidivist who, as a single woman, was prosecuted ten times for assault during the years 1882–1892. Her offenses included several assaults on the father of her illegitimate child, a man who left her during pregnancy and provided little financial support: she told the magistrates during one prosecution that, since the father’s child support was “inadequate to what she had lost and suffered,” she would “wait…for him to make a home for her.” After two subsequent prosecutions for assaulting him, she appears to have desisted from crime when the couple reached a reconciliation; she married the father shortly after her last assault prosecution (1893) and was never brought back to court.

Meanwhile, just as forming marriages could influence a cessation of offending, the loss of a partner may have prolonged women’s offending careers; two women became widows after the onset of offending, both of whom continued to commit acts of violence. Indeed, widowhood could exacerbate recidivist behavior since women could face particularly difficult circumstances when “left alone after having someone to provide for them, particularly if they had children who were still dependent on them” (Turner, 2009). Marital separations may also have contributed to offending; four of the five recidivists who separated from their husbands were subsequently summoned for assault.

DESISTANCE FROM VIOLENT CRIME: THE CRIMINAL JUSTICE SYSTEM

Contemporaries became increasingly concerned by the efficacy of the criminal justice system as the nineteenth century progressed. A rise in reoffending rates resonated with fears that prisons were not only failing to reform criminals but also acting as “breeding grounds” whereby new offenders became influenced by members of the “criminal class” (Williams & Godfrey, 2015). Historians have similarly questioned the utility of the Victorian prison system: while some argue that the short nature of custodial sentences provided little deterrence, others suggest that incarceration merely resulted in convicts being released to the original conditions under which their crimes had been committed and thus failed to disincentivize re-offending (Weiner, 2008; Turner, 2011).

The following discussion on the impact of penal policies on female desistance from crime is based on the eventual rather than initial punishments received by women. Of the 258 prosecutions against recidivists in this study, 177 resulted in conviction and either a financial or custodial punishment (fines being considered the lighter option). Although the magistrates issued financial sentences in most cases, many women were unable or unwilling to pay their fines and, in default of not making payment, were incarcerated. Evidence from court minute books and newspapers indicates that, although financial penalties were initially issued in 81.4% of these assault convictions, in reality only 42.4% of cases resulted in women actually paying fines. Women who underwent prison sentences thus went to jail either because they had been issued with a custodial sentence outright or because they had been sent there in lieu of paying a fine.

From a top-level examination, financial sentences fostered little deterrence among violent female reoffenders. Cheltenham and Exeter women who paid fines were often charged for subsequent assaults; over 89% of recidivists’ convictions which resulted in fines were followed by at least one subsequent prosecution. However, the reoffending rate varied by women’s personal circumstances. Married recidivists had a higher reoffending rate after paying fines than their unmarried counterparts (91.5% compared to 70%); possibly married women had stabler financial circumstances and were therefore able to rely on husbands to pay their fines. One recidivist was convicted three times for assault and on each occasion paid the imposed fine. During the second hearing, her husband agreed to pay her fine when telling the magistrates to “settle it which way [they] like[d].” Another woman opted for a prison sentence over a five-shilling fine for an assault in 1894, but her husband ultimately paid the fine; she was subsequently convicted for two further acts of assault later that same year.

Sources indicate that fines were more penalizing of unmarried defendants. Their greater reluctance to reoffend after a fine may have been due to some recidivists being sole earners – at a time in which women’s wages were typically piecemeal – and lacking the support of a male breadwinner’s salary (Vincent, 1991). An unmarried dressmaker from Cheltenham was unable to pay fine for two assault convictions and faced short prison sentences in default; however, she had been able to pay her one-pound fine for a third conviction and was not brought back before the magistrates again. An even larger proportion of unmarried women desisted after a fine if they had been unemployed, likely a result of their financial circumstances rendering financial sentences particularly taxing.

Custodial sentences appear to have been a greater deterrent. Whilst the reoffending rate after a fine was 89%, only 51.5% of convictions resulting in a prison sentence were followed by another assault prosecution. Many recidivist women paid numerous fines in succession but desisted after their first experiences of incarceration. A Cheltenham recidivist, for example, continued to offend after paying four fines for assaults but ultimately desisted after undergoing her first (and only) prison sentence. Like financial penalties, the impact of custodial sentences varied by women’s backgrounds. Age may have influenced women’s responses to incarceration; reoffending rates after custodial sentences were lower among older recidivists, only 20% of those aged 50 or over compared to 54.5% of those aged 16–30 having reoffended after prison. Several recidivists pleaded with magistrates due to their age: on receiving her third assault conviction, one Cheltenham recidivist pleaded with the magistrates on account of being a “poor, lone, old woman” and hoped they would reconsider her sentence if she promised not to reoffend; although her request was declined and she was sent to prison for one week with hard labor, she was not subsequently brought back before the Bench. Familial ties may also have affected women’s reactions to custodial sentences. Married reoffenders living with their husbands were the least likely to desist after incarceration: only 49% of these women reoffended after prison sentences compared to 58.3% of unmarried recidivists and 75.1% of those who were separated. Their reluctance to reoffend after incarceration may have been due to the prospect of spending time away from families; one woman was fined three times for assault but, on receiving her first custodial sentence in 1898, she allegedly “burst out crying and pleaded to say goodbye to her husband and little girl” before going to prison. She was not brought to the magistrates’ court again.

Although prison sentences appear to have been more effective than fines, it must be noted that some women continued to offend after repeated prison terms – and some women even chose prison over a fine. A prolific Cheltenham offender opted for a prison sentence rather than a fine, stating to the Bench that she “w[ouldn’t] pay a half-penny [and would] go down” and that her children “may go to the workhouse.” Another recidivist, sentenced to either a fine of five shillings or a week with hard labor, told the magistrates that she “chose to do the seven days” because she had “nowhere else to go.” Although uncommon, such cases indicate that some recidivists were unaffected by the local prison system and may have been caught in a cycle of offending and incarceration.

PERSISTENT FEMALE OFFENDERS

Criminological evidence indicates that, while most recidivists eventually desist from crime, there is a group of “persistent” offenders in every criminal population who resist both informal social controls and the more formal controls of the criminal justice system. At Cheltenham and Exeter, this group of persistent offenders – defined in this study as those who perpetrated violent crime until within two years of their deaths – was very small. Out of the forty-seven recidivists under examination, only six were classified as persistent violent offenders. These offenders differed from other recidivists: their offending careers were longer; their prosecutions more frequent; their punishments stricter; and their domestic circumstances more unstable. The “persistent offender” group can be broadly split into two sub-groups: those who experienced particularly turbulent familial lives; and those who fell into a cycle of alcoholism, poverty, and distress. The former, “turbulent lives” category consisted of four female recidivists, all of whom were from Exeter: “Emma” (who offended in the years 1881–1908); “Maria” (1896–1907); “Sally” (1897–1909); and “Agnes” (1907–1914) [all names have been anonymised for ethical purposes].

The most immediate similarity between these four women was their shared experiences of domestic abuse. Each had prosecuted her husband for assault numerous times, the records of which indicate long histories of marital violence. Agnes prosecuted her husband at least eight times during the years 1902–1912, each time testifying to his violent conduct toward her. In 1904, she told the Bench that she was “so much in danger of his violence that she had to lock her door to protect herself from him” and further stated that she “lived in fear of him” as he had “continually ill-used her since the first day of their marriage.” Several assaults later, she requested a separation order against him: despite her testimony that she had suffered “cruel ill-usage during the whole four and a half years of her married life” due to her husband frequently being drunk, breaking things and threatening to kill her; however, the magistrates refused with the response that the couple must “try to bear with each other.” No peace was found; the Western Times subsequently published an article entitled “NINE YEARS’ MISERY” which reported a further assault by Agnes’ husband and alleged that he was “constantly ill-treating her” when drunk.

Emma experienced an especially unstable marriage, one characterized by numerous assaults, separations and reconciliations. Within fourteen weeks of their marriage (1880), she prosecuted her husband for threats of violence and testified that she was afraid to live with him. By the end of their first year together, she had moved back to her parents’ house with their son. The couple had reconciled by 1883, although she applied for a separation order against him later that year: claiming that she had returned since he “promised to behave properly and be a teetotaler,” she told the magistrates that “after about a month…[he] was worse than he had been before.” Her request for separation was refused and, ultimately, she took out four further assault prosecutions against him during the years 1883–1888. The couple were separated again by 1889, in which year Emma prosecuted her husband for moving to London and failing to maintain the family. Although he later returning to the family by 1891, he was warned by the Bench that he should “refrain from molesting his wife.” Ultimately the couple had separated again by 1901 and Emma spent the remainder of her life separated from her husband.

The women in the “turbulent lives” group were not only victims of domestic abuse but they also reciprocated on occasion. Agnes charged her husband for assault on numerous occasions but was also convicted for assaulting him in 1907 and 1909 amid domestic disputes. The first assault, which she claimed to be an act of self-defense, resulted in a fine. The second prosecution resulted in a month-long prison sentence, Agnes having used a razor to cut her husband’s face when he “knocked her down” and “told her to keep her mouth shut.” Maria was also convicted for assaulting her husband during a domestic argument. An article in Trewman’s Flying Exeter Post entitled “MATRIMONIAL MISERY” reported that she had threatened him, kicked him, and punched him during a row. She stated in defense that she had been responding to “years of unhappiness and wretchedness”: claiming that she had had “a rough time of it” during their 27-year marriage, she testified that he would often stay out with other women before returning home and “beat[ing] her until she was bruised.”

In Maria’s case, marital discord manifested in violence not only toward her partner but also against other women; adulterous behavior by her husband incited conflict with female rivals. During the years 1887–1889, she was convicted three times for assaulting her husband’s alleged mistress. For the first assault, she was described as a “jealous wife” who assaulted the mistress for being “annoyed at the familiarity between [the woman] and her husband.” Her third assault on this woman resulted in a prison sentence alongside reprimand from the magistrates, yet Maria replied that there would be “a hundred more convictions” if the magistrates “continued to allow her to live with [her] husband.” She was subsequently charged for assaulting a different alleged mistress in 1896: stating that she had “faced [the mistress] like a woman” on account of her being “too intimate” with her husband, she claimed that it was “all through he and nothing else.” She was also convicted at the Assizes in 1896 and sentenced to nine months’ hard labor after setting fire to the house of another alleged mistress.

This group of persistent offenders also shared financial difficulties during their offending careers; all were working-class women who experienced ill-paid employment and overcrowded housing. Sally and her husband were both intermittently employed as hawkers, had several children to support and were living in a lodging-house on Coombe Street – then one of Exeter’s slum districts – throughout their lives. They were also convicted together at Cullompton magistrates’ court in 1889 for stealing a brush, Sally testifying that she “had no situation” and “needed to feed her family.” Emma also expressed financial difficulty, her intermittent separations from her husband necessitating her to support their eight children through a combination of welfare funds and her earnings as a hawker. Following her husband’s move to London in 1889, she told the magistrates that, since her husband had sent no money despite making “good business,” she was forced to feed the children through poor relief. Maria, meanwhile, was committed to the workhouse after separating from her husband and, finding herself destitute on departure, subsequently moved into an almshouse for elderly pensioners.

As aforementioned, the second category of persistent offenders in this study consisted of two women whose life courses featured alcoholism, poverty and distress. The first was “Eva” from Cheltenham, whose offending career extended over the years 1883–1909. At the onset of offending, Eva appears to have had a relatively stable domestic life: she was living with her husband and their children; both she and her husband were employed; and the family were able to remain in the same, non-shared house throughout their marriage. In addition, no evidence was found to indicate that her husband had a criminal or violent history.

Yet Eva’s pathway into crime opened when she became a prolific drinker. Before appearing in court for violent crimes, she received numerous summonses for drunk and disorderly behavior. On one occasion, a constable reported that he had found her drunk in the road with “no hat or bonnet” on, her hair “loose down her back.” A subsequent charge alleged that she had been intoxicated when she broke into a neighbor’s house and smashed their windows. In 1884, by which time she had received seven drinking summonses, a magistrate commented that she had appeared “too often” for alcohol-related offenses.

Drunken conduct frequently brought Eva into conflict, especially with police officers. In 1883, she allegedly pushed and bit an officer when he attempted to arrest her for drunkenness. The encounter resulted in two convictions, one for assaulting an officer and one for drunkenness. On another occasion, an officer had been called to a pub to “remove” the inebriated Eva and, although she “went quietly” when arrested, she used “shocking language” when arriving at the police station and “scratched and kicked [the officer] several times.” She was again convicted on charges of drunkenness and assaulting an officer. Drunkenness also resulted in conflict with members of her community. In 1883, she drunkenly challenged two neighbors – a husband and wife – to fight her. When a constable arrived at the scene, he found her “insensibly drunk”, standing before “a huge crowd of over 100 people” with her sleeves rolled up. Eva was involved in at least five more street fights with neighbors during her offending career, her final appearance occurring in 1909: after a drunken row with another woman, she agreed to settle the matter through by fighting, a scuffle which the arresting officer later described as a row between two “drunken women” who were “scratching like cats.”

The second woman in the category of persistent offenders was “Martha,” an Exeter recidivist whose offending career spanned the years 1871–1890. Like Eva, Martha appears to have had a relatively stable domestic life prior to offending: her husband had no known previous convictions; they had several children; and her family had resided in the same, (again) non-shared house for years. Although she was unemployed, her husband was consistently employed as a laborer until his death in 1888. Census records also indicate that some family members were living on her street, suggesting that she may have had a relatively strong support network closeby.

Martha’s earliest assaults were relatively run-of-the-mill offenses prompted by disputes with her neighbors. Most rows were rooted in either territorial or slander-related concerns, issues likely triggered by the cramped living conditions of late Victorian working-class housing. One of her first offenses involved assaulting a neighbor in 1881 when she overheard them calling her names, whilst an assault committed the following year was provoked when a neighbor’s child had been disruptive by playing outside of her house. In 1885, she was charged with assaulting another neighbor after telling him he had “no right” to be on her path and throwing a bucket of ashes over him.

Maria’s propensity for alcohol contributed to her criminality. Her earliest drinking charges, occurring in the 1870s, involved conflict with neighbors, publicans, and police officers alike. In 1874, when she was refused a drink by a publican who thought she had “had quite enough,” she made “offensive epithets” and broke the pub windows. Two years later, she was charged with being drunk and using abusive language after making drunken threats toward her neighbor and his mother. She was subsequently summoned for abuse after drunkenly attacking her neighbor in a public house and using offensive language. Evidence indicates that she continued to drink after the death of her husband in 1888, resulting in further court appearances: in 1890, for example, she assaulted a publican by throwing a plate at him when he refused to serve her another drink.

Her widowhood alongside alcohol consumption may have caused significant financial difficulties and contributed to further criminality. By 1891, three years after her husband’s death, she was unemployed and living in the Exeter workhouse. She relocated to Plymouth after her release, in which city she continued to offend: alongside two further assault convictions, she was convicted of stealing some brushes in 1897 and stated in her defense that she was an “old widow” with “nothing and no-one” to support her. Indeed, Maria’s socio-economic instability had notable impacts on her later life. In 1931, she was charged at Exeter for shoplifting, a case for which a newspaper reported under the headline of “THE INFLUENCE OF DRUGS: A SAD CASE” that she had been an “invalid” for over seven years and had stolen “a large dose of drugs which she was accustomed to taking for her nerves.” She pleaded in defense that she was a “poor, old woman” and asked for magisterial leniency, the Bench thus agreeing to avoid incarceration “on account of her age” and instead binding her over in the sum of £10 for two years.

The six persistent offenders in this study all experienced difficult living conditions, social isolation, and economic marginalization. Some faced turbulent, violent marriages: for these women, separations, abandonment, and adultery were not only a part of ordinary marital life but also a contributor to criminal behavior. The remaining two women appear to have struggled with alcohol use, a propensity which resulted in habitual petty offending alongside conflicts with members of their communities. Ultimately, all six women appear to have become caught in a cycle of offending that made it difficult for them to desist: indeed, in a study of nineteenth-century female recidivism, Turner (2011) notes that some female offenders could become “so excluded from society that their reintegration into their families or communities was not feasible”; even after leaving prison “the harsh existence of working-class life or homelessness only exacerbated, or at least continued, their offending behavior.” Leigh Straw’s study (2013) on female recidivism in twentieth-century Australia similarly argues that some women had “minimal or no familial or financial support” and were thus “caught in a cycle of offending and incarceration” such that they were “confined to a life more or less decided by legislation, the courts, police and press.”

LIFE-COURSE TRAJECTORIES OF VIOLENT RECIDIVISTS AND NON-RECIDIVISTS

As yet, only one historical study has paid notable attention to the post-offending lives of female reoffenders. Focusing on the life-course trajectories of nineteenth-century Swedish recidivists, Lotta Vikström (2011) has indicated that the post-offending experiences of young reoffenders and non-reoffenders differed in two fundamental ways: first, reoffenders were more likely to experience an “untimely” death; second, they were more likely to be ostracized and thus to migrate from their hometowns. Vikström's study thus provided much-needed insight into the experiences of women recidivists and opened new lines of enquiry for historical research.

Findings regarding the lives of nineteenth-century Swedish recidivists do not, however, resemble the experiences of the violent female recidivists in late Victorian and Edwardian England. There were insubstantial differences in death rates among reoffenders and non-reoffenders, the average age of death having been 69 and 70 respectively. It is near-impossible from the available sources to ascertain the causes of these women’s deaths, although there is a record of one recidivist who, after having become unemployed and separating from her husband, ultimately committed suicide due to the “lot of trouble” she had suffered. Nor were migration rates particularly differential, with 25% of recidivists compared to 24.3% of non-recidivists in the study having left the towns of Cheltenham or Exeter after offending.

However, several other traits can help to distinguish the post-offending experiences of recidivists and non-recidivists. In particular, habitually violent women may have had more difficulty in forming partnerships: only one recidivist (compared to four non-recidivists) got married in their post-offending life; possibly a related finding, recidivists were also less likely to have children in their post-offending lives. Reoffenders also appear to have had more difficulties in maintaining relationships: 19% of recidivists compared to 2.3% of non-recidivists became separated from their husbands in later life. Separations may have resulted either from marital unhappiness or from being stigmatized as a criminal. One recidivist had been living with her husband throughout much of her offending career, although a series of prosecutions against her for assault and drinking ultimately harmed their relationship; her husband stated in one prosecution that he “hardly knew how to deal with her” and the couple soon separated, with the woman subsequently moving to London with their daughters by 1911. By contrast, most non-recidivists who separated from their husbands later reached reconciliation; for example, one woman committed her only assault in 1906 when she was separated and living with her parents, yet subsequent census records indicate that she later reunited with her husband.

Living conditions also mark a further divergence between these groups of women. Recidivists more frequently experienced cramped living conditions not only prior to offending but also in their post-offending lives: eight recidivists – but only one non-recidivist – moved into shared lodgings after their court appearances, most having relocated due to familial difficulties. Additionally, many recidivists found themselves without stable or permanent residences after their offending careers, two ultimately becoming homeless in later life. A Cheltenham habitual offender with a long history of assault, theft and alcohol-related offenses became homeless after her cessation from crime in 1887: describing her as a “drunk [and] homeless woman” who spent her time “in and out” of the workhouse, an article in Cheltenham Chronicle reported that she “annoyed” the Cheltenham Guardians by asking for money and that it was “undesirable that a Guardian should be subjected to such annoyance.” In addition, while none of the non-recidivists appear to have been institutionalized after offending, five recidivists spent time in workhouses, almshouses and/or asylums in their post-offending lives. One Exeter recidivist was known in her community as a prolific drinker: in one assault case, her neighbor (a witness) alleged that she drank two bottles of stout a day; another said that “after drinking [she] had the impression that she was the Queen of England [and that] everybody must bow down to her.” She was ultimately institutionalized, the 1901 and 1911 censuses having recorded her as an inmate at the Gloucester County Lunatic Asylum.

CONCLUSION

Historical research on crime makes a significant contribution to the fields of women’s history and criminology by providing insight into the interactions, struggles, and, at times, desperate actions of criminalized women. Discussions of habitual offending are no exception since they cast light on how marginalized women navigated social and legal structures in their journeys toward desisting from and persisting in crime. Yet historical research on female recidivism has remained scant. This article has therefore offered an original contribution to discussions on female reoffending, and to the experiences of marginalized women more generally, by exploring the onset, cessation and aftermath of female habitual violence in late Victorian and Edwardian Cheltenham and Exeter.

Violent female recidivists constituted only a minor proportion of all women charged for acts of petty violence, yet they were responsible for a substantial amount of women's assaults. These women experienced a large degree of marginalization: many lived in cramped housing, were unmarried mothers, and/or were undertaking unskilled labor. Despite these circumstances, most recidivists eventually desisted from crime. The criminal justice system – particularly incarceration – appears to have had a stronger influence on desistance than informal social control factors; however, marriage, and the greater economic and social stability that marriage brought, also appears to have had a positive impact. There was a small minority of women for whom social and judicial impetuses had little impact, especially those experiencing abusive marriages, alcohol use, and/or destitution.

Although many recidivist offenders eventually desisted, prospects were typically bleaker for these women in their post-offending lives. They experienced more difficulties than non-recidivists in both the relationship and labor markets, more frequently finding themselves without husbands or occupations and therefore with more limited support networks. Recidivists were also more likely to move into shared lodgings, to become homeless or to be institutionalized. The life courses of habitually violent women therefore demonstrate that repeat offending – and perhaps the act of being labeled a habitual criminal – may have had significant repercussions on women’s later lives both socially and economically. This supports contemporary evidence on female recidivism in indicating that women habitual offenders can struggle to “drift back” into conventional life after offending due to being “marginalized from family, friends, children and work…[and] los[ing] their traditional life structures” (Sommers et al., 1994).