• 3 European Sourcebook of Crime and Criminal Statistics (2010, p. 195).
• 4 Ruthenfrans (1989) ; Heidensohn (1996, 2^nd ed.) ; Dow Burkhead (2006, p. 50) ; Silvestry, Crowther- (...)
• 5 Feeley, Little (1991) ; Feeley (1994) ; Van de Pol (1987, pp. 149-151) ; Van der Heijden (1995).
• 6 King (2008, pp. 216 & 221) ; Zedner (2002, p. 1) ; Roets (1985, pp. 363-368) ; Faber (1983, pp. 253 (...)
• 7 Van der Heijden (1995, pp. 4-10).

1Criminologists and historians agree that women usually are responsible for a smaller proportion of recorded offences than men : approximately 13 per cent of all prosecutions in Europe3. However, there is much less agreement about patterns of change over time. Criminologists generally assume that sex differences in recorded crime rates are invariant over time and space, stressing the continuity of men’s excessive contribution to criminality rather than discontinuity4. Data on early modern Europe show that in France, England, and the Netherlands between 1600 and 1800 women played a much more prominent role in crime than in the twentieth century. Case studies on London and Amsterdam show that in the seventeenth century 50 per cent of all recorded criminals were female5. The data for various English counties and Dutch towns suggest that females represented 20 to 40 percent of the recorded offences in the seventeenth and eighteenth centuries6. Such data show that women’s involvement was not limited to distinctively female offences such as infanticide, witchcraft, and prostitution. Women also constituted a large part of the perpetrators of crimes that are typically associated with males, such as property offences and violence7.

• 8 Beattie (1975, p. 81, 1986).
• 9 Silvestry, Crowther-Dowey (2008, p. 27) ; Pollak (1961) ; Adler (1975) ; Arnot, Usborne (1999).

2In his study on crime and the courts in England between 1660 and 1800 J.M. Beattie was one of the first to acknowledge that female crime patterns changed over time. In his path-breaking work Beattie paid particular attention to rural and urban differences, arguing that female crime rates were higher when women gained more independent positions and when their participation in public life increased8. Like many criminologists, Beattie linked strong gender differences in criminal behaviour to contrasts in their presence and functions in public life : women’s limited freedom and opportunity may lead to less involvement in crime and more lenient treatment by the courts9.

• 10 Feeley (1994, pp. 250-259).

3The presumed link between women’s participation in public life and their involvement in crime has led to a vigorous debate among historians about the precise timing of increase or decrease in women’s freedom and opportunities as well as about the long-term trends in female crime. In several provocative articles Malcolm Feeley stirred the debate by arguing that in Europe the female percentage in crime dramatically declined over the eighteenth and nineteenth centuries. Feeley linked the decline of female crime to the (re-)emergence of patriarchal structures and the separation of public and private spheres which caused fundamental differences in male and female patterns of crime10.

• 11 Feeley, Little (1991, p. 719) ; Feeley, Hadar Aviram, unpublished online paper (2008) : ‘Where have (...)
• 12 Feeley, Little (1991, pp. 724-740).

4Feeley and his co-authors Little and Aviram based their findings on the proceedings of the Old Bailey in London between 1687 and 1912 and on secondary literature on female crime in various European countries, particularly The Netherlands11. Though recognizing that factors of demography, jurisdiction, and war might have played a role in the changing proportion of women in crime rates, they dispute the relevance of such factors, arguing that demographic developments and changes in the judicial system would not be sufficient to explain long term trends in female crime12.

• 13 King (2008, pp. 198-199) ; Shoemaker (1988) ; Emsley (2^nd edn, 1996, p. 152) ; Arnot, Usborne (1991 (...)
• 14 King (2008, pp. 199-220).
• 15 King (2008, pp. 202-206-220).

5While most historians assume that a separation of private (female) and public (male) spheres in the eighteenth and nineteenth centuries may have reinforced the gendered crime pattern, other scholars raised substantial doubts about the idea of the ‘Vanishing Female’ as suggested by Feeley13. Peter King maintained that the English evidence points to long-term stability in female recorded crime in the eighteenth and nineteenth centuries rather than a long-term decline14. King agreed that the proportion of female crime rates strongly fluctuated between 1750 and 1850, though he links such changes to complex short-term developments, such as war and exceptional circumstances. Furthermore, there were wide variations in the proportion of offenders who were female15.

• 16 Zedner (2002, p. 36).
• 17 King (2008, p. 220).

6His findings confirm Beattie’s argument of a close relationship between the degree of urban involvement in a region and its percentages of female offenders. Although spatial variations and the complex geography of female involvement in indictable crime make it difficult to draw general conclusions, there appears to be very little evidence of long-term decline, at least not until the second half of the nineteenth century. The work of Lucia Zedner on nineteenth-century England showed that there was no decline in the proportion of indicted women before around 1850. Furthermore, the numbers of suspected persons between 1893 and 1910 demonstrate that women’s proportion declined as late as the twentieth century. Before 1905 women represented 22 to 23 percent of those suspected of crime16. King concludes that “the proportion of females to be found amongst indicted offenders showed very little, if any, sign of significant long-term change. On closer inspection therefore, the vanishing female offender vanishes”17.

• 18 Arnot, Usborne (1999, p. 11) ; King (2008, p. 199) ; Feeley (1994) ; Van der Heijden (1995) ; Beatt (...)

7Although it has been stressed that more quantitative and comparative data on the period between 1750 and 1900 are needed to determine the changing nature of recorded male and female crime, there is still no consistent evidence about long-term trends in female crime in Europe18. This article aims at contributing to the debate on long-term developments in male and female criminal behaviour by presenting new evidence on patterns in the prosecution of Dutch female crime over the eighteenth and nineteenth centuries. As Feeley argued that a similar decline in female crime occurred in the Netherlands, the Dutch case is highly relevant to the general debate.

• 19 Our examinations are based on Jaarboekje over 1836-1846, uitgegeven op last van Z.M. den Koning’ (J (...)

9We will first make some remarks about the arguments being raised and the methods used by Feeley and his co-authors. We will then turn to the Dutch case and present our findings on female crime in the Netherlands over the eighteenth and nineteenth centuries. The findings are based on recent examinations of the criminal records of various Dutch towns between 1750 and 1838, as well as national prison records and houses of correction between 1839 and 188019.

• 20 Van der Heijden (1995, pp. 1-10) ; Van de Pol (1987, pp. 148-155) ; Faber (1983, pp. 243-254).

10As King rightly argued, Feeley does not take into account regional variation of gender differences in crime, nor does he explain the impact of particular or exceptional circumstances, such as short-term socio-economic transformations and changes to law or procedural practice. There are no comparative studies on early modern Dutch crime, but the findings of various case studies – particularly on the period between 1600 and 1750 – confirm the suggestion that there were strong regional variations in crime rates. Data on various towns and cities in the Low Countries suggest that local circumstances did have an impact on gender differences in crime. The female contribution to crime varied greatly between regions and towns : generally the proportion of female criminals was much larger in big cities, particularly in port towns. In several towns the women’s share in crime was approximately 20 to 30 percent (Ghent) while in other cities the participation of women in prosecuted crime was much larger (in Amsterdam, Leiden, and Rotterdam ranging between 35 and 50 percent)20.

• 22 Schwerhoff (2000), unpublished online paper ‘Gender and Criminal Justice : the German Case’, 19^th I (...)
• 23 Van Zanden (1997, p. 22).

13Besides the lack of sufficient evidence for the period around 1800, the concept of ‘the vanishing female’ is problematic in itself. Gerd Schwerhoff already pointed to Feeley’s poor elaboration of the concept and the need for a more critical and comprehensive interpretation of the sources. It is not clear to what extent processes of industrialization would have resulted in a decline in women’s labour participation, nor is there evidence for a strong relationship between female crime rates and women’s labour participation. In Germany the labour participation of women increased in the course of the nineteenth century, but increasing female employment rates did not result in higher female crime rates22. There is also no evidence that points to a link between developments in labour relations in the Netherlands and the decline in the proportion of women in crime. The process of industrialization started in Holland – the region that Feeley focuses on – as late as 1850 while the reliable Dutch data in Feeley’s analysis cover the period before 181123. If Feeley’s suggestions are correct in the sense that there was an actual decline of Dutch women in the criminal process after 1800, it seems very unlikely that such a trend can be explained by the process of industrialization.

• 24 King (2008, pp. 218-219).

14There is another point regarding the alleged decline of female crime rates that deserves much more attention : the significant jurisdictional changes after 1800. For England, King pointed to the relationship between changes to the patterns of prosecution and procedural practice from 1850 to 1890 and the lower proportion of women brought to trial for indictable offenses in the new statistical series after 1857. The judicial system of the Netherlands underwent fundamental changes after 1800 as well. In the early modern period crimes were prosecuted by local criminal courts, primarily in towns and cities. In 1811 the prosecution of crime became a national matter, resulting in the replacement of the old local courts and prisons by regional courts and national prisons. The Dutch sources used by Feeley, Little and Aviram do not include the correctional courts that existed between 1811 and 1838 and the national prison records after 1839 – an omission of data that may be crucial to their argument that women’s decline began after 1800 – nor do they give a careful explanation for the crucial differences between the local jurisdictions of the early modern period and the new national statistical series of the nineteenth century. To quote King, “Feeley and Little’s data on the period from the 1850s to the 1890s is clearly deeply suspect given the jurisdictional changes”²⁴.

• 25 King (2008, p. 112).
• 26 Christiaens (1994, pp. 149-169, 1999).
• 27 Christiaens (1999, pp. 149-169 & 157-160).
• 28 Christiaens (1994, pp. 158-159).

15There were also significant changes in the treatment of criminals by courts and prosecutors. In the course of the nineteenth century prosecutors began to focus on education and juvenile crime, and subsequently age became an important factor in the prosecution process. The changing attitudes to childhood were linked to broader public concerns about crime, poverty and juvenile offenders. King has shown that in England in the nineteenth century “there was a new desire to discipline rather than ignore juvenile offenders”²⁵. The examinations of Jenneke Christiaens on Belgian criminal children confirm this trend. Growing concern about the behaviour of poor and ‘immoral’ families led to a broad societal movement that linked family intervention to crime prevention. In order to bring these new values into practice, the state created legal instruments that allowed criminal justice administrators to increasingly intervene in the private sphere of the household²⁶. From the first decades of the nineteenth century onwards, Belgian children were increasingly removed from their families and placed in houses of correction in order to educate and discipline them²⁷. A considerable proportion of children who were confined at the request of their fathers, were female. Between 1861 and 1875, 35 percent of the 2157 confined youngsters were girls. By the end of the nineteenth century 200 to 300 children were incarcerated every year at the request of their parents²⁸.

• 29 Groenveld, Dekker, Willems (1997, pp. 324-331) ; Dekker (2001).
• 30 Leonards (1995, p. 279) ; Bakker (2010, p. 432).

16Like in Belgium, the increase in the proportion of juvenile offenders in the Netherlands was linked to broader societal concerns about crime and children, and there was a growing awareness among public authorities that the state ought to take responsibility in such matters. Children were increasingly confined in houses of correction in order to discipline and re-educate them, regardless their age and sex. Next to state prisons, private initiatives were undertaken to establish houses of correction in which children from poor and often broken families were ‘re-educated’29. The examinations of Chris Leonards on the nineteenth-century Dutch penitentiary institutions show that boys were increasingly imprisoned. The proportion of girls amongst the juvenile prisoners appeared not to decline between 1860 and 190030.

• 31 Shore (1999) ; Cox, Shore (2002).

17The studies on England, Belgium, and the Netherlands may offer an indication of the numbers and percentages of young women and girls who were imprisoned, but none of these works include gender differences in their analyses. The data on juvenile delinquents concerns almost exclusively boys, except for the studies of Heather Shore and Pamela Cox on juvenile offenders in nineteenth-century Britain. Such explorations may offer explanations for changing attitudes of prosecutors and courts towards male and female crime, and subsequently to changing male and female crime rates. Shore found that institutional responses to juvenile crime were based largely on a concern of boys rather than girls. However, such considerations are unnoticed in the work of Feeley and his co-authors³¹.

• 32 Faber (1983, p. 254).

18Dutch evidence on male and female crime is mainly available for large towns in Holland during the period between 1600 and 1750. Various studies on Amsterdam, Rotterdam, and Leiden have shown that the proportion of women who were prosecuted by the criminal courts was – at times – exceptionally high. The study of Sjoerd Faber on Amsterdam between 1680 and 1811 showed that females represented 35.2 percent of those sentenced by the Amsterdam criminal court32. In the last quarter of the seventeenth century women’s involvement in crime in Amsterdam was generally 49 percent, though there were continuous short term fluctuations, and in some years the female crime rate was as low as 20 percent. The occasional low figures at the end of the eighteenth century gave Feeley reason to believe that there was a long-term decline of female crime in Amsterdam, which set in around 1800. Although the proportion of women amongst those who were prosecuted appeared to fall to 20 percent at the end of the eighteenth century, in some years between 1750 and 1800 female crime rates continued to be exceptionally high (more than 50 percent). Recent examinations of the new criminal courts of Amsterdam after 1811 make it possible to extend the period and expand the information about the prosecution pattern in the nineteenth century. As we shall see below, the figures appear to show a picture of long-term stability in the sense that there were continuous short-term fluctuations in the proportion of female defendants.

• 33 Kloek (1990, p. 133).

20Reliable data on Leiden is presented by Kloek’s examinations in 1990 on the confessions of defendants between 1678 and 1794. She found that in this period women’s share in all offenses was approximately 41 percent, and there was no decline over the course of the period. Between 1781 and 1790 women represented 44 percent of those who were prosecuted, and between 1791 and 1794 women’s share in crime was 58 percent33. Women were particularly represented in property offenses (47%), moral offenses (67%), and offenses against public order (31%). It may seem obvious that women were heavily represented in moral offenses because they were prosecuted for prostitution, though most of those cases against women were related to adultery and fornication.

• 34 Tegelaar (2012b, pp. 31-48).

21Recently, Tegelaar provided new data on the long-term patterns of male and female crime in the criminal records of Leiden in the eighteenth and beginning of the nineteenth century (sentence books), and her findings confirm the female crime trends as described by Kloek34. Tegelaar shows that there may have been years with lower female rates in the 1770s, but as Kloek already revealed, there was no decline in the share of women in crime after that period. Tegelaar presents data on the early nineteenth century, and demonstrates that there were extensive continuities in women’s involvement in crime between 1700 and 1804. Between 1700 and 1702, for example, 40 percent of the offenders were female, between 1800 and 1802 women’s involvement in the criminal process was around 43 percent.

• 35 Tegelaar (2012b, p. 48).

23Women were mostly prosecuted for property crime, and in the course of the period 1675-1811 this offence became increasingly important in the cases that were handled by the criminal court. Detailed information on the criminal cases shows that between 1600 and 1604 only 21.5 percent of the women who were prosecuted were held for a property offense, yet between 1700 and 1704 almost 70 percent of the cases involving women concerned property, and between 1800 and 1804 that percentage rose to 83.3. Clearly, in the course of the period women were increasingly prosecuted for committing property crimes, and less for moral offenses. Between 1775 and 1804 women rarely came before the criminal court for committing a moral crime. There may be a link between increasing percentages of property offenses and poverty, because in the course of the eighteenth century women increasingly pointed to their poor circumstances as the reason for their offenses. Indeed, in the second half of the eighteenth century the textile industry of Haarlem declined, and a considerable proportion of the employees working in this sector became unemployed. Haarlem’s population declined considerably because many textile workers left the town to find work elsewhere. The people who were brought before the courts at the end of the eighteenth century were increasingly poor migrants and wage labourers whose work was related to the textile industry of Haarlem35.

• 36 Van der Heijden (1995, 1998, pp. 46, 75 & 154-155).

24Lastly, there are new figures on the prosecution of men and women in Rotterdam during the second half of the eighteenth century. In earlier works about women’s criminality, Van der Heijden showed that between 1700 and 1750 women represented on average 35 percent of those brought before the criminal courts of Rotterdam. The information on the Rotterdam criminals included both the criminal sentence books in which felony cases were recorded and the correction books which consisted of misdemeanours, particularly violence. The latter were referred to as fight books (vechtboeken), though the records included a much wider variety of offenses, including felony cases. The figures on women’s share in the total number of offenses (the number of persons was lower) was slightly higher, on average 38 percent of the misdemeanours were committed by women. Clearly, the figures on female crime in Rotterdam between 1700 and 1750 resemble the fluctuations in female crime in other cities. Women were prosecuted for a variety of offenses : 33 percent of them for a moral offense ; 23 percent for a crime related to property ; 23 percent for crime against public order ; and 21 percent of female offenders for violence. Those moral offenses were not related to prostitution, but primarily to adultery and to a lesser extent fornication. The fact that Rotterdam was an important town of the East Indian Company had a significant impact on female crime ; like in Amsterdam, most adulterous women were left behind by their husbands who worked as a sailor for the company36.

• 37 Stads Archief Rotterdam, Oude Rechterlijke Archieven (ORA), Vechtboeken 1750-1795.

25What about the proportions of female offenders in Rotterdam in the second half of the eighteenth century, the period about which Feeley has no information ? New data on the fight books between 1750 and 1795 reveal that the proportion of women’s involvement in crime did not decline in the second half of the eighteenth century. It may not be a surprise that the figures on female crime rates continued to be stable in the years between 1750 and 1795 – considering the figures of the other towns in Holland – but it is remarkable when one looks at the nature of the offenses that were recorded in the fight books. Although in this period all crime categories were recorded in the fight books, 40 percent of the prosecuted persons had committed a crime related to violence. Surprisingly, women made up a substantial portion of those who were held because they displayed violent behaviour. On average women’s share in crime in the fight books between 1750 and 1795 was 36 percent, though in some periods women were responsible for 44 percent of the offenses37.

• 38 Beattie (1975, p. 81) ; Shoemaker (1991, pp. 208-209).
• 39 King (2008, p. 215).

26How can we explain the stable high rates of female crime in the early modern period ? The most obvious explanation is the extent of urbanization in Holland, a link that is particularly explored by Beattie in his work about the criminality of women in England. Beattie found women’s contribution to crime to be generally much higher in cities than in rural communities or small towns. The lack of economic and social support from the traditional community caused women in towns to lead more independent, public and exposed lives³⁸. In his survey on the London data on female crime, King also suggested that the wide variation in the proportion of offenders who were female may be explained by the degree of urban involvement in a region. The most urbanized regions, such as Surrey and Kent, had the highest proportions of female offenders ; the rural areas had much lower percentages of women who were prosecuted.³⁹.

• 40 De Vries, Van der Woude (1995, p. 83) ; Clark (2009, pp. 119-123 & 128).
• 41 Van der Heijden, Van den Heuvel (2007).
• 42 Shoemaker (1991, pp. 208-209).
• 43 Moch (2003, p. 146).

27The close relationship between the degree of urbanization and the percentage of female offenders is particularly relevant to the highly urbanized region of Holland. At the end of the eighteenth century, 30 to 40 percent of the Dutch population lived in towns, in England 20 percent, in Italy 18 percent, and in Germany 5.5 percent. In Holland, the level of urbanisation was as high as 70 percent⁴⁰. Furthermore, the towns in Holland were characterised by an exceptionally high ratio of women. Compared to many other towns in Western Europe, Dutch urban centres, and particularly those in the province of Holland, were characterised by very large numbers of women living alone – either permanently or temporarily. The relatively independent position of women and consequently their greater public presence may explain the exceptionally high female crime rates in the towns of Holland⁴¹. Being involved in public activities, women had more opportunities to make their own decisions, to move around freely, and to work outside the home. This freedom also caused greater risks of getting on the wrong side of the law. Because of their more public lifestyles, the insecurity of their employments, and their lack of family support, independent women – especially if they belonged to the lower classes – were more likely to enter into disputes and to commit crime in times of hardship⁴². For many young lower-class women migrating from the countryside to the city, criminal activities became a logical survival strategy⁴³.

• 44 Leonards (1995, p. 110) ; Dekker (2009, pp. 17-36) ; Petersen (1978, pp. 119-146).

30As the young nation state of the Netherlands began to assume responsibility for most public facilities, prisons became the sole responsibility of the central authorities that cooperated with a new institution for the moral improvement of prisoners : ‘Genootschap tot Zedelijke Verbetering der Gevangenen’ (translation : Society for the Moral Improvement of Prisoners). From 1833 onwards, the distinction between adult crime and juvenile crime resulted in the establishment of national youth prisons that where initially meant for both sexes, but which became separate institutions for boys and girls after 183644.

• 45 De Smet (2005, p. 3).
• 46 Leonards (1995, pp. 57,175 & 182, 1989, p. 108).

31The Dutch penal codes of 1810 and 1867 distinguished between children below and above the age of sixteen. Children under sixteen were not held accountable for their actions while children over 16 were sentenced as adults. However, the law made exceptions for younger children who were believed to have consciously committed crimes (nullum crimen sine culpa). Although such children were not convicted in the same way as children over 16, they did not escape imprisonment. Those who received the verdict nullum crimen sine culpa were placed in houses of correction until the age of twenty45. Special prisons were established in order to distinguish between convicted juveniles and children under 1646.

• 47 Ruitenbeek (2010, pp. 66-67).
• 48 Ruitenbeek (2010, p. 68) ; Aerts (2006, pp. 242-243).

34The court records of Amsterdam between 1815 and 1835 clearly show that the rates of accusations and trials against women fluctuated, just as they did in the previous periods. Furthermore, the overall number of individuals prosecuted increased considerably during the first half of the nineteenth century, particularly in the period after 1830.47. This trend was not linked to demographic changes in the city of Amsterdam. In the period between 1815 and 1836 the population of Amsterdam slightly increased from 183,000 inhabitants to 206,000, while the numbers of the accused individuals grew considerably faster48. The female crime rates as revealed by the examinations of the Amsterdam correctional court were slightly lower than the figures for the Amsterdam criminal court before 1811. That might reflect a small decline in female involvement in crime, although jurisdictional changes may be a more plausible explanation. The records of the correctional court between 1811 and 1838 include defendants from various towns and villages in the north of Holland, not just Amsterdam.

• 49 Van Rutter (1989).

35Ruitenbeek also presents information about the records of the Court of Assizes of The Hague (1811-1838) whose jurisdiction covered felony cases in the south of Holland49. This court handled crimes such as theft and burglary, assault, homicide, manslaughter, infanticide, patricide, poisoning, and rape. Around 10 percent of the defendants were eventually acquitted, and released without punishment. Those who were convicted by the Court of Assizes received prison sentences that could range from a few months to 20 years, often in combination with a fine or scaffold punishment. Van Ruller examined the records of the South-Holland Court of Assizes for the years between 1811 and 1838 during which he counted 3,704 defendants. The involvement of women in felony cases appeared to be rather high, as 19 percent (699 women) of the defendants were female. Most of these women (around 80 percent) were held because they had committed theft in combination with other offenses. Around 20 percent of the defendants were prosecuted for violence and assault, and occasionally for manslaughter or murder. Much additional research needs to be done on the period between 1811 and 1838 to draw definitive conclusions. However, the present evidence suggests that there were extensive continuities in the proportion of female offenders amongst the various courts between c. 1750 and 1838.

• 50 Our sources : Jaarboekje over 1836-1846, uitgegeven op last van Z.M. den Koning (Jaarboekje van Lob (...)

36What about the national statistics on male and female crime after 1838 ? This survey includes our own examinations on the national statistics of the prison population between 1839 and 1886. The information includes the records of ten national prisons and four state-run houses of correction. The prisons were located in towns across different regions in the Netherlands, and its populations came from all over the Netherlands. There is no information available about the original place of residence of the offenders. Between 1839 and 1886 a total of 147,352 individuals were committed to prison, an average of 3,000 prisoners a year. Between 1836 and 1886 on average 400 children were imprisoned every year. During the first period after the introduction of the Dutch Penal Code children constituted only a small part of the total prison population, generally about 6 to 10 percent. However, in the course of the nineteenth century the proportion of imprisoned children rose to over 25 percent50.

• 51 [www.bevolkingstellingen.nl] (September 2012) ; Kok (2003) ; Leonards (1989, pp. 98-99).
• 52 Van der Woud (2010, pp. 42-45).
• 53 Van der Woud (2010, p. 58).

39The decline in numbers of adult prisoners in these statistics over the course of this period makes it even more difficult to interpret the figures on female crime rates. Between 1845 and 1880 the number dropped from 211 convicted women a year to 45, while the number of convicted men dropped from 1523 to 502. This appears to reflect a permanent trend of steady decrease in the number of adult prisoners between 1845 and 1880. The decrease in prosecutions between the 1840s and 1880s was not related to a decline in population growth in the course of the nineteenth century51. Between 1840 and 1900, the size of the population in many towns in Holland doubled or even tripled. The population of Amsterdam rose from 211,349 inhabitants in 1840 to 510,853 in 1900, while during the same period of time the population of Rotterdam rose from 78,098 to 318,50752. Around 1880 about half of the Dutch population (2 million people) belonged to the working class, a quarter of which received poor relief53.

• 54 Weevers, Bijleveld (2010a, pp. 351-364, 2010b).

40What happened to criminal men and women after 1845 ? There is much further research needed, but it may be that those criminals who were previously sent to prison, were increasingly sent to the state-run reformatory institutions and workhouses (Rijksopvoedingsgesticht and Rijkswerkinrichting). Recent work of Bijleveld and Weevers on the state-run workhouse in Leiden shows that between 1886 and 1892 approximately 380 female beggars and vagabonds were sent to the workhouses, many of whom were recidivists who later returned to the institution. Around 54 percent of these women were sentenced by the regional courts in Amsterdam, The Hague and Heerenveen, and apparently they were sentenced to the workhouse as an alternative to prison. Their male counterparts were most probably sent to the male workhouse in Veenhuizen (in the north-east of Netherlands). However, the increasing importance of workhouses is not a sufficient explanation for the sharp decline in the proportion of women in the Dutch prisons after 1877. The study of Bijleveld and Weevers indicates that the number of women in the workhouses was much lower than the number of men. In the last decades of the nineteenth century, 2,700 to 3,000 men were brought to the workhouse in Veenhuizen every year, against only 100 women to the workhouse in Leiden54

• 55 These data are based on the following sources : Jaarboekje over 1836-1846, uitgegeven op last van Z (...)

41The data on the inmates of the Dutch main state-run prison makes it possible to also examine the number and percentages of convicted boys and girls between 1836 and 1886. The figures represent boys and girls under 20 years of age who were confined to a prison or a house of correction. The evidence points to a rather static pattern in the proportion of imprisoned girls, suggesting a considerable amount of continuity during this period of time. Girls constituted between 9 and 21 percent of inmates during the period under examination. Between 1837 and 1886 a total of 3,306 girls were sentenced to one of the Dutch prisons, an average of 70 girls per year. The number of the imprisoned boys was considerably higher, on average 346 boys per year. Between 1836 and 1886 a total of 16,785 boys under twenty were confined in one of these prisons. The proportion of the boys ranged between 79 and 91 percent. The evidence therefore suggests that there was no decline in the proportion of young girls in the juvenile prisons between 1840 and 188055.

• 56 Leonards (1995, pp. 48, 58, 64-65) ; King (2008, p. 214).

42The prison records and the records from the houses of correction show that the proportion of boys and girls was rather stable in the period 1836-1886. For instance, in 1846 about 19 percent of the imprisoned juveniles were girls, while in 1886 the proportion of girls amongst the prisoners was similarly 19 percent. There were however substantial changes in the numbers of the imprisoned juveniles. Between the 1840s and 1880s the numbers of prosecutions involving both boys and girls increased considerably, from 216 juveniles in 1836 to 638 in 1882. The number of boys under twenty increased from 167 in 1836 to 551 in 1882, while the number of young girls rose from 33 to13256. The large proportion of boys was also reflects increased patterns of recidivism. Girls were generally prosecuted only once in their lives while many more boys appeared to return to prison repeatedly. Each of these returns was counted in the records. Therefore, the real proportion of the convicted girls was higher than the above mentioned figures. Much more information is needed on the proportion of boys and girls in juvenile prisons before we can draw definite conclusions about long-term trends in the prosecution rates of girls, but so far the evidence indicates that there was no decline in the proportion of imprisoned girls.

• 57 Leonards (1995, p. 64).

43Over the nineteenth century new juvenile institutions were built leading to a considerable increase in the capacity of prisons to accommodate young criminals. However, the rise in the number of juvenile prisoners was not necessarily linked to such improvements. In 1859 the government opened a new penal institution for young female prisoners (Montfoort), but the increased capacity did not lead to a notable rise in the number of juvenile prisoners in the following years57. Instead the numbers and the proportion of juveniles who were committed to prison gradually increased in the course of the nineteenth century. This trend represented a rise in the prosecution numbers of both boys and girls.

• 58 Groenveld, Dekker, Willemse (1997, pp. 300-303) ; Van der Vlis (2002, pp. 43-47) ; Leonards (1989, (...)

46From the 1870s onwards the category of juveniles who were confined under section 66 of the criminal code increased considerably. In the first decades after the introduction of the criminal code of 1810, only a small number of juveniles under the age of 16 were sent to prison for correction. During the nineteenth century the prosecution and imprisonment of juveniles under 16 increased both in number and proportion. By the 1880s the majority of imprisoned juveniles under 20 were confined under section 66 : 71 percent of the girls and 65 percent of the boys. These figures indicate a remarkable shift in the prosecution policies of the national government. Prosecutors and courts increasingly followed the principle that the national government should be responsible for the education and well-being of criminal juveniles. The trend also indicates a growing concern for the criminal behaviour of children under 1658.

• 59 Jaarboekje over 1836-1846, uitgegeven op last van Z.M. den Koning (Jaarboekje van Lobatto), ’s-Grav (...)

47The growing focus of prosecutors on children of all ages involved girls as well as boys, and girls were not treated differently. Girls did not get more lenient treatment by the courts but received similar punishments for similar types of crime. In the period between 1840 and 1880 approximately 26 percent of the convicted children by Dutch courts received a prison sentence of between 12 months and two years, while around 20 percent of the children were confined for 5 to 10 years59.

• 60 Some studies on women’s work in the nineteenth century : Schilstra (1976) ; Stavenuiter, Dorsman (1 (...)
• 61 Van Poppel, Van Dalen, Walhout (2006) ; Braspennincx, Kalkhoven, 2009.
• 62 Van der Heijden, Schmidt (2010).

48What information do we have about the employment and occupations of the prosecuted women and girls ? It is often suggested that in the course of the eighteenth and nineteenth centuries prevailing ideologies about domesticity led to fundamental differences in the labour participation of men and women, with lower participation of women in the labour market. Feeley assumed that the exclusion of women and children from productive work and their confinement to the household led to a decline in female and juvenile crime, without providing evidence for any link between female crime rates and female participation in the labour market. There are only a few studies on the labour participation of Dutch women in the nineteenth century, and there is no consistent evidence about the labour patterns of women in this period60. However, recent work on women’s participation in the labour market indicates that there was no decline in the labour participation of women in the course of the nineteenth century. Demographic historian Van Poppel suggests that the Dutch industrialization around 1850 had little impact on women’s labour participation. Thus while industrialisation did not necessarily create employment for women, it did cause an increase in the exclusion of women either61. Recent examinations on the labour participation of Dutch women point to a different image, that of high levels of female labour participation well into the nineteenth century. The growing bureaucracy provided ever more employment opportunities, and women profited from this as much as men did62.

• 63 Ruitenbeek (2009, pp. 18-19 & 45).
• 64 Van der Heijden, Van den Heuvel (2007) ; Van der Heijden, Van Nederveen Meerkerk, Schmidt (2009) ; (...)

49There is substantial evidence to suggest that the alleged disappearance of women from the public sphere in the Netherlands over the nineteenth century simply did not occur. Evidence from the criminal court records of Amsterdam shows that most of the women who were prosecuted were in employment. Between 1811 and 1835, 62 percent of all female delinquents had an occupation at the time of arrest. Increasingly these women were prosecuted and convicted because of violence against neighbours, and less often because they had committed property offences or moral crimes such as prostitution or adultery63. These figures indicate that men and women’s activities in the public sphere do not reflect prevailing ideologies of domesticity. In practice women’s activities transcended the realm of the household, and many of them were not able – even if they wanted – to uphold the ideal of the husband as the breadwinner64.

• 65 In 1880 70 percent of the incarcerated women were employed, a proportion similar to the year of 185 (...)

50We have also information from the national prison records about occupations of those women who were sentenced to prison. These figures reveal a remarkably high degree of labour participation by women between 1835 and 1880. Throughout the period 70 to 90 percent of the women were employed by the time of their arrest, and there was no decline in the labour participation of these women. On average 28 to 39 percent of the female convicts were wage earners, often working in the textile industry, while around 30 percent of them worked as domestic servants. The rest earned their living through independent activities, such as traders, retailers, innkeepers or food sellers. The data show that there was no link between the degree of labour participation and women’s involvement in crime or female prosecution rates. According to the national prison statistics the proportion of women who were confined fell to 8 percent after 1877, though the proportion of women who worked before their detention continued to be high65.

• 66 Clark (2007, pp. 272-283) ; De Vries (2008, pp. 199-201).

51How can we explain the sharp decline in the proportion of incarcerated women in the last quarter of the nineteenth century ? Obviously, more research on the prisons records and workhouses is needed to draw definitive conclusions about the patterns of female prisoners after 1870. Nevertheless, there may be a plausible explanation for a decline in female crime rates towards the end of the nineteenth century. Economic historians, such as Gregory Clark and Jan de Vries, have argued that after 1850 women’s position in North-West Europe generally improved as a result of increasing living standards. Clark suggests that overall the equality between men and women in societies after 1850 increased. In addition to material income, there were also improvements in the quality of life in terms of life expectancy, health, numbers of surviving children, consumption and social care66.

• 67 De Haan (2000).
• 68 Weevers, Bijleveld (2010a).

52Dutch historian De Haan draws similar conclusions when she links improving living conditions to declining female crime rates and lower levels of female recidivism. De Haan points to the rise in women’s involvement in societies and committees devoted to the improvement of the living conditions and re-socialisation of female prisoners. From the second half of the nineteenth century onwards, changing legislation on illegitimate children and child support also improved women’s social and economic position, which may have diminished the despair which sometimes prompted women to commit theft or infanticide67. The evidence on the nineteenth-century female workhouses shows that those women who were sentenced to the workhouses were older women who lived alone, often with children, and employed in the lowest-paid jobs. Bijleveld and Weevers describe them as a rather exceptional and isolated group of women who committed crime as their final option to survive. These women were not bad but poor68.

• 69 Brugmans (1978 pp. 92-107) ; Smit (1996) ; De Kruik (2002) ; Schmidt, Van Nederveen Meerkerk (2006) (...)
• 70 Schenkeveld (2008, p. 31).

53How about the link between juvenile crime and the level of children’s labour participation ? Little work has been done and there is no general agreement among historians about the level of labour participation of children in various sectors during the nineteenth century69. Most historians argue that in Europe, child labour – both in industry and agriculture – declined over the nineteenth century because of the increasing focus on education and the introduction of laws that prohibited child labour70. The previous paragraphs showed that there was no decline in the Netherlands in the prosecution of young girls at criminal trials. Is there evidence of a stable pattern of labour participation of the boys and girls who were prosecuted ?

• 71 See for this problem also : Habermas, Schwerhoff (2009).

60The English and Dutch data show that there is no evidence for a long term decline in female crime from the end of the eighteenth century onwards. A complicating factor is the caesura in the study of crime. Historians and criminologist tend to focus on either the early modern period or the period after ca. 1800, and not much work has been done on the period of transition between 1800 and 192071. What we need is more consistent and comparative evidence that includes local and regional variation, differences between urban and rural areas, and jurisdictional differences between regions within Europe and beyond. Only then may we find an answer to the question whether, when and why female crime rates declined after 1850.