Gender Judgments: An Investigation of Gender Differentiation in Sentencing and Remand in New Zealand By Samantha Jeffries Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy Department of Sociology University of Canterbury Christchurch, New Zealand 2001 II ( ii ( Dedication To Dorothy Esme Jeffries 111 Acknowledgements I would like to thank Dr Greg Newbold who I have admired and respected since first hearing him lecture when I was a stage one sociology student. As an undergraduate, Dr Newbold encouraged me to undertake post-graduate study and as my Ph.D. supervisor, his enthusiasm, support, guidance and belief in my ability, ultimately brought this thesis to its fruition. Thanks to Associate Professor Garth Fletcher, not only for his guidance on statistical matters (for which he has been indispensable) but also for his ongoing support and concern for my welfare and my future. I would like to thank Alex Broom, Simon Campbell, Glennis Dennehy, Louise Humpage, Belinda Mortlock, and Brigid Thompson for their practical suggestions (including the allimportant question of fonts and heading styles), emotional support and, most importantly, their friendship. Thank you to Jessica (my beautiful and inspiring daughter) for putting up with a mum who was not always on this planet. My appreciation is further extended to Grant, my partner, for often bearing the brunt of my stress but always remaining calm and understanding. Thanks to my parents Phillip and Deanna Jeffries. Thank you to Lewis for de-stressing me with the simple things in life. Thanks to my brother Cy Jeffries - your sarcasm and cynicism always makes me laugh. My appreciation is extended to the University of Canterbury (including the Departments of Sociology and Psychology) and the New Zealand Federation of University Women. This thesis would not have been undertaken without the financial support you all so kindly provided. Finally, thank you to Dr Arnold Parr for many things but most importantly for telling me I could do it. iv ble Of Contents Page Dedication Acknowledgements Table of Contents List of Tables List of Figures List of Appendices Abstract 11 iii IV viii x Xl XlI Chapter One - Gender and the Criminal Justice System 1 Chivalry and the Criminal Justice System 1 Sex Differences in Criminality and Variance Across Judicial Decision-making Points 3 Sex Differences in the Decision-making Process Rather Than the Outcome 4 Towards a Social-Structural Approach to Judicial Decision-making 5 Gender Constructions - Troubled and Treatable Women, Acting and Punishable Men Ethnicity, Age, Class, and the Fulfilment of Gender Role Expectations Moving away from Chivalry Social Control Daly's Research - Social Control and Cost 5 6 10 12 14 New Zealand Research on Criminal Justice Processing 17 The Current Research 23 Research Setting Access Sampling Procedure Data Gathering Independent (predictor) Variables Dependent (criterion) Variables Sentencing Outcomes Imprisonment Suspended Sentences of Imprisonment Work-Based Sentences Care-Based Sentences Monetary Penalties Remand Outcomes Closing Comments 24 24 25 26 27 27 28 28 28 28 28 29 29 30 v Page Chapter Two - Wide Sample Characteristics, Descriptive Data And Sex Differences In Independent Variables Independent Variables Socio-Oemographic Factors Sex and Ethnicity Age Employment Status Overall Familial Situation Criminal History Variables Seriousness of Criminal History Bail Breaches Period Since Last Criminal Conviction Offence Seriousness Major Offence Category Seriousness of Violent Offence Seriousness of Drug Offence Seriousness of Property Offence Offender's Role and Co-offenders Offence Location Use of Weapons Value of Property Affected Victim-Offender Relationship - Violent Offenders Sex of Victim - Violent Offenders Victim Injury - Violent Offenders Court Process Factors Court of Sentencing Plea Number of Conviction Counts Pre-Sentence Reports Pre-Sentence Recommendations Overall Health Substance Abuse Negative Life Experiences Dependent Variables Remand Outcomes Remand Status Length of Custodial Remand Bail Conditions Sentencing Outcomes Imprisonment Sentencing Decision Length of Imprisonment Term Non-Custodial Sentences Suspended Sentences of Imprisonment Work-Based Sentences Care-Based Sentences Monetary Penalties 32 34 34 34 35 36 36 39 39 41 41 42 43 43 44 44 44 45 46 47 47 48 48 48 49 49 49 50 51 51 53 53 55 55 55 56 57 58 58 58 59 60 60 60 61 Closing Comments 61 Chapter Three - Hierarchical Multiple Regression Analysis 64 Sentencing Outcomes 65 67 Imprisonment Sentencing Oecision vi Page Length of Imprisonment Term Suspended Sentences of Imprisonment Work-Based Sentences Care-Based Sentences Remand Outcomes The Remand Status Decision Length of Custodial Remand Bail Conditions 72 75 76 77 78 79 82 85 Closing Comments 86 Chapter Four - Path Analysis 90 Sentencing Outcomes Imprisonment Sentencing Decision Length of Imprisonment Term Sentencing Outcomes - A Theoretical & Methodological Note Remand Outcomes The Remand Status Decision Length of Custodial Remand Bail Conditions 92 92 93 95 98 98 98 99 Closing Comments 100 Chapter Five - Regression Analysis: The Decision-Making Process 103 Sentencing Outcomes 104 Imprisonment Sentencing Decision Length of Imprisonment Term Suspended Sentences of Imprisonment Remand Outcomes The Remand Status Decision Length of Custodial Remand Bail Conditions 104 105 106 106 107 108 109 Closing Comments 110 Chapter Six - Offenders' Crime Stories 112 Characteristics of the Case-Study Sample Drug Offences Violent Offences Property Offences Crime Stories 112 113 113 114 115 Drug Offences Joan and Gary, Possession of Class C Drug for Sale/Supply Elaine and Alex, Selling/Supplying a Class B Drug Jill and Denis, Selling/Supplying a Class B Drug Sally and Warren, Selling/Supplying a Class C Drug Mary and Tim, Selling/Supplying a Class B Drug Kath and Chad, Selling/Supplying a Class C Drug Karri and Jim, Manufacturing a Class B Drug 121 124 125 126 128 129 130 131 Vll Violent Offences Sue and Mark, Aggravated Burglary Maria and Karl, Wounding with Intent to Cause Grievous Bodily Harm David and Jane, Assault with a Weapon Margo and Ben, Possession of an Offensive Weapon Property Offences Sally and Mike, Arson Cindy and James, Receiving Stolen Property Bridget and Rob, Arson Glenis and Todd, Receiving Stolen Property Page 131 134 135 137 138 140 142 143 144 145 Closing Comments 146 Chapter Seven - Pre-Sentence Reports, Judges' Sentencing Remarks, and Gender 148 New Zealand - Our Gendered Society? 151 The Construction of Gender in Probation Officers' Pre-Sentence Reports 154 156 158 160 161 162 166 Female Offenders and Domesticity Female Offenders and Family Pressures Female Offenders, Support Networks and Informal Social Control Women as Essential Family Figures The 'Family Man' and the 'Man Alone' Women's Pathology and Men's Action The Construction of Gender in Judges' Sentencing Remarks Female Offenders and Domesticity Female Offenders and Family Pressures The 'Family Man' and the 'Man Alone' Women's Pathology and Men's Action 173 175 177 178 181 Closing Comments 186 Chapter Eight - Closing Comments 190 Current Research Findings 191 191 194 196 Statistical Analysis and Findings Case-Study Analysis and Findings The Current Results and New Zealand's Criminal Justice System Theoretical Explanations 196 Equality or Difference 198 'Female' Vs 'Male' Focussed Justice 201 Closing Comments - Humane Justice 202 References 206 V111 List of Tables Page Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 Table 10 Table 11 Table 12 Table 13 Table 14 Table 15 Table 16 Table 17 Table 18 Table 19 Table 20 Table 21 Table 22 Table 23 Table 24 Table 25 Table 26 Table 27 Table 28 Table 29 Table 30 Table 31 Table 32 Table 33 Table 34 Table 35 Table 36 Table 37 Table 38 Table 39 Table 40 Table 41 Table 42 Table 43 Table 44 Ethnicity Employment Status Childcare Responsibilities Marital Status Living Arrangements Overall Familial Situation Bail Breaches During Present Trial or in The Past Major Offence Category Seriousness of Violent Offences Seriousness of Drug Offences Seriousness of Property Offences Co-Offenders Offender's Role in Offence Offence Location Use of Weapons Victim-Offender Relationship - Violent Offenders Sex of Victim - Violent Offenders Victim Injury - Violent Offenders Court of Sentencing Plea Number of Conviction Counts Pre-Sentencing Recommendations Physical Health Mental Health Overall Health Substance Abuse Negative Childhood Experiences Adulthood Victimisation Negative Life Experiences Remand Status Bail Conditions Imprisonment Sentencing Decision Suspended Sentences of Imprisonment Work-Based Sentences Care-Based Sentences Monetary Penalties Description of Variables Used For The Sentencing Analyses Zero Order Correlations - Imprisonment Sentencing Decision Imprisonment Sentencing Decision, Combined Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Imprisonment Sentencing Decision, Drug Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Imprisonment Sentencing Decision, Property Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Zero-order Correlations - Length of Imprisonment Term Length of Imprisonment Term, Combined Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Length Of Imprisonment Term, Drug Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses 35 36 37 37 38 39 41 43 43 44 44 45 45 46 46 48 48 48 49 49 50 51 51 52 53 53 54 54 55 56 57 58 60 60 61 62 67 69 70 71 72 73 74 75 ix Page Table 45 Table 46 Table 47 Table 48 Table 49 Table 50 Table 51 Table 52 Table 53 Table 54 Table 55 Table 56 Table 57 Table 58 Table 59 Table 60 Table 61 Table 62 Table 63 Table 64 Table 65 Table 66 Table 67 Table 68 Table 69 Table 70 Table 71 Zero-order Correlations - Suspended Sentences of Imprisonment Zero-order Correlations - Work- Based Sentences Zero-order Correlations - Care-Based Sentences Description of Variables Used For The Remand Analysis Zero Order Correlations - The Remand Status Decision The Remand Status Decision, Combined Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses The Remand Status Decision, Property Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Zero Order Correlations - Length of Custodial Remand Length Of Custodial Remand, Combined Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Length of Custodial Remand, Drug Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses Zero Order Correlations - Bail Conditions Bail Conditions, Combined Offences - Standardised Regression Coefficients From Hierarchical Regression Analyses The Imprisonment Sentencing Decision - Sex Differences In Predictor Variable Correlations Length of Imprisonment Term - Sex Differences In Predictor Variable Correlations Suspended Sentences of Imprisonment - Sex Differences In Predictor Variable Correlations The Remand Status Decision - Sex Differences In Predictor Variable Correlations Length of Custodial Remand - Sex Differences In Predictor Variable Correlations Bail Conditions - Sex Differences In Predictor Variable Correlations Seriousness of Drug Offence Seriousness of Violent Offence Seriousness of Property Offending Categories of Judicial Remand Responses and Outcomes for 21 Pairs of Drug Offenders Categories of Judicial Sentencing Responses and Outcomes for 21 Pairs of Drug Offenders Categories of Judicial Remand Responses and Outcomes for 14 Pairs of Violent Offenders Categories of Judicial Sentencing Responses and Outcomes for 14 Pairs of Violent Offenders Categories of Judicial Remand Responses and Outcomes for 15 Pairs of Property Offences Categories of Judicial Sentencing Responses and Outcomes for 15 Pairs of Property Offences 76 77 78 79 80 81 82 83 84 84 85 86 105 106 106 107 109 109 113 114 114 123 124 133 133 141 141 x List of Figures Page Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Figure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 Age, Combined Offences Period (In Months) Since Last Criminal Conviction, All Offenders, Combined Offences Period Since Last Criminal Conviction, All Offenders, Combined Offences (Log Transformation) Value of Property Affected Value of Property Affected - (Log Transformation) Number of Conviction Counts, Combined Offences Length of Custodial Remand Combined Offences Length of Custodial Remand Combined Offences (Log Transformation) Length of Imprisonment term Combined Offences Length of Imprisonment Term Combined Offences (Log Transformation) Path Analysis - Imprisonment Sentencing Decision, Combined Offences Path Analysis - Length of Imprisonment Term, Combined Offences Path Analysis Model One - The Remand Status Decision, Combined Offences Path Analysis - Length of Custodial Remand, Combined Offences Path Analysis - Bail Conditions, Combined Offences Path Analysis - Bail Conditions, Property Offences Daly's and Heidensohn's Models of Justice Path Analysis - Imprisonment Sentencing Decision, Drug Offences Path Analysis - Imprisonment Sentencing Decision, Property Offences Path Analysis - Length of Imprisonment Term, Drug Offences Path Analysis - Length of Imprisonment, Violent Offences Path Analysis - The Remand Status Decision, Property Offences Path Analysis - Length of Custodial Remand, Drug Offences 35 42 42 47 47 48 56 57 59 59 93 95 98 99 100 100 201 221 222 222 223 223 224 Xl List of Appendices Page Appendix One Additional Path Analysis Models 221 Appendix Two Supplementary Case-Study Results, Offenders' Crime Stories 225 xii Abstract This thesis addresses the complex and often contentious matter of gender difference in criminal court processing. In New Zealand, little substantive research has been conducted on the issue and this study attempts to remedy the deficit by providing an understanding of how gender operates in New Zealand's criminal justice system. To achieve this, a statistical analysis of 388 individuals, or 194 matched male/female pairs, was conducted. This was accompanied by a casestudy analysis of 50 matched male/female pairs, including investigation of individual crime stories, together with Judges' sentencing remarks and Probation Officers' pre-sentencing reports. It was found that: a) sentencing and remand outcomes often differed for adult men and women, with the former usually receiving 'harsher' sanctions, b) different factors were often considered when determining men's and women's judicial outcomes and, c) certain 'types' of men and women were more likely to be extended judicial leniency. In explanation, gendered ways of viewing, understanding and judging offenders indicated the manner in which judicial processing came to be differentiated by sex. For example, constructions of women as dependent, emotional, and traumatised by victimisation often appeared as explanations and excuses for women's offending. In contrast, such 'troubles' appeared as simply unbelievable or irrelevant in the case of men. In this way, men were denied justifications for their offending and they tended to be held fully responsible for their actions. These findings ultimately led to the question of whether male and female offenders should be treated the same or differently by the criminal courts. I argue that before this can be fully addressed, a more balanced understanding of gender and its impact on men's lives is required. I challenge feminist criminologists to transcend the boundaries of the equality/difference debate by problematising criminal justice processing as it relates to both sexes, rather than simply in terms of women versus men. 1 Chapter One Gender and th Criminal Justice stem "One of the outstanding concomitants of the existing inequality between the sexes is chivalry and the general protective attitude of man toward woman. This attitude exists on the part of the male victim of crime as well as on the part of the officers of the law, who are largely male in our society. Men hate to accuse women and thus indirectly send them to their punishment:, police officers dislike to arrest them, district attorneys to prosecute them, judges and juries to find them guilty, and so anN (Pollak, 1950 cited in Eaton, 1986: 22). After years of relative neglect, the female offender has, since the 1970s, become the subject of considerable international debate. A supposed increase in female crime during this period spulTed theories relating to women's liberation and criminality (Pollock-Byrne, 1990: 22-23). This was particularly the case in The United States of America, where throughout the 1970s and 1980s a number of writers concerned with the effects of the feminist movement on female criminality questioned whether the movement had inspired a female crime wave (see for example, Adler, 1975; Simon, 1975). Referred to as the 'female emancipation', or 'liberation' hypothesis, this perspective suggested that female crime was increasing, becoming less traditional, and more masculine in nature (Chesney-Lind, 1987: 114). This was seen as the result of increased female participation in the labour market and liberalising attitudes concerning traditional female gender roles that not only affected female criminality, but also the reactions of the public and the criminal justice systems to it. The once 'chivalrous' criminal justice system was now suggested to be hardening, or at least equalising, its stance towards the female offender (Steffensmeier, 1980; Kempinen, 1983: 3; Heilburn & Heilburn, 1986) and this set the scene for a second debate regarding the existence of gender disparity in judicial processing. 1 Chivalry and the Criminal Justice System In 1950, Otto Pollak claimed in his book, The Criminality of Women, that female offenders were preferentially treated in a criminal justice system dominated by men and thus characterised by male notions of chivalry. 1 Pollak argued that men are socialised to behave towards women in a fatherly and protective manner and female offenders are often compared to mothers and wives, whom the male judiciary cannot imagine behaving in a criminal way. 1 The word chivalry is a relic of medieval Europe where it represented a system of values, ideals, and a more refined 'gentlemanly' conduct. Essentially men were required to protect and fight for women who were the weaker more vulnerable sex (Moulds, 1978: 417; Julian, 1993: 344-345). 2 Pollak presumed that offending women were placed on pedestals, treated gallantly and protected from punishment, with the result that their criminal activity was less likely to be detected, reported, prosecuted, or sentenced harshly (cited in Tjaden & Tjaden, 1981: 74-75; Eaton, 1986: 22; Cain, 1990: 2; Culliver, 1993: 4; Belknap, 1996: 70-71, Newbold, 2000: 6566).2 Official crime statistics, both internationally and in Australasia, appear to support Pollak's thesis. 3 In Great Britain, women are more likely than men to be cautioned and less likely to be remanded into custody or sentenced to imprisonment (Hedderman & Hough, 1994). In the United States, women constitute only a small proportion of offenders arrested, convicted, and imprisoned (Feinman, 1980: 22; Daly & Tonry, 1997: 201-202). Similar findings are noted in Australia where the raw statistics suggest leniency is afforded to women. For example, a recent Australian study on 'homicides between sexual intimates' found that women are more likely to be released on bail, less likely to be convicted of murder and generally received lighter sentences than men (Easteal, 1993 cited in Alder 1994: 150). In New Zealand, national statistics show women are less likely than men to be convicted of an offence or sentenced to imprisonment but are more likely to have their cases discharged (Spier, 1997: 45; Newbold, 2000: 67). Once imprisoned, New Zealand women receive shorter terms than men and are more likely to be granted early release on parole (Lash, 1998:83; Newbold, 2000: 67). International analysis of Second and Third United Nations Crime Surveys by Harvey, Burnham, Kendall and Pease (1992: 208) further shows that men are "disproportionately suspected, apprehended, prosecuted, convicted, and imprisoned throughout the world." Despite what the 'raw statistics' may indicate, research addressing the question of gender difference in criminal justice processing has produced mixed findings, thus challenging the idea of 'chivalry' (see Simon, 1975: 49-67; Armstrong, 1977: 109-110; Chesney-Lind, 1987: 124131; Laster, 1994: 67-68; Daly & Bordt, 1995). Sex differences in legal variables (such as criminal history or current crime seriousness) and extra-legal factors, particularly gender related characteristics 4 such as family or economic situations, have been noted to account, at least partially, for disparate judicial outcomes. In the case of extra-legal factors, sex is placed in a "social framework" with researchers looking not just at sex itself, "but at the statuses The idea that women are treated chivalrously by the judiciary was actually noted first by Thomas (1907) in his book Sex and SOciety. However, this notion of chivalry was largely ignored until Pollak revived it in 1950 (cited in Nagel & Hagan, 1983: 113). Since then Pollack's name has become synonymous with the chivalry thesis. 3 Pollak's conclusions were also reached after examining official conviction rates (see Curran, 1983: 42). 4 Defined as "any aspect of social life that is particularly characteristic of one sex" (Kruttschnitt, 1984: 214). 2 3 associated with gender (such as parent and economic provider) that may influence judicial decision-making" (Flavin, 1998: 156). Sex-based variability has also been noted across judicial decision-making points (e.g. plea, pre-trial release, conviction, sentencing) and some researchers have found that the judicial decision-making process, rather than the outcome, differs for men and women. For example, the weighting given to criminality could vary, according to whether the offender is male or female. Finally, sex differences in the treatment of certain 'types' of men and women have been identified. For example, preferential treatment may only be extended to white middle class women and/or women who are care givers (Belknap, 1996: 73-77; Roberts, 1995). Each of these issues will be discussed in turn. 2 Sex Differences in Criminality and Variance Across Judicial Decisionmaking Points One of the most pervasive attacks on the 'chivalry thesis' is that women's law breaking is usually less serious or fundamentally different from men's and these legal characteristics do, or should, legitimately result in less severe judicial outcomes for women (see for example, Kruttschnitt & McCarthy, 1985; Raeder, 1993; Daly, 1994; Chesney-Lind, 1997; Gelsthorp & Loucks, 1997; Carlen, 1998; Flavin, 1998: 149). Studies failing to consider more than one judicial decision-making point (e.g. those looking at sentencing but not remand) may also be problematic because sex differences may vary or cumulate across different decision points (Wilbanks, 1986: 518). Research which has examined sex differences in judicial outcomes at various points in the criminal justice process and/or controlled for sex differences in legal variables yields mixed results (see for example, Fenster & Mahoney, 1981; LaBeff, 1981; Nagel, 1981; Tjaden & Tjaden, 1981; Nagel, Cardascia & Ross, 1982; Steffensmeier & Kramer, 1982; Croyle, 1983; Curran, 1983; Farrington & Morris, 1983; Frazier, Bock, & Hemetta, 1983; Kempinen, 1983; Gruhl, Welch & Spohn, 1984; Kruttschnitt, 1984; Kruttschnitt & Green, 1984; Zingraff & Thomson, 1984; Spohn, Welch & Gruhl, 1985; Ghali & Chesney-Lind, 1986; Heilburn & Heilburn, 1986; Wilbanks, 1986; Daly, 1987b; Douglas, 1987; Johnston, Kenndy & Shuman, 1987; Mair & Brockington, 1988; Daly, 1989a; Lindquist, White, Tutchings & Chambers, 1989; Bickle & Peterson, 1991; Crew, 1991; Boritch, 1992; Champion, 1993; Steffensmeier, Kramer & Streifel, 1993; Daly, 1994; Farnworth & Raymond, 1995; Spohn & Spears, 1997; Mustard, 1999). For example, Curran (1983) examined judicial processing at four levels of the criminal justice process (negotiations, prosecution, conviction and sentence) and controlled for the effects of legal factors (including current offence seriousness and criminal history) and 4 non-legal factors (race, age and occupational status). Results indicated inconsistent differences in treatment by sex at each stage of the process, and it could be concluded only that preferential treatment was afforded to women at sentencing (Curran, 1983: 54). Gruhl, Welch & Spohn (1984) examined the decisions to charge, plead guilty, convict, and imprison. Extensive legal controls measuring criminal history and current crime seriousness were introduced along with non-legal controls including age, ethnicity, employment status, plea and attorney type. Findings showed that female offenders were treated more leniently than men within the prosecution's decision to dismiss the charges and in the Judge's decision to incarcerate. No differences were found between men and women regarding the decision of the Judge and Jury to convict and offenders' decisions to plead guilty. Increasingly, evidence has suggested that preferential outcomes for criminal women were not as pervasive as the chivalry hypothesis proposed. In their review of the research, Parisi (1982) and Nagel and Hagan (1983) concluded that the relationship of sex with court processing varied from stage to stage. Nevertheless, women still tended to receive preferential treatment at the point of sentencing and pre-trial release. In a more recent analysis of statistical sentencing studies, sex differences were found to range from eight to 25 percent, with women's sentences being less severe than men's. Incarceration periods also varied, with men being sentenced to an average of about twelve months longer than women when appearing before the court under supposedly similar circumstances (Daly, 1994: 32; see also Daly & Bordt, 1995 for a review of the statistical literature). 3 Sex Differences in the Decision-making Process Rather Than the Outcome Different factors may also be considered when determining men's and women's culpability and punishment (see for example, Kruttschnitt, 1982a; Kruttschnitt, 1982b; Visher, 1983; Kruttschnitt, 1984; Sarri, 1986; Bickle & Peterson, 1991; Crew, 1991; Steffensmeier, Kramer & Streifel, 1993; Farnworth & Raymond, 1995; Belknap, 1996: 72). For example, some studies have found that legal variables such as offence characteristics, judicial processing variables or criminal history impact on men's judicial sanctioning more than women's (see for example, Nagel, 1981; Farrington & Morris; 1983; Kruttschnitt & McCarthy, 1985; Boritch, 1992; Channels & Herzberger; 1993; Farnworth & Raymond, 1995). In one study, Nagel (1981: 112) found that offence severity and criminal history had no significant effect on the likelihood of incarceration for women. In contrast, men were adversely 5 affected by these factors - thus, committing serious offences and/or having prior convictions substantially increased the likelihood of imprisonment. Kruttschnitt & McCarthy, (1985: 168) found that the "most consistent evidence of gender bias" in their study appeared in the differential weighting of predictor variables. In this case, pending criminal cases and prior offence history were generally found to be more significant predictors of pre-trial release for men than women over the 15 year study period. A larger number of pending cases and more serious criminal histories adversely affected men's chances of remaining free between arrest and case disposition. Similarly, in research on pre-trial decision-making by Channels and Herzberger (1993), the criteria used to make bond decisions for men and women differed. For men, bond type and amount was significantly affected by the characteristics of the current crime and prior record, while for women none of these variables were significant. It has also been suggested that victim characteristics may influence the judicial sanctioning of men and women differently (see for example, Visher, 1983; Daly, 1994). Visher's (1983; 18) study found that the victim-suspect relationship was an important factor in police officers' arrest decisions but while knowing the victim increased men's chances of being arrested it did not impact on female suspects. Prior studies have also found that crimes involving female victims may be punished more harshly if the offender is a male but it is suggested that this may not be relevant for female offenders whose female victims are less vulnerable in comparison to them (Daly, 1994: 108-110). 4 Towards a Social-Structural Approach to Judicial Decision-making Extra-legal factors (such as ethnicity, age, class, economic position, domestic responsibilities and mental health) appear in the studies presented below and reflect research focussing on a more social structural understanding of sex differences in judicial sanctioning. First, these studies show how gender operates in the criminal justice system to make punitive sanctioning more likely for men. Second, they demonstrate how gender operates to ensure differences in the treatment of certain 'types' of men and women. 4.1 Gender Constructions Troubled and Treatable Women, Acting and Punishable Men m Frazier, Bock, and Henretta (1983) examined the relationship between convicted offenders' sex and sentencing severity. Using a statistical analysis, these researchers found 6 original sex differences, which favoured women, disappeared once pre-sentence report recommendations were controlled for. Further investigation showed imprisonment recommendations were less likely for women, even after statistically controlling for important offender characteristics, legal variables and prior decisions, because strained familial relationships or mental health problems explained women's crime. In comparison, male crime was seen to be related to irresponsible behaviour and men were presented as more culpable and dangerous than women. Thus sex differences in sentencing recommendations could be rationalised. Incarceration sentences were deemed more appropriate for women who needed treatment, while imprisonment was needed for men because they were seen as dangerous and irresponsible (Frazier, Bock and Henretta 1983: 315). Allen's research (1987a; 1987b; 1987c) found that psychiatric rather than penal rationales were used to decide outcomes for women more often than for men. Like those studied by Frazier, Bock, and Henretta (1983), female offenders in Allen's research were presented more often as 'mad', as victims of personal misfortune and thus not altogether responsible for their criminality. In contrast, men were more likely to be presented as 'bad', and "active, intentional creatures, who are inherently responsible for their actions" (Allen, 1987c: 109). These gendered constructions resulted in women being seen as weak, troubled, unable to control their behaviour and as less culpable than men. Under these circumstances, the 'disordered' female offender was more likely to be judged as needing help rather than punishment. It would seem that that the 'sexual politics of sickness' is a common feature of criminal justice discourse (Heidensohn, 1996: 95). In another study conducted by Worrall (1990), criminal women were again constructed within discourses of pathology. Worrall (1990) argues that reconciling women's criminality within a dominant discourse of appropriate femininity, like pathology, allows criminal women to be refeminised as 'real' women who subsequently deserve treatment rather than punishment. Domestic problems were also prominent in judicial discourses about women and presented as an explanation or excuse for female offending. Worrall (1990: 60) states that "women might be reduced to breaking the law either directly by insufferable husbands ... or indirectly by the pressures of family life." 4.2 Ethnicity, Age, Class, and the Fulfilment of Gender Role Expectations Gender Differences in the Treatment of Certain 'Types' of Men and Women Research grappling with the possible interplay between race/ethnicity, gender and judicial processing is a recent phenomenon (see for example, Kruttschnitt, 1982a; Kruttschnitt, 7 1982b; Visher 1983; Gruhl, Welch & Spohn, 1984; Spohn, Welch & Gruhl, 1985; Sarri, 1986; Spohn, Gruhl, & Welch, 1987; Daly, 1989a; Lindquist, White, Tutchings & Chambers, 1989; Bickle & Peterson, 1991; Spohn & Spears, 1997). Research results indicate that ethnic minority women may be treated more harshly than their European counterparts. Daly and Tonry (1997: 229-230) comment that research findings generally indicate sex "differences within race groups but not racial or ethnic differences within gender groups." The age of a female offender has also been found to affect her at all stages of the judicial process. Indeed, the treatment of young juvenile women in the criminal justice system is often cited as one of the strongest challenges to the idea that preferential treatment is being afforded to female offenders (see for example, Saunders & Wattenberg, 1954; Gibbons & Griswold, 1957; Morris, 1965; Chesney-Lind, 1973; Kratcoski, 1974; Chesney-Lind, 1977; Hancock & Hiller, 1981; Sheldon, 1981; Teilman & Landry, 1981; Mahony & Fenster, 1982; FigueiraMcDonough, 1985a, Mann, 1985; Chesney-Lind, 1987; Figueira-McDonough, 1987). The need to care for and protect juvenile female offenders for their own good is seen to result in harsher treatment in comparison to males. Writers have noted the operation of a double standard in the criminal justice system, which has consistently resulted in the harsher treatment of girls, and has subjected them to controls which are related more to dominant gender ideals than their offence (Chesney-Lind, 1973; Smart, 1976: 131-134; Chesney-Lind, 1977; ChesneyLind, 1995: 87-92). Eaton (1986: 25) notes that girls who come to the attention of the criminal justice system: ... are more likely than their male counterparts to be questioned about their sexual activity, more likely to have their offences viewed as an aspect of sexual promiscuity, and more likely to lose their liberty for activities which would not be against the law if committed by an adult. Chesney-Lind (1987: 117) argues that the justice system's treatment of female juveniles may well constitute "one of the clearest examples of institutionalized sexism in contemporary society." This is largely due to the juvenile courts' "ability to take youth into custody for a variety of non-criminal status offences", which include, for example, such activities as "running away from home, curfew violation", or "being a person in need of care and protection." Despite the fact that self report studies indicate otherwise, 5 research has tended to indicate that girls are over represented in the status offence category (Chesney-Lind, 1995: 88). In comparison to boys, girls are more likely to be arrested, charged, and punished with status 8 offences (see for example, Wattenberg & Saunders 1954; Gibbons & Griswolds 1957; Kratcoski, 1974; Chesney-Lind 1977; Hancock & Hiller, 1981: 92-126, Teilrnan & Landry, 1981). Like ethnicity, class and gender may intersect to benefit some women over others. Ideologically speaking, economic dependence, particularly on a spouse or husband, may be seen as a desired feature of womanhood. Any class-based leniency given to a woman by the judiciary appears to be dependent on the earnings of her husband rather than her own individual income. Worrall (1990: 88) reveals that "respectable, middle-class wives and mothers" are assumed to be more susceptible to reform, thus needing little if any punishment while working class women are perceived as 'tough', less feminine and thus in need of harsher punishment. Studies show that judicial treatment is generally harsher for women working outside the horne than for those working within it. Men, in contrast, benefit from involvement in paid work and if employed are treated less harshly than their unemployed counterparts (see for example, Farrington & Morris, 1983; Kruttschnitt, 1984; Kruttschnitt & McCarthy, 1985; Crew, 1991). For offending women, domesticity and dependence appear to be the traits mitigating punishment. Although possibly subject to change over time and across judicial decision- making points, some researchers have found that marital status has relatively little impact on men's judicial processing compared with women (see for example, Nagel, 1981; Nagel, Cardascia & Ross, 1982; Farrington & Morris, 1983; Boritch, 1992; Channels & Herzberger, 1993). An American study by Nagel, Cardascia, & Ross (1982) found that married women were less likely than unmarried women to spend any time imprisoned. Farrington and Morris (1983), in their study of Cambridge Magistrates' Courts in England, discovered that divorced or separated, and childless women received relatively harsh sentences in comparison with other more domesticated women. Worrall (1990) argues that domesticity, (being a wife, mother and working within the horne) is another way in which the judiciary can 'referninise' criminal women. Offending women who are wives and mothers or otherwise express acceptable feminine attributes such as dependence can be repositioned as real women and thus be deserving of leniency. However, in cases where offending women's lives are at odds with acceptable femininity, they may be treated more harshly. Similarly, when women commit 'unfeminine' crimes, they may also receive little judicial sympathy. 5 Figueria-McDonough (1985a: 277) for example found that virtually the same proportion of girls and boys were involved in status offences, apart from sexual activities, which were reported more by males. Criminal offences, such as violent offences and property offences, however were characteristiced by greater male involvement. 9 In addition to women's non-criminal behaviour, Edwards (1986: 80) argues that treatment in the criminal justice system may depend on the degree to which a woman's criminal behaviour can be reconciled within dominant gender discourses. Offending which deviates too far from 'femaleness' or 'femininity' may result in women being sanctioned at least, if not more, harshly than men convicted for similar crimes. A review of the research indicates that preferential treatment is greatest for women when more serious offences are analysed, but when minor offences are considered, sex differences are minimal. However, there is still evidence of gender differences in judicial treatment for specific types of crime within these broader serious/minor offence categories (Chesney-Lind, 1987: 129-131). More specifically, it appears that 'traditional' female crimes (i.e. prostitution or shoplifting), may result in judicial leniency for women while committing 'non-traditional' crimes (e.g. certain types of violence) may result in harsh treatment (Simon, 1975: 52, Feinman, 1980: 24; Parisi, 1982: 209, Wilbanks, 1986; Chesney-Lind, 1995: 91-92). Bishop & Frazier (1984: 386) explain why this may be: It is argued that crime is viewed symbolically as masculine behaviour and that when a woman commits an offence - especially a violent offence - she is punished with a vengeance. She is punished both for the crime and for exceeding the bounds of genderappropriate behavior. . Several studies provide empirical support for the argument that women whose crimes conflict with dominant gender ideals will not be extended preferential judicial treatment. For example, Nagel, Cardascia and Ross (1982: 269) examined sex effects on criminal justice decisionmaking within offence types and found a strong negative effect for females charged with personal crimes (interpersonal violence), but no significant effect for males similarly charged. Figueira-McDonough's (1985b: 113) study on the processes of charge bargaining and sentence reduction found that the presence of a weapon was "significantly associated with pleas only for women, increasing significantly the likelihood of bargaining." She concludes that the "use or possession of a gun adds seriousness to an offence only when perpetuated by a woman." These cases suggest that preferential treatment is only being reserved for a particular group of women - those who commit less 'masculine' crimes. However, further research has since found no evidence of sex differences in the judicial processing of specific more 'masculine' type offences or less 'masculine' crimes (see for example, Bishop & Frazier, 1984; Wilbanks, 1986; Steury & Frank, 1990; Farnworth & Raymond, 1995; Spohn & Spears, 1997). 10 5 Moving Away From Chivalry In the literature, chivah'y is often presented as having only positive effects on women. A "courtly gentlemanliness" is implied on the part of Judges, who are depicted as wanting to protect women from punishment (Odubekun, 1992: 352). Women are seen to benefit from this relationship in that 'chivalry' results in men placing women on pedestals and treating them gallantly (Belknap, 1996: 71). However, as noted, the research indicates that not all women are extended 'preferential' treatment over all men all the time. Thus judicial decision-making involves a more complex process than originally suggested by the 'chivalry thesis'. In the literature, the term 'chivalry' is often used interchangeably with the term 'paternalism'. However, I feel that it is important to distinguish between the two because the latter appears to have emerged partly as a critique of the former. Unlike chivalry, paternalism is often depicted as having both positive and negative effects. It is acknowledged that preferential treatment is only given to some women, over some men, at some stages of the criminal justice process. Moulds (1978: 418) argues that the concept of 'chivalry' is only helpful in so far as it describes "the superficial elements in male-female relationships, namely, the social amenities." Women are seen to benefit from this relationship in that 'chivalry' results in men placing women on pedestals and treating them gallantly (Belknap, 1996: 71). In contrast, 'paternalism' acknowledges gender-based power relationships (Odubekun, 1992: 353). Whether women benefit from their exchanges with predominantly male law enforcement officials, varies and depends on the fulfilment of gender role expectations. 6 Unlike 'chivalry', 'paternalism' thus proposes that there are costs as well as benefits to the 'differential' treatment of female offenders. Although the initial benefits to women may be their freedom, the major cost is the reinforcement of dominant gender role ideology, which keeps women dependent and powerless, denies female agency, and holds that women are less able than men and in need of special protective treatment. 'Paternalism' represents an improvement over 'chivalry' but both have been criticised as similarly misguided and overly simplistic. 'Paternalism', like 'chivalry', assumes that sexism is overtly practised by the judiciary; it is seen as something that is 'done to' offenders. As such, unequal treatment is considered problematic because protecting women is construed as an ideological front for patriarchy rather than an expression of judicial concern for a subordinate In the context of paternalism and gender role expectations, some commentators have explained contradictions in the lenient treatment of some women over others by juxtaposing 'chivalry' against what is called the 'evil woman' thesis. However, like Nagel and 6 11 group (Daly, 1989b: 12). Conversely, even if some women appear to be treated preferably, why do we presume this reflects concern for the female offender? It could be argued that differential treatment is not necessarily misplaced because differences in men's and women's social positioning can, and should, legitimately result in varied judicial treatment. To expand, it may be the case that judicial concern is being directed at the welfare of innocent third parties, such as children and other family members who would be adversely affected by the harsh treatment of women. This sentiment is reflected in the following statement from New Zealand District Court Judge Ron Young (1997: 22), who notes that: As a broad principle women cannot avoid imprisonment simply because they have a family to care for. However the fact a woman is a care giver is often highly relevant to sentencing choice. It can reduce a sentence of imprisonment on the rationale that the consequences for the offender and family are greater. Further, the criminal justice system's differential treatment of men and women could reflect fundamental differences in the lives of men and women. To expect, or want, a gender-neutral approach could harm women who already occupy a subordinate position in the social structure. The criminal justice system (like society) is dominated by men, based on male notions of justice, and designed primarily to deal with male offenders. Thus, as long as it does not put women in a negative position, treating men and women differently is viewed positively by some feminist legal scholars who have put forward a "sort of separate but equal argument" based on the idea that men are not the same as women (see Daly & Chesney-Lind, 1988: 524-526; Raeder, 1993; Chesney-Lind, 1997: 162). Reflecting on social structural differences between men and 'women has moved researchers beyond notions of chivalry, gender roles, and overt sexism, toward a more fully integrated understanding of gender and the criminal justice system (Flavin, 1998). Research conducted by Daly (1987a, 1987b, 1989a, 1989b, 1994), Kruttschnitt (1982b & 1984; Kruttschnitt & Green, 1984; Kruttschnitt & McCarthy, 1985) and Eaton (1986, 1987), for example, considered linkages between gender ideology, social control, social cost and how general understandings and constructions of gendered difference in men's and women's criminal and non-criminal lives can explain differential treatment. Hagan (1983: 135-136) I do not see these two perspectives as being in opposition to each other. Rather, they are complimentary and I feel are brought together under the concept of 'paternalism" which acknowledges gendered variation iri judicial treatment. 12 5.1 Social Control Proponents of the social control hypothesis assert that there is an inverse relationship between formal (state) control, and informal control. While past researchers argued that preferential treatment was dependent on fulfilling gender role expectations, social control theorists instead hypothesised that the more an individual was subject to informal social control, the less need there was for the judiciary to impose formal controls to prevent future offending (Steury & Frank:, 1990: 418-419). Women are more likely than men to experience a high degree of informal social control via their dependent status in the family (see Hagan, Simpson & Gillis, 1979; Kruttschnitt; 1982b: 496-498). In comparison to men, for example, women's lives are characterised by economic dependence on a husband, partner, or the state and the high level of supervisory activity which comes from living with others (such as a husband/partner, children, other relatives) (Bickle & Peterson, 1991: 373). To test this informal social control thesis, Kruttschnitt (1982b: 498) statistically analysed Probation Officers' reports for 1034 female offenders processed by a Californian probation department between 1972 and 1976. All offenders were convicted of either disturbing the peace, assault, forgery, or drug offences. Kruttschnitt (1982b: 496) hypothesised that economic dependency is a gender related status which "connotes informal social control" and "that the quantity of informal social control inherent in a dependency status may explain the sentences that female criminal defendants incur." To test this assertion, Kruttschnitt (1982b: 502-503) created a statistical measure of informal social control, by assessing levels of economic dependency in women's lives. Subsequent statistical analysis supported Kruttschnitt's (1982b) thesis, as did a further qualitative analysis of Probation Officers' reports which showed evaluations often focused on women's dependency. Overall, less formal (state-imposed) control was recommended to the courts by Probation Officers when the women were in a situation of strong economic dependence. Kruttschnitt (1982b: 508-510) concludes that these findings suggest that the legal system prefers to exert little formal control over women whose lives presently contain a degree of daily informal social control, such as that of economic dependency. Taking this research further, Kruttschnitt (1984), Kruttschnitt & Green (1984), and Kruttschnitt and McCarthy (1985), conducted analyses of criminal court sanctioning (pre-trial release and sentencing) in Minnesota, t'a test the hypothesis that fundamental gender differences in the social positioning of men and women could explain women's less severe judicial 13 outcomes. Analyses conducted by Kruttschnitt & McCarthy (1985) again introduced measures for familial social control. Pre-trial release statuses for offenders over a 16 year period were examined and women were found to receive preferential outcomes in twelve of these years. Once familial social control was entered into the statistical equations, women's preferential treatment was eliminated in five of the twelve years where sex was shown to be a significant predictor of the pre-trial release decision. Kruttschnitt (1984) and Kruttschnitt & Green (1984) also held constant sex-based familial statuses in their analyses to see if initial sex differences in criminal sanction would disappear and thus be explained by gender differences in family circumstance. Kruttschnitt (1984) continued to find sex differences in sentencing and remand with more 'severe' outcomes remaining more likely for men regardless of these controls. Kruttschnitt and Green (1984) discovered that the decision to incarcerate was significantly affected by pre-trial release status, and that women were often the beneficiaries of pre-trial freedom, but holding constant structural gender differences significantly reduced this apparent leniency (Kruttschnitt and Green, 1984: 550). Like Kruttschnitt, Eaton (1986: 29) thought that differences in the degree of formal control imposed on offending men and women might reflect differences in the amount of informal social control in their lives. Eaton (1986, 1987) used a case-study approach to assess the importance of gender on bail and sentencing outcomes in a Magistrates' Court outside London. After extensive court observation, interviews with judicial personal, and document analysis, Eaton (1986: 97) concluded that men and women were treated similarly when they appeared under similar circumstances. However, it was rare for men and women to appear before the court under similar circumstances, so differential treatment was not surprising. In the courtroom, male and female offenders' economic and familial circumstances not only varied but were also expected to differ substantially. Men were expected to provide economically for their families. In comparison, women were expected to be responsible for childcare and emotional support. Eaton (1986: 95-98) thus identified the criminal justice system as a site of cultural reproduction where a dominant model of the 'ideal' family was supported and actively reinforced by courtroom discourse. Furthermore, this model of the family was acknowledged by the judiciary as an important source of control comparable to that offered by the criminal justice system (Eaton, 1987: 107). At best, Kruttschnitt and her colleagues found only partial support for the thesis that the quantity of informal social control inherent in women's dependency may explain why men 14 receive more severe judicial outcomes. However, Eaton and Kruttschnitt directed attention to how men's and women's social locations might explain treatment in the criminal justice system. Daly (1987a, 1987b, 1989a, 1989b) developed this social control hypothesis further. Daly revised the work of Kruttschnitt and added a dimension of social cost into her theory of gendered judicial decision-making. 5.2 Daly's Research - Social Control and Cost; Gendered Criminality, Lives and Judicial Decisions. Daly (1987b: 154) argues that there were two problems with Kuttschnitt's social control argument: "the locus of informal social control is misspecified, and gender differences in court outcomes cannot be adequately explained by it." While economic dependency within the confines of the family may be the reality for many women, she says, it is also the case that women will have others who are dependent on them for their care-giving labour. Daly (1987b: 154) subsequently argues that "when theorizing about the impact of familial relations as informal social control, greater emphasis should be given to whether male and female defendants have dependents, not solely whether they are dependent on others." In addition to widening the locus of informal social control, Daly (1987a, 1987b, 1989a, 1989b) also argues that there are social cost components to sentencing. Court officials had indicated concern for families, rather than for women, and this Daly says, could explain sex differences in judicial outcomes. Statistical analyses of pre-trial release decisions, sentencing and dismissals, found that "net of case severity, charge severity, the type of offence charged, prior record, and other defendant characteristics, male and female defendants are treated differently on the basis of their ties to and responsibilities for others" (Daly, 1987b: 167). Men and women with strong familial ties spent shorter periods in pre-trial custody and were sentenced less severely than non-familied offenders. Sex differences in judicial outcomes were subsequently seen to involve both social control and social cost. First, the courts did not need to formally control offenders who were seen as informally controlled within families, which promoted responsible behaviour. Second, the judiciary extended preferential treatment to familied offenders because of the social cost of removing them from families (Daly, 1987b: 155, 1989b: 27). It is argued that observed sex differences in judicial outcomes, demonstrate the higher value that the criminal court places on the caring/nurturing role, usually performed by the mother, compared with the economic 15 breadwinner role, usually performed by the father (Daly, 1987a, 1987b, Daly, 1989a, Daly, 1989b). Daly (l989b: 27) notes from her interviews with Judges that: The judges justified different sentences for the family men and family women, based on a labor hierarchy, in which the care of children was primary and Eilconomic support was secondary for maintaining families. Most judges said that the only factor that distinguished their sentencing of men and women (other case factors being equal) was that women cared for children. Indeed, familied men who were responsible for the care of others were found to be treated the same as their female counterparts. Thus it is concluded that any apparent leniency afforded to female offenders is in fact the result of concern for minimising the cost to society and to innocent third parties, of removing caregivers from families. In 1994, Daly conducted further research into the question of gender and judicial decisionmaking. This research set a new methodological standard by closely meshing statistical and case-study material to describe and compare offenders' criminal cases and lives. In doing so, this study provided a clear understanding of how gender differences in criminality and constructions of men's and women's biographies may offer justifiable explanations for differences in judicial treatment. First, Daly (1994: 25-35) conducted a statistical analysis which looked at the impact of sex on sentencing while controlling for other legal and non-legal factors typically included in statistical investigations of this kind. Like many other statistical studies, it was subsequently found that women were less likely to be sentenced to imprisonment, and once they received imprisonment were given shorter terms than men. Second, Daly (1994) selected a smaller case-study sample from the larger statistical sample and conducted a pair-wise comparative analysis. The case-study sample was a good representation of the larger data set and consisted of 40 men and women who were as alike as possible with respect to statutory charges, age, race or ethnicity and pre-trial status. Crime details for each offender pair were analysed to determine how alike women's and men's crimes actually were. Offenders' biographies, as presented to the court, were also examined to determine whether "gendered pathways" into the judicial system might explain differences in sentencing. In the latter biographical analyses, gender was viewed as an exchange where cultural ideals of masculinity and femininity were constructed between offenders and other judicial actors. Rather than seeing the judiciary as overtly sexist, as suggested by proponents of 16 'chivahf or 'paternalism' theory, the criminal justice system is instead seen as a site where gender is actively constructed. Daly (1994: 259) explains: When comparing the pathways of women and men, the role of social construction becomes clear: we cannot be sure what part of gender difference is real and what part is amplified or attenuated through social construction. For example, the PSI reports suggested that three times more women than men were either physically or sexually abused while growing up ...These differences may reflect real differences in women's and men's lives. Simultaneously, they may reflect how experiences are elicited in interactions between a probation officer and a defendant, and how biographical story lines take gendered forms. Daly's (1994) research findings showed that women were more likely to be the sole caregivers of children, but while familial circumstances were important, they were secondary to the gender differences in the crime and defendant's prior record (Daly & Bordt, 1995: 161). Men's and women's criminality may have been coded for the purposes of statistical analysis as similar but further case-study investigations revealed they were in fact different. Once the details of the crimes were read more closely "only one pair evinced unwarranted gender disparity" (Daly & Bordt, 1995: 161). Generally, women's criminality was simply less serious than the men's and this explained why they received less severe sentencing outcomes. Furthermore, "boundaries between victim and offender were more often blurred in the women's social histories." In contrast to men, women were more likely to be constructed as 'scared by victimisation' and this rendered them "less blameworthy, more a product of past or current problems than a chosen course of action" (Daly, 1994: 259-260). Finally, Judges expressed more optimism in women's potential to reform (Daly & Bordt, 1995: 163). Daly (1994: 263) concludes that ... in the New Haven court, the gender story has proven to be largely one of men's failures, men's greater blameworthiness, and men's somewhat poorer prognosis for change. Measured against the biographies and offences of men, women looked better. Given the new methodological standards set by Daly (1994), future research should consider gender differences both in offenders' criminality and in judicial constructions of biography in more detail. First, a comparative content analysis of men's and women's lawbreaking should be undertaken to ensure that apparently 'like' criminality is in fact alike (Daly; 1994: 260, Daly & Bordt, 1995: 162). Second, researchers need to document and understand how gender-based determinations enter into criminal justice decision-making (Daly; 1994: 260). For example, how are men's and women's pathways into the criminal justice system constructed and how do judicial decision-makers determine sentencing outcomes differently for men and women? In short, researchers need to go further than statistical analysis alone by employing case-study 17 techniques if a comprehensive understanding of how gender operates III criminal justice decision-making is to be achieved (Daly, 1994: 258-269). 6 New Zealand Research on Criminal Justice Processing In New Zealand, little is known about the impact gender has on judicial processing. Aside from women's victimisation and prostitution, questions about women, gender and criminal justice have only been asked in recent years (see Deane, 1995; 1997; 2000; Barwick, Burns and Gray, 1997; Triggs, 1999). These New Zealand studies are discussed below but none are as thorough or methodologically robust as Daly's (1994) research or many other international studies. Deane's work (1995; 1997; 2000) is the first academic consideration of gender and sentencing in New Zealand. Deane (1995; 1997; 2000) used both a statistical and case-study approach to examine two different samples of offenders convicted of what would be considered less serious offences. The ftrst sample included all offenders sentenced in Wellington and Porirua District Court between 1989 and 1991 for unlawful taking of property, all drug offenders, and all those convicted of false pretences (n=955). The second sample, identified as a case-study sample, was taken from the Wellington District Court and included all female offenders (n=52) and a sample of male offenders (n=165) who, irrespective of offence, were sentenced over a three month period during 1993. Deane undertook courtroom observation during the study period, supplemented observations with court records and files, and analysed Probation Officers' presentence reports. In Deane's 1995 publication, she states that after conducting a step-wise multiple regression of shoplifting and simple cannabis possession, sex did not significantly contribute to predicting sentences (Deane, 1995: 113). Given the minor nature of these offences, such a result may not be surprising. 7 Details about her regression analysis are not provided in this pUblication but the data set used was taken from Deane's (1997) doctoral thesis, which I subsequently examined to gain a better understanding of her analytical approach. In Deane's (1997) doctoral thesis, sex differences in sentence severity for the ftrst sample (n=955) of offenders were assessed using a ranked sentencing scale. 8 Recall that this sample As previously noted, international research shows that sex differences are greatest when more serious offences are analysed but when minor offences are considered sex differences are minimal (see Chesney-Lind, 1987: 129). 8 I am unsure if this ranked sentencing system was used throughout the analyses. For example, at some points a logistic regression analyses is conducted which is a technique most often used when the dependent variable is dichotomous. 7 18 included property and drug offenders. Analysis was conducted for all sentencing outcomes (imprisonment and all community-based sentences) and community based sentences only. Deane (1997: 125 & 127) used what appeared to be an Ordinary Least Squares (OLS) stepwise regression on all sentencing outcomes and argued that sex "did not playa significant role in predicting the sentencing of drug offenders [or] the sentencing of the property offenders when other factors were held constant." A further step-wise logistic regression was run on property offenders to see if there were sex differences in community based sentences. Deane (1997: 128-129) found that sex was influential in "sentencing offenders to a community based sentence ... women were more likely to be sentenced to a CSO 9 and men were more likely to be sentenced to Periodic Detention PD." 10 Deane (1997: 130) then argues that sex differences in Periodic Detention reflect sex differences in criminality and that "males were more likely to be sentenced to PD reflecting the fact they were more likely to be convicted of serious offences than women." It is later concluded, "when men and women ... appeared ... under similar circumstances for similar offences, they were treated in a similar way ... However, offenders rarely appeared under similar circumstances" (Deane, 1997: 316). The purpose of a multiple regression analysis is to hold constant differences in circumstance. Deane apparently conducted such an analysis and found evidence of sex differences in the likelihood of receiving periodic detention and community service orders, so I am unsure how she reaches this conclusion (see Deane 1997: 128-130). The use of step-wise regression is also problematic because, as Deane (1997: iii) herself points out, "the order of entry of variables is based solely on statistical criteria [which is more useful] for determining which varibl~s predict the outcome rather than the relative strengths of each variable in the subset." The sentencing process is a social process and there is a possible causal order to it, so allowing purely statistical criteria to determine the order in which variables are entered into a model is of concern. Furthermore, Deane's (1997: iii) research question asks, "is gender a factor in determining sentencing severity?" I am unsure how this question can be adequately addressed when the step-wise method used does not tell us the 'relative strengths' of the relationship between sex and sentencing outcomes. It is highly probable that offence characteristics and criminal history are, relative to sex, strongly Community Service Order - A sentence requiring between 20 and 200 hours unpaid work for a community group. This sentence has a strong rehabilitative component to it (Hesketh & Young, 1994: 12). 10 Periodic Detention - considered the next most serious sentencing outcome after imprisonment. Offenders are required to partake in supervised unpaid work for a set period of time each week. 9 19 correlated with sentencing outcome, which would explain how these types of factors emerged as influential in Deane's step-wise regression. However, not knowing what the relationship between sex and sentencing was after controlling for these factors, makes claims of equal treatment premature. After acknowledging that the use of step-wise multiple regression is potentially problematic, Deane (1997: Appendix IV) reanalysed her property offending data for community based sentences. Although Deane (1997: 125 -130) had previously appeared unsure of her original findings, (that men were more likely to receive periodic detention and women a community services order), after the reanalysis she concludes that sex differences remain. Obtaining similar results in this reanalysis is seen by Deane (1997) as evidence that outcomes would be similar in every case, so no further reanalyses of the data took place. However, her decision only to reanalyse sentencing outcomes where sex differences were originally found and ignore incidences where men and women appeared to be treated the same, is difficult to justify. Given that step-wise regressions cannot tell us the 'relative strength of the relationship between sex and sentencing it would have been better to check cases where no sex differences were previously evident in the step-wise regressions in order to determine what the strength of the relationship actually was. The fact that she failed to do this weakens the validity of her results. In section two of Deane's (1997) doctoral thesis a further analysis, including more step-wise regressions, was undertaken on a second sample. Recall that this sample was taken from the Wellington District Court and included all female offenders (n=52) and a sample of male offenders (n=165) who, irrespective of offence, were sentenced over a three month period during 1993. "Two men for every women sentenced by the same judge," during this period were included in the case-study sample (Deane, 1997: 161-162). This resulted in two quite different groups of men and women being selected. It is well known that sex differences exist in the severity and type of crimes committed by men and women. Subsequently, it was not surprising to fmd that Deane's (1997: 174 & 183) case-study sample consisted mainly of violent men and dishonest women, with the men's overall offending being ranked more serious. Comparing offenders in this setting will obviously result in different, yet easily explainable sentencing differences, because men and women are appearing before the court under completely different conditions. Deane (1997: 183) herself recognises this and notes that "under these circumstances, it is anticipated that women would be sentenced to more lenient penalties than men." After conducting another step-wise multiple regression, which, as noted, IS 20 problematic in itself, Deane (1997: 192) concludes that sex does not influence sentencing and that the only variables of any importance "in accounting for variation in sentencing patterns" were offence seriousness, number of previous convictions and current convictions. However, comparing sentences under these circumstances is clearly a redundant exercise. What is needed is a sample of closely matched male and female offenders so that sentencing outcomes for a comparative group can be analysed (see for example, Daly, 1994). Deane (1997; 2000) continued her case-study analysis by describing court processes for the same group of offenders. Once again, this is problematic, given the vast differences between the criminality of men and women in her sample. With regard to gender and its impact, Deane finds (1997: 249) that in contrast to Pakeha men, women in the case-study sample were more likely to experience delays in court processing ... This may have been because Pakeha men were better able to recognise what was required and as a result their lawyers requested they be dealt with quickly." Aside from this sex difference, Deane's fmdings are often contradictory. For example, in her discussion section summarising the findings of the case-study analysis, she (Deane, 1997: 252253) states there is no evidence of "differential treatment of women and men through informal social control via their role in the nuclear family." This comment is made without further explanation, despite earlier and later comments suggesting familial circumstances were more important for women e.g. "family circumstances ... were more often mentioned in relation to women" (Deane, 1997: 237); "the family situation, although identified as a mitigating factor for women, did not apply to the same extent for men. Of the 53 men sentenced without a PSR it was only in three cases that the family was mentioned at all" (Deane, 1997: 243); "Women more than men were presented as having family ... related problems" (Deane, 1997: 256). Analyses of pre-sentence reports found that compared to men, women received less severe sentencing recommendations (Deane 2000: 99) but given the vast differences in the criminality of men and women in this sample Deane's fmding is hardly surprising. Once again, Deane's (1997) discussion of the gendered use of familial circumstances in courtroom discussion is contradictory. On the one hand, Deane (1997: 306) notes that "women's reports were more likely to stress family ... related problems", but on the other hand she argues that women are rarely judged by their familial positions (Deane, 1997: 306-307). Deane (2000: 103) later 21 changes her mind once again, and states that "gender roles for women were reinforced in some cases in the reports." Finally, Deane (1995; 1997; 2000) concludes that no gender bias exists in New Zealand's criminal justice system. However, "gender bias" it not defined and it appears that what she actually means is that New Zealand's criminal justice system is virtually unaffected by gendered considerations. In my opinion, the contradictions, inaccuracies, sample and statistical problems evident in Deane's (1995; 1997; 2000) work make her research and these conclusions highly problematic. New Zealand society is a gendered one; a culture in which structures of masculinity and femininity are central to the formation of society as a whole (James & Saville-Smith: 1994). New Zealand's criminal justice system operates in and thus, one would expect, is logically affected by this wider gendered context. That being the case, I am surprised that in Deane's (1995; 1997; 2000) findings, dominant ideas about masculinity and femininity are held not to feature in criminal justice discourse. Deane's (1995; 1997; 2000) results also conflict with international research which shows consistent statistical sex differences in sentencing and more recently, have identified gendered discourses operating in the criminal justice system as a reason. In explanation for why her research seemingly contradicts international [mdings, Deane (1997: 330; 2000: 105) argues that New Zealand's Department of Justice has addressed gender issues through the implementation of a Gender Equity Committee and gender sensitive policy such as the EEO.11 Deane's (1995; 1997; 2000) conclusions that gender is essentially unimportant in New Zealand's criminal justice system is further at odds with other New Zealand sentencing research. For example, Triggs (1999) from the Ministry of Justice has undertaken a statistical, and compared with Deane's analysis, far more rigorous, study looking at sentencing. It found substantial differences between men and women. Triggs (1999) had access to a large Ministry of Justice database and used simultaneous logistic regression to assess which factors were influential at sentencing. All proved cases nation-wide involving an imprisonable offence for the years 1983, 1987, 1991 and 1995 were included in the sample (n=300,000) and every sentencing outcome available in New Zealand was assessed. Triggs (1999: 31-35) included 11 Equal Employment Opportunities programmes are designed to "eliminate barriers to the advancement in employment of women, minority ethnic groups and people with disabilities" (Sayers, 1992: 143). 22 controls for the following independent variables: current offence seriousness, current offence type, number of current charges, plea, accumulated seriousness of previous cases, time since the most recent case, total number of charges ever proved against an offender, previous offence types, previous breaches of sentence, most recent sentence, previous sentence, age, ethnicity. With all independent variables controlled, results showed that men were more likely than women to be imprisoned, to receive periodic detention or a monetary penalty. In contrast, women were more likely to receive community service, community programme or no sentence (Triggs, 1999: 123). This essentially meant that men were sentenced more harshly than women, even when they appeared for sentencing under seemingly similar circumstances. Although Triggs (1999) has established statistical evidence of sex differences in sentencing, making defInitive conclusions at this point may still be premature because, as Triggs (1999: 123) notes: ... it is possible that at least some of this [sex] difference may be due to factors that could not be measured in this study, such as ... gender differences in the relative gravity of offending even within specific offence types or gender differences in the actions or circumstances of the average offender. Triggs (1999) recognises the limitations of her statistical analysis and understands that an alternative research design would ideally include a case-study component. This would allow a more in-depth analysis of offenders' crime stories and ensure that apparently 'like' criminality was in fact alike. Triggs (1999) statistical analysis did not control for gender related statuses in offenders' biographies (e.g. child care, marital status, health) so it is also possible that structural variations in men's and women's lives could explain the sentencing differences noted. For example, another New Zealand study conducted by Barwick, Burns and Gray (1996:44) found that 51% of the Judges they surveyed took different factors (to those for men) into account for women defendants, and parenting responsibilities was one of the most common factors cited. However like crime stories, biographical constructions of men's and women's lives are perhaps best examined using a case-study approach in addition to statistical analyses. Subsequently, the methods used for the present research combine both a statistical and casestudy analysis. 7 The Current Research The current research addresses the complex question of gender difference in New Zealand's criminal justice system It is recognised that the treatment of men and women could 23 vary at different decision-making points and that prior judicial decisions may impact on later ones so both remand and sentencing are investigated. This study hopes to fill an obvious gap in New Zealand research while also gaining a better understanding of how gender operates in this country's criminal justice system. As noted, little substantive research has been conducted on this issue in New Zealand. What has been done is either fundamentally flawed (Deane, 1995; 1997; 2000) or limited to statistical analysis only (Triggs, 1999). A few international researchers have combined both statistical and case-study analyses (see for example, Frazier, Bock and Henretta, 1983; Crew, 1991; Steffensmeier, Kramer & Streifel, 1993), but Daly's (1994) study is one of the most comprehensive, and is instrumental in the current research design. In particular, I draw on Daly's (1994) case-study investigation of lawbreaking and 'gendered pathways' into the criminal justice system, along with Worrall's (1990) and Allen's (1987a, 1987b, 1987c) qualitative analyses of judicial discourse. The current research combines statistical analysis, including multiple regression (see Chapters 2-5), with case-study investigations of individual crime stories (see Chapter 6), Judges' sentencing remarks and Probation Officers' pre-sentencing reports (see Chapter 7). By combining these methods it is hoped that the following three questions can be answered: e First, are sentencing and remand outcomes different for adult 12 men and women? • Second, are the criteria used for deciding these outcomes gendered (i.e. are different factors considered when determining men's and women's judicial outcomes, are certain 'types' of men and women more likely to be extended judicial leniency)? • Third, if sex differences in sentencing and remand do exist, how can they be accounted for? 7.1 Research Setting The fieldwork for this study .was conducted in Christchurch, New Zealand. New Zealand has only one criminal court system, and the courts of Christchurch are part of this national network. Adult criminal trials are conducted either in the District Court, which has Although it is important to acknowledge that gender differences may exist for both adult and juvenile offenders in New Zealand, this research focuses solely on the former. Recent juvenile justice system reform, by way of the Children, Young Persons and Their Families Act (1989), has seen a radical change in how New Zealand deals with it juvenile offenders. Juvenile justice has shifted away from the adult criminal justice system and is now dealt with by the Department of Child, Youth and Family Services, where a new emphasis has been placed on the involvement of community, whanau (family), iwi (tribe) and hapu (sub-tribe) in the processing of young offenders. These changes are exemplified through the use of family group conferences and diversion, which has resulted in the majority of offenders being filtered out of the adult criminal justice process (Maxwell & Morris, 1993; Brown, Goddard & Jefferson, 1989). Such developments make it difficult to conduct research into areas of processing and outcomes. Sentencing and punishment for example, will vary considerably, often resting on discussion outcomes between the police, families, community groups, youth justice co-ordinator, and the youth court. Access to data on youth offenders is also difficult to obtain due to polices of privacy and 12 24 jurisdiction over all summary offences and indictable offences which may be tried summarily, or in the High Court, which can only try by indictment. A District Court trial may be heard by a Judge alone or by a Judge and jury, but trials in the High Court normally require jury determination. Juries determine guilt or innocence but have no say in sentencing. Once a guilty verdict has been registered, sentencing is usually carried out by the trial Judge, after hearing submissions from defence, prosecution, probation authorities, and so on. District Courts can only sentence up to a maximum of three years imprisonment. If a longer term is contemplated, a defendant must be sent to the High Court (see Hodge, 1991). The current research analysed a sample of indictable cases 13 sentenced in the Christchurch District Court and the Christchurch High Court between January 1990 and February 1997. Indictable cases usually include more serious types of offending, so it is possible that sex differences in judicial outcomes could be greater in this case (see Chesney-Lind, 1987: 129). Focusing on more serious offences further distinguishes this research from Deane's (1995; 1997; 2000), which only considered more minor offending. Whilst sentencing decisions were the major focus, remand decisions for the same sample of men and women were also analysed. 7.2 Access The research required access to documentary material in the form of trial files. Trial files are a source of personal information on offenders that are compiled by criminal justice personnel. The sensitive nature of this documentation required careful measures to ensure offenders' privacy and confidentiality. The implausibility of obtaining the written informed consent of participants, due to transitory lifestyles and the low probability of response, meant that protecting confidentiality and privacy was even more pressing. Access to court files was granted subject to conditions aimed at maintaining confidentiality.14 Research information was not used in any form that would reasonably allow individual offenders to be identified, their names did not appear on study files (printed or electronic), and the files did not leave the court premises. protection as advocated by the Department of Child, Youth and Family Services this makes research into the area of youth offending impractical. 13Purely summary offences are excluded from the sample because files kept on Summary District Court matters contain little information. The Ministry of Justice's (1996a) "List of Offences with Maximum Penalties by Act and Section," was a useful tool in identifying whether offences were summary or indictable. 14 Ethical approval was obtained from the Department For Courts, and the University of Canterbury Human Ethics Committee. 25 7.3 Sampling Procedure The sample was selected from two court registers;15 the High Court's "Return of Prisoners Tried and Sentenced," and the District Court's "Return of Persons Committed for Trial and Sentence." Because of the large numbers involved, only cases involving drug offences, violent offences, and property offences for adult offenders (aged 17 years and over) of both sexes were included. All women sentenced in the three offence categories during the seven year study period were included in the initial sample used for the statistical analysis, while the male portion of the sample was selected using a matched sampling method because of their relatively large numbers. First, men were matched with women by court of sentencing, major statutory offence, and year of disposition. For example, a woman convicted of burglary in the High Court in 1990 would be matched with all males convicted of burglary in the High Court in the same year. Next, women with more than one male match on the above three variables were matched with men who were similar on plea, number of charges, ethnicity, age, and sentencing Judge. In cases where more than one match remained, previous convictions were examined and the male who had the most similar number of previous convictions to his female match, was retained. Thus a 1: 1 male to female ratio was obtained. If a male match by statutory offence could not be found for a female within the same year, then a match was sought over all years. In eight cases this search was successful but in ten it was not. Four women from the Christchurch District Court sample were excluded because a male match could not be found across all years, and in the High Court six women were excluded. In the end, 388 offenders, or 194 matched pairs of men and women, were assessed. 16 This matched sampling method was chosen to ensure that male and female offenders were as 'similar' as possible from the outset. As discussed earlier, sex differences in the severity and type of crimes committed by men and women are well documented. If a random sampling, method had been chosen, as Deane (1997) did, the male and female samples would not have 15 The registers are paper files, which have been collected and bound by the final hearing or sentence date, The following information is contained in the registers: Name of offender, Sex of offender, offender's country of birth, ethnic origin of offender, offences committed (including Act and Section), committal date, court of committal, plea, date of plea, verdict, date of verdict, result of hearing/sentence date, trial/sentence number. 16 Incomplete information in both the High Court and the District Court registers often made matched sampling difficult. At times, major offence would be recorded with incomplete reference to exact sections of an Act - for example, receiving stolen property, which is broadly defined in section 258 of the New Zealand Crimes Act 1961, also has a number of SUb-sections attached to it with varying penalties, If the offences were recorded by Act and section only, defining the seriousness of the offence was difficult until access to the more detailed files began, 26 been comparable because men's criminality would likely to have been more serious than the women's. Approximately one quarter (50 matched pairs; 100 offenders) of the overall sample used for the statistical analysis was then selected for in-depth case studies. To allow generalisations between the case-study sample and the original statistical sample, cases selected were representative of the statistical sample in terms of the court of sentencing and offence categories. The offence characteristics (role, number of co-offenders, number of counts convicted, use of weapons, victim offender relationship, sex of victim, injury to victim and value of property involved), conviction histories, and plea for each pair, were examined and the closest matched pairs selected. Ethnicity and age were also considered in the matching process but legal variables took precedence. 7.4 Data Gathering Data for the statistical sample was coded from court trial files, and a number of social and legal factors thought to affect sentencing and remand decisions were recorded. Offenders whose files were missing were removed from the sample, along with the matched male or female. Three pairs of participants were removed for this reason. Complete files generally contained the following information, depending on the type of offence committed: police charge sheets, indictment sheets, summary of facts (offence characteristics), Probation Officers' reports, psychiatric reports (where requested), copy of the offenders' criminal/traffic offence history, Judges' comments at sentencing, computer data outlining the history of the trial (including remand status and sentencing outcome), letters from the offender, letters from the offenders' friends and/or family, copies of the bail notice, reparation statement, victim impact report, legal aid applications, and trial transcripts. All these documents were used to collect data for the statistical analysis, but criminal history sheets, summaries of facts (offence characteristics), victim impact reports, Judges' sentencing remarks, and Probation Officers' reports, were utilised in detail for the case studies and subsequently entered verbatim into a lap top computer on Court premises (because the photocopying of documents was not permitted). The later documents provided detailed accounts of offenders' criminality, characteristics, and circumstances. A particularly comprehensive source of information was the Probation Officers' reports, which contained 27 details of family circumstance, employment history, physical health, mental health, offending history, response to past sentences, and personal circumstances sUlTounding the offending. In most cases, a sentencing recommendation was also made. Judges' sentencing remarks (made in open court) were also important because they outlined the rationale behind sentencing decisions. 7.5 Independent (predictor) variables 17 For the statistical analysis, legal and social variables believed to impact on sentencing and remand decisions were selected after an extensive review had been made of past research and literature. The independent variables were grouped into seven categories: sociodemographic variables (sex, ethnicity, age, employment status and familial situation), criminal history variables (number of prior criminal convictions, number of prior convictions in the same offence category, number of prior imprisonment terms, period since last criminal conviction and prior convictions for breaches of bail), offence variables (offender's role, number of co-offenders, offence location, weapon use, value of property affected, victimoffender relationship, sex of victim, victim injury), court process variables (plea, number of conviction counts) remand status (length of time remanded in custody), pre-sentence report recommendations and the culpability/blameworthiness variables identified in these reports (health, substance abuse and negative life experiences). 18 7.6 Dependent (criterion) Variables There are a number of different remand and sentencing options available to the court and the dependent variables in the statistical analyses were selected to reflect this. To maintain workable numbers, these variables were grouped into sentencing outcomes (the imprisonment decision, imprisonment length, suspended sentences of imprisonment, work-based sentences, care-based sentences, and monetary penalties) and remand outcomes (remand status, length of custodial remand, and bail conditions). In all cases, the sample was also broken down into specific offence categories to assess interplays between sex and offence type for each of the judicial outcomes. 17 'Traditionally', the use of the terms 'dependent' and 'independent' variable have typically been confined to experimental research designs. However, this usage has become more liberal recently and is often used in correlational designs and I will use the term 'dependent' and 'independent' in this thesis. Independent variables are the presumed causes of the effects or the variables which are thought to predict a certain outcome (Reaves, 1992: 353). A dependent variable is the factor, which is explained or affected by the independent variable (Argyrous, 1996: 396). In the present research sentencing and remand outcomes are dependent variables and factors such as sex and seriousness of criminal history are the independent variables. 18 1t is important to note that the sex of Probation Officers and Judges may also impact sentencing. However, male judges sentenced all offenders in the current study and the sex of Probation Officers could not be recorded because it was difficult to establish this from the court's trial files. 28 7.6.1 Sentencing Outcomes 7.6.1.1 IMPRISONMENT Not surprisingly, imprisonment is considered the most serious sentencing outcome in New Zealand (Triggs, 1999: 47). The maximum imprisonment term for each type of offence is laid down in New Zealand by statute, but actual sentence is often dependent on the court. Offenders sentenced in the High Court and those found guilty at trial in the District Court, can be sentenced to imprisonment terms up to the maximum statutory amount. However, offenders who plead guilty in the District Court, may only be sentenced to a maximum of three years imprisonment (see Hodge, 1991: 7-11). 7.6.1.2 SUSPENDED SENTENCES OF IMPRISONMENT Suspended sentences of imprisonment were introduced by the Criminal Justice Amendment Act 1993. After a Judge decides that an offence requires imprisonment the term may be suspended if circumstances are seen to warrant this decision. If and when suspensions are appropriate is not outlined in legislation but sentencing precedence suggests youthfulness, minimal criminal histories, rehabilitation, diminished culpability, and co-operation with authorities may be circumstances under which suspending an imprisonment term might occur (Hesketh & Young, 1994: 29). Only imprisonment terms of six months to two years may be suspended, and then for no more than a period of two years. If another offence is committed during the suspension period, then the imprisonment sentence may be activated (Triggs, 1998: 84-85). 7.6.1.3 WORK-BASED SENTENCES - PERIODIC DETENTION AND COMMUNITY SERVICE After imprisonment, periodic detention is considered the next most serious sentencing outcome and thus has a strong punitive component: Periodic detention requires offenders to partake in supervised, unpaid, work for a set period of time. Usually offenders sentenced to periodic detention must report on a Saturday to undertake work between 8am and 5pm (Hesketh & Young, 1994: 1~ Triggs, 1998: 86). Similar to periodic detention, community service also requires offenders to work but in this case, 20 to 200 hOurs of unpaid work for a community group is required. This sentence also has a punitive component but is considered to have "a stronger rehabilitative element than periodic detention" (Hesketh & Young, 1994: 12). 7.6.1.4 CARE-BASED SENTENCES - SUPERVISION AND COMMUNITY PROGRAMMES Sentences of supervision all come with standard conditions (e.g. offenders are required to inform their Probation Officers when they move house) and offenders must report to a 29 Probation Officer on a regular basis, for a set period of between six months and two years. Special conditions may also be attached to supervision such as attending educational training, treatment for drug addiction, and receiving counselling or psychological/psychiatric treatment as directed by the Probation Officer. Sentences of supervision have a punitive component because they restrict offenders' lives but are generally considered more rehabilitative because the punishment component is at the lower end of the scale (Hesketh & Young, 1994: 12-13; Triggs, 1999: 97). Offenders sentenced to community programmes are placed in the care of a community group and participate in rehabilitative type programmes for a period not exceeding six months if residential, or twelve months if non-residential. Although the constraints of a community programme are quite substantial, this sentence is considered primarily rehabilitative with the programmes designed to address offenders' problems. Very few offenders are sentenced to community programmes, and their use as a viable sentencing alternative to imprisonment has never exceeded 1.3% of the total cases prosecuted in New Zealand (Triggs, 1998: 89). 7.6.1.5 MONETARY PENALTIES - FINES, REPARATION, COURT COSTS, CONFISCATION/FORFEITURE Less serious offences are likely to result in the imposition of a monetary penalty. Monetary penalties include fines, reparation, court costs, and confiscation. Reparation can be imposed when an offender has caused any loss or damage to any property in the commission of an offence and also in the cases where victims have suffered harm The court can also order offenders to pay court costs or confiscate property derived from criminal activity. In the District Court, filles must not exceed $4000. 7.6.2 Remand Outcomes Remand is important not only as an outcome but also as a factor in sentencing because it may affect sentencing severity (Nagel, 1980: 107-108; Frazier, Bock and Henretta, 1983: 308; Kruttschnitt & Green, 1984). A long custodial remand may be viewed as a sanction in itself and Judges may reduce sentences accordingly (Nagel, 1980: 107), or it may indicate how dangerous an offender might be (Frazier, Bock & Henretta, 1983: 308). Remand is subsequently considered both a dependent and independent variable in this study, particularly custodial remand length. 30 In New Zealand, remand outcomes may vary, and there are three remand options open to the court: a remand at large, on bail, or in custody.19 If an offender is remanded at large, a time and place for the next court hearing is arranged, but no other restrictions are imposed. If an offender is remanded on bail, incarceration is avoided but a number of other conditions are imposed. Remand status can change throughout the judicial process. For example, bail may initially be declined but then awarded upon appeal, or it may be awarded and then revoked if, for example, bail conditions are ignored or further charges are laid. Additionally, especially if a prison sentence is contemplated, a person granted bail pending trial who is then convicted, may be held in custody pending sentence. This being so, the remand outcome assessed in this study was the most serious outcome received (at large, on bail or in custody) throughout the court process. 20 The period spent in custodial remand (in days) and bail conditions (standard or special) were two other remand outcomes analysed. 8 Closing Comments From the research reviewed in this chapter, it becomes clear that the importance of being male or female in the criminal justice system is the subject of some debate both theoretically and methodologically. Theoretically, claims of preferential treatment benefiting women, as suggested by the 'chivalry thesis', have been discredited after further research findings showed preferential treatment was not extended to all women, over all men, all the time. 'Chivalry' has been criticised for being simplistic and reducing understandings of gender in judicial processing to "the superficial elements in male-female relationships, namely, the social amenities" (Moulds, 1978: 418). The 'paternalism' argument sought to explain these varied findings; it moved beyond the social amenities, acknowledged gender-based power relationships and argued that judicial treatment may depend on conforming to dominant gender role expectations. In this regard, 'preferential' treatment may not be so preferential after all because gender roles, which, in the wider scheme of things, disadvantage women, are being actively reinforced. However, 'paternalism', like 'chivalry', is now considered a theoretically simplistic concept. Researchers have begun to reflect more on differences in men's and women's place in the social structure (e.g. as care givers, paid workers, victims, dependants) to ascertain if, how, and whether or not, the sexes should be treated differently. ,Linkages between gender ideology, social control, social cost and 19 Offenders may also be remanded to a psychiatric institution, but only four offenders in the current study received this remand option. All of these offenders were also remanded in custody, and were subsequently coded under the custodial remand category. Remand on Home Detention is also possible but no offenders in this study received this outcome. 31 gendered differences in men's and women's criminal and non-criminal lives, have been highlighted as plausible and possibly justifiable reasons for differential treatment. Methodologically, statistical research designs have been criticised for failing to consider both relevant legal and non-legal factors but despite this, high-quality statistical studies continue to fmd sex differences in some judicial outcomes. Daly and Bordt (1995: 160) argue that "the crux of the matter is whether we can trust the statistical literature and how the findings can be interpreted". It is possible that the sex differences reported in statistical analyses are artificial in some way or reflect an inability to capture gendered differences in offenders' criminality and lives using numerical coding schedules. It may well be that after further, more in-depth investigation, gender differences in judicial processing are found to exist for other (possibly legitimate) reasons. For example, it may be that women's offending causes less harm than men's, that their actual offending roles are minor, that the judiciary does not want to harm families, that women are scared by past or current problems and are subsequently less blameworthy, or that Judges are sensitive to the degree of informal social control in women's lives. Subsequently, it is important that statistical and case-study approaches be combined to enable a better understanding of men's and women's criminal cases and lives. This approach is taken in the current thesis, which not only seeks to uncover whether or not men and women are treated differently but whether this differential treatment can be warranted as fair or just. Seriousness was ranked according to the degree of individual freedom lost under each type of remand. A remand in custody was considered the most serious remand outcome, followed by a remand on bail, and finally a remand at large. 20 32 Chapter Two Wide Sample Characteristics, Descriptive Data And Sex Differences In Independent Variables This chapter outlines the sample and variable (independent and dependent)1 characteristics for data used in the statistical study. The statistical study is an important first step toward answering the three research questions posed: whether sentencing and remand outcomes differ for men and women, whether the criteria used for deciding these outcomes are gendered and if sex differences in sentencing and remand do exist, can they be accounted for? To answer these questions, multiple regression analysis,2 including path analysis,3 will be conducted in later chapters. However, before undertaking this, it was necessary to complete two tasks. First, the data set was assessed for problems of skewed distributions,4 multicollinearity 5 and range restriction problems. 6 All of these problems can adversely affect later statistical analysis but standard solutions do exist for dealing with these issues and these are discussed where appropriate. Second, it was important to establish if sex differences existed in the independent variables. For example, it is quite likely that men's criminal histories are more serious than women's. This would be an important factor to take into account if sex differences in sentence severity were later found and seriousness of criminal history was also shown to impact on judicial outcomes. In this chapter, independent variables (such as seriousness of criminal history) were broken down by sex into cross tabulations/ for both combined offences and within each offence category (drug, violent, property) to assess how similar male and female offenders actually were. As noted, identifying significant sex differences in independent variables is an important Refer to footnote 17 in Chapter One for a definition of an independent and dependent variable. In multiple regression there is a single dependent variables (judicial outcome) and multiple independent variables (e.g. health, plea, seriousness of criminal history). The multiple regression equation contains a statistical measure of the unique relationships (regression coefficient) between independent and the dependent variables while controlling for all other independent variables (Hinkle, Wiersma & Jurs, 1994: 454). 3 Path analysis is an extension of multiple regression in that it entails the use of multiple regression in relation to explicitly formulated causal models. Path analysis cannot establish causality but it can examine the pattern of relationships between three of more variables (Bryman & Cramer, 1997: 268). 4 Variables which are not normally distributed. 5 "This problem arises when two independent variables are so highly correlated [statistically related], that there is little variance left to correlate with the dependent variable" making the regression coefficients "unreliable or useless" (Fletcher, 1999). 6 For example, if only three people out of 100 were sentenced to imprisonment this would constitute a range restriction problem. Such a pattern suppresses the subsequent correlation (the statistical relationship between variables). 7 A table displaying the joint frequency distribution of two variables - it is also known as a bivariate table or contingency table (Argyrous, 1996: 396). 1 2 33 first step toward explaining any sex differences in sentencing and remand outcomes which might later be found. Statistically significant sex differences are noted only when different at the p < .05 level. s If significance levels are not cited, it can be taken that sex differences were not statistically significant. Distributions for both independent and dependent variables were checked for normality. Skewed distributions are considered problematic because they can depress or inflate correlation coefficients. 9 Where distributions of variables were found to be badly skewed, results of distributions are presented after appropriate statistical transformations have been completed. The aim, in these cases, was to produce reasonably normal distributions for subsequent multiple regression and path analysis (which assume normality of distributions). Where dependent variables were dichotomous,10 the distributions were also examined to make sure they were reasonably even. Variable distributions more extreme than a 70%/30% split are problematic because of restriction of range problems. The strength of the relationships among independent variables were also examined to detect multicollinearity. Multicollinearity is considered a problem because it can produce unreliable regression coefficients (Bryman & Cramer, 1997: 257). Although correlations of .70 are considered the benchmark for identifying multicollinearity problems, moderate-to-high statistical relationships between independent variables may also indicate that similar constructs are being measured. Interpreting high correlations (in part) as evidence of conceptual overlap, the independent variables were constructed to avoid multicollinearity problems. The rule followed was one variable for each theoretical construct, and the aim was to limit the number of independent variables as far as possible, consistent with sound theoretical and methodological reasoning. Sample distributions for the independent variables used in later statistical analysis are outlined first and include the following variables: socio-demographic (sex, ethnicity, age, employment status, overall familial situation), criminal history (seriousness of criminal history, bail breaches, period since last criminal conviction), offence seriousness (major offence categories, offender's role, co-offenders, offence location, use of weapons, value of property affected, The criterion or cut off point beyond which we assume that a result is due to simple chance is referred to as the significance level and is usually set at probability or p level of 0.05. This means that the probability of a result occurring purely by chance is only five out of 100 (Reaves, 1992: 359; Bryman & Cramer, 1997: 107). 9 "A number between -1.0 and +1.0 that describes the relation between two variables. A coefficient of 0 indicates an absence of relationship. A correlation of -1.0 means there is a perfect negative relationship; a correlation of +1.0 means there is a perfect positive relationship. Other coefficients imply relationships of differing strengths" (Reaves, 1992: 349). S 34 offender/victim relationship, sex of victim, victim injury), court process (court of sentencing, plea, number of counts convicted), pre-sentence reports (pre-sentence report recommendations, overall health, substance abuse, negative life experiences). Secondly, distributions for the dependent variables (sentencing and remand outcomes) will be outlined including: the remand status decision, length of custodial remand, bail conditions, imprisonment sentencing decision, length of imprisonment term, suspended sentences of imprisonment, work-based sentences, care-based sentences and monetary penalties. 1 Independent Variables 1.1 Sodo-Demographic Factors 1.1.1 Sex and Ethnicity As previously noted a 1:1 male-to-female ratio was obtained via the matched sampling method, so 50% (n=194) of the sample was male and 50% (n=194) female. The majority of the overall sample was identified as European and 20% of the sample was identified as non-European (see Table 1). The category non-European included Pacific Island, Asian, and Maori people. Only three offenders were identified as Pacific Islanders, and a further three as Asian, so most of the 20% were Maori. Ethnicity was a possible match variable, so the number of offenders in each ethnic group was heavily influenced by the ethnicity of female offenders and thus was different from national crime statistics. 11 However, the ethnic distribution of female offenders in the current sample also differed from national crime statistics. For example, in 1996, 59% of female offenders sentenced for property, drug or violent crime, were non-European (Spier, 1997: 45) as opposed to 23% in the current case. The most likely reason for this disparity is the ethnic demography of the Canterbury region, which has a small female Maori population (6%) compared to Auckland (12%) and Wellington (12%) (Statistics New Zealand, 1996a: 78 - 94). Hence, the proportion of Maori offenders in this sample is roughly what one might expect, given Canterbury's regional demographic characteristics. 10 11 Had only two possible values or categories (also known as Binomial) such as imprisoned/not imprisoned. Refer to Spier, 1997: 45 35 1- TABLE ETHNICITY Female n Male n % Total % n % Combined Offences 150 44 194 77 23 100 161 33 194 83 17 100 388 80 20 100 78 13 91 88 14 100 85 6 91 93 7 100 163 19 182 90 10 100 27 22 49 55 45 100 30 19 49 61 39 100 57 41 98 58 42 100 46 8 54 85 15 100 45 9 54 83 17 100 91 17 108 84 16 100 European Non-European Total 311 77 Drug Offences European Non-European Total Violent Offences European Non-European Total Drug Offences European Non-European Total 1.1.2 Age The overall age distribution (combined offence categories) can be seen in Figure 1 and is reasonably normally distributed. Age was another possible match variable so the mean (m)12 ages were similar for men and women: combined offence categories women m = 29 years (§.,Q. = 8.00)13 and men m = 30 years (s.d. = 9.00); for drug offending women m = 30 years (s.d. = 8.49), men m = 31 years (s.d. = 8.41); for violent offending women m = 25 years (s.d. = 6.73), men m = 26 years (s.d. = 8.31); for property offending women m = 30 years (s.d. = 7.53), men m = 31 years (s.d. = 9.90). This age distribution is similar to that found nation wide in New Zealand (see Spier, 1997: 44). FIGURE 1 - AGE, COMBINED OFFENCES 100 80 60 40 = 20 Frequency • o 20.0 12 ill 40.0 Std. Dev 8.58 Mean = 29.6 _ _ _gN = 388.00 50.0 60.0 is the symbol used to represent the mean which is defined as the sum of all the scores in a distribution divided by the total number of cases - otherwise known as the average (Argyrous, 1996: 397). 13 The §.Jl or standard deviation calculates the average distance of scores in a distribution from the mean (Reaves, 1992: 360; Argyrous, 1996: 398). It is a measure of how widely or narrowly the scores are distributed. 36 1.1.3 Employment Status Table 2 shows the employment status of offenders. Few offenders were involved in paid work and violent offenders were less likely than other offenders to be in paid work. Women were less likely than men to be in paid work over all. These differences were only statistically significant for combined offence categories (x 2 (1) offending (x 2 (1) = 4.61, p = 7.14, p < .05Y4 and for drug < .05). This sex difference is not surprising because New Zealand women are generally less likely to be involved in paid employment (Statistics New Zealand, 1993: 82) and more likely to work in the home for no financial gain (McKinlay, 1992). Similar patterns have been noted amongst New Zealand's sentenced inmates. For example, during 1995, approximately 15% of women were in paid employment prior to entering prison, compared to 31 % of men (Lash, 1996: 55). TABLE 2- EMPLOYMENT STATUS Combined Offences In paid work Not in paid work Total Drug Offences In paid work Not in paid work Total Violent Offences In paid work Not in paid work Total Property Offences In paid work Not in paid work Total 1.2 n Female % n Male % n Total % 28 166 194 14 86 100 49 145 194 25 75 100 311 388 20 80 100 14 91 15 65 100 26 65 91 29 71 100 40 142 182 22 78 100 5 44 49 10 90 100 9 40 49 18 82 100 14 84 98 14 86 100 9 45 54 17 83 100 14 40 54 26 74 100 23 85 108 21 79 100 77 77 Overall Familial Situation Previous researchers have noted that an offender's familial situation may influence judicial decision-making and explain differences in the outcomes men and women receive (see Chapter One). To assess this possibility, the current research collected a number of different measures of familial situation (childcare responsibilities, marital status, living arrangements) at the time of sentencing, and then combined them to produce an overall measure. This combined family variable, henceforth referred to as 'overall familial situation', was subsequently used in later statistical analysis. Of those offenders who had mothered/fathered children, men were significantly more likely than women to have little or no responsibility for children sUillillt:fd across offence categories (x 2 (1) 14These figure represent the results of a chi-square test (x 2) which was used to compare frequency distributions for male and female offenders to determine whether they were significantly different (Reaves, 1992: 361). The significance level was set at or p < .05. 37 = 36.95, p <.001) and also within each specific offence category: drug offences (x 2 (1) = 21.13, p <.001); violent offences (x 2 (1) = 13.63, p <.001); property offences (x 2 (1) = 4.75, p <.05) (see Table 3). TABLE 3- CHILDCARE RESPONSIBILITIES Female Male Total n % n % n % 84 110 194 43 57 100 143 51 194 74 26 100 227 161 388 59 41 100 31 60 91 34 66 100 62 29 91 68 32 100 93 89 182 51 49 100 25 24 49 51 49 100 42 7 49 86 14 100 67 31 98 68 32 100 28 26 54 51 48 100 39 15 54 72 28 100 67 41 108 62 38 100 Combined Offences No Yes Total DruR Offences No Yes Total Violent Offences No Yes Total Offences Pro~ety No Yes Total The majority of offenders were single 15 and the sexes were equally likely to be single (see Table 4). TABLE 4- MARITAL STATUS Female Male Total n % n % n % 116 78 194 55 45 100 120 74 194 62 34 100 236 152 388 61 39 100 55 36 91 60 40 100 54 37 91 59 41 100 109 73 182 60 40 100 31 18 49 63 37 100 28 21 49 57 43 100 59 39 98 60 40 100 34 20 54 63 37 100 34 20 54 63 37 100 68 40 108 63 37 100 Combined Offences Sin!:Jie MarriediOefacto Relationship Total Drug Offences Single Married/Oefacto Relationship Total Violent Offences Single MarriediOefacto Relationship Total Property Offences Single MarriediOefacto Relationship Total Having ties to, and responsibilities towards, a spouse and/or children may be an important factor for court decisions. In addition, familial ties may also be relevant for offenders who live 15 Identifying the marital status of offenders may have been problematic because of offenders' financial dependence on income support from the government. In New Zealand, income support payments are grounded in the concept of the family wage, and subsequently assessed depending on marital status; if a person has a partner payment rates will be assessed on both the individuals economic position and their partner's. This can be especially troublesome for single parent families reliant on the Domestic Purposes Benefit and in the present research more women than men were identified as having sole responsibility for children. Financial necessity, and fear of being caught 'ripping off' the system may have resulted in women over reporting their marital status as single. Men may also have over reported their status as single to protect partner's who receive the Domestic Purposed Benefit or to protect their Unemployment Benefits which they would not be entitled to if their partner was working. 38 with parents, siblings, or other family members. The majority of offenders, both male and female, were found to be living with family members (see Table 5). TABLE 5- LIVING ARRANGEMENTS Combined Offences Lives alone or with people other than family Lives with family Total Drug Offences Lives alone or with people other than family Lives with family Total Violent Offences Lives alone or with people other than family Lives with family Total Property Offences Lives alone or with people other than family Lives with family Total n Female % n Male % n Total % 26 168 194 15 86 100 39 155 194 20 80 100 65 323 388 17 83 100 7 84 91 8 92 100 13 78 91 14 86 100 20 162 182 11 89 100 9 40 49 18 81 100 12 37 49 25 75 100 21 98 21 79 100 10 44 54 18 82 100 14 40 54 26 74 100 24 84 108 22 78 100 77 Some offenders with childcare responsibilities may also have been married and would thus be living with family members. Therefore, it was not surprising to find moderate-to-high statistical relationships (r=. 35 to r=. 50, p <. 001) between childcare responsibilities, marital status, and living arrangements. On theoretical grounds, and to avoid collinearity, an overall measure of familial ties was constructed by summing childcare responsibilities (0 = no child care responsibilities, 1 = yes, has child care responsibilities), marital status (0 = single, 1 = marriedldefacto relationship) and living arrangements (0 = lives alone or with people other than family, 1 = lives with family). In this sample, women's (m= 1.82, s.d. = .94) overall familial ties were generally stronger than men's (m = 1.45, s.d. = .91) and women were more likely to be categorised with strong familial ties (see Table 6). This sex difference was statistically significant for combined (p <.001) 16 and drug offence categories (p <.05). 16 Because I consistently adopt a correlational regression analysis strategy throughout the thesis, I used correlations between mean scores of men and women to test for sex differences. It should be noted that such tests are mathematically equivalent to L- tests}. The Significance level was set at a probability of p.D5 and results showed that the differences between male and females were significant at this level. 39 TABLE 6 m OVERALL FAMILIAL SITUATION n Combined Offences No familial ties Minimal familial ties Moderate familial ties StronQ familial ties Total Drug Offences No familial ties Minimal familial ties Moderate familial ties StronQ familial ties Total Violent Offences No familial ties Minimal familial ties Moderate familial ties StronQ familial ties Total Property Offences No familial ties Minimal familial ties Moderate familial ties Strong familial ties Total Female % n Male % n Total % 26 29 94 45 194 13 15 49 23 100 39 44 93 18 194 20 23 48 9 100 65 73 187 63 388 17 19 48 16 100 7 11 50 23 91 8 12 55 25 100 13 20 50 8 91 14 22 55 9 100 20 31 100 31 182 11 17 55 17 100 9 7 24 9 49 18 14 49 18 100 12 13 20 4 49 24 26 41 8 100 21 20 44 13 98 21 20 45 13 100 10 11 20 13 54 22 20 37 24 100 14 11 23 6 54 26 20 43 11 100 24 22 43 19 108 22 20 40 18 100 1.3 Criminal History Variables Criminal history may be one of the most influential variables in judicial processing, so numerous measures of it were gathered including: number of prior criminal convictions, number of prior criminal convictions in a similar offence category to the current offence, number of prior imprisonment terms, prior bail breaches, and period since last criminal conviction. The former three variables were combined to generate one variable called 'seriousness of criminal history', which was used for all subsequent statistical analyses. Prior bail breaches and period since last criminal conviction will also be entered into later statistical equations. 1.3.1 Seriousness of Criminal History Seriousness of criminal history may adversely affect judicial outcomes, with serious criminal histories increasing the severity of judicial outcome. New Zealand statistics indicate that women offenders tend to have fewer previous convictions than men (Triggs, 1999: 54). Of all those prosecuted in New Zealand's criminal courts during 1996, for example, two-thirds of women, compared to half the men, had no prior convictions in the previous six years (Triggs, 1998: 43). Similar differences between men and women were evident in the current sample; 15% of men were identified as first offenders compared with 25% of women. The current research focus on more serious offences can possibly account for the higher proportion of offenders in this sample with prior criminal convictions. 40 The mean number of prior convictions for the whole sample (combined offence categories) was 18.47 (s.d. = 26.38). The number of prior convictions was significantly larger for men (m = 22.78, s.d. = 28.48) than women (m = 14.16, s.d. = 23.38) (x 2 (75) = 77.73, p = .001). Similar, but non-significant, sex differences were also noted within each offence category (drug, violent, property). While the number of prior convictions may increase the severity of a judicial outcome, so too might the nature of these past offences, especially if there is a history of similar offending. Thus, the numbers of prior criminal convictions in the same offence category as the present offence were also recorded. For all offenders in the sample with prior criminal convictions, the mean number of convictions in the same offence category as their present offence was 5.73 (s.d. = 14.22). Men (m = 7.82, s.d. = 17.32) were more likely than women (m = 3.64, s.d. = 9.81) to have convictions in the same offence category as their present conviction. The number of prior imprisonment terms served may further inform us about the depth of an offender's past criminality. Of those offenders with prior criminal convictions, the mean number of prior imprisonment terms served by offenders overall (combined offence categories) = 3.72) served a significantly larger number of prior imprisonment terms than women (m = .66, s.d. = 1.77) summed across offences (combined offence categories) (x 2 (16) = 61.69, p <.001), for drug offences (men m = 2.07, s.d. = 3.41, women m = .34, s.d. = .82, x 2 (11) = 32.68, p = .001) and violent offences (men ill = 3.24, s.d. = 3.58, women m = 1.02, s.d. = 2.04, x 2 (11) = 22.96, p <.05). Similar, but nonwas 1.64 (s.d. = 3.07). Men (m = 2.61, s.d. significant, sex differences were also noted for property offenders. Some offenders with prior convictions had convictions in an offence category similar to their present offence and may also have served prior terms of imprisonment. Therefore, it was not surprising to find moderate to high correlations (r.45 to [.65, p < .001) between number of prior criminal convictions, number of prior criminal convictions in a similar offence category, and number of prior imprisonment terms. Thus, an overall measure of criminal history was produced by summing the standardised scores ~ scores) for number of prior criminal convictions, number of prior criminal convictions in a similar offence category, and number of prior imprisonment terms. Not surprisingly, men's (m = 2.47, s.d. = .53) overall criminal histories (for combined offence categories) were significantly (p < .001) more serious than women's (m= 2.18, s.d. = .50). 41 1.3.2 Bail Breaches A history of breaching bail may reduce the likelihood of it being granted in the future. However in this sample, generally few differences were found in men's and women's propensities to breach bail (during the present trial or in the past). The exception was property offences, where men were significantly more likely than women to have breached bail (x 2 (1) = 4.47, p <.05) (see Table 7). TABLE 7- BAIL BREACHES DURING PRESENT TRIAL OR IN THE PAST Female Male Total n n n % % % Combined Offences Yes No Total 44 150 194 23 100 50 144 194 26 74 100 94 294 388 23 76 100 17 74 91 19 81 100 17 74 91 19 81 100 34 148 182 19 81 100 20 29 49 41 59 100 17 32 49 35 65 100 37 61 98 39 62 100 7 47 54 13 87 100 16 38 54 30 70 100 23 85 108 21 79 100 77 Drug Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total 1.3.3 Period since Last Criminal Conviction Short periods between past and present offending could indicate a likelihood of reoffending. For combined offence categories, women (54%) were 8% more likely than men (46%) to have offended in the previous twelve months.17 However, no significant sex differences could be found in the mean periods between current and prior conviction. As shown in Figure 2, the distribution for the period since last criminal conviction was badly skewed, so before conducting further statistical analysis, a log transformation was conducted to achieve a more normal distribution. Figure 3 illustrates that this transformation was successful and a normal distribution was achieved. 17 Slightly more men than women were actually serving sentences of imprisonment during this 12 month period (25 men and ten women), So the real rate of re-offending in the previous twelve months was about equal. 42 FIGURE 2 PERIOD (IN MONTHS) SINCE LAST CRIMINAL CONVICTION, ALL OFFENDERS, COMBINED Q OFFENCES Std. Dev = 41.41 Frequency ..1IIIIa:=-.IIIIIiIII...............__...........J Mean = 28.6 N = 309.00 Months Since Last Criminal Conviction FIGURE 3- PERIOD SINCE LAST CRIMINAL CONVICTION, ALL OFFENDERS, COMBINED OFFENCES (LOG TRANSFORMATION) 30 20 10 Std. Dev = .55 Frequency Mean = 1.12 o N = 309.00 Log of Months Since Last Criminal Conviction 1.4 Offence Seriousness The sampling method in this research enabled male offenders to be matched one-to-one with female offenders by statutory offence: thus there were no sex differences in the incidence of each crime. The numbers of offenders between and within each offence category (drug, violent, property) are presented in this section. However, by itself, statutory offence is not an adequate measure of offence seriousness, because the circumstances of particular offences may vary greatly. Other measures of offence seriousness used were: offender's role, number of co- 43 offenders, offence location, use of weapons, value of property affected, victim-offender relationship, sex of victim, and victim injury. These variables are discussed later in this section. 1.4.1 Major Offence Category The largest proportion of offenders in this study were convicted of a drug offence, followed by property and violent offences (see Table 8). Nation-wide data indicates that the majority of women offenders are convicted of petty property crimes, followed by drug and violent offences. During 1996, property offending accounted for approximately 66% of all female convictions in New Zealand (across the three offence categories), followed by drug (18%), and violent offending (17%) (Spier, 1997: 45). The difference between national statistics and this sample is probably explained by the focus on indictable offences in this research. Less serious offences tend to be summary offences which were not included the current sample (see Chapter One). TABLE 8- MAJOR OFFENCE CATEGORY Violent Offence Property Offence Drug Offence Total Female no % 25 49 28 54 91 47 100 194 Male no 49 54 91 194 Total no 98 108 182 388 % 25 28 47 100 % 25 28 47 100 1.4.2 Seriousness of Violent Offence The violent offences were categorised into seven offence classes according to maximum statutory penalty, as listed in Table 9. TABLE 9 - SERIOUSNESS OF VIOLENT OFFENCES Classification Class Class Class Class Class Class Class Total 1 2 3 4 5 6 7 violent offence violent offence violent offence violent offence violent offence violent offence violent offence Maximum imprisonment Life 14 years 10 years 7 years 5 years 2 years 1 year Female Male n % n % n % 3 33 1 5 4 1 2 49 6 68 2 10 8 2 4 100 3 33 1 5 4 1 2 49 6 68 2 10 8 2 4 100 6 66 2 10 8 2 4 98 6 68 2 10 8 2 4 100 Total Class 1 violent offence = Murder (Crimes Act, s 172) Class 2 violent offences = Wounding with intent to cause grievous bodily harm (Crimes Act, s 188 {1}}, kidnapping (Crimes Act, s 209), aggravated robbery (Crimes Act, s 235), aggravated burglary (Crimes Act, s 240A) , extortion by certain threats (Crimes Act, s 236). Class 3 violent offences = Robbery (Crimes Act, s 234) Class 4 violent offences = Accessory after the fact to murder (Crimes Act, s176), conspiracy to commit aggravated robbery (Crimes Act, s 235 & s310), wounds with intent to injure (Crimes Act, s 188 {2}}, Class 5 violent offences = Cruelty to a child (Crimes Act, s 195), Assault with weapon (Crimes Act, s202C). Class 6 violent offences Assault on a child (Crimes Act, s 194), Possession of offensive weapons (Crimes Act, s 202A). Class 7 violent offences = Common Assault (Crimes Act, s196). 44 1.4.3 Seriousness of Drug Offence Using the maximum statutory penalty attached to each offence, SIX drug offence classes were established (see Table 10). TABLE 10 m SERIOUSNESS OF DRUG OFFENCES Classification Class 1 DruQ Offence Class 2 Drug Offence Class 3 DruQ Offence Class 4 Drug Offence Class 5 DruQ Offence Class 6 Drug Offence Total Maximum imprisonment Life 14 Years 10 Years 8 Years 7 Years 3 Years Female no % 16 18 55 60 2 2 13 14 4 4 1 1 100 91 Male no 16 55 2 13 4 1 91 Total no 32 110 4 26 8 2 182 % 18 60 2 14 4 1 100 % 18 60 2 14 4 1 100 Class 1 Drug Offence = ImporVexport, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class A drug (Misuse of Drugs Act, s6 2a). Class 2 Drug Offence = Import/export, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class B drug (Misuse of Drugs Act, s6 2b). Class 3 Drug Offence = Conspire to commit an offence against s6 (1) of the Misuse of Drugs Act a Class B drug (Misuse of Drugs Act, s6 2Ab). Class 4 Drug Offence = ImporVexport, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class C drug (Misuse of Drugs Act, s6 2c). Class 5 Drug Offence = Permit Premises to be used in the commission of a Class B drug offence (Misuse of Drugs Act, s12 2b) Class 6 Drug Offence = Permit premises to be used in the commission of a Class C drug offence (Misuse of Drugs Act, s12 2c). 1.4.4 Seriousness of Properly Offences Four property offence categories were assessed according to maximum statutory penalty (see Table 11). TABLE 11 - SERIOUSNESS OF PROPERTY OFFENCES Classification Class 1 Property Offence Class 2 Property Offence Class 3 Property Offence Class 4 Property Offence Total Maximum imprisonment 14 Years 10 Years 7 Years 3 months Female Male n % n % n % 7 8 38 1 54 13 15 70 2 100 7 8 38 1 54 13 15 70 2 100 14 16 76 2 108 13 15 70 2 100 Total = = Class 1 Arson (Crimes Act, s 294) Class 2 Burglary (Crimes Act, s 241) Class 3 = Theft (Crimes Act, s 222, s 223, s 227), Fraud (crimes act, s 246 1, s 246 2(a), s 229A) and Receiving Stolen Property (Crimes Act, s 258 (a)). Class 4 = Fraud (Crimes Act, s 246 (C)). 1.4.5 Offender's Role and Co-offenders Males and females acted alone about half of the time. Of those who acted in concert with others, males and females were equally likely to be principals (see Tables 12). In 87% of the cases, offenders played an active or equal role in the crime (see Table 13). 45 TABLE 12 - CO-OFFENDERS n Combined Offences Acted alone Acted with others Total Drug Offences Acted alone Acted with others Total Violent Offences Acted alone Acted with others Total Property Offences Acted alone Acted with others Total TABLE Female % Male % n n Total % 88 106 194 45 55 100 97 97 194 50 50 100 185 203 388 48 52 100 43 48 91 47 53 100 49 42 91 52 46 100 92 90 182 51 49 100 13 36 49 27 73 100 18 31 49 37 63 100 31 67 98 32 68 100 32 22 54 59 41 100 30 24 54 56 44 100 62 46 108 57 43 100 13 - OFFENDER'S ROLE IN OFFENCE n Combined Offences Active/equal role Secondary role Total Drug Offences Active/equal role Secondary role Total Violent Offences Active/equal role Secondary role Total Property Offences Active/equal role Secondary role Total Female % n Male % n Total % 166 28 194 86 14 100 172 22 194 89 11 100 338 50 388 87 13 100 83 8 91 91 9 100 80 11 91 88 12 100 163 19 182 90 10 100 37 12 49 75 25 100 41 8 49 84 16 100 78 20 98 80 20 100 46 8 54 85 15 100 51 3 54 94 6 100 97 11 108 90 10 100 1.4.6 Offence Location Minimal differences in offence were found between the sexes. Men and women offended with similar frequency in private and public locations (see Table 14). 46 TABLE 14 - OFFENCE LOCATION Total n Female % n Male % n % 68 126 194 35 65 100 71 123 194 37 63 100 139 249 388 36 64 100 15 76 91 17 83 100 13 78 91 14 86 100 28 154 182 15 85 100 21 28 49 43 57 100 24 25 49 49 51 100 45 53 98 46 54 100 32 22 54 59 41 100 34 20 54 63 37 100 66 42 108 61 39 100 Combined Offences Public Private Total Drug Offences Public Private Total Violent Offences Public Private Total Property Offences Public Private Total 1.4.7 Use of Weapons In the current sample, 72% of violent offenders used weapons. Women were as likely as men to use weapons (see Table 15). TABLE 15 - USE OF WEAPONS Use of Weapon Female n % Yes 35 71 No 14 29 Total 49 100 1.4.8 n 36 13 49 Male % 73 27 100 n 71 27 98 Total % 72 28 100 Value of Property Affected For property offences, the mean amount of financial loss in dollars to victims was $7,945 (s.d. = $14,105). The mean loss of property to women's victims was higher than for men ($10,265 compared to $5624) but this difference was not statistically significant possibly due to a small sample size (n=108). As demonstrated in Figure 4, the distribution was strongly skewed, so before further statistical analysis, a log transformation was conducted to achieve a more normal distribution. Figure 5 shows that this transformation was successful and a more normal distribution was achieved. 47 FIGURE 4 - VALUE OF PROPERTY AFFECTED 50 ~-. 40 30 20 f>' c 10 Std. Dev = 14105.36 Q) ::J Mean 0- ~ LL =7945.2 .1IIiiIIiIL.,..-..,.-=;;:::::;=>IIIijllllllllI\IIII--.--,-JiIiI!IIiillllll'jlIIL..,,...,IIiI1jIiI N =108.00 0 6''tl ~ ">CfJ.."><tJ O''tl 0'<tJ . Vqo Vqo-'t!o Vqo Vqo Vqo Value of Property FIGURE 5 - VALUE OF PROPERTY AFFECTED - (lOG TRANSFORMATION) 40~-, 30 20 10 (;' cQ) Std. Dev = .63 Mean = 3.49 ::J 0- ~ LL N = 108.00 0 2.00 2.50 2.25 1.4.9 3.00 2.75 3.50 3.25 4.00 3.75 4.50 4.25 5.00 4.75 Victim-Offender Relationship - Violent Offenders 18 In this sample, little difference was found between the sexes in their relationships with victims men were only slightly more likely than women to have known their victims prior to their offence (see Table 16). 18 The victim-offender relationship could not be considered for property offenders because this information was rarely given. Drug offending was considered a victimless crime. 48 16 - TABLE VICTIM-OFFENDER RELATIONSHIP- VIOLENT OFFENDERS n Female % 20 29 49 Knew Victim Unknown Victim Total 41 59 100 n Male % 24 25 49 49 51 100 Sex of Victim - Violent Offenders 1.4.10 n 44 54 98 Total % 45 55 100 19 In New Zealand, men are more likely both to be the perpetrators and the victims of crime (Statistics New Zealand, 1996b: 30 & 63). The dominance of this male-male, victimoffender relationship was evident in the current research. Male offenders were significantly more likely than females to victimise men, and women were more likely to victimise women (x 2 (1) = 7.80, p < .05) (see Table 17). TABLE 17 - SEX OF VICTIM - VIOLENT OFFENDERS Sex of Victim Female n % n 22 Male Victim 45 33 Female Victim 27 55 16 Total 49 49 100 Male % 67 33 100 Victim Injury - Violent Offenders 1.4.11 n 55 43 98 Total % 56 44 100 20 The majority of violent offences resulted in emotional distress or physical injury to victims. Slightly more women than men inflicted physical injury on victims (see Table 18). TABLE 18 - VICTIM INJURY - VIOLENT OFFENDERS n No iniurv Emotional Distress Physical iniurv Total 7 12 30 49 Female % 14 25 61 100 n 15 11 23 49 Male % 31 22 47 100 n 22 23 53 98 Total % 22 24 54 100 1.5 Court Process Factors There are a number of court process factors which can potentially impact on judicial outcomes. In New Zealand, for example, District Courts can only sentence up to a maximum of three years imprisonment while the High Court can sentence up to the statutory maximum. In addition to court of sentencing, plea and number of charges may be influential in judicial decision-making. These factors are discussed in this section, but few sex differences were noted because offenders were matched on these variables. The sex of victims was only considered for violent offending because this information was rarely available for property offenders. Drug offending was considered a victimless crime. 20 Victim injury was only considered for violent offending because this information was rarely available for property offenders. Drug offending was considered a victimless crime. 19 49 1.5.1 Courl of Sentencing In addition to statutory offence, the sampling method used matched male and female offenders one-to-one by Court of Sentencing, so equal numbers were achieved within offence category by court. Offenders were distributed fairly evenly between the District Court and the High Court: 208 (104 pairs) were sentenced in the Christchurch District Court, and 180 offenders (90 pairs) were sentenced in the Christchurch High Court (see Table 19). TABLE 19 - COURT OF SENTENCING Male Female District Court (Christchurch) High Court (Christchurch) Total Total n % n % n % 104 90 194 54 46 100 104 90 194 54 46 100 208 180 388 54 46 100 1.5.2 Plea Plea was a possible match variable. The majority of offenders in this sample pleaded guilty and the sexes entered this plea with similar frequency, regardless of offence type (see Table 20). TABLE 20 - PLEA n Female % n Male % n Total % Combined Offences Guilty Not Guilty Total 130 64 194 67 33 100 132 62 194 68 32 100 126 388 68 33 100 64 27 91 70 30 100 68 23 91 75 25 100 132 50 182 72 28 100 32 17 49 65 35 100 32 17 49 65 35 100 64 34 98 65 35 100 34 20 54 63 37 100 32 22 54 59 41 100 66 42 108 61 39 100 26~ Drug Offences Guilty Not Guiltv Total Violent Offences Guilty Not Guilty Total Property Offences Guilty Not Guilty Total 1.5.3 Number of Conviction Counts Given that the number of conviction counts was a possible match variable, it is not surprising that the mean number of conviction counts was similar, when summed across offence categories, for men(m = 2.07, s.d. = 3.58) and women (m = 2.42, s.d = 3.32) as well as within specific offence categories: drug (men m = 1.76, s.d. = 1.39, women m = 1.94, s.d. = women m = 1.83, §Ji. = 1.73), property (men m = = 1.57, s.d. = .64, 1.35, s.d. = .48, women m = 1.48, s.d = .50). 1.79), violent (men m Overall the number of conviction counts, as seen in Figure 6, was skewed and considered problematic because the relationship between variables 50 in later statistical analysis could be adversely affected. To achieve a more normal distribution, this variable was converted into a dichotomous variable (one conviction count or more than one conviction count) before conducting further statistical analysis. The sexes were equally likely (approximately) to have more than one conviction count (see Table 21). FIGURE 6 _ NUMBER OF CONVICTION COUNTS, ~ COMBINED OFFENCES _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _, ~m 300 200 100 Std. Dev = 3.46 Mean = 2.2 o 5.0 TABLE N = 388.00 -~ 10.0 0.0 20.0 15.0 30.0 25.0 40.0 35.0 45.0 21 _ NUMBER OF CONVICTION COUNTS Female Combined Offences 1 count More than 1 count Total Drug Offences 1 count More than 1 count Total Violent Offences 1 count More than 1 count Total Pro~ety Offences 1 count More than 1 count Total 1.6 Male % n Total % n % n 105 89 194 54 46 100 116 78 194 60 40 100 221 167 388 57 43 100 51 40 91 56 44 100 56 35 91 61 38 100 107 75 182 59 41 100 26 23 49 53 47 100 25 24 49 51 49 100 51 47 98 52 48 100 28 26 54 52 48 100 35 19 54 65 35 100 63 45 108 58 42 100 Pre-Sentence Reports Probation officers often make sentencing recommendations to Judges in pre-sentencing reports and also discuss reasons behind these recommendations. In addition to offenders' criminal backgrounds, reports may point out a history of drug addiction, victimisation, and poor physical or mental health to a sentencing Judge. Specialist psychiatric/psychological reports may also be requested which usually focus on the latter group of factors. In this section, sex differences in the information provided by these reports are presented and 51 discussed. Sample numbers vary because of report availability and the information provided in them. 1.6.1 Pre-Sentence Recommendations Men were significantly more likely than women to receive an imprisonment recommendation for combined offence categories (x 2 (1) (x 2 (1) = 5.07, p = 9.34, p < .05) and for drug offences <.05). Male violent and property offenders also received imprisonment recommendations more often than women, but this was statistically non-significant (see Table 22). TABLE 22 m PRE-SENTENCING RECOMMENDATIONS Female Combined Offences Imprisonment Other Total Drug Offences Imprisonment Other Total Violent Offences Imprisonment Other Total Property Offences Imprisonment Other Total Male Total n % n % n % 42 145 187 22 78 100 68 116 184 37 63 100 110 261 371 30 70 100 17 73 90 19 81 100 30 59 89 34 66 100 47 132 179 26 74 100 19 29 48 40 60 100 27 20 47 57 43 100 46 49 95 48 52 100 6 43 49 12 88 100 11 37 48 23 77 17 80 97 18 82 100 100 1.6.2 Overall Health In the current sample, poor physical health was identified for men and women at similar rates (see Table 23). TABLE 23 - PHYSICAL HEALTH Female Male Total n % n % n % 62 126 188 33 67 100 54 132 186 29 71 100 116 258 374 31 69 100 36 53 89 40 60 100 31 59 90 34 66 100 67 112 179 37 63 100 6 43 49 12 87 100 9 39 48 19 81 100 15 82 97 16 85 100 20 30 50 40 60 100 14 34 48 29 71 100 34 64 98 35 65 100 Combined Offences Poor Good Total Drug Offences Poor Good Total Violent Offences Poor Good Total Property Offences Poor Good Total 52 Comments on poor mental health were made significantly more often about women, with approximately twice as many women as men identified with poor mental health for combined offence categories (x 2 (1) = 34.00, p = < .001). This pattern was also seen within specific offence categories: drug (x 2 (1) = 14.97, p <. 001), violent (x 2 (1) = 6.51, p < .05) and property (x 2 (1) = 13.67, p <.001) (see Table 24). TABLE 24 - MENTAL HEALTH n Female % n Male % n Total % Combined Offences Poor Good Total 115 75 190 61 39 100 58 131 189 31 69 100 173 206 379 46 54 100 50 41 91 55 45 100 24 66 90 27 73 100 74 107 181 41 59 100 29 20 49 59 41 100 16 32 48 33 67 100 45 52 97 46 54 100 36 14 50 72 28 100 18 33 51 35 65 100 54 47 101 53 47 100 Drug Offences Poor Good Total Violent Offences Poor Good Total Property Offences Poor Good Total A moderate correlation (r.40, p <. 001) was found between physical health and mental health, showing that offenders with poor physical health tended to have poor mental health. Accordingly, an overall measure of health was produced by adding the mental health (O=poor, l=good) and physical health (O=poor, l=good) scores together (producing scores from 0 to 2). Higher scores equated to better health. This overall measure of health was the subsequent variable used in the statistical analysis presented in coming chapters. Table 25 shows that men were more likely to be categorised with good overall health (m = 1.40, s.d. = .65) and women were more likely to be categorised with poor overall health em = 1.05, s.d. = .68). These sex differences were statistically significant for combined offence categories (p < .001), drug offences (p <. 05) and property offences (p <. 05). Internationally, the 'sexual politics of sickness' is noted to be a common feature of criminal justice processing (see Heidensohn, 1996: 95). Sickness and pathology are frequently noted to characterise judicial discourse for women more often than men (see Chapter One). With this being the case, the findings presented in Table 25 do not come as a surprise. 53 TABLE 25 m OVERALL HEALTH Female % Male n n Total % n % Combined Offences Poor Average Good Total 39 99 52 190 21 52 27 100 18 77 94 189 10 41 49 100 57 176 146 379 15 47 38 100 19 48 24 91 21 53 26 100 7 41 42 90 8 46 47 100 26 89 66 181 14 49 36 100 2 31 16 49 4 63 33 100 3 19 26 48 6 40 54 100 5 50 42 97 5 51 43 100 18 20 12 50 36 40 24 100 8 17 26 51 16 34 50 100 26 37 38 101 27 36 37 100 Drug Offences Poor Average Good Total Violent Offences Poor Average Good Total Property Offences Poor Average Good Total 1.6.3 Substance Abuse The majority of offenders were identified in the pre-sentence report as having a substance abuse problem. Minimal and statistically non-significant differences were noted between the sexes (see Table 26). The exception was male property offenders, who were significantly more likely than women to be identified with a substance abuse problem; x 2 (1) = 8.78, p <. 05). TABLE 26 m SUBSTANCE ABUSE Female Male Total n % n % n % 116 73 189 61 39 100 124 66 190 65 35 100 240 139 379 63 37 100 59 31 90 66 34 100 59 31 90 66 34 100 118 62 180 66 34 100 38 11 49 78 22 100 30 18 48 62 38 100 68 29 97 70 30 100 19 31 50 38 62 100 35 17 52 67 33 100 54 48 102 53 47 100 Combined Offences Yes No Total Drug Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total 1.6.4 Negative Life Experiences Negative childhood experiences relating to unhappiness and/or trauma to offenders were defined as: abuse (sexual, physical and emotional), neglect or a deviant family of origin (criminal family and/or alcoholic/drug abusive parents). More women than men were identified 54 as having negative childhood experiences (see Table 27). However, this difference was only statistically significant for combined offence categories (x property offences (x 2 2 (1) = 4.80, p < .05) and for (1) = 5.45, p <. 05). It was not significant for either violent or drug offence categories. In this study, adulthood victimisation was defined as physical and sexual assaults from spouses, other known persons, or strangers. Female offenders were significantly more likely to have experienced victimisation in adulthood overall (combined offence categories) (x 62.69, p < .001), and within each specific offence category: drug violent (x 2 (1) TABLE (x 2 (1) = 30.95, p 2 (1) = <. 001), = 18.57, P < .001) and property (x 2 (1) = 13.59, p < .001) (see Table 28). 27 - NEGATIVE CHILDHOOD EXPERIENCES Female % n Combined Offences No Yes Total Male n Total n % % 111 79 190 58 42 100 131 58 189 69 31 100 242 137 379 64 36 100 59 32 91 65 35 100 65 25 90 72 28 100 124 57 181 68 32 100 20 29 49 41 59 100 23 25 48 48 52 100 43 54 97 44 56 100 32 18 50 64 36 100 43 8 51 84 16 100 75 26 101 74 26 100 Drug Offences No Yes Total Violent Offences No Yes Total Property Offences No Yes Total TABLE 28 - ADULTHOOD VICTIMISATION n Female % n Male % n Total % Combined Offences No Yes Total 125 65 190 66 34 100 184 5 189 97 3 100 309 70 379 82 18 100 62 29 91 68 32 100 89 1 90 99 1 100 151 30 181 83 17 100 29 20 49 59 41 100 46 2 48 96 4 100 75 22 97 77 23 100 34 16 50 68 32 100 49 2 51 97 4 100 83 18 101 82 18 100 Drua Offences No Yes Total Violent Offences No Yes Total Property Offences No Yes Total Offenders identified in pre-sentencing reports as experiencing traumatic childhoods also tended to experience abuse during adulthood (!: = .40, p < .001), so to aid later statistical analysis, a new variable was constructed to measure negative experiences across the life course. Adding 55 the scores for negative childhood experiences (O=no, l=yes) and adulthood victimisation (O=no, l=yes) produced scores from 0 to 2 with higher scores equating to more negative life experiences. Not surprisingly, women's negative than men's em = .33, s.d. em = .76, s.d. = .70) life experiences were more = .49) (see Table 29). Sex differences were statistically significant for combined offence categories (p = <.001) and within each specific offence category: drug (p < .001), violent (p = .001) and property (p < .001). TABLE 29 - NEGATIVE LIFE EXPERIENCES Female n % Combined Offences No victimisation 40 75 Childhood or adulthood 86 45 Childhood and adulthood 29 15 Total 190 100 Drug Offences No victimisation 39 43 Childhood or adulthood 43 47 Childhood and adulthood 9 10 Total 91 100 Violent Offences No victimisation 13 27 Childhood or adulthood 23 47 Childhood and adulthood 13 27 Total 49 100 Property Offences No victimisation 23 46 Childhood or adulthood 20 40 Childhood and adulthood 14 7 Total 50 100 2 n Male % n Total % 128 59 2 189 68 31 1 100 203 145 31 379 54 38 8 100 65 24 1 90 72 27 1 100 104 67 10 181 57 37 6 100 22 25 1 48 46 52 2 100 35 48 14 97 36 50 14 100 41 10 0 51 80 20 0 100 64 30 7 101 63 30 7 100 Dependent Variables The next section explores the distributions of dependent variables in preparation for the regression and path analysis presented in forthcoming chapters. The following remand and sentencing outcomes are included: The Remand Status Decision, Length of Custodial Remand, Bail Conditions, Imprisonment Sentencing Decision, Length of Imprisonment Term, Suspended Sentences of Imprisonment, Work-Based Sentences, Care-Based Sentences and Monetary Penalties. Sex differences in this set of variables are reported in the next chapter. 2.1 Remand Outcomes 2.1.1 Remand Status Offenders' remand status was coded by the most serious remand outcome (at large, on bail or in custody) received throughout the trial process. None of the offenders received a remand at large as the most serious remand outcome, because all offenders who were remanded at large were also remanded on bail or in custody at some point. Overall, the majority of 56 offenders received a custodial remand. Drug offenders were more likely to be remanded in custody than violent offenders or property offenders (see Table 30). TABLE 30 m REMAND STATUS 21 Total Combined Offences Custody Bail Total Drug Offences Custody Bail Total Violent Offences Custody Bail Total Property Offences Custody Bail Total n % 210 159 369 57 43 100 129 47 176 73 27 100 54 36 90 60 40 100 31 72 103 30 70 100 2.1.2 Length of Custodial Remand The mean custodial remand length for the whole sample (combined offences) was 40 days (s.d. = 56) (see Figure 7). A log transformation was conducted and an approximately normal distribution was achieved (see Figure 9). Violent offenders received the longest custodial remand lengths (m followed by property offenders (m = 47 days, s.d. = 60 days, s.d. = 72), = 65) and then drug offenders (m = 30 days, s.d. =42). FIGURE 7 m LENGTH OF CUSTODIAL REMAND COMBINED OFFENCES Std. Dev Mean • •II!IIIIIl....___...._..,..I 20.0 60.0 N ~ ~ ~ 55.58 39.9 210.00 100.0140.0 180.0220.0 260.0 300.0 Days spent in custody prior to sentencing The sample size was reduced because 10 women and 9 men were already serving terms of imprisonment when they were charged, in these cases bail was not an option. 21 57 FIGURE 8 - LENGTH OF CUSTODIAL REMAND COMBINED OFFENCES (LOG TRANSFORMATION) ______________ 50~- ~ 40 30 20 10 Frequency o 0.00 1.00 .50 .25 1.50 1.25 .75 2.00 1.75 2.50 2.25 Log of days spent in custody prior to sentencing 2.1.3 Bail Conditions A remand on bail requires offenders to sign a standard bail bond guaranteeing that they will personally attend their next court hearing. If an offender fails to appear, she/he commits an offence and is liable, on summary conviction, to imprisonment or a fine (Summary Proceedings Act 1957 s. 54). A Judge may also impose special bail conditions and these will also be recorded on the bail bond in addition to the standard bail conditions. Bail conditions were considered for all offenders remanded on bail during the court process, regardless of whether they had also been remanded in custody. Overall, 51 % the sample received the standard condition (guarantee of attendance) as the only condition of bail, and 49% received special conditions. Dividing the data by offence category showed that violent offenders were more likely to be given special bail conditions (see Table 31). TABLE 31 - BAIL CONDITIONS n Combined Offences Special Standard Total Drug Offences Special Standard Total Violent Offences Special Standard Total Property Offences Special Standard Total Total % 151 159 310 49 51 100 78 76 154 51 49 100 38 25 63 60 40 100 35 58 93 38 62 100 58 2.2 Sentencing Outcomes 2.2.1 Imprisonment Sentencing Decision Approximately half of the sample received a sentence of imprisonment. Separating the data into specific offence categories showed violent offenders were more likely to be imprisoned than either drug or property offenders (see Table 32). TABLE 32 -IMPRISONMENT SENTENCING DECISION n Combined Offences Imprisoned Not Imprisoned Total DruQ Offences Imprisoned Not Imprisoned Total Violent Offences Imprisoned Not Imprisoned Total Property Offences Imprisoned Not Imprisoned Total 2.2.2 Total % 202 186 388 52 48 100 102 80 182 56 44 100 68 30 98 69 31 100 32 76 108 30 70 100 Length of Imprisonment Term The mean imprisonment term overall was 25 months ~. = 28) and the distribution for combined offence categories can be seen in Figure 9. The overall distribution was badly skewed, with some extreme outliers. A log transformation was thus conducted and an approximately normal distribution was achieved (refer to Figure 10). Violent offenders were given the longest imprisonment terms (m = followed by drug offenders (m = 21 months, s.d. s.d. 39 months, s.d. = 37),22 = 21) and property offenders (m = 12 months, = 9). 22 In New Zealand, people convicted of murder can only receive one sentence and that is life imprisonment. The average amount of time an offender imprisoned for murder can expect to serve is 11 or 12 years (Newbold, 2000: 146). Subsequently, for offenders convicted of murder in this sample, imprisonment terms were coded as 11 112 years. 59 FIGURE 9- LENGTH OF IMPRISONMENT TERM COMBINED OFFENCES Std, Dev Frequen Mean N 0.0 20.0 10.0 40.0 30.0 80.0 60.0 50.0 70.0 100.0 90.0 120.0 110.0 = 28.19 = 25.3 = 202.00 140.0 130.0 Months Sentenced FIGURE 10 - LENGTH OF IMPRISONMENT TERM COMBINED OFFENCES (LOG TRANSFORMATION) Std. Dev = .41 Mean = 1.21 Freque N = 202.00 0.00 .25 .50 .75 1.00 1.251.50 1.75 2.00 2.25 Log of Months Sentenced 2.2.3 Non-Custodial Sentences This section outlines non-custodial sentencing alternatives (suspended sentences of imprisonment, work-based sentences, care-based sentences and monetary penalties). The Criminal Justice Act (1985, s. 26) states that a monetary penalty may run concurrently with a sentence of imprisonment, so this penalty is available to all offenders in the sample (n=388). However, other non-custodial sentences, i.e. suspended sentences of imprisonment, work-based sentences and care-based sentences, cannot run concurrently with imprisonment, so sample sizes reduced (n=186) in these cases and included only those offenders given a non-custodial sentence (see Table 32). 60 2.2.3.1 SUSPENDED SENTENCES OF IMPRISONMENT As part of their non-custodial sentence, 30% of offenders in the current sample received suspended prison terms. Drug offenders were more likely than violent and property offenders to get a suspended imprisonment term (see Table 33). TABLE 33 - SUSPENDED SENTENCES OF IMPRISONMENT Total % n Combined Offences Yes No Total 55 131 186 30 70 100 39 41 80 51 49 100 8 22 30 27 73 100 8 68 76 10 90 100 Drug Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total 2.2.3.2 WORK-BASED SENTENCES Work-based sentences included both periodic detention and community service. In the current sample, the majority of offenders who received a non-custodial alternative received work as part of that sentence. Separating the data by offence category revealed that work-based sentences were equally likely in all offence categories (see Table 34). TABLE 34 - WORK-BASED SENTENCES n Total % Combined Offences Yes No Total 110 76 186 59 41 100 49 31 80 61 39 100 13 17 30 57 43 100 44 32 76 58 42 100 Drull Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total 2.2.3.3 CARE-BASED SENTENCES Care-based sentences included supervision, community programme and psychiatric committal. In the current sample, 41% of all offenders received care as part of their non- 61 custodial sentence. Breaking the data down into offence categories showed violent and drug offenders were more likely than property offenders to be given care-based sentences (see Table 35). TABLE 35 m CARE-BASED SENTENCES n Combined Offences Yes No Total Drug Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total 2.2.3.4 Total % 109 186 41 59 100 33 47 80 59 41 100 18 12 30 60 40 100 26 50 76 34 66 100 77 MONETARY PENALTIES Monetary penalties included fmes, reparation, confiscation/forfeiture and payment of court costs. Very few offenders received monetary penalties as part of their sentence and property offenders were more likely than other offence categories to get this penalty (see Table 36). TABLE 36 - MONETARY PENALTIES n Combined Offences Yes No Total Drug Offences Yes No Total Violent Offences Yes No Total Property Offences Yes No Total Total % 62 326 388 16 84 100 15 167 182 8 92 100 3 95 98 3 97 100 44 64 108 41 59 100 3 Closing Comments The aims of this section were to establish and provisionally document the variables to be used in the later multiple regression and path analysis. Distributions were checked for normality, and the distributions of dichotomous dependent variables were examined to ensure 62 they were relatively even. Independent variables were also examined to detect multicollinearity: sex differences in the independent variables were presented and discussed. The following variables were adjusted and relatively normal distributions were achieved: period since last criminal conviction (Figures 2, 3), value of property (Figures 4, 5), number of conviction counts (Figure 6, Table 21), length of custodial remand (Figures 7, 8), length of imprisonment term (Figures 9, 10). Results showed that the majority of dichotomous dependent variables were relatively even. However, the following variables had grossly uneven splits so are excluded from any further statistical analysis: remand status drug offences (Table 30), suspended sentences of imprisonment for violent offences (Table 33), suspended sentences of imprisonment for property offences (Table 33), and monetary penalties (Table 36). One further aim was to refine the measurement and number of independent variables consistent with sound theoretical and methodological reasoning. The following variables were identified for inclusion in subsequent statistical analysis: sex, age, ethnicity, overall familial situation, employment status, seriousness of criminal history, period since last criminal conviction, bail breaches, offender's role in the offence, co-offenders, offence location, use of weapons, property value, victim-offender relationship, sex of victim, victim injury, plea, number of conviction counts, length of custodial remand,23 pre-sentence recommendations, overall health, substance abuse, negative life experiences. 24 Finally, identifying sex differences in independent variables was an important first step toward exploring the impact of gender on judicial decision-making. Having matched the sample of male and female offenders according to a number of important variables, an examination of independent variables showed some similarities and differences between men and women. Court processing and the content and context of offending was similar for men and women. This was partially due to the matched sampling method, which allowed men and women to be matched one-to-one by statutory offence and court of sentencing. In addition, plea and number of conviction counts were also possible match variables, so few sex differences were noted in these factors. In the current sample, men and women were also equally likely to act alone or with others, to take a primary or secondary role, and to act in a public or private location. Male 23 The variable length of custodial remand, was both an independent and dependent variable. 24 Major offence categories (see Tables 8 t011) and court of sentencing (see Table 19) were not selected for inclusion in further analysis because men and women were matched one-to-one on these variables. The samples used in later statistical analyses vary at times (e.g. when assessing length of imprisonment term only those offenders sentenced to imprisonment were included) but statistical 63 and female violent offenders were equally likely to use weapons, victimise strangers or known persons, victimise their own sex, and cause emotional or physical harm to victims. Matched sampling also meant that few ethnic or age differences existed between the men and the women in this sample but sex differences were noted in other non-matched sociodemographic factors. Women were significantly less likely than men to be involved in paid work, were more likely to have childcare responsibilities, and generally experienced stronger familial ties. Because of differences in gender-role expectations, this result was not surprising: New Zealand women are expected, and are more likely, to be responsible for the care of children (Craig, 1992; James & Saville-Smith, 1994). As expected, men's criminal histories were significantly more extensive and senous than women's. Male property offenders also breached bail more often than women did but generally speaking, the sexes had an equal propensity to breach bail. Sex differences in criminal history and breaching bail may be important, because one factor a Judge must give consideration to when sentencing or deciding remand status is the likelihood of re-offending (Green, 1961; Chesney-Lind, 1978; Farrington & Morris, 1983; Eaton, 1986; Eaton, 1987; Daly and Bordt, 1995; Young, 1997). The content of pre-sentence reports revealed that many offenders experienced health difficulties (mental and physical), substance abuse problems, and negative life experiences, but poor health and negative life experiences were identified significantly more often for women. This difference presented women as less culpable and blameworthy, perhaps explaining why these female offenders were significantly less likely to receive pre-sentence imprisonment recommendations. Finding differences in male and female offenders' past criminality and lives is consistent with past research (see Chapter One). In contrast to men, women's less serious offence histories, strong familial ties, positive pre-sentencing recommendations, health problems and negative life experiences could be used to justify less severe judicial treatment. However, what must now be established is whether or not there are in fact sex differences in sentencing and remand outcomes and if so, whether or not these other identified gendered differences can plausibly explain this. These issues are the focus of the next chapter. tests showed that in all cases men and women remained closely matched. 64 Chapter Three Hierarchical Multiple Regression Analysis Thus far, it has been established that male and female offenders may appear before the court with different background circumstances (see Chapter Two). This result is consistent with past research which has further found sex differences in some judicial outcomes, for some offences, even when these background circumstances are controlled (see Chapter One). In the present chapter, I investigate whether there are sex discrepancies in remand and sentencing in the current sample. I ask if previously identified variance, (e.g. familial ties, seriousness of criminal history, pre-sentencing recommendations, health problems, negative life experiences), between men and women could plausibly explain any differences found. Given past research findings, I expected to find some degree of sex-based variance in remand and sentencing outcomes but realised that this may depend on the judicial outcome and/or the offence category (drug, violent, property) being investigated. This chapter presents zero-order correlations 1 and the results of some hierarchical multiple regression analyses. Conducting these statistical investigations helped develop hypotheses about how predicted sex differences in judicial outcomes might be explained. Using path analyses, these hypotheses are tested in the next chapter. Thus, results reported in the current chapter were particularly important for building later statistical investigations. The aims of the hierarchical multiple regression analyses were: 4& To establish the extent of sex differences in sentencing and remand outcomes (when controlling for other socio-demographic variables). ® To identify other major variables affecting sentencing and remand outcomes, which might explain sex differences in judicial outcomes. Results are presented in sections, according to judicial outcome. Section one considers the above questions with regard to sentencing and includes the imprisonment decision (imprisoned or not imprisoned), length of imprisonment term, work-based sentences and care-based sentences for combined and specific offence categories (drug, violent, property). Suspended 1 Zero-order correlations describe the strength of a relationship between an independent and dependent variable before other factors are controlled. 65 sentences of imprisonment were analysed for combined offence categories and for drug offences (but not for violent or property offences because of restricted variance). Zero-order correlations are presented for all judicial outcomes but regression results are only reported if significant sex differences were found and samples were large enough to make reasonable inferences given the number of independent variables (this is a problem with power). Hierarchical regression analyses were thus not conducted on samples smaller than 95. In Section two, regressions for remand outcomes are outlined and the above questions were considered with regard to length of custodial remand and bail conditions for combined and specific offence categories (drug, violent, property). The remand status decision is analysed for combined offence categories, for property offences and violent offences (but not for drug offences because of restricted variance). Again, hierarchical regressions were not conducted on samples smaller than 95. Ordinary least squares regression was utilised for the hierarchical r~gesion analyses. Concern has been expressed regarding the use of ordinary least squares regression when dichotomous dependent variables are used (Daly, 1989a: 149). Research conducted by Steffensmeier, Kramer and Streifel (1993: 426), for example, found that ordinary least squares regression presented a rather conservative measure, and the effects of some independent variables (including sex, age and race) were under-estimated. In response to this concern, logistic regression analyses were also conducted on the current data. However, the results were very similar. Hence, only the ordinary least squares regression results are presented, because these are more readily interpretable than those produced with logistic regression. 1 Section One: Sentencing Outcomes Twenty-three independent variables and five dependent variables were analysed. The independent variables were grouped into six categories: 1. Socio-demographic variables (sex, ethnicity, age, employment status, overall familial situation). 2. Criminal history variables (seriousness of criminal history, period since last criminal conviction). 3. Offence variables (offender's role in the offence, co-offenders, offence location, use of weapons, value of property, victim-offender relationship, sex of victim, victim injury). 4. Court process variables (plea, number of conviction counts) 66 5. Remand status (length of custodial remand) 2 6. Pre-sentence report variables (recommendations, overall health, substance abuse, negative life experiences). Hagan & Bumiller (1983: 3) note that "sentencing is an end result of a decision-making process that involves offenders moving through a series of ... stages." Farrell and Swigert (cited in Hagan & Bumiller 1983:3) further explain: ''The highly structured nature of the judicial system lends itself to a systematic analysis of legal processing. The discrete ordering of events - the social characteristics of the defendants prior to their entry into the system, their accumulated criminal histories, ... pre-trial release ... and final disposition - constitutes a series of stages that allows the researcher to assert the causal sequence of relationships." Subsequently, the possible causal ordering of the groupings of independent variables in the current study made it appropriate and meaningful to use hierarchical regression analysis. For example, socio-demographic factors could causally influence criminal history, offence characteristics, court processes, remand status and pre-sentence report variables,3 but not vice versa. Subsequently, the six variable groups were entered into the regression equation in the order listed above. Thus, for example, the effect of sex on sentencing was calculated while controlling for the other demographic variables, but not controlling for criminal history (criminal history cannot cause sex). However, the effects of criminal history on sentencing did control for sex and other socio-demographic variables given that sex could causally influence criminal history. The methodological logic used in setting up and interpreting the associated regression analyses is standard practice, but will be explained further where necessary. Dependent variables not only measured the most commonly analysed imprisonment decision by the Judge (imprisoned or not imprisoned), but also included the length of imprisonment term and the different types of non-custodial sentence available to Judges. The types of noncustodial sentence were grouped into three categories: The variable, length of custodial remand, is employed as an independent variable in this case and differs from the dependent variable length of custodial remand described previously (Chapter 2 - Figure 9). Coded in days from zero upwards, it is an overall measure of remand status, unlike the latter, which only included offenders remanded in custody (that is, it includes people not remanded in custody). 3 Some debate exists about the appropriateness of including remand status and pre-sentence recommendations into regressions assessing sentencing outcomes. Prior research has found that the relationships between these two variables and sentencing are often strong enough for one to question the "methodological separateness of the independent and dependent variable" (Hagan & Bumiller, 1983: 33). With regard to remand status and sentencing, Daly (1989a: 148) argues that "typically, the same variables predict each decision, indicating that court officials use the same logic in deciding if a defendant should be segregated from the community, whether before or after conviction." In contrast, Kruttschnitt's & Green's (1984: 546-547) research found that remand status acted as a mediating variable and note that "a sizeable part of the influence of sex on the decision to incarcerate can be accounted for by the pretrial release status." Such a finding is seen to call "into question those studies which fail to control for pre-trial release." Given the debates, excluding remand status and pre-sentence report recommendations from the analysis at this point would be inadvisable However, if current regression results show that remand status and pre-sentencing recommendations impact on sentencing then the suitability of including or removing them from further path analyses will be discussed in the following chapter. 2 67 <II Suspended sentences of imprisonment. @! Care-based sentences (supervision, community programme and psychiatric committal). <II Work-based sentences (periodic detention and community service). Coding of the independent and dependent variables is presented in Table 37. For reference purposes, the table or figure number where variable descriptions and distributions can be found in the previous chapter are presented next to each variable in the table. TABLE 37- DESCRIPTION OF VARIABLES USED FOR THE SENTENCING ANALYSES Independent Variables Socio-demographic variables Sex Ethnicitv Age Employment Status Overall Familial Situation Criminal History Variables Seriousness of criminal history Period since last criminal conviction # Offence Variables Offender's role Co-offenders Offence location Use of weapons Value of Property Victim-offender relationship Sex of victim Victim injury Court Process Variables Plea Number of Conviction Counts Remand Status Length of custodial remand Pre-Sentence Reports Pre-sentence recommendations Overall Health (physical and mental) Substance abuse Negative life experiences Dependent variables Imprisonment sentencinQ decision Length of imprisonment term Suspended sentence of imprisonment imposed Work-based sentence imposed Care-based sentence imposed Description Reference 0= male, 1= female 0= European, 1= Other In years 1= in paid employment, 2- not in paid employment A continuous variable, 0= no familial ties, 1= minimal familial ties, 2= moderate familial ties, 3= strong familial ties Table 1 Figure 1 Table 2 Table 6 P~e3 A continuous variable (number of prior criminal convictions, number of prior criminal convictions in a similar offence terms). category, number o/PJior im~sonet A continuous variable, in months (log transformation) Figure 3 0= primary/equal, 1- secondary 0= acted alone, 1= acted with others location 0= public location, 1=~ivate o~ yes, 1- no A continuous variable, in dollars (log transformation) O=stranQer, 1=known person 0= male, 1= female 0- no injury, 1- emotional distress, 3~ physical injurl Table 13 Table 12 Table 14 Table 15 Figure 5 Table 16 Table 17 Table 18 0= Guilty, 1= NolJluilli' 0- 1 count, 1- More than 1 count Table 20 Table 21 A continuous variable, in dfll!sJlQa transformation) Figure 8 0= imprisonment, 1= other 0- poor mental and physical health, 1- has either mental or phvsical health problems, 2= has no health.2l'oblems 0= yes, 1= no Table 22 Table 25 0= no victimisation, 1= victimised either in childhood or adulthood, 2= victimised both in childhood and adulthood Description 0= imprisoned, 1= Not imprisoned A continuous variable, in months Jlog transformationL 0= yes, 1- no Table 29 Page 38-39 Table 26 Table 32 Figure 10 Table 33 0= yes, 1= no Table 34 0= no, 1= yes Table 35 # Period since last Criminal conViction related only to offenders preViously conVicted. Including thiS variable meant that a different sample was used. Separate regressions were run to allow its inclusion but all results were non-significant so will not be presented. 1.1 Imprisonment Sentencing Decision The zero-order correlation matrix for the Imprisonment Sentencing Decision with independent variables is presented in Table 38. These zero-order correlations describe the statistical strength of the relationship between each independent variable and the decision to imprison without controls. The first column in Table 38 incorporates all offence categories, 68 while the following three are offence-specific and show zero-order correlations for drug, violent and property offence categories. Zero-order correlations (with sex) for the imprisonment decision represent the percentage differences 4 between women and men. For combined offence categories, there were significant correlations between sex and the imprisonment decision: women were 17% less likely than men to receive a sentence of imprisonment. Patterns also showed that sex differences were offence specific. Women were 16% less likely than men to be imprisoned for drug offences, and 30% less likely to be imprisoned for property offences. In contrast, women who committed violent crime were only 8% less likely than men to receive a sentence of imprisonment, but this was statistically non-significant. Zero-order correlations between the imprisonment decision and other independent variables indicate the complexity of sentencing decisions. For combined offence categories, the three most important factors for the imprisonment decision were pre-sentence recommendations, remand status, and the seriousness of criminal history. Imprisonment was significantly more likely if the Probation Officer recommended imprisonment in the pre-sentence report, the offender was remanded in custody for long periods and he or she had a more serious criminal history. An examination of specific offence categories revealed that these three factors were important regardless of offence. Correlations between categorical variables allows the use of the Binomial Effects Size Display to describe the results (see Rosenthal & Rubin, 1982). These cited percentages are correct, even though the traditional way of computing effect size is by squaring the correlation. 4 69 TABLE 38 ~ ZERO ORDER CORRELATIONS ~ IMPRISONMENT SENTENCING DECISION SOCIO-DEMOGRAPHIC VARIABLES Sex Aqe Ethnicity Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Offender's role in the offence Co-offenders Offence location Use of weapons Property Value Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts REMAND STATUS Lenqth of custodial remand PRE-SENTENCE REPORTS Pre-sentence Recommendations Overall Health (physical and mental) Substance abuse Negative life experiences .. * Significant at p <.05 Combined Offences 11=388 Drug 11=182 Violent 11=98 Property 11=108 .17 • .01 -.06 .02 -.13* .16' -.16* .11 .22 -.10 .08 .08 -.20' .11 -.04 .30* .10 .05 .05 -.18' -.31* -.40* -.26' -.40* .06 -.12* -.10* NA NA NA NA NA .13 -.07 -.05 NA NA NA NA NA .04 -.21 .03 .08 NA .26* .10 .06 .08 .02 -.14 NA -.02 NA NA NA .06 -.21* -.03 -.24* .07 -.41* .14 -.02 -.41* -.33* -.25* -.41' .51* -.13* .20' -.03 .45* -.03 .07 -.01 .52* -.20* .29* -.01 .46' -.15 .25' .06 5 As previously described, a hierarchical regression analysis by groups of independent variables was next carried out for combined, drug and property offence categories. No further analysis was conducted for violent offences because the zero-order correlations showed that men and women had approximately equal chances of being imprisoned. For combined offence categories, hierarchical regression results revealed that the sex differences previously noted at the zero-order level increased slightly in strength and remained significant (see results in Table 39). Regardless of other socio-demographic factors, the regression coefficient 6 for sex showed that men were still 20% more likely than women to be imprisoned. Results also showed that whilst controlling for all preceding variable groups, the likelihood of imprisonment significantly increased for offenders with more serious criminal histories, a larger number of conviction counts, and for those who acted alone or in a private setting. Offenders who spent a long time in custodial remand also had their imprisonment chances significantly increased (even whilst controlling for socio-demographic, criminal history, offence and court process variables). In addition to the length of custodial remand, a positive pre-sentence report recommendation substantially decreased the probability of 5 The codes for the imprisonment decision are: O=imprisoned, 1=not imprisoned The regression coefficient estimates the amount of change in the dependent variable (imprisonment decision) resulting from change in the independent variable (e.g. sex) with all other factors in the equation controlled (in this case other socio-demographic variables). A regression coefficient is represented as a number between -1.0 and +1.0. A coefficient of 0 indicates no relationship between an dependent and independent variable. A correlation of -.10 means there is a perfect negative relationship; a correlation of +1.0 means there is a perfect positive relationship. Other coefficients imply relationships of differing strengths" (Reaves, 1992: 349). 6 70 imprisonment. Poor health, another factor noted in the pre-sentence report, also significantly reduced the probability of imprisonment. TABLE 39 - IMPRISONMENT SENTENCING DECISION, 7 COMBINED OFFENCES STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOCIO-DEMOGRAPHIC VARIABLES R (multiple) Sample Size .23· n= 3BB ,20· Sex Age -,01 Ethnicity -,06 ,00 Overall Familial Situation Employment status -,15* CRIMINAL HISTORY VARIABLES R (multiple) .35* Sample Size n= 3BB -,28* Seriousness of criminal history OFFENCE VARIABLES R(multiple} .39* Sample Size n= 3BB ,08 Role -,14* Co-offenders -,11* Location COURT PROCESS VARIABLES R(multiple} .46* Sample Size n= 3BB -,01 Plea -,25* Number of Conviction Counts REMAND STATUS ,53* R (multiple) Sample Size n= 384 -,26* Length of custodial remand PRE-SENTENCE REPORTS R (multiple) .62· Sample Size n= 371 ,36* Pre-sentence recommendations -,09* Overall Health (physical and mental) ,02 Substance abuse -,03 Negative life experiences " * Slgmflcant at p <,05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included, Similar to combined offence categories, women convicted of drug offences remained 17% less likely than men to be imprisoned once other socio-demographic factors were controlled (see Table 40). Furthermore, a smaller number of conviction counts, a non-custodial pre-sentencing recommendation, a less serious criminal history and shorter custodial remand periods significantly decreased the probability of imprisonment. 7 The codes for the imprisonment decision are: O=imprisoned, 1=not imprisoned 71 TABLE 40 m IMPRISONMENT SENTENCING DECISION, B DRUG OFFENCES m STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOCIO-DEMOGRAPHIC VARIABLES R (multiple) Sample Size .29' n=182 Sex .17' Age -.16' Ethnicitv .10 Overall Familial Situation -.02 Employmenl status -.16' CRIMINAL HISTORY VARIABLES R (multiple) .42' Sample Size n=182 Seriousness of criminal history -.32' OFFENCE VARIABLES R (multiple) .46' Sample Size n=182 Role .16' Co-offenders -.15' Offence location .06 COURT PROCESS VARIABLES R (multiple) .52' Sample Size n=182 Plea -.09 Number of Conviclion Counts -.24' REMAND STATUS R (multiple) .53' Sample Size n=182 Length of custodial remand -.14' PRE-SENTENCE REPORTS R (multiple) .60' Sample Size n=179 Pre-sentence recommendation .29' Overall Health -.06 Substance abuse -.10 Negative life experiences -.05 .. , Significant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. For property offences, the introduction of other socio-demographic factors into the equation actually increased the zero-order sex difference by 5% (see Table 41) and women remained significantly less likely than men to be imprisoned. Once again, with all preceding variables controlled, results showed that a community-based pre-sentencing recommendation, a less serious criminal history, and shorter custodial remand periods, significantly decreased imprisonment chances. BThe codes for the imprisonment decision are: O=imprisoned, 1=not imprisoned 72 TABLE 41 = iMPRISONMENT SENTENCING DECISION, 9 PROPERTY OFFENCES = STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOClo-DEMOGRAPHIC VARIABLES R (multiple) Sample size .40* 11=1 OS ,36* ,06 ,02 -.04 -.21* Sex AQe Ethnicity Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES R(multiple) .4S· Sample size n=10S -.2S· Seriousness of criminal history OFFENCE VARIABLES .51* R (multiple) Sample size n=10S Role .04 Co-offenders -.01 Location -.08 Value of Property -.12 COURT PROCESS VARIABLES R (multiple) .54' Sample size n=10S Plea .07 -.15 Number of Conviction Counts REMAND STATUS R (multiple) .60' Sample size n=10S Length of custodial remand -.34* PRE-SENTENCE REPORTS R (multiple) .66' Sample size 11=97 Pre-sentence recommendation .40* Overall Health -.08 Substance abuse .09 .. • Significant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included, 1.2 Length of Imprisonment Term Examining zero-order correlations between imprisonment term and each of the independent variables represents the first step toward establishing whether sex differences in imprisonment terms exist and whether other variables might explain this.10 The zero-order correlation matrix for length of imprisonment term with dependant variables is presented in Table 42. The first column includes all offence categories, while the following three are offence specific and separate into drug, violent and property offence categories. For combined offence categories, zero-order correlations showed that men were sentenced to significantly longer terms em = 28 months, s.d = 27) than were women em = 21 months, s.d. = 30). Patterns also showed that the sex difference was offence specific. length for men (m = 46 months, s.d. = 35) longer than for women (m = 31 months, s.d. The imprisonment convicted of a violent offences was significantly = 39). Men convicted of drug offences (m = 23 The codes for the imprisonment decision are: O=imprisoned, 1=not imprisoned this sample only included those sentenced to imprisonment, zero-order correlations between sex and the independent variables revealed similar sex differences to those noted for the whole sample as outlined in Chapter One. 9 10 Although 73 months, s.d. = 19) were also sentenced to significantly longer terms than women (m months, s.d. = 22) and this difference was also statistically significant (p < .05). = 17 11 For combined offence categories, a large number of conviction counts, a long custodial remand period, and a pre-sentence recommendation of imprisonment, all significantly increased the imprisonment term given. An examination of specific offence categories further revealed favourable pre-sentence recommendations reduced imprisonment terms regardless of offence type. A long custodial remand period significantly increased imprisonment terms for offenders convicted of violent and drug crimes, but had little impact on property offenders. Imprisonment terms increased significantly for property offenders if they were non-European and in good health, but were identified with negative life experiences. A large number of conviction counts significantly increased imprisonment terms for drug offenders. Playing a secondary role in the offence, pleading guilty, being convicted of a small number of counts and having negative life experiences significantly decreased imprisonment terms for violent offenders. TABLE 42 - ZERO-ORDER CORRELATIONS - LENGTH OF IMPRISONMENT TERM Combined Offences Drug Violent Property n=202 n=102 n=68 n=32 -.20' -.02 .05 -.02 -.06 -.26' .10 -.15 .11 -.00 -.36' .07 -.01 -.15 -.15 -.06 .00 .39* .17 -.12 .06 .12 .04 .18 -.04 .10 -.05 NA NA NA NA NA .09 .16 -.02 NA NA NA NA NA -.27* -.17 .02 -.08 NA .12 -.10 .09 -.28 -.09 -.12 NA .11 NA NA NA .11 .24' .01 .33' .18* .18* .20 -.07 .26' .31' .32' .04 -.42' .03 -.00 .07 -.33' .11 -.02 .02 -.44' -.04 .09 -.16' -.59* .36' .05 -.32' SOCIO-DEMOGRAPHIC VARIABLES Sex Age Ethnicitv Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Offender's role in the offence Co-offenders Offence location Use of weapons Property Value Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts Remand Status Length of custodial remand PRE-SENTENCE REPORTS Pre-sentence Recommendations Overall Health (physical and mental) Substance abuse Negative life experiences .. Significant at p <.05 . The identical hierarchical regression analyses, with the same four groups of independent variables, utilised for the imprisonment decision were again conducted for combined and drug offence categories. The purposes of these regressions were twofold: first, to ascertain the extent to which sex impacted on imprisonment terms once other socio-demographic factors were Significance tests for correlations are mathematically identical to t-tests of differences between means. The means are cited to give information about of the magnitude of the sex differences. 11 74 controlled; second, whilst controlling for all preceding variables, to identify other significant sentencing factors which might explain sex differences in imprisonment terms. No further statistical analysis was conducted for property offences because of minimal zero-order sex differences. Although significant sex differences were found in imprisonment terms for violent offences, further hierarchical regressions were not conducted because of small sample sizes (see Table 42). However, the sex difference noted for violent offences will be investigated using a path analysis in the next chapter because only major, and thus far fewer, independent variables need be included. Table 43 shows that the sex difference found at the zero-order level for combined offence categories remained. Even with other socio-demographic factors controlled, women's imprisonment terms were approximately seven months shorter than men's. Aside from sex, it was found that with all preceding factors controlled; a serious criminal history, a small number of conviction counts, a short custodial remand period and a non-custodial sentencing recommendation, were all conducive to shorter imprisonment terms. TABLE 43- LENGTH OF IMPRISONMENT TERM, COMBINED OFFENCES - STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOCIO-DEMOGRAPHIC VARIABLES R (multiple) Sample Size .22' n=202 Sex -.20' Age -.03 Ethnicity .07 Overall Familial Situation .01 Emplovment status -.05 CRIMINAL HISTORY VARIABLES R (multiple) .25' Sample Size n=202 Seriousness of criminal history .15' OFFENCE VARIABLES R (multiple) .29' Sample Size n=202 Role -.09 Co-offenders .14 Location -.04 COURT PROCESS VARIABLES R(multiple) .39' Sample Size n=202 Plea .14 .26* Number of Conviction Counts REMAND STATUS R (multiple) .45' Sample Size n=202 Length of custodial remand .23* PRE-SENTENCE REPORTS R (multiple) .61' Sample Size n=195 -.33* Pre-sentence recommendations -.01 Overall Health (physical and mental) Substance abuse .0 Negative life experiences -.1 * Significant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that pOint included. .. 75 Regardless of other socio-demographic factors, female drug offenders received terms of imprisonment approximately seven and a half months shorter than men's (see Table 44). Acting alone, pleading guilty, a minimal criminal history, a small number of conviction counts and a non-custodial pre-sentencing recommendation also significantly decreased the length of imprisonment terms. TABLE 44 - LENGTH OF IMPRISONMENT TERM, DRUG OFFENCES - STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOClo-DEMOGRAPHIC VARIABLES R (multiple) Sample size .36' n=102 -,30' Sex Aqe ,09 -,12 Ethnicity ,16 Overall Familial Situation ,07 Employment status CRIMINAL HISTORY VARIABLES R (multiple) .35' Sample size n=102 ,15' Seriousness of criminal history OFFENCE VARIABLES R (multiple) .40' n=102 Sample size -,01 Role ,20' Co-offenders -,05 Location COURT PROCESS VARIABLES R (multiple) .57* Sample size n=102 ,20' Plea .44' Number of Conviction Counts REMAND STATUS R (multiple) .58' Sample size n=102 ,11 Lenqth of custodial remand PRE-SENTENCE REPORTS R (multiple) .62' Sample size n=98 -,22' Pre-sentence recommendation ,10 Overall Health ,08 Substance abuse ,12 Neqative life exPeriences " * Slgmflcant at p <,05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. 1.3 Suspended Sentences of Imprisonment The zero-order correlation matrix for suspended sentences of imprisonment was examined to determine if sex differences existed in the likelihood of receiving this sentencing outcome. No significant sex differences were found and results showed that men and women received suspended sentences of imprisonment at approximately similar rates, regardless of offence type (see Table 45). Subsequently, no further statistical analysis was conducted. 76 TABLE 45 - ZERO-ORDER CORRELATIONS - SUSPENDED SENTENCES OF IMPRISONMENT 12 Combined Offences Drug Violent Property n=186 n=80 n=30 n=76 -.06 .10 .04 .10 -.05 -.13 .24* .04 .05 -.12 .07 -.05 .31 .15 .06 .10 .12 -.04 .03 .01 .01 .13 .31 -.13 -.04 -.02 .22* -.06 -.02 .02 .00 -.04 .02 NA NA NA NA NA NA NA NA NA NA -.15 -.04 .22 -.26 -.08 .06 SOCIO-DEMOGRAPHIC VARIABLES Sex Age Ethnicity Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Ollender's role in the ollence Co-offenders Ollence location Use of weapons Property Value Victim-ollender relationship Sex of victim Victim iniury COURT PROCESS VARIABLES Plea Number of Conviction Counts REMAND STATUS Lenqth of custodial remand PRE-SENTENCE REPORTS Pre-sentence Recommendations Overall Health (phvsical and mental) Substance abuse Negative life experiences .. * Significant at p <.05 1.4 NA NA .07 .12 -.02 .35* NA NA NA .10 .15 -.33 -.06 -.04 .15 .19* .08 .05 .08 -.01 -.08 -.14* .02 .00 -.09 -.10 -.11 -.05 -.32 .00 .08 -.10 -.12 -.13 .24* Work-Based Sentences The zero-order correlation matrix for work-based sentences is presented in Table 46. Correlations were examined to determine whether sex differences existed in the likelihood of being sentenced to work. For combined, drug and property offences, men and women received work-based sentence at approximately similar rates. Men were 35% more likely than women to have received a work-based sentence for violent offending but this difference was nonsignificant probably due to small sample size. Given these results, no further statistical analysis was conducted for work-based sentences. The codes for Suspended Sentences of Imprisonment are; O=yes, received a suspended sentence of imprisonment, 1=no, did not receive a suspended sentence of imprisonment. 12 77 TABLE 46= ZERO=ORDER CORRELATIONS m WORK= BASED SENTENCES 13 Combined Offences Drug Violent Property n=186 n=80 n=30 n=76 -.09 .09 -.08 .04 -.03 .01 .07 .02 .05 -.08 -.35 .24 -.08 -.02 -.21 -.08 .06 -.17 .04 .09 .01 -.07 .14 .06 .07 .17' -.02 NA NA NA NA NA .19 .10 -.21 NA NA NA NA NA .16 .39* .05 .04 NA .07 -.45' -.06 -.10 .15 .05 NA .04 NA NA NA .09 -.11 -.05 -.23' .16 .03 .20 -.05 -.04 .08 -.27 -.11 .01 .21' .10 -.07 -.12 .17 .19 -.06 -.06 .29 .06 -.26 .17 .21 .04 .02 SOCIO-DEMOGRAPHIC VARIABLES Sex Age Ethnicity Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Offender's role in the offence Co-offenders Offence location Use of weapons Property Value Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts REMAND STATUS Length of custodial remand PRE-SENTENCE REPORTS Pre-sentence Recommendations Overall Health (physical and mental) Substance abuse Negative life experiences ., • Slgmflcant at p <.05 1.5 Care-Based Sentences The zero-order correlation matrix presented in Table 47 shows minimal sex differences in care-based sentences for combined offences and drug offences. Women convicted of violent and property offences were more likely than men to receive sentences of care but this was nonsignificant so no further statistical analysis was conducted. 13 The codes for Work-Base Sentences are: O=yes, received a work-based sentence, 1=no, did not receive a work-based sentence. 78 TABLE 47 = ZERO-ORDER CORRELATIONS - CARE-BASED SENTENCES 14 Combined Offences Drug Violent PrO£erk n=186 n=80 n=30 n=76 .09 -.21' .13 -.12 .18' -.02 -.24' .18 -.24' .28' .24 -.12 .33 .25 .18 .18 -.15 -.03 -.15 .05 .07 .27' -.38' .07 .01 -.07 -.01 NA NA NA NA NA -.06 -.19 -.04 NA NA NA NA NA .12 .05 .11 .14 NA -.02 .02 -.01 -.00 -.01 -.11 NA -.01 NA NA NA -.09 -.22' -.09 -.36' .11 .00 -.18 -.24' -.14' -.14 -.19 -.07 -.00 -.12 -.28' .23' -.02 -.12 -.41 • .22' -.27 .04 -.13 .28 .18 -.21 -.20 .17 SOCIO-DEMOGRAPHIC VARIABLES Sex Age Ethnicity Overall Familial Situation Employment status CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Offender's role in the offence Co-offenders Offence location Use of weapons Property Value Victim-offender relationship Sex of victim Victim iniury COURT PROCESS VARIABLES Plea Number of Conviction Counts REMAND STATUS Length of custodial remand PRE-SENTENCE REPORTS Pre-sentence Recommendations Overall Health (physical and mental) Substance abuse Negative life experiences .. • Slgmflcant at p <.05 2 Section Two: Remand Outcomes In this section zero-order correlation matrices and hierarchical regressions are examined to determine whether sex differences in remand exist and if so, which variables might account for this. Twenty independent variables and three dependent variables were examined at the zero-order level in addition to being included into the remand regressions. The independent variables were grouped into four categories: 1. Socio-demographic variables (sex, ethnicity, age, employment status, overall familial situation). 2. Criminal history variables (seriousness of criminal history, period smce last criminal conviction, bail breaches). 3. Offence variables (offender's role in the offence, co-offenders, offence location, use of weapons, value of property, victim-offender relationship, sex of victim, victim injury). 4. Court process variables (plea, number of conviction counts) As was the case with sentencing outcomes, the possible causal order of these groupings made it meaningful to use hierarchical regression analyses. Factors were entered into the regression by variable group, in the order presented above. 14 The codes for care-based sentences are: 0= no, did not receive a care-based sentence, 1=yes did receive a care-based sentence. 79 The dependent variables measured the following remand outcomes: ED Remand status ED Length of custodial remand ED Bail conditions Coding of the independent and dependent variables are presented in Table 48. Table or Figure numbers from the previous chapter where variable description and distributions can be found are presented next to each variable in the table for reference purposes. TABLE 48 m DESCRIPTION OF VARIABLES USED FOR THE REMAND ANALYSIS Independent Variables Socio-demographic variables Sex Ethnicity Age Employment Status Overall Familial Situation Criminal History Variables Seriousness of criminal history Period since last criminal conviction # Bail breaches Offence Variables Offender's role Co-offenders Offence location Use of weapons Value of Propertv Victim-offender relationship Sex of victim Victim injury Court Process Variables Plea Number of Conviction Counts Dependent variables Remand status Length of custodial remand Bail conditions Description Reference O=male, 1=female O=European, 1=Non-european A continuous variable - in years O=in paid employment, 1=not in paid employment A continuous variable: O=no familial ties, 1=minimal familial ties, 2=moderate familial ties, 3=strong familial ties. Page 33 Table 1 Figure 1 Table 2 Table 6 A continuous variable (number of prior criminal convictions, number of prior criminal convictions in a similar offence category, number of prior imprisonment terms A continuous variable - in months (log transformation) 0=ves,1=no . Page 38-39 FiQure 3 Table 7 O=primary/equal, 1=secondary O=acted alone, 1=acted with others O=public location, 1=private location 0=yes,1=no A continuous variable, in dollars (I0Q transformation) O=stranger, 1=known person 0=male,1=female A continuous variable: O=no injury, 1=emotional distress, 2=physical injury Table 13 Table 12 Table 14 Table 15 Figure 5 Table 16 Table 17 Table 18 O=guilty, 1=not Quiltv A continuous variable, number of counts charged Table 20 Table 21 O=remanded in custodv, 1=Remanded on bail A continuous variable, in days (log transformation) O=special conditions, 1= standard Table 30 Figure 8 Table 31 # Note: Period since last criminal conViction related only to offenders preViously conVicted. Including thiS variable meant that a different sample was used. Separate regressions were run to allow its inclusion but all results were non-significant so will not be presented. 2.1 The Remand Status Decision The zero-order cOlTelation matrix for the Remand Status Decision (remanded in custody or remanded on bailY s with independent variables is presented in Table 49. These zeroorder cOlTelations describe the statistical strength of the relationship between each independent variable and the remand decision, before controls are introduced. The first column lS In Chapter One I explained that offenders' remand status was coded by the most serious remand outcome (at large, on bail or in custody) received throughout the trial process. However, none of the offenders received a remand at large as the most serious remand outcome, because all offenders who had been remanded at large had also been remanded on bail or in custody at some point. 80 incorporates all offence categories, while the next three are offence specific and separated into drug, violent and property offence categories. For combined offence categories, there were significant zero-order correlations between sex and remand status. In this case, the correlations represent percentage differences and showed women were 17% less likely than men to be remanded in custody. Patterns also revealed that the sex difference was offence-specific. Women were 35% less likely than men to be remanded in custody for property offences and 13% less likely to be remanded in custody for both violent and drug offences, but this was only statistically significant in the latter case. For violent offenders, overall familial situation emerged as a significant socio-demographic factor, and offenders with weak familial ties were more likely to receive a custodial remand. Young drug offenders were less likely to be remanded in custody while young property offenders were more likely to receive custodial remand. For combined offences, a serious criminal history, breaching bail, pleading guilty, and offending in a private location, all significantly increased the probability of custodial remand. The former two variables similarly impacted on remand outcomes for property offenders. Seriousness of criminal history also significantly increased custodial remand chances for drug offenders. Violent offenders who victimised strangers were significantly more likely to be remanded in custody than those who victimised persons known to them. TABLE 49- ZERO ORDER CORRELATIONS - THE REMAND STATUS DECISION Combined Offences Drug Violent Pr~ek 11=369 11=176 11=90 11=103 .17' -.04 .02 -.03 .06 .13* .03 -.19* .05 .02 .13 -.09 .17 -.08 .26* .35* -.17 .31* -.16 .14 -.18* .11 * -.19* .02 -.17 .14 -.36* .30* .00 .00 -.12* NA NA NA NA NA -.00 .07 .12 NA NA NA NA NA .10 .02 -.02 .02 NA .32* .14 .15 -.08 -.04 -.02 NA .17 NA NA NA .11' -.02 -.03 -.09 .17 -.05 .16 .12 SOCIO-DEMOGRAPHIC VARIABLES Sex Ethnicity Age Emplovment Status Overall Familial Situation CRIMINAL HISTORY VARIABLES Seriousness of criminal historv Bail breaches OFFENCE VARIABLES Offender's role Co-offenders Offence location Use of weapons Value of Propertv Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts .. * Significant at p <.05 16 The codes for remand status are: O=remanded in custody, 1=Remanded on bail 16 81 Further hierarchical regression analyses were conducted for combined and property offence categories because of the significant zero-order sex differences found (see Table 49). However, violent offences were excluded because of small sample size and non-significant sex differences at the zero-order level. Although zero-order correlations showed that women were less likely than men to be remanded in custody for drug offending, further analysis could not be conducted because of low variance in the dependent variable (refer to Table 30 in Chapter Two). After conducting the hierarchical regression analyses for combined offence categories, results found zero-order sex differences did not change after socio-demographic variables were controlled. Regardless of ethnicity, age, employment status and overall familial situation, women continued to be remanded in custody significantly less often than men (see Table 50). With all preceding variables controlled, it was also found that custodial remand was more likely for those offending in private locations and/or with serious criminal histories. TABLE 50 - THE REMAND STATUS DECISION, 17 COMBINED OFFENCES - STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOClo-DEMOGRAPHIC VARIABLES R (multiple) Sample size .19' n=369 Sex .17* -.05 Ethnicity Age .01 Employment Status -.05 Overall Familial Situation .03 CRIMINAL HISTORY VARIABLES R (multiple) .24' Sample size n=369 -.11* Seriousness of criminal history Bail breaches .06 OFFENCE VARIABLES R (multiple) .28' Sample size n=369 Offender's role -.01 Co-offenders .00 -.14* Offence location COURT PROCESS VARIABLES R(multiple) .29' Sample size n=369 Plea .06 Number of Conviction Counts -.04 .. * Slgmficant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. Once other socio-demographic variables were controlled, female property offenders also remained significantly less likely to be remanded in custody. Older offenders, those with serious offence histories and prior bail breaches were also significantly more likely to be remanded in custody (see Table 51). 17 The codes for the remand status decision are: 0= remanded in custody, 1= remanded on bail. 82 TABLE 51 ~ THE REMAND STATUS DECISION, 18 PROPERTY OFFENCES m STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SOClo-DEMOGRAPHIC VARIABLES R (multiple) Sample size .53' n=103 .3S' Sex Ethnicity -.15 Age .2S' Employment Status -.12 Overall Familial Situation -.12 CRIMINAL HISTORY VARIABLES R (multiple) .5S' Sample size n=103 Seriousness of criminal history -.13' Bail breaches .14' OFFENCE VARIABLES R (multiple) .60' Sample size n=103 Offender's role -.11 Co-offenders .04 Offence location .OS Value of Property .00 COURT PROCESS VARIABLES R (multiple) .60' Sample size n=103 Plea .04 .04 Number of Conviction Counts .. * Significant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. 2.2 Length of Custodial Remand Length of custodial remand measures the amount of days spent in incarceration prior to sentencing, and thus only applies to those offenders who were remanded into custody. Zeroorder correlations for this remand outcome are presented in Table 52. These correlations represent the statistical relationship between Length of Custodial Remand and each independent variable without controls and are the fIrst step toward establishing whether sex differences in remand exist and whether other variables might explain this. 19 The first column in Table 53 incorporates all offences, while the next three are offence specific and separated into drug, violent and property offence categories. Zero-order correlations revealed that men spent significantly longer periods in custodial remand = 53 days s.d. = 65, m women = 22, s.d. = 31), drug offences (mmen =41 days, s.d. = 51, m women = 17 days, s.d. = 22) and property offences (m men = 54 days s.d. = 70, m women = 14 days s.d. = 17). For combined offence than women for combined offence categories, (m men categories, custodial remand periods significantly increased for those with serious offence histories. For drug offenders, serious criminal histories and acting with others signifIcantly increased custodial remand lengths. Prior bail breaches significantly increased custodial codes for the remand status decision are: 0= remanded in custody, 1= remanded on bail. this sample only includes those remanded into custody, zero-order correlations between sex and the independent variables reveal similar sex differences to those for the whole sample as outlined in Chapter One. 18The 19 Although 83 remand periods for property offenders while using weapons significantly impacted on violent offenders. TABLE 52 - ZERO ORDER CORRELATIONS - LENGTH OF CUSTODIAL REMAND Combined Offences Drug Violent Properly 11=210 11=129 11=54 11=31 -.29* .03 -.05 .05 -.08 -.30* -.09 .02 -.04 .02 -.23 .12 .09 .09 -.19 -.38* -.22 -.18 .27 .07 .25* -.09 .29* -.06 .20 .15 .15 -.43* -.09 .10 -.09 -.04 .18* .01 -.09 .09 -.10 NA NA NA NA NA NA NA NA NA NA -.22 -.14 -.10 -.50* NA NA .15 -.14 -.01 .13 NA NA NA .07 .06 .05 .14 .19 .04 -.11 -.23 SOCIO-DEMOGRAPHIC VARIABLES Sex Ethnicitv Age Employment Status Overall Familial Situation CRIMINAL HISTORY VARIABLES Seriousness 01 criminal history Bail breaches OFFENCE VARIABLES Offender's role Co-offenders Offence location Use of weapons Value of Property Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts .. * Significant at p <.05 Hierarchical regressions were further conducted for combined and drug offence categories. However, further analyses were not conducted on either property or violent offence categories because of small sample sizes (see Table 52). For combined offence categories, introducing controls for other socio-demographic factors did not reduce the sex differential previously noted at the zero-order level. Women's custodial remand periods remained approximately 43 days shorter than men's (see Table 53). Once all preceding variables were controlled, it was further revealed that significantly longer custodial remand periods were given to offenders with serious offence histories, those who played a primary offending role and acted with others. 84 TABLE 53 = LENGTH OF CUSTODIAL REMAND, COMBINED OFFENCES = STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES SoclO-DEMOGRAPHIC VARIABLES R(multiple) .30' Sample size n=21 0 -.30' Sex Ethnicity .04 -.03 AQe Employment Status .. 06 Overall Familial Situation .00 CRIMINAL HISTORY VARIABLES R(multiple) .34' Sample size n=21 0 .18' Seriousness of criminal history -.00 Bail breaches OFFENCE VARIABLES R (multiple) .39' Sample size n=210 -.14' Offender's role .16' Co-offenders -.06 Offence location COURT PROCESS VARIABLES R (multiple) .41' Sample size n=210 .10 Plea .11 Number 01 Conviction Counts .. , Slgmflcant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. For drug offences, hierarchical regression results found that men continued to serve custodial remand periods approximately 26 days longer than women, even with other socio-demographic factors controlled. Acting with others or having a serious criminal history also significantly increased custodial remand periods, regardless of all other factors (see Table 54). TABLE 54 - LENGTH OF CUSTODIAL REMAND, DRUG OFFENCES - STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES Socio-DEMOGRAPHIC VARIABLES R (multiple) .33' Sample size n=129 -.32' Sex Ethnicity -.OS Age .03 -.00 Employment Status Overall Familial Situation .10 CRIMINAL HISTORY VARIABLES R (multiple) .40' Sample size n=129 Seriousness of criminal history .25' -.01 Bail breaches OFFENCE VARIABLES R (multiple) .4S' Sample size n=129 -.13 Offender's role .26' Co-offenders -.08 Offence location COURT PROCESS VARIABLES R (multiple) .51' Sample size n=129 .07 Plea Number of Conviction Counts .19 .. , Slgmflcant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included. 85 2.3 Bail Conditions The zero-order correlation matrix of bail conditions (special or standard) with independent variables appear in Table 55. Obviously, the type of bail conditions imposed by the court can only be assessed for those offenders who are remanded on bail. 20 The first column incorporates combined offence categories, while the next three are offence specific and separated into drug, violent and property offence categories. Statistically significant sex differences were found in the types of bail conditions imposed. For combined offence categories, men remanded on bail receive special conditions 11 % more often than women. Patterns also showed that this sex difference was offence specific. Male property offenders were 41 % more likely than women to have special conditions imposed on bail. For combined offence categories and drug offences, the likelihood of special conditions being imposed on bail significantly increased if criminal histories were serious or offenders acted with others. For property offences, a serious criminal history similarly increased the pro~abilty of special bail conditions. No variables reached statistical significance for violent offeIlders. TABLE 55 - ZERO ORDER CORRELATIONS - BAIL CONDITIONS 21 Combined Offences Drug Violent Property 11=310 11=154 11=63 11=93 .11' -.03 .02 .01 .06 .04 -.05 -.14 -.07 .12 -.07 .00 .08 .20 -.07 .41' .05 .16 .06 .12 -.14' -.22' -.02 -.16 -.01 -.15' -.06 NA NA NA NA NA -.01 -.24' -.04 NA NA NA NA NA .04 .01 -.09 -.02 NA -.00 .14 -.17 .02 -.02 .03 NA -.01 NA NA NA .00 -.00 -.09 -.03 .20 .13 -.02 -.05 SOCIO-DEMOGRAPHIC VARIABLES Sex Ethnicitv Age Employment Status Overall Familial Situation CRIMINAL HISTORY VARIABLES Seriousness of criminal history OFFENCE VARIABLES Offender's role Co-offenders Offence location Use of weapons Value of Property Victim-offender relationship Sex of victim Victim injury COURT PROCESS VARIABLES Plea Number of Conviction Counts .. at p <.05 , Significant Violent and drug offence categories produce non-significant zero-order sex differences and sample sizes were too small for further analyses to be conducted. Zero-order sex differences for property offences were significant but small sample size prevented further analysis at this 20Correiations between sex and the independent variables for this sub-sample of offenders revealed similar sex differences to those noted for the whole sample as previously outlined in Chapter Two. 21 The codes for bail conditions are: 0= special, 1= standard. 86 point (see Table 55). Sex differences in bail conditions for property offenders are assessed in the next chapter. Thus, further hierarchical regressions were only conducted for combined offence categories and results showed that zero-order sex differentials remained. Women were significantly less likely than men to have special bail conditions imposed even with other socio-demographic factors controlled (see Table 56). With all preceding variables controlled, having a serious criminal history also significantly increased the probability of receiving special bail conditions. TABLE 56 - BAIL CONDITIONS, COMBINED OFFENCES - STANDARDISED REGRESSION COEFFICIENTS FROM HIERARCHICAL REGRESSION ANALYSES 22 SOCia-DEMOGRAPHIC VARIABLES R(multiple) .27 Sample size n=310 .11 • Sex Ethnicity -.04 Aqe .02 Employment Status -.00 Overall Familial Situation .03 CRIMINAL HISTORY VARIABLES R (multiple) .30' Sample size n=310 Seriousness of criminal history -.11' Bail breaches .03 OFFENCE VARIABLES R(multiple) .33' Sample size n=310 Offender's role .05 Co-offenders .07 Offence location .07 COURT PROCESS VARIABLES R (multiple) .33' Sample size n=310 Plea -.04 Number of Conviction Counts -.03 .. • Significant at p <.05 Note: The Multiple R assesses the multiple correlation with all independent variables up to that point included 3 Closing Comments Otto Pollak (1950 cited in Eaton, 1986: 22) once asserted that the criminal justice system gave women preferential treatment over men because of chivalry. Pollak's claim of chivalry was later critiqued. It was argued that sex differences in offenders' criminality and lives might explain disparate treatment which may also vary across judicial decision-making points and offence categories (see Chapter One). These later comments appear, at least partially, to be supported by the results presented thus far. First, significant sex differences were found in the criminality and lives of both male and female offenders (e.g. criminal history, health) (see Chapter Two). Second, results presented in this chapter show that women only received less severe sanctions than men sometimes and for certain types of offences. Third, 22 The codes for bail conditions are: 0= special, 1= standard. 87 other major variables affecting sentencing and remand outcomes were identified and some of these factors might explain sex differences in judicial outcomes. More specifically, the statistical results presented in this chapter show that men and women had approximately equal chances of receiving suspended sentences of imprisonment, work-based sentences and care-based sentences. However, pronounced sex differences in judicial outcomes were found for the imprisonment sentencing decision, length of imprisonment term, the remand status decision, length of custodial remand and bail conditions. For combined offence categories (drug, violent, property), women were less likely to be imprisoned than men, and if sentenced to imprisonment, were given shorter terms. These outcomes were also offence specific. Women in this study were less likely to be imprisoned for property and drug offences but equally likely to be imprisoned for crimes of violence. This suggests that 'preferential' treatment is being received for a particular group of women - those who commit less 'masculine' (in this case non-violent) crime (see Nagel, Cardascia & Ross, 1982; Figueira-McDonough, 1985b). However, it is more likely that negligible sex differences in the imprisonment decision for violent offenders reflects statutory requirements which state that the Court must impose a full-time custodial sentence on violent offenders unless satisfied that there are "special circumstances" (The Criminal Justice Act 1985 s.5 [1]). Judges have more discretion when imposing imprisonment terms on violent offenders, and this is reflected by a significant zero-order correlation showing that if imprisoned, female violent offenders were given terms 15 months shorter than men. Female drug offenders were also given substantially shorter terms of imprisonment than men, but property offenders receive equal terms regardless of sex. Overall (combined offences), women were remanded in custody less often than men, once remanded in custody women further spent shorter periods of time there, and if remanded on bail they were less likely than men to be given special bail conditions. Sex differences in remand status were more pronounced for property offences. Once remanded in custody, zero-order correlations show that female drug, violent and property offenders remained there for shorter periods than men. A regression analysis was only conducted for drug offences (small sample sizes for violent and property offences prevented further investigation), and fmdings show that women continued to spend significantly shorter periods in custodial remand, regardless of other socio-demographic factors. Zero-order correlations reveal that special bail conditions were more likely for male property offenders, but were equally likely for men and women convicted 88 of violent or drug offences. Small sample size prevented further analysis of sex differences in bail conditions for property offences. This outcome along with others similarly affected by small sample size will be addressed by path analyses in the next chapter which only incorporate key, and thus far fewer, independent variables so smaller samples may be investigated. Current results show that previously identified sex differences in familial ties and employment status could not explain why women's sanctions were sometimes less severe than men's. Where sample size permitted, hierarchical regression analyses controlled for socio-demographic factors but the relationship between sex and judicial outcomes was virtually unaffected. In the few cases where zero-order correlations were all that was examined, sex was the only sociodemographic factor of any statistical significance to the judicial outcome. Thus at this stage, results do not support arguments that sex-based variation in informal social control or social cost can explain sex differences in judicial outcomes. 23 Overall, fIndings showing sex to be influential at the point of remand and sentencing were not unexpected. Previous researchers have also found that, in comparison to men, women generally receive less 'severe' sentencing and pre-trial release (remand) outcomes (e.g. see reviews conducted by Parisi, 1982; Nagel & Hagan, 1983; Daly & Bordt, 1995). It has further been noted that a number of other influential independent variables (particularly criminal, court processing, health and life histories factors) are also differentiated by sex (see for example, Frazier, Bock & Henretta, 1983; Kruttschnitt & Green, 1984; Allen, 1987a, 1987b, 1987c; Daly, 1994). Similarly, the present analysis found that less serious criminal histories, short custodial remand periods, non-custodial pre-sentence recommendations, poor health, playing a secondary offence role, negative life experiences and no prior bail breaches all reduced sanction severity in some instances. All these factors characterised women more often than men, so sex differences in judicial outcomes might in fact be mediated through these other key independent variables. That is, Judges may be making decisions based on these factors rather than on offenders' sex per se. Past research findings suggest that, once we control for these other sexbased factors, statistical evidence of gender disparity in judicial outcomes may reduce. 23 It will be recalled from Chapter One that social control theorists assert that there is an inverse relationship between formal (state) control, and informal control. Compared with men, women are often economically dependent on the state or a male breadwinner and/or have others (usually children) who are dependent on them. As a result, women experience a high degree of informal social control within the family. Social cost proponents argue that the judiciary extends preferential treatment to familied offenders because of the social cost of removing them from families (Daly, 1987b: 155, 1989b: 27). It is argued that observed sex differences in judicial outcomes, demonstrate the higher value that the criminal court places on the caring/nurturing role, usually performed by the mother, compared with the economic breadwinner role, usually performed by the father (Daly, 1987a, 1987b, Daly, 1989a, Daly, 1989b). 89 Therefore, further analysis is needed to establish how sex influences judicial outcomes. We need to know whether sex exerts a direct or indirect effect on sentencing and remand. For example, do Judges treat women more 'leniently' because they are women? Or do Judges treat women more leniently because their criminality and lives differ substantially from men's (which just happens to vary as a function of gender)? Or does later judicial treatment in fact reflect earlier experiences of sex-based differentiation by other judicial actors (e.g. probation officers, Judges who decide remand). These questions can be answered by path analysis, which is the topic of the next chapter. 90 Chapter four Path Analysis When considering sentencing and remand outcomes, results presented in the last chapter showed that men's judicial outcomes are often 'harsher' than women's. This seemingly disparate treatment could not be explained by previously identified sex differences in familial ties or employment status (see Chapter Two). However, other major variables affecting sentencing and remand outcomes, which might plausibly explain how men and women come to receive different sanctions, have also been identified (see Chapter Three). Criminality, remand status, and the content of Probation Officers' pre-sentence reports, all significantly impacted on certain judicial outcomes and each in turn differed by sex. Thus a critical question is raised namely: to what extent does sex exert direct effects on judicial decision-making, or indirect effects through its impact on mediating variables (such as criminal history, pre-sentence reports and remand status)? To put this another way, do Judges treat men and women differently simply because they are men and women, or does disparity arise solely on the basis of variables like seriousness of criminal history - which just happen to differ according to sex? Results from previous statistical studies generally show that sex-based differences in independent variables do not always provide a complete explanation for why men and women receive different judicial outcomes. In many studies, sex differences in sentencing and remand remain even when factors such as seriousness of criminal history, are statistically controlled for (see Chapter One). I predicted that some degree of sex-based variance would be found in some judicial outcomes even when other influential sex-based factors were controlled. To test this, a number of path analyses have been conducted and the results are presented in this chapter. The procedure known as path analysis tests the fit of a priori causal model, using multiple regression. In relation to the criminal justice system, the premise is that there is a "discrete ordering of events" which involves offenders moving though a number of different judicial stages (Hagan & Bumiller, 1983: 3). Similar to the hierarchical regressions discussed earlier, these plausible causal orderings make it appropriate and meaningful to use path analysis. The path models tested in this section are based on both theoretical considerations 1 and on the results of the analyses already presented. 2 1 Theoretical rationale is an important part of path analyses because patterns of causation are made explicit via the development of causal models and subsequently tested by the researcher. Pedhazur (1982: 181) notes that: "The choice of a model is, of course, not 91 To summarise, the aims of the path analysis are to establish: II The importance of sex by looking at its direct 3 and indirect 4 effects on sentencing and remand outcomes. ~ To identify other factors which might account for sex differences in sentencing and remand outcomes. II To examine the direct effects of other variables, on sentencing and remand outcomes. Variables were selected for the path analysis from previous analysis according to: a) the size and significance of correlations between independent variables and sentencing/remand at the zero-order level, b) the size and significance of regression coefficients between independent variables and sentencing/remand as indicated by regression results, and c) the sex differences in major independent variables as indicated in prior cross-tabulationslzero-order correlations. 5 This technique excludes minor or 'irrelevant' variables. 6 An important assumption of path analysis is that causality can plausibly flow in one direction only (from left to right) (see Land, 1969: 34; Pedhazur, 1982: 582). This assumption is met in the models tested and presented below. Non-significant path coefficients were excluded from these models if smaller than 13 =.10, 7 to render the diagrams more readily interpretable. arbitrary, nor is it determined by considerations regarding the analytic approach one wishes to use. A model reflects one's theory about interrelations among variables being studied, and the process by which the independent variables effect the dependent variable." 2 Recall that legal and social variables previously shown or believed to impact on sentencing and remand decisions were selected after extensive reviews of past research and literature (see Chapter One); this process carried with it implications of causality based on knowledge and theory. Subsequently, the rationale behind variable selection presented in this section began with the research design. However, the final tested models in the path analyses were developed after identifying which variables significantly correlated with judicial outcomes and whether sex differences appeared to be involved. 3 The relationships between sex and sentencing/remand once other independent variables in a path model are controlled (Klepper, Nagin, Tierney, 1983: 93). 4The proportion of the relationship between two variables which is mediated, or transmitted, by other variable/s (Pedhazur, 1982: 181). For example, sex differences in sentencing may be mediated through other factors such as seriousness of criminal history, health or ~re-sntc report recommendations. Length of imprisonment term only includes offenders who were sentenced to imprisonment so zero-order correlations between sex and other independent variables for this sentencing outcome were examined. However, similar sex differences to those noted for the whole sample were found. Similarly, length of custodial remand and bail conditions only include specific groups of offenders so zeroorder correlations between sex and other independent variables for these remand outcomes were examined but similar sex differences to those noted for the whole sample were found (see Chapter Two). 6 Past researchers have usually not used path analyses to investigate the impact of sex on judicial outcomes. Instead, step-wise multiple regression with the simultaneous inclusion of all variables has often been utilised. This technique is problematic as it includes irrelevant independent variables and the exact point where sex effects are mediated is difficult to establish. However, given this tradition I conducted additional regressions, where sample size permitted, to look at the effects of sex on each sentencing outcome once all other independent variables were controlled. In all cases, the impact of sex on sentencing outcomes was unchanged from that in the path analyses so these results are not presented. 7 The symbol (3 represents the standardised regression coefficient. The regression coefficient estimates the amount of change in the dependent variable (e.g. imprisonment terms) for one unit change in the independent variable (e.g. sex) with all other factors in the equation controlled. In other words, it indicates the unique effect of the independent variable on the dependent variable. 92 1 Section One: Sentencing Outcomes 1.1 The Imprisonment Sentencing Decision A series of regressions was run to test models in which the effects of sex on imprisonment sentencing decisions for combined, drug, and property offences were mediated through the other major variables (i.e. seriousness of criminal history) which are themselves differentiated by sex. In all cases, these models show that the direct effect of sex on imprisonment decisions (combined r =.17, drug r = .16, property r = .30) reduced to nonsignificant proportions (combined ~ = -.006, drug ~ = .002, property ~ =.14)8 after controlling for the remaining variables. The path model for combined offences is presented and discussed below. Because similar results were obtained for drug and property offences, no further discussion is provided but path models for these offence categories appear in Appendix One. The results (in Figure 11) suggest the following causal sequences: Judges were more inclined to prefer imprisonment if the person charged was a) in good health, b) had a more serious criminal history, c) had a longer custodial remand period, and d) was recommended for a prison sentence by a Probation Officer in the pre-sentence report. Thus, the reason Judges were more likely to imprison men was that men (compared to women) a) had more serious criminal histories, b) had lengthier custodial remand periods, c) had stronger recommendations for imprisonment, and d) were identified as being in better health. Interestingly, this path analysis also throws light on the origins of the pre-sentence recommendations. The results showed that such recommendations were not based on sex per se, but on variables that co-vary with sex. That is, recommendations of imprisonment were stronger if the length of custodial remand was longer and the offender had a more serious criminal history. 9 Length of custodial remand was in turn a function of criminal history seriousness and sex. Men and offenders with serious criminal histories were remanded into custody for longer periods. Why custodial remand lengths were longer for men is not explained by the model, which shows that men were remanded into custody for longer than women, even when seriousness of criminal history was controlled. A sex difference of 14% for property offenders is quite substantial - men were 14% more likely than women to be imprisoned. Although it is probable that small sample size impacted on measures of statistical significance, I can not conclude disparity in this case because the result is non-significant. 9 Discussion of offenders' health appeared in pre-sentence reports but results showed that while poor health was recorded here it had little impact on actual pre-sentencing recommendations for combined offences (~;-.03) 8 93 FIGURE 11 • PATH ANALYSIS· IMPRISONMENT SENTENCING DECISION, COMBINED OFFENCES -.24 Health -.22 -.09 -.26 Sex of Offender Seriousness of Criminal History ~ .33 1 (-.03) -.15 -.33 Length of Custodial Remand ~ Pre-Sentence Recommendations .40 Imprisonment Decision -----1j 11---1- - : : : : Multiple R = .57 (p < .001) All path coefficients are statistically significant at p <.05 unless in brackets Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal histories), Length of Custodial Remand (coded in days from zero upwards), Health (O=poor mental and physical health, 1=has either mental or physical health problems, 2=has no health problems), Pre-Sentence Recommendations (O=imprisonment, 1=other), Imprisonment Decision (O=imprisoned, 1=not imprisoned). 1.2 Length of Imprisonment Term Although the direct impact of sex on the decision to imprison was generally small and non-significant, it is still possible that sex might directly affect imprisonment terms. Aside from the crime of murder, which carries a mandatory sentence of life imprisonment, the setting of imprisonment terms is only governed by statutory maximums. In contrast, more general statutory guidance is provided to Judges with regard to the imprisonment decision. 10 Thus differential treatment according to sex may be more likely when imprisonment terms are set than when imprisonment or not is decided. Results presented in the previous chapter showed that men's imprisonment terms were longer than women's for combined, drug and violent offences, but that this may be explained by sex differences in criminality (e.g. seriousness of criminal history, offender's role), length of custodial remand, and/or the contents of Probation Officers' pre-sentence reports (e.g. For example, certain offences involving Class A drugs carry a statutory presumption of imprisonment (see Misuse of Drugs Act 1975 s. 6[4]) as do serious violent offences (see Criminal Justice Act 1985 s. 5 [1]). In contrast, section 6 of the Criminal Justice Act provides a general presumption against the use of imprisonment for property offenders. 10 ) 94 recommendations, references to negative life experiences). All these factors were found to influence imprisonment terms in some cases and to be differentiated by sex. Subsequently, a series of regressions was run to test models in which the effects of sex were mediated through these other variables. Results demonstrated that part of Judges' decision-making in relation to sentence length was based on offenders' sex alone. Once other key variables were controlled, men's outcomes were still more 'harsh' even when they appeared before the court under similar circumstances to women. For combined offence categories, the zero-order correlation between sex and imprisonment term (r = .20) actually increased in size after the other variables were controlled (see Figure 12). The results showed that women's imprisonment terms were approximately 8.5 months shorter than were men's. Similar results were found for violent and drug offences: female violent offenders received imprisonment terms twelve months shorter than men and female drug offenders' terms were around five months shorter. The path analysis for combined offences is presented and discussed below. However, no further discussion is provided for drug and violent offences because fmdings were similar to those obtained for combined offence categories. Path models for drug and violent offences appear in Appendix One. The results for combined offences are presented in Figure 12. This path model suggests the following causal sequences: Judges will increase imprisonment terms if the person being sentenced had a) a serious criminal history, b) spent a long time in custodial remand, c) received a non-custodial sentencing recommendation in the pre-sentence report and, d) was a man. Interestingly, the direct impact of sex on imprisonment terms was more substantial than any other factor in the model. This shows that being male was more detrimental (imprisonment terms increase) than having a serious criminal history, long custodial remand period andlor a custodial pre-sentencing recommendation. As is the case with the imprisonment sentencing decision, this path analysis also shed some light on how pre-sentencing recommendations were reached. Results show that serious criminal influenced histories and long custodial remand periods pre-sentencing recommendations. An imprisonment recommendation was more likely for offenders with serious criminal historiesandlor those who had spent long periods in custodial remand before sentencing. The direct effect of sex on Probation Officers' pre-sentencing recommendations was minimal (combined these characteristics. ~= -.04) because men were more likely than women to have both of Thus, the results showed that sex impacted on pre-sentence 95 recommendations indirectly because Probation Officers' decision-making was not based on sex per se, but reflected sex differences on other key variables. This path model does not explain sex differences in remand status. Men's custodial remand periods remained longer than women's, even with seriousness of criminal history controlled. FIGURE 12 - PATH ANALYSIS - LENGTH OF IMPRISONMENT TERM, COMBINED OFFENCES -.24 -.34 Sex of Offender 1 Length of Custodial Remand Seriousness of Criminal History .19 -.28 -.14 Pre-Sentence Recommendations -.37 Imprisonment Term r .19 -.23 .Multiple R = .53 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal histories), Length of Custodial Remand (coded in days from zero upwards), Pre-Sentence Recommendations (O=imprisonment, 1=other), Imprisonment Term (a continuous variable in months, log transformation). 1.3 Sentencing Outcomes - A Theoretical and Methodological Note Before proceeding further, it is important to acknowledge that including remand status (Length of Custodial Remand) and pre-sentencing recommendations in the above models may be problematic. Hagan and Bumiller (1983: 33) argue that the relationship between these two variables and sentencing are often strong enough for one to question the "methodological separateness of the independent and dependent variable." In other words, it is difficult to establish the true nature of the relationship between these factors. Strong correlations among remand status, pre-sentencing recommendations and sentencing could indicate that these variables a) have causal relationships, b) essentially represent the same decision or c) have a degree of both causality and similarity. 96 Zero-order relationships between remand status, pre-sentencing recommendations and the two sentencing outcomes (the imprisonment decision and length of imprisonment term) ranged from r = .26 to r = .5l. These relationships would not usually be considered large enough to question the "methodological separateness of the independent and dependent variable" (see Chapter Two for a discussion about multicolliniarity). Furthermore, results from the path analyses showed that custodial remand length had a unique impact on both pre-sentencing recommendations and sentencing outcomes while pre-sentencing recommendations had a substantial independent impact on sentencing. The relationship between pre-sentencing recommendations and sentencing also makes theoretical sense. Similar factors may be considered when reaching both of these decisions, but each one is also made at a different point in the judicial process and involves different judicial actors, who may use different decision-making frameworks. Probation Officers' decisionmaking tends to be grounded in a social work framework, which gives primacy to the offenders and their welfare (Hagan, 1975; Reed & Thier, 1981: 234; Shapland, 1987: 83). In contrast, Judges' decision-making must consider a diverse number of judicial aims including victims' rights, public sentiment and community safety. Probation Officers may be aware of, perhaps even influenced by, Judges' sentencing exceptions but the different frameworks used leads me to conclude that pre-sentencing recommendations and sentencing outcomes are conceptually distinct. In contrast to the pre-sentencing/sentencing relationship, sentencing and remand outcomes are more difficult to distinguish from each other because a similar decision-making framework is used to decide both outcomes. Judges are usually the judicial actors responsible for both decisions 11 and similar sets of concerns are raised when deciding if an offender should be segregated from the community during or after a criminal trial (Daly, 1989a: 148). However, it is also possible that one of these 'similar sets of concerns' could be whether the offender is male or female. If women are treated favourably at the earlier stage of remand, and if this favourable outcome then leads to favourable sentencing, then sex-based differences in the latter judicial stage are essentially being compounded by the differential treatment which occurred in the former (Kruttschnit & Green 1984: 546-547). So far, the path results indicate that remand had an independent effect on sentencing and that sex, in tum, directly impacted on remand: long custodial remand periods increase sentence 11 In fact, sometimes the same Judge may decide remand and sentencing. 97 severity and men spent more time in custody than women. However, first, the richness of social life is difficult to code completely in numerical terms. Second, the independent variables almost certainly contain some degree of measurement error. Third, in the case of remand, it is quite possible that some important influencing factors were not measured at all. 12 To test empirically whether the close connection between remand and sentencing really mattered, all path analyses were re-run, minus the variable 'length of custodial remand'. The direct effect of sex on both the decision to imprison and length of imprisonment term then increased. 13 Sex differences in imprisonment terms also increased once length of custodial remand was removed from the models. In all cases, women's imprisonment terms continued to be significantly shorter than men's.14 To appease the more sceptical, I also re-ran the path models again without the pre-sentence recommendation and direct sex differences in all sentencing outcomes lengthened even further. Excluding length of custodial remand from the path models also resulted in sex having a direct impact on pre-sentencing recommendations. In this case, women were found between 9% and 14% more likely than men to receive a non-custodial sentencing recommendation, which in turn, directly decreased the actual chances of imprisonment. In the original path models, sex did not directly affect pre-sentencing recommendations so it would seem that Probation Officers are influenced by whether or not an offender has been remanded into custody. Perhaps remand status indicates to the Probation Officer what a Judge's sentencing expectations might be. In this case, there may be little point recommending a non-custodial sentence when a long custodial remand period makes an imprisonment term almost inevitable. Increases in the direct impact of sex on both pre-sentencing and sentencing outcomes once length of custodial remand was removed from the path models suggests that ignoring the possibility of remand as a mediating variable is inadvisable. Although similar factors (e.g. seriousness of criminal history) may influence all three judicial outcomes, the evidence shows that sex-based decision-making earlier in the judicial process may also make a unique contribution to sex-based outcomes latter in the process. Remand outcomes are further investigated in Section Two, presented below. 12 I could only measure what is in the trial files and in contrast to sentencing, little information is available on how remand decisions are reached. 13 Women were found to be 2%, 4% and 15% less likely to be imprisoned for combined, drug and property offences respectively. This compares with the -.6%, .2% and 14% sex difference noted before removing remand status from the equation. 14 Women's imprisonment terms are 9 112 months shorter for combined offences (previous result = 8 112 months), 17 months shorter for violent offences (previous result = 12 months) and 5 112 months shorter for drug offences (previous results = 5 months). 98 2 Section Two: Remand Outcomes 2.1 The Remand Status Decision Regressions were conducted for combined and property offences to investigate whether the effects of sex on remand were direct, or mediated through other factors. Results in both cases were similar. Zero-order sex differences in remand status reduced, when all other variables were controlled, for combined (r = .17, ~= .14) and property offences (r = .35, ~= .24), but significant sex differences still remained, showing that offenders' sex directly impacted on the remand status decision. The path results for both combined and property offences are similar, so this section only presents the analysis for the former. The path model for property offences appears in Appendix One. The results (in Figure 13) for combined offences showed that Judges made remand decisions based on both seriousness of criminal history and offenders' sex. Offenders with serious criminal histories were more likely to be remanded into custody but even with this factor controlled, men remained more likely than women to receive a custodial remand. Interestingly, sex influenced Judges' remand decisions as much as seriousness of criminal history did. FIGURE 13 - PATH ANALYSIS - THE REMAND STATUS DECISION, COMBINED OFFENCES -.17 Sex of Offender Seriousness of Criminal History -.14 Remand Status t .14 = Multiple R .23 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Remand Status (O=remanded in custody, 1=remanded on bail). 2.2 Length Of Custodial Remand Causal models were developed to test whether the effects of sex on length of custodial remand were mediated through other variables for combined and drug offence categories. In both cases, the effects of sex on custodial remand length reduced (combined offences drug offences ~ = -.24) ~ = -.23, from that found at the zero-order level (combined offences r =-.29, drug offences r =-.30) because of sex differences in the seriousness of criminal history. In other words, part of the reason why women spent less time in custodial remand was that they had 99 less serious criminal histories than did men. Despite this, the direct impact of sex on length of custodial remand remained substantial. Regardless of sex differences in seriousness of criminal history, men were still remanded in custody for 42 days longer than women (on average) for combined offences and 26 days longer for drug offences. Furthermore, when it came to custodial remand length, simply being male could be as detrimental (longer custodial remand) as having a serious criminal history. Given similarities in findings for combined and property offences, I only present the path model for combined offences below (see Figure 14) (refer to Appendix One for property offence results). FIGURE 14 - PATH ANALYSIS - Sex of Offender -.27 LENGTH OF CUSTODIAL REMAND. COMBINED OFFENCES Seriousness of Criminal History .18 -.23 Length of Custodial Remand j Multiple R = .34 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Length of Custodial Remand (a continuous variable, in days, a log transformation) 2.3 Bail Conditions In the previous chapter, zero-order correlations and hierarchical regressions showed that for combined and property offences, having a serious criminal history and being male increased an offender's chance of receiving special bail conditions. Men's criminal histories were further found to be more serious than women's. It is thus possible that sex differences in bail conditions may be mediated through seriousness of criminal history, so path models were developed to test this proposition. When this was done for combined offences, the direct effect of sex on bail conditions reduced, but only slightly, from the zero-order level (r = .11 to /3 = .08). Although non-significant, sex was not being mediated, (at least in any substantial way), through seriousness of criminal history (see Figure 15) - men were still more likely than women to be given special bail conditions. Furthermore, male property offenders remained 40% more likely than women to be given special bail conditions, even with seriousness of criminal history controlled. Path analyses for both combined and property offences are presented below (see Figure 16). THE LIBRARY UNIVERSITY OF CANTERBURY CHRISTCHURCH, N.Z. 100 FIGURE 15 - Sex of Offender PATH ANALYSIS - BAil CONDITIONS, COMBINED OFFENCES Seriousness of Criminal History -.12 Bail Conditions Multiple R = .17 (p < .001) All path coefficients are statistically significant at p .05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Bail Conditions (O=special conditions, 1=standard conditions). FIGURE 16 - PATH ANALYSIS - BAil CONDITIONS, PROPERTY OFFENCES Sex of Offender Seriousness of Criminal History Bail Conditions .40 i Multiple R = .41 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Bail Conditions (O=special conditions, 1=standard conditions). 3 Closing Comments Judicial decision-making is a complex process involving the movement of offenders through a series of stages in which they are assessed by a number of different judicial actors. My results show that sex differences in criminality, (e.g. seriousness of criminal history), often provide a partial explanation for why women's remand and sentencing outcomes were less serious than men's. For example, part of the reason why women were less likely than men to be imprisoned was that they had less serious offence histories. In addition, sex-based disparity at the point of sentencing was affected by sex-based decision-making earlier in the judicial process. Women's custodial remand periods were shorter than men's regardless of sex differences in other key factors (e.g. seriousness of criminal history). This in turn increased women's chances of receiving a non-custodial pre-sentencing recommendation, which then mitigated the final sentence severity. Sex differences in the nature of pre-sentence reports (e.g. being more likely to identify women as being in poor overall health) further reduced the severity of final sentence for women. 101 In addition to these indirect sex effects, results from the path analysis also show that some of the sex differences found in some judicial outcomes were direct. That is, sex differences in variables previously found to be influential (i.e. seriousness of criminal history, length of custodial remand, pre-sentencing report recommendations) failed fully to explain why women's judicial outcomes were generally less severe than men's. In short, Judges appear to be treating men more 'severely' than women. Compared with the decision to imprison, Judges have more discretion when imposing imprisonment terms and this is perhaps reflected in findings which show that with other factors statistically controlled, women's imprisonment terms were substantially shorter than men's for combined, drug and violent offences. Women were also less likely than men to be remanded in custody for combined and property offences. Women's custodial remand periods were also shorter than men's for combined and drug offences. Female property offenders were less likely than men to be given special bail conditions, as were women overall (combined offences). The results produced in the current study are not surprising. Past researchers have found that judicial outcomes often differ for men and women. Sometimes (as is in the current research) this is explained by factors such as differences in men's and women's criminality, lives and treatment at earlier stages in the judicial process. At other times, men's judicial outcomes often remain more severe than women's even when sex differences in other key variables are controlled (e.g. see reviews conducted by Parisi, 1982; Nagel and Hagan; 1983; Daly & Bordt, 1995). In Chapters Six and Seven, a case-study investigation of individual crime stories, Judges' sentencing remarks and pre-sentencing reports is undertaken to confirm, explain and interpret the statistical results presented in this chapter. First, this case-study analysis will be useful because a more subtle, complex and meaningful account of the judicial processing will be obtained. Second, the case-study analysis will give us a way of confirming (or otherwise) the patterns disclosed in the statistical analysis. Third, case studies will allow us to link more strongly the statistical [mdings to societal processes (particularly gendered processes) and criminal justice processing. Before conducting the case-study analyses, however, a further statistical investigation is undertaken in Chapter Five. This [mal statistical analysis will ascertain whether the decisionmaking processes, rather than outcomes, are differentiated by sex. First, I will ask whether 102 there is statistical evidence to suggest that the criteria used for deciding judicial outcomes differ for men and women. For example, do legal variables such as offence characteristics, judicial processing variables or criminal history impact on men's judicial sanctioning more than women's? Second, are certain 'types' of men and women more likely to be extended judicial leniency? For example, it may be the case that lenient treatment is reserved for offenders who fulfil gender-role expectations. 103 Chapter Five Regression Analysis: The DecisionMaking Process Up to this point, I have investigated sentencing and remand outcomes only. I have established that in this sample, men and women often received different sentencing and remand outcomes. In the present chapter, I investigate whether the decision-makillg process, rather than the outcome, differs for men and women. Essentially, I want to know if the criteria used for deciding sentencing and remand are gendered. First, I ask whether different factors are considered when determining men's and women's judicial outcomes. Second, I ask whether certain 'types' of men and women are more likely to be extended judicial leniency. Past researchers have found that legal variables such as offence characteristics, judicial processing variables or criminal history impact on men's judicial sanctioning more than women's (e.g. Nagel, 1981; Farrington & Morris; 1983; Kruttschnitt & McCarthy, 1985; Boritch, 1992; Channels & Herzberger; 1993; Farnworth & Raymond, 1995). Furthermore, the operation of gender in judicial decision-makillg has been shown to result in different treatment for certain 'types' of men and women. In particular, age, class, ethnicity and the fulfilment of gender-role expectations (in both criminality and everyday life) have been found to impact on judicial outcomes (see Chapter One). Given these results, I expected to find evidence of sex differences in the weighting of predictor variables (particularly legal variables such as seriousness of criminal history) when sentencing and remand outcomes were decided. I also expected to find that offenders who conformed to normative ideals (e.g. being European and fulfilling genderrole expectations) would be extended judicial leniency . In data-analytic terms, investigating these expectations translates into whether interaction effects exist between sex and other variables in relation to the decisions made by a Judge. The following judicial· decisions were investigated: imprisonment sentencing decisions, length of imprisonment terms, suspended sentences of imprisonment, work-based sentences, care-based sentences, remand status decisions, length of custodial remand and bail conditions. Although prior results only showed sex differences in some of these judicial outcomes, it is still possible that the decision-makillg process differed between the sexes, because different factors could be considered for men and women when making them. 104 To investigate whether sex differences existed in the decision-making process, the sample was broken down by sex for combined and specific offence categories (drug, violent, property). Separate zero-order correlation matrices were then examined for men and women to identify sex differences in the extent to which the independent variables predicted outcomes. A multiple regression approach was used to test whether any sex differences in these correlations were statistically significant. If the interaction effect (sex multiplied by the predictor variable) explained significant variance over and above the main effects (sex and predictor variable), this indicated a significant sex difference in the correlations. For example, if health correlated +.50 with imprisonment for men and -.50 for women, the difference would probably be significant. This analysis was done twice. In the first stand-alone analysis, the test was conducted with only sex, the other predictor variable, and the interaction variable in the equation. In the second analysis, the significance level of the interaction was tested with all other independent variables also added (and thus controlled for) into the regression equation. 1 Section One: Sentencing Outcomes Significant sex differences in the weighting of predictor variables for sentencing outcomes are presented below. 1.1 The Imprisonment Sentencing Decision Results showed that only two correlations were significantly different for men and women (see Table 57). For combined offences, a serious criminal history increased the likelihood of imprisonment considerably for men, but was of little detriment to women. This difference was significant in both the stand-alone regression and when all other independent variables were included in the regression equation. In other words, with all other independent variables controlled, a serious criminal history remained more detrimental to men than it was for women. This result is consistent with international research (e.g. Nagel, 1980; Kruttschnitt & McCarthy, 1985; Channels & Herzbeger, 1993). In comparison to women, weighting criminal history so highly for men presents them as potentially dangerous and likely to re-offend. Thus, the Judge can easily justify harsher sentencing. Marked sex differences were also found for violent offending. Age was significantly more important for female violent offenders, with young women more likely to be imprisoned than older offenders. This result was also statistically significant in both the stand-alone regression and with all independent variables controlled (see Table 57). For males, the statistical 105 relationship between age and the imprisonment decision was small and non-significant. Other studies have also found judicial sanctioning to be more severe for young women than older ones. (e.g. Wattenberg & Saunders, 1954; Gibbons & Griswold, 1957; Morris, 1965; Chesney-Lind, 1973; Kratcoski, 1974; Chesney-Lind, 1977; Hancock & Hiller, 1981; Sheldon, 1981; Teilman & Landry, 1981; Mahony & Fenster, 1982; Figueira-McDonough, 1985a, Mann, 1985; Chesney-Lind, 1987; Figueira-McDonough, 1987). Some argue that the reason for this can be linked to a judicial need requiring the care and protection of juvenile female offenders for their own good. Commentators note the operation of a double standard in criminal justice processing, which consistently results in the harsher treatment of girls, and subjects them to controls which are related more to dominant gender ideals than their offence (ChesneyLind, 1973; Smart, 1976: 131-134; Chesney-Lind, 1977; Chesney-Lind, 1995: 87-92). Whether or not this is the case in New Zealand is beyond the scope of this research, which is primarily concerned with the treatment of adult women. However, the present results suggest a need for future researchers to compare the treatment of female and male juvenile offenders in this country. TABLE 57, THE iMPRISONMENT SENTENCING DECISION - SEX DIFFERENCES IN PREDICTOR VARIABLE CORRELATIONS 1 Predictor Variables Combined Offences M W Violent M W Significant Sex Differences Between Correlations Combined Offences Violent Offences Stand-alone All variables Stand-alone All variables Regression 1 controlled 2 Regression 1 controlled 2 Age -.08 .28* -.38* -.17 p<.05 Seriousness of criminal history Sample Size n= 194 n= 194 n=49 n=49 1. AnalYSIs conducted with only sex, the other predictor vanable and the interaction van able. 2. Analysis conducted with sex, the other predictor variable and all other independent variables. p<.05 p<.05 1.2 Length Of Imprisonment Term International studies have found that crimes involving female victims may be punished more harshly than those involving male victims because female victims may appear more vulnerable (see Daly, 1994: 108-110). Support for this idea is found in Table 58. In the standalone regression it was found that men who victimised women were given longer terms than those who victimised men. 2 In contrast, female violent offenders who victimised men were given longer terms than those who victimised women. Daly (1994: 110) notes that "a female victim may not be more vulnerable in comparison to a female offender." Thus a female who victimises another female may not be viewed as harshly as a male who victimises a female. The current results support the argument that men are treated harshly for victimising 'vulnerable' women but sex-based victim vulnerability is relatively unimportant in the case of The codes for the sentencing outcome and each significant independent variable are as follows: Imprisonment Sentencing Decision (O=imprisoned, 1= not imprisoned), Age (in years), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher scores representing more serious criminal histories). 1 p<.05 106 female offenders. Women were treated more harshly for victimising men who, ideologically speaking, are considered 'stronger' and not as 'vulnerable' as women. Maybe female offenders who victimise men are seen to be overstepping 'traditional' boundaries. It may be the case that women who 'stand-up' to men are seen as 'tough', more 'masculine' and subsequently in need of harsher sanctioning because they have acted outside acceptable gender bounds. Similarly, men who victimise women may overstep 'traditional' gender-role boundaries, which have ideologically situated them as the protectors of 'vulnerable' women. TABLE 58, LENGTH OF IMPRISONMENT TERM - SEX DIFFERENCES CORRELATIONS IN PREDICTOR VARIABLE 3 Predictor Variables M Violent W Significant Sex Differences Between Correlations Violent Offences Stand-alone Regression Socio-demographic variables -.32* Sex of victim .28* n= 32 Sample Size n= 36 1. AnalysIs conducted with only sex, the other predictor variable and the interaction variable p<.05 1.3 Suspended Sentences of Imprisonment Separate zero-order correlations by sex for combined and drug offence categories were examined and regressions were conducted to determine if sex differences were statistically significant. Analyses could not be conducted for property or violent offences because of range restrictions in the data (see Table 33 in Chapter Two). In the stand-alone regressions, differences between the sexes were found for combined offence categories only (see Table 59). The results showed that ethnicity was a strong predictor of sentence for men but not women: European men were much more likely than non-European men to receive a suspended imprisonment term. In the second analysis, however, the significance level for ethnicity fell below statistical significance. TABLE 59, SUSPENDED SENTENCES OF IMPRISONMENT - SEX DIFFERENCES IN PREDICTOR VARIABLE CORRELATIONS 4 Combined offences W M Significant Sex Differences Between Correlations Combined Offences Stand-alone Regression Socio-demographic variables -.05 .22* Ethnicity n= 110 Sample Size n= 76 1. AnalYSIs conducted with only sex, the other predictor variable and the interaction variable. p<.05 2 Section Two: Remand Outcomes Significant sex differences in the weighting of predictor variables for remand outcomes are presented below. A second regression controlling for all independent variables could not be conducted because of small sample sizes (n = 68). The codes for the sentencing outcome and each significant independent variable are as follows: Length of Imprisonment Term (in months), Sex of Victim (0= male, 1=female). 4 The codes for the sentencing outcome and each significant independent variable are as follows: Suspended Sentences of Imprisonment (O=yes, received a suspended sentence of imprisonment, 1=no, did not receive a suspended sentence of imprisonment), Ethnicity (O=European, 1=other). 2 3 107 2.1 The Remand Status Decision It has been proposed that when offending women's lives conform to notions of 'ideal femininity' the judiciary may treat them less severely. In particular, it has been suggested that domesticity may mitigate sanction severity for women (see Chapter One). Whether this is because Judges are reinforcing gender-role expectations, recognising that domesticity contains a degree of informal social control, or are concerned about the social costs of removing care givers from families, is a subject of debate (see Chapter One). Results in the current study showed that for combined and violent offences, having strong familial ties substantially decreased the possibility of custodial remand for women, but not for men. If informal social control and social cost are considered when making remand decisions, these apparently relate only to women because familied men receive no judicial leniency. The statistical interaction between sex and familial situation was significant in stand-alone regressions. Once all other independent variables were included, statistical significance was not maintained for combined offences. For violent offences, sex differences in familial situation could not be tested using the second regression equation because of small sample size (n = 90) (see Table 60). In addition to familial situation, ethnicity was also found to predict remand status for female violent offenders but not for male. It was shown that non-European women were substantially more likely to be remanded in custody than European women. This result was statistically significant in the stand-alone regression, but again small sample sizes prevented further analysis (see Table 60). Many researchers have noted that 'lenient' treatment is only reserved for 'certain kinds of women'. In the United States, some studies have shown that "African American women and other women of color tend to receive more severe responses by the system than Anglo women" (Belknap, 1996: 73). The current result suggests that this may also apply, in particular cases, to ethnic minority women in New Zealand. TABLE 60 - THE REMAND STATUS DECISION - CORRELATIONS Predictor Variables SEX DIFFERENCES IN PREDICTOR VARIABLE 5 Combined offences M W M Violent W Significant Sex Differences Between Correlations Combined Offences Violent Offences Stand-alone Regression Stand-alone Regression p<.05 p<.05 p<.05 Socio-demographic variables -.30* .13 Ethnicity .04 .45* Familial Situation -.07 .14* n= 184 n=45 n= 45 Sample Size n= 185 1. AnalYSIs conducted with only sex, the other predictor vanable and the interaction vanable codes for the remand outcome and each significant independent variable are as follows: The Remand Status Decision (O=remanded in custody, 1=remanded on bail), Ethnicity (O=European, 1=other), Overall Familial Situation (O=no familial lies, 1=minimal familial lies, 2=modrale familial lies, 3=slrong familial lies). 5 The 108 2.2 Length Of Custodial Remand Only a small number of property offenders (n=31) were remanded into custody so analyses were only conducted for combined, drug and violent offence categories. Results for the stand-alone regressions revealed several sex differences in predictor variables for violent offences (see Table 61). Causing serious injury and using weapons substantially increased custodial remand periods for men, but not for women. For combined offence categories, a not guilty plea increased custodial remand periods for men, but not women. These results are consistent with research findings that show offence characteristics and judicial processing variables impact on men's judicial sanctioning more strongly than women's (e. g. Nagel, 1981; Kruttschnitt & McCarthy. 1985; Channels & Herzberger, 1993). In comparison to women, judicial focus on the characteristics of men's crimes and pleas aggravates their dangerousness and general lack of remorse. This in turn can be used to legitimate long custodial remand periods. Offence location is the one crime characteristic found to predict long custodial remand periods for female offenders. Unlike men, committing a drug offence in a public setting increases women's custodial remand periods. This finding is possibly linked to gender-role expectations, which situate women in the private rather than public sphere. Women who commit offences in public deviate from their traditional and appropriate place within the home and harsher treatment may result because the female offender is punished not only for the crime but also for "exceeding the bounds of gender appropriate behavior" (Bishop & Frazier, 1984: 386). Second regressions were conducted for combined and drug offence categories. The additional regression was not conducted for violent offences because of small sample size (n = 54). Sex differences remained statistically significant after all independent variables were included in the second regression equation, for combined offences only. 109 TABLE 61, LENGTH OF CUSTODIAL REMAND CORRELATIONS Predictor Variables p SEX DIFFERENCES IN PREDICTOR VARIABLE 6 Combined Offences M W Violent Drug W M Significant Sex Differences Between Correlations Combined Offences W M Stand-alone Regression 1 All variables controlled 2 Offence location .17 -.22' -.73' -.16 Use of weapons .39' -.18 Victim Injury -.07 Plea p<.05 .18' !F60 !F30 !F89 !F69 !F24 Sample size !F121 1. AnalysIs conducted with only sex, the other predictor variable and the interaction variable. 2. Analysis conducted with sex, the other predictor variable and all other independent variables. Drug Offences Stand-alone Regression 1 p<.05 Violent Offences Stand-alone Regression 1 p<.05 p<.05 p<.05 2.3 Bail Conditions Where bail conditions were concerned, no significant sex differences were found in the weighting of predictor variables for combined offence categories but statistically significant sex differences were found within specific offence categories. Separate zero-order correlations by sex for drug, violent and property offence categories are subsequently outlined below along with the regression results (see Table 62). Stand alone regression results again showed that certain offence characteristics were weighted differently for men and women. With property offences, playing a primary role in the offence increased men's chances of getting special bail conditions but not for women's. For violent offences, weapon use was a strong predictor of special bail conditions for men but not for women. Acting with others significantly increased the probability of special bail conditions for male drug offenders but had no impact on women. Regressions including all other variables could not be conducted for violent or property offences because of small sample sizes and sex differences did not maintain statistical significance for drug offences. TABLE 62, BAIL CONDITIONS - SEX DIFFERENCES IN PREDICTOR VARIABLE CORRELATIONS Drug Predictor Variables M Offender's role Co-offenders Use of weapons -.31' Sample Size n=70 W Violent M W Property M W .24* -.28' I -.19 n= 84 I .39' n= 25 -.23 n- 38 7 Significant Sex Differences Correlations Drug Violent Offences Offences Stand-alone Stand-alone Regression 1 Regression 1 Between Property Offences Stand-alone Regression 1 p<.05 p<.05 p<.05 n-47 n= 46 I 1. AnalYSIS conducted With only sex, the other predictor variable and the interaction variable. 2. Analysis conducted with sex, the other predictor variable and all other independent variables. 6 The codes for the remand outcome and each significant independent variable are as follows: Length of Custodial Remand (in days), Offender's Role (O=primary/equal, 1=secondary) , Offence Location (O=public, 1=private), Use of Weapons (O=yes, 1=no), Victim Injury (O=no injury, 1=emotional distress, 2=physical injury), Plea (O=guility, 1=not guilty). 7 The codes for the remand outcome and each significant independent variable are as follows: Bail Conditions (O=special conditions, 1=standard conditions),Offender's Role (O=primary/equal, 1=secondary) , Co-offenders (O=acted alone, 1=acted with others), Use of Weapons (O=yes, 1=no). 110 3 Closing Comments In Chapters Three and Four I investigated sentencing and remand outcomes and it was established that these often differed for men and women. In the current chapter, I considered whether there were sex differences in the decision-making process rather than the outcomes. In other words, I wanted to investigate whether the criteria used when deciding remand and sentencing were gendered. First, I asked if different factors were considered when determining men's and women's judicial outcomes. Second, were certain 'types' of men and women more likely to be extended judicial leniency? I expected to fmd sex differences in the weighting of predictor variables (particularly legal variables such as seriousness of criminal history) when sentencing and remand outcomes were decided. I also predicted that offenders who conformed to normative ideals (e.g. fulfilled gender-role expectations - both criminally and in everyday life) would be more likely to receive judicial leniency than those who did not. The results presented in the current chapter generally support these predictions. The tendency of the criminal justice system to focus on men's legal, rather than social situations, increased the likelihood of receiving punitive sanctions. Gendered decision-making also ensured that certain 'types' of men and women were extended judicial leniency . Results show that different factors were sometimes considered when determining men's and women's judicial outcomes. Support is provided for the argument (of past researchers) that women whose crimes or lives conflict with dominant gender ideals are less likely to be extended judicial leniency (e.g. Nagel, Cardascia & Ross, 1982; Figueira-McDonough, 1985; Edwards, 1986: 80). For example, committing a drug offence in a public, as opposed to a private setting, was shown to increase the length of time that women in this study spent in custodial remand, while victimising men further extended sentences of imprisonment for female violent offenders. Aside from these two factors, legal variables generally affected men more than women, whose judicial outcomes were more likely to be aggravated by social factors. 8 Results show that only certain 'types' of women were extended judicial leniency. In comparison to their domesticated, older, European counterparts, younger ethnic minority women with weak familial ties received more severe judicial sanctions. As noted, legal variables were generally more likely to impact on men's judicial outcomes. Serious criminal histories, causing injury to victims, victimising women, using weapons, playing an active role. in the offence, having co-offenders and pleading not guilty, all substantially affected men's judicial outcomes in some cases. Compared with women, men 8 One exception was found in the case of suspended sentences of imprisonment where European men were more likely to have 111 characterised by these factors received harsh judicial outcomes because culpability, dangerousness and blameworthiness were aggravated. This result is consistent with previous research findings (e.g. Nagel, 1981; Farrington & Morris, 1983; Kruttschnitt & McCarthy, 1985; Boritch, 1992; Channels & Herzbeger, 1993; Farnworth & Raymond, 1995). What emerges from the evidence thus far is that gender affects sentencing and remand, but further investigation is still required. A case-study analysis is undertaken in the following chapters to confIrm, explain and interpret these results. imprisonment terms suspended than non-European men. 112 Chapter Six Offenders' Crime Stories This section investigates the crime stories of 100 individual offenders (50 pairs) who were selected from the previous sample used in the statistical analyses (n = 388).1 Pairs sentenced for the same statutory offence, in the same court (District or High) with similar offence characteristics, 2 criminal histories, 3 pleas and biographical details (age and ethnicity) were examined and the best matched pairs were then selected for inclusion in the case-study sample. In this process, all pairs selected were sentenced for the same statutory offence, in the same court after which matching priority was given to offence characteristics, criminal history, and plea. To enable generalisations from the case-study sample to the larger original sample, statutory offence and court distributions similar to those in the staic~ sample were sought. These are outlined before the case-study results are discussed. Primarily, the crime stories were taken from the following documents: police summanes of facts, charge sheets and Court Registers. 4 In addition, specific sections of Judges' sentencing remarks outlining crime details were also used. 5 1 Characteristics of the Case-study Sample Sixty percent of offenders were sentenced in the District Court and 40% in the High Court. This is roughly proportional to the larger sample used for the statistical analyses in Chapters 2 to 5 where approximately 54% of offenders were sentenced in the District Court and 46% in the High Court (see Table 48 in Chapter Two). In the case-study sample, 42% of offenders were sentenced for a drug conviction, 30% for a property conviction, and 28 % for a violence conviction. This distribution is similar to the statistical sample, where 47% of offenders were drug offenders, 28% were property offenders and 25% were violent offenders (see Table 8 in Chapter Two). lTime constraints made it impossible to look at all 388 cases in detail. Including - role, number of co-offenders, number of counts convicted, use of weapons, victim offender relationship, sex of victim, injury to victim and value of property involved. 3 Including - number of prior criminal convictions, number of prior convictions in a similar offence category, number of prior imprisonment terms. 4 The Court Registers are: The High Court's "Return of Prisoners Tried and Sentenced," and the District Court's "Return of Persons Committed for Trial and Sentence." These registers are paper files that have been collected and bound by the final hearing or sentence date. The following information is contained in the registers: Name of offender, Sex of offender, offender's country of birth, ethnic origin of offender, offences committed (including Act and Section), committal date, court of committal, plea, date of plea, verdict, date of verdict, result of hearing/sentence date, trial/sentence number. 5 Offender's names, place names, times, dates and other trivial information has been changed to protect the identity of those involved. 2 113 1.1 Drug Offences In the statistical sample, the majority of drug offenders were sentenced for Class 2 drug offences, followed by Class 1, Class 4, Class 5, Class 3, and finally Class 6 offences. In the narrow case-study sample, similar proportions were achieved (see Table 63). TABLE 63 - Class 1 Drug Class 2 Drug Class 3 Drug Class 4 Drug Class 5 Drug Class 6 Drug Total SERIOUSNESS OF DRUG OFFENCE Offence Offence Offence Offence Offence Offence Case-Study Sam Ie n % 10 23 42 18 2 4 23 10 2 4 2 4 100 42 Statistical Sample n 32 110 4 26 8 2 182 % 18 60 2 14 4 1 100 Class 1 Drug Offence = ImporVexport, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class A drug (Misuse of Drugs Act, s6 2a). Class 2 Drug Offence = ImporVexport, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class B drug (Misuse of Drugs Act, s6 2b). Class 3 Drug Offence = Conspire to commit an offence against s6 (1) of the Misuse of Drugs Act a Class B drug (Misuse of Drugs Act, s6 2Ab). Class 4 Drug Offence = ImporVexport, sell, give, supply, administer, deal, possess for supply, produce, manufacture, distribute a Class C drug (Misuse of Drugs Act, s6 2c). Class 5 Drug Offence = Permit Premises to be used in the commission of a Class B drug offence (Misuse of Drugs Act, s12 2b) Class 6 Drug Offence = Permit premises to be used in the commission of a Class C drug offence (Misuse of Drugs Act, s12 2c). 1.2 Violent Offences Although murder IS recognised as a gendered crime in both New Zealand and internationally (see Jones, 1980; Browne, 1987; Jurik & Winn, 1990; Rapaport, 1991; Dobash, Dobash, Wilson & Daly, 1992; Benekos, 1995; Follingstad, Brondino & Kleinfelter, 1996; Roberts, 1996; Gauthier & Bankston, 1997; McDonald, 1997; Pratt & Deosaransingh, 1997; Bungay, 1998) which mayor may not warrant differential treatment by the courts, it could not be included in the case-study sample (see Table 64). New Zealand has a small population and the in-depth coverage given to murder cases by the media (especially those committed by women) meant that the anonymity of offenders could not be assured. All other classes of violent crime were included. The largest number of individuals sentenced for violence in the statistical sample were Class 2 offenders. The majority of violent offenders in the case-study sample were also Class 2 offenders. For the remaining classes, as many pairs as possible were retained in the case-study sample. Only two offenders were convicted of a Class 3 offence and both were included in the case-study sample. In the statistical sample, ten offenders were identified as committing a Class 4 offence. Three Class 4 pairs were originally identified for inclusion but files for two pairs were missing when I returned to the Court after completing my statistical analyses. Subsequently, only one pair of Class 4 offenders could be included. To maintain numbers proportional to the statistical sample, an extra pair of Class 5 offenders was selected to make up 14% of the case-study sample. All Class 6 offenders from the statistical 114 sample were included but only one pair of Class 7 offenders was considered, again because of a missing file. TABLE 64 - SERIOUSNESS OF VIOLENT OFFENCE Case-Study Saml Ie Class Class Class Class Class Class Class Total 1 Violent 2 Violent 3 Violent 4 Violent 5 Violent 6 Violent 7 Violent Offence Offence Offence Offence Offence Offence Offence Statistical Sample n % n % 0 16 2 2 4 2 2 28 0 57 7 7 14 7 7 100 6 66 2 10 8 2 4 98 6 68 2 10 8 2 4 100 Class 1 violent offence = Murder (Crimes Act, s 172) Class 2 violent offences = Wounding with intent to cause grievous bodily harm (Crimes Act, s 188 (1)), kidnapping (Crimes Act, s 209), aggravated robbery (Crimes Act, s 235), aggravated burglary (Crimes Act, s 240A) , extortion by certain threats (Crimes Act, s 236). Class 3 violent offences = Robbery (Crimes Act, s 234) Class 4 violent offences = Accessory after the fact to murder (Crimes Act, s176), conspiracy to commit aggravated robbery (Crimes Act, s 235 & s31O), wounds with intent to injure (Crimes Act, s 188 (2)), Class 5 violent offences = Cruelty to a child (Crimes Act, s 195), Assault with weapon (Crimes Act, s202C). Class 6 violent offences = Assault on a child (Crimes Act, s 194), Possession of offensive weapons (Crimes Act, s 202A). Class 7 violent offences = Common Assault (Crimes Act, s196). 1.3 Property Offences Similar to the statistical sample, the majority of property offenders in the case-study sample had been convicted of a Class 3 property offence. The only property offenders to be sentenced in the High Court were Class 1 property offenders. In order to achieve proportional numbers by court of sentencing, a relatively large number of Class 1 offenders were selected. Of the original 16 Class 2 property offenders in the statistical sample, two were selected in the current case. In the statistical sample, two offenders were convicted of a Class 4 property offence but both had to be excluded from the case-study sample because the court files contained minimal information about them. This reflects the minor nature of the crime, which only carries a maximum penalty of three months imprisonment 6 (see Table 65). TABLE 65 - SERIOUSNESS OF PROPERTY OFFENDING Case-Study sample n Class Class Class Class Total 1 2 3 4 Property Property Property Property Offence Offence Offence Offence 8 2 20 0 30 Statistical Sample n % 27 7 66 0 100 14 16 76 2 108 % 13 15 70 2 100 = = Class 1 Arson (Crimes Act, s 294) Class 2 Burglary (Crimes Act, s 241) Class 3 = Theft (Crimes Act, s 222, s 223, s 227), Fraud (crimes act, s 246 1, s 246 2(a), s 229A) and Receiving Stolen Property (Crimes Act, s 258 (a)). Class 4 = Fraud (Crimes Act, s 246 (C)). 6 Minor fraud which involves obtaining property not exceeding $100 by false pretence (see Crimes Act, s 246 (C)). 115 2 Crime Stories The statistical design used in previous chapters to investigate whether there were sex differences in judicial outcomes included a wide range of factors, making it a fairly rigorous statistical investigation. The data presented in Chapters Three and Four show that sex differences in sentencing and remand outcomes existed in some cases. However, coding schemes used in statistical research designs have been criticised as limited in their ability to capture the full character of each individual case (Conley & O'Barr, 1987; Crew, 1991; Triggs, 1999: 123). Daly (1994: 264-265) argues that statistical analysis, and the logicoscientific reasoning driving it, should be used alongside narrative analysis or modes of reasoning because; "The narrative pole offers depth and meaning. For sociolegal research, it gives us the full texture of the case. We may try and quantify elements in the narrative, but the uniqueness of the crime story ... may resist being captured by the uniform scheme. It is not simply a matter of technical incompetence, for one could code and quantify many details. Rather it is matter of how the story is larger and more meaningful than the sum of its parts." In response to Daly's (1994) concerns, I examined the crime details or stories for pairs of offenders (i.e. male and female) to determine whether judicial outcomes for like crimes or crimes of similar seriousness, were similar. Before this could be achieved, it was necessary to establish criteria to judged the 'seriousness' of crimes. Crime senousness IS not a completely objective value. Crime seriousness IS (in part) a subjective perception, which varies according to time, place and context 7 (Rossi, Waite, Bose, Berk, 1974; Boritch, 1992; Daly 1994: 89-91). Dominant ideas about crime seriousness within a given society, during a particular time, are usually reflected in law. New Zealand's criminal justice legislation lays down general remand and sentencing principles 8 including the prescription of maximum sentences and outlining who is bailable as of right. 9 Apart from this, legislation provides only minimal guidance10 as to what factors should govern remand and sentencing decisions (Hodge, 1991: 71; Hesketh & Young, 1994: 37).11 For example, some offences carry a statutory presumption of imprisonment, while a non-custodial sentence is presumed for other offences, unless there are special circumstance (Hall, 1994: B/3-B/7). But 7 For example, in New Zealand public and political outrage at certain times has led to 'getting tough' on domestic violence, drunk driving and more recently home invasion. 8 See Crimes Act [1960J, Misuse of Drugs Act [1975]) Criminal Justice Act [1985J 9 Section 319 of the Crimes Act 1961 outlines the conditions under which a person is either bailable as of right, or bailable at the discretion of the court. However, everyone in this study sample was bailable at the discretion of the court so this has no relevance. 10 In New Zealand a Sentencing Digest is also available to Judges, lawyers and other legal researchers. This document is not a set of guidelines as such, but is based on sentencing precedent and provides summaries of sentencing judgements; the majority of which originate from the Court of Appeal (Sentencing Digest, 1999: 1). Given that the present analysis is trying to ascertain if like crimes are treated alike, and that the individuals in each pair were sentenced in the same year, it is presumed that any precedent set is applicable to both offenders. 11 This approach is viewed positively because, in contrast, rigid statutory regulations could restrict discretion and result in unfairness from treating alike cases which are unalike (Ashworth, 1995: 32). New Zealand's legislation is thus seen to ensure each person before the courts is given individualised consideration (Ministry of Justice, 1997: 27). 116 this guidance is only provided for some offences, and what constitutes 'special circumstances' is not altogether clear (Hesketh & Young, 1994: 43).12 Statute also requires that in some cases, a High Court Judge must make remand decisions and that custodial remand is preferable in cases of serious violence 13 but once again, there is a large degree of judicial discretion. 14 As the case studies are outlined, offence specific statute which provides sentencing and remand guidance, where available, is noted. Crime seriousness is thus difficult to define using statutory definitions alone, but it is generally acknowledged that the most serious crimes are seen to cause major harm and are performed with a high degree of culpability, while the least serious crimes cause minor harm and are preformed with a low degree of culpability (Husak, 1998: 192). In other words, the "seriousness of an offence may be analysed in terms of harmfulness or potential harmfulness of the conduct, and the culpability of the offender" (Ashworth, 1995: 134). Factors routinely identified as implying a more serious or harmful quality to a criminal act and those impacting on offender culpability are presented below. In the case-study analyses, these factors were used to measure the overall seriousness of offenders' criminality. e Offences causing high levels of distress to victims: The damaging and distressing effects on crime victims is an important factor in judicial decision-making and is used to assess crime seriousness (Daly, 1994: 89; Hall, 1994: B/122; Wasik, 1998: 104). The physical injury suffered, and the extent to which it is permanent or temporary, will affect the seriousness of a violent offence. Psychological or emotional consequences and the extent to which they are lasting or transitory are also relevant to property offenders (Hesketh & Young, 1994: 47-48). CI Offences where victims are unknown to the offender: Research indicates that offenders who victimise strangers, rather than known persons, are sentenced more harshly, particularly in the case of violent offending (Daly, 1994: 89; Simon, 1996: 104). Thus, crime seriousness is aggravated by victimising a stranger and could potentially impact on remand decision-making. 12 Hesketh and Young, (1994: 43), writing in a New Zealand context, argue that it is "impossible to predict with any certainty what the Courts will find to be a special circumstance." 13 The following violent offences are identified as serious in the Crimes Act 1961 s. 318 (only those relevant to the current research are outlined): wounding with intent to cause grievous bodily harm, robbery and aggravated robbery. 14 Hodge (1991: 71) states that "other than stating that the court has discretion to grant or refuse bail, s. 319 of the Crimes Act 1961 is silent as to when and how that discretion ought to be exercised." 117 <111 Offences involving victims perceived to be vulnerable: Crimes committed against individuals are generally considered more serious than those committed against corporations or organisations. First, crimes involving the intrusion of victims' living space and invasion of privacy may be deemed more serious than the invasion of commercial establishments (von Hirsch, 1993:31). Second, commercial victims may appear less vulnerable in their ability to "bear the loss" because "the loss is not (usually) suffered by a single individual and is more likely to be insured" (Wasik, 1998: 116.) In general, offending against individuals who are defmed as 'vulnerable' increases crime seriousness. 'Vulnerable victims' are defined as less "able to defend themselves through one or another form of physical or psychological limitation," and the impact of the crime on them is seen to have a "deeper and more prolonged effect than it would on a non-vulnerable victim" (Wasik, 1998: 112). There is a widelyheld view that it is worse to take advantage of people who are particularly 'helpless', so offences against the elderly, the very young, disabled or women 15 may be judged as particularly serious (Daly, 1994: 89-90; Hall, 1994:B/122; Hesketh & Young, 1994: 49; Ashworth, 1995: 130; Wasik, 1998: 115). 49 Offences that are planned or premeditated: Intent, motive and circumstance determine the extent of an offender's culpability and blameworthiness. Deliberately planned, premeditated criminal activity which demonstrates a degree of professionalism is usually considered more serious than criminality which occurs impulsively on the 'spur of the moment' or in the 'heat of passion' (Hall, 1994: B/115-B/116; Hesketh & Young, 1994: 49-50). Ashworth (1995: 131) explains that: "Planned lawbreaking constitutes a great threat to society, since it betokens a considered attack on social values, with greater commitment and perhaps continuity than a spontaneous crime." 49 Offences where a position of power or trust has been misused: Crimes that involve a breach of trust or abuse of power will often aggravate crime seriousness (Hall, 1994: B/137 - B/140; Hesketh & Young, 1994: 48). Breaches of trust or authority are viewed as additionally harmful. Trust is an important part of many It is noted by Daly (1994: 110) that defining crimes involving female victims as more serious may not hold when the offender is also female because "a female victim may not be more vulnerable in comparison to a female offender." 15 118 social relationships and networks of authority and also rnsures the cohesive operation of society (Ashworth, 1995: 131). o Offences where there was a long history of law breaking, particularly law breaking of similar nature to the current offence: Culpability is mitigated by a lack of previous convictions because it illustrates that an offender is generally of 'good character', that the current offence is thus 'out of character' and the offender may not have been fully aware of their actions. In these cases, offenders may be extended leniency. In contrast, those with extensive criminal histories may be viewed by the judiciary as remorseless with little respect for law and more culpable (Hesketh & Young, 1994: 52). The number of prior convictions an offender has is not the only component of criminal history. When making sentencing decisions courts will also "weigh and interpret" criminal records according to the type and seriousness of prior convictions alongside the time between present and past convictions. A conviction-free period may reduce the effect of previous convictions on judicial outcomes by reducing overall criminality. On the other hand, a history of similar offending, especially if it is of a more serious nature, will increase culpability (Hesketh & Young, 1994: 53; Ashworth, 1995: 162-164). • Offenders playing a minor role or being seen to be following orders: When assessing appropriate judicial outcomes, offender culpability can be determined by degrees of offence participation. Offenders seen to be 'under the influence' of a cooffender or 'coerced into' committing a crime may be held less culpable than those seen to be acting under their own initiative. distinguishes between leaders and followers. In other words, the judiciary This perception in tum may be influenced by gender (Hall, 1994: B1200-B1202). In a mixed sex group of offenders, dominant ideological views of women as "fragile, naive, weak-willed and compliant, and often as the witless thralls of men" (Newbold, 2000: 69) will increase the possibility of them being perceived as the followers while men will be seen as the leaders. o Group Action: Particularly where violent crime is concerned, offences committed in groups may aggravate crime seriousness simply because a group of offenders is perceptually more 'scary' than an individual offender (Ashworth, 1995: 129-130). Furthermore, the gender composition of such groups is also noted to have possible 119 effects in that men working in all male groups might be perceived as more dangerous than mixed-gendered groups (Daly, 1994: 95). The latter argument regarding gender is problematic, and while noted in the casestudy analyses, is not, in its own right, considered a legitimate reason for judicial disparity. It is not, therefore, used here to assess criminal seriousness. Taken to its extreme, this line of reasoning could be used to legitimise the harsher judicial sanctioning of all criminal men over all criminal women, simply because we fear men more than women, whether 'rationally' or not. Most crimes are unimaginable without the presence of men, so to a large extent the well-documented 'fear of crime' is, in effect, a fear of men (Collier, 1998:2). Objection to this line of reasoning is best illustrated by the shock that would result from suggesting groups of Maori or Pacific Island offenders should be sanctioned more severely than groups of Europeans simply because the fear of crime is equated with certain ethnic or racial characteristics. @ Victim Provocation: In some cases, the victim/offender boundary can be blurred because a victim may provoke an offender's behaviour by "waylaying, assaulting, or taunting them" (Wasik, 1998: 118). In these circumstances: "80th are substantially at fault in their behaviour and, at the end of the exchange between them it is largely a matter of chance which of the participants tums out to be the victim, and which is the offender. Or, if not actually initiating with the exchange, the victim may have been perfectly willing to trade insults with the offender, and may have chosen to stand their ground, a meet force with force, rather than taking the chance to flee or retreat. Or the victim may have disengaged initially, perhaps even left the scene, but then returned to pursue the matter" (Wasik, 1998: 118). In cases such as these an offender's culpability may be reduced because the victim's behaviour has contributed to the offence in some way (Hall, 1994: B/129). @ Cooperating with justice authorities: Admitting an offence before it is discovered, co-operating with the police, and/or entering a plea of guilty demonstrates to the judiciary that an offender is remorseful, owning up, or taking responsibility for their actions which in tum mitigates cUlpability. Furthermore, a guilty plea saves the state time and expense of a contested hearing while also sparing victims and witnesses the ordeal of attending court (Hall, 1994: B/222-B/232; Hesketh & Young, 1994: 5456; Ashworth, 1995: 136-140). 120 (II Restitution to victims: Another source of mitigation derives from an offender's offer to pay compensation or to make other amends to the victim. This is particularly powerful when an offender has compensated victims before case outcomes are finalised in court, or made arrangements with the victim to do so (Hesketh & Young, 1994: 54; Ashworth, 1995: 141). The current analysis utilises a pair-wise (man vs woman) comparison of criminality, using the criteria above,16 and judicial outcomes for each offence category (drug, violent and property). Daly (1994: 236-254) originally designed this method and cases were grouped according to whether judicial responses were similar, different or disparate. A similar judicial response was highlighted when offenders' criminality was similarly serious and they received similar sentencing and remand outcomes. A different judicial response was highlighted when offenders' criminality differed and they received different outcomes. A disparate response was highlighted when it was not immediately plain why two people received different punishments for criminality of comparable seriousness, or the same punishment for criminality of different seriousness. The similar, different and disparate framework used at this point is based on equal treatment assumptions; that is, does like criminality result in like judicial outcomes? This assessment is based on concepts of proportionality and offence-seriousness with an assessment of offenders' criminality being undertaken. However, other personal 17 or social circumstances 18 may justify lenient judicial outcomes and positive forms of judicial discrimination (including power differences between men and women) (Daly, 1994: 237; Hesketh & Young, 1994: 59-64). These factors are considered when pre-sentencing reports and Judges' sentencing remarks are investigated (in the next chapter), at which point the analysis will shift from focussing primarily on judicial outcomes to the criteria which influence the legal processing of men and women. 19 In this section, the case-study results are presented by offence category (drug, violent and property) and categorised by judicial response (disparate, different, similar). Judicial 18 New Zealand's Ministry of Justice (1995) has also devised a Seriousness of Offence Scale which assigns scores to imprisonable offences according to the "average number of days imprisonment imposed on every offender convicted of that offence between 1990 and 1994, where the average is taken over both imprisoned and non-imprisoned offenders" (Spier, 1996: 129). In the current analysis, this scale generally added little to the criteria already outlined and was not very useful because the offenders were convicted of the same statutory offence and matched closely on many of the crime characteristics noted in the scale. For example, the seriousness score for aggravated robbery using a weapon (988) is higher than that for an aggravated robbery done manually (846) but offender pairs were usually match according to weapon use so this difference in seriousness was already accounted for. Nevertheless, I checked all offences in my case-study sample against this seriousness of offence scale and will note outcomes when they further aided my analysis. 17 For example, the good or reasonably good character of an offender (Samuels, 1987: 67). 18 For example, the impact of a particular outcome on an offenders' family, health and rehabilitative prospects. 19 At this point the qualitative analyses will also complement the statistical results presented in Chapter 5. 121 responses are summarised in table form and examples of disparate responses presented. 21 20 are then Case studies are introduced by noting how I coded them for previous statistical analysis (see Chapters, 2, 3, 4, 5). This is followed by a detailed description of current offence seriousness, criminal history seriousness and how, overall, this was used to classify judicial responses as disparate. As already noted, statistical coding schedules have been criticised for being limited in their ability to capture the "uniqueness of the crime story" (Daly, 1994: 265). Thus, the purpose of this chapter is to confirm (or otherwise) the patterns disclosed in my prior statistical analysis. Before proceeding further, it is important to acknowledge that a qualitative analysis of individual crime stories may also be problematic. Without intending to enter the complex debate over qualitative vs quantitative methods, I wish to state that I do not consider the casestudy analysis to be somehow 'better than' my statistical analysis. It could be argued that the case-study analysis presented in this chapter merely repeats, in a more 'subjective' way, the statistical analysis because many of the variables used previously (e.g. criminal history seriousness) are again utilised to assess judicial outcomes. The case study investigations also took place after the statistical analysis had been conducted and I was the only person to assess the data. 22 This raises questions about 'objectivity' and researcher bias. I do not intend to enter into debate about positivistic vs post-modern notions of 'objectivity'. However, as the research methods I have chosen demonstrate, I recognise both sides of the qualitative/quantitative argument. I acknowledge the value of conducting a case-study analysis of offenders' crime stories but also accept that this method could be criticised for lacking 'scientific' rigor. I present the reader with the same information I used when· categorising judicial responses as disparate. Some of the disparate cases are outlined in the present chapter. All remaining cases of disparity and some examples of different and similar judicial responses are presented in Appendix Two. To some extent, this may curb criticisms of 'objectivity' and 'bias' (as discussed above), by essentially allowing my assessment to be reviewed. 2.1 Drug Offences The Misuse of Drugs Act (1975) is the key statute governing penalties for drug offending in New Zealand. In addition to maximum statutory penalties, there are two other key sections in the Act which offer Judges direct sentencing guidance. In section 6 [4] there is a statutory presumption of imprisonment for offenders who import, produce, manufacture, All remaining disparate cases appear in Appendix Two. Examples of different or similarresponses are not presented in this chapter but do appear in Appendix Two. 22 Protecting offenders' privacy and anonymity was paramount and the Department For Courts granted access under the condition that I would be the only person to access the trial files. 20 21 122 supply, administer, offer to supply, offer to administer a Class A drug, or who possess a Class A drug, unless; Having regard to the particular circumstances of the offence or of the offender, including the age of the offender if he is under 20 years of age, the Judge or Court is of the opinion that the offender should not be so sentenced (Misuse of Drugs Act 1975 s 6 [4]). Furthermore, where any person is convicted of importing, producing, manufacturing, supplying, administering, offering to supply, offering to administer, or possessing a Class A or B drug and a Judge decides to impose a full-time custodial sentence, he/she should also consider whether to give a fme. Section 7 of the Misuse of Drugs Act 1975 also states that when sentencing an offender for supplying, administering, offering to supply or administer, or otherwise dealing any Class C drug to persons aged over 18 years, imprisonment should not be imposed unless, "by reason of the offender's previous conviction or of any exceptional circumstances relating to the offence or offender, the Judge is of the opinion that such a sentence should be imposed." We have seen that female offenders tended to receive less serious judicial outcomes but that this is sometimes explained by men's more serious criminal histories (see Chapter 4). In the case-study analysis, 21 pairs of drug offenders (see Table 66 & 67) were studied to determine if sentencing and remand outcomes were similar, different or disparate. Table 66 and Table 67 summarise judicial remand and sentencing responses for offenders by grouping them into eight categories according to similarity, difference and disparity. The types of offences committed and the judicial outcomes received within these categories are also presented in these tables. The case-study results presented in Tables 66 and 67 confIrm my statistical analysis (see Chapters 3 & 4). It is shown that in comparison to women, men were generally 'disadvantaged' when it comes to remand and sentencing outcomes. Table 66 summarises judicial remand responses for offenders by grouping them into seven categories and presents details of remand outcomes in these categories. As seen in Table 66, of the 21 pairs, 12 received disparate outcomes. In all but two cases, the disparity 'favoured' women. Difference was recorded in four cases and five cases were assessed as similar. Table 67 summarises judicial sentencing responses for offenders by grouping them into eight categories and presents details of sentencing outcomes in these categories. Results again show that of the 21 pairs assessed, 12 received disparate outcomes and in all but two cases, the disparity 'favoured' women. Difference was recorded in six cases and similarity in a further three cases. 123 TABLE 66 m CATEGORIES OF JUDICIAL REMAND RESPONSES AND OUTCOMES FOR 21 PAIRS OF DRUG OFFENDERS Remand Outcome CATEGORY 1 (D~PARrY) MeQan and Stew Jackie and Sven Joan and Gary Elaine and Alex Penni and Tony CATEGORY 2 (D~PARrY) Jill and Denis Mary and Tim Helen and Richard CATEGORY 3 (D~PARrY) Alison and Ray CATEGORY 4 (D~PARrY) Kath and Chad Monica and Ross Megan = Bail with special conditions, Stew = Bail with special conditions, custodial remand 7 days Jackie - Bail with standard conditions, Sven - Bail with special conditions, custodial remand 29 days Joan = Bail with special conditions, Gary = Bail with standard conditions, custodial remand 24 davs. Elaine = Bail with standard conditions, custodial remand 14 days, Alex = Bail with special conditions, custodial remand 65 days Penni = Bail with standard conditions, Tony= Bail with special conditions Jill = Bail with special conditions, custodial remand 7 days, Denis = Bail with standard conditions, custodial remand 31 days Mary = Bail with standard conditions, Tim - Bail with special conditions, custodial remand 8 days Helen = Bail with standard conditions, custodial remand 20 days, Richard = Bail with special conditions, custodial remand 42 days Alison = Bail standard conditions, Ray = Bail standard conditions Kath = Bail with special conditions, Chad = Bail with special conditions Monica = Bail with standard conditions, Ross = Bail with standard conditions CATEGORY5(D~Pr) Amanda and Callum Amanda = Bail with special conditions, custodial remand 22 days, Callum = Bail with standard conditions, custodial remand 7 days. CATEGORY 6 (DIFFERENCE) Melinda and Cameron Melinda = Bail with special conditions, Cameron = Custodial remand 176 days Jane and Greg Jane = Bail with special conditions, Greg = Custodial remand 168 days. Patricia and Ron Patricia = Bail with special conditions, custodial remand 12 days, Ron = Bail with special conditions, custodial remand 22 days Beth = Bail with standard conditions, Stephen = Bail with special conditions Beth and Stephen CATEGORY? (SIMILAR) Kari and Jim Jenni and Ron Sally and Warren Debbie and Bobby Carmen and Steve Kari = Bail with standard condition, Jim = Bail with standard condition Jenni = Custodial remand 35 days, bail with special conditions, Ron = Custodial remand 56 days, bail with special conditions Sally = remanded on bail with special conditions, Warren = remanded on bail with special conditions Debbie = Bail with special conditions, custodial remand 7 days, Bobby = Bail with standard conditions, custodial remand 9 days Carmen = Bail with special conditions, Steve = Bail with special conditions Category 1 - men's and women's criminality was judged similarly serious but men received more severe remand outcomes. Category 2 - women's criminality was judged more serious than men's but men received more severe remand outcomes. Category 3 - women's criminality was judged more serious than men's but they received similar remand outcomes. Category 4 - men's criminality was judged more serious than women's but they received similar remand outcomes. Category 5 - women's and men's criminality was judged similarly serious but women received more severe remand outcomes. Category 6 - men's criminality was judged more serious than women's and men received more severe remand nld,..nmac:: 124 TABLE 67 D CATEGORIES OF JUDICIAL SENTENCING RESPONSES AND OUTCOMES FOR 21 PAIRS OF DRUG OFFENDERS Sentence CATEGORY 1 (D~PARrY) Megan and Stew Jackie and Sven Jenni and Ron Amanda and Callum Joan and Gary Elaine and Alex Megan = 6 months imprisonment, Stew = 12 months imprisonment Jackie = Twelve months supervision, Sven - 9 months imprisonment Jenni = 12 months imprisonment, Ron = 3 years 9 months imprisonment Amanda = 18 months imprisonment suspended for two years, eight months periodic detention, $800 forfeiture, Callum = 3 years imprisonment I Joan = 6 months imprisonment, Gary= 16 months imprisonment Elaine = 12 months imprisonment suspended for 18 months, 6 months periodic detention, Alex = 12 months imprisonment CATEGORY2(D~Pr) Jill and Denis Sally and Warren Alison and Ray CATEGORY 3 (D~PARrY) Mary and Tim CATEGORY 4 (D~PARrY) Kath and Chad Jill = 18 months supervision, Denis = 15 months imprisonment, $2,000 fine, forfeiture of Motor Vehicle. Sally = 9 months imprisonment suspended for 12 months, 6 months periodic detention, Warren = 9 months imprisonment Alison = $500 fine, $95 court costs, Ray = $2,000 fine, $95 court costs Mary = 18 months imprisonment, Tim = 18 months imprisonment Kath = 2 years imprisonment, $9,300 forfeiture, Chad = 2 years imprisonment, $9,300 forfeiture CATEGORY5(D~Pr) Kari and Jim CATEGORY 6 (DIFFERENCE) Melinda and Cameron Jane and Greg Monica and Ross Ron and Patricia Beth and Stephen Kari = 6 months periodic detention, 12 months supervision, Jim = 8 month periodic detention Melinda = 13 months imprisonment suspended for 18 months, Cameron = 4 years imprisonment Jane = 12 months supervision, Greg = 18 months imprisonment Monica = 9 months imprisonment, Ross = 24 months imprisonment Patricia = 12 months supervision, Ron = 18 months imprisonment suspended for 2 years, 9 months periodic detention, 12 months supervision Beth = 9 months imprisonment suspended for 2 years, 5 months periodic detention, Stephen = 9 months imprisonment suspended for 18 months, 6 months periodic detention, $800 forfeiture CATEGORY 7 (DIFFERENCE) Helen and Richard CATEGORY 8 (SIMILAR) Debbie and Bobby Penni and Tony Carmen and Steve Category 1 Category 2 Category 3 Category 4 Category 5 Category 6 Category 7 Category 8 - Helen = 6 months imprisonment, Richard = 24 months supervision Debbie = 18 months imprisonment, Bobby = 12 months imprisonment Penni = 60 hours community service, $400 fine, Tony = 200 hours community service Carmen = 5 months imprisonment, Steve = 5 months imprisonment men's and women's criminality was judged similarly serious but men received more severe sentences. women's criminality was judged more serious than men's but men received more severe sentences. women's criminality was judged more serious than men's but they received similar sentences. men's criminality was judged more serious than women's but they received similar sentences. women's and men's criminality was judged similarly serious but women received more severe sentences. men's criminality was judged more serious than women's and men received more severe sentences. women's criminality was judged more serious than men's and women received more severe sentences. men's and women's criminality was judged similarly serious and they received similar sentences. Examples of case studies where male and female drug offenders were judged as receiving disparate outcomes are now presented. Included are the following offender pairs: Joan and Gary (Category I remand outcome, Category 1 sentencing outcome), Elaine and Alex Category I remand outcome, Category 1 sentencing outcome), Jill and Denis (Category 2 remand outcome, Category 2 sentencing outcome), Sally and Warren (Category 7 remand outcome, Category 2 sentencing outcome), Mary and Tim (Category 2 remand outcome, Category 3 sentencing outcome), Kath and Chad (Category 4 remand outcome, Category 4 sentencing outcome), Kari and Jim (Category 7 remand outcome, Category 5 sentencing outcome) (see Tables 66 & 67). 2.1.1 Joan and Gary, Possession of Class C Drug for Sale/Supply These two offenders were coded fairly similarly - Joan (35 years, Maori) and Gary (27 years, Maori) were sole offenders, they acted in private locations and pled not guilty to one count of possessing Class C drugs for supply. Neither offender had prior drug, bail breach 125 convictions or a history of imprisonment but Joan's criminal convictions were more numerous than Gary's: 18 prior convictions as opposed to eight. Joan's last conviction was also more recent than Gary's: three years ago compared to eight. In both cases, long crime-free periods may have reduced criminal seriousness. After examining the crime details further it was found that in contrast to Joan, Gary's crime involved larger amounts of cannabis, 70 foils compared with 40. Conversely however, Joan had $700 in her possession, the proceeds of drug sales, compared with Gary's $200. Taking $20 as the usual price for a cannabis foil, Gary's offending involved $1,600 of cannabis while Joan's involved $1,500. Gary's crime also had gang connections and he appeared to be 'taking the rap' in line with gang rules so on one hand, Gary's gang affiliations may have increased his blameworthiness by attaching a degree of professionalism to his criminal act. On the other hand, 'taking the rap' may have reduced Gary's culpability; he could be seen as being 'under the gang's influence'. Alternatively, general 'prejudice' toward gang members could have resulted in harsh judicial treatment - yet this can hardly legitimate it. After an undercover operation police executed a search warrant on Joan's house. On the execution of the warrant police found forty foils of cannabis and $700 in cash. When questioned by police she denied that the cannabis was for supply and claimed that is was for the use of her husband and herself. (Joan' Possession of a Class C Drug for Sale/Supply, Extracted from Trial Transcripts and Judge's Sentencing Remarks) Gary obtained seventy cannabis foils and $200 cash from an associate and concealed them in the boot of his car. He then drove to a local gang house. The police subsequently arrived, searched the car and found the cannabis. Gary claimed that the cannabis was for his own use and refused to supply the police with the name of his friend. During the course of the trial Gary sought release from the admission that the cannabis was his and claimed that he had lied to the police and that in accordance with gang rules he had accepted responsibility for the cannabis because he was the driver of the car. (Gary· Possession of a Class C Drug for Sale/Supply, Extracted from Trial Transcripts and Judge's Sentencing Remarks) While perhaps overly cautious, slightly higher drug values and gang connections in Gary's case meant that I judged his current criminality to be marginally more serious than Joan's. However, Joan's substantial criminal history meant that I fmally judged this pair's overall criminality as similarly serious. Despite this, Joan's judicial outcomes were less harsh than Gary's: Joan was remanded on bail with special conditions and sentenced to six months imprisonment; Gary was remanded into custody for 24 days, in addition to bail with standard conditions, and sentenced to 16 months imprisonment (see Tables 66 & 67). Subsequently, I concluded that disparity existed in this case because no legitimate reason for these judicial discrepancies could be found. 2.1.2 Elaine and Alex, Selling/Supplying a Class B Drug Elaine (40 years, European) and Alex (26 years, European) both entered pleas of not guilty to a count of supplying a Class B drug and acted alone. One difference in the coded 126 crime elements was found in offence location: Elaine's drug dealing took place in a commercial establishment but Alex's occurred at a private location. Elaine and Alex both had past criminal convictions. In contrast to Alex, Elaine's criminal history was more serious primarily because she had more convictions: 24 compared to 14. Furthermore, it was only two months since Elaine's last criminal conviction - compared with eleven months in Alex's case. Elaine and Alex had similar drug offending histories, five and six convictions respectively, but neither had prior convictions for breaching bail, nor had either served imprisonment terms in the past. Examination of the crime details showed more drugs were involved in Alex's case, but this was the only difference between his and Elaine's offending. That factor, weighted against Elaine's more substantial criminal history, caused me to consider this pair's overall criminality to be similar. Despite this, Alex's judicial outcomes were more severe: Alex was remanded on bail with special conditions, remanded in custody for 65 days, and sentenced to twelve months imprisonment; Elaine was remanded on bail with standard conditions, remanded in custody for 14 days, and sentenced to twelve months imprisonment which was suspended, plus six months periodic detention (see Tables 66 & 67). I concluded there was disparity in this case. Elaine was legitimately prescribed morphine sulphate tablets by her general practitioner. Upon receiving her prescription she went to a local massage parlour and supplied a female friend with six of the thirty milligram morphine sulphate tablets she had been given of prescription. (Elaine - Supplying a Class B Drug, Extracted from Trial Transcripts and Judge's Sentencing Remarks) Alex was legitimately prescribed morphine sulphate tablets by his general practitioner. Upon receiving his prescription he went to his girlfriend's house and supplied her with ten of the sixty milligram morphine sulphate tablets he had been given on prescription. (Alex - Supplying a Class B Drug, Extracted from Trial Transcripts and Judge's Sentencing Remarks) 2.1.3 Jill and Denis, Selling/Supplying a Class B Drug Jill (23 years, European) and Denis (32 years, European) entered pleas of guilty to more than one count of selling Class B drugs from private locations. Both offenders acted alone. After looking at the crime stories in more detail I decided that Jill's current offending was more serious than Denis's. First, Jill was convicted on 13 counts compared to three counts in Denis's case. Second, different types of drugs were involved. Denis was convicted on two counts of selling a Class B drug (cannabis oil) and one count of selling a Class C drug (cannabis plant). Jill sold morphine (a Class B drug) and cannabis oil in addition to eleven counts of selling a Class C drug (cannabis plant). Compared with selling cannabis oil, New Zealand Judges usually consider the selling of morphine to be more serious. 23 Thirdly, Jill's 23 The Ministry Of Justice Seriousness of Offence scale scores selling morphine (270) higher than selling cannabis oil (209). 127 drug dealing was monetarily more lucrative than Denis's (see extracts from the police report below). There was one possible mitigating factor in Jill's case: her partner was involved, to a minimal extent, in some of the drug dealing activity but at best, this reduced Jill's culpability only marginally. The police made arrangements for two plain clothed police officers to work within the Christchurch area. During the course of their duty these officers became aware of Jill's drug dealing activity. Over the course of six months Jill had approached the police officer and sold him Class C drugs on eleven occasions and Class B drugs, namely Morphine, on two occasions. The police officer received a phone call from Jill at his home address and asked whether the constable wished to buy some MST's (Morphine Sulphate Tablets). Jill stated she had 50 pills for sale at $4 each. Later that day the officer drove around to Jill's address, purchased the 50 pills and handed her $200 cash. Jill then offered the officer two foils containing cannabis leaf in return for 5 of the original morphine tablets sold. The deal was subsequently carried out. A few days later the officer was visiting Jill's address. While there Jill asked him whether he wished to buy 30 caps of cannabis oil. She stated that the price was $20 per cap; $600 was given to her by the police officer. A week later, the officer was visited by Jill at his home address; she asked whether the officer was interested in purchasing an ounce of cannabis. The officer indicated that he was and Jill made a phone call. A short time later two male persons arrived at the address and were introduced to the police officer by Jill. Following a brief discussion, one of the males left the address and returned a short time later with the cannabis. The ounce bag was shown to the officer and a price was discussed with Jill. The officer gave the male person $190 in cash for the ounce bag. A few weeks later the officer was at his business address when he received a phone call from the de facto of Jill. There was a short discussion regarding the selling of cannabis. Later that day Jill arrived at the officer's address and removed a plastic bag containing cannabis leaf from her handbag. A price was discussed - $280 - for the ounce bag of cannabis. Jill placed the money in her handbag and left. Approximately 1 week later, Jill contacted the officer again at his home address by phone. She asked the officer if he would like to visit her home and look at an ounce of cannabis. A short time later the officer visited Jill's address and was escorted into the bedroom by Jill. Once in the bedroom she produced a plastic bag containing cannabis from her bedside draw. The bag weighted approx one ounce and the price of $250 was decided upon. The officer handed Jill $215, which she placed in her tracksuit pocket. She then asked the officer whether he wished to buy a further two ounces of cannabis, which the officer stated he would like to see. She then left the bedroom, returning a short time later with a further two ounce bag of cannabis, which she produced from under her jersey. A price was finally decided upon, a further $500 in cash, which he obtained from his car. He then handed Jill the money. Some weeks later the officer received another phone call from Jill. She asked him if he would like to buy another ounce of cannabis. Later that day Jill arrived at the officer's home and handed him a plastic bag containing cannabis leaf. The bag weighed one ounce and after a phone call by Jill, the price of $200 for the cannabis was decided upon. The officer gave Jill the money, which she placed in her handbag. Ten days later, the officer spoke to Jill on the phone. She stated that she would have some cannabis at her address within the next 10 minutes. The officer subsequently went along to her home address and was met by Jill at the front door. Jill's de facto produced two plastic bags containing cannabis leaf from a bag that was sitting at his feet under the dining room table. The officer looked at the cannabis and the price of $550 for the two ounce bags was decided upon and handed to Jill's partner. Over the next 2-3 weeks Jill made more phone calls to the officer offering him ounces of cannabis. In all these cases the officer picked Jill up from her home address and was directed to the addresses of other persons. The deals were subsequently negotiated between the officer and these other persons, all monies were given to the people at these addresses. (Jill - Selling/Supplying a Class B Drug - Extracted from Police Summary of Facts) Under cover police officers became aware that Denis was involved in drug dealing and they went to his address in Christchurch. At the house police spoke to Denis and he offered fourteen caps of cannabis oil to the officers at a price of $40 each. After agreeing on this price, the Police officer paid Denis $560 in cash and received the cannabis oil. A few days later, Denis called at the police officer's address and told him that he had in his possession 10 caps of cannabis oil. He stated that he was selling the caps for $40 each but agreed to sell them for $30 each provided the officer purchased all 10 caps. The police officer paid Denis $300 in cash and was handed the cannabis oil. A week later Denis called at the police officer's address again. He had in his possession 50 foils of cannabis plant. He offered to sell the foils at $20 each provided the police officer purchased all 50. The officer agreed and paid $1,000 in cash to Denis. When perpetuating these offences, Denis drove to the police officer's house in a motor vehicle. After completing the sale, he left the address in the same vehicle. When spoken to by police the Denis stated that he had committed the offences because he needed the money. (Denis - Selling/Supplying a Class B Drug, Extracted from Police Summary of Facts) As far as differences in criminal history were concerned, Denis had nine prior convictions and Jill had five, but four of Jill's prior criminal convictions were for drug offending compared with 128 two in Denis's case. Neither offender had prior bail breaches or imprisonment histories. I subsequently decided that their criminal history seriousness was comparable so, given differences in current criminality, Jill's overall criminality was judged more serious. However, Denis's judicial outcomes were harsher than Jill's. Denis was remanded in custody for 31 days in addition to being bailed with standard conditions and he was sentenced to 15 months imprisonment, $2,000 fine, and forfeiture of his motor vehicle. In contrast, Jill was remanded on bail with special conditions, spent just seven days in custodial remand and was sentenced to 18 months supervision (see Tables 66 & 67). Thus, disparity was concluded because it was not immediately clear why Jill, whose criminality was far more serous than Denis's, had been treated less harshly in this case. 2.1.4 Sally and Warren, Selling/Supplying a Class C Drug The coded elements of Sally's (38 years, European) and Warren's (25 years, European) cases were the same: both offenders entered pleas of guilty to one count, and acted alone in private locations. The details of Sally's and Warren's cnmes were further considered and minor differences found. As opposed to Warren, Sally's case involved smaller amounts of cannabis (five foils compared with eleven), but money from the sale of an extra eight foils (approximately) was found in Sally's possession, making the act of drug dealing similar in both cases. One mitigating factor, which could have exonerated Sally to some extent, was her cooperation with police; Sally admitted supplying drugs, unlike Warren. The police executed a search warrant at Sally's home address. Sally was asked if she had any cannabis at the address, whereby she led the police to a small plastic bag containing 5 cannabis bullets rolled in tinfoil. When asked what the cannabis was for she stated she was going to sell it. In explanation, Sally claimed she "owed a lot of money for hire purchases." She decided to make quick money from selling drugs, but claimed she had only been doing it for a maximum of 2 weeks. $160 cash was also located at the address, which she accepted was the proceeds of previous sales. (Sally -Selling/Supplying a Class C Drug, Extracted from Police Summary of Facts) Police observed a person walk into a central city gang address and a short time later, they saw the same person come out. A constable then approached the person and spoke to him. As a result, this person admitted to purchasing a cannabis foil for $20 from a person inside the address. The following day, the police executed a search warrant on the address. Warren was located in a bedroom at the rear of the address (and later identified by the purchaser of the cannabis). When spoken to, Warren denied selling any form of drugs. A subsequent search resulted in a shampoo container being located in a bathroom next to Warren's bedroom. The container had 11 cannabis foils wrapped in a plastic bag inside it. When spoken to, Warren admitted that the cannabis was his and stated that it was for his own personal use. (Warren -Selling/Supplying a Class C Drug, Extracted from Police Summary of Facts) Erring on the side of caution, I judged Warren's current criminality as marginally more serious than Sally's but her past criminality was convincingly more serious than his. It was 14 months 129 smce Sally's last criminal conviction, she had 31 prior convictions including 15 for drug offending and had served three prior imprisonment terms. Warren had also served three prior imprisonment terms. However, in contrast to Sally, Warren had 26 prior convictions, which included only one for drug offending and it was over four years since his last conviction. Unlike Sally, Warren had a prior conviction for breaching bail. The more extensive and serious nature of Sally's past offending outweighed any small differences in current criminality and I concluded that her overall criminality was more serious than Warren's. Both offenders' remand outcomes were the same (bail with special conditions) but Sally was sentenced to a suspended imprisonment term plus six months periodic detention while Warren was sentenced to nine months imprisonment (see Tables 66 & 67). Judicial sentencing responses were considered disparate because it was not clear why Warren's sentence was harsher than Sally's. However, Warren's past bail breach meant similar remand outcomes had been expected, so in this respect judicial responses were similar. 2.1.5 Mary and Tim, Selling/Supplying a Class B Drug Mary (25 years, European) and Tim (26 years, European) both entered pleas of guilty to selling Class B drugs. Both offenders played primary/equal roles in their offences as sole . perpetrators and committed their crimes at private locations. Mary was convicted on more than one count while Tim was convicted of just one count. Mary's and Tim's criminal histories differed, but Mary's was more serious. Tim had five prior drug convictions, compared with Mary's four, and his last conviction was less than six months prior to the current offence, compared with two years in Mary's case. However, this was outweighed by the extent of Mary's criminal history: she had 132 prior convictions; Tim only had 31. Both offenders had a prior bail breach. Mary had received imprisonment sentences on three previous occasions, while Tim had only been imprisoned once. The crime details show Mary's and Tim's drug dealing was prolonged and organised but I decided Mary's current criminality was more serious primarily because she was convicted on two counts (one of supplying a Class B Drug and one for possessing a Class B drug for supply) while Tim was convicted on one count of supplying a Class B drug. Furthermore, drug values in Mary's case were nearly ten times higher than those in Tim's. Police executed a search warrant at Mary's address. Located in her bedroom were 5 x 100 mg and 38 x 30 mg Morphine Sulphate Tablets. These were in unlabelled pill bottles. Also located was a diary in Mary's name which contained calculations relating to morphine sulphate tablets, both 30 mg and 100 mg, the numbers she had had in her possession at various times and lists of people they has been sold to, including amounts owed to her as a result. The calculations indicated that Mary had sustained a 130 supply of the drugs and was charging $140 for a 100 mg tablel and $45 for a 30 mg tablet. $2,400 was the street value of the tablets located. Mary made partial admissions regarding her drug dealing activities to police, saying that she sold 30 mg tablets to friends of hers who were intravenous drug users at slightly cheaper than normal 'street' price. (Mary -Selling/Supplying a Class B Drug, Extracted from Police Summary of Facts) The police drug squad became aware that Tim was selling Morphine Sulphate tablets from an address in the central city. Police investigations into the activities established that between 30 and 50 people were observed entering the address for short periods of time and then leaving. A number of these people were known to the Police to be drug addicts. This occurred on a daily basis. Tim was located by the Police and spoken to about the activities at the address. He admitted selling morphine sulphate tablets for a person he would not identify. For the sale of the Morphine Sulphate Tablets Tim received $5 for each one sold. He said he received approximately $250. Tim admitted the reason for selling the Morphine Sulphate was to earn money but stated he used the tablets intravenously hirnself, which caused money problems. (Tim - Selling/Supplying a Class B Drug, Extracted from Police Summary of Facts) Overall, Mary's criminality was thus considered more senous than Tim's but they both received 18 months imprisonment and Mary's remand outcome was less serious: she was remanded on bail with standard convictions while he was remanded into custody for eight days, in addition to bail with special conditions (see Tables 66 & 67). 2.1.6 Kath and Chad, Selling/Supplying a Class C Drug Coded crime elements showed Kath's (28 years, European) and Chad's (26 years, European) offending was similar: both offenders entered pleas of guilty to one count of selling a Class C drug, they were co-offenders but each played active roles. Their offending occurred in a private location. Crime details supported the coded crime elements. Although Kath owned the house in which the drug dealing took place, it was clear Chad spent most of his time there and their role in selling drugs was similar. This pair was involved in a fairly organised and large-scale drug dealing operation and neither offender was overly cooperative with police. I concluded that Kath's and Chad's current criminality was similarly serious. The police carried out daily observations of Kath's and Chad's address after receiving a number of complaints from local residents regarding the large number of visitors to the address on a daily basis. Observations showed that between daylight hours 50 to 100 persons visited the offenders' address. The visitors would be let into the address through the front door, they would then only stay at the address a short time before leaving. On occasions money could be seen exchanging hands. When these persons were stopped they were found to be in possession of Temgestic tablets. The police executed a search warrant on the offender's address and located 75 Temegestic tablets along with approximately $8, 000 in cash. After executing the search warrant at the address, visitors continued to arrive and when police answered the door requests for drugs were made to them. The day after the police search people continued to purchase drugs from the address. Kath was the owner and occupier of the house. On many occasions during the operation, she was seen to open the door and invited persons into her house. During observations Kath was seen to be home on all days during the police operation. When spoken to by police Kath denied ever selling drugs from her home. In explanation for the large number of visitors, she stated that she had "a lot of friends." Chad is Kath's partner and during the operation, observations showed that he was at the address each day and appeared to actively live there. He was seen opening the door to visitors and inviting them into the house. When spoken to Chad stated that he only stays at the address approximately 2-3 nights during the week and denied any knowledge of being involved in the sale of drugs from the address, nor was he aware that drugs were being sold from the address. In explanation for the large number of visitors coming to the address, Chad stated he had a lot of friends, that he had no knowledge of any drug deal taking place, nor was he actively involved in receiving money for drugs. (Kath and Chad - Selling ISupplying a Class C Drug, Extracted from Police Summary of Facts) 131 It may have been that Kath's and Chad's current criminality was similar, but Chad's criminal history was more serious than Kath's. Chad and Kath each had 13 prior convictions for drug offending but overall Chad had 47 prior convictions in all, compared to 24 in Kath's case, and he had been sentenced to imprisonment on 15 separate occasions while Kath had only been imprisoned twice. Neither Kath nor Chad had prior convictions for breaching bail and it was approximately six months since they had been last convicted. I considered Chad's overall criminality to be more serious than Kath's. However, they received the same remand and sentencing outcomes; bail with special conditions and two years imprisonment, plus forfeiture of $8,000 so disparity was concluded this time in favour of Chad - the male (see Tables 66 & 67). 2.1. 7 Kari and Jim, Manufacturing a Class B Drug Both Kari (18 years, Maori) and Jim (20 years, European) entered pleas of guilty to one count of manufacturing a Class B drug, they were co-offenders, had active/equal offence roles and manufactured drugs in a private location. Kari and Jim both appeared as first time offenders and further investigation of the crime details failed to uncover any differences in their offending. Kari and Jim co-operated with police and admitted manufacturing drugs: Police executed a search warrant upon Kari's and Jim's motel room. As a result of this search police located cannabis material, used syringes, traces of cannabis oil, along with a pressure cooker containing cannabis leaf soaked in a alcohol based liquid. When spoken to by police both Jim and Kari admitted possession of the material and syringes as essential ingredients to produce cannabis oil. The offenders admitted that it was a joint enterprise to produce cannabis oil and that it was for their own use and not for sale. Further, the offenders were fully co-operative and indicated that it was more of an experiment. (Jim and Kari - Manufacturing a Class B Drug, Extracted from Police Summary of Facts) I subsequently judged Kari's and Jim's overall criminality as similar, but while they received the same remand outcome (bail with standard conditions), they were sentenced differently: Kari was sentenced to six months periodic detention plus twelve months supervision with the condition to receive counselling, Jim was sentenced to eight months periodic detention (see Tables 66 & 67). Although supervision is primarily rehabilitative and periodic detention is a punitive sanction, I still considered Kari's sentencing outcome to be marginally more serious because she was under judicial control for a longer period than Jim. Given similarities in this pair's criminality, sentencing outcomes were considered disparate and slightly in favour of the male. 2.2 Violent Offences The Crimes Act 1961 and the Criminal Justice Act 1985 provide specific guidance to Judges when dealing with violent offenders. Remand guidance is found in Section 318 of the Crimes Act 1961 where it is noted that persons found guilty of, or pleading guilty to a serious 132 violent offence and who have one or more previous convictions for a serious violent offence shall, while waiting to be sentenced, not be granted bail or allowed to remain at large. Section 5 (1) of the Criminal Justice Act 1985 further creates the presumption of imprisonment for serious violent offenders in the following terms; "Where "(a) An offender is convicted of an offence punishable by imprisonment for a term of two years or more, and (b) The Court is satisfied that in the course of committing the offence, the offender used serious violence against or caused serious danger to the safety of, any person The Court shall impose a full-time custodial sentence on the offender, unless the Court is satisfied that, because of the special circumstances of the offence or of the offender, the offender should not be so sentences." Of further relevance to the current research is the parole legislation governing the release of prisoners jailed24 for serious violent offences. Persons sentenced from two years to 15 years for a serious violent offence 25 are ineligible for parole and are generally required to serve twothirds of their sentence before being released on remission (Criminal Justice Amendment Act 1987 [s93]).26 In contrast, those sentenced to imprisonment for more than one year but less than two years, are eligible for parole after serving one-third of their ~entc (Criminal Justice Act 1993 [s89] 27 and persons sentenced to one year or less are released at their final release date after serving half their sentence (Criminal Justice Act 1985 [s95]). The case-study results presented in Tables 68 and 69 confirm previous statistical analysis (see Chapters 3 & 4). In comparison to women, men were generally 'disadvantaged' when it came to remand and sentencing outcomes. Table 68 summarises judicial remand responses for offenders by grouping them into six categories and presents details of remand outcomes in these categories. Of the 14 pairs assessed, four cases were found to be disparate. In all but one case this disparity 'favoured' women. Difference was recorded in six cases and similarity in four. Table 69 summarises judicial sentencing responses for offenders by grouping them into six categories and presents details of sentencing outcomes in these categories. Of the 14 pairs assessed, nine cases were considered disparate. In all but one case this disparity 'favoured' women. Different judicial outcomes were highlighted in a further five cases. 24 In New Zealand, and other Commonwealth nations, the terms 'jail' and 'prison' are used interchangeably. 25 The following violent offences are identified as serious in the Crimes Act 1961 s. 318 (only those relevant to the current research are outlined): wounding with intent to cause grievous bodily harm, robbery and aggravated robbery. 26 The Criminal Justice Amendment Act 1987 [s93J actually states that "an offender shall not be eligible to be released on parole in respect of any sentence of imprisonment for a term of more than two years imposed on the offender." However, prisoners sentenced to two years imprisonment for a serious violent offence generally do not appear before the District Prisons Board (parole board) until they have served two thirds of their sentence. 27 Before 1993 offenders were not eligible for parole until they served half of their sentence (Criminal Justice Act 1985 [s93]). However, this is not relevant for the case-study analyses because all 'serious violent offenders' were sentenced after 1993. 133 TABLE 68 14 CATEGORIES OF JUDICIAL REMAND RESPONSES AND OUTCOMES FOR Q PAIRS OF VIOLENT OFFENDERS Remand Outcome CATEGORV1 (D~PARITV) Selina and Roger Joanne and Grant Selina, = 7 days in custodial remand, bail with special conditions, Roger -295 days in custodial remand Joanne remanded on bail with special conditions, Grant = 210 days in custody CATEGORV2(o~PI) Margo and Ben Category 3 (D~PARITV) Rebecca and Jason CATEGORV 4 (DIFFERENCE) Sue and Mark Nicole and Aaron Michelle and Paul Sarah and Simon CATEGORV 5 (DIFFERENCE) Kate and Phil Joan and Andrew CATEGORV6 (SIMILAR) Maria and Karl Tania and Allan Jane and David Belinda and Regan Margo = bail with special conditions, Ben = bail with special conditions Rebecca = remanded on bail with special conditions, Jason = Remanded on bail with special conditions Sue = 2 days in custodial remand, bail with special conditions, Mark = 36 days in custodial remand Nicole = remanded on bail with soecial conditions Aaron = 120 days in custodial remand Michelle = Bail with special conditions, Paul - Custodial remand Sarah = 7 days in custodial remand, bail with special conditions, Simon = 71 days in custodial remand Kate = 32 days in custodial remand, Phil = Bail with special conditions Joan = 64 days in custodial remand, Andrew = 59 days in custodial remand Karl = 28 days remanded in custody, Maria - 29 days remanded in custody Tania = remanded on bail with special conditions Allan = remanded on bail with soecial conditions Jane = remanded on bail with special conditions, David = remanded on bail with special condition Belinda = remand on bail with standard conditions, Regan = remand on bail with standard conditions Category 1 • men's and women's criminality was judged similarly serious but men received more outcomes. Category 2 • women's criminality was judged more serious than men's but they received similar remand Category 3 • men's criminality was judged more serious than women's but they received similar remand Category 4 • men's criminality was judged more serious than women's and men received more outcomes. Category 5 • women's criminality was judged more serious than men's and women received more outcomes. TABLE 69 - severe remand outcomes. outcomes. severe remand severe remand CATEGORIES OF JUDICIAL SENTENCING RESPONSES AND OUTCOMES FOR 14 PAIRS OF VIOLENT OFFENDERS Sentence CATEGORV1 (O~PARITV) Tania and Allan Sue and Mark Selina and Roger Joanne and Grant CATEGORV2 (D~PARITV) Joan and Andrew Maria and Karl CATEGORV 3 (D~PARITV) Jane and David CATEGORY 4 (D~PARITV) Margo and Ben Belinda and Regan CATEGORV 5 (DIFFERENCE) Nicole and Aaron Michelle and Paul Sarah and Simon Rebecca and Jason CATEGORV 6 (DIFFERENCE) Kate and Phil Tania = 11 months imprisonment, Allan = 2 years, 9 months imprisonment Sue = 6 months imprisonment, Mark = 5 years imprisonment Selina, = 9 months imprisonment, Roger - 2 years, 6 months imprisonment Joanne = 9 months imprisonment suspended for 12 months, 9 months periodic detention, Grant = 12 months imprisonment Joan = 18 months imprisonment, Andrew = 3 years, 9 months imprisonment Maria = 1 year imprisonment, Karl = 4 years imprisonment Jane = 6 months periodic detention, 18 months supervision, David = 5 months periodic detention Margo = 12 months imprisonment suspended for 6 months, Ben = 8 months periodic detention, 12 months supervision Belinda = 5 months periodic detention, Regan = 6 months imprisonment suspended for 18 months. Nicole = 6 months periodic detention, Aaron = 13 months imprisonment Michelle = 6 months periodic detention, 12 months supervision, Paul = 4 years imprisonment Sarah = 1 year imprisonment, Simon = 2 years imprisonment Rebecca = 1 year 6 months imprisonment, Jason = 2 years 6 months imprisonment Kate = 5 years imprisonment, Phil = 2 years imprisonment Category 1 • men's and women's criminality was judged similarly serious but men received more severe sentences. Category 2 • women's criminality was judged more serious than men's but men received more severe sentences. Category 3 • men's and women's criminality was judged similarly serious but women received more severe sentences. Category 4 • women's criminality was judged more serious than men's but they received similar sentences. Category 5 • men's criminality was judged more serious than women's and men received more severe sentences. Category 6 • women's criminality was judged more serious and women received more severe sentences. Examples of cases where male and female violent offenders were judged as receiving disparate outcomes will now be presented. Included are the following offender pairs: Sue and Mark (Category 4 remand outcome, Category 1 sentencing outcome), Maria and Karl (Category 6 remand outcome, Category 2 sentencing outcome), David and Jane (Category 6 remand 134 outcome, Category 3 sentencing outcome), Margo and Ben (Category 2 remand outcome, Category 4 sentencing outcome) (see Tables 68 & 69). 2.2.1 Sue and Mark, Aggravated Burglary The coded elements of Sue's (24 years, Maori) and Mark's (22 years, European) aggravated burglary were very similar. Both acted with others, played active roles in the offence, invaded private residences, victimised women they did not know, caused physical harm, and used weapons. Sue and Mark had both been convicted on a single count and entered pleas of guilty to aggravated burglary. However, further investigation revealed differences between Sue's and Mark's crime. Both offenders had knives in their possession during the offence but unlike Sue, Mark used his knife to cause injury. Despite this, I judged Sue's offending as more severe and brutal; she only threatened her victim with the knife but then proceeded to methodically punch and kick her throughout the burglary. In comparison, the injury Mark inflicted on his victim could have been construed as unintended. Property of a higher value was also taken during Sue's burglary. In the early hours of the morning Sue and her two co-offenders went to a house where a party was taking place. A woman (the victim) came to the door and Sue verbally abused her, showed her a knife she was carrying and punched her in the head with a closed fist. She grabbed the victim by the throat with hand and pushed her back against the wall. The female co-offender then approached and punched the victim in the head with a closed fist. The male co-offender demanded money, and grabbed her handbag (value $180). Then all three offenders punched and kicked the victim; pulled her hair and tried to put her head through a window. The attacked stopped and Sue demanded the victim's jewellery. While she was taking her jewellery off, Sue and her female co-offender punched her in the head. The male co-offender took her jewellery valued at $500. All three offenders then dragged the victim onto the road, punching, kicking and pulling out her hair. (Sue - Aggravated Burglary, Extracted from Police Summary of Facts) The victims, a woman and a man, in this matter were at home watching television in their lounge when Mark and his co-offender, wearing balaclavas and dressed in black clothing, knocked in the back door. Mark was armed with a knife and his co-offender was armed with a piece of wood. The victims refused them entry. The female victim tried to stop them from entering but Mark and his associate pushed past her. Mark cut her arm when he pushed through the door. The male victim, armed with a knife, tried to protect himself and his wife. Mark's co-offender over powered him. Both offenders then demanded money and were handed over a small amount of cash ($50). The offenders then left the house. (Mark - Aggravated Burglary, Extracted from Police Summary of Facts) Sue's victim could also be perceived as more vulnerable than Mark's. Mark's victims were known drug dealers, with their deviant status perhaps making them less worthy of judicial sympathy. Both offenders' victims were harmed physically and emotionally by the attack; but physical injury was slightly more serious for Sue's victim. Unlike Mark, Sue was also uncooperative with police. Keeping in mind that defIning crimes involving female victims as more serious may not hold when the offender is also female (see Daly, 1994:110) I still concluded that Sue's current criminality was more serious than Mark's. 135 As a result of the attack the victim received bruising, and grazing to her head, face, and arms. She was not hospitalised but required medical attention to clean up her injuries. After the attack the victim also suffered from emotional distress and was fearful for her Iife ... When approached by police all three offenders struggled violently and tried to escape. (Sue - Aggravated Burglary, Extracted from Police Summary of Facts) As a result of the offending Mark was responsible for inflicting a small cut to the female victim's arm. This cut did not require medical attention. However, after the attack the female victim suffered emotionally. She was unable to sleep and lost her appetite. Became apprehensive about noises around the house at night and both her and her husband were forced to move from the address as a result. ... Mark admitted his part in the robbery, stating he had carried the knife and threatened the victims with it. (Mark - Aggravated Burglary, Extracted from Police Summary of Facts) Differences in Mark's and Sue's criminal histories were also found. Mark had more prior convictions (35) than Sue (16) and unlike Sue, he had also served a prior imprisonment term. But both had equal numbers of past convictions for violence. Sue's criminal record showed a prior conviction for assaulting a police officer and aggravated assault. Mark had a conviction for possessing an offensive weapon and common assault. It was four months since Sue's last conviction, compared with 15 months for Mark. While Sue was far from a ftrst time offender and her capacity for violence equaled Mark's, differences in the number of prior convictions and Mark's prior term of imprisonment led me to conclude that Mark's criminal history was marginally more serious than Sue's. Shorter periods between Sue's last conviction and Mark's, balance their criminal histories somewhat. After weighing Mark's criminal history against the more serious nature of Sue's present offence, I judged this pair's over all criminality to be similar. However, Mark's judicial outcomes were far harsher than Sue's. Mark was remanded in custody throughout the entire court process, spending a total of 36 days there and was then sentenced to five years imprisonment. Sue spent two days in custodial remand, the majority being spent on bail with special conditions and she received just six months imprisonment (see Tables 68 & 69). Although remand differences were explained by Mark's prior bail breaches, I found no reason for such vast sentencing differences, so sentencing disparity was concluded. 2.2.2 Maria and Karl, Wounding with Intent to Cause Grievous Bodily Harm Maria (40 years, Maori) and Karl (25 years, Maori) each entered a plea of guilty to one count of wounding with intent to cause grievous bodily harm. They had both acted alone, in private locations, and used knives to cause physical injury. There were differences in coded crime elements: Karl victimised a man he knew while Maria victimised a female stranger. This suggested Maria's victim was more vulnerable than Karl's, but defining crimes involving female victims as more serious may not hold when the offender is also female because "a 136 female victim may not be more vulnerable in comparison to a female offender" (Daly, 1994: 110). Further investigation of the crime details and criminal histories for these offenders showed Maria's overall criminality was marginally more serious than Karl's. Karl and Maria both had long criminal histories, including prior imprisonment terms, convictions for violence and breaching bail. Although Karl had served more terms of imprisonment than Maria (three compared with one), I judged Maria's offence history to be more serious than Karl's: she had 27 prior convictions including three for violence (assaulting a police officer, aggravated robbery and common assault) and four for breaching bail. In contrast, Karl had 26 prior convictions, including two for violence (assaulting a police officer and possession of an offensive weapon) and two for breaching bail. Furthermore, it was only nine months since Maria's last criminal conviction but Karl's last criminal conviction was over three years before. This crime-free period in Karl's life should have reduced the effect of prior convictions on judicial outcomes, by reducing overall seriousness. Not only did I consider Maria's past offending more serious than Karl's, so too did I judge her present offending to be more serious. Crime details showed that neither crime was overly premeditated and victims in both cases provoked the assaults to some degree, but Maria's victim, unlike Karl's, received life-threatening injuries. Maria's ex-girlfriend (Sue) was at a party when an argument developed between her and another female partygoer (the victim) who slapped Sue's face. Sue was upset by the argument and while in tears rung Maria to ask for a ride home. Another person grabbed the phone and told Maria what had happened and Maria immediately went to the address. When Maria arrived at the party she approached the victim and repeatedly punched her about the head and body, causing her to fall on the floor. When on the floor she kicked her about the head and body. While she was doing this she grabbed a kitchen knife which was nearby, stabbed the victim's arms and chest. Maria then stood over the victim and told her never to hit any of her friends again. Maria was then restrained by other persons. The victim was taken to hospital. She received abrasions, bruising and swelling to the head and body. She also received stab wounds to her arms and chest. As a result of the stab wounds she suffered a collapsed lung which, without surgery, would have resulted in death. She was hospitalised for five days. (Maria - Wounding with Intent to Cause Grievous Bodily Harm, Extracted from Police Summary of Facts) Karl was a close friend of his male victim (Brad). On the evening of the offence Karl was drinking with friends at a party. During the evening, and by chance, he met up with another friend (Don) who was from out of town and staying at Brad's house. Don told Karl that his girlfriend (Sue) had been at Brad's house earlier in the evening and that "Brad and Sue were getting into each other" (sexually). Karl was upset by this so he went back to the Brad's house with Don. Once there, Karl went into a bedroom to find Brad and Sue in bed together asleep. Angered by what he saw, Karl went into the kitchen and found a carving knife. He then returned to the bedroom and began to stab Brad with the knife. Karl then ran off. Karl's victim was taken to hospital where he received treatment for knife wounds. The victim required surgery on his arm. The wounds received although requiring stitching and surgery, were not regarded as life threatening. (Karl· Wounding with Intent to Cause Grievous Bodily Harm, Extracted from Police Summary of Facts) 137 When spoken to by police, Maria was cooperative and admitted the assault. Karl was cooperative too, but police records do not identify him as remorseful; Karl also claimed he could not remember the assault. When approached by the police Maria admitted the assault and said in explanation she thought her friend was severely hurt and this had caused her to become extremely angry and violent. She said there was no justification for the assault. Maria was noted to have been remorseful. (Maria - Wounding with Intent to Cause Grievous Bodily Harm, Extracted from Police Summary of Facts) Later that moming, the Karl was located at a local bar. In the course of an interview he admitted going to the Brad's house but stated that he could not remember seeing Brad or attacking him with a knife. (Karl - Wounding with Intent to Cause Grievous Bodily Harm, Extracted from Police Summary of Facts) While remorse may be a mitigating factor in judicial outcomes, I considered it an insufficient explanation for the substantial difference in Karl's and Maria's sentences. Karl received four years imprisonment while Maria only received one year (see Table 69). Maria's sentence was all the more questionable because wounding with intent to cause grievous bodily harm is defined in statute as a serious violent offence (Criminal Justice Act 1987 [s93]).28 There were no differences in remand outcome, both offenders were remanded into custody for similar periods (see Table 68) which is not surprising given the fact that both Maria and Karl had previously breached bail. 2.2.3 David and Jane, Assault with a Weapon David (20 years, Maori) and Jane (23 years, European) were both convicted on one count each of assault with a weapon. David entered a plea of guilty to this charge but Jane's plea was not guilty. Jane and David both acted alone as the primary perpetrators in the assaults, used knives and committed the crimes in private dwelling houses causing physical injury to male victims. David did not know his victim, while Jane did. A further look at the crime stories revealed that Jane had assaulted her ex-partner (Gavin), who received moderate injuries as a result. There was a degree of incitement from Gavin, who provoked the assault initially by hitting Jane in the face. In contrast, David assaulted an unsuspecting stranger in the back yard of his home, so compared with Jane's victim, David's appeared more innocent. Although blurred by intoxication, David's offence involved some planning: he went out with the intent to assault a person who had earlier insulted him. Unfortunately, David's drunkenness led him to assault the wrong person. David's victim was 28 Karl would be ineligible for parole, he would need to serve two-thirds of his sentence before being released on remission with possible conditions (Criminal Justice Act 1993 [s90], [sI01]). Maria would not appear before the parole board being released unconditionally after half her sentence was served (Criminal Justice Act 1993 [s90]). 138 seriously injured as a result of his violence and required surgery. It would seem that Jane's current criminality was less serious than David's. Jane had recently separated from her de-facto partner (Gavin). At the time of the incident Gavin had partially moved out of the house that they had shared together. On the day in question, after talking to Jane on the phone, Gavin went around to Jane's house to pick up the last of his possessions. They started arguing in the bedroom and Gavin punched Jane in the face. He then went into the kitchen, followed by Jane who proceeded to push him against the wall. At this time Jane took a knife from on top of the refrigerator and attacked Gavin, slashing his wrist in the process. (Jane - Assault With a Weapon, Extracted from Trial Transcripts and Judge's Sentencing Remarks) On the day of the assault David had been supporting a friend who was at a tenancy tribunal hearing. When he left the hearing David was verbally abused by an old acquaintance involved in his friend's case. That night, after drinking and becoming completely intoxicated, David tired to find this man's house but he was drunk so went to the wrong house. Making a lot of noise David was discovered by the owner of the house, a man unknown to him, in his back yard. A struggle occurred, and David stabbed his victim with a knife he had been carrying. (David - Assault With a Weapon, Extract from Police Summary of Facts) While David's current criminality may have been marginally more serious than Jane's, her criminal history (18 prior convictions) was more extensive than his (seven prior convictions) and Jane's last conviction was eleven months ago compared with seven years in David's case. Neither offender had prior convictions for violence or breaching bail, nor previously served terms of imprisonment. I subsequently concluded that David's and Jane's overall criminality was similarly serious. In this case I considered differences in present offending to be balanced by differences in past offending. However, Jane was sentenced more harshly than David - she received 18 months supervision (with the special condition to attend counselling) in addition to six months periodic detention, while David only received five months periodic detention (see Table 69). The domestic nature of Jane's offence may explain the additional sentence of supervision. The court may have thought she needed extra help and guidance. Both offenders were remanded on bail with special conditions (see Table 68). 2.2.4 Margo and Ben, Possession of an Offensive Weapon Margo (21 years, European) and Ben (24 years, Maori) were both convicted on one count of possession of an offensive weapon. Margo entered a plea of guilty to this offence but Ben's plea was not guilty. These were the only Class 6 offences in the statistical sample, so were included here even though the coded aspects of Margo's and Ben's offences were very different - suggesting Margo's criminality was more serious than Ben's. Margo had committed her offence alone when she physically injured a woman she knew. Ben acted with others against an unknown male victim who was not injured. Both offenders had 139 active or equal roles in the offence. Exploration of the crime stories confIrmed that Margo's offending was more serious than Ben's. Margo had originally been charged with wounding with intent to injure and possession of an offensive weapon, but the former more serious charge was discharged on the day of the trial because the victim refused to give evidence. In the early hours of the morning Margo was standing on a central city street corner. Margo was working at the time (sex worker). She was approached by another sex worker, the victim in this matter, who pulled up along side her in a motor vehicle. The victim got out of her vehicle, picked up a small stick and walked towards Margo. As the victim got closer she noticed a Police car approaching so dropped the stick and walked back towards her car. While the victim was approaching her car Margo had pulled a small knife from her handbag. She then used the knife to stab the victim in the back. After the stabbing they both began fighting and fell to the ground. The Police arrived and separated them. When spoken to by the Police Margo admitted using the knife, however claimed she just held it out to protect herself and said the victim "must have walked into it." As a result of the incident the victim received a stab wound to her back which narrowly missed her lung. The victim remained in hospital for three days. (Margo· Possession of an Offensive Weapon, Extracted from Police Summary of Facts) On the evening of the incident Ben and a friend (Troy) were drinking at a local bar. As the evening progresses a fight broke out between Ben, Troy, and another group of people. This opposing group left the bar only to be confronted by Ben and Troy in the car park. Ben was carrying a wheel brace while Troy was carrying a wooden batten. The victims ran back into the bar to telephone the police. Meanwhile Ben and Troy left the scene and were found by police a short time later. The wheel brace, and wooden batons were found in the boot of their car. When apprehended by police, Ben refused to make a statement. (Ben - Possession of an Offensive Weapon, an Extracted from Trial Transcripts and Judge's Sentencing Remarks) I also considered Margo's prior criminal history to be more serious than Ben's. She had 32 prior criminal convictions, including six for violence, and had served a prior term of imprisonment. Ben had nine convictions overall but none for violent offending and had never been imprisoned. Furthermore, Margo's last criminal conviction was less than six months prior to the current conviction; Ben had not offended in over three years but unlike Margo, he did have a prior conviction for breaching bail. The more serious nature of Margo's overall criminality led me to expect different judicial outcomes. However, disparity was concluded in this case because Margo received the same remand outcome as Ben (bail with special conditions) and a similar sentence: Margo was sentenced to twelve months imprisonment, suspended for 6 months, Ben received eight months periodic detention and twelve months supervision (with conditions to undertaking courses and counselling as directed by the probation offIcer) (see Tables, 68 & 69).29 29 Suspended sentences of imprisonment act as a final warning to offenders; re-offending within the specified period will, (in most cases), result in imprisonment. However, by itself this sentence impacts little on offender's liberty. Periodic detention and supervision on the other hand do impact on offenders' liberty. The former requires the offender to work at a detention centre once a week and in Ben's case the latter required regular visits with his probation officer. 140 2,3 Property Offences Section 6 of the Criminal Justice Act 1985 provides a general presumption against the use of imprisonment for selected property offences: "Where an offender is convicted of an offence punishable by imprisonment for a term of 7 years or less, the court shall not impose a full-time custodial sentence on the offender unless the court is satisfied that, because of the special circumstances of the offence or of the offender, any other syntence that it could lawfully impose would be clearly inadequate or inappropriate." In the current research, this presumption was relevant to all Class 3 property offenders (theft, fraud and receiving stolen property) (see Tables 70 & 71). Fifteen pairs of property offenders were studied to determine whether sentencing and remand outcomes were similar, different or disparate. The case studies results presented in Tables 70 and 71confrrm previous statistical analyses (see Chapters 3 & 4). It is shown that in comparison to women, men were generally 'disadvantaged' when it came to remand and sentencing. Table 70 summarises judicial remand responses for offenders by grouping them into five categories and presents details of remand outcomes in these categories. Results show that of the 15 pairs examined seven received disparate outcomes. In all cases, this remand disparity 'favoured' women. Difference was found in a further six cases and similarity in two. Table 71 summarises judicial sentencing responses for offenders by grouping them into five categories and presents details of sentencing outcomes in these categories. As seen from Table 71 of the 15 pairs assessed, eight received disparate sentencing outcomes. In all these cases disparity was found to 'favour' women. sentencing outcomes. Difference was recorded in the remaining seven 141 TABLE 70 m CATEGORIES OF JUDICIAL REMAND RESPONSES AND OUTCOMES FOR 15 PAIRS OF PROPERTY OFFENCES Remand Outcome CATEGORY 1 (D~PARITY) Ruth and Ricki Cindy and James CATEGORY 2 (D~PARITY) Bridget and Rob 70 d'!Y§i Ruth = remanded on bail with special conditions, Ricki = remanded in custo~fr Cindy = remanded on bail with standard conditions, James = remanded on bail with ~ecial conditions Bridget = remanded on bail with special conditions, Rob = remanded on bail with special conditions, 7 days in custodial remand Susan = Remand on bail with standard conditions, Brad - Remand on bail with special conditions conditions Vicki = Remand on bail with standard conditions, Jeff = Remand on bail with ~ecial Susan and Brad Vicki and Jeff CATEGORY3(D~PI) Glenis and Todd Sandra and Terrv CATEGORY 4 (DIFFERENCE) Glenda and Shane Annabel and Mitch Emma and Luke Jessica and Graham Ronda and Karl Violet and Wayne CATEGORY 5 (SIMILAR) Judyand Pete Sallvand Mike Glenis = remanded on bail with standard conditions, Todd = remanded on bail with standard conditions Sandra = remanded on bail with standard conditions, Terry = remanded on bail with standard conditions Glenda = Bail with standard conditions, Shane = Bail with special conditions, custodial remand for 20 days Annabel = Bail with special conditions, Mitch - Bail with special conditions, custodial remand for 14 d'!Y§i. Emma = Bail with standard conditions, Luke = Bail with special conditions conditions Jessica = Bail with standard conditions, Graham = Bail with ~ecial Ronda = Bail with standard conditions, Karl - Bail with special conditions Violet = remanded on bail with special conditions, Wayne = remanded on bail with standard conditions, 9 days in custodial remand. Judy= Bail with standard conditions, Pete = Bail with standard conditions Sally = 25 days in custodial remand, Mike = 18 d<i}'S in custodial remand Category 1 • men's and women's criminality was judged similarly serious but men received more severe remand outcomes. Category 2 • women's criminality was judged more serious than men's but men received more severe remand outcomes. Category 3 • women's criminality was judged more serious than men's but they received similar remand outcomes. Category 4 • men's criminality was judged more severe and men received more severe remand outcomes. r:!:lt.o.nnFU /:\ _ rnQ'~ TABLE 71 m anrl ulnrnan'e t"rirninaliht llIHle illrlna'" c:irnil.ar anrl thou r.ol"ahu:ui c:irnil.ar roman" nIJt,..nrnoc: CATEGORIES OF JUDICIAL SENTENCING RESPONSES AND OUTCOMES FOR 15 PAIRS OF PROPERTY OFFENCES Sentence CATEGORY 1 (D~PARITY) Ruth and Ricki Sally and Mike Cindy and James Judvand Pete CATEGORY 2 (D~PARITY) BridQet and Rob Sandra and Terry CATEGORY 3 (D~PARITY) Glenis and Todd Susan and Brad CATEGORY 4 (DIFFERENCE) Glenda and Shane Annabel and Mitch Emma and Luke Jessica and Graham Violet and Wayne Ronda and Karl CATEGORY 5 (DIFFERENCE) Vicki and Jeff Ruth = 8 months periodic detention, Ricki = 3 years irTlfll'isonment Sally - 10 months imprisonment, Mike - 18 months imprisonment Cindy = 12 months imprisonment suspended for 12 months, James - 3 months imprisonment Pete = 12 months imjJrisonment Judy = 24 months s~ervion, BridQet = 18 months imprisonment, Rob = 24 months imprisonment Sandra = $250 reparation, Ter~ = $250 r~ation and 4 months~eridc detention Glenis = 5 months periodic detention, Todd = 5 months~eridc detention Susan - $500 reparation, $75 court costs $600 fine, Brad - 100 hours community service Glenda = 5 months periodic detention, reparation Shane = 6 monthsJl..eriodic detention, r~ation Annabel-18 months sU(:lervision, Mitch -12 months imprisonment Emma = $500 fine, $65 court costs, Luke - 4 months imprisonment Jessica = 12 months supervision, Graham = 6 months periodic detention, $8, 500 r~ation. Violet = 4 months periodic detention, Wa'jrle = 6 months irTlfll'isonment Ronda = 6 months supervision, $2,000 reparation, Karl - 18 months imprisonment Vicki = 9 months imprisonment suspended for 2 years, 5 months periodic detention, $26,000 reparation, Jeff = 60 hours community service Category 1 • men's and women's criminality was judged similarly serious but men received more severe sentences. Category 2 • women's criminality was judged more serious than men's but men received more severe sentences. Category 3 • women's criminality was judged more serious than men's but they received similar sentences. Category 4 • men's criminality was judged more serious than women's and men received more severe sentences. Category 5 • women's criminality was judged more serious than men's and women received more severe sentences. Examples of case studies where male and female property offenders were judged as receiving disparate outcomes will now be presented. Included are the following offender pairs: Sally and Mike (Category 3 remand outcome, Category 1 sentencing outcome), Cindy and James (Category 1 remand outcome, Category 1 sentencing outcome), Bridget and Rob (Category 2 142 remand outcome, Category 2 sentencing outcome), Glenis and Todd (Category 3 remand outcome, Category 3 sentencing outcome) (see Tables 70 & 71). 2.3.1 Sally and Mike, Arson Sally (20 years, European) and Mike (21 years, European), both entered pleas of guilty to more than one offence count. Both offenders played active or equal roles and offended against organisations. Two of the coded crime elements suggested that Sally's offence maybe more serious than Mike's. First, Sally was the sole perpetrator, unlike Mike, who had a cooffender, which could have made him less blameworthy. Second, Sally's arson caused $9,600 damage, compared with $3,800 in Mike's case. Sally was convicted on two counts of arson, compared with Mike who was convicted on a count of arson and a count of burglary. According to statute and the Ministry of Justice's Seriousness of Offence Scale (1995), Sally's offending was more serious than Mike's. The maximum statutory penalty for burglary is ten years imprisonment and this offence has a seriousness score of 76.1, while arson carries a maximum term of 14 years imprisonment and has a seriousness score of 221. These initial thoughts regarding the more serious nature of Sally's current offending were further confirmed once the crime details were investigated. While this pair's ftrst count of arson was similar, Sally's second count of arson was more serious than Mike's burglary. Sally set ftre to a property adjacent to her boyfriend's house to "show her boyfriend how much she was hurting and to get some attention." This arson resulted in $6,000 damage. In comparison, Mike burgled a working men's club, taking alcohol and a cooked chicken; estimated value was $700. In the early hours of the moming Sally was in the grounds of a school. She went to a classroom at the rear of the school where there was a hole in the window. (She had broken the window the previous day). Sally struck several matches and individually held them against the curtain adjacent to the broken window. When the fire took hold she left the scene. The resulting damage did not spread far from the window and the damage was minor. Estimated of damage was $3,700. When spoken to by police, Sally stated she did it mainly to get some attention .... A couple of days after the first arson Sally committed her second arson. In the early hours of the moming Sally was at her boyfriend's house. She had consumed a large amount of alcohol and had become upset over the current state of their relationship. Sally decided to set a fire to the building next door to her boyfriend's house. She went to the rear of the house, and using a motorcycle helmet smashed a window and set fire to the curtain on the other side. Once the fire had taken hold she left the scene. When spoken to by police Sally said she did it to show her boyfriend how much she was hurting and to get some attention. Extensive damage was done to the building, which was an assessment centre for persons with psychiatric disorders. Estimated damage $6,000. (Sally - Arson, Extracted from Police Summary of Facts) In the early hours of the morning Mike and a co-offender walked to a school and talked about burning it down. They stacked rubber mats and a plastic bag up against some doors in the school and then both ignited the mats with cigarette lighters. Once the blaze took hold of the mats Mike and his co-offender left the area. Estimated damage was $3,500. One week after committing the arson, Mike committed a burglary. In the early hours of the morning, Mike and a co-offender went to a working men's club where they forced a door open with the aid of a screwdriver. Once inside they loaded kegs of beer, four dozen quarts of beer, a cask of wine, and a cooked chicken into their car: Reparation was sought for the food, alcohol and damaged door. Estimated value was $700. (Mike - Arson and Burglary, Extracted from Trial Transcripts and Judge's Sentencing Remarks) 143 Criminal history seriousness differed for this pan- of offenders. Sally was a flrst time offender while Mike had four prior convictions for petty drug and property crime. 30 Although Sally's criminal history may have been less serious than Mike's, her CUlTent crime was substantially more serious than his. Thus this pair's overall criTninality was considered to be siTnilar. Mike and Sally received similar remand outcomes (Sally was remanded into custody for 25 days and Mike for 18 days) but sentencing outcomes were considered disparate because Mike was sentenced to 18 months imprisonment while Sally only received ten months (see Tables 70 & 71). 2.3.2 Cindy and James, Receiving Stolen Property Cindy (38 years, European) and James (25 years, European) each entered pleas of guilty to a count of receiving stolen property. Both offenders acted alone, had primary or equal roles in the offences and the property received belonged to commercial organisations. Property received by Cindy was valued at $3,000, compared to $350 in James's case. Thus differences in property value situated James's offending, compared with Cindy's, at the lower end of the scale. 31 Further exploration of this pan-' s offences provided further evidence of Cindy's higher criminality. Cindy had received computer equipment taken from a local school while James had received a power tool originally lifted from a hardware store. It could be argued that school premises are more likely to evoke judicial sympathy because they are more vulnerable than a commercial business, thus making Cindy's act more serious. Furthermore, all property in James's case was recovered because he was caught when he tried to sell the power tool to a second hand dealer, but only half of the computer equipment was recovered from Cindy, suggesting she had sold the other half. Both offenders told police that they were unaware the property was stolen. A school was broken into and computer equipment stolen. The following week a search warrant was issued on Cindy's house and half of the computer equipment stolen from the school was recovered by police. In explanation Cindy claimed the computer belonged to an ex-flat mate who had subsequently moved overseas. Cindy said she had been holding on to the computer, as this person owed her money for rent. (Cindy - Receiving Stolen Property, Extracted from Police Summary of Facts) A power tool was shoplifted from a hardware store in Christchurch. A week later James took the stolen tool into a Christchurch second hand shop and sold it. He gave a false name and address when selling the tool. When interviewed by the Police, James claimed he had been given the power tool by a person he met in the pub. (James - Receiving Stolen Property Extracted from Police Summary of Facts) His last criminal conviction was over 12 months ago, he had no prior convictions for breaching bail and had never been imprisoned in the past. 31 The Ministry of Justice's Seriousness of Offence Scale (1995) confirms this: receiving stolen property valued between $100·$1000 has a score of 13.3 while receiving property valued at over $1000 has a score of 32.5. 30 144 I subsequently considered Cindy's current criminality to be more serious than James's because of higher monetary amounts, the fact that only half the property was recovered, and that the victims were school children. Conversely, James's criminal history was more serious; overall he had more prior convictions than Cindy (16 compared to 27) and while they both had prior convictions for property offending, James had 13 as opposed ten in Cindy's case. It was six months since Cindy's last criminal conviction and four months since James's. Balanced against the more serious nature of Cindy's current offence, I judged the overall criminality of these offenders to be similar. But James's judicial outcomes were more severe: James was remanded on bail with special conditions and sentenced to three months imprisonment, Cindy was remanded on bail with standard conditions and given a suspended sentence of imprisonment. Thus disparity was concluded in this case because it was not immediately clear why James was treated differently from Cindy (see Tables 70 & 71). 2.3.3 Bridget and Rob, Arson Bridget (32 years, European) and Rob (30 years, European) both entered pleas of guilty to one count of arson. These offenders acted by themselves, set fire to the property of people they knew and caused around $6,000 damage. Bridget and Rob had extensive criminal histories but Bridget's was more serious. It was only three months since Bridget's last criminal conviction; she had a total of 42 prior convictions including six for arson, two for breaching bail and had served two prior terms of imprisonment. But in contrast, Rob's last conviction was over 12 months ago, he had 23 prior convictions, including one for arson, none for breaching bail and had never received an imprisonment term. Further exploration of the crime details revealed that Bridget's arson was in fact marginally more serious than Mike's. Both arsons were, to a certain extent, acts of revenge; Bridget and Rob were trying to 'get back' at people who had upset them. In contrast to Rob, Bridget's arson had endangered people's lives. 32 When questioned by police Bridget's and Rob's explanations were similarly vague and confused. Rob could not remember committing the crime and Bridget said she remembered the arson but could not stop herself from doing it. A couple of weeks before the incident Bridget was evicted from a hostel by her landlord. She was forbidden to return to the hostel but one evening she returned to visit a friend who lived there. During the visit she went to hostels lounge room and set a fire in a cupboard. The fire spread up the walls of the lounge. When the fire service arrived smoke was pouring out of the lounge area and had also gone up to the first floor where a family with two young children resided. The ambulance was called to treat these people for smoke inhalation. When spoken to by police Bridget admitted lightening the fire. In explanation she stated that she did not know why she had done it, she just could not stop herself. (Bridget - Arson, Extracted from Police Summary of Facts) The Ministry of Justice's Seriousness of Offence Scale scores Arson which endangers life higher (score=360) than that which does not (score=221). 32 145 Rob and his wife (Judy) separated after having been married for a number of years and she purchased a property for herself and her new partner. Rob became aware that his ex-wife was living in a de facto relationship. He went around to Judy's house and a verbal dispute occurred between them concerning Judy's new relationship. The following day Rob contacted his wife by telephone. He said he loved her, wanted her back and could only live if he knew one day he might get her back. Judy told him that they would never get back together and hung up on him. She then went out shopping. While she was out Rob went to her house and set a fire in the garage. Neighbours rung the fire service and moderate damage was caused to the address. When spoken to by police Rob said he didn't know if he had gone to his ex-wile's house. (Rob - Arson, Extracted from Police Summary of Facts) I decided that Bridget's overall criminality was thus more serious than Rob's because of the extent and seriousness of her criminal history and the fact that the current crime endangered people's lives. However, Bridget's judicial sanctions were not as harsh as Rob's. Bridget was remanded on bail with special conditions while Rob was remanded in custody for seven days in addition to a period on bail with special conditions. Both offenders received a sentence of imprisonment but Bridget's imprisonment term was shorter than Rob's: 18 months as opposed to 24 months (see Table 70 & 71). 2.3.4 Glenis and Todd, Receiving Stolen Property Glenis (32 years, Maori) and Todd (33 years, European) entered pleas of not guilty to a count of receiving stolen property; both were sole perpetrators. Property recovered from Todd' ($1,500) had a higher value than that recovered from Glenis's crime ($870). In both cases, the property received belonged to individuals rather than organisations. Court records noted that Glenis and Todd sold a proportion of the stolen goods. Apart from coded differences in property values (making Todd's offence marginally more serious), this pair's crimes were very similar. The day before the offence a dwelling house in Christchurch was burgled and property, including jewellery and compact discs, were taken. The following day Glenis went to a second hand record shop and attempted to sell the compact discs that had been taken in the burglary. Many of the compact discs were rare collectors items and when questioned by the store manager about where she had got them from Glenis became nervous, ran from the shop and left some of the compact discs behind. The store manager subsequently called the police. After lifting fingerprints from the compact discs the police were able to establish the identity of Glenis and a search warrant was executed on her house. The remaining compact discs were not recovered. (Glenis - Receiving Stolen Property, Extracted from Trial Transcripts and Judge's Sentencing Remarks) Six months prior to the offence a motor vehicle was stolen from a Christchurch address. A search warrant was executed at Todd's home address. The stolen motor vehicle was located in the garage at the rear of Todd's property. Minus its engine and its wheels. When spoken to by police Todd claimed that he has answered an advertisement in the local newspaper. He claimed a person came to his home with the car on a trailer and he paid $500 cash for the car without its motor and wheels. Enquires with the local newspaper established that there was no advertisement for the car or the body as outlined by Todd. (Todd - Receiving Stolen Property, Extracted from Trial Transcripts and Judge's Sentencing Remarks). 146 Todd's present offence may have been marginally more serious than Glenis's but her prior criminality was substantially worse than his. Glenis had 30 prior convictions including 28 for property offending, her last criminal conviction was only seven months before her current conviction and she had served two terms of imprisonment in the past. In contrast, Todd was virtually a first time offender; he had one prior conviction for theft but this occurred nearly seven years before to his current conviction, for which he received a non-custodial sentence. I thus considered Glenis's overall criminality to be more serious than Todd's and expected more severe judicial outcomes in her case. Yet both offenders were remanded on bail with standard conditions and each received a sentence of five months periodic detention (see Tables 70 & 71). 3 Closing Comments Coding schemes used in statistical research designs have been criticised for being limited in their ability to capture the full character of each individual case (Conley & O'Barr, 1987; Crew, 1991; Daly 1994; Triggs, 1999: 123). In order to address this, I have examined the crime details or stories for 50 matched pairs of offenders (i.e. male and female) to determine whether my prior statistical patterns that showed sex differences in sentencing and remand outcomes (see Chapters 3 & 4) might be confirmed or otherwise. The case-study results presented in this chapter confirm and strengthen my statistical fmdings, while also illustrating the way in which disparities are manifested. In some cases, differences in men's and women's judicial outcomes were explained by differences in current and past criminality, while in others similar judicial outcomes resulted from similar criminality. However, disparate outcomes resulting in 'leniency' for women remained evident in the majority of cases, with men either receiving different or more severe punishments than women, for criminality of comparable seriousness, or the same punishment as women for less serious criminality. Thus far, the framework used to assess the case-study group of offenders has been based on equal treatment assumptions by asking whether like criminality resulted in like judicial outcomes. In most cases, the answer is no. 33 However, sex differences in other personal or social circumstances were excluded from this part of the case-study analysis. Such information is found in the pre-sentencing reports and will be assessed in the following chapter. Judges' sentencing remarks and pre-sentencing reports will also be investigated to see how presentencing recommendations and actual sentencing outcomes are rationalised for men and women. 34 This shifts the analysis to the question of how the criteria used for deciding judicial 33 It could be argued that no two cases are the same and that my attempts to draw comparisons are therefore misguided but the analyses was conducted using factors recognised as important in judicial decision-making (Ashworth, 1987: 26). 34 Both Judges' sentencing comments and pre-sentencing reports are specifically concerned with sentencing. No documents of this kind exist for remand decisions so these cannot be assessed further. 147 outcomes are gendered (i.e. are different factors considered when determining men's and women's judicial outcomes, and are certain 'types' of men and women more likely to be extended judicial leniency?) (see Chapter One). Thus we seek to know not only why men and women receive different judicial outcomes, but also whether sex differences in the decisionmaking process can explain this. The next stage of the case-study analysis will therefore be useful because 1) a more subtle, complex and meaningful account of the judicial processing will be obtained and, 2) it will enable us to link current research findings more strongly with societal processes (particularly gendered processes) and with criminal justice processing. 148 Chapter Seven Pre-Sentence Reports, Judges' ntencing Remarks, and Gender Prior statistical analysis showed that although some sex differences in judicial outcomes could be explained by sex differences in criminality (e.g. seriousness of criminal history), women often received more 'lenient' judicial outcomes than men (see Chapters Three & Four). In the previous chapter, these statistical patterns were confirmed using a case-study analysis, which compared the criminality of male and female offenders (see Chapter Six). Furthermore, the current research has produced statistical evidence that the decision-making process differs for men and women, with prior results showing that different factors were sometimes considered when determining men's and women's judicial outcomes. Social rather than criminal or legal variables were more important for women, while the opposite was generally more true for men (see Chapter Five). The purpose of the current chapter is to analyse Probation Officers' presentence reports and Judges' sentencing remarks, in order to 1) obtain a more subtle, complex and meaningful account of judicial processing and, 2) enable links to be made between research fmdings, societal processes (particularly gendered processes) and criminal justice processing. Understanding how or why there were sex differences in judicial outcomes and the decisionmaking process, is the primary aim of this chapter. In the statistical analysis presented in earlier chapters, gender related statuses such as familial situation, income, health and negative life experiences were coded, measured and controlled for both male and female offenders. We have seen that these factors impacted on judicial decisions in some cases. The difficulty with the coding schedule used in the statistical analysis, however, was that it only recorded the presence or absence of particular gender statuses. This could be problematic because the construction of discussion around these statuses may have varied, left different impressions about an offender, and potentially impacted on sentencing. The cases of James and Kate illustrate this point well. In my statistical analysis both were coded as soleparents, but as the probation reports show, the importance of this parenting was only emphasised for Kate. James is a solo parent who looks after his two year old daughter. (James, Violent Offence - Probation Officer's Pre-Sentence Report) 149 Kate appears a capable and committed parent toward her three children. The drawn out period of the charges has placed considerable stress on her and her family and a sentence of imprisonment would further add to this situation. She intends on enrolling her latest child at school which will entail Kate being involved in the school. (Kate, Violent Offence - Probation Officer's Pre-Sentence Report) To achieve a comprehensive explanation for how men and women come to receive different judicial outcomes we need to have a more subtle, complex and meaningful account of judicial processing. We need to make links with societal processes (particularly gendered processes), criminal justice processing and criminological theory (e.g. social control, social cost, paternalism). This is best achieved by analysing 'discourse' (Spencer, 1984: 208).1 Such an analytical approach recognises the need for an understanding of how gender, as opposed to sex, 'gets done' for both men and women. Sex is recognised as important "to the extent that it gains meaning from the social, cultural or historical context within which it is placed" (Odubekun, 1992: 355). To achieve a structural understanding of gender, this analysis uses a conceptual framework grounded in understandings of gender in feminist thought, although it is understood that there is no common feminist perspective. Thus, as Daly & Chesney-Lind (1988: 504) point out, it is recognised that: It Gender is "a complex social, historical, and cultural product; it is related to, but not simply derived from, biological sex difference and reproductive capacities"; It "Gender and gender relations order social life and social institutions in fundamental ways"; It "Gender relations and constructs of masculinity and femininity are not symmetrical" or universal, but are based on an organising principle reflecting the superiority of some groups of men over others and over women (Daly & Chesney-Lind, 1988: 504). gender focus goes beyond women and sexism. The Not only are the power relations between men and women important, but so too are those between different groups of men (Law, Campbell & Dolan, 1999, Collier, 1998). It Systems of knowledge are gendered. This chapter looks at the impact of gender in the presentations of offenders in sentencing decisions. (Unfortunately, remand outcomes cannot be considered because of limited file documentation). Pre-sentence reports and Judges' sentencing remarks are examined for the same 1 It needs to be pOinted out that the analysis of discourse in this chapter is by no means as thorough as that produced by Allen (1987b). Eaton (1986) or Worrall (1990) whose analyses were the subjects of entire books. The purpose here is simply to gain further insight into the operation of gender in criminal justice decision-making. 150 case-study sample (100 individual offenders or 50 pairs) used in the previous chapter (Chapter Six). As we have seen, this case-study sample was taken from, and is representative of, the original wide sample (n=388) used in the statistical analyses. 2 Furthermore, through the course of this research I have read the pre-sentence reports and Judges' sentencing remarks for all 388 offenders. While the focus and examples given in the present chapter are from the case-study sample, I can say with certainty that the results presented are an accurate reflection of the entire sample. Both the pre-sentence reports and Judges' sentencing remarks are viewed as records of exchanges between offenders and other judicial actors, as sites where cultural ideals of masculinity and femininity are constructed or, quite simply, where gender is 'done'. I therefore conceptualise gender as something that is actively constructed rather than as a role, stereotype or ideal that is passively incorporated by individuals. For example, the exchange between offenders and Probation Officers is just that, an exchange or interactive process whereby gender is constructed by both the offender and the Probation Officer. When analysing these reports I not only consider what is said but what is not said. I ask if . information is screened or presented in contradictory ways, according to the gender of the offenders (Reed & Thier, 1981: 238; Worrall, 1990: 9). It is hoped that this perspective will provide a structural understanding of gender and criminal justice processes which may in turn explain not only why men and women receive different sentencing outcomes but why there are sex differences in the decision-making process, as has been suggested by numerous other authors (e.g. Kruttschnitt, 1982a; Kruttschnitt, 1982b; Kruttschnitt, 1984; Kruttschnitt & Green, 1984; Eaton, 1986; Allen, 1987a; Allen, 1987b; Allen, 1987c; Eaton, 1987; Jackson & Smith, 1987; Mair & Brockington, 1988; Worrall, 1990; Raeder, 1993; Pearson, 1997; McIvor, 1998; Heberle, 1999). Pre-sentence reports and Judges' sentencing remarks are analysed, and presented separately in the following discussion. 3 Unlike the previous chapter, (which looked at cases for pairs of offenders), the current results are ordered and presented according to analytical themes. This is Refer to Chapters One and Two for discussion about this sample and an outline of sample characteristics. It is important to note that the sex of Probation Officers and Judges may impact on sentencing. However, male Judges sentenced all offenders in this study and the sex of Probation Officers could not be recorded because it was difficult to establish from the court's trial files. 2 3 151 done to protect offenders' identities and ensure their privacy. Names in this section are also different from those previously assigned. Offenders' crimes are only identified as either violent, property or drug and criminal histories are discussed in general terms. This ensures offenders' privacy and identity by making it difficult to link crimes, criminal justice proceedings and personal circumstances. Other minor personal details were also changed such as the sex of dependent children, places, dates, and lawyers' names. Before this analysis can take place it is important to understand if, and how, gender operates in contemporary New Zealand society. It is recognised that doing justice to, and developing an understanding of this, is an enormous task. 4 However, it is imperative that a general overview is provided because the criminal justice system is an institution operating within, and thus affected by, gender constructions in the wider social context. 1 New Zealand - Our Gendered Society? In New Zealand, female culture and dominant ideas about femininity are typically linked to a woman's place in the home; her unpaid work, fulfilment of her family's needs both materially, through house work; and psychologically, through emotional support. Women are held primarily responsible for the success and happiness of the family unit and are required to manage emotions and distress of family members. Situating women as the 'experts' on relationships, feelings and emotions also ensures that as individuals, they are seen to exist within a realm of emotionality. A woman's position in the private sphere is ideally that of a stay-at-home mother, usually economically dependent either on the state or more preferably on a man (see for example Park, 1991; Briar, 1992; McKinlay, 1992; James & Saville-Smith, 1994; Thompson, 1998; Thompson, 2000). When women are involved in paid work it is more likely, than is the case for men, that work will be 'fitted around' child care and other familial responsibilities (Novitz, 1987: 46). Women in paid employment earn less than men, on average, irrespective of industry, occupational group or level of qualification. They also lack widespread representation in high status occupations, are more likely to work part-time, and to be found in traditionally female employment with low pay rates such as teaching (primary), nursing, machinists, personal Refer to the following sources for more in-depth discussions of gender in New Zealand: Brookes, Macdonald & Tennant (1986), Cox (1987), Phillips, (1987). Park (1991). Briar, Munford & Nash (1992), Statistics New Zealand (1993), James & Saville-Smith (1994), Du Plessis & Alice (1998), Daley & Montgomerie, (1999), Law, Campbell & Dolan (1999). 4 152 assistants/private secretaries to (usually) male managers and clerical/sales positions (Novitz, 1987: 37; Statistics New Zealand, 1993: 113-115). Regardless of women's involvement in paid public work, they still remain responsible for the majority of unpaid work within the family home. Furthermore, the types of unpaid private work women do differ from men. Women tend to spend more time cooking, cleaning and caring for children while men do repairs, maintenance and garden work (Novitz, 1987: 45; McKinlay, 1992; Statistics New Zealand, 1993: 100). Women also are involved in more voluntary unpaid work outside the home (Statistics New Zealand, 1993: 100-102). This voluntary work is usually of a caring nature; an extension of familial duties such as caring for elderly family members, participation in community projects involving children and negotiation with schools (Craig, 1992: Ill). Women's roles as caregivers both within and outside the family in New Zealand result in a particular construction of femininity. Women's alleged nurturing and maternal capabilities are emphasised through a 'cult of domesticity' (see James & Saville-Smith: 1994) which in turn structures women's lives as dependent and privatised (Craig, 1992: 108).5 Historically, the 'cult of domesticity' emerged in New Zealand during the 1880's and 1890's. This was a time of social disorder and unrest in which the professional middle classes and landed gentry felt threatened. 6 The state responded by actively constructing women as the instruments through which social order could be imposed on men and children via the sanctity of the family (Phillips, 1987: 50-51; Craig, 1992: 108; James & Saville-Smith, 1994: 28; Olssen, 1995; Dalley, 1999: 63; Olssen, 1999). Thus, paradoxically, it is within the family that women gained a certain degree of power (James and Saville-Smith, 1994: 54-55). Social control was actively re-established through the promotion of a gendered culture and a particular construction of femininity that presented women as 'moral redeemers'. While being economically dependent on men, women were seen as morally the more responsible sex. The 'cult of domesticity' has since remained a powerful discourse in contemporary New Zealand and renders women both powerless and powerful. The New Zealand ideology of the 'cult of domesticity' is similar to the American 'cult of true womanhood' which "extolled the virtues of piety, sexual purity, submissiveness, and domesticity" and required that "women dedicate themselves to the private sphere, nurturing and serving their families and transforming their homes into a Haven in a Heartless World" (Kerber & De Hart·Mathews, 1987: 15). 6 Violence, drunkenness, gangs of street kids, theft, gambling and 'idleness' were seen as endemic (James & Saville-Smith, 1994:28, see also Phillips, 1987: 44-80). Dalley (1999: 63) notes; "Nineteenth century New Zealand was a dangerous place. Contemporaries saw danger, especially of the moral kind, in many places and in many guises. Alcohol, prostitution, venereal disease, gambling, intemperance, and desecration of the Sabbath headed a IisLAs in the Old World, most peril lurked in New Zealand's burgeoning towns and cities ... Discourses of danger abounded, heightening and reproducing a pervasive sense of moral danger that focussed on sexuality and gender relations". 5 153 In New Zealand, the portrayal of dominant masculinity and appropriate roles for men stand in contrast to those of women. They are also defined, in part, by familial connections or lack of them, and work. James and Saville-Smith (1994) and Phillips (1987) identify two opposing yet complementary masculinities in New Zealand society, the 'man alone' and the 'family man'. Both masculinities allow independence for men in the public sphere of paid work. According to James and Saville-Smith (1994) and Phillips (1987) the 'man alone' identifies himself through male 'mateship', 7 rather than familial ties and represents a threat to social order. This was particularly so during the 19th century, a time when many men flocked to gold fields and sheep stations to find paid labour without their families. Through mateship these groups of men developed unity, but their "disorderly reckless habits" were at odds with capitalism and bourgeois ethics of hard work, savings and discipline (Phillips, 1987: 49-50). Subsequently, the concept of 'family man' was actively promoted by the State as a way to socially control the 'man alone'; constructing the 'family man' became "instrumental in defining a less disruptive role for men" in New Zealand society (James & Saville-Smith, 1994: 37). For example, allowing women the political vote early in this country (1893) is often considered by New Zealanders as a "shining example of our free and liberal traditions", but parliamentary debates during this period show little concern for sexual equality. Indeed, Parliament hoped that the political power of the 'family man', over the 'man alone', would be increased by allowing women to vote because the 'family man' would get two votes (his and his wife's) and "so counteract the influence of those men who have no abiding place in the colony." Kate Sheppard herself (the leading female activist behind the fight for New Zealand women's right to vote) argued that giving women the vote "would strengthen the Home vote-that is the vote of the more settled and earnest minded part of the community" (Phillips, 1987: 53-54). Despite state initiatives, mates hip continued to be an integral part of New Zealand culture so what was required was a male role that was both domesticated and mateship orientated. James and Saville-Smith, (1994: 38) note that out "of this imperative emerged a masculinity in which the 'Man Alone' was preserved and celebrated, but subordinate to the virtues of the Family Man". Although both the 'family man' and the 'man alone' were meant to be hard working, dependable, pragmatic, self-reliant and loyal, only the 'man alone' could resist authority. The 'family man' had to accept the regulation imposed by waged work because he needed to 154 provide economic support for his wife and children. He needed to be the 'breadwinner' and a 'good provider'. The maintenance of the dominant social order thus required primacy be given to the 'family man' over the 'man alone' and to varying degrees throughout New Zealand's history the State actively promoted the former masculinity;8 society aided its construction and men have fiercely protected (usually from female encroachment into paid work) their 'right' to economically provide for their families (Novitz, 1987: 37-42; Saville-Smith, 1987; James & Saville-Smith, 1994: 39; Andrewes, 1999; Frank, 1999). In contemporary New Zealand society, the public domain of waged work is still men's territory. Employment commitments 'keep men in line' and are thus an important source of social control. The 'family man' is still evident and men's roles within the family generally remain that of the major breadwinner and provider. Even in families where women work, men tend to earn more money and women's paid work is generally seen as supplementary to men's. Sports clubs (particularly rugby football and league clubs), working men's clubs, gang 'pads' and pubs 9 remain venues in which men can celebrate and maintain mateship. They are places where men can be 'Alone'. Constructions of masculinity in New Zealand continue to situate men as actors in the public sphere with male culture being shaped by economic independence and mateship maintenance outside the home. In contrast, female culture is situated in the private sphere and shaped by economic dependency and relationship maintenance within the home (James & Saville-Smith, 1994: 49-50). 2 The Construction of Gender in Probation Officers' Pre-Sentence Reports In New Zealand, pre-sentence reports are compiled by Probation Officers at the request of a Judge, after an offender has been convicted but before sentencing has taken place. Legislation for pre-sentence reports is found in the Criminal Justice Act (s. 15) and dictates that pre-sentence reports are only required where an offence is punishable by imprisonment. In Friendship between men. "Attracting men into the family required more than the benefits of having an unpaid domestic and sexual servicer in the home, and the co-optation of qualities associated with the Man Alone. It required direct financial incentive. The State's family policy has, since the late nineteenth century, actively redirected income from men and women outside nuclear family structures to men with dependent wives and children. Agricultural settlement and urban housing polices have traditionally favoured married men with dependants, and have been facilitated by the State's accumulation of Maori lands, either through confiscation or appropriation. Taxation and welfare have similarly been orientated towards supporting the Family Man ... in the private sector, award rates for waged and salaried men have advantaged the married over the single" (James & Saville-Smith, 1994: 39). 9 Bars or hotels. 7 8 155 these cases, a Judge may request a Probation Officer to report to the court on the "social circumstances, and (where appropriate) the personal history and personal characteristics, of the offender," and to give advice about sentencing alternatives. This type of information may also be conveyed to Judges by prosecutors, defence counsel and offenders,10 but pre-sentence reports are unique because they are construed as an independent and authoritative source of information provided by 'experts' (Wilkie, 1993: 8-9). Compared to other information sources, this idea of 'expert information' presents Probation Officers' information as 'less biased'. However, since most of the information gathered for pre-sentence reports is 'extra-legal' the process of compiling a pre-sentence report is potentially laden with bias (Reed & Thier, 1981: 233). The pre-sentence reports used in the present study provided a formal written account of the rationales used when Probation Officers made decisions. These reports are not "literal descriptions of the actual events or lines of reasoning that produce decisions" (Drass and Spencer, 1987: 278), but remain a worthy topic of study because preconceptions about gender may influence sentencing outcomes. After all, sentencing Judges are not privy to the process of writing a pre-sentence report. All that is of concern to them is the written information presented by the Probation Officer. The pre-sentence reports used in the cases I studied appeared in a standard format and usually included discussion or notes under the following topic areas: III Sources of Information CD Family • Employment and Finances CII Health CII Factors relating to the current offence • Responses to previous sentences and criminal history CII Sentencing recommendations 10 Comments from these three groups of judicial actors were not fully assessed by this research because an ethnographic method was not used. However, prosecutor, defence counsel and offenders statements were often referred to by Judges in their sentencing remarks and at this level, are assessed later in this chapter. 156 The pre-sentence report is compiled after interviewing an offender but in addition to this, a Probation Officer will often seek information about the offender from other sources. These other sources may include: family members (spouses, ex-spouses, siblings, children, parents) health professionals (counsellors, psychologists, psychiatrists, general practitioners, drug rehabilitation centre staff), other criminal justice personnel (periodic detention centre wardens, lawyers), criminal justice records, employers, teachers, pastors and budget advisers. 2.1 Female Offenders and Domesticity In the pre-sentence reports, the majority of women offenders were easily portrayed as situated within the realm of domesticity because their lives were rooted in financial dependence on men or the state and they were primary care givers of children. However, many women challenged this ideal of femininity. For some, the ability to fulfil familial needs both materially through, 'keeping house' and psychologically, through emotional support was difficult to reconcile with the drug addiction, crime, violence, and material deprivation which characterise their households. For others, primary responsibility for childcare or financial dependence did not feature in their lives. Worrall (1990: 31) notes that crime is perceived as primarily a masculine activity. Since the law and criminal justice system are also male dominated, the female offender is seen to challenge notions of acceptable femininity. Female offenders are, therefore, seen either as 'not real women' or as 'not real criminals'. Many women who break the law are also wives or mothers, or otherwise express acceptable feminine attributes such as dependency. In these situations, women's offending can be easily reconciled within dominant discourses of femininity. However, there will be cases where this is difficult because women's lives do not slot easily into ideas about acceptable femininity. Julie, from one case-study example, did not have children to care for but did own a number of dogs. The relationship between Julie and her dogs was emphasised throughout the pre-sentence report in terms of the worry they caused her and the care she gave them. Julie lived independently, in her own flat, but the pre-sentence report presented this as something that had been thrust upon Julie against her will. Julie's partner had been imprisoned and while 'coping with her independence', Julie was noted to be uncomfortable and clearly distressed by it. She was thus 'refeminised' as dependent on her husband after all. The pre-sentence report reads: Her partner received a prison term six months ago. She depends on this relationship to a very great extent, and visits him three times weekly as a matter of priority. She takes the care of her three dogs very seriously, she has raised them from birth. She says one of these dogs is so disabled that 157 she is obliged to carry her everywhere ... lt seems that with her partner in prison she has coped with her independent living situation with a regime of regular visits to him. (Julie· Violent Offence) Julie's offence is defined in New Zealand Statute as one of serious violence. She is a serial offender with a long list of criminal convictions, including many for previous violence. The pre-sentence report, in presenting her as fundamentally domesticated, dependent, and caring, minimised her extensive and serious criminality. Her criminal character was downplayed and she was 'refeminised' in terms of the traditional image of womanhood. Another example is Sharon, who was convicted of selling Class B drugs in a sophisticated operation. She had over 100 prior criminal convictions, many of which were drug related, and she could legitimately be described as a 'career criminal'. But the pre-sentence report points away from this fact. Instead she is described as a 'good mother', and her drug dealing is said to have arisen out of a benevolent desire to help her friends, who are intravenous drug users. Thus her drug dealing is recast in terms of her 'womanly' role as carer and nurturer. She is a good mother and someone who likes to help people but it often backfires on her... several of her close friends use drugs. She was able to purchase some surplus morphine from a friend. It appears that she bought them in order to pass them on at the same price and thereby help her friends avoid resorting to prostitution to support their habits ... Her interests centre around parenting her sons and caring for her dogs. (Sharon· Drug Offence) In three similar cases, Sarah, Joanne and Ruth were all childless, without male partners and in conflict with other family members. Sarah is described as openly independent. During their interviews with the Probation Officers, these women are depicted as wanting to be involved in family relationships. In fact, their inability to achieve domesticated femininity is presented as the reason behind their predicament and the Probation Officers recommend sentencing should include instruction, guidance or supervision to help with 'family identification'. This 'refeminisation' process detracted from the seriousness of their current and past criminality. Sarah was convicted of a 'serious violent offence' and had a long history of violent offending. Joanne had prior criminal convictions and her property offence was at the more serious end of the statutory scale. In contrast, Ruth was a first time offender. There appears to be a second factor in her offending. While initially Sarah's attitude was one of independence - almost bravado, in the course of the interview she made some quite revealing comments regarding her family involvement. After the death of her parents, her brother was given guardianship of the youngest children, Sarah returned home to help care for the family ... during this period, within a setting of immediate and extended family, Sarah stayed off drugs and alcohol. However, her brother 158 eventually moved to another county. Rather than separate the children it was decided that both should accompany him. Sarah was back in prison within months. It is difficult not to conclude, that in spite of her strong assertions of independence and responsibility for her actions, that this loss, has contributed substantially to Sarah's aimless lifestyle. (Sarah - Violent Offence) She has difficulty in relating to people, and finds it hard to trust or confide in anyone. She is pessimistic about the prospect of developing any close relationships, particularly with members of the opposite sex, and she reports that in such situations she becomes extremely anxious about the prospect of separation and her behaviour deteriorates accordingly. Further probing revealed that she has a very poor self image ... she expressed regret about the deteriorating relationship which she had with her mother and expressed a strong desire to overcome her difficulties with relationships by way of treatment. (Joanne - Property Offence) It would seem that Ruth has had to largely fend for herself and by her own accounts has had difficulties establishing meaningful relationships with family members ... she seems to be lacking guidance and structure in her life. I consider a term of Supervision could provide this ... (Ruth - Drug Offence) 2.2 Female Offenders and Family Pressures Identifying the family, or lack of one in Sarah's case, as the source or cause of women's offending or the reason for substance dependence, which in turn links with criminality, was common in women's pre-sentence reports. This fmding was similarly identified in Worrall's research (1990:60) where she found "domestic problems were seen to explain or excuse female crime (which, of itself, was assumed to be unnatural)." These constructions were another way to 'refeminise' women, to deal with the contradiction between criminality and femininity, to reconstitute criminal women as 'real women' via their familial relationships. The impact of familial problems on women's offending was presented in the presentence reports as either direct or indirect. Worrall (1990:60) explains that women "might be reduced to breaking the law either directly by insufferable husbands (,Women aren't naturally criminal - it's the men that force them into it'), or indirectly by the pressures of family life." In the next three cases, we see familial relationships being used in a direct way to construct women as not entirely responsible for their actions. In all these cases responsibility or blame is essentially removed from the offending women and placed on the men in their lives. In the first case, Jill's husband is constructed as the 'real culprit' - the 'insufferable' husband. To a certain extent, this removed Jill's criminal responsibility because she was merely "embroiled in the circumstances surrounding her partner's lifestyle." Jill is currently caring as a sole parent for her five year old child and six month old baby. At the time of the offence she was living in a de facto relationship with Glen (the co-offender). Jill states that Glen was addicted to intravenous drugs but generally maintained his addictive lifestyle outside the home. However, his lifestyle has caused significant stress for Jill and she described the circumstances of this offence as the last straw. Jill says that since her arrest she and Glen have separated although she still has regular contact with him ... Jili presents as a person who has previously been embroiled in 159 circumstances surrounding her partner's addictive lifestyle. However, she is now distancing herself from him and that lifestyle; she has successfully addressed her own drug difficulties and is coping well with the demands of parenting. (Jill - Drug Offence) Maria's drug involvement is held to be directly linked to her ex-partner; a relationship which is noted to "dictate her actions." Maria lives in her home with her two children. She is in contact with the children's father but no longer considers that she is in a relationship with him. However, her links with him in terms of involvement in the drug-subculture are still evidenced and, to a certain extent, dictate her actions ... Maria's involvement in the drug sub-culture stretches back many years. She has not entirely severed her links, the strongest of which appears to be the father of her children, who visits regularly and shares his drugs with her. (Maria - Drug Offence) Sally's victim was her ex-partner, Brad, but even here Brad is held responsible for Sally's actions and indeed for his own victimisation. In this case, the victim-offender relationship is recast with Brad being presented as the offender and Sally as the victim. During the past six years Sally has been in a relationship with Brad (the victim). She described Bad as jealous, possessive and violent and related instances of Brad's violence. In interview Sally said she had tried to end the relationship at various times, but usually took Brad back after harassment. Sally stated she finished with Brad six months ago ... On the day of the offence Sally went to Brad's house to tell him to stop hassling her. On leaving Brad's house she saw a present she had previously given him inside his garage, upset, she set it on fire ..... there is some concern concerning her relationship with Brad ... her relationship appeared to be negative yet Sally had in the past, failed to finish completely with him. (Sally - Property Offence) The following cases illustrate how women's criminality is explained and excused by their caregiving role, by indirect familial pressures from families of origin, or current familial circumstances. The result is that familial circumstances are seen to act on women who are, in turn, presented as having relatively little control over their criminality. Pam's and Bridget's offending was excused because it occurred in the course of their care-giving activities. Rather ironically, Pam is noted to have good domestic skills and it is argued that her drug dealing occurred with the best of intentions. Bridget's intentions are also presented as caring and honourable. She offended to lighten her "husband's burden, to bring down his stress load." Present matters seem to have arisen from her role as the central carer to both family and friends .... While she was aware that sale and possession of drugs was illegal, her own experience of its positive effects on her husband ... appear to have contributed to her lack of appreciation of the seriousness of her actions. She now understands the gravity of her position .... Pam does not appear to abuse drugs and demonstrated good skills in general management of her domestic circumstances. (Pam - Drug Offence) 160 When she was sixteen she met a thirty year old man while on holiday. The couple continued to correspond and were married ten months later. They lived with her husband's sister. However, this situation was far from ideal. Bridget felt indebted to her husband's sister, a fact she states played a part in initiating the offence in that amongst other things she felt like she owed it to her sister-in-law. Bridget was frustrated that she was unable to contribute in any financial way. She also felt guilty about her proposed visit to her father in Scotland. There was also the notion that she 'owed something to her sister-in-law'. Thus, in order to lighten her husband's burden, to bring down his stress load and to do something in return for all the kindness shown her she hit on the idea of property offending after seeing how easy it was to get rnoney from the company previously.... There is no doubt in rny mind that her main airns were to relieve the stress for her husband, to pay back her husband and sister-inlaw in that she felt she owed them. (Bridget - Property Offence) Sue's and Jessica's offending is presented as a response to traumatic familial experiences; these experiences are noted to have "shaped their adult thinking" and "had a bearing" on the course of "the current offences." Sue described a wretched childhood full of various forms of neglect and abuse which has shaped her adult thinking, and left her with a legacy of emotional problems ... these problems ... appear to have aggravated her sense of misfortune and helped support her offending ... Sue impressed as a person who experiences, and is burdened by, a magnitude of personal misfortunes. She appears to have been damaged by her childhood circumstances, an the consequent complex emotional problems have shaped her attitudes, that have in turn been an integral factor in her offending. (Sue - Property Offence) Her parents separation, her subsequent disappointment in and rejection of her father, her associates, and de facto, her relationships with young men with addiction problerns, may all have a bearing on her course to the current offences. (Jessica - Drug Offence) 2.3 Female Offenders, Support Networks and Informal Social Control The examples given so far demonstrate how familial constructions of women as nurturing, caring, dependent and relatively powerless can detract from their perceived criminality and reduce the need for punitive sentences. Social control within the family was alluded to and women's future sentencing was often linked to treatment via the establishment of 'good familial relationships'. Furthermore, where good relationships existed in families they were seen as rehabilitative in their own right, as the following examples demonstrate. Julie's and Freda's rehabilitation is seen as dependent on the supportive home environment created by a sister in the former case, and husband in the latter. For the last two months Julie has been living with her sister and her four children. It is clear that Julie's sister has a considerable stabilising influence on her and contributes to the care of her children. Julie has been abusing drugs and related substances since she was eleven years old. Up until the time of the offence she was using drugs intravenously, as well as consuming prescription pills and large quantities of alcohol. Since she has moved to her sister's home her drug intake has decreased dramatically. Although she does admit using occasionally, her sister's influence has clearly seen her take a more responsible attitude towards herself and the care of her children. (Julie - Violent Offence) 161 It may be that, with the move from her previous well known address, her awareness of the distress that her absence causes her daughter, son and husband and the affectionate support of her husband will motivate her to seek treatment and attempt to gain a drug free status. (Freda - Drug Offence) Prue's father, rather than Prue herself, is said to be taking "responsibility for" his daughter's problems and for assisting her to get treatment. For the first time to her parents Prue has acknowledged her drug addiction that she needs help in overcoming it...Her father accepts responsibility for his daughter, and is taking her to a drug rehabilitation centre for assessment. (Prue· Drug Offence) In the past, Tasha relied on her own resources to overcome drug addiction and is currently expressing determination to do so again. Yet despite expressions of independence, Tasha's recovery is still pinned to her husband's support. The impression gained is that Tasha's background has required her to rely on her own resources .... when asked how she managed to overcome her drug dependence in the past she said she achieved this by removing herself briefly from her social and geographical environment and detoxifying herself. Whilst she expresses determination to cease drug use in the future, it would appear that this is more likely to be achieved with the help and support of her partner than through conventional programmes. (Tasha • Violent Offence) 2.4 Women as Essential Family Figures Paradoxically, the family was also constructed as a site where familied women had a certain degree of power and control. This is hardly surprising, given women's place as 'moral redeemers' within the 'cult of domesticity' which presents them as the instruments through which social order can be imposed on men and children (James & Saville-Smith, 1994: 28). In the pre-sentence reports, concerns were often raised about the impact or social cost of removing women from families. Generally, women were presented as the only ones who could keep families happy and together. In extreme cases, women were presented as the saviours of partners and children who were 'troubled' themselves. Juxtaposing the 'moral redemptress' against criminality in this way reduced the suitability of incarceration for women by detracting from offending and highlighting how sentencing may impact on families. General concerns regarding the social cost of removing women from families are shown in the examples of Debbie and Kate presented below. The inevitability of a custodial sanction is acknowledged. In light of the punishment which such a sanction would impose on her children and husband it is recommended that her sentence be as short as possible. (Debbie - Drug Offence) 162 Kate appears a capable and committed parent toward her three children. The drawn out period of the charges has placed considerable stress on her an her family and a sentence of imprisonment would further add to this situation. She intends on enrolling her latest child at school which will entail Kate being involved in the school. (Kate - Violent Offence) The second group of cases, which included Sandra, Kim, Cindy and Jill, illustrate how women are presented as the saviours of partners or children who were 'troubled' themselves. Sandra's fourteen year old son faces admission to a drug rehabilitation .... a non-custodial sentence would enable her to address her own drug related difficulties first and then she will be able to contribute positively to her son's treatment. (Sandra - Drug Offence) Kim's children have recently displayed problematic behaviour and the youngest has recently changed schools ... a case worker from the Children and Young Person's Service has informed me that in his role as case worker for the youngest child, he has been in close contact with the family and has observed a deal of strain among them regarding the possibility of Kylie being imprisoned. He noted that the youngest child has showed improvement in his behaviour since gaining involvement with that service... He also stated that although Kim's partner has offered to care for the children should she be imprisoned, Kylie has been the main care giver and her partner has not assumed that role in the past. He stated that he believed there would be a significant impact on the family should she be imprisoned. Kylie gave the impression of being closely focussed on her family and was obviously anxious about their care should she be imprisoned. (Kim - Drug Offence) Cindy has resided at the same address for the past two years. Presently she has the difficult task of caring for three children on her own while her husband serves a prison term. She seems well organised, tidy and making a strong effort to keep herself and her children on an even keel. Cindy's efforts towards a healthier lifestyle, may increase her husband's motivation to make changes, upon his release .. She has the enormous task of reuniting her family ... she has come a long way and has a long way to go... it is my opinion that Cindy would respond positively to the further support and assistance of the probation service. (Cindy - Property Offence) Jill's ten year old son is presently undertaking counselling with a school guidance counsellor, because of anxiety problems and possible sexual molestation in the past. Jill has been an active supportive participant in this process. In my opinion her son needs his mother's ongoing care and involvement in order to maintain progress. Separation at this time would be detrimental to her son's emotional stability. Jill expressed great fears for her children and her son in particular, in the case of separation by imprisonment. (Jill - Property Offence) 2.5 The 'Family Man' and The 'Man Alone' The portrayal of dominant ideals of masculinity stood alongside femininity in the pre- sentence reports. Like women, men were defined in terms of familial connections or lack of them through two constructions of masculinity: the 'family man' and the 'man alone'. Just as maintaining familial needs via the 'cult of domesticity' was ideologically important for women, men's criminality was more easily detracted from if they were successful 'Family Men' rather than 'Men Alone'. The 'cult of domesticity' controlled women socially and was used by judicial actors to 'refeminise' and reconstruct them as 'real' women rather than criminals. 163 Similarly, the successful 'family man' represents socially controlled masculinity while the 'man alone' threatens the social order. Men's criminality is thus aggravated if they are constructed as unsuccessful family men because they pose an ongoing threat to society. This makes punitive sanctioning more likely for the 'man alone' than it is for the successful 'family man'. Discussion about employment status was more prevalent in men's pre-sentence reports, which is not surprising given links between paid work and the ideal of 'bread winning' masculinity. This is not to say that talk about employment or unemployment was absent from women's reports but for women, discussion was included only to fullfill standard report requirements. All pre-sentence reports have an employment/income heading but only a few lines were ever allocated to discussion about women's (un)employment, as the following examples illustrate. Rebecca is in receipt of the unemployment benefit (Rebecca - Property Offence) Susan is presently on a sickness benefit. (Susan - Violent Offence) Jane has most recently been employed as a hairdresser, a position she sought as a way of supplementing the family income. (Jane - Drug Offence) Julie has mostly been a full-time mother who has done sewing and fruit picking to supplement her income. (Julie - Drug Offence) Unlike women, detailed discussion about men's employment status was usually provided in the pre-sentence reports. Men's employment was often presented as an indication of responsibility, stability and worth but unemployment was invariably associated with a criminal lifestyle. These constructions reflect general ideologies about men in New Zealand, where primacy is given to the 'family man', if he is a successful provider, over the 'man alone'. Women's presentence reports often framed them as domesticated, not really criminal and thus not really deserving of punitive sanctions. Constructing men as 'breadwinning family men' similarly detracted from their criminality, possibly reducing the need for more punitive sanctions. For example, Ben offended because he was trying to provide for his children and presents as "a responsible member of the community" who "actively seeks work." Until recently Ben lived with his five year old son in rental accommodation. He has five other sons who live with their mother. Ben retains close contact with his children ... apparently Ben was prompted to sell drugs in an attempt to better provide for his children. He stated that his children, who live with their mother, are often without food and other basic necessities. Despite his outward appearance, Ben is a 164 pleasant and co-operative man. In many respects, he appears to be a responsible member of the community in that he actively seeks work and visits his children on a regular basis. (Ben· Drug Offence) Allan's criminal status is similarly minimised because of his status as a "hardworking family man" which is difficult to reconcile with his offending. Allan impresses as a pleasant, articulate man and an unlikely offender. It is difficult to reconcile the current offence with the man who presents, given that those canvassed expressed their shock that the stable, hardworking family man should have become addicted to 'homebake' and been involved in the current offence. (Allan· Drug Offence) Presentations of criminal men as 'family men' were the exception rather than the rule. Instead, men were more often portrayed as financially unable to support their families and were generally seen as the source of all familial problems. Jim has been married for 20 years and is the father of four children. Jim's wife described her marriage as in name only, citing her husband's inability to provide adequately, his behaviour and continued offending as reason why their marriage has been under stress... Jim is a carpenter by trade but he appears to operate his business in a haphazard manner, and it is equally apparent that he cannot rely on it as a regular source of income... At the age of forty, Jim seems unfortunately long past the stage where something more positive can be made of his life, especially with regard to his familial responsibilities. (Jim - Property Offence) Although some women's criminality was seen as being caused by dysfunctional family relationships, women were still seen as necessary to maintaining familial harmony. In contrast, the pre-sentence reports constructed many men as being destructive in terms of family harmony. The examples of Andrew, Kyle, Matthew and Simon clearly illustrate this: Andrew's use of drugs and alcohol is believed to further isolated and alienated him from those who care about him. He was inadequate as a partner and father, preferring to pursue his ow!1 selfish interests and there is no chance of resuming his ten year relationship with his ex-partner. (Andrew - Property Offence) Kyle has for a significant part of his adult life been actively involved in drug abuse with predictable adverse results for his life and his families. When not caught up in a cycle of addiction Kyle is able to see that his drug using lifestyle has been very hard upon his partner of some ten years, and their five year old son ... Kyle's partner's loyalty must have been overly tested by Kyle's drug abuse over the years and it is remarkable that she continues to profess her love for him and her commitment to him. (Kyle - Drug Offence) Matthew acknowledges that his relationship with his partner has been turbulent...he seems willing to examine his destructive patterns of communication which have at times culminated in abusive behaviour. (Matthew· Violent Offence) 165 There is a degree of animosity between Simon and his wife ... he accuses her of trying to alienate the children from him. She worries about his effect on their son, especially, and she states that he was violent, aggressive, and sexually deviant during their marriage ... his current girlfriend is unsure if she will persevere with their current relationship. She has discovered that Simon has been unfaithful during their relationship and was lying to her... she is concerned about his alcohol dependence and sexual practises. (Simon - Property Offence) Men who failed to achieve the ideal of the 'breadwinning family man' were presented as 'Men Alone'. This less-desirable and less-controlled form of masculinity commonly featured in presentence reports both for familied and unfamilied men. Like 'family men', unemployment remained unacceptable for single men even if they did not have families to support. Unruly, disruptive and generally criminal behaviour was blamed for single men's joblessness. Grant has been unemployed since leaving school. While he does not lack intelligence, Grant has little sense of direction and stated that since he began using drugs, he has lost all interest in leisure or employment pursuits. He described the collection and use of drugs as a full-time job, in that when he was not in the act of searching out supplies, he was either planning or thinking about it. (Grant - Drug Offence) Aaron was previously employed for approximately eight months. His former employer said that Aaron virtually presented as a conscientious and reliable worker, particularly when supervised. However, his work habits deteriorated from late last year and his ex-employer believed the change in attitude could be alcohol and drug related. (Aaron - Property Offence) For men, achieving a more socially acceptable masculinity VIa work, and ultimately 'breadwinning' for the family, may have reduced the need for more formal control at sentencing. This contrasted with women for whom caring for the family was presented as paramount. In fact, the potential importance of men's care-giving ability was all-but ignored in the pre-sentence reports. This is starkly illustrated in the cases of Brad, Hyden and Wayne presented below. All these men were solely or primarily responsible for childcare but the importance of their child caring was not discussed, as was the usual practice in women's reports. For example, speculation regarding the impact imprisonment may have on the children was not provided. In Brad's case, his inability to obtain adequate financial support for his child, rather than his care-giving role, dominates the discussion. It would seem that removing fathers from families has few social costs, maybe because financial provision is easier to replace than emotional care. Brad lacks insight into his life in general. As a joint-custodial parent living on a single unemployment benefit he remains financially vulnerable. Although he was aware that he could obtain further funding to help care for his daughter, he had not taken any action to obtain it. He appears very unmotivated. 166 When asked about possible education courses or employment training programmes that he felt he might benefit from, he stated that he did not feel the need to make changes in his life. (Brad· Drug Offence) Hyden lives in rented accommodation with his six month old son. He currently has sole care of the child as his partner is serving a prison sentence... the relationship between Hyden and his partner has at times been difficult but they are planning to resume living together upon her release. (Hyden· Violent Offence) Wayne is a solo parent who looks after his two year old daughter. A son is in the custody of his former partner. Wayne separated from his partner despite her efforts to reinstate the family unit. (Wayne· Property Offence) Finally, it should be noted that men's criminality was rarely excused by reference to dysfunctional familial relationships, as was the case for women. This did not mean that men had not experienced traumatic childhoods, because pre-sentence reports did note incidences where men's family backgrounds were troubled. However, such discussions occurred less often than they did for women and had less emphasis upon it. Although talking about familial problems created a link to men's offending by offering possible explanations, problematic childhoods were rarely presented as an excuse for criminality. For example, it was rare to find reference to familial experiences having 'shaped' men's adult thinking or 'bearing on' the course of their current offences as it did for women. Instead, remarks tended to be offhand and. detached from men's crimes, as the following examples illustrate. Joe's childhood and adolescence were characterised by his parents' marital discord and difficulties in parenting. Previous records indicate that he was the victim of parental neglect and abuse. (Joe - Violent Offence) Barry describes having lived with his mother until aged 14 at which time he was placed in a foster home for a brief period. Since that time he has largely lived independently in a number of flatting situations. (Barry· Drug Offence) Richard grew up in Auckland and there were seven children in the family. His father was a violent man who drunk excessively. (Richard· Property Offence) Carl is the youngest of five children and his parent and siblings have all been involved in criminal offending. He spent most of his childhood in foster care. His foster mother remains supportive of him. (Carl· Drug Offence) Ron was bought up in a disruptive home environment characterised by his father's abuse of alcohol and consequent violence towards family members. As a result, Ron was placed in foster care at the age of fourteen. (Ron· Violent Offence) 2.6 Women's Pathology and Men's Action The general tendency to construct pathology or mental illness as predominantly a female problem has been noted by feminist scholars (Us sher , 1991; Tavris, 1992; Prior, 1999). 167 More specifically, criminological research shows offending women are often reframed as 'sick' rather than criminal (Allen, 1987a; Allen, 1987b; Allen, 1987c; Worrall, 1990; Ussher, 1991: 171-173). Allen (1987a: 82) notes that pathologising criminal women neutralises their guilt, responsibility and dangerousness and undercuts demands for punitive or custodial sentences; women are essentially 'rendered harmless'. Pathology thus functions similarly to domesticity; women are again 'refeminised' but this time as 'mad' rather than 'bad' and their criminality is reconciled within a dominant feminine discourse which detracts from their offending (Ussher, 1991: 172). Research has also found that judicial responses to male offenders are "bound by stereotypical assumptions concerning their manhood" (Walklate, 1995:140). Judicial sympathy is rarely given to men because they are understood to be 'bad' rather than 'mad'. Badness and disruption are synonymous with maleness; criminality and masculinity are ideologically intertwined. Thus the ascription of criminal labels to men somehow seems more 'natural' than for women. Ussher (1991: 172) explains that "we know that men commit crimes, that they are sometimes bad, and thus we merely punish them for their anti-social actions, without looking any deeper for an explanation for their behaviour. The man who stabs his lover is merely acting like a man. He must be punished, but his behaviour causes no particular surprise." In the current study, strictly 'pathological' or 'medicalised' reasons for offending provided by mental health professionals (psychologists or psychiatrists) and presented in Probation Officers' reports were rare for men: one man was diagnosed with post-traumatic stress disorder, two with anti-social personality disorders and one with a suicidal tendency. In comparison, 15 women were diagnosed as 'disordered' by psychologists: one woman had posttraumatic stress disorder, seven had suicidal intentions and seven had physiological 'hormone imbalances' relating to reproduction, including post-natal depression (five cases), menopause (one case) and hormonal imbalances during pregnancy (one case). In addition to gender differences in the frequency of reported pathology, the discussions presented around mental health tended to vary between the sexes. Causal links between offending and mental health problems were made more often for women, which sets up pathology as a mitigating factor. In the examples below, Sally's post-natal depression is shown indirectly to be the "cause of the current offences." 168 Post-natal depression resulted in her admission to hospital following the birth of her daughter. .. she was discharged before fully recovered and this attributed to the subsequent acceleration of her drug abuse. This increasingly serious drug addiction is seen as the main cause of the current offences. (Sally -Violent Offence) Community service is recommended for June because she had been under "considerable stress" after the birth of her child. The period after the birth of her child was difficult for June, in that she became significantly depressed. This post-natal depression lasted as long as two years .... in terms of sentencing and noting the considerable stress June has been under it is recommended that she be sentenced to community service. (June - Drug Offence) Men's diagnoses on the other hand, were rarely presented in the same light as women's. Regardless of 'pathological' depression and suicidal thoughts, John's probation officer concludes that there is "no evidence to suggest that John is not responsible for his actions." It would seem that "a lengthy prison term is inevitable." It is not entirely a surprise that John again appears before the court for a very serious offence. His background of emotional, physical and possible sexual abuse has left a legacy... he still carries around a great deal of anger. Also it is apparent that he retains a propensity to abuse alcohol and drugs. In addition his periods of depression are pathological and include suicidal thoughts. Having said all that, there is no evidence to suggest that John is not responsible for his actions. He is very much aware that a lengthy prison term is inevitable. (John' Violent Offence) Peripheral mental health problems 11 appeared less frequently in men's reports: 16 compared to 31 in women's reports. Once again, gender differences in the nature of these reports were found. Women were frequently identified as having general 'emotional problems' and were described as anxious, stressed, distressed, lacking self esteem, lacking assertiveness and emotionally unstable. Men were noted to have 'behavioural problems'. They were disruptive, impulsive and/or angry, but explanations for this behaviour were rarely given. Presenting men this way constructed them more as 'subjects of action' with problems linking to what they did rather than what they felt (Allen 1987b: 40-41). In contrast, women's thoughts and feelings were richly elaborated on in reports. Inner turmoil, usually caused by events outside of women's control, were frequently used to explain offending. This essentially removed criminal women from the realm of human action, invoked sympathy and reduced their blameworthiness (Allen, 1987a: 84). 11 Discussed by Probation Officers but not diagnosed as such by mental health professionals. 169 These gender differences are exemplified in the examples presented below. All the men are presented as actors, masters of their own destiny and responsible for changing their own deviant behaviour. In contrast, Linda appears stricken by inner turmoil. Nathan impresses as being a person who has struggled to deal with his impulsively and anger for many years and who has self medicated with substances. This has tended to reduce his ability to control his behaviour. (Nathan· Violent Offence) Simon acknowledged at the interview that he has resumed heavy daily consumption of alcohol and cannabis and now accepts that he lacks the self-control needed to modify his substance use and criminal behaviour. .. A clear pattern of behaviour has become apparent over the last few years. He lapses back into an aimless life-style dominated by substance abuse during periods of liberty, before reoffending and further incarceration ensues. (Simon· Drug Offence) Craig has failed to address his addiction problem ... prior to his last prison sentence he confidently predicted that alcohol and drugs no longer represented any threat to his stability in the community. Craig acknowledged at interview that he resumed heavy daily consumption of alcohol and cannabis immediately after his release and accepts that he lacks the self-control needed to modify his substance use and criminallifestyle ... no community based sentence can be recommended. (Craig· Violent Offence) Linda's husband died five years ago, and since then she has experienced difficulty raising her children ... this has caused her a lot of anxiety and emotional stress ... she has been seeing a counsellor to try and get her life back together again ... ln interview Linda was very emotional and it is clear that the last year has indeed been most traumatic for her. It is recommended that the court consider a period of supervision with the special condition that Linda continue counselling. (Linda· Property Offence) The case of Rangi and Jane is particularly striking because both offenders were similarly convicted of a serious violent offence and had previously spent time in psychiatric institutions. Rangi's report constructs him as an actor by noting that he "made little progress" and continually "acted out" while in the psychiatric institution. Rangi is presented as not really 'sick', he knows his behaviour is "morally wrong" but is unlikely to change. An explanation for why Rangi, or indeed Nathan, Simon or Craig (above), may have 'acted out' is not offered by the Probation Officers. In Rangi's case the only real concern is community protection. In contrast, the only person Jane poses a danger to is herself because while there is no evidence of mental illness, "her ability to cope with emotional tensions is grossly deficient" and "suicidal feelings" are noted. Rangi no longer suffers from any psychiatric illness ... he is able to understand that his behaviour is morally wrong but acts out when he is angry. His acting out takes the dangerous form of violence. I regard him as a danger to the community.... Rangi spent four years as a committed patient but made little progress during this time ... he is unlikely to change his behaviour, and a consequent risk to the community. (Rangi • Violent Offence) 170 Jane states that she had suicidal feelings at the time the offence was commiUed ... Although there is no evidence of mental illness, her ability to cope with emotional tensions is grossly deficient...it is possible if not likely that when under great emotional pressure she would inhibit symptoms of depression .... at the time of the offence she would have been under severe pressure; her ability to establish a home, the rejection of her boyfriend and eviction from her home ... she would have been desperate in her efforts to function ...Jane's problems are seen to stem from her perception of herself as unlovable and unloved. (Jane· Violent Offence) The examples of Nathan, Simon, Craig and Rangi illustrate that men are not given the potential to experience inner pain. It is not acknowledged that men's "acting out" can in fact be the result of personal distress bought on by traumatic life circumstances in the way that it is for women. The result is that men are held responsible for their criminal actions. In contrast, the emotional pain leading women into a criminal life are sought, outlined and emphasised by Probation Officers in their reports. The result is women appear less responsible for their actions, as the examples of Linda and Jane show. Gendered constructions ill the pre-sentence reports support a masculine ideology which minimises men's feelings, vulnerability and weakness. We gain the impression that criminal men are unaffected by traumatic life events because women are used as the yardstick for mental illness. 'Talking about it' and expressing emotion are normative expressions of femininity and psychological distress (Tavris, 1992: 259). Offending women are thus expected to talk about feelings while also displaying emotion. However, culturally approved masculinity does not support similar responses from men and alternative expressions of psychological trauma from criminal men are ignored. For example, it is never asked if Craig's 'lack of self control', Nathan's 'struggle with anger,' Rangi's 'acting out' and Simon's 'aimless lifestyle' are in fact expressions of grief, depression, or stress. Instead it is easier to presume that men suffer less than women, are less deserving of understanding and thus more worthy of harsh punishment. Emotionality and talking about problems enable Probation Officers to 'refeminise' criminal women who were presented as passive victims of circumstances and undeserving of punitive sanctions. In most cases, women actively participated in this talk; they constructed gendered judicial discourse as much as they were constructed by it. However, some women actively contradicted culturally approved femininity by refusing to talk about their feelings, by being unemotional ancl!or openly denying their victim status. For example, Karri's father had recently died but she denies needing help to 'deal' with this and instead states she will only go to counselling if it will help her to avoid imprisonment. Despite this admission, the Probation 171 Officer claims Karri does not know what she is saying because she is in the "throws [sic] of a depressive episode" and in need of supervision rather than imprisonment: Karri is obviously in the throws of a depressive episode. This could account for her apparent lack of self motivation and her present perspective; for example she said that she would attend grief counselling only if it was to her advantage in sentencing. This is more likely to reflect her psychological state than to be a well thought out comment...sentencing is complicated by the seriousness of the charge .. .Despite this it is evident that a sentence of supervision would seem the most appropriate for Karri. (Karri - Violent Offence) In Jenni's case, failure to show emotion is seen as problematic and she is sent for a psychiatric assessment. Jenni was tense and stressed at the initial interview and described various personal traumatic experiences while exhibiting little or no emotions. She was subsequently referred to Dr Brown for a psychological assessment .... as the interview progressed she was able to talk more freely, and I gained the impression that she was being as honest as her insight allowed ... it is apparent that she is suffering from a degree of distress ... (Jenni - Property Offence) Emotional denial or silence from women is thus interpreted by Probation Officers as further evidence of pathology and the 'mad' rather than 'bad' label is again reinforced. For women in 'emotional denial' , help is needed to get them 'in touch with' and to 'talk about', their feelings. Female criminality, it would seem, is integrally intertwined with pathology because it provides another way to reconcile women's offending within an overriding feminine discourse. Even women who failed to express 'normal' feminine emotionality or talk nonetheless had 'normal' female emotions attributed to them, albeit in pathological form (Allen, 1987a: 105-106). In contrast, men who tried to break with culturally acceptable masculinity by showing emotion, talking about mental health problems and/or attributing their behaviour to circumstances beyond their control, tended to be rebuffed or ignored, as the example below demonstrates. Bill described the symptoms of apparently manic depressive episodes, which he said, thrust him back into pill taking in an attempt to calm himself down. However, past psychiatric reports conclude that Bill shows no evidence of psychological disturbance. (Bill - Property Offence) Any attempts by male offenders to negate responsibility for their actions were quickly silenced as illustrated in the following case of Glen. While troubled childhoods mitigated women's criminal responsibility by invoking judicial sympathy, Glen's childhood, which was characterised by violence, instead only provided further proof of an entrenched and unchanging 172 criminal identity. Glen is described as having actively "adopted an anti-social lifestyle" rather than being a victim of childhood trauma. Glen's father is also dying of cancer and Glen tells the Probation Officer that this has been upsetting for him but it is not considered a mitigating circumstance, given Glen's criminal background. Finally, the desire for treatment is made clear by Glen and rejected by the Probation Officer as either not genuine or not possible, making an interesting contrast to the case of Karri presented previously. It would seem that women need help whether they want it or not, whereas men do not need help even if they ask for it because "talking about feelings, experiences and inner problems is regarded as foreign, irrelevant or simply unhelpful" to them (Allen, 1987a: 105-106) or seen merely as a ploy to shift responsibility for their actions elsewhere. Glen is the fourth child of seven. He reports a home life characterised by parental violence. He is described as having learning difficulties at school and was asked to leave high school after three months, being exempt from then on. He left home at the age of fourteen. Glen's family have a history of offending ... Glen was co-operative during interview. He is a man who from an early age adopted an antisocial lifestyle characterised by criminal associates and substance abuse ... over the last few years records indicate the development of a more genuine desire to change ... although he is going though a difficult time with his father's illness I am reluctant even in the face of his desire to pursue treatment to recommend this option given his history. Issues surrounding treatment could be addressed prior to his release from prison. (Glen - Drug Offence) An emphasis on the inner turmoil of female offenders tended to construct them as relatively harmless and as not really responsible for their actions. This improved women's chances of being judged suitable for community-based sentences or short imprisonment terms. Furthermore, women's involvement in the criminal justice system was often seen as a punishment itself having added to both the offender's and their family's turmoil. The whole process from arrest through to sentencing was often seen as something 'being done to women and their families', rather than the result of something the offending women had caused. For example, Joan was "fearful" of imprisonment, Cindy was "upset and shocked" by her remand in custody, "imprisonment would be a significant blow to" Sue's self-esteem, Carol was "anxious" about incarceration, Jane was "under considerable stress" because of the whole process, and Bridget's custodial remand period was "harrowing." Joan is fearful of being imprisoned because of the impact on her children ... lf the court considers that a prison sentence is essential I recommend that this be suspended ... both her doctor and counsellor consider Joan could benefit from some ongoing over sight and support and accordingly supervision is recommended to provide this. (Joan -Drug Offence) 173 Cindy was upset and shocked at finding herself on remand in custody. (Cindy· Drug Offence) Jane appeared very stressed by the circumstances surrounding this offence and the weekend she spent in custody... she appears to be under considerable stress, no doubt largely as a result of her conviction for this offence. (Jane· Violent Offence) Jill is under considerable stress and strain at present, partially because of the current offence. (Jill· Property Offence) Imprisonment would be a significant blow to her self-esteem (Sue - Drug offence) Carol presented at interview as a woman under a noticeable amount of pressure and stress ... she is closely focussed on her family and was obviously anxious about their care should she be imprisoned. (Carol - Drug Offence) Bridget spent three quite harrowing days in custody and found the separation from her children upsetting. (Bridget - Drug Offence) In contrast to the pre-sentence reports on women, the idea that men could also find the criminal justice process and possibility of imprisonment scary, upsetting, stressful, distressing, harrowing or a "significant blow to the self-esteem" was rarely mentioned, if at all, in the reports I read. 3 The Construction of Gender in Judges' Sentencing Remarks The comments made by Judges at sentencing are recorded and stored on offenders' trial files. Remarks are presented in a somewhat standardised format and usually include the following information: circumstances of the offence, offenders' criminal history, offenders' social and personal characteristics/histories as outlined in the pre-sentence report, by prosecutors, defence counsel and offenders, and the sentencing decision. The remarks made by Judges at sentencing are useful because they provide a formal account of the rationales used when sentences are decided. The court process is a complex interplay between the Judge and many other actors, but the printed sentencing records can provide a window into how cultural ideals of masculinity and femininity operate and impact on sentencing (Gelsthorpe & Loucks, 1997). 174 Although Probation Officers make sentencing recommendations, their reports are constructed within a social work framework, which generally focuses on offenders and their welfare (Hagan, 1975; Reed & Thier, 1981: 234; Shapland, 1987: 83). While aware of, and perhaps even influenced by, Judges' sentencing expectations (Roberts, 1987: 94), the decision-making framework used by Probation Officers differs from Judges, who are required to consider a diverse number of sentencing aims including retribution, denunciation, incapacitation, deterrence, rehabilitation and restitution (Ashworth, 1995: 57-81; Ministry of Justice, 1997: 37-80). Retribution is punishment-focussed and requires that the seriousness of an offence be replied to by the state's imposition of a punishment proportional to the criminal harm caused. Denunciation is also proportional, with harsher penalties being used to publicly condemn less acceptable forms of behaviour; punishment in this case operates to reinforce societal norms (Ministry of Justice, 1997: 37-39). Incapacitation ensures that offenders are incapable of offending again for set period of time; this principle operates under the proviso of public protection and crime prevention (Ashworth, 1995: 67-68). As far as deterrence is concerned the hope is that future levels of offending will reduce if fear of the consequences can be instilled in both the individual offender and society at large (Ministry of Justice, 1997: 39). Rehabilitative sentencing aims to reduce future crime by changing the behaviour, attitudes, or skills of the offender. Rehabilitation assumes that offending has specific causes and the focus is on identifying and remedying these factors (Ministry of Justice, 1997: 62). Restitution, developing from an increased focus on the rights and needs of victims, is a more recent sentencing aim (Ashworth, 1995: 73-74). This principle requires the offender to put right the wrong done and usually takes the form of monetary payment and/or apology. Reparation places the victim at the centre of the sentencing process rather than the offender or society (Ministry of Justice, 1997: 62). Some of these sentencing aims are exclusive, some overlapping, some incompatible and it is argued that the weight given to each by sentencing Judges inevitably changes (Samuels, 1987: 66, Kapardis, 1987: 20). In this study, Judges frequently used retribution, denunciation, incapacitation, deterrence, and rehabilitation as the framework for rationalising sentencing decisions. I was struck by the changing emphasis given to these principles when reading through the Judges' sentencing remarks. Of particular interest was how constructions of 175 gender within this framework produced different sentencing rationales for men and women. Rehabilitation tended to take centre stage for women, while the remaining sentencing aims were paramount for men. Like the pre-sentence reports, gender was constructed around dominant ideologies linked to family, work and mental health. 3.1 Female Offenders and Domesticity Presenting women within a framework of domesticity was common in many Judges' sentencing remarks. Women's familial relationships were usually discussed in some depth by Judges at sentencing and provided a way to rationalise less severe sentences for women on rehabilitative grounds, as long as the women were seen as conforming to dominant ideals about familial femininity. For example, Judges expressed concern about the cost of women's imprisonment on children. Sentencing in respect of this offence today is complicated by the fact that you are presently approximately eight months pregnant. If I sentence you to a term of imprisonment with immediate effect, the likely outcome would be that you would be released from prison at the time of or shortly before the birth of your child. Even if that outcome was not to occur, the last thing that your unborn child needs is to start his or her life in the environment of prison. (Sarah - Violent Offence: Sentence = Suspended Sentence of Imprisonment) Judges also identified the family as an important source of social control in women's lives. In the cases of Julie and Sue presented below, social cost and social control were used to mitigate sentences of imprisonment in favour of supervision; a sentence based primarily on rehabilitation. Julie was convicted of a serious violent offence and Sue had been dealing in drugs. Julie's violent offence was particularly serious, carrying a presumption of imprisonment under New Zealand law. In both cases the Judges commented on principles of retribution and denunciation in that they acknowledged the seriousness of the offending. This was marked further in Julie's case, where the Judge also sentenced her to a term of periodic detention, in addition to supervision. However, both offenders avoided imprisonment because of 'special circumstances' including their statuses as mothers and the stabilising influences of family members on their lives. The women's offending was thus reconciled within discourses of domesticity, which saw women as controlled within the family but as also central to its maintenance. This ideological yet often material positioning of women at the centre of the family required greater efforts to be made to rehabilitate them so they could fulftl their 'proper' 176 gender-roles. As can be seen, Julie's Supervision Order required her to attend are-parenting course while Sue's drug rehabilitation treatment was seen to enable her to be a 'good mother'. You are the mother of three children. You have recently been living with your sister and that has bought an element of stability into your life ... your lawyer submits that the interests of society as well as your family's would be best served by a non-custodial sentence...the general circumstances of the offence, which were gone into at trial are by no means excusable.. it is my view that by a narrow margin there are sufficient special circumstances here to justify a sentence of imprisonment... I am going to give you the benefit of the doubt but I think your case stands a narrow line between imprisonment and not .. special circumstances will entitle the court to stop short of the otherwise mandatory imprisonment... I agree with the Probation Officer's observation that supervision would be an appropriate course to follow, it is my view that there must also be something on top of supervision to mark the seriousness with which the court views the use of knives in any sort of situation. You have escaped imprisonment by a narrow margin. (Julie - Violent Offence: Sentence attend a re-parenting course). = Periodic detention, Supervision with the condition to You are a solo mother with a young child ...the Probation Officer's report says that you impress as a likeable and vulnerable young woman who has difficulty in trying to cope with the responsibilities of independent living and motherhood ... in the mean time your father has accepted responsibility for you while you have been on bail ...The things that have influenced me are that you have the support of your parents and you have responsibility for your daughter. You will have learned in the past few months particularly that her future, just how stable she is, what she grows up to be, will depend on you and for that reason you are being given the chance to see that you make something of yourself; if you rid yourself of the drug habit and that for both your sake, your parents' sake and the community's sake, you will be able to act as a good mother and a good citizen. I want to emphasise to you that normally the sentence imposed on you would have been imprisonment. You are being given this opportunity. It is up to you to make the most of it... (Sue - Drug Offence: Sentence = Supervision with the condition to attend drug treatment) In the pre-sentence reports it was rare to find critical comments about women's mothering abilities. Women were nearly always presented as the moral redeemers of families; responsible for keeping families together and keeping them happy. Such constructions were popular because they provided a way to 'refeminise' criminal women as 'real women' rather than as 'criminals'. These patterns were also noted in Judge's sentencing remarks, as the cases of Sue and Julie above demonstrate. In contrast to Probation Officers, Judges would sometimes emphasise punishment through retribution, denunciation and incapacitation when women failed to conform to what they perceived to be domestic ideals. Incapacitation was particularly important and Judges would not only highlight the need to protect the public from female offenders but, more specifically, the need to protect children from 'bad mothers'. Child protection and social cost, it seemed, worked both ways by keeping 'good mothers' out of jail and by putting 'bad mothers' in. I am quite amazed to have a pre-sentence report recommending supervision ... the seriousness of this type of offending is such that a prison sentence in my view is inevitable ...the community is to be protected from activities of this kind. I have been asked to have regard to the fact that you are the mother of small children ... however, there are difficulties in making special allowances because of children. It would be quite unjust to allow a lesser term for children so far as you have become a drug and alcohol addict and have resorted to violence ... 1must confess that I think your children will be much better served being away from you for some while than being left in your custody ... (Kim - Violent Offence: Sentence = Imprisonment) 177 Her personal circumstances are that she is a solo parent with three children. All of the witnesses describe her as a good mother, and whilst I accept that, one is bound to say that she could not have thought very much about the children when she embarked upon this course of offending ...she must have been aware that her behaviour could result in criminal charges and impact on her children. It is all very well to protest now... but it seems to me somewhat in hindsight... I have thought carefully about the strong mitigating submission by counsel. ...this defendant is obviously a person that one side of her character has a lot to offer. On the other side, however... she continues to offend. In my view, I am satisfied that a custodial sentence must be imposed. (Karla - Drug Offence: Sentence = Imprisonment) 3.2 Female Offenders and Family Pressures Similar to the pre-sentence reports, direct or indirect domestic problems were used by Judges to mitigate women's criminality and reduce sentence severity. The pressures of family life and/or the actions of insufferable husbands were used to construct women as not entirely responsible for their actions. In these circumstances, criminal women could be reframed as 'real women' and not criminal after all, because explanations that generate compassion and understanding were found for their offending behaviour. In the case of Viv, for instance, focus on the damaging effects of past familial crises in conjunction with present family relationships (she was also a mother and had a supporting family) dominated the Judge's sentencing remarks. This drew attention away from Viv's current and past drug offending. The Judge discusses Viv's current offence only briefly and fails altogether to mention her prior drug convictions. Instead, her background is noted to have had a "bearing" on the course of her life and the Judge wants to give Viva chance to heal. Sally is described as an "accidental" offender whose offending is the result of a destructive relationship. The "real culprit" is her husband. I intend to accept the recommendation made by the Probation Officer but I want to explain why to you and also indicate publicly why I took that view ... Your background has had a bearing on the course of your Iife... your parents' separation and your rejection by various people ...you impress as a vulnerable young woman ... I want to emphasise to you that imprisonment would normally be given ... you are being given this opportunity. It is up to you to make the most of it... (Viv· Drug Offence: Sentence = Supervision) I agree with the Probation Officer that you were an accidental offender. You were in a destructive relationship with the real culprit ... a sentence of imprisonment is not required. (Sally Drug Offence: Sentence::: Supervision) m Shirley was convicted of a serious violent offence, which would normally result in a sentence of imprisonment unless 'special circumstances' existed. In this case, 'special circumstances' were found and Shirley's offending was directly and quite explicitly blamed on her partner (cooffender) who the Judge described as a "dangerous criminal." Indeed, most of the Judge's 178 remarks in this case were dedicated to reinforcing just how 'bad' a criminal Shirley's cooffender was. No reference was even made to the fact that Shirley herself had nine prior convictions. The Judge also quickly dismissed the possibility that Shirley could have influenced her partner. She is presented as a "victim of circumstance" and as a puppet with no will of her own. This appeared to reduce the need for punitive sanctioning in favour of rehabilitation. The aims of deterrence, incapacitation and denunciation are subsequently rejected by the Judge, who argues that "the community's interests are better served" by a sentence which will rehabilitate Shirley. The need for sentencing to reflect the seriousness of Shirley's offence is recognised by the Judge and a degree of retribution is acknowledged but the 'special circumstances' of the case mean this is achievable outside the prison walls in the community. In mitigation of penalty it must be recognised that...your co-offender and partner was a headstrong and dangerous criminal ... in your relationship he was not only the dominant partner, he was the dominating partner. Your addiction to drugs, your obvious need for a loving and supportive relationship where your could find it, left you vulnerable to him and heavily reliant upon him ... 1 accept as a general rule that people involved in this type of offending should be imprisoned .. .ln your case, in my judgement, imprisonment is not required. There is no need to deter you from similar activity in the future. To a degree you were a victim of circumstance. The circumstance of your drug dependency and your reliance on your partner. He, as I said, was a dangerous criminal whose conduct you could not in the end influence, not in a determinative way .. .ln the end I think that the community's interests are better served by a sentence which rehabilitates you rather than a sentence which denounces your activity and operates as a general deterrent. A punitive aspect of your sentence is however called for but that can be addressed by incorporating it into a community based sentence.... 1am confident that you will not reoffend, at least not in a way such as this. (Shirley - Violent Offence: Sentence = Periodic Detention and Supervision) 3.3 The 'Family Man' and The 'Man Alone' For women, the family was a central theme in most Judges' sentencing remarks, with familial connections usually mitigating, but sometimes aggravating, sentence severity. In contrast, discussion about men's families rarely featured. The relative silencing of discussion about men's familial relationships is best illustrated by the exclusion of all talk pertaining to the care given by the three solo fathers in this study. Pre-sentence reports mentioned these men's status as sole care-givers but the focus was on their (in)ability to provide in an economic rather emotional and practical sense, for their children. In contrast to the women, Judges did not even mention these men's status as solo parents in their sentencing remarks. Men's care giving abilities did not seem to be valued highly perhaps because caring for children is not seen as a component of prevailing masculinity; men are meant to provide for, rather than care for, the family. 179 When Judges did talk about men's families this discussion usually appeared in the context of rebuffing suggestions that particular familial circumstance (past or present) should mitigate sentencing or to demonstrate how damaged men's families had become as a result of their criminality. In Greg's case, the Judge expresses surprise that he can maintain a relationship with his partner, given his criminal behaviour. I am surprised that your partner remains loyal to you ... That loyalty is continually abused by you and by your offending, in circumstances where you know that apprehension will inevitability lead to imprisonment. (Greg - Drug Offence: Sentence::: Imprisonment) A possible mitigating circumstance for Jim is the upset caused by a relationship break up. This would normally reduce women's blameworthiness and mitigate sentence severity but in Jim's case, these relationship problems are turned back onto him. The Judge argues that Jim's unruly behaviour did in fact cause the relationship problems to begin with. Both Greg and Jim are reframed as actors and makers of their own destiny, denying the need for rehabilitation. Undoubtedly you were, as you commonly are, affected by alcohol, possibly drugs, and certainly were in a high emotional state following the breakdown of a long standing relationship, that break down, however, really occurred for the very reasons I have just mentioned; drugs and alcohol. Your counsel suggests that I should chose a community based sentence .. I am afraid I do not consider it a viable option. (Jim· Property Offence: Sentence::: Imprisonment) Like current familial relationships, discussion about 'dysfunctional' childhoods was extremely rare in the case of men. If mentioned, presenting men as actors rather than victims quickly marginalises possible childhood trauma. Troubled childhoods did not excuse male offending and any attempts by men to present themselves as 'victims of circumstance' were quickly dismissed. Being held responsible for their criminality once again allowed sentencing to be rationalised on punitive, rather than rehabilitative, grounds. Harry, for example, had a 'difficult upbringing' but his suffering is constructed as rather trivial and unimportant in light of his serious offending: he goes to jail. I am told you had difficulties in your upbringing, but perhaps you have not suffered quite as greatly as others ... There is a limit in drug offending as to how much personal circumstances can be taken into account .. you were involved in serious offending ... 1sentence you to 12 months imprisonment. (Harry - Drug Offence: Sentence::: Imprisonment) 180 Wayne's childhood sexual abuse is also trivialised by the Judge and he is subsequently presented as a culpable actor rather than a victim. According to the Judge, the only reason Wayne is distressed by the abuse is because of his addiction problems; if he stops abusing substances, the Judge argues, then the sexual abuse would not be a problem and he would not offend. Wayne is held responsible for his own trauma, his own actions and is subsequently sentenced to imprisonment. After all, he committed a serious offence and the punishment given should reflect this: Counsel refers to some matters on your behalf - the sexual abuse which you suffered as a child; the head injury which you subsequently suffered; and the substance abuse ... 1 suspect that if you stopped abusing substances, including alcohol, neither of those other two factors would necessarily be influential in your continued offending. There is no doubt that this is serious offending and that imprisonment is the appropriate sentence. (Wayne· Violent Offence: Sentence = Imprisonment) Thus, the family is seen as either relatively unimportant in criminal men's lives or is simply better off without them. The former is not surprising given that masculinity in New Zealand situates men as actors in the public, rather than private, sphere. The latter is explained by most criminal men's inability to conform to dominant expectations of what 'Family Men' should be doing, such as providing financially for their families through paid employment. Instead, men are more easily positioned as 'men alone' because of their criminality, general unruliness, . unemployment, and lack of commitment to 'breadwinning'. The Probation Officer's report makes bleak reading. Your wife says you spend little time at home ... 1 suspect because you would rather pursue activities which you know she would disapprove of. Your previous employment obviously was affected by your alcohol and drug usage...The Probation Officer's recommendation is that I should chose a community based sentence, that periodic detention is the only viable option. I am afraid I do not consider it viable at all. (Jim· Property Offence: Sentence = Imprisonment) The 'man alone' represents a less controlled form of masculinity and must be kept in check. Punitive sanctions are easier to legitimate in these cases. However, the closer men got to fulfilling 'breadwinning' roles, by entering the sphere of paid public work, the more likely Judges were to speak favourably of them. Mark, for example, committed a serious violent offence and had a number of prior convictions, but despite this the Judge does not imprison him because Mark has a "good work record." Whilst the offence is serious ... 1 think you have made a favourable impression on the Probation Officer and I am prepared to have regard to your good work record ... I am prepared to accept that you have learnt your lesson .. (Mark· Violent Offence: Sentence = Periodic Detention) 181 Grant also had prior convictions but his criminality was similarly detracted from by his employment record and the Judge states he is willing to suspend the imprisonment term. Somewhat unusual for a person who faces a charge of this nature, you are also a person with a relatively good employment record. You have worked for most of the last eight years ... the managing director of the company speaks well of you ...there is no suggestion that your cannabis use habit has affected your ability to retain your positions with that company ... the courts have made it clear over the years that this type of offending must generally result in a sentence of imprisonment...a sentence of imprisonment is therefore, in my view, the appropriate sentence for this offence ... 1 must nevertheless consider whether a suspended sentence of imprisonment would be an adequate response ... in your case the factors which must be taken into account weight in the balance of your favour. In particular, your present and recent employment history. (Grant· Drug Offence: Sentence = Suspended Sentence of Imprisonment & Periodic Detention) 3.4 Women's Pathology and Men's Action Domesticity dominates the Judges' sentencing remarks and presented as the most popular way of reconstituting 'criminal women' as 'real women'. Less popular, but nevertheless evident, was 'refeminisation' via pathology and mental unwellness which constructed women's criminality as caused and excused by some sort of inner turmoil. Perhaps presenting women as 'troubled' rather than 'bad' allowed Judges to accommodate the criminal women before them within dominant ideologies about appropriate female emotionality and dependence. For example, Kay's and Anne's offending is presented as incomprehensible unless mental health problems exist, and while psychological assessments reject evidence of pathology in Anne's case, the Judge disputes these. Seemingly, to him, women's crime cannot occur as a rational response to a particular situation. In their roles as wives and mothers, women are held responsible for handling the emotions and distress of family members but as the repository of emotions, women are also easily depicted as "irrational creatures who cannot cope and need help" (Tavris, 1992: 275). It seems over a good number of years you have been treated as having some psychological problems. Nobody could understand anybody behaving in this particular fashion as you do unless they have some mental illness. (Kay· Property Offending) I have been troubled by the psychiatric report evidence. The report says - There is nothing in her history nor on examination that would suggest any form of psychiatric illness. She has not been depressed. There is nothing to suggest recent psychiatric epileptic activity and her medical history is otherwise uneventful'. I find that conclusion in itself surprising because the report refers to the fact that you attempted to commit suicide at the age of thirteen and if that is not an eventful episode in medical history I don't know what is. I also accept that you were depressed, and it may well be that the doctor's adopted a medical definition of depression beyond that of what laymen would regard as a state of depression, which I accept you were under at the time ... (Anne - Property Offending) 182 Focussing on their traumas constructed criminality as something outside of women's control. This could mitigate sentence severity because culpability was reduced. Although Judges may note the seriousness of women's offending at sentencing, references appeared as lip service. This made demands for retribution, denunciation, incapacitation and deterrence less likely. Rehabilitative sentencing aims become the primary focus, because female offenders present and are presented as needing help rather than punishment. Sherry, for example, was convicted of a serious violent offence and had a number of prior criminal convictions. The seriousness of her offending is noted by the Judge as "the type of crime" for which people are usually imprisoned. Still, Sherry is apparently not a 'usual person' and "imprisonment is not required" because there was no need to "deter her from similar activity in the future." Sherry has problems with poor self esteem and drug addiction; everyone's interests, the Judge argues, are best served by a non-custodial sentence. I accept as a general rule that people involved in this type of crime should be imprisoned ... ln your case, in my judgement, imprisonment is not required. There is no need to deter you from similar activity in the future ... ln the end I think the community's interest is better served by a sentence which rehabilitates you rather than a sentence which denounces your activity and operates as a general deterrenl...Your recovery, from your addiction to drugs and your restoration of some self-esteem are the important features and if that can be achieved then I am confident that you will not re-offend. (Sherry, Violent Offence: Sentence = Supervision) Sue had numerous prior convictions and her current offence is defined by statute as a serious violent offence, carrying with it a presumption of imprisonment. The Judge begins sentencing by acknowledging the seriousness of Sue's criminality, but then turns to her drug abuse problems. Sue's blameworthiness is subsequently reduced, she cannot be held responsible for her actions because she is a drug addict who was "clearly intoxicated" at the time of the offence. It is never suggested that Sue should in fact be held responsible for her drug abuse, instead it appears as something that just happened to her. You appear for sentence on a serious charge ... the complainant suffered a nasty injury...The presentence report demonstrated that you have a long history of drug abuse, as so many people who come before the court for sentence unhappily have. You were obviously intoxicated on the occasion with which I am concerned. There were elements of intoxication which are not relevant to guilt but are relevant to the question of sentence ...you were clearly grossly intoxicated .... that is not in itself a mitigating factor in accordance with the Act but I do not think I should wholly overlook it for the purposes of sentencing ... you escape imprisonment by a narrow margin. (Sue - Violent Offence: Sentence = Supervision) Judy's drug offending is also acknowledged by the Judge as a serious crime and one in which little weight is given to personal circumstance. Despite this, Judy's drug addiction is used to 183 mitigate sentence severity. Judy is not held responsible for her drug addiction, which is presented as an unfortunate situation in which Judy just found herself. You have a long history of drug abuse ... 1have regard to the pre-sentence report and to what your lawyer has said on your behalf. You are still apparently using drugs ... this is a dilemma in which you have unfortunately found yourself. The principles in cases of this sort are normally very clear ... people who commit these types of crimes go to jail and little weight can be given to personal circumstances ... but in this case I accept you have not offended for a SUbstantial period and that in combination with your personal circumstances leads me to the conclusion that in this case a term of imprisonment is appropriate but suspended. (Judy - Drug Offence: Sentence = Suspended Sentence of Imprisonment) Even when women were sentenced to imprisonment, reduced terms were rationalised by Judges via constructions of women as 'troubled' rather than 'bad'. Dianne had an extensive criminal history including numerous convictions for violence and her present victim nearly died as a result of Dianne's assault on her. In this case, it was practically impossible for the Judge not to imprison Dianne because statute requires it. However, even here the Judge rationalises the sentence in terms of rehabilitation. He states that he does not "ignore" Dianne's "needs" because "help is available in prison." It would seem that framing women in terms of more punitive sentencing aims is difficult when dominant feminine ideology ensures women are seen as weak, dependent and in need of help. Despite having little choice in imprisoning Dianne, discretion in setting the imprisonment term is still available to the Judge and it is at this point that her "deep seated emotional problems" and "drug dependency" are identified as mitigating factors: It is acknowledged by your lawyer that s.S. of the Criminal Justice Act applies and that the circumstances are such that a custodial sentence is appropriate ... 1do not ignore the various factors that have been very properly urged upon me but I think that their proper effect is to make shorter the imprisonment term that would otherwise be appropriate .... She has a longstanding drug dependency and deep seated emotional problems ... 1do not ignore that the prisoner needs help but a custodial sentence does not shut the door on that. Help is available in prison ... After taking into account the mitigating factors ... 1sentence you to twelve months imprisonment. (Dianne - Violent Offence: Sentence = Imprisonment) Similar to Dianne's situation, the Judge says that Sarah's crime is serious and one for which personal circumstances would not usually be considered. Still another exception is made and Sarah's imprisonment term is "much shorter than would otherwise have been the case" because of her "long standing drug problem," "the effect of the arrest" and her general mental state. This construction disassociates Sarah from her offending, subsequent arrest, and her drug problem which are all presented as having been 'done to her' rather than as something Sarah caused. Sarah's blameworthiness and criminal responsibility are subsequently reduced because she is not construed as an acting, thinking, human being; she is successfully 'refeminised'. 184 I have regard to your counsel's submission in relation to the fact of your long-standing drug problem, what is said to have been the effects of the arrest on you, and in general terms, your state of mental health. The sentencing principles in cases like this are abundantly clear. The first is that the courts are to attach little weight to personal circumstances and secondly, that drug dealers or those who prepare to deal in drugs, almost invariably go to jail. .. 1take the view that that a sentence of imprisonment is the only appropriate sentence in this case but, because of the mitigation factors I have mentioned, the sentence will be much shorter than would otherwise have been the case .... (Sarah - Drug Offence: Sentence = Imprisonment) In the Judge's sentencing remarks, men were more likely than women to be presented as actors and definers of their own destiny, often devoid of feeling, vulnerability and weakness. Once again I present the case of Wayne, who by all accounts suffered much trauma in his life. Wayne has a substance abuse problem but this did not reduce blameworthiness, as was often the case with women's sentencing. Instead, Wayne is held responsible for his own trauma, his substance abuse, and subsequently he is made accountable for his criminality. Counsel refers to these matters on your behalf - the sexual abuse which you suffered as a child; the head injury which you subsequently suffered; and substance abuse ... 1 suspect if you stopped abusing substances, including alcohol, neither of those other two factors would necessarily be influential in your continued offending ... there is no doubt that imprisonment is the appropriate sentence given the seriousness of this assault. (Wayne, Violent Offence: Sentence = Imprisonment) Men were more likely to be constructed as 'bad' rather than 'troubled', possibly because 'badness' and criminality are ideologically intertwined with masculinity itself. Judges need not reconcile criminal men with dominant gender ideology because criminality is consistent with manliness. Reconciliation is instead required when psychiatrists, Probation Officers, or defence lawyers present criminal men as pathological because this challenges dominant ideas about masculinity. While female offending was presented as virtually incomprehensible unless mental health problems exist, men's mental health problems appeared as simply unbelievable to Judge's who either rejected or ignored them. For example, David was convicted of a serious violent offence, he had a number of prior criminal convictions and a Psychiatrist recommended supervision because he was "a young man who has been deeply and adversely affected by traumatic childhood experiences." The Psychiatrist believed that "imprisonment would further damage an ah"eady severely damaged young man." The Judge acknowledges this recommendation but rejects it because there is no specific psychiatric disorder and the public interest would be best served if David was imprisoned. I have carefully read more than once the Psychiatric report but I am afraid that I cannot possibly agree to the proposition that you should receive Supervision ... you have no major psychiatric disorder and the 185 doctor has no specific psychiatric recommendation to make, although he did urge me to adopt the Supervision approach ..... a sentence of imprisonment must in my view be imposed in the public interest. (David, Violent Offence: Sentence::: Imprisonment) When men's mental health problems were recognised by the Judges they were, in contrast to women, rarely seen as a mitigating circumstance. Women's offending was often detracted from with references to drug abuse, and general mental health problems. Men's drug abuse and mental health problems in contrast, were minimised with reference to the seriousness of their offending. This can be seen in the cases of Andrew, Ben, Simon and Mike as rehabilitation is rejected because the offences were serious enough to make imprisonment inevitable, regardless of drug addiction, depression, mood swings, or being under the influence of drugs and alcohol at the time of the offence. It is perfectly clear from the psychiatric reports that have been received that you have suffered a depressive illness and the court is always sympathetic to that kind of circumstance. Never the less this was a serious offence and you will be imprisoned. (Andrew, Property Offence: Sentence::: Imprisonment) It is recognised that he had a substance abuse problem and is prone to mood swings and depressive episodes, which may require treatment. However, s.S. of the CJA apples and a sentence of imprisonment is inevitable. (Ben, Violent Offence: Sentence = Imprisonment) There is a psychiatric report, which confirms that the prisoner is not suffering from any mental illness but recommending that help be given in connection with his alcohol problem. Other personal circumstances are set out in the pre-sentence report. By s.12 of the Criminal Justice Act the court is not to take into account the fact that the offending occurred under the influence of alcohol or drugs ... it is my view that the total effective sentence ought to be one of four years imprisonment. (Simon, Violent Offence: Sentence = Imprisonment) You clearly have a long term drug addiction but of course it is well known that very little weight can be given to personal circumstances in cases of this nature...you with your connection with the drug scene would be well aware of the consequences you faced when you engaged in this sort of activity and the court has to have regard to the deterrent aspect...1 am of the view that a sentence of imprisonment is called for in cases of this nature unless there are exceptional circumstances requiring leniency... 1 am satisfied that there is no such special circumstances. (Mike, Drug Offence: Sentence::: Imprisonment) Focussing on men's criminality denied their experiences of mental infIrmity, denied them a legitimate reason for offending, held them accountable for their actions and presented them as more dangerous than women. These constructions helped to rationalise punitive sentencing aims over rehabilitation. When reading the Judges' sentencing remarks for men I was constantly struck by the focus given to the seriousness of men's criminality, the danger they 186 posed to public, the need to pull men out as cases worthy of public condemnation, and the need to teach men a lesson so they would not offend again. The case of Bill, presented below, provides a good example of how, in the case of men, rehabilitative sentencing ideals are rejected in favour of retribution, deterrence, denunciation and incapacitation with the resulting outcome being imprisonment. I have read the pre-sentence report and I have carefully read more than once the Psychiatric Report ... 1 agree with counsel that I must weigh your personal circumstances against the public interest. I have done that. I have given this case anxious thought as to whether I could properly adopt the drug rehabilitation option. I agree that unless your drug habit is beaten you will remain a public danger ... ln any event I am satisfied that a prison term must be imposed in the public interest... I agree essentially with the Crown that although rehabilitation and assistance are important for you personally, the public interest must be considered and punishment and deterrence must come first...When all is said and done this was a serious offence ... 1accept that it was somewhat unsophisticated but in my view this type of offending is a serious public evil...You represent a danger to the public who are entitled to be protected from your activities for a period ...The sentence of the court is accordingly that you be imprisoned for four years. (Bill, Violent Offence: Sentence = Imprisonment) Thus we can conclude that while personal traumas may mitigate women's sentences, they are relatively unimportant for men who were more often seen as actors and controllers of their own destiny. It follows then, that to be granted judicial leniency, men would need to demonstrate action in changing their deviant behaviour. This is illustrated in the example of Andrew who is seen to be making positive changes in his life and is subsequently able to avoid imprisonment. The Probation Officer has recommended a suspended prison sentence ... of course, it does not take me to tell you that you are facing a serious charge and that in the normal course of event a charge of that nature results in a term of imprisonment without suspension. In your situation, however, there are mitigating circumstances to be taken into account including the positive features of the probation report which indicate that you may have made a change in direction which will lead you away from offending ... it seems you are keen to find employment...it seems you have also taken steps to distance yourself from those involved with drugs; the Probation Officer thinks that it is likely that you are no longer involved with drugs. That, of course, is capable of representing the turning point in your Iife... 1 am prepared to give you the benefit of the doubt on this occasion. (Andrew, Drug Offence: Sentence = Suspended Imprisonment Term) 4 Closing Comments In previous chapters it was found that men and women sometimes received different judicial outcomes and that the decision-making process was also differentiated by sex. In the current chapter, it was concluded that an understanding of these sex differences might be obtained by investigating how gender impacted on Probation Officers' pre-sentence reports and Judges' sentencing remarks. Although prior statistical analysis coded and controlled for gender related statuses (such as familial situation, income, health and negative life experiences), I was concerned that my coding schedule may not have fully captured the complexity of gender and 187 how it operated in judicial processing. Thus a discourse analysis of both the pre-sentence reports and sentencing remarks was undertaken. The purpose was to understand how constructions of gender in judicial decision-making impacted on sentencing outcomes. Remand outcomes could not be considered because of limited documentation, but prior statistical analysis suggests that gender constructions similar to those found at sentencing would also be found at remand (see Chapter 5). 12 New Zealand society is a gendered society; a culture in which structures of masculinity and femininity are central to the formation of society as a whole (James & Saville-Smith: 1994). New Zealand's criminal justice system is an institution operating in, and thus logically affected by, this wider gendered context. It is not surprising therefore to find evidence that judicial discourse is also a gendered discourse; dominant ideas about masculinity and femininity were found to permeate Probation Officers' and Judges' discussions. This impacted not only on what was said but what was not said with Probation Officers and Judges screening or presenting information differently for men and women in accordance with gender ideologies. In particular, the family and mental health were identified as key sites of variance where explanations for sex differences in judicial outcomes could be found. Female offenders challenge appropriate ideas of femininity through their criminality and involvement in the criminal justice system, both of which are traditionally the domain of men. Thus, when confronted with criminal women, the justice system may see them as either 'not women' or 'not criminals' (Worrall, 1990: 31). Constructing women within dominant ideals of femininity in relation to the family and mental illness provides a way to 'refeminise' offending women as 'real' women and not really criminal after all. For example, women were presented as nurturers, dependants, pathological and victims of circumstance. This neutralised their dangerousness, blameworthiness and responsibility, making punitive sanctions seem less appropriate. Judicial constructions of male offenders were also bound by dominant masculine assumptions but these usually made punitive sanctions more, rather than less, likely. Dominant discourses of masculinity focus on badness, disruption, and criminality. There is no need to reconcile men within dominant gender ideology because criminality is consistent with 'manliness'. 12 Thus, For example, in some cases familial situation was found to strongly predict women's remand status but not men's (see Table 64). 188 judicial sympathy was rarely extended to men because most were seen as a threat to the social order and in need of state-controlled regulation. A few men were able to demonstrate a more socially controlled yet acceptable form of masculinity via their involvement in paid public work. Employment was beneficial to men especially if they had families to support financially. Being seen as a 'hard worker' and controlled by 'breadwinning' commitments often decreased men's chances of imprisonment. Acceptable ideals of masculinity require men to be providers rather than carers of families so, unlike women, men's child care responsibilities were rarely discussed or used to legitimate sentencing leniency. Similarly, constructions of acceptable femininity presented women as carers rather than providers, so women's employment was rarely discussed or used to mitigate sentence severity. Dominant discourses of femininity also ensured that pathology, emotionality, inner turmoil and trauma were discussed and used to excuse women's criminality, while detracting from their potential to be dangerous. These constructions helped to rationalise rehabilitation over punishment. Judicial presentations of men, on the other hand, supported a masculine ideology, denying men of feeling, vulnerability, weakness and the general right to experience mental unwellness. Men were instead placed in the domain of human action, being presented as actively adopting an offending lifestyle or at fault for not ridding themselves of their criminality. Constructing men as powerful actors by presenting them as definers of their own destiny, meant that criminal men were more likely to be held responsible for their actions and to be seen as dangerous. Primacy could therefore be given to punitive sanctioning over rehabilitative measures. Thus, judicial discourse constructs men and women differently and in accordance with dominant gender ideologies. The criminal justice system, it would seem. finds it difficult to see offenders as genderless, which is not surprising given the centrality of gender in our society. This does not mean that sexual disparity in judicial outcomes is the result of direct or conscious sexism on the part of either Judges or Probation Officers. The relationship between gender and sentencing is socially constructed, being actively produced through an interactive process that includes offenders, Probation Officers, Judges, other judicial actors and the public. This process results in different theoretical and practical criteria being fulfilled in the process of 'being a woman' or 'being a man' which in turn enables different judicial outcomes (Allen, 1987b: 114). 189 This does not mean that the gendered nature of criminal justice decision-making is necessarily 'correct' or 'fair'. Indeed, both men and women are to different extents and in different ways, debilitated by social constructions of gender. Men are denied reasons for their offending and excluded from feeling, caring, nurturing and talking. The possibility that criminal men are powerless is ignored and this ultimately denies men understanding and the right to heal. Women, on the other hand, are denied the right to be thinking, acting and purposeful human beings. For women, powerlessness and dependency are embraced and they are essentially denied the right to self-determination. 190 Chapter Eight Closing Comments ".. .ideological and political processes which assert and sustain the authority ofnormative heterosexuality... have powerful consequences for both men and women. They demand that men be tough, that they 'take it like a man' and are 'tougher than the rest~ And whether this be expressed in terms ofphysical prowess or mental prowess, it has a damaging effect on men, their emotions and the quality of the relationships they form around them. It also has a profoundly damaging effect on women. // (Walklate, 1995: 186-187) At the beginning of this thesis I posed three research questions: First, are sentencing and remand outcomes different for adult men and women? Second, are the criteria used for deciding these outcomes gendered (i.e. are different factors considered when determining men's and women's judicial outcomes and are certain 'types' of men and women more likely to be extended judicial leniency)? Third, if sex differences in sentence and remand do exist, how can they be accounted for? These research questions arose from a more general criminological debate, namely whether or not women are, in comparison to men, treated preferably in the criminal justice system. While this debate has continued for many years outside of New Zealand, in this country little substantive research or discussion has taken place. Traditionally, international argument about gender difference in the criminal justice system is rooted in Otto Pollak's (1950) claim that the male judiciary extends chivalry to women and as a result, sanctions women less harshly than men (see, Tjaden & Tjaden, 1981: 74-75; Eaton, 1986: 22; Cain, 1990: 2; Culliver, 1993: 4; Belknap, 1996: 70-71, Newbold, 2000: 65-66). Pollak based his conclusions on raw crime statistics and thus failed to take into account differences in men's and women's criminality and lives. Women's law breaking is generally less serious than men's and there are fundamental gender differences in the lives of the two sets of offenders (e.g. child care responsibilities). Pollak's critics have subsequently argued that these factors may account for the apparent leniency which Pollak saw being extended to women. Since the 1970's, many researchers have examined judicial sanctioning to ascertain whether sex based disparities in court outcomes are real or simply an artefact of other differences between men and women. After controlling for differences in men's and women's lives and criminality, this research has produced mixed findings. Some have shown that 'preferential' 191 treatment may not be extended to all women, over all men, all the time (e.g. it could depend on fulfilling gender-role expectations). Yet most of this research finds that a degree of differential treatment on the basis of sex remains. First, men appear generally to be treated more 'harshly' than their female counterparts, particularly at the point of pre-trial release (remand) and sentencing (see Parisi, 1982, Nagel & Hagan, 1983, Daly and Bordt; 1995). Second, different factors are often considered when determining men's and women's judicial outcomes (Edwards, 1986: 80; Belknap, 1996: 71-85). 1 Current Research Findings Using both a statistical and a case-study analysis, the current research found patterns similar to those identified in the international literature. Overall, it was found that a) sentencing and remand outcomes often differed for adult men and women, with the former usually receiving 'harsher' sanctions and, b) the criteria used for deciding these outcomes were also gendered (i.e. different factors were considered when determining men's and women's judicial outcomes and certain 'types' of men and women were more likely to be extended judicial leniency). In explanation, gendered ways of viewing, understanding and judging offenders indicated how men and women come to receive different treatment. The results are summarised below. 1.1 Statistical Analysis and Findings The statistical study presented in Chapters 2-5, compared the sentencing and remand outcomes of 388, or 194 matched pairs, of adult men and women sentenced in the Christchurch High and District Courts in New Zealand between January 1990 and February 1997. The analysis was conducted and presented in four stages. In addition to assessing problems of skewed distributions, multicollinearity and range restriction, stage one of the statistical analysis (see Chapter Two) identified similarities and differences between men and women in the statistical sample. It was found that factors relating to court processing (e.g. plea) and the content and context of offending (e.g. offender role) were generally similar for men and women. This was partially due to the matched sampling method which also allowed offenders to be matched closely by age and ethnic group. Despite similarities, some sex differences were still found in the statistical sample. Women were significantly less likely than men to be involved in paid work, were more likely to have childcare responsibilities, and generally experienced stronger familial ties. Men's criminal histories were significantly more extensive and serious than women's. Sex differences were 192 also noted in the content of pre-sentence reports. Poor health and negative life experiences were identified significantly more often for women. In stage two of the statistical analysis, zero-order correlations and hierarchical multiple regression results were presented to ascertain the extent of sex differences in sentencing and remand outcomes (when controlling for other socio-demographic variables) and to identify other major variables affecting sentencing and remand outcomes, which might explain sex differences in judicial outcomes. Results showed that men and women had approximately equal chances of receiving suspended sentences of imprisonment. Small sex differences were noted for work and care-based sentences but the results were not statistically significant. Pronounced and statistically significant sex differences were found in the decision to imprison, length of imprisonment term, the remand status decision, length of custodial remand and bail conditions (see Chapter Three). In general, women received less severe sanctions than men - they were less likely to be remanded into custody, if remanded into custody they remained there for shorter periods, if remanded on bail they were less likely to be given special bail conditions, at the point of sentencing women were less likely to be jailed, and if imprisoned their terms were shorter than men's. However, other major variables affecting sentencing and remand outcomes, which might plausibly explain how men and women came to receive these different sanctions, were also identified at this point in the statistical analysis (see Chapter Three). Criminality, remand status, and the content of Probation Officers' pre-sentence reports (e.g. pre-sentence recommendations, offenders' life experiences, health problems), all significantly impacted on certain judicial outcomes and each in turn differed by sex. Thus a critical question was raised, namely: to what extent was sex exerting direct effects on judicial decision-making, or indirect effects through its impact on mediating variables (such as criminal history, presentence reports and remand status)? In other words, were Judges treating men and women differently simply because they were men and women, or was disparity arising out of variables like seriousness of criminal history - which just happen to differ according to sex? Path analysis was subsequently used in stage three of the statistical analysis to establish whether sex exerted a direct or indirect effect on sentencing and remand (see Chapter Four). In all but one case (the imprisonment sentencing decision), sex continued to impact directly on the judicial outcomes investigated (length of imprisonment term, remand status, length of custodial remand, bail conditions). That is, sex differences in variables previously found to be influential (i.e. seriousness of criminal history, length of custodial remand, pre-sentencing 193 recommendations) failed fully to explain why women's judicial outcomes were generally less severe than men's. With other factors statistically controlled, women's imprisonment terms remained substantially shorter than men's. For combined offences (drug, violent, property) women's imprisonment terms were approximately 8.5 months shorter than were men's. Female violent offenders received imprisonment terms approximately twelve months shorter than men and female drug offenders' terms were around five months shorter. Women were 14% less likely than men to be remanded in custody for combined offences and 24% less likely for property offences. Of those offenders remanded into custody, men remained there for 42 days longer than women (on average) for combined offences and 26 days longer for drug offences. Finally, for those offenders remanded on bail male property offenders were 40% more likely than women to be given special bail conditions and although non-significant, men were 8 % more likely than women to be given special bail conditions overall (combined offences). Sex-based disparity at the point of sentencing (length of imprisonment term and the decision to imprison) was also found to be affected by gendered decision-making earlier in the judicial process which, at least to some degree, impacted on final sentence. Path analysis results showed that men's remand outcomes were more severe than women's regardless of sex differences in other key factors (e.g. seriousness of criminal history) and this often increased the severity of men's sentences. Severe remand outcomes also increased men's chances of receiving a custodial pre-sentencing recommendation which in tum aggravated final sentence severity. The gendered nature of pre-sentence reports (e.g. being more likely to identify women as having poor health or as having had negative life experiences) reduced the severity of offenders' final sentences in some cases. These general patterns were evident when imprisonment terms were set and when Judges decided to sentence an offender to jail. In the latter case, as already noted, sex did not have a direct impact on the decision to imprison because sex was mediated through other key variables. Results showed that Judges were more likely to imprison men because men, compared to women a) had more serious criminal histories, b) had lengthier custodial remand periods and, c) had less favourable pre-sentence reports (e.g. sentencing recommendations, references to health problems). When these factors were controlled, sex differences in the decision to imprison became statistically non-significant (combined .6%, drug .2%, property 14%).1 Therefore, unlike imprisonment terms, gender 1 It is probable that small sample size impacted on measures of statistical significance in the case of property offenders. While results showed that men were 14% more likely than women to be imprisoned, this was non-significant so disparity could not be concluded. 194 difference in the initial decision to imprison could not be attributed to sentencing Judges, but to gender differentiation occurring earlier in the process. In the last stage of the statistical analysis, an investigation was undertaken to ascertain whether the decision-making processes, rather than outcomes, were differentiated by sex. First, results suggested that the criteria used for deciding judicial outcomes differed for men and women. Second, there was evidence that certain 'types' of men and women were more likely to be extended judicial leniency. Support was provided for the argument that women whose crimes conflict with dominant gender ideals will not be extended judicial leniency (Edwards, 1986: 80). Committing a drug offence in a public, as opposed to a private setting, substantially increased the length of time women spent in custodial remand. Women who victimised men also had their chances of getting an imprisonment sentence increased. Aside from these two factors, criminality and judicial processing variables were generally more likely to disadvantage men. Serious criminal histories, causing injury to victims, victimising women, using weapons, playing an active role in the offence, having co-offenders and pleading not guilty, impacted substantially on men's judicial outcomes. It is possible that the sex differences reported in my statistical analyses were artificial in some way or reflected an inability to capture gendered differences in offenders' criminality and lives using a numerical coding schedule (see Daly, 1994; Daly & Bordt, 1995). It may actually be the case that after a further, more in-depth inspection, gender differences in judicial processing are found to exist for other reasons (for example, Judges may be sensitive to the degree of informal social control in women's lives). 1.2 Case-Study Analysis and Findings To investigate this possibility and illustrate the manner in which disparities occur, a closer case-study analysis was conducted. One-hundred individual offenders (50 pairs) were selected from the larger sample used in the statistical analyses (n = 388).2 Pairs sentenced for the same statutory offence, in the same court (District or High) with similar offence characteristics, criminal histories, pleas and biographical details (age and ethnicity) were examined and the best matched pairs were then chosen for inclusion in this case-study sample. In this process, all pairs selected were sentenced for the same statutory offence, in the same court after which matching priority was given to offence characteristics, criminal history, and plea. To enable generalisations from the case-study sample to the larger original sample, 2Time constraints made it impossible to look at all 388 cases in detail. 195 statutory offence and court distributions similar to those in the statistical sample were maintained. In the first stage of the case-study analysis (see Chapter Six) I examined the crime details or stories for the 50 pairs of offenders to determine whether criminality of similar seriousness resulted in similar judicial outcomes for men and women. The purpose of this analysis was simply to confrrm (or otherwise) the patterns disclosed in my prior statistical analysis. A pairwise (men vs women) comparison of criminality was undertaken using factors recognised as important in judicial decision-making (e.g. victim provocation and vulnerability of victims). Cases were grouped using a framework similar to that utilised by Daly (1994). Judicial responses were either considered similar, different or disparate. A disparate response was highlighted when it was not immediately plain why two people had received different punishment for criminality of comparable seriousness, or the same punishment for criminality of different seriousness (Daly, 1994: 239). Disparate sex-based outcomes were found in the majority of case studies, thus confrrming my statistical [mdings. In most of these disparate cases, men either received different or more severe punishments for criminality of comparable senousness to women, or women received the same punishment as men for more serious criminality. Second, in order to reveal the process by which disparity creeps into the judicial process, I conducted a case-study analysis of Probation Officers' pre-sentence reports and Judges' sentencing remarks. I considered how Probation Officers and Judges screened and/or weighted factors differently according to gender when they made their decisions. In comparison to my prior statistical analyses, these case studies produced a more subtle, complex and meaningful account of how gender operates in judicial processing, producing different outcomes for men and women. Results showed that dominant ideas about masculinity and femininity permeated Probation Officers' and Judges' discussions. Gender impacted not only on what was said but on what was not said. Probation Officers and Judges screened and presented information differently for men and women in accordance with dominant gender ideologies. What emerged were two gendered ways of viewing, understanding and judging offenders and this explained how men and women came to receive different judicial outcomes. In particular, the family and mental health were identified as key sites of gendered variance. Here, female offenders tended to appear as nurturers, dependent, pathological, emotional, traumatised and as victims of circumstance. This neutralised their dangerousness, blameworthiness and 196 responsibility, making punitive sanctions less appropriate. In contrast, male offenders were typically bound by dominant masculinity which presented them as dangerous, responsible for their own actions and thus in need of harsh judicial sanctioning. Men were essentially denied excuses for their offending. Trauma, victimisation, powerlessness and the general right to express emotion and talk about feelings were usually absent for discourses sUlTounding men. Instead, criminal men were presented as bad, disruptive, powerful and self-determining. 1.3 The Current Results and New Zealand's Criminal Justice System It could be argued that the current research findings are specific to the Christchurch area. Indeed, a study conducted by Heather Deane (1995, 1997, 2000) in the Wellington and Porirua District Court produced no evidence of gender difference in sentencing. However, Deane's [mdings are vitiated because of serious flaws in her research design and analysis (see Chapter One). Her results, therefore, cannot be relied upon. Furthermore, Deane's data contradict a more thorough New Zealand-wide statistical analysis (n = 300,000) conducted by the Ministry of Justice which showed that men were sentenced more 'harshly' than women, even when they appeared for sentencing under similar circumstances (Triggs, 1999). The Ministry's research supports the findings in the CUlTent study and suggests that my results are applicable nation-wide. 2 Theoretical Explanations The human world is fundamentally gendered. Thus, it is hardly surprising that New Zealand's criminal justice system is not a gender-neutral institution but instead presents as a site where gender, and its associated role expectations, are actively produced in the same way as in other social spheres. An analysis of the pre-sentence reports and Judges' sentencing remarks makes it clear that the relationship between gender and judicial decision-making is something that is being actively generated though an interactive process which includes offenders, Probation Officers, Judges, other judicial actors and the public. Essentially, different social processes and criteria are being fulfilled in the process of 'being a woman'or 'being a man' and this in turn results in Judges and Probation Officers making different decisions for male and female offenders (Allen, 1987b: 114). How then does my analyses fit with theoretical notions of chivalry, paternalism, social cost and social control? To recap the theoretical debates presented in Chapter One, it will be recalled that the 'chivalry' thesis proposes that men are socialised to behave towards women in a fatherly and protective manner and female offenders are often compared to mothers and wives, whom the male judiciary cannot imagine behaving in a criminal way. The presumption is that offending 197 women are placed on pedestals, treated gallantly and protected from punishment, with the result that their criminal activity is less likely to be detected, reported, prosecuted, or sentenced harshly. Theoretically, claims of preferential treatment benefiting women, as suggested by the 'chivah'y thesis', have been challenged after researchers found that preferential treatment was not extended to all women, over all men, all the time. For example, an American study by Nagel, Cardascia, & Ross (1982) found that married women were less likely than unmarried women to spend any time imprisoned. In response, 'chivalry' was criticised for being simplistic and reducing understandings of gender in judicial processing to "the superficial elements in male-female relationships, namely, the social amenities" (Moulds, 1978: 418). The 'paternalism' argument sought to explain these varied findings by moving beyond the social amenities and acknowledging gender-based power relationships. Judicial treatment, this perspective argues, may depend on conforming to dominant gender role expectations. In this regard, 'preferential' treatment may not be so preferential after all because gender roles, which, in the wider scheme of things, disadvantage women, are being actively reinforced. However, 'paternalism', like 'chivalry', may also be theoretically simplistic. Some researchers now reflect more on differences in men's and women's place in the social structure (e.g. as care givers, paid workers, victims, dependants) to ascertain how, and whether or not, the sexes should be treated differently. Linkages between gender ideology, social control, social cost and gendered differences in men's and women's criminal and non-criminal lives, have been highlighted as plausible and possibly justifiable reasons for differential treatment. With regard to the current research, each theoretical perspective is arguably useful because none is necessarily contradictory to the other (Crew, 1991: 61). It is conceivable that female offenders benefit, in part, from legal officials who treat them gallantly and are reluctant to inflict harm on them - that is, who are chivalrous. Furthermore, my research shows that sanction severity is not always reduced for all women, over all men, all the time. Treatment often depends on whether or not men and women appear as 'appropriately' masculine or feminine, which in turn links closely to dominant gender-role expectations (paternalism). Finally, differences in the constructions of men's and women's lives also playa part in judicial sanctioning (social cost and control). Women often avoid harsh sanctions because of their status as nurturers and because their position within the family unit is both stabilising and restrictive in its own right. Conversely, men avoid harsh sanctions if they are employed and committed to 'breadwinning'. However, the locus of social control and cost is clearly gendered. As demonstrated in the cases we saw of men who were solely or primarily responsible for childcare, their capacity to give care and experience familial social control was all but ignored 198 in judicial discussions. Similarly, (un)employment and breadwinning is not prioritised in discussions about women's lives (in contrast to men's lives). Thus, evidence from the current research suggests that elements of paternalism, social control and cost may operate in conjunction to produce different outcomes for men and women. Furthermore, the current study finds, as Daly (1994: 260) did, that "boundaries between victim and offender were more often blurred in the women's social histories". This result is consistent with international research which also [mds that gendered constructions of women as 'troubled' can partially account for why judicial sanctions are less severe than men's (see for example, Allen, 1987a, 1987b, 1987c; Daly, 1994). However, chivalry, paternalism, social cost or social control do not directly address the gendered nature of the 'troubled' vs 'untroubled' offender and the subsequent impact this has on judicial decision-making. While beyond the scope of the current thesis, it appears that a more integrated theory of gender and criminal justice decision-making is required. 3 Equality or Difference Showing that courts treat men and women differently ultimately leads to the question of so-what (Cain, 1990: 2)? Do we interpret these gender-based differences as warranted or unwarranted? Such questions have divided feminist scholars whom, until recently, have continued to debate whether or not gender equality is necessarily a good thing. On one hand, it is argued that since fundamental differences between the sexes actually do exist, treating men and women the same may be problematic because it will further disadvantage an already disadvantaged group. Reflected in social control and cost discussions, this viewpoint calls for recognition that men's and women's societal positioning is different; that there are "genuine physical and social differences" between the sexes which tend to disadvantage women over men (Smith, 1993: 7). For example, women's lives are more likely to be characterised by poverty, domesticity, victirnisation and dependency. Thus, in recognising the differential needs of men and women, it is proposed that women should receive differential treatment "so long as women are not placed in a more negative position" as a result i.e., as long as it does not disadvantage them (Chesney-Lind & Pollock, 1995: 156). On the other hand, some feminists consider differential treatment as problematic in that it reaffirms men's dominance over women. Reflected in the previous discussion of paternalism, this view posits the judicial protection of women as an ideological front for patriarchy in that 199 traditional ideals about women as 'naturally' domestic, dependent, weak and emotional are perpetuated (Daly, 1989b: 12; Smith, 1993: 7; Smart, 1980; 293). Ultimately, this may result in extensive personal, psychological, social, economic and political damage to women's fight for self-determination and equality. MacKinnon (1987: 38-39) argues that for "women to affIrm difference, when difference means dominance, as it does with gender, means to affirm the qualities and characteristics of powerlessness." Equalisation with men is subsequently proposed because to accept difference may result in women being seen as "different from" and thus "less than" men (Chesney-Lind & Pollock, 1995: 156). More recently, there has been a call from feminist writers to transcend the old equality/difference debate because both approaches present men as the standard against which "both actions by and treatment of females are measured" (Cain, 1990: 2; see also Smart, 1995: 42). Men are presented as the "norm, as the human standard", while women appear as nothing more than the "interlopers into a world organised by others" (Naffine, 1995: 14-15). Smith (1993: 8) describes this as follows: ... Iook back at the statement that the question is whether women, being basically similar to men, require equal treatment, or being significantly different from men, require special treatment. What may not be obvious is that this essentially means, Heads I win, tails you lose. That is, it assumes the outcome in advance, for to agree that if women are 'different' (Le. different from men) they will require 'special treatment' is to assume a male or patriarchal standard of what normal treatment is. In the style of Catch-22, the male-centred equality/difference debate IS now considered problematic because whether they are treated differently or similarly to men, women are ultimately disadvantaged (Henderson, 1991: 414). The difference stance "nourishes a crude socio-biology" whereas the equal treatment stance can and has been used to the detriment of women (Smart, 1989: 84). For instance, in parts of the United States, recent sentencing reforms based on male models of justice have been implemented to eliminate 'disparate' sentencing outcomes. 3 Mandatory sentencing minimums, 'get tough attitudes' and Draconian sentencing guidelines which seldom allow consideration of offenders' social situations (such as age, education, vocational skills, mental and emotional condition, physical condition, previous employment record, family or community ties, family or community responsibilities) have virtually eliminated sex differences in sentencing. As a result, there has been a dramatic increase in women's imprisonment and incarceration terms (see Raeder, 1993; Chesney-Lind & Pollock, 1995; Chesney-Lind, 1997; Hartley, 1999) Raeder (1993: 922) notes that "both the 3 These reforms were intended to reduce class and race disparities. 200 number and percentage of sentenced women offenders" are "growing at a faster rate than that of males" and this cannot be explained by increases in women's crime or arrests. Furthermore, the idea that prison is a "harsher and more unusual punishment for women than it is for men" (Heidensohn, 1986: 292) is suggested by many criminal justice researchers both internationally (see for example Singer, 1973; Goetting & Howsen, 1983; Elliot & Morris, 1987; Genders & Player, 1987; O'Dwyer, Wilson & Carlen, 1987; Pollock-Byrne, 1990; Chesney-Lind & Pollock, 1995; Chesney-Lind, 1997; Carlen, 1998; Flavin, 1998) and in New Zealand (see Phillips, 1992; Hamilton, 1995; Caird, 1999). The reasons for this argument are several. First it is said that, educational, vocational, recreational programmes and facilities available to males far exceed those available to females. Prison programmes and the everyday regime of the prison further tend to emphasise a form of repressive refeminsation where conformity to 'conventional femininity' is enforced (Heidensohn, 1986: 292). Second, the location of many women's prisons has also been identified as a problem, with the majority of women, compared to men, serving their sentences in prisons many kilometres from home, and away from their family and friends. Phillips (1992: 227) notes that the "inevitable consequence is that they [women] are unable to maintain links with the people who could offer them support or to have regular visits from partners or children." On release it is thus more difficult for the women to re-establish these links. Efforts are made to hold male inmates close to family and friends and women need to be extended the same option. Third, rules and regulations within women's prisons are notably stricter, and cover more petty details than for men (see for example, Singer, 1973; 302; Pollock-Byrne, 1990: 98; Hamilton, 1995: 12-16 & 114-115). Fourth, histories of abuse, particularly sexual abuse amongst the female prison population are said to make certain prison regimes particularly harmful. For example, strip searches are claimed to be especially traumatic for women (see Chesney-Lind, 1997: 165). Finally, the sexual abuse of female prisoners by male wardens is raising concerns internationally (see Chesney-Lind, 1997: 166). However, as far as I am aware the treatment of female prisoners by male correctional officers has not been researched in New Zealand and it is possible that the situation may well be different here. Concerns in this country have recently been raised about 'consensual' sexual relations between female inmates and male prison officers (Dominion, 12-8-98, Christchurch Press, 29-9-98) but to my knowledge the issue of sexual abuse per se has not been identified as a major problem in our women's prisons. 201 4 'Female' Vs 'Male' Focussed Justice To transcend the old equality/difference debate a reconceptualisation of criminal justice has been called for by feminist criminal justice commentators (see for example, Heidensohn, 1986; Smart; 1989; Daly, 1989c; Chesney-Lind, 1997; Carlen, 1998). It is argued that feminist focus should now be directed at disadvantage rather than difference (Smith, 1993; 8); that equity rather than equality should be sought through developing a social-based rather than a justice-based approach to criminal justice processing (Chesney-Lind & Pollock, 1995: 157). Instead of a 'male version' of justice which emphases "fairness, equal treatment, and rationality" in deciding judicial outcomes, a 'female version' of justice is proposed, which emphasises "needs, motives, and relationships" (Chesney-Lind & Pollock, 1995: 169). For example, Daly (1989c: 6-7) suggests that criminal justice processing needs to move towards an "ethic of care" as opposed to a "logic of justice" and Heidensohn (1986) proposed that a "Persephone", rather than a "Portia" approach to justice is required (see Figure 17). FIGURE 17 - DALv's AND HEIDENSOHN'S MODELS OF JUSTICE Daly's (1989: 6) 'Care' and 'Justice' Based Models Et~ni.6 " of Care Airll§ of Punishment: Rehabilitation, special deterrence Decision criteria: Forwardlooking (based on a prediction of future behaviour). Ideological elements: Equity, fairness, rationality (formal and substantive equality). Perspective practices: Tailor the sentence to the crime and to offender characteristics; personalise sentencing. Social unit of punishment: Family based; a person in relation to others Concept of justice: procedural and substantive equality, through greater emphasis on the latter Sentencing scheme: Individualised Logic of Justice Aims of Punishment: Retribution, general deterrence Decision criteria: Backwardlooking (based on the offence committed). Ideological elements: Equality, fairness, rationality (formal equality). Perspective practices: Equal treatment for those convicted of the same offence; depersonalise sentencing Individual based: a person not connected to others Heidensohn's (1986: 293) 'Portia' and 'Persephone' Models of Justice Portia Persephone Values and characteristics: Values and characteristics: Masculine, Rationality, Feminine, caring and Individualism personal System: civic rights, rule of law System: networks, informal Concept of justice: legal, equality, procedural. Concept of Justice: responsibility, co-operation Features: norm is male Features: norm is female Concept of justice: Emphasis on procedural equality. Sentencing scheme: Just Deserts Aspects of these 'female', 'care', or 'Persephone-based' justice models are already present in New Zealand's criminal justice system. For example, our treatment of youth offenders stresses the involvement of family (whanau), iwi (tribe), hapu (sub-tribe), as well as victims and the community in the decision-making process. Care-based sentencing options, such as supervision and community programmes (see Chapter 1), are also available to Judges when sentencing adult offenders. Furthermore, the cunent research has shown that, to a certain extent, courts 202 already apply 'female' versions of justice when sentencing women. For example, familial commitments, responsibilities, poor health and histories of victimisation have been found to mitigate sentence severity. Ideally, some feminists writers envisage a "separate, gentler, more sympathetic justice system exclusively received for women", but this is problematic for a number of reasons, all of which relate to the fact that criminal justice cannot work in isolation from the broader social context. First, continuing power over women by some groups of men in society at large, makes this proposition unworkable. It is unlikely that men would agree to such an arrangement, and the potential for women to be 'infantilised' by it is great (Heidensohn, 1986: 296). Second, a climate of 'just deserts' or 'getting tough' on crime has more recently emerged in New Zealand (e.g. a presumption of imprisonment for serious violent offenders), and while policy measures reflecting this are not as extreme or rigid as those in the United States (e.g. mandatory sentencing minimums, 'three strikes and you're out' policies, truth in sentencing), notions of retribution and punishment now dominate public and political sentiment. Concern for care and rehabilitation appear to be evading New Zealanders who increasingly desire to see offenders (especially men) 'locked up' and the 'key thrown away'. Ironically, this stance of 'getting tough' and 'making offenders pay' has evolved (in part) from feminist-based movements, such as the Women's Refuge and Rape Crisis, who are concerned with raising public awareness about and punishing more harshly, the violent and sexual victimisation of women by men. Thus, while we must 'get tough' on criminal men, we must 'care more' about offending women. 5 Closing Comments - Humane Justice My argument is that while feminist criminologists have extended our understanding of gender in relation to women's lives, we also need to develop a sympathetic understanding of gender and its impact on men's lives. Men are not universally powerful, and their criminality, just like women's, relates to their social circumstances. Women may 'deserve' a more caring justice system because of familial responsibilities and because of having lives scarred by victimisation, but many men fit into the same category. Recalling my earlier point as it relates to women, evidence also suggests that imprisonment may be somewhat 'harsh' and 'unusual punishment' for men as well. The physical abuse of male inmates in Mangaroa prison, for example, has recently resulted in the government being forced to pay tens of thousands of dollars compensation to male inmates who endured 203 "systematic beatings by hit squads of guards" (New Zealand Herald, 8-9-00). There has been no reported equivalent in women's prisons. Second, while the variety of educational, vocational and recreational programmes available to men may exceed those available to women, men are less likely than women to take advantage of them: in 1997, 80% of female prison inmates in New Zealand were enrolled in prison programmes compared to 45% of men (Lash, 1998: 55). Thus, men and women are both disadvantaged albeit in different ways. Women may lack variety but they benefit from being involved in prison programmes. Men, on the other hand, may benefit from programme variety but miss out when it comes to involvement. Third, while the everyday regime of the women's prison may emphasise a form of repressive 'refeminsation', so too is it likely that dominant masculinity (which can itself be repressive) is also reproduced in men's prisons. Fourth, a recent Ministry of Justice (1996b) study of male prison inmates primarily responsible for the care of children found that the men were clearly traumatised by being separated from their children. Inmates thus deprived described themselves as: "miserable, missing their children, frustrated, desperate, unhappy, helpless, guilty, and devastated" (Ministry of Justice, 1996b: 27). Feminist scholars point out that women's care-giving is valued more highly than men's breadwinning by the courts and that this can partially explain why women receive less severe sentences (see for example, Daly 1987a, 1987b, 1989a, 1989b). The current research has also indicated that there is judicial concern about the impact of removing women from families, but little concern for the impact of removing men from families. How do we know that the trauma or social cost of removing women from the family is somehow greater than that caused by removing men? The Ministry of Justice (1996b) study mentioned above found that there were clear social costs to removing offending men from their children. It notes that: "the prison system does not make it easy for children to maintain contact with their imprisoned father. Children who live some distance from where their father is imprisoned can be doubly punished - toli calis are expensive and distance can make visits impractical. ..improvements or alternatives may need to be considered if children are not to be punished as much as or more than their fathers" (Ministry of Justice, 199Gb: 31). However, there are few studies of this kind and the reality is that scant attention has been given to the way in which a man's imprisonment impacts on his children or his family in general (Davis, 1991: 27). Research that has been conducted suggests that wives/partners and children are being adversely affected emotionally, financially, mentally and physically by men's imprisonment (Davis, 1991: 41-42). 204 In the current research, dominant feminine discourses ensured that offending women were often presented and accepted as 'troubled'. Histories of victimisation and the subsequent effects of this emerged as explanations and often excused women's offending. In contrast, such 'troubles' appeared as simply unbelievable in the case of men and if, for example, presented to Judges, these were often rejected or ignored. In this way, men were denied reasons for their offending and they were held fully responsible for their actions. While it may be the case that histories of victimisation are more common in the case of women, both current and international research results show that criminal men also experience victimisation. Arguably, this in turn has led to impaired personal functioning and ultimately criminality (Sheridan: 1996). Furthermore, the extent of men's victimisation and its subsequent impacts could be greater than we think. Men may be less likely than women to report abuse or neglect because revealing such sensitive information runs counter to dominant ideas about masculinity. A societal inability to acknowledge men as victims further perpetuates men's silence and ultimately adds to their trauma (see for example, Groth & Wolbert-Burgess, 1980; Kaufman, Divasto, Jackson, Voorhees & Christy, 1980; Anderson, 1982; Sanel & Masters, 1982; Goyer & Eddleman, 1984; Masters, 1986; Calderwood, 1987; Mezey & King, 1989; Poropat & Rosevear, 1993; Hickson, Davies, Hunt, Weatherburn, McManus & Coxon, 1994; StuckmanJohnson & Stuckman-Johnson; 1994; Donnelly & Kenyon, 1996; Sheridan: 1996; Stermac, Sheridan, Davidson & Dunn, 1996). I would like to see feminist criminologists transcend the boundaries of the equality/difference debate by problematising criminal justice processing as it relates to both sexes rather than simply in terms of women against men. As a societal group, criminal men and women both tend to come from disadvantaged circumstances. Men's criminality, just like women's, does not exist in a "social, political or economic vacuum" unaffected by unemployment, poverty, drug and alcohol addiction, victimisation, general mental and physical illness (Bagley & Merlo, 1995: 147). This is not to deny that certain circumstances which are relatively unique to women's experience may explain, excuse, or mitigate their criminality (e.g. care giving and dependence). What I am arguing is that there are also certain circumstances, which are relatively unique to men's experience, which could also explain, excuse, or mitigate their criminality. Thus, in agreement with Daly (1989c: 15), I would like to see a "feminist conception of criminal justice which maintains a focus on women's lives and on redressing harms to women, but which does not ignore those men who have been crippled by patriarchal, class, and race relations." illtimately, perhaps, a more caring criminal justice system might not treat our criminal women more like men, but treat our criminal men more like women. 205 In New Zealand, further gender-based offender research is therefore required. Feminist criminologists in this country (of which there are very few) have spent the majority of their time focussing on female victimisation. While this may be a noble endeavour, we need to be careful not to 'feminise' gender studies with the pervasive 'women question', thereby ignoring the 'men question' and relegating males to the "unexplored de-sexed norm" (Collier, 1998: 12). We should stop rendering masculinity invisible by universalising all men as powerful and acknowledge that "ideological and political processes which assert and sustain the authority of normative heterosexuality ... have powerful consequences for both men and women" (Walklate, 1995: 186-187). New Zealand society, the criminal justice system, and to some extent, feminist discussions (especially those surrounding the victimisation of women by men) continue to present most men in terms of thinking, acting, powerful human beings, while simultaneously embracing women's powerlessness and dependency. As long as this continues, women's right to self determination and power will be ignored and men will continue to vent frustration at a society that refuses to acknowledge their weaknesses and extend them understanding. If this occurs, destructive consequences will continue to be felt, not only by men, but also by women. 206 References Adler, F. (1975). Sisters in Crime: The Rise of the New Female Criminal. McGraw-Hill. New York: Alder, C. (1994). 'Women and the Criminal Justice System', in Chappell, D., Wilson, P. (eds.). The Australian Criminal Justice System the Mid 1990's. Sydney: Butterworths. Allen, H. (1987a). 'Rendering Them Harmless: The Professional Portrayal of Women Charged With Serious Violent Crimes', in Carlen, P., Worrall, A (eds.). Gender, Crime and Justice. Philadelphia: Open University Press. Allen, H. (1987b). Justice Unbalanced: Gender, Psychiatry and Judicial Decisions. Philadelphia: Open University Press. Allen, H. (1987c). 'The Logic of Gender in Psychiatric Reports to the Courts', in Pennington, D.C., Lloyd-Bostock, S. The Psychology of Sentencing: Approaches to Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. Anderson, c.L. (1982). Males As Sexual Assault Victims: Multiple Levels of Trauma. Journal of Homosexuality, 7, 2-3, pp 145-162. Andrewes, F. (1999). 'The Man in the Grey Flannel Suit: White-Collar Masculinity in PostWar New Zealand', in Daley, C., Montgomerie, D. (eds.). The Gendered Kiwi. Auckland: Auckland University Press. Argyrous, G. (1996). Statistics for Social Research. Melbourne: MacMillan. Armstrong, G. (1977). Females Under the Law - "Protected" but Unequal. Crime and Delinquency, 23, pp 109-120. Ashworth, A (1987). 'Disentangling Disparity', in Pennington, D.C., Lloyd-Bostock, S. The Psychology of Sentencing: Approaches to Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. Ashworth, A (1995). Sentencing and Criminal Justice. 2nd Edition., London: Butterworths. Barwick, H., Burns, J., Gray, A (1997). 'Gender Equity in the New Zealand Judicial System: Judges' Perceptions of Gender Issues', in Judicial Working Group on Gender Equity. Judicial Seminar on Gender Equity. Rotorua, 15-19 May, 1997 (unpublished). Bagley, K., Merlo, AV. (1995). 'Controlling Women's Bodies', in Merlo, AV., Pollock, J.M. Women, Law, and Social Control. Boston: Allyn and Bacon. Belknap, J. (1996). The Invisible Women: Gender, Crime and Justice. Belmont: Wadsworth Publishing. Benekos, P.J. (1995). 'Women as Victims and Perpetrators of Murder' in Merlo, AV., Pollock, J.M. Women, Law, and Social Control. Boston: Allyn and Bacon. 207 Bickle, G.S., Peterson, RD. (1991). The Impact of Gender-Based Family Roles on Criminal Sentencing. Social Problems, 38, 3, pp 372- 394. Bishop, D.M., Frazier, C. (1984). The Effects of Gender on Charge Reduction. Sociological Quarterly, 25, pp 385-396. The Gender and Criminal Court Outcomes: An Historical Analysis. Boritch, H. (1992). Criminology, 30, 3, pp 293-325. Briar, c., Munford, R, Nash, M. (1992). (eds.). Superwoman Where Are You? Policy and Women's Experience. Palmerston North: Dunmore Press. Social Briar, C. (1992). 'Women, Economic Dependence and Social Policy', in Briar, c., Munford, R, Nash, M. (eds.). Superwoman Where Are You? Social Policy and Women's Experience. Palmerston North: Dunmore Press. Brookes, B., Macdonald, C., Tennant, M. (1986). (eds.). Women in History: Essays on European Women in New Zealand. Wellington: Allen & Unwin. Brown, M., Goddard, L., Jefferson, S. (1989). Children, Young Persons and Their Families Act: New Zealand Law Society Seminar. Wellington: New Zealand Law Society. Browne, A. (1987). When Battered Women Kill. New York: The Free Press. Bryman, A., Cramer, D. (1997). London: Routledge. Quantitative Data Analysis with SPSS for Windows. Bungay, R (1998). Scarecrows Why Women Kill. Auckland: Random House. Calderwood, D. (1987). The Male Rape Victim. Medical Aspects of Human Sexuality, 21, 5, pp 53-55. Cain, M. (1990). Towards Transgression: New Directions in Feminist Criminology. International Journal of the Sociology of Law, 18, pp 1-18. Caird, B. (1999). Speaking Sentences: Inside Women's Experience of Gender, Punishment and Rehabilitation. Unpublished MA thesis in Social Policy, Massey University, Palmerston North. Carlen, (1998). Sledgehammer: Women's Imprisonment at the Millennium. MacMillian Press. Houndmills: Champion, D.J. (1993). 'A Comparative Analysis of Female Felons: Some Recent Trends in Sentencing Severity Compared with a Matched Sample of Male Offenders', in Culliver, C.c. (ed). Female Criminality: The State of the Art. New York: Garland. Channels., N.L., Herzberger, S.D. (1993). 'The Role of Gender in Pretrial Decision Making', in Culliver, c.c. (ed.). Female Criminality: The State of the Art. New York: Garland Publishers. Chesney-Lind, M. (1973). Judicial Enforcement of the Female Sex Role: The Family Court and the Female Delinquent. Issues in Criminology, 8, 2, pp 51-69. 208 Chesney-Lind, M. (1977). Judicial Paternalism and the Female Status Offender: Training Women to Know Their Place. Crime and Delinquency, 23, pp 121-130. Chesney-Lind, M. (1978). 'Chivalry Reexamined: Women and the Criminal Justice System', in Bowker, L.H. (ed.). Women, Crime, and the Criminal Justice System. Lexington: Lexington Books. Chesney-Lind, M. (1987). 'Female Offenders: Paternalism Reexamined', in Crites, L.L., Hepperle, W.L. (eds.). Women, the Courts and Equality. New Park: Sage Publications. Chesney-Lind, M. (1995). 'Women and Crime: The Female Offender', in Naffine, N. (ed.). Gender, Crime and Feminism. Dartmouth: Aldershot. Chesney-Lind, M. (1997). Oaks: Sage Publications. The Female Offender: Girls, Women, and Crime. Thousand Chesney-Lind, M., Pollock, J.M. (1995). 'Women's Prisons: Equality with a Vengeance', in Merlo, A.v., Pollock, J.M. Women, Law, and Social Control. Boston: Allyn and Bacon. Collier, R. (1998). Masculinities, Crime, and Criminology: Men, Heterosexuality, and the Criminal(ised) Other. London: Sage Publications. Conley, J.M., O'Barr, W.M. (1987). Fundamentals of Jurisprudence: An Ethnography of Judicial Decision Making in Informal Courts. North Carolina Law Review, 66, 3, pp 467-507. Cox, S. (1987). (ed.). Public and Private Worlds: Women in Contemporary New Zealand. Wellington: Allen and Unwin. Craig, W. (1992). 'The Politics of Caring', in Briar, c., Munford, R., Nash, M. (eds.). Superwoman Where Are You? Social Policy and Women's Experience. Palmerston North: Dunmore Press. Crew, K.B. (1991). Sex Differences in Criminal Sentencing: Chivalry or Patriarchy? Justice Quarterly, 8, 1, pp 59-83. Croyle, J.L. (1983). Measuring and Explaining Disparities in Felony Sentences: Courtroom Work Group Factors and Race, Sex, and Socioeconomic Influences on Sentence Severity. Political Behavior, 5, 1, 135-153. Culliver, c.c. (1993). 'Women and Crime: An Overview', in Culliver, c.c. (ed.). Female Criminality: The State of the Art. New York: Garland Publishers. Curran, D.A. (1983). Judicial Discretion and Defendant's Sex. Criminology, 21, 1, pp 4158. Daley, C., Montgomerie, D. (1999). (eds.). University Press. The Gendered Kiwi. Auckland: Auckland Dalley, B. (1999). 'Criminal Conversations: Infanticide, Gender and Sexuality in NineteenthCentury New Zealand', in Daley, c., Montgomerie, D. (eds.). The Gendered Kiwi. Auckland: Auckland University Press. 209 Daly, K (1987a). Structure and Practice of Familial-Based Justice in a Criminal Court. Law and Society Review, 21, 2, pp 267-290. Daly, K (1987b). Discrimination in the Criminal Courts: Family, Gender, and the Problem of Equal Treatment. Social Forces, 66, 1, pp 152-175. Daly, K (1989a). Neither Conflict Nor Labeling Nor Paternalism will Suffice: Intersections of Race, Ethnicity, Gender, and Family in Criminal Court Decisions. Crime and Delinquency, 35, 1, pp 136-168. Daly, K (1989b). Rethinking Judicial Paternalism: Gender, Work-Family Relations, and Sentencing. Gender and Society, 3, 1, pp 9-36. Daly, K (1989c). Criminal Justice Ideologies and Practices in Different Voices: Some Feminist Questions about Justice. 1nternational Journal of the Sociology of Law, 17, pp 1-18. Daly, K (1994). Gender, Crime, and Punishment. New Haven: Yale University Press. Daly, K, Chesney-Lind, M. (1988). Feminism and Criminology. Justice Quarterly, 5, 4, pp 497-538. Daly, K, Bordt, RL. (1995). Sex Effect and Sentencing: An Analysis of the Statistical Literature. Justice Quarterly, 12, 1, pp 142-175. Daly, K, Tonry, M. (1997). 'Gender, Race, and Sentencing', in Tonry, M. (ed.). Crime and Justice: A Review of Research. Chicago: The University of Chicago Press. Davis, A. (1991). The Costs of Surviving Men's Prison Sentences. Women and Criminal Justice, 2, 2, pp 27-44. Deane, H. (1995). 'Race, Gender, and the Sentencing Process in a New Zealand District Court', in Hazlehurst, KM. (ed). Perceptions of Justice: Issues in Indigenous and Community Empowerment. Sydney: Avebury. Deane, H. (1997). Race, Gender and the Sentencing Process: A Study of Sentencing in the District Court in New Zealand. Unpublished PhD thesis in Criminology, Victoria University, Wellington. Deane, H. (2000). The Influence of Pre-Sentence Reports on Sentencing in a District Court in New Zealand. The Australian and New Zealand Journal of Criminology, 33, 1, pp 91-106. \ Dobash, RP., Dobash, E.R, Wilson, M., Daly, M. (1992). The Myth of Sexual Symmetry in \ Marital Violence. Social Problems, 39, 1, pp 71-91. Donnelly, D.A., Kenyon, S. (1996). 'Honey, We Don't Do Men': Gender Stereotypes and the Provision of Services to Sexually Assaulted Males. Journal of Interpersonal Violence, 11, 3, pp 441-448. Drass, KA., Spencer, J.W. (1987). Accounting for Pre-Sentencing Recommendations: Typologies and Probation Officers' Theory of Office. Social Problems, 34, 3, pp 277-293. 210 Douglas, R. (1987). Is Chivalry Dead? Gender and Sentence in the Victorian Magistrates' Courts. The Australian and New Zealand Journal of Sociology, 23, 3, pp 343-357. Du Plessis, R., Alice, L. (1998). (eds.). Feminist Thought in Aotearoa New Zealand: Differences and Connections. Auckland: Oxford University Press. Eaton, M. (1986). Justice for Women? Family; Court and Social Control. Milton Keynes: Open University Press. Eaton, M. (1987). 'The Question of Bail: Magistrates' Responses to Applications for Bail on Behalf of Men and Women Defendants', in Carlen, P., Worrall, A (eds.). Gender, Crime and Justice. Milton Keynes: Open University Press. Edwards, S.S.M. (1986). Neither Bad nor Mad: The Female Violent Offender Reassessed. Women's Studies International Forum, 9,1, pp 79-87. Elliot, L., Morris, R. (1987). 'Behind Prison Doors', in Adelberg, E., Currie, C. (eds.). To Few To Count: Canadian Women in Conflict With The Law. Vancouver: Press Gang Publishers. Farnworth, M., Raymond, H.C.T (1995). Gender Differences in Felony Court Processing: Three Hypotheses of Disparity. Women and Criminal Justice, 6, 2, pp 23-44. Farrington, D.P., Morris, AM. (1983). Sex, Sentencing and Reconviction. British Journal of Criminology, 23, 3, pp 229-248. Feinman, C. (1980). Women in the Criminal Justice System. New York: Praegar Publishers. Fenster, c., Mahoney, AR. (1981). The Effect of Prior Record Upon The Sentencing of Male-Female Co-Defendants. The Justice System Journal, 6, pp 262-271. Figueira-McDonough, J. (1985a). Are Girls Different? Gender Discrepancies Between Delinquent Behaviour and Control. Child Welfare, 64, 4, pp 273-289. Figueira-McDonough, J. (1985b). Gender Differences in Informal Processing: A Look at Charge Bargaining and Sentence Reduction in Washington, D.C. Journal of Research in Crime and Delinquency, 22, 2, pp 101-133. Figueira-McDonough, J. (1987). Discrimination or Sex Differences? Criteria For Evaluating Juvenile Justice System's Handling of Minor Offences. Crime and Delinquency, 33, 2, pp 403-424. Flavin, J. (1998). 'Razing the Wall: A Feminist Critique of Sentencing Theory, Research, and Policy', in Ross, J.I. (ed.). Cutting the Edge: Current Perspectives in Radical/Critical Criminology and Criminal Justice. London: Praeger. Fletcher, G. (1999). Multiple Regression Workshop. Department of Psychology, University of Canterbury, Cln-istchurch. Follingstad, D.R., Brondino, M.J., Kleinfelter, KJ. (1996). Reputation and Behavior of Battered Women Who Kill Their Partners: Do These Variables Negate Self-Defence? Journal of Family Violence, 11, 3, pp 251-267. 211 Frank, T. (1999). 'Bread Queues and Breadwinners: Gender in the 1930s', in Daley, e., Montgomerie, D. (eds.). The Gendered Kiwi. Auckland: Auckland University Press. Frazier, e.E., Bock, E.W., Henretta, J.e. (1983). The Role of Probation Officers in Determining Gender Differences in Sentencing Severity. The Sociological Quarterly, 24, pp 305-318. Gauthier, D.K., Bankston, W.B. (1997). Killing. Criminology, 35, 4, pp 577-600. Gender Equality and the Sex Ratio of Intimate Gelsthorpe, L., Loucks, N. (1997). 'Magistrates' Explanations of Sentencing Decisions', in Hedderman, e., Gelsthorpe, L. (eds.). Understanding the Sentencing of Women. London: Home Office. Genders, E., Player, E. (1987). 'Women in Prison: The Treatment, the Control and the Experience', In Carlen, P., Worrall, A (eds.). Gender, Crime and Justice. Philadelphia: Open University Press. Ghali, M., Chesney-Lind, M. (1986). Gender Bias and the Criminal Justice System: An Empirical Investigation. Sociology and Social Research, 70, 2, pp 164-171. Gibbons, D., Griswold, M.J. (1957). Sex Differences Among Juvenile Court Referrals. Sociology and Social Research, 42, 2, pp 106-110. Goetting, A, Howsen, R.M. (1983). Women in Prison: A Profile. Prison Journal, 63, pp 2746. Goyer, P.F., Eddleman, H.e. (1984). Same-Sex Rape of Nonincarcerated Men. American Journal of Psychiatry, 141, 4, pp 576-579. Green, E. (1961). Judicial Attitudes in Sentencing: A Study of Factors Underlying the Sentencing Practice of the Criminal Court of Philadelphia. London: MacMillan. Groth, N., Wolbert-Burgess, A (1980). Journal of Psychiatry, 137, 7, pp 806-810. Male Rape: Offenders and Victims. American Gruhl, J., Welch, S., Spohn, C. (1984). Women as Criminal Defendants: A Test for Paternalism. Western Political Quarterly, 37, pp 456-467. Hagan, J., Simpson, J.H., Gillis, A.. (1979). The Sexual Stratification of Social Control: A Gender-Based Perspective on Crime and Delinquency. British Journal of Sociology, 30, 1, pp 25-38. Hagan, J. (1975). The Social and Legal Construction of Criminal Justice: A Study of The Pre-Sentencing Process. Social Problems, 22, 5, pp 620-637. Hagan, J., Bumiller, K. (1983). 'Making Sense of Sentencing: A Review and Critique of Sentencing Research', in Blumstein, A, Cohen, J., Martin, S.E., Tonry, M.H (eds.). Research on Sentencing: The Search for Reform. Washington: National Academy Press. Hall, G.G. (1994). Sentencing Guide. Wellington: Butterworths. 212 Hamilton, A (1995). The Reporting and Punishment of Internal Offences at Christchurch Women's Prison: A Descriptive Study. Unpublished MA thesis in Psychology, University of Canterbury, Christchurch. Hancock, L., Hiller, AE. (1981). 'The Processing of Juveniles in Victoria', in Mukherjee, S.K., Scutt, J.A (eds.). Women and Crime. Sydney: George Allen and Unwin. Hartley, RD. (1999). The Impact of Sentencing Guidelines on Sentencing Disparities in Federal Narcotics Cases. Unpublished Conference Paper Presented at the 1999 American Society of Criminology Annual Meetings, November 16-20, Ontario, Canada. Harvey, L., Burnham, RW., Kendall, K., Pease, K. (1992). Gender Differences in Criminal Justice. British Journal of Criminology, 32, 2, pp 208-217. Heberle, R (1999). Disciplining Gender; Or, Are Women Getting Away with Murder? Signs, 24,4. Hedderman, C., Gelsthorpe, L. (eds.). London: Home Office. (1997). Understanding the Sentencing of Women. Hedderman, C, Hough, M. (1994). Does the Criminal Justice System Treat Men and Women Differently? London: Home Office Research and Statistics Department. Heidensohn, F. (1986). Models of Justice: Portia or Persephone? Some Thoughts on . Equality, Fairness and Gender in the Field of Criminal Justice. International Journal of the Sociology of Law, 14, pp 287-298. Heidensohn, F. (1996). Women and Crime. 2nd Edition, London: MacMillan Press. Heilburn, AB., Heilburn, M.R (1986). The Treatment of Women Within the Criminal Justice System: An Inquiry into the Social Impact of the Women's Rights Movement. Psychology of Women Quarterly, 10, pp 240-251. Henderson, L. (1991). Law's Patriarchy. Law and Society Review, 25, 2, pp 411-444. Hesketh, B., Young, W. (1994). Sentencing and Plea Making. Wellington: New Zealand Law Society. Hickson, F.CI., Davies, P.M., Hunt, AJ., Weatherburn, P., McManus, T.J., Coxon, A.P.M. (1994). Gay Men as Victims of Nonconsensual Sex. Archives of Sexual Behavior, 23, 3, pp 281-294. Hinkle, D.E., Wiersma, W., Jurs, S.G. (1994). Applied Statistics for the Behavioral Sciences. 3rd edition. Boston: Houghton Mifflin Company. Hodge, W.C (1991). Criminal Procedure in New Zealand. 3rd Edition, Sydney: The Law Book Company. Husak, D.N. (1998). 'Desert, Proportionality, and The Seriousness of Drug Offences', in Ashworth, A, Wasik, M (eds.). Fundamentals of Sentencing Theory. Oxford: Clarendon Press. 213 Jackson, H., Smith, L. (1987). Female Offenders: an Analysis of Social Inquiry Reports. Home Office Research Bulletin, 23, pp 8-11. James, B., Saville-Smith, K. (1994). Gender, Culture and Power: Challenging New Zealand's Gendered Culture. 2nd Edition. Auckland: Oxford University Press. Johnston, J.B., Kennedy, T.D., Shuman, G.!. (1987). Gender Differences in the Sentencing of Felony Offenders. Federal Probation, 51, 1, pp 49-55. Jones, A (1980). Women Who Kill. New York: Hort, Rinehart and Winston. Julian, F.H. (1993). 'Gender and Crime: Different Sex, Different Treatment?' In Culliver, c.c. (ed.). Female Criminality: The State of the Art. New York: Garland Publishers. Jurik, N.C., Winn, R. (1990). Gender and Homicide: A Comparison of Men and Women Who Kill. Violence and Victims, 5, 4, pp 227-242. Kapardis, A (1987). 'Sentencing by English Magistrates As A Human Process', in The Psychology of Sentencing: Approaches to Pennington, D.C., Lloyd-Bostock, S. Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. Kaufman, A, Divasto, P., Jackson, R., Voorhees, D., Christy, J. (1980). Male Rape Victims: Noninstitutionalized Assault. American Journal of Psychiatry, 137, 2, pp 221-223. Kempinen, C. (1983). Changes in the Sentencing Patterns of Male and Female Criminal Defendants. Prison Journal, 63, pp 3-11. Kerber, L.K., De Hart-Mathews, J. (1982). (eds.). Women's America Refocusing the Past. 2nd Edition. New York. Oxford University Press. Klepper, S., Nagin, D., Tierney, L.J. (1983). Discrimination in the Criminal Justice System: A Critical Appraisal of the Literature, in Blumstein, A, Cohen, J., Martin, S.E., Tonry, M.H (eds.). Research on Sentencing: The Search for Reform. Washington: National Academy Press. Kratcoski, P.C. (1974). Differential Treatment of Delinquent Boys and Girls in Juvenile Court. Child Welfare, 53, pp 16-22. Kruttschnitt, C. (1982a). Respectable Women and the Law. The Sociological Quarterly, 23, pp 221-234. Kruttschnitt, C. (1982b). Women, Crime, and Dependency: An Application of the Theory of Law. Criminology, 19, 4, pp 495-513. Kruttschnitt, C. (1984). Sex and Criminal Court Dispositions: The Unresolved Controversy, Research in Crime and Delinquency, 21, 3, pp 213-232. Kruttschnitt, C., Green, D.E. (1984). The Sex-Sanctioning Issue: Is it History? American Sociological Review, 49, pp 541-551. 214 Kruttschnitt, c., McCarthy, D. (1985). Familial Social Control and Pretrial Sanctions: Does Sex Really Matter? The Journal of Criminal Law and Criminology, 76, 1, pp 151-172. LaBeff, B. (1981). Race, Sex, and Prison Sentence Disparity. Free Inquiry in Creative Sociology, 9, 1, pp 35-38. Land, K.c. (1969). 'Principles of Path Analysis', in Borgatta, E.F. (ed.). Methodology. San Francisco: Jossey-Bass Inc Publishers. Sociological Lash, B. (1996). Census of Prison Inmates 1995. Wellington: Ministry of Justice. Lash, B. (1998). Census of Prison Inmates 1997. Wellington: Ministry of Justice. Laster, K. (1994). Arbitrary Chivalry: Women and Capital Punishment in Victoria, Australia 1842-1967. Women and Criminal Justice, 6, 1, pp 67-95. Law, R, Campbell, H.R, Dolan, J.C. (1999). (eds.). Zealand. Palrnerston North: Dunmore Press. Masculinities in AotearoaiNew Lindquist, J.H., White, O.Z., Tutchings, T., Chambers, C.D. (1989). Judicial Processing of Males and Females Charged with Prostitution. Journal of Criminal Justice, 17, pp 277-291. MacKinnon, C. (1987). Feminism Unmodified: Discources on Life and Law. Harvard University Press. London: Mahoney, AR, Fenster, C. (1982). 'Female Delinquents in a Suburban Court', in Rafter, N.H., Stanko, B.A, (eds.). Judge, Lawyer, Victim, Thief" Women, Gender Roles, and Criminal Justice. Boston: North Eastern University Press. Mair, G., Brockington, N. (1988). Female Offenders and the Probation Service. The Howard Journal, 27, 2, pp 117-126. Mann, C.R (1985). Female Crime and Delinquency. Alabama: University of Alabama Press. Masters, W.H. (1986). Sexual Dysfunction As An Aftermath of Sexual Assault of Men by Women. Journal of Sex & Marital Therapy, 12, 1, pp 35-45. Maxwell, G.M., Morris, A (1993). Family, Victims and Culture: Youth Justice in New Zealand. Wellington: Wellington Institute of Criminology/Social Policy Agency. McKinlay, R (1992). 'Women and Unpaid Work', in Briar, c., Munford, R, Nash~ M. (eds.). Superwoman Where Are You? Social Policy and Women's Experience. Palrnerston North: Dunmore Press. McDonald, E. (1997). 'Criminal Defences for Women', in Ablett Kerr, J., McDonald, B., Young, R (eds.). Women in the Criminal Justice System: New Zealand Law Society Seminar. Wellington: New Zealand Law Society. McIvor, G. (1998). Jobs for the Boys?: Gender Differences in Referral to Community Service. The Howard Journal, 37, 3, pp 280-290. 215 Mezey, G., King, M. (1989). The Effects of Sexual Assault on Men: A Survey of 22 Victims. Psychological Medicine, 19, pp 205-209. Ministry of Justice. (1995). Seriousness of Offence Scale. Wellington: Ministry of Justice. Ministry of Justice. (1996a). List of Offences with Maximum Penalties by Act and Section Current. Wellington: Ministry of Justice. Ministry of Justice. (1996b). Male Inmates Who Were Their Children's Primary Care Givers. Wellington: Ministry of Justice. Ministry of Justice. (1997). Sentencing Policy and Guidance: A Discussion Paper. Wellington: Ministry of Justice. Moulds, B.F. (1978). Chivalry and Paternalism: Disparities of Treatment in the Criminal Justice System. Western Political Quarterly, 31, 3, pp 416-430. Morris, RR (1965). Attitudes Toward Delinquency by Delinquents, Non-Delinquents and Their Friends. The British Journal of Delinquency, 5, 3, pp 249-265. Mustard, D.B. (1999). Racial, Ethnic and Gender Disparities in Sentencing: Evidence from the US Federal Courts. University of Georgia Economics, Working Paper 97-458. Naffine, N. (ed.). (1995). Gender, Crime and Feminism. Dartmouth: Aldershot. Nagel, I. (1981). 'Sex Differences in the Processing of Criminal Defendants', in Morris, A., Gels thorpe, L. (eds.). Women and Crime: Cropwood Conference Series No 13. Cambridge: University of Cambridge Institute of Criminology. Nagel, I., Cardascia, J., Ross, c.B. (1982). 'Sex Differences in the Processing of Criminal Defendants', in Weisberg, D.K. (ed.). Women and the Law: The Social Historical Perspective. Cambridge: Schenkman. Nagel, I., Hagan, J. (1983). 'Gender and Crime: Offense Patterns and Criminal Court Sanctions', in Tonry, M., Morris, N. (eds.). Crime and Justice, Vol. 4. Chicago: The University of Chicago Press. Newbold, G. (2000). Crime In New Zealand. Palmerston North: Dunmore Press. Novitz, R (1987). 'Bridging the Gap: Paid and Unpaid Work', in Cox, S. (ed.). Public and Private Worlds: Women in Contemporary New Zealand. Wellington: Allen and Unwin. Odubekun, L. (1992). A Structural Approach to Differential Gender Sentencing. Criminal Justice Abstracts, 24, 2, pp 343-360. O'Dwyer, J., Wilson, J., Carlen, P. (1987). 'Women's Imprisonment in England, Wales and Scotland: Recurring Issues', In Carlen, P., Worrall, A. (eds.). Gender, Crime and Justice. Philadelphia: Open University Press. Olssen, E. (1995). 'Towards a New Society', in Rice, G. (ed.). The Oxford History of New Zealand. Auckland: Oxford University Press. 216 Olssen, E. (1999). ' Families and the Gendering of European New Zealand in the Colonial Period, 1840-80', in Daley, C., Montgomerie, D. (eds.). The Gendered Kiwi. Auckland: Auckland University Press. Parisi, N. (1982). 'Are Females Treated Differently? A Review of Theories and Evidence on Sentencing and Parole Decisions' in Rafter, N.H., Stanko, E.A. (eds.). Judge, Lawyer, Victim, Thief Women, Gender Roles, and Criminal Justice. Boston: North Eastern University Press. Park, J. (1991). (ed.). Ladies a Plate: Change and Continuity in the Lives of New Zealand Women. Auckland: Auckland University Press. Pearson, P. (1997). When She Was Bad: Violent Women and the Myth of Innocence. New York: Viking Penguin. Pedhazur, E.J. (1982). Multiple Regression in Behavioural Research: Explanation and Prediction. 2nd Edition, New York: Holt, Rinehart and Winston. Phillips, C, (1992). 'Women in Prison', in Briar, C., Munford, R., Nash, M. (eds.). Superwoman Where Are You? Social Policy and Women's Experience. Palmerston North: Dunmore Press. Phillips, J. (1987). A Man's Country? Auckland: Penguin Books. \ Pollock-Byrne, J.M. (1990). Publishing Company. The Image of the Pakeha Male - A History. Women, Prison and Crime. Pacific Grove: Brooks Cole Poropat, P., Rosevear, W. (1993). 'Sexual Assault of Males', in Weiser-Easteal, P. (ed.). Without Consent: Confronting Adult Sexual Violence - Conference Proceedings Number 20 of the Australian Institute of Criminology, Canberra 1993. Canberra: Australian Institute of Criminology. Pratt, c., Deosaransingh, K. (1997). Gender Differences in Homicide in Contra Costa County, California: 1982-1993. American Journal of Preventive Medicine, 13, 6, pp 19-24. Prior, P.M. (1999). Gender and Mental Health. London: MacMillian. Raeder, M.S. (1993). Gender and Sentencing: Single Moms, Battered Women, and Other Sex-Based Anomalies in the Gender-Free World of Federal Sentencing Guidelines. Pepperdine Law Review, 20, 3, pp 905-990. Rapaport, E. (1991). The Death Penalty and Gender Discrimination. Review, 25, 2, pp 367-383. Law and Society Reaves, C.c. (1992). Quantitative Research for the Behavioral Sciences. New York: John Wiley & Sons, Inc. Reed, W.L., Thier, J.D. (1981). The Presentence Investigation and Discretion in Criminal Sentencing. California Sociologist, 4, 2, pp 233-245. 217 Roberts, A. R (1996). Battered Women Who Kill: A Comparative Study of Incarcerated Participants with a Community Sample of Battered Women. Journal of Family Violence, 11, 3, pp 291-304. Roberts, C. (1987). 'The Probation Officer's Dilemma', in Pennington, D.C., Lloyd-Bostock, S. The Psychology of Sentencing: Approaches to Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. Roberts, D.E. (1995). Motherhood and Crime. Social Text, 42, pp 99-123. Rosenthal, R, Rubin, D.B. (1982). A Simple, General Purpose Display of Magnitude of Experimental Effect. Journal of Educational Psychology, 74, 166-169. Rossi, P.R., Waite, E., Bose, C.E., Berk, RE. (1974). The Seriousness of Crimes: Normative Structure and Individual Differences. American Sociological Review, 39, pp 224-237. Samuels, A. (1987). 'Consistency in Sentencing', in Pennington, D.C., Lloyd-Bostock, S. The Psychology of Sentencing: Approaches to Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. I Sarrel, P.M., Masters. W.R. (1982). Sexual Molestation of Men by Women. Archives of Sexual Behavior, 11,2, pp 117-131. Sarri, RC. (1986). Gender and Race Differences in Criminal Justice Processing. Women's Studies International Forum, 9, 1, pp 89-99. Saville-Smith, K. (1987). 'Women and the State', in Cox, S. (ed.). Public and Private Worlds: Women in Contemporary New Zealand. Wellington: Allen and Unwin. Sayers, J. (1992). Equal Employment Opportunities: Directions for the Future? in Briar, c., Munford, R, Nash, M. (eds.). Superwoman Where Are You? Social Policy and Women's Experience. Palmers ton North: Dunmore Press. Sentencing Digest [Computer Based Document]: Wellington: Status Publishing. Shapland, J. (1987). 'Who Controls Sentencing? Influences on the Sentencer', in Pennington, D.C., Lloyd-Bostock, S. The Psychology of Sentencing: Approaches to Consistency and Disparity. Oxford: University of Oxford Centre for Socio-Legal Studies. Sheldon, RG. (1981). 'Sex Discrimination in the Juvenile Justice System: Memphis Tennessee, 1900-1917', in Warren, M.Q., (ed.). Comparing Male and Female Offenders. Los Angeles: Sage Publications. Sheridan, M.J. (1996). Comparison of the Life Experiences and Personal Functioning of Men and Women in Prison. Families in Society: The Journal of Contemporary Human Services, 77, 7, pp 423-434. Simon, L.M.J. (1996). Legal Treatment of the Victim-Offender Relationship in Crimes of Violence. Journal of Interpersonal Violence, 11, 1, pp 94-106. Simon, RJ. (1975). Women and Crime. Lexington: Lexington Books. 218 Singer, L.R. (1973). Women in the Correctional Process. American Criminal Law Review, 11, pp 295-308. Smart, C. (1976). Women, Crime and Criminology: A Feminist Critique. Routledge & Kegan Paul. London: Smart, C. (1980). 'Criminological Theory: Its Ideology and Implications Concerning Women', in Weisberg, D. (ed.). Women and the Law: The Social Historical Perspective. New York: Schenkman. Smart, C. (1989). Feminism and the Power of Law. London: Routledge. Smart, C. (1995). Publications. Law, Crime and Sexuality: Essays in Feminism. London: Sage Smith, P. (1993). (ed.). Feminist Jurisprudence. New York: Oxford University Press. Spencer, J.W. (1984). Conducting Presentencing Investigations: From Discourse to Textual Summaries. Urban Life, 13, 2-3, pp 207-227. \ Spier, P. (1996). Conviction and Sentencing of Offenders in New Zealand: 1986 to 1995. Wellington: Ministry of Justice. Spier, P. (1997). Conviction and Sentencing of Offenders in New Zealand: 1987 to 1996. , 'Wellington: Ministry of Justice. Spohn, C.C., Spears, J.W. (1997). Gender and Case Decisions: A Comparison of Case Outcomes for Male and Female Defendants Charged with Violent Felonies. Women and Criminal Justice, 8, 3, pp 29-59. Spohn, C., Gruhl, J., Welch, S. (1987). The Impact of the Ethnicity and Gender of Defendants on the Decision to Reject or Dismiss Felony Charges. Criminology, 25, pp 175191. Spohn, C.C., Welch, S., Gruhl, J. (1985). Women Defendants in the Court: The Interaction between Sex and Race in Convicting and Sentencing. Social Science Quarterly, 66, pp 17885. 'Statistics New Zealand. (1996a). New Zealand Census of Population and Dwellings. Wellington: Statistics New Zealand. 'Statistics New Zealand. (1996b). Zealand. New Zealand Now Crime. Wellington: Statistics New 1Statistics New Zealand. (1993). All About Women in New Zealand. Wellington: Statistics New Zealand. Steffensmeier, D. (1980). Assessing the Impact of the Women's Movement on Sex-Based Differences in the Handling of Adult Criminal Defendants. Crime and Delinquency, 26, pp 344-357. 219 Steffensmeier, D., Kramer, J. (1982). Sex-based Differences in the Sentencing of Adult Criminal Defendants: An Empirical Test and Theoretical Overview. Sociology and Social Research, 66, 3, pp 289-304. Steffensmeier, D., Kramer, J., Streifel, C. Criminology, 31, 3, pp 411-446. (1993). Gender and Imprisonment Decisions. Stermac, L., Sheridan, P.M., Davidson, A, Dunn, S. (1996). Sexual Assault of Adult Males. Journal of Interpersonal Violence, 11, 1, pp 52-64. Steury, E.H., Frank, N. (1990). Gender Bias in Pretrial Release: More Pieces of the Puzzle. Journal of Criminal Justice, 18, pp 417-432. Struckman-Johnson, C, Struckman-Johnson, D. (1994). Men Pressured and Forced Into Sexual Experience. Archives of Sexual Behavior, 23, 1, pp 93-114. Tavris, C. (1992). The Mismeasure of Woman. New York: Touchstone. Teilman, K.S., Landry, P.H. (1981). Gender Bias in Juvenile Justice. Journal of Research in Crime and Delinquency, 18, 1, pp. 47-80. Thompson, B. (2000). 'Poverty, Social Assistance, and the Employability of Mothers: Restructuring The Welfare States, Book Review'. New Zealand Sociology, forthcoming. Thompson, B. (1998). Welfare Mothers: A Comparative Analysis. Unpublished Honours Dissertation: University of Canterbury. Tjaden, P.G., Tjaden, CD. (1981). 'Differential Treatment of the Female Felon: Myth or Reality?' In Warren, M.Q. (ed.). Comparing Female and Male Offenders. Beverly Hills: Sage. 'Triggs, S. (1999). Sentencing in New Zealand: A Statistical Study. Wellington: Ministry of Justice. (Triggs, S. (1998). From Crime to Sentence: Trends in Criminal Justice, 1986 to 1996. Wellington: Ministry of Justice. (1991), Women's Madness: Misogyny or Mental Illness? Ussher, J. HarvesterlWheatsheaf New York: (1983). Gender, Police Arrest Decisions, and Notions of Chivahy. Visher, C.A Criminology, 21, 1, pp 5-28. von Hirsch, A (1993). Censure and Sanctions. Oxford: Clarendon Press. Walklate, S. (1995). Gender and Crime: An Introduction. London: Prentice Hall Harvester Wheat sheaf. Wasik, M. (1998). 'Crime Seriousness and the Victim-Offender Relationship in Sentencing', in Ashworth, A, Wasik, M. Fundamentals of Sentencing Theory. Oxford: Clarendon Press. I 220 Wattenberg, W.W., Saunders, F. (1954). Sex Differences Among Juvenile Offenders. Sociology and Social Research, 39, 1, pp 24-31. Wilbanks, W. (1986). Are Female Felons Treated More Leniently By The Criminal Justice System? Justice Quarterly, 3, 4, pp 517-529. Wilkie, M. (1993). Sentencing Women: Pre-Sentence Reports and Constructions of Female Offenders. Nedlands: The University of Western Australia Crime Research Centre. Worrall, A. (1990). Offending Women: Female Lawbreakers and the Criminal Justice System. London: Routledge. Young, R. (1997). 'Bail and Sentencing Issues', in Ablett-Kerr, J., McDonald, E., Young, R. (eds.). Women in the Criminal Justice System: New Zealand Law Society Seminar. Wellington: New Zealand Law Society. Zingraff, M., Thomson, R. (1984). Differential Sentencing of Women and Men in the U.S.A. International Journal of the Sociology of Law, 12, pp 401-413. Newspaper References New Zealand Herald, 8-9-2000 'Row Erupts Over Mob Compo - Prebble: Payouts up to $90,000'. Dominion, 12-8-98 'No Sex in Jail, Says Former Guard'. The Christchurch Press, 29-9-98 'Call for Investigation of Relationships in Prison' . 221 Appendix One Additional Path Analysis Models FIGURE 18 - PATH ANALYSIS - IMPRISONMENT SENTENCING DECISION, DRUG OFFENCES -.22 -.18 Sex of Offender Seriousness of Criminal History .20 I Length of Custodial Remand -.26 Pre-Sentence Recommendations .37 Imprisonment Decision j -.23 = Multiple R .55 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher scores representing more serious criminal histories), Length of Custodial Remand (coded in days from zero upwards), Pre-Sentence Recommendations (O=imprisonment, 1=other), Imprisonment Decision (O=imprisoned, 1=not imprisoned). 222 FIGURE 19 - PATH ANALYSIS - IMPRISONMENT SENTENCING DECISION. PROPERTY OFFENCES -.20 -.26 Sex of Offender ~ Seriousness of Criminal History .17 ~ 1 Length of Custodial Remand -.41 ~ Pre-Sentence Recommendations .38 Imprisonment Decision ~ i1 -.29 !.....------(.14) Multiple R = .60 (p < .001) All path coefficients are statistically significant at p <.05 unless in brackets Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher scores representing more serious criminal histories), Length of Custodial Remand (coded in days from zero upwards), Pre-Sentence Recommendations (O=imprisonment, 1=other), Imprisonment Decision (O=imprisoned, 1=not imprisoned). FIGURE 20 - PATH ANALYSIS - LENGTH OF IMPRISONMENT TERM, DRUG OFFENCES -.27 .19 11 Sex of Offender -.25 ~ Seriousness of Criminal History ~ .11 Length of Custodial Remand ~ -.23 Pre-Sentence Recommendations -.20 Multiple R = .44 (p < .001) ~ -.25 1 Imprisonment Term i All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal histories), Length of Custodial Remand (coded in days from zero upwards), Pre-Sentence Recommendations (O=imprisonment, 1=other) , Imprisonment Term (a continuous variable in months, log transformation). 223 FIGURE 21 - PATH ANALYSIS - LENGTH OF IMPRISONMENT, VIOLENT OFFENCES .26 ... Negative Life Experiences -.19 .19 .20 Pre-Sentence Recommendations Sex of Offender -.35 ----. .28 .30 .20 -.38 Multiple R = .62 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal histories), Offenders Role (O=primary/equal, 1=secondary), Length of Custodial Remand (coded in days from zero upwards), Negative Life Experiences (O=no victimisation, 1=victimised either in childhood or in adulthood), Pre-Sentence Recommendations (O=imprisonment, 1=other), Imprisonment Term (a continuous variable in months, log transformation). FIGURE 22 - PATH ANALYSIS - THE REMAND STATUS DECISION, PROPERTY OFFENCES -.21 Sex of Offender -.34 -.31 Seriousness of Criminal History Bail Breaches .24 .15 1 Remand Status j Multiple R = .46 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Bail Breaches (O=yes, 1=no), Remand Status (O=remanded in custody, 1=remanded on bail). 224 FIGURE 23 - PATH ANALYSIS - LENGTH OF CUSTODIAL REMAND. DRUG OFFENCES Sex of Offender -.21 Seriousness of Criminal History .24 Length of Custodial Remand j -.24 = Multiple R .38 (p < .001) All path coefficients are statistically significant at p <.05 Coding for the variables in this model are as follows: Sex (O=male, 1=female), Seriousness of Criminal History (a continuous variable with lower scores representing less serious criminal histories and higher score representing more serious criminal), Length of Custodial Remand (a continuous variable, in days, a log transformation) 225 Appendix Two Supplementary Case-Study Results, Offenders' Crime Stories This appendix supplements Chapter Six where I presented some examples of cases judged as disparate. Presenting all the case-studies in Chapter Six would have been cumbersome, so all remaining cases of judicial disparity for both sentencing and remand are presented in this appendix. An example of a different and a similar judicial response is also provided to give the reader a better understanding of how such cases were assessed. 1 Different JUdicial Outcomes An example of a different, as opposed to disparate or similar, judicial response is presented below. A different judicial response was highlighted when offenders' criminality differed and they also received different outcomes. 1.1 Annabel and Mitch, Burglary Annabel (20 years, European) and Mitch (20 years, European) were both coded as pleading not guilty to more than one count each of burglary. Both offenders played active/equal roles in the offences and had co-offenders. There were also differences in the coded elements of these cases: Annabel had burgled commercial establishments and property losses were valued at $3,150, Mitch had burgled private homes and property losses were $1,300. It may be that burglaries of private homes are perceived more seriously than those of commercial properties, firstly because of "the extent to which an intrusion into the home affects privacy" (von Hirsch, 1993: 31) and secondly, because commercial victims, opposed to individuals, may be better equipped to "bear the loss" of crime (Wasik, 1998: 116). However, this was balanced somewhat by larger property losses in Annabel's case. Further investigation of the crime details showed that Mitch and Annabel committed three burglaries each and played similar roles. Mitch (in two burglaries) and Annabel (in three burglaries) helped remove property items from premises after their co-offenders had 'broken in' , although in one burglary Mitch waited in the car. In both cases, property from the burglaries was loaded into cars and both offenders were responsible for driving. Annabel and her cooffender were seen by police and pursued through town at high speeds; police stated that 226 Annabel's reckless driving endangered the public. Unlike Mitch's burglary, all property was recovered in Annabel's case. Annabel ,and her co-offender (Richard) went by car to shop premises in the central city. Annabel was driving the motor vehicle. The Richard gained entry into the premises by smashing a plate glass window with a baseball bat. Annabel and Richard then reached through the broken window and took merchandise valued at $100 dollars. The damage caused to the premises was valued at $1,000. Both offenders then returned to the car and Annabel drove them to a second shop and Richard gained entry by smashing the window. Richard and Annabel then took merchandise valued at $350.00. They returned to the vehicle and Annabel drove them to a third shop. The male co-offender subsequently smashed the window of this shop and both he and Annabel took merchandise valued at $1,700 dollars. However, two policemen passing by observed them so they ran back to their car. The car made off at high speed with Annabel driving. Police patrol cars chased Annabel for over 20 minutes through the central city. Annabel drove in a dangerous manner to aid their escape, failed to stop at stop signs, ran red lights and reached speeds of up to 120 km per hour. Police finally forced her off the road. All the property from the three burglaries was recovered by police from inside the vehicle. (Annabel - Burglary, Extracted from Trial Transcripts and Judge's Sentencing Remarks) Mitch and his male co-offender (Gary) went to a private home in the Christchurch area. Gary climbed over the back fence of the house while Mitch waited in the park behind the property. Gary broke the rear window of the house and gained entry. He placed jewellery and other household goods including a compact disc player into a bag. The total value of the goods taken was $1000.00. Gary then passed the goods over the fence to Mitch who hid the goods in the park. Mitch and Gary met back at their car and Gary drove the car to the park to uplift the proceeds of the burglary. Both offenders then left the area. The police did not recover the goods. A few days later Mitch and Gary drove to another house; Gary left Mitch in the car and broke into the address. Gary then came out of the house and asked Mitch to give him a hand. Mitch helped his associate pack the car with property from the house. They then went back to Gary's house where they both unload the property from the car. Then the Gary and Mitch drove to another private address. Gary stayed in the car, and his associated gained access to the house through an open window. He removed property from the address and packed it into the car. These goods were also taken back to Gary's house. The total value of goods taken was $300; they were recovered from Gary's flat. (Mitch - Burglary, Extracted from Trial Transcripts and Judge's Sentencing Remarks) I finally judged this pair's current offence seriousness as similar after weighing Annabel's driving and higher property values against differences in offence location (commercial establishment as opposed to private residence) and whether or not property was recovered. However, Mitch's offence history was more serious than Annabel's so his overall criminality was judged more severe: Mitch's had 13 prior convictions, all for property offending (including nine burglary convictions), his last conviction was ten months ago and he had served two prior terms of imprisonment. Annabel only had three prior convictions, none were for property offending, it was 17 months since her last conviction, and she had never been imprisoned previously. Given that Mitch's overall criminality was more serious than Annabel's it was not surprising to find that he received harsher judicial sanctions. Annabel was remanded on bail with special conditions and sentenced to 18 months supervision, Mitch was remanded on bail with special conditions and into custodial remand for 14 days, he was sentenced to twelve months imprisonment (see Tables 70 & 71 in Chapter 6). 227 2 Similar Judicial Outcomes An example of a similar, as opposed to a disparate or different, judicial response is presented below. A similar judicial response was highlighted when offenders' criminality was similarly serious and they received similarly severe outcomes. 2.1 Debbie and Bobby, Possession of a Class B Drug for Supply Debbie (45 years, European) and Bobby (24 years, European) each entered pleas of guilty to a count of possessing a Class B drug for supply. Both offenders played active roles and offended in private locations but Debbie had a co-offender while Bobby acted alone, possibly making him more blameworthy. Debbie had 16 prior criminal convictions including two prior convictions for possession of a Class B drug for supply, no prior convictions for breaching bail and it was three months since her last conviction. Debbie had been imprisoned once in the past. Bobby had two prior drug convictions (possession of a Class C drug) and it was six months since his last conviction. However, he had 30 convictions in total, including two for breaching bail, and had served two prior terms of imprisonment. Thus, Bobby's criminal history was considered more serious than Debbie's. A further look at the crime details showed Debbie's current criminality was in fact more serious than Bobby's. Debbie's drug operation was more sophisticated - she was clearly in the business of drug dealing and she was particularly uncooperative with police. Furthermore, Debbie was convicted of selling morphine while Bobby only dealt cannabis oil. The Ministry of Justice Seriousness of Offence Scale (1995) scores selling morphine (270) higher than selling cannabis oil (209). Balanced against Bobby's more serious criminal history, it was concluded that these offenders had similar criminality. A search warrant was executed by police on Debbie's house. When police arrived at her property Debbie and her male co-offender were located in the kitchen area of the house. Debbie was seen by police officers to be concealing an item in her mouth which she was attempting to swallow. After a struggle the police managed to extract the item from her mouth. It was found to be a package wrapped in gladwrap which contained 13, 60 milligram morphine sulphate tablets and 17, 30 milligram morphine sulphate tablets. A bottle containing five 30 milligram morphine sulphate tablets was also found in her bedroom. Other items found by police at Debbie's house included a radio scanner, some empty morphine sulphate bottles, a video camera set up to monitor the front door, a cell phone and money totalling $200 in $20 notes. (Debbie· Possession of a Class B Drug for Supply, Extracted from Police Summary of Facts) Police executed a search warrant on Bobby's house. The police found 48 capsules of cannabis oil hidden in various locations throughout the house and money totalling $265 hidden under the bed in a plastic bag. When spoken to by police Bobby claimed that the drugs were for his own use. (Bobby - Possession of a Class B Drug for Supply, Extracted from Police Summary of Facts) 228 Although this pair's criminality was considered similarly serious, they received different judicial outcomes. Bobby was remanded on bail with standard conditions and into custody for 95 days. Debbie was remanded on bail with special conditions and only spent seven days in custodial remand. However, Debbie was sentenced to 18 months imprisonment while Bobby only received 12 months. I could have categorised this case as disparate because offenders with similarly serious criminality received different judicial outcomes. However, judicial sanctions were finally considered comparative because Bobby's long custodial remand period meant his total incarceration period was similar to Debbie's (see Tables 66 & 67 in Chapter Six). 3 Disparate Judicial Outcomes All remaining 19 cases of judicial disparity for both sentencing and remand not presented in Chapter Six are presented below. 3.1 Megan and Rangi, Selling/Supplying a Class A Drug Megan (25 years, Maori) and Rangi (22 years, European) were both convicted on more than one count of supplying a Class A drug. Both offenders acted alone and supplied drugs in public locations but Rangi pled guilty while Megan pled not guilty which could have made her appear more blameworthy than Rangi. Rangi had more criminal convictions than Megan did (six compared with two) but in both cases offences were minor (e.g. shoplifting, obscene language, minor found in a bar) and none were for drug offending or breaching bail. Megan's last conviction was only six months before compared with two years in Rangi's case, which could have reduced the impact of his additional convictions, somewhat. Neither offender had previously received an imprisonment sentence. At the most, this meant that Rangi's criminal history was marginally more serious than Megan's. , Further investigation revealed similarities and differences in Rangi's and Megan's offending. Both offenders were convicted on two counts and had supplied Class A drugs to undercover police officers. Financial gain from the drug sales was higher in Rangi' s case but I decided this was balanced somewhat by Megan's uncooperative attitude toward police and the fact that she had sold LSD, while Rangi had only sold psychedelic mushroorns. 1 Undercover police officers were deployed in the Christchurch area for the purposes of obtaining evidence of the sale and supply of controlled drugs. During the course of their duties one of these police officers visited the work premises of Megan. Megan and the police officer had a general conversation during which she told the police officer she had been tripping on the previous weekend. She told the police officer what type of LSD trip she had been using and said that "there was plenty of it around if you knew the right people." He asked her if she could get him any and she said "sure I used to As far as selling/supplying is concerned, Cocaine and Heroin are the only drugs ranked higher than LSD in The Ministry of Justice's Seriousness of Offence Scale (1995). The drug in psychedelic mushrooms is psilocyhine, which is similar to LSD in its effects. 1 229 sell heaps" she informed the officer that they would cost $35 each tab; he asked for 5 tabs and paid her $175. The following day Megan telephoned the police officer and asked him to met where she worked. When he arrived she handed him a folded piece of paper containing 5 LSD tabs. The police officer retumed to Megan's place of work some weeks later and indicated to her that he would like to buy some more LSD. He handed Megan $175 and told her that it was for 5 LSD tabs. He then made arrangements to meet with Megan at her place of work the following day. The next day Megan gave the police officer $35 and 4 LSD tabs and apologised that she could only get four tabs. When approached by police Megan refused to answer any questions and contacted her lawyer. (Megan, Selling/Supplying a Class A drug, Extracted from Police Summary of Facts) Undercover police officers were deployed in the Christchurch area. During the course of their deployment two of these officers became involved with Rangi. One afternoon both officers visited a local hotel. While sitting in the hotel they were approached by Rangi and after a short conversation Rangi offered to sell the officers capsules of psilocine which is commonly known as magic mushrooms. They officers left the hotel with Rangi and drove to an address where the Rangi sold 5 capsules of psilocine to one of the officers for a total of $60. The following day, Rangi telephoned the other police officer at his home claiming he had 300 magic mushrooms for sale. It was arranged for the Rangi to met the officer that afternoon at a hotel. That afternoon Rangi brought 45 small and 10 larger capsules of psilocine with him. The price of the drugs were discussed and the police officer gave Rangi $400. When interviewed by police Rangi admitted selling drugs to the police officer. By way of explanation Rangi said that he "was not a drug dealer, never had been and never will be." He said "he just thought the police officer was a mug who would buy these things; he couldn't believe that anybody would pay money for them." (Rangi - Selling/Supplying a Class A drug, Extracted from Police Summary of Facts) In this case it was difficult to decide whose overall criminality was more serious, but I finally concluded that it was similarly severe. This decision was reached after weighing Rangi's extra criminal conviction against the shorter time frame since Megan's last conviction, the larger financial gains made by Rangi against Megan's not guilty plea, generally uncooperative attitude and the fact she dealt LSD. However, there were vast differences between Megan's and Rangi's judicial outcomes. Megan's imprisonment term (six months) was only half that of Rangi's term (twelve months) (see Table 67 in Chapter Six). Furthermore, Rangi was remanded in custody for seven days in addition to receiving bail with special conditions while Megan was only remanded on bail with special conditions (see Table 66 in Chapter Six). Disparity was subsequently concluded in this case. 3.2 Jackie and Sven, Manufacturing a Class B Drug Jackie (24 years) and Sven (23 years) were identified as European, both entered pleas of guilty, played active/equal roles, were sole offenders and manufactured drugs in private locations, but Jackie had more than one conviction count while Sven only had one. A further look at the crime details show both offenders manufactured cannabis oil for their own use and were noted to be co-operative with police. Jackie was convicted on two counts (manufacturing a Class B controlled drug and cultivating cannabis) but aside from this, there were no further differences in offending. I thus considered Jackie's current criminality to be more serious than Sven's. 230 Police executed a search warrant on Jackie's home address. Located growing in various places around the garden were 21 small cannabis plants. Another sixteen plants and seedlings were located growing in a bedroom at the rear of the address. Jackie admitted they were hers. Located in the garage was a large coffee jar containing cannabis leaf and Iso Propyl Alcohol. Iso Propyl Alcohol is the chemical used to manufacture cannabis oil from cannabis plant material. Jackie admitted that the cannabis and Iso Propyl Alcohol was hers and that she had mixed the two substances to make cannabis oil for her own use. (Jackie - Manufacturing a Class B Drug, Extracted from Police Summary of Facts) Police executed a search warrant on Sven's girlfriend's address. He was not present during the search, however his car was parked at the rear of the property. A search of the kitchen cupboards revealed a glass jar with a quantity of cannabis plant material in the bottom of it. An alcohol-based substance had been added to the cannabis material in order to produce cannabis oil. Latter that day, Sven arrived at the police station. He admitted exclusive possession of all the items seized by the police during the search. Sven admitted attempting to produce cannabis oil for his own use. He claimed to receive a greater effect from using cannabis oil than cannabis leaf. (Sven - Manufacturing a Class B Drug, Extracted from Police Summary of Facts) Neither offender had a history of drug offending, breaching bail or prior imprisonment terms but Sven had five prior convictions while Jackie only had one. Furthermore, Sven's last conviction was 13 months ago but it was over seven years since Jackie's. These differences meant Sven's past criminality was more serious than Jackie's. Although I do not judge this pair's overall criminality as equally serious, I do consider them comparable because Jackie's extra count compensated somewhat for Sven's more serious criminal history. Despite this, judicial outcomes differed: Sven was remanded into custody for 29 days in addition to receiving bail with special conditions and was sentenced to nine months imprisonment. Jackie was remanded on bail with standard conditions and sentenced to twelve months supervision (see Tables 66 & 67 in Chapter Six). Disparity was concluded because it was not overly clear why Sven had received substantially harsher judicial outcomes than Jackie. 3.3 Jenni and Ron, Conspiring to SelVSupply a Class B Drug Jenni (27 years, European) and Ron (28 years, European) were co-offenders, they entered pleas of guilty to one offence count each, both played secondary roles and supplied drugs from a private residence. Ron's past criminality was marginally more serious than Jenni's; she had seven prior convictions which included four for drug offending, Ron had twelve prior convictions, including nine for drug offending and had served one prior imprisonment term. Neither offender had breached bail in the past and it was approximately 14 months since either individual was last convicted. Thus the coded elements of this case suggested that Ron's overall criminality was marginally more serious than Jenni's. conclusion was unchanged after further looking at the crime details of these two cases. This 231 Both offenders were identified by police as playing a secondary role which involved telling people over the phone that drugs were available so I considered Jenni's and Ron's current criminality to be similar. Extensive electronic surveillance of two separate properties was undertaken by the police, this included recorded telephone conversations. Ron along with another male co-offender, girlfriend and child, resided at one of these address and Jenni resided with her children and another female co-offender at the second property. Ron's and Jenni's co-offenders were convicted of supply a Class B drug and were the major offenders in this case. However, the operation also revealed that Ron and Jenni had indicated to telephone callers that it was all right to come around for a grey, an orange, a sixty, a Napp, a single or a thing. These all being references to Morphine Sulphate Tablets. Telephone conversations between Ron, Jenni, and the other two co-offenders were also intercepted, and contained reference to drug supply. (Jenni and Ron· Conspiring to sell/supply a Class B Drug, Extracted from Police Summary of Facts) At best, Ron's overall criminality was marginally more serious than Jan's, given differences in criminal history, so slightly different judicial outcomes were expected. Small differences were in fact found for remand outcomes. Ron was remanded on bail with special conditions, into custody for 56 days and Jenni was remanded on bail with special conditions and into custody for 35 days. However, Ron was sentenced to three years, nine months imprisonment compared to twelve months imprisonment in Jenni's case. Differences in overall criminality may have explained minor difference in this pairs remand outcomes but I could not see how Ron's sentenced term of imprisoned came to be over three times longer than Jenni's. Thus disparity was concluded in sentencing but not remand for this pair of offenders (see Tables 66 & 67 in Chapter Six). 3.4 Amanda and Callum, Selling/Supplying a Class A Drug Amanda (45 years, European) and Callum (22 years, European) each pled guilty to more than one conviction count. Both offenders played active offence roles and dealt drugs from private locations, but Callum had co-offenders while Amanda supplied drugs by herself. Callum and Amanda were both fIrst time offenders. Thus the coded crime elements suggested that Callum's and Amanda's overall criminality was of similar seriousness. Further exploration of the crime stories revealed general differences in this pair's crimes but I still considered Callum's and Amanda's criminality to be similarly serious because differences were minor. Amanda was convicted on one count of supplying a Class A drug (LSD), two counts of possessing a Class C drug (Cannabis) for supply and one count of selling a Class C drug (Cannabis). Similarly, Callum was convicted on one count of supplying a Class A drug (LSD), two counts of supplying a Class C drug (Cannabis) and one count of selling a Class C drug (Cannabis). Amanda's drug dealing took place in her own home and police found over 232 $1,000 in her possession; proceeds from drug sales. Reports and documentation did not identify monetary amounts in Callum's case. Callum was found to be dealing drugs for a local gang who were involved in large scale commercial drug dealing at numerous locations throughout Christchurch. This gang connection possibly made Callum's offending appear more serious than Amanda's. Yet Callum was only a 'small time' player in a much larger drug dealing enterprise and harsh judicial treatment because of gang affIliation is, in itself, a discriminatory practice. Furthermore, Amanda's drug dealing was clearly an organised operation with some degree of professionalism. Callum was co-operative with police and admitted supplying drugs. Amanda admitted some drug dealing activity but denied others. Police executed another search warrant at Amanda's home. Amanda was located sitting on a bed in her bedroom. When informed of police presence she produced a plastic bag containing one ounce each of cannabis and a third contained two-thirds of an ounce of cannabis. Amanda's handbag was found to contain $780 in cash. A significant portion of this money was in $20 notes. A set of ounce scales was located on a shelf in Amanda's bedroom. A visitor at the address was found to be in possession of a cannabis bullet. Amanda openly admitted she had purchased the cannabis that morning. She had paid $400 per ounce for it and had sold two bullets from that supply. Amanda added that she had sold cannabis over a eight year period and had never been apprehended by Police. Amanda stated in explanation that the proceeds of the sales had been used to supplement her household expenses. One week later the Police intercepted a parcel awaiting courier delivery to Amanda's address. The parcel was opened by Police and found to contain 50 doses of the Class A controlled drug LSD. The following day the parcel was delivered to Amanda address. Amanda accepted the package, took it inside and opened it. The Police executed a search warrant at the address and located the LSD. Also found hidden in the pantry of the kitchen were two ounces of cannabis and $400 made up in $20 notes. In explanation, Amanda admitted receiving the package which had been sent to her. She admitted that the LSD was for sale, and that she had paid $11 per trip and expected to sell them for $25 each. In explanation for the cannabis found, Amanda said that she had obtained the cannabis from a friend. She denied that the $400 cash located was the proceed of drug sales. The police did not accept this. (Amanda - Selling/Supplying a Class A Drug, Extracted from Police Summary of Facts) The **** gang had been increasing it's drug dealing activities around the Christchurch area at the time of Callum's arrest. According to police, the gang had been setting up a number of drug dealing houses. These houses were run by gang associates and set up for the sole purpose of selling LSD, cannabis and cannabis oil. Overall, the police identified seven separate houses and arrested fifteen people (including Callum). Callum was identified as being involved in the selling of drugs from two houses. When spoken to by police he admitted dealing both cannabis and LSD from the houses. (Callum - Selling/Supplying a Class A Drug, Extracted from Police Summary of Facts) Regardless of judging this pair as having similarly senous criminality, Callum's judicial outcomes were more severe than Amanda's. Callum was sentenced to three years imprisonment while Amanda received a suspended imprisonment term, eight months periodic detention and an $800 confiscation order (see Table 67 in Chapter Six). In contrast, Amanda's remand outcomes were more serious than Callum's. She was remanded into custody for 22 days and received bail with special conditions while Callum was remanded into custody for seven days and received bail with standard conditions (see Table 66 in Chapter Six). Disparity was concluded in this case because criminality was comparative and it was unclear why Callum, a first time offender, was sentenced so harshly in comparison to Amanda. It was also unclear why, conversely, Amanda's remand outcome was harsher than Callum's. 233 3.5 Alison and Ray, Permit Premises to be Used in the Commission of a Class C Drug Offence Alison was aged 26 years and identified as Maori. Ray was 31 years and identified as European. The coded elements of this pair's crimes were identical, both offenders entered pleas of not guilty and were convicted on one count of allowing their homes to be used in the commission of a Class C Drug offence. It was two years since Ray's last conviction, compared to four years in Alison's case, but the extent and seriousness of Alison's overall criminality outweighed this. Alison had eleven prior convictions, including one prior conviction for cultivating cannabis but none for breaching bail. Ray had one prior conviction for property offending. Neither offender had received an imprisonment term in the past. A further look at the crime details showed that Alison's and Ray's crimes were very similar. Both offenders were guilty of allowing a similar number of cannabis plants to be grown on their property and neither offender was overly co-operative with criminal justice authorities. A search warrant was issued on Alison's home and eighteen small cannabis plants were found growing in individual pots in the garden. Alison was not home at the time and the police did not interview her until the next day. When questioned about the plants she claimed that it must have been done while she was away for a holiday. When questioned further about the ownership of the cannabis, Alison claimed that they might have belonged to her friend who had been staying at the house in her absence. It was later established at trial that Alison had not been away for a holiday and her friend admitted to possession of the cannabis. (Alison· Permits Premises to be Used in the Commission of a Class C Drug Offence, Extracted from Police Summary of Facts) A search warrant was issued on Ray's address and sixteen small cannabis plants were found growing in a green house. Ray was the owner of the property and lived with four male boarders. When questioned by police Ray denied any knowledge of the plants growing at his property. Ray's boarders later confessed that the cannabis plants belonged to them. (Ray· Permits Premises to be Used in the Commission of a Class C Drug Offence, Extracted from Police Summary of Facts) Although I considered Ray's and Alison's current criminality to be similar, Alison's overall criminality was judged more serious than Ray's because of differences in criminal history. Despite this, both offenders were remanded on bail with special conditions and Ray's sentence was more severe than Alison's. Ray received a $2,000 fine and $95 court costs while Alison was only fined $500 plus $95 court costs. Thus, disparity was concluded in this case because Ray was treated harshly in comparison to Alison (see Tables 66 & 67 in Chapter Six). 3.6 Monica and Ross, Manufacturing a Class B Drug Monica (39 years, European) and Ross (34 years, European) each entered pleas of guilty, to one count of manufacturing a Class B drug, they acted with others in private locations and played primary/equal roles. Both offenders had criminal histories but Ross's was 234 more serious than Monica's. Ross had 38 convictions in all, including eight for drug offending and two for breaching bail plus he had served a prior term of imprisonment. In contrast, Monica had 13 prior convictions including seven for drug offending, none were for breaching bail, and she had never been imprisoned. It was 13 months since Monica's last conviction and only seven since Ross's and he was imprisoned for this offence. Ross's criminal history was subsequently more severe than Monica's. Weighed against similarities in the coded crime elements, Ross's overall criminality thus appeared more serious than Monica's. Crime stories further supported this conclusion. Ross and Monica's offending was fairly similar but Ross was uncooperative with police. The police executed a search warrant on Monica's house. They located Monica and her male cooffender in the kitchen in the process of manufacturing morphine. A number of articles normally associated with the manufacture of morphine through what is commonly known as 'homebake' were clearly visible. The police asked Monica a number of questions and she stated she was manufacturing homebake to sustain her drug dependency. (Monica· Manufacturing a Class B Drug, Extracted from Police Summary of Facts) The police executed a search warrant on Ross's address. They located Ross and his male co-offender in the bathroom trying to flush substances down the toilet. Ross had to be restrained by police from doing so. Located in the Kitchen and Laundry areas of the house were glassware containers and buckner funnels and various other glassware. Also associated with this glassware were different chemicals which Ross had ready access to. In the kitchen area there were a number of large Panadine tablets, as well as the Panadine Ross was trying to dispose of down the toilet. These chemicals and glassware equipment are required for the manufacturing of the Class B controlled drug Morphine. When spoken to in relation to his actions Ross stated that all he was going to do was extract the codeine from the Panadine tablets and then drink the mixture. In relation to him tipping certain items down the toilet, Ross stated that he thought what he was doing was against the law, however he was not sure. Only traces of morphine were located at the address, these being empty sachets on the stove area. (Ross· Manufacturing a Class B Drug, Extracted from Police Summary of Facts) I expected Ross to receive more serious judicial outcomes because his overall criminality was judged more serious than Monica's. However, this only occurred with sentencing: Ross was sentenced to 24 months imprisonment, Monica received nine months imprisonment. In the case of remand, I concluded disparity because Ross and Monica were both remanded on bail with standard conditions (see Tables 66 & 67 in Chapter Six). 3.7 Helen and Richard, Manufacturing a Class B Drug Helen and Richard were both Europeans aged 22 years. Each offender pled guilty, played active roles as sole offenders and committed crimes in private locations. Helen and Richard were both convicted on more than one count. The coded elements of their criminal acts suggested similar criminality and further analysis of the crime details revealed only one small difference in this pair's offending. Both offenders were convicted on three counts (which included one count for producing a Class B, drug, possessing cannabis and cultivating cannabis), both co-operated with police and stated that all drugs involved were for their own 235 use. However, Richard was convicted of manufacturing Opium while Helen had produced cannabis butter. The Ministry of Justice Seriousness of offence scale scores the manufacture of Opium (181) slightly higher than cannabis butter (163). Thus, I considered Richard's current offence to be marginally more serious than Helen's. Police executed a search warrant on Helen's home. Helen and her two children occupied the address. In the kitchen at the address police found approximately two pounds of cannabis butter which had been recently produced by the defendant. The defendant admitted making the butter and stated that she intended using it. In pots at the rear of Helen's property police located twenty-three cannabis plants, which varied in height between one and two metres. The majority of the plants were large and mature. Police also located eighty-three cannabis seeds in the defendants home, along with a small amount of dried cannabis leaf. The defendant admitted growing the cannabis and stated that it was going to be for her own use. (Helen - Manufacturing a Class B Drug, Extracted from Police Summary of Facts) Police executed a search warrant of Richard's home. Richard and his son occupied the address. Two tablespoons containing a brown powdery substance were located in a cupboard in the kitchen. A third spoon containing several stained filters was lying beside the tablespoons. When spoken to about the spoons Richard told police the powdery substance was opium. He stated that the substance in the filters was opium and that he had used them to inject the drug. Richard told the police he had manufactured the opium. He told police he was about to inject the opium when police disturbed him. Located at the rear of the address were twelve potted cannabis plants. The plants were approximately 12-13 centimetres in height. A small jar of cannabis leaf was located on a table in the kitchen. Richard told police the plants and leaf belonged to him and were for his own use. (Richard - Manufacturing a Class B Drug, Extracted from Police Summary of Facts) It was 13 months since Helen's last criminal conviction and six months since Richard's but neither offender had previously been imprisoned. Helen had no prior drug or bail breaching convictions and although Richard also had no prior convictions for breaching bail, he did have one for cultivating cannabis. Despite Richard's prior drug conviction, and the fact tnat he had manufactured opium, Helen's overall criminal history was more serious because she had 109 prior criminal convictions, compared to ten in Richard's case. I considered this an explanation for why Helen's sentence was more severe than Richard's: Helen was sentenced to six months imprisonment; Richard to 24 months supervision (see Table 67 in Chapter Six). However, Richard's remand outcomes were more serious than Helen's: Richard was remanded into custody for 42 days and remanded on bail with special conditions, Helen was remanded into custody for 20 days and received bail with standard conditions (see Table 66 in Chapter Six). Thus, while varied sentencing outcomes were categorised as different and explainable, remand outcomes were classed as disparate; the woman's criminality being more serious, but the man's remand outcome was more serious. 3.8 Penni and Tony - Permit Premises to be Used in the Commission of a Class B Drug Offence Penni (27 years, European) and Tony (34 years, European) both entered pleas of guilty to one count of permitting their premises to be used in the commission of a Class B Drug 236 Offence. They both played primary roles in the actual offence, in which they were essentially accessories to their co-offenders' more serious drug crimes. I considered Tony's criminal history to be marginally more serious than Penni's because he had one extra conviction. Penni had one prior conviction for cultivating cannabis 27 months before her current conviction. Tony had two prior convictions including one for possession of cannabis twelve months prior to the current conviction. Neither offender had been previously sentenced to imprisonment. Crime details showed that Tony's and Penni's drug offending was linked to their spouses. I judged Tony's and Penni's overall criminality to be similar after reading the crime details because Tony was more co-operative with police which balanced small differences in criminal history. Police executed a search warrant on Penni's address. Penni answered the door and police went to the kitchen area. Penni's partner and a friend were located in the kitchen in the process of manufacturing methamphetamine. The police asked Penni if she knew what had been going on in the kitchen before they arrived. Penni claimed that she had been asleep in the bedroom and did not realise what had been going on. (Penni - Permits Premises to be Used in the Commission of a Class B Drug Offence, Extracted from Police Summary of Facts) Police executed a search warrant on Tony's address. They located glassware, chemicals and codine tablets used in the manufacture of homebake morphine. Tony said that the equipment belonged to a friend of his and that his friend had intended to 'do a bake' there later that day. Tony said that this friend had been using his house for a few months to manufacture morphine. Tony said that in return for hiding the equipment and letting his friend use his house he received morphine to support his wife's drug habit. (Tony - Permits Premises to be Used in the Commission of a Class B Drug Offence, Extracted from Police Summary of Facts) The similarity of Tony's and Penni's overall criminality is reflected by similarity in sentencing: Tony was sentenced to 200 hours community service and Penni received 60 hours community service plus a $400 fine. However, there were disparate remand outcomes: Tony was remanded on bail with special conditions while Penni received bail with standard conditions (see Tables 66 & 67 in Chapter Six). 3.9 Tania and Allan, Aggravated Burglary Allan (25 years, European) and Tania (19 years, Maori) both entered pleas of guilty to aggravated burglary and many coded aspects of their offences were similar. Both had primary roles in the offences which took place in private dwelling houses, they used knives and did not know their victims. However, the coded element of this case suggested Tania's offence was more serious than Allan's. Tania was convicted on more than one count, acted alone, and 237 caused her female victim physical injury. In contrast, Allan was convicted of one count, he had a co-offender and his male victim was unharmed. It was found, after further exploration of the cnme details, that Allan's male victim was unharmed because he was not in the house at the time of the offence. A security camera had captured Allan's offending. Allen and his male co-offender took property valued at $1,000 and this was never recovered. I considered the coded elements of Tania's offence as more serious than Allen's because she was convicted on more than one count and she caused harm to the victim. Despite acting alone (which could potentially make her less threatening than Allan), Tania's offending remained more serious once the crime stories were analysed. Tania was convicted on two counts; the second count being for attempted burglary. Although Tania's burglaries did not end with financial gain like Allan's, I judged Tania's crimes to be more serious. Both crimes were planned and in the case of the aggravated burglary, Tania invaded the home of a particularly vulnerable woman, attacked her with a knife in the presence of her infant and caused physical injury. The victim impact statement also noted Tania's victim suffered emotional trauma. Tania made plans to go to a dwelling in Christchurch to commit a burglary. Taking a bag and a screwdriver to assist her in the crime, she drove to a local house. Tania went onto the property and began to jemmy windows and doors to break in. When this did not work Tania picked a large rock out of the garden and threw it through a window to gain entry into the victim's house. Once inside she went into the kitchen drawer and removed a large kitchen knife. Carrying the knife with her she moved about the house accumulating property. Eventually Tania entered the bedroom where the female victim and infant son were sleeping. The victim in this matter is disabled. The victim awoke in the darkened room, to see Tania before her. Tania turned on the victim while she was still in her bed and thrust the knife, point first, towards her face and throat. Tania pushed the knife downward with both hands, the victim strenuously held the knife at bay by pushing away with her hands. This struggle went on for a couple of minutes, during which time the point of the knife pushed onto the victims face cutting her but this injury was not serious. Tania abandoned the struggle leaving the property she had accumulated behind. (Tania - Aggravated Burglary, Extracted from Police Summary of Facts) The victim left his address early in the evening and locked his house. The victim also had a security camera, which he activated before he left. The victim arrived home the next morning to find he had been burgled. The total value of the property stolen was $1,000. The police attended the burglary and took possession of the security videotapes. These tapes showed the offenders entering the premises at approximately in the early hours of the morning. One of the offenders (Allan's co-offender) was carrying a baseball bat and Allan was carrying a knife. Both offenders were heavily disguised and were wearing full-face balaclavas. During the course of the burglary the offenders removed their balaclavas and were captured on the security camera. (Allen - Aggravated Burglary, Extracted from Police Summary of Facts) When spoken to by police, Tania was noted to have been: "co-operative, remorseful and confessed to the facts as they were outlined." She told police she was going through a difficult stage in her life and was under the influence of illegally obtained medication at the time." Tania's co-operation with police and her 'difficult time' could be seen as mitigating factors, 238 but the next day she attempted a second burglary. This tended to negate Tania's remorseful, cooperative attitude in the earlier instance. "Once again, taking a screw driver to assist she went onto a property. Before doing anything further, she knocked on the door to establish if anybody was home; the occupant of the house was home, answered the door and spoke to Tania. Tania pretended she was looking for a friend and upon being told she was at the wrong address, she lef!." (Tania - Attempted Burglary, Extracted from Police Summary of Facts) Allen was also co-operative with police. The explanation given for his offending possibly reduced Allan's culpability because he claimed his co-offender had coerced him into the crime with threats: When interviewed Allan readily admitted his involvement in the aggravated burglary. He stated that his co-offender had pressured him into it; telling him he would rape his wife. (Allan - Aggravated Burglary, Extracted from Police Summary of Facts) Despite Tania's young age (19 years), I considered her current criminality more serious than Allan's because she was convicted on two counts, attacked a particularly vulnerable person and caused a great deal of trauma. However, their overall criminality was finally assessed as similar because of differences in criminal history: Allan had 50 prior convictions, including three for violence, and one for breaching bail plus a history of imprisonment. 2 Tania had seven convictions including one for breaching bail, none for violence and no history of imprisonment. Tania and Allan both had prior convictions within the previous six months. The similarity of this pair's overall criminality was reflected in similar remand outcomes. Tania and Allen were both remanded on bail with special conditions. However, sentencing outcomes were disparate. Allan received two years nine months imprisonment compared to Tania who received eleven months (see Tables 68 & 69 in Chapter Six). 3.10 Selina and Roger, Kidnapping and Aggravated Robbery Roger (23 years, Maori) and Selina (22 years, Maori) each pled not guilty to more than one conviction count. They were co-offenders, played active/equal roles and caused physical harm to a man unknown to them in a private location. Further analysis of the crime details showed that these offenders were convicted on two counts (kidnapping and aggravated robbery), their offence was planned, and the victims were assaulted repeatedly. At times Roger appeared to take more control of the situation than Selina 239 (this is italicised in the narrative below) but this was not substantial so I concluded that this pair's current criminality was similarly serious. Being a man, the judiciary may have presumed Roger was the primary culprit and that Selina was 'under his influence'. His physical size and strength, compared with Selina, could also have made him appear more 'scary', and directly, he could have caused more physical damage. However, it would have been unjust to punish Roger more harshly because of physical size, strength and judicial perceptions of gender relations between 'typical men' and 'typical women'. In the early hours of the morning the victim met Roger and Selina at a local bar. He had been drinking and was intoxicated. Later in the evening the victim got into a taxi with the offenders and went to Roger's house where they continued to drink alcohol. While the victim was in the bathroom the offenders discussed the possibility of robbing the victim and as a result took him for a walk to a local park. Once there Roger and Selina took turns kicking and punching the victim in the head and stomach. They took him back to Roger's house, where the beatings continued both inside and outside the house by both offenders. The victim was hit in the head and face during the course of these assaults. Roger and Selina demanded that the victim give them the pin number to his credit card. When the victim finally told the offenders his pin number he was physically forced into a car and driven to an automatic teller machine where Roger attempted to make a cash withdraw from his account, but this was unsuccessful. The offenders then took the victim to where Selina lived and he was physically forced into the house where they again assaulted him and demanded to know where his money was. However, the victim had no money so the offenders again forced him into their car. They drove to a petrol station where Roger purchased some petrol and hot chips using the victim's credit card. They drove the victim into the countryside; the victim felt he was going to be murdered. In a quiet country area the offenders stopped the car. Roger ordered the victim out of the car and to follow him; they walked towards some trees at the side of the road, but the victim seeing an opportunity to escape, ran into the trees. He then found a farmhouse and telephoned the police ... As a result of the attack the victim received bruising to his face, head and body; he suffered headaches for some weeks after. Since the attack the victim is afraid to leave his house. When apprehended both offenders were cooperative. (Selina and Roger - Kidnapping and Aggravated Robbery, Extracted from Police Summary of Facts, Trial Transcripts and Judge's Sentencing Remarks) Roger's and Selina's criminal histories differed slightly: Selina had two prior convictions while Roger had ten, but both had one prior conviction for violence (common assault). Neither offender had prior convictions for breaching bail or had received imprisonment sentences in the past. It was approximately twelve months since either offender was last convicted. In contrast to Selina, Roger's extra criminal convictions required harsher judicial treatment but I was not prepared for the extent of the differences incurred. Roger was remanded in custody throughout the course of the trial (295 days) but Selina only spent seven days in custodial remand; the remaining time she was on bail with special conditions (see Table 68 in Chapter Six). Further differences were found in sentencing: Roger was sentenced to two and a half years imprisonment (added to his custodial remand period, Roger's imprisonment term was over three years long). Selina, on the other hand, received nine months imprisonment (see Table 69 in Chapter Six). Already substantial sentencing differences were made worse in this case because aggravated robbery is a serious violent offence (Criminal Justice Act 1987 [s93]). 2 He had served three prior terms of imprisonment. 240 Roger, (being sentenced to over two years imprisonment), would be ineligible for parole and required to serve two-thirds of his sentence before being released on remission with conditions (if the parole board thought this necessary) (Criminal Justice Act 1993 [s99, s101]). On the other hand, Selina would not appear before the parole board - she would be released unconditionally after half her sentence was served (Criminal Justice Act 1993 [s90]). Disparity was concluded in this case because it was not clear why such vast differences in incarceration time were warranted (see Table 69 in Chapter Six). 3.11 Joanne and Grant, Robbery Joanne (32 years, Maori) and Grant (23 years, Maori) both pled guilty to on one count of Robbery, played primary roles in their offences, caused physical injuries and victimised men who were unknown to them in private dwellings. However, Joanne had a co-offender while Grant acted alone and it has been noted that groups could be more intimidating than sole offenders. However, it is also possible that individual culpability is more likely to be questioned and reduced if a group is involved in the offending. Further investigations of the crime stories showed that Joanne had attacked a man she did not know on his private property, causing him and his daughter (who had seen the attack on her father) much distress, and property taken was not recovered. In contrast, Grant's offending took place at a party. His male victim was physically injured, but there was no evidence of long-term emotional trauma and police recovered the victim's property. The presence of a male co-offender in Joanne's case may have diminished her blameworthiness in the eyes of the court, particularly since her co-offender was also her partner (suggesting she was 'under his influence'), but the crime details clearly showed Joanne was the primary perpetrator. Neither Joanne nor Grant were noted to be uncooperative with the police. Joanne and her male co-offender forced their way into a house through a window. Once inside, they took some money and alcohol from a bedroom but were subsequently disturbed by the male owner of the house and his daughter coming home so went back out the window. Leaving the alcohol and money behind Joanne and her co-offender ran out onto the street but the homeowner, having seen them went into the house and telephoned the police. Whilst on the phone to the police, Joanne came back to the front door and told the man she had left something behind. The man blocked the door to the house and Joanne proceeded to push, pull and punch him in the face in an attempt to get back into the house. It was noted that Joanne was a woman of quite big build, while her victim was described as a small man. The victim retaliated to Joanne's attack and punched her in the chest. As this struggle was going on Joanne's co-offender came back to the house and told the victim to leave Joanne alone. The victim did as he was told and Joanne went back into the house to pick up the money and alcohol that they had left behind when they had been interrupted earlier. The victim went to hospital and was treated for bruising to his head, he also suffered headaches for about a week after the attack. The victim's daughter, who was present at the time, was emotionally affected by seeing her father attacked; experiencing nightmares and anxiety for months afterwards. She subsequently attended counselling to deal with her fears. The police did not recover the money and alcohol taken by Joanne and her co-offender. (Joanne - Robbery, Extracted from Police Summary of Facts) 241 Grant's robbery took place at a party in the victim's home. Grant walked up to the victim who was sitting down, punched him in the face and demanded his jacket. The victim replied "no" so Grant continued to punch him in the face and head; he then pulled the jacket of his victim and told him to keep his month shut. Due to the injuries received, the victim was taken to hospital. His injuries included bruising to his neck, back, legs, an eye, and chipped teeth. The Jacket was latter recovered from Grant by police who found him wearing it. When questioned, Grant stated that the man had hit him first, that he retaliated and took his jacket to teach him a lesson. (Grant - Robbery, Extracted from Police Summary of Facts). Joanne's past criminality was considered marginally more serious than Grant's. Joanne had fewer prior convictions than Grant overall (33 compared to 42) and had served fewer imprisonment terms (three compared to five). However, Joanne had an extensive history of violent criminal behaviour including: assaulting a police officer, common assault, assault with intent to rob, indecent assault, and wounding with intent to cause grievous bodily harm. Grant only had one prior conviction for violence (common assault) and the majority of convictions were for property offending. Furthermore, Joanne was convicted of a criminal offence only one year prior to her current court appearance but Grant had not appeared before the court in over two years. Grant and Joanne both had a prior conviction for breaching bail. Sentencing and remand outcomes differed substantially between Joanne and Grant: Joanne was remanded on bail with special conditions, received a suspended sentence of imprisonment and nine months periodic detention. Grant was remanded in custody for 210 days and imprisoned for twelve months (see Tables 68 & 69 in Chapter Six). Group offending versus individual offending provides a possible explanation for these differences because Joanne's culpability may have been questioned particularly since her co-offender was also her partner. However, given her substantial history of violence, this should have mitigated Joanne's culpability only slightly. A disparity in judicial sanctioning was concluded because Grant was imprisoned and Joanne was allowed to stay in the community. 3.12 Joan and Andrew, Aggravated Robbery Joan (19 years, European) and Andrew (21 years, Maori) each entered pleas of guilty to one count of aggravated robbery. Both had primary roles in the robberies, which occurred in public locations, and they used knives to threaten their victims. In her robbery, Joan victimised a woman and had a co-offender. In contrast, Andrew acted alone and his victim was male. Crime stories revealed differences in Joan's and Andrew's current offending but these factors balanced each other to produce crimes of comparable seriousness and offenders of comparative culpability. Both Joan's and Andrew's offences could be viewed as "typical stickups" which Daly (1994: 99) defines as occurring "on the street or against a commercial establishment; the offender uses a weapon and does not know the victim." Andrew's offence took place in a 242 commercial establishment (a service station) and Joan's offence was a street robbery. Joan's robbery netted her $100 compared to Andrew's $1,000 but this difference was outweighed by four other factors: All monies taken by Andrew were recovered within minutes of the robbery but Joan's takings were not found, Joan used physical force unlike Andrew, Joan victimise two women but there was only one male victim in Andrew's case, and Joan was uncooperative with the authorities. In the early hours of the morning Joan and her female co-offender robbed two women walking in the centre of the city. Joan grabbed one of the women from behind, held a knife to her throat and demanded money. She threatened the victim saying she would "really do it" unless the women handed over her purse. Joan then pushed the victim's face up against the wall of a building and demanded she hand her purse over to her co-offender which she did. Joan's co-offender then went through the purse taking out money and credit cards. Joan theFl pointed the knife at the second woman and demanded money. The second victim handed over what money she had. Joan then ordered the two women to run away. When questioned by police Joan denied any involvement, and gave the police a false name. She continued to plea not guilty until the trial date was set and then she changed her plea to gUilty. (Joan - Aggravated Robbery, Extracted from Police Summary of Facts) Andrew walked into a service station carrying a sports bag, wearing sunglasses and a hooded jacket; with the hood pulled over his head. He walked up a man working at the counter, leaned across, pushed the bag towards him and said, "fill it up'. The service station attendant replied "what?" and Andrew again said "fill it up." He then produced a knife, which he had been holding in his hand. The service station attendant then complied with Andrew's request and placed a sum of money into Andrew's bag. Andrew then left the service station. Two other male service station workers, a member of the public and a traffic officer followed Andrew from the service station until the traffic officer apprehended him. When spoken to by police, Andrew admitted being responsible for the robbery. He said he needed the money to payoff some debts and get out of the city. (Andrew - Aggravated Robbery, Extracted from Police Summary of Facts) Joan's criminal history was arguably more serious than Andrew's. She had 31 prior convictions compared to Andrew's seven. Joan also had two prior convictions for breaching bail, eight for violence (including six for common assault, one for assault with a weapon and one for assault on a police officer) and had previously served three terms of imprisonment. Andrew had no prior convictions for breaching bail, had two prior convictions for violence (aggravated robbery and common assault) and served two prior imprisonment terms. Andrew's prior conviction for aggravated robbery is serious because it is the same as the current conviction, meaning Andrew could be seen as more culpable. However, I considered this was balanced, if not exceeded, by Joan's more extensive history of violence. It was less than three months since either Joan or Andrew had been convicted of a criminal offence. I concluded that Joan's overall criminality was more serious than Andrew's because while their current offending was similar, Joan's criminal history was worse. However, Andrew's imprisonment term was three years and nine months as opposed to 18 months in Joan's case (see Table 69 in Chapter Six). Aggravated robbery is defined in law as a serious violent offence (Criminal Justice Act 1987 [s93]) so substantial differences in imprisonment terms are 243 compounded further in this case. 3 Joan (64 days) was remanded into custody for slightly longer than Andrew (59 days) so disparity was not evident in remand. However, disparity was concluded in the case of sentencing because, in contrast to Joan, Andrew had been punished more severely for less serious criminality. 3.13 Belinda and Regan, Common Assault Belinda (24 years, European) and Regan (26 years, European) entered pleas of guilty to one count each of common assault. Both acted with others, had primary or equal roles in the assaults, caused physical injury and victimised men they did not know in public locations. After further exploration of the two assaults, I considered Belinda's current crime as more severe than Regan's. Belinda's victim was particularly vulnerable (mentally impaired), he had innocently bumped into Belinda at a music concert and was assaulted as a result. Roger's offending took place in the context of a bar fight. His victim provoked the assault and I considered the incident an act of mutual aggression. As a result of these incidences, Regan's and Belinda's victims received bruising to the facial area but neither required hospitalisation. Belinda and her male co-offender were at a music concert in the central city when two disabled men walked in front of them and knocked Belinda's drink out of her hand. Belinda and the co-offender started punching and kicking the two men in response. At one stage while her partner held down one of the men Belinda was seen to kick him in the head and punched him repeatedly. A by-stander pulled Belinda of her victim. (Belinda· Common Assault, Extracted from Police Summary of Facts) Regan's offending took place in a local hotel while he was drinking with a couple of male acquaintances. Regan and one of his acquaintances began arguing with another man in the pub; a fight eventually broke out and in the course of this fight Regan punched the man in the face causing him to fall to the ground. (Regan· Common Assault, Extracted from Police Summary of Facts) I judged Regan's past offending to be marginally more serious than Belinda's. Neither offender had prior convictions for violence or served previous terms in prison but Regan had more prior convictions than Belinda (54 compared with 49). Both offenders had prior convictions within the last six months. Small differences in the number of Regan's prior convictions were exceeded by the more serious nature of Belinda's current crime and I finally judged Belinda's overall criminality to be marginally more serious than Regan's. Yet despite this, similar sentences Offenders sentenced to more than 2 years imprisonment for aggravated robbery are ineligible for parole and serve two-thirds of their sentences before being released on remission; this release may have conditions attached, if deemed necessary by the parole board (Criminal Justice Act 1993 [s99, s101]). Offender's sentenced to imprisonment terms of 12 months or less do not appear before the parole board and are released unconditionally after half their sentence is served (Criminal Justice Act 1985 [s95]), 3 244 were given: Belinda was sentenced to five months periodic detention, Regan received months imprisonment suspended for 18 months (see Table 69 in Chapter Six). SIX 4 Both Regan and Belinda were remanded on bail with standard conditions (see Table 68 in chapter Six). For the purposes of sentencing, Belinda's criminality was considered more serious than Regan's. However, Regan had six prior convictions for breaching bail. Thus in the case of remand, the seriousness of this pair's criminality was balanced and similar remand outcomes were expected. 3.14 Rebecca and Jason, Aggravated Burglary Rebecca (32 years, European) and Jason (35 years, Maori) were co-offenders and entered pleas of not guilty on two counts (aggravated burglary and assault with a weapon). The coded elements of their crimes were similar. Both carried weapons into a private location and physical injury was caused to a woman known to them. However, Rebecca, unlike Jason, was coded as playing a secondary role in the offence which may have mitigated crime senousness. Further exploration of the crime stories confirmed Jason's offending was more serious than Rebecca's. Although both offenders had broken into their victim's home carrying weapons, Rebecca only used her weapon to threaten the victim while Jason used his to cause injury. It was concluded that Jason's current criminality was more serious than Rebecca's. Rebecca and Jason went to the address of a former a female associate who they believed had stolen drugs from them. They wore balaclavas and carried baseball bats to the scene. On reaching the front door they proceeded to smash in the window with the baseball bats. Once they had gained entry Jason attacked the victim with his baseball bat. Rebecca threatened the victim - telling her that she was a thief and was going to pay. (Rebecca and Jason Aggravated Burglary, Extracted from Trial Transcripts and Judge's Sentencing Remarks) Rebecca's criminal history was more extensive than Jason's: 90 prior convictions, compared with 56. Her last criminal conviction had been 13 months before her current appearance, while Jason had not been convicted in over three years. Although Rebecca's offence history initially appeared more serious than Jason's, I finally considered Jason's criminal history to be more serious because he had four prior convictions for violence and had been sentenced to imprisonment on eleven previous occasions. In contrast, all Rebecca's previous convictions A suspended sentence of imprisonment is a final warning to offenders (re-offending within the specified period will, in most cases, results in imprisonment) and is subsequently a serious sentencing outcome. However, aside from restricting criminal behavior, by itself a suspended sentence impacts little on an offender's liberty. Periodic detention on the other hand does impact on offenders' liberty. The offender is required to work at a detention centre once a week. 4 245 were for petty property offending and she had only served three prior imprisonment terms. Differences in offending roles and history of violence subsequently explained why Rebecca's imprisonment term (18 months) was shorter than Jason's (30 months). However, there was no difference in remand status; both offenders were remanded on bail with special conditions. Thus, while difference rather than disparity was concluded for sentencing, disparity was concluded for remand (see Tables, 68 & 69 in Chapter Six). 3.15 Ruth and Rieki, Arson Ruth (21 years, European) and RicId (20 years, European) each entered pleas of guilty to one count of arson; they had both acted alone and victimised organisations. Ruth's arson resulted in substantially more damage than RicId's ($70,000 compared to $1,000). Ruth was a fIrst time offender, RicId was not. RicId's last conviction was over three years ago, he had never been imprisoned but had five prior convictions in total, including one for attempted arson. However, larger monetary losses in Ruth's case were compounded when crime details were analysed and I concluded that Ruth's current criminality was more serious than RicId's. Ruth had set fire to her business property, methodically planning and executing the arson to obtain insurance. In comparison, RicId's fire had charred the wall of a small tool shed and only a small degree of planning was evident. Ruth and her husband took possession of a new business property, which they had purchased. Later that year, the fire service was called to this property because a fire had started in the waiting room. The fire service investigated the fire and concluded the investigation as an accidental cause. The following week the fire service was again called to the property. On this occasion the dwelling was ablaze. During the fire someone had heard coughing coming from the burning building. This person crawled through the building (which was well ablaze) and saw a woman lying on the floor. This person then assisted the woman, being Ruth, out of the building. When initially spoken to by the fire service and pOlice, Ruth claimed she had returned to the property that morning to clean the property following the previous fire. She said that on entering the address she was hit over the head with a hard object and had fallen to the floor. Ruth then stated she heard a male voice talking and felt rough hands fondle her between the legs. She then stated she was dragged through the house; finally going into a state of unconsciousness. As a result of Ruth's very serious allegations, specialists were engaged to examine the scene of the fire. A medical practitioner also examined Ruth and a large number of detectives were deployed to carry out an investigation. It was latter revealed that Ruth had in fact set fire to the business property for the insurance. She claimed that her husband was under a lot of financial pressure. (Ruth - Arson, Extracted from Police Summary of Facts) Ricki was at a local school in the early evening. He approached a wooden tool shed, picked up a rock threw it through the window, reached inside and opened the window. Ricki was carrying a tin of lighter fluid; he poured this onto some rags adjacent to the window and ignited them. He then left. In explanation the Ricki stated he was angry for things he had done in the past and he wanted to punish himself. The fire was extinguished before any extensive damage was done. The shed suffered charring along one wall. (Ricki - Arson, Extracted from Police Summary of Facts) RicId's past offending, particularly his past conviction for attempted arson, made him more blameworthy than Ruth and different judicial responses were expected. However, after 246 compensating for the seriousness of Ruth's current offence, I judged their criminality to be similarly serious. Thus the judicial outcomes, which were far more severe in Ridei's case, were considered disparate. Ruth was remanded on bail with special conditions and sentenced to eight months periodic detention. In contrast, Ricki was remanded into custody for 70 days and was sentenced to three years imprisonment (see Tables 70 & 71 in Chapter Six). 3.16 Judy and Pete, Fraud In this case, both offenders had acted with others, played primary/equal roles in the offending, committed fraud against organisations rather than individuals, and were convicted on more than one count. Judy (46 years, European) entered a plea of guilty and her crime netted her $3,000, and while Pete (48 years, European) also entered a plea of guilty but the proceeds of his crime were only valued $1,000. Thus, Judy's offending was arguably more serious than Pete's because of larger monetary amounts. Exploration of the crime details further extenuated the seriousness of Judy's current criminality compared with Pete's. She was convicted on nine counts of fraud, compared with eight counts in Pete's case and she did not co-operate with police, while Pete did. Using a stolen chequebook Judy claimed she obtained from a male boarder, Judy, accompanied by the male boarder, went to nine separate retail outlets over a three day period. The types of goods obtained were food from a supermarket, cosmetics, jewellery and women's clothing from numerous stores throughout the city. Judy signed all cheques. When questioned by police Judy denied the offence and refused to give samples of her handwriting. (Judy - Fraud, Extracted from Police Summary of Facts) Using a stolen credit card Pete and a male acquaintance purchased paint and other DIY accessories over the period of one month. When questioned about his actions Pete admitted his actions and stated that he "couldn't help himself because he has a drinking problem." Pete signed all credit card slips. (Pete - Fraud, Extracted from Police Summary of Facts) Judy's current criminality may have been more serious than Pete's but his criminal history (63 prior convictions) was more extensive (25 prior convictions). Pete also had 48 prior convictions for fraud, as opposed to 13 in Judy's case and had previously been imprisoned on four occasions while Judy had only been sentenced to imprisonment once. Both offenders had prior convictions for breaching bail and it was approximately six months since their last criminal convictions. Overall, I considered this pair's criminality to be similar once the more serious nature of Judy's current offence was balanced against Pete's criminal history and this was reflected in similar remand outcomes - both offenders were remanded on bail with standard conditions, regardless of prior bail breaches (see Table, 70 in Chapter Six). However, Pete's sentence was more 247 severe than Judy's: twelve months imprisonment, as opposed to 24 months supervision (see Table 71 in Chapter Six), so I concluded disparity in sentencing outcomes. 3.17 Sandra and Terry, Fraud Sandra (38 years, European) and Terry (37 years, European) were co-offenders and each entered pleas of not guilty to one count of fraud. They both played an active/equal role in the offence, and victimised someone known to them, from whom $500 was illegally taken. Sandra had six prior criminal convictions, including three for property offending and it was four years since her last conviction. In contrast, Terry had seven prior convictions, none were for property offending and it was over seven years since his last conviction. Neither offender had prior convictions for breaching bail or had served prior terms of imprisonment. Compared with Terry, Sandra's history of property offending and the shorter time frame between current and past offending, suggested her criminality was marginally more serious. Crime details provided no evidence suggesting Terry's offence was more serious than Sandra's. Terry and Sandra lived together in rented accommodation. After failure to pay rent their landlord sought compensation from the Tenancy Tribunal. At the Tribunal hearing Terry and Sandra produced a receipt which they claimed showed that they had paid two months rent. As a result, they were awarded the sum of $500 by the tribunal. However, after further investigation it was found that the receipt had been altered. When spoken to by police both Terry and Sandra denied altering the document. (Sandra and Terry - Fraud, Trial Transcripts and Judge's Sentencing Remarks) However, Sandra and Terry received the same remand outcome (bail with standard conditions) but Terry was given a harsher sentence: He received $250 reparation plus four months periodic detention; Sandra was sentenced to just $250 reparation with no Periodic Detention Order (see Tables 70 & 71 in Chapter Six). Thus I concluded judicial disparity because different outcomes did not appear warranted. 3.18 Susan and Brad, Theft Susan (31 years, Maori) and Brad (23 years, European) entered pleas of guilty to theft. Both offenders acted alone taking property of similar value, $500 in Susan's case and $700 in Brad's. However, Susan was convicted on more than one count while Brad was only convicted on one count. This suggested that Susan's offending was more serious than Brad's but Brad's victim was possibly more vulnerable than Susan's - Brad stole from a known individual while Susan victimised an organisation. I thought this factor balanced crime seriousness somewhat, so the coded crime elements suggested that crimes of similar seriousness were committed. 248 After looking at the crime details it was found that in addition to her theft conviction, Susan was also convicted for possession of cannabis. In contrast, Brad was only convicted on one count of theft. Susan stole money from her place of work and Brad from his flatmate, so in both cases a position of trust was abused. However, the seriousness of Susan's theft was aggravated further because she also abused a position of power. She had used her position as a bank worker to steal money. Brad and Susan were both uncooperative with police but property taken by Brad, unlike that taken by Susan, was recovered. Balancing larger monetary amounts in Brad's case against un-recovered property in Susan's, her additional conviction and abuse of power led me to conclude Susan's current offence was marginally more serious than Brad's. Susan was employed as a cashier in a bank. She wrote down the balance of her cash draw at the end of the day had the total checked and signed on a balance sheet by her supervisor. Susan then started to place all the money in the bank safe. As she did so she placed $500 in her pocket. The safe was then locked. The following day, the money were again checked and found to be deficient by $500. Police were advised and they executed a search warrant at Susan's house. During the search Susan voluntarily gave the police four grams of Cannabis plant. The stolen money was not located. When interviewed Susan refused to answer any questions. (Susan - Theft, Extracted from Police Summary of Facts) A stereo belonging to Brad's flatmate (Will) was taken from their home address. Brad moved from this flat into another flat with a mutual friend of both Brad's and Will's. Brad's 'new flatmate questioned him about the stereo and Brad admitted taking it from Will at his prior flat. Brad's new flatmate contacted a relative who came to the address and removed the stereo. Police were called and Brad was spoken to in relation to the theft of the stereo. Brad denied the facts and refused to speak to police. (Brad - Theft, Extracted from Police Summary of Facts) Neither offender had a particularly long list of prior criminal convictions. Susan had three priors for property offending and Brad had two. Susan's last conviction was less than a year prior to her current theft conviction. In comparison, Brad's last conviction occurred when he was 18 years old, over five years before. Neither offender had previously breached bail or served imprisonment terms. Susan's criminal history, like her current offence, was subsequently judged as marginally more serious than Brad's. Despite this Brad's remand outcome (remanded on bail with special conditions), was worse than Susan's (remanded on bail with standard conditions) so disparity was concluded in this case. This pair of offenders also received different sentences: Brad was sentenced to 100 hours community service; Susan received $500 reparation, $75 court costs and a $600 fine (see Tables 70 & 71 in Chapter Six). I found it difficult to identify sentence severity in this case. A community service sentence and fine are both at the lower end of the sentencing severity scale and require quite different things from an offender - the loss of money, as opposed to time. However, in some cases fines are repaid over months or years and this was probable in the case of Susan, who was unemployed as a result of her offending. Although not the same, I decided that the imposition caused by these 249 two sentences was similar so disparity was concluded because different criminality had resulted in like sentencing outcomes. 3.19 Vicki and Jeff, Class 3 Property Offence, Fraud Vicki (40 years, European) and Jeff (44 years, European) were both fIrst time offenders, they acted alone and had committed frauds against organisations. There were differences in the coded elements of these cases. Vicki was convicted on two counts compared to Jeff's four counts, but her crime was more serious because she netted over $50,000 while Jeff only received $8,800. Compared to Jeff, Vicki was also uncooperative throughout the criminal justice process, she refused to talk to police and entered a plea of not guilty at trial. In comparison, Jeff co-operated with police, entered a plea of guilty and showed remorse. Vicki's criminality was subsequently judged to be more serious than Jeff's harsher judicial outcomes were expected. Vicki suffered an injury to her back while employed as a meat packer in a supermarket. As a result of the injuries she was supposedly unable to work and was granted compensatory earnings from the ACC. Recipients of earnings related to compensation are required to advice the ACC if their ability to work changes and if they undertake paid or unpaid employment. They are also required to provide medical certificates which verify their work status. Vicki filed medical certificates with the ACC which resulted in the continued payment of her compensation and her work status varied from 'unfit for any type of work' to 'fit for selected work' ... However, ACC established that Vicki was employed as a meat packer in two different establishments while certified as 'unfit for any type of work'. Vicki did not advice ACC of that work for 8 months and when she did Vicki understated her earnings and the hours she worked. She regularly claimed she was working 3 hours per week when hours were as high as 30 hours per week. She was further granted home help assistance and was entitled to claim up to $260 per week for that assistance. Recipients of home help are required to pay for their home help, complete a claim detailing the duties and hours worked by the home helper and have the home helper sign the claim form, acknowledging they are being paid the amount claimed. The form is then submitted to the ACC for payment. Vicki presented 89 claims over a 22 month period. Vicki employed the house keeping services of an elderly woman and told her she would pay her $60 per week to undertake house work. The women undertook 2-3 hours of general housework per day 3-4 days per week. Vicki would get the women to sign ACC claim forms ahead of time and Vicki did not complete the amount claimed before having the woman sign the claim forms. When spoken to by police regarding the home help claims, Vicki stated that she had paid for the elderly woman's groceries and doctors bills. She then rung the woman to ask her to support her story with police but the woman refused. Vicki refused to discuss the other claims with police. (Vicki • Fraud, Extracted from Police Summary of Facts, Trial Transcripts and Judge's Sentencing Remarks) Jeff was a insurance salesman. During the course of selling retirement plans to clients he submitted bogus plans, making initial payments, and claimed commission on them. When spoken to by police Jeff stated At the he was experiencing financial difficulties and used the bogus plans to receive commission. Jeff expressed remorse (Jeff - Fraud, Extracted from Trial Transcripts) Vicki's sentence was harsher than Jeff's which is not surprising given the more serious n<iture of her offending. Vicki received nine months imprisonment suspended for two years, five months periodic detention plus $26,000 reparation and Jeff was sentenced to 60 hours community service (see Table 71 in Chapter Six). However, Jeff's remand outcome was more severe than Vicki's. He received bail with special conditions while Vicki was given bail with 250 standard conditions. Thus, disparity was concluded for in remand outcomes (see Table 70 in Chapter Six).