View this issue as a PDF: 
CrimHighlightsV9N6.pdf
This issue of Criminological Highlights addresses the following questions:
- Why are courts designed the way they are?
- How much of a problem is ‘fear of crime’?
- Why are many community supervision programs ineffective?
- How can community supervision programs be made effective?
- Are ‘risk’ scales useful in assessing youths for recidivism?
- Is the neighbourhood in which a very young child grows up important?
- What is the impact of laws requiring arrests in domestic violence cases?
- Did the legalizing of abortion in the U.S. in 1973 contribute to the reduction in crime in the 1990s?
Courtrooms are designed in a fashion that has (purposefully?) led to the demise of the notion of the ‘public’ trial.
Perhaps because of lawyers’ “obsession with the word” (p. 384) there has been little research on the internal space of the courtroom. This paper argues that the configuration of the criminal courts, including such matters as the nature and height of various barriers, reflects a particular view of the role of the various participants. More specifically, this paper suggests that ‘the public’ has been marginalized by the architecture of the courtroom.
Certain symbols are simple and their meaning is unambiguous. “When a royal coat of arms is placed behind a judge’s chair it makes clear that the full authority of the state and legitimate force is behind the judge” (p. 385). Such placement is not accidental: In England, for example, there is an 813 page guide on court standards and design that imposes a detailed template on designers of new courts. Less obvious than the placement of the coat of arms is the manner in which the space for the public has become more peripheral and contained over time. Indeed it is argued that as the role of the press has increased over the years, the role of the public has been diminished. For example, while the author of this paper was sometimes questioned about taking notes, she never noticed members of the press being required to explain their note-taking.
The design of courts suggests that courts are more concerned with the visibility of the spectators than they are with the visibility of the proceedings by the public. One exception is that “spectators are expected to have a clear view of the judge but are destined to get no more than a ‘general view’ of the proceedings” (p. 396). Indeed, English courts are designed so as to minimize the ability of the public to have direct eye contact with jurors, just as they are designed so as to make lawyers and accused almost unidentifiable. Courts are also designed to prevent the public from seeing the defendant while seated (p. 396). When electronic screens are used to display evidence, they are often placed in a way that makes it impossible for the public to view the evidence. In addition, it would appear that courthouses are constructed on the basis of fear of the public: the English guide to court architecture includes separate ‘zones’ for various groups, most of which are to restrict the accessibility of the public. “The sophisticated forms of segregation and surveillance employed allow things to be arranged in such a way that the exercise of power is not added on from the outside but is subtly present in ways which increase its efficiency and transform spectators into docile bodies” (p. 399).
Conclusion. “Since the only person a member of the public is sure to have a clear view of is the judge, it would seem to be the case that the observation of justice is now limited to observation of the adjudicator rather than evaluation of evidence and the weight which should be afforded it. It is process rather than substantive argument that the public is encouraged to observe” (p. 396). The author argues that “the use of space within the courtroom tells us much about the ideologies underpinning judicial process and power dynamics in the trial…. Perhaps most significantly it helps members of the judiciary to maintain control over who, and what, is likely to be heard” (p. 398).
Reference: Mulcahy, Linda (2007). Architects of Justice: The Politics of Courtroom Design. Social & Legal Studies, 16(3), 383-403.
Is ‘fear of crime’ really a pressing problem for most people in western countries?
The supervision of offenders on probation or parole has been an important part of corrections’ responsibility for over a century. In both Canada and the United States, the portion of the total correctional population that is being supervised in the community far outnumbers the number in prison.. Yet a summary of 15 studies published in the past 30 years would suggests that, overall, community supervision per se or the amount of supervision has very little impact overall and perhaps no impact in reducing violent recidivism. This paper attempts to understand why community supervision is so ineffective, and by implication, what can be done to increase its effectiveness.
This paper starts from the principle that the first necessary step in creating effective community supervision is to identify those offenders being supervised in the community who have a reasonably high likelihood of reoffending in the first place. There is no point in focusing resources on those offenders not in need of intervention. The next step is to identify the goals of intervention – what the offender needs to change in order to reduce the likelihood of reoffending. Finally, effective interventions have to be provided.
In order to understand why probation services may not be effectively employed in ordinary probation supervision, a case study was carried out in Manitoba (Canada). Probation files were examined and audio recordings of probation officer-client meetings were made. The goal was simple: do probation officers (in this jurisdiction) follow what is seen as good correctional practice? There is no reason to believe that Manitoba probation officers were different in any important way from officers in other Canadian jurisdictions.
Clients were seen during the first three months of probation an average of 4.3 times. There was a very small relationship (r=0.22) for adults, but not for youth, between the ‘risk’ level of the client and the number of meetings with the probation officer. A risk assessment tool was generally administered, but for only 40% of the identified needs was there a corresponding intervention plan in place for the offender. The likelihood of an intervention plan being formulated appeared to vary with the need. For example, 80% of those identified as having a substance abuse problem had an action plan noted in their file. On the other hand, 40% of the adult offenders had employment problems noted, but only 10% of these cases had a verifiable action plan designed to address this problem. But in addition, “Analysis of the audiotapes showed that identified criminogenic needs were not discussed in the majority of cases” (p. 267).
Conclusion. Since the data suggest that ‘ordinary’ community supervision has little impact on re-offending, is the only justification for probation the cost saving in comparison with incarceration? This paper would suggest that before coming to that conclusion, one needs to implement, systematically, what is known about reducing offending. “It is clear that probation officers can learn to do more and to do it better. [It has been] demonstrated that training in [various techniques] can make a difference and the beneficiaries of such training efforts will be the staff, the offenders and the community” (p. 268). For an example of an effective program, see Criminological Highlights V9N6#4).
Reference: Bonta, James, Tanya Rugge, Terri-Lynne Scott, Guy Bourgon, and Annie K. Yessine. (2008). Exploring the Black Box of Community Supervision. Journal of Offender Rehabilitation, 47(3), 248-270.
Community supervision programs are less effective than they could be in large part because probation officers do not adhere to basic principles of effective supervision.
The supervision of offenders on probation or parole has been an important part of corrections’ responsibility for over a century. In both Canada and the United States, the portion of the total correctional population that is being supervised in the community far outnumbers the number in prison.. Yet a summary of 15 studies published in the past 30 years would suggests that, overall, community supervision per se or the amount of supervision has very little impact overall and perhaps no impact in reducing violent recidivism. This paper attempts to understand why community supervision is so ineffective, and by implication, what can be done to increase its effectiveness.
This paper starts from the principle that the first necessary step in creating effective community supervision is to identify those offenders being supervised in the community who have a reasonably high likelihood of reoffending in the first place. There is no point in focusing resources on those offenders not in need of intervention. The next step is to identify the goals of intervention – what the offender needs to change in order to reduce the likelihood of reoffending. Finally, effective interventions have to be provided.
In order to understand why probation services may not be effectively employed in ordinary probation supervision, a case study was carried out in Manitoba (Canada). Probation files were examined and audio recordings of probation officer-client meetings were made. The goal was simple: do probation officers (in this jurisdiction) follow what is seen as good correctional practice? There is no reason to believe that Manitoba probation officers were different in any important way from officers in other Canadian jurisdictions.
Clients were seen during the first three months of probation an average of 4.3 times. There was a very small relationship (r=0.22) for adults, but not for youth, between the ‘risk’ level of the client and the number of meetings with the probation officer. A risk assessment tool was generally administered, but for only 40% of the identified needs was there a corresponding intervention plan in place for the offender. The likelihood of an intervention plan being formulated appeared to vary with the need. For example, 80% of those identified as having a substance abuse problem had an action plan noted in their file. On the other hand, 40% of the adult offenders had employment problems noted, but only 10% of these cases had a verifiable action plan designed to address this problem. But in addition, “Analysis of the audiotapes showed that identified criminogenic needs were not discussed in the majority of cases” (p. 267).
Conclusion. Since the data suggest that ‘ordinary’ community supervision has little impact on re-offending, is the only justification for probation the cost saving in comparison with incarceration? This paper would suggest that before coming to that conclusion, one needs to implement, systematically, what is known about reducing offending. “It is clear that probation officers can learn to do more and to do it better. [It has been] demonstrated that training in [various techniques] can make a difference and the beneficiaries of such training efforts will be the staff, the offenders and the community” (p. 268). For an example of an effective program, see Criminological Highlights V9N6#4).
Reference: Bonta, James, Tanya Rugge, Terri-Lynne Scott, Guy Bourgon, and Annie K. Yessine. (2008). Exploring the Black Box of Community Supervision. Journal of Offender Rehabilitation, 47(3), 248-270.
There are ways of supervising offenders in the community that can be more effective than ordinary community supervision.
Community supervision – both probation and parole – is often seen as consisting of an intertwining of law enforcement and social work modes of dealing with offenders. Large numbers of people in many countries are on some form of community supervision. In the U.S. there are about 6 million offenders on community supervision, about three times the number of people incarcerated in prisons and jails. Hence there is a need to determine what strategies are most effective in working with offenders in the community. This study examined the difference in outcomes between a “nail ’em and jail ’em” (law enforcement model) and an approach which first identified criminogenic needs of offenders, developed case plans that responded to those needs, provided services that used social learning or cognitive behavioural interventions, and provided an appropriate learning environment for the offender in which the offender could learn prosocial behaviours and successfully complete supervision. In sum, it tested whether this ‘Proactive Community Supervision’ (‘PCS’) model worked more effectively than a traditional model.
The PCS approach required special (continuing) training for staff in, among other things, how to relate to the offenders they were supervising. The purpose of offender-supervisor meetings was to share information and to assess, refine, and restate program goals. Performance measures were developed for staff and offenders, and attempts were made to change the organizations themselves to insure that they were supportive of the goals of PCS.
A group of 274 PCS clients in four sites in the (U.S.) state of Maryland were compared to a matched set of controls receiving ordinary probation and parole supervision in four other sites. Prior to the implementation of the PCS program, offenders in the PCS and control sites had been arrested at the same rates. The PCS program was implemented reasonably effectively for the majority, but by no means all, of the clients.
Looking at the groups as a whole, those clients who were assigned to receive PCS supervision (whether or not they actually received the full treatment) were less likely to be re-arrested (30%) than were the control clients (42%). There were no significant differences in the number of technical violations (35% for the PCS group; 40% for the controls) which is somewhat surprising given that the PCS offenders had more contact with their supervisors.
Conclusion. This example of a new generation of community supervision programs attempts to provide supervision staff with “tools of motivational enhancement, social learning environments, and targeted emphasis on core criminogenic needs” (p. 294), and moves away from a strict “accountability” (or enforcement) model. The fact that it was successful in reducing re-arrests does not mean that any similar ‘treatment’ oriented program will also be effective. This was an integrated program in which, for example, treatment based interventions were focused on those moderate and high risk offenders most likely to benefit from them. It appears to demonstrate that it is not ‘workload’ per se or the amount of attention given to the offender that is important. Instead, it appears to be the targeted nature of the community based interventions delivered in a “correctional milieu where offender change is supported” (p. 297) that had positive impacts on offenders.
Reference: Taxman, Faye S. (2008). No Illusions: Offender and Organizational Change in Maryland’s Proactive Community Supervision Efforts. Criminology and Public Policy, 7 (2), 275-302.
A commonly used ‘risk’ scale for youths is shown to predict recidivism for probationers but only for some youths and, for them, at a very low rate of accuracy.
Identifying those young offenders who are likely to re-offend may be useful in part because the practice can, in theory, identify those youths who are in need of rehabilitative programs that address aspects of the youth’s life related to offending. Generally speaking, formal risk assessments have been shown to be more accurate than subjective assessments of future risk. This paper examines the predictive validity of the Youth Level of Service Case Management Inventory (YLS/CMI) which is “one of the most widely used risk assessment measures for youth” (p. 477).
All of the youths (n=328) in a Midwestern U.S. city who received sentences of probation for at least a year were assessed using this risk assessment tool. The main measure of recidivism was the presence of a new criminal charge during that year. The YLS/CMI assesses 8 domains (e.g., criminal history, family circumstances, education/ employment, substance use/ abuse, personality/ behaviour) on the basis of the youth’s responses to 42 interview questions.
In this sample, 26% of the youths re-offended in the 12 month period. Dividing the youths into three groups, 11% of the low risk, 26% of the moderate risk, and 39% of the high risk youths reoffended (differences that were statistically significant). On the surface, then, these results might be seen as encouraging, but if the purpose was to identify the ‘high risk’ offender, these results need to be examined more carefully. In this study, 79 of the 328 youths in the study (24%) were identified as being ‘high risk.’ Even among these 79 ‘high risk’ youths – youths who had already been involved with the youth justice system and who were assessed as being high risk – the measure correctly identified only 31 of them. Said differently, 48 of these 79 youths were assessed as being high risk, but did not re-offend. The fact that something is ‘statistically significant’ does not, obviously, guarantee that it is highly accurate.
Other findings also limit the usefulness of the measure. There was a significant correlation between the YLS/CMI score and the number of charges for white youths. However, for girls the YLS/CMI score was not a significant predictor of recidivism. Nor was it a significant predictor of recidivism for youths identifying themselves as African Americans or of Hispanic/ Latino background.
Conclusion. This study serves as a reminder of the problems we have in predicting recidivism for youths. “Cumulative YLS/CMI scores explained a relatively small amount of the recidivism rate between offenders” (p. 482). In other words, although there was a significant relationship between the ‘risk’ scores and actual re-offending, the size of the relationship was so small that it may not be of practical use. Furthermore, the lack of a significant relationship for girls, African Americans and youths of Hispanic/ Latino background lends support to the conclusion that such measures may be of very limited practical use in a youth justice system.
Reference: Onifade, Eyitayo, William Davidson, Christina Campbell, Garrett Turke, Jill Malinowski, and Kimberly Turner (2008). Predicting Recidivism in Probationers with the Youth Level of Service Case Management Inventory (YLS/CMI). Criminal Justice and Behavior, 35(4), 474-483.
Growing up in severely disadvantaged neighbourhoods is associated with behaviour problems for boys just as they start school.
It is well established that children who grow up in poor families are at heightened risk for behaviour problems and involvement in crime. But in addition to the characteristics of their own families, there are reasons to believe that young children in poor neighbourhoods may be especially at risk, since these neighbourhoods may be more likely to be characterized by conditions (e.g., large numbers of unsupervised youths) associated with offending. This study looks at the impact of neighbourhood disadvantage on the behaviour of boys during early childhood.
Mothers who were participants in a monetary and nutritional support program for children in low-income families were recruited at a time when they had a boy who was about 1-1.5 years old. The mothers were interviewed four times, the last time when this child was about 6 years old. The level of economic disadvantage for each family’s neighbourhood was estimated from census data. An index of neighbourhood disadvantage was created from a combination of economic characteristics as well as the proportion of single mother households and households headed by people with relatively low levels of education. Behaviour problems were assessed by questions such as whether the boy “gets in many fights,” “physically attacks people,” or “has temper tantrums or a hot temper.”
Obviously, mothers and children who, themselves, were disadvantaged tended to live in disadvantaged neighbourhoods. And boys who showed behaviour problems when they were very young were more likely to show behaviour problems when they were six. However, above and beyond these factors and various other controls, coming from a relatively disadvantaged neighbourhood was associated with high levels of behaviour problems for the six year-old boys. It also appeared that the effect of neighbourhood disadvantage was not linear. As neighbourhood disadvantage increased, there was no impact on behaviour problems until a threshold was reached. Above that point, behaviour problems were considerably higher, but additional increases in neighbourhood disadvantage had no impact. Interestingly, however, neighbourhood disadvantage was not associated with overt behaviour problems reported by the mothers prior to age 6. Instead, these problems emerged around the time of the transition to school.
Conclusion. The fact that relatively severe neighbourhood disadvantage affected boys’ behaviour at age six, above and beyond characteristics of the youth and the family, suggests that focusing resources on those living in especially disadvantaged neighbourhoods can have a multiplicative impact on boys as they move from home to school. Those receiving the benefits of an economic support program will be helped as will those who live in a neighbourhood that, overall, is no longer severely disadvantaged. If society were interested in reducing school violence or aggression, it might want to focus resources on those severely disadvantaged neighbourhoods from which its problem children are likely to come.
Reference: Winslow, Emily B. and Daniel S. Shaw (2007). Impact of Neighbourhood Disadvantage on Overt Behaviour Problems During Early Childhood. Aggressive Behavior, 33, 207-219.
Statutes that require or advise police to arrest suspects in domestic violence incidents increase the likelihood of arrest not only in those cases but also in incidents with other victim-offender relationships.
Over the past 30 years, there have been changes in the laws governing the arrest of suspects in domestic violence incidents in many jurisdictions. For example, in the U.S., some states allow police the same discretion that is allowed in any other criminal case, while others have laws that instruct police that they must arrest. Other states instruct police that arrest is the preferred outcome. Concern has been expressed about possible ‘dual arrest’ outcomes of such laws – in which both partners are arrested when there is an allegation of violence from both . The rate of ‘dual arrest’ outcomes has been reported to be as high as 23% in some studies and as low 5% in others. This study examines the impact of mandatory and preferred arrest laws on arrest not only in domestic violence cases, but also in assaults involving other victim-offender relationships.
The data that were examined were from police ‘incident reports’ in 8 states with mandatory arrest statutes, 4 states with preferred arrest laws, and 7 states with ordinary ‘discretionary’ laws. Four types of violent incidents were examined: those in which the violence occurred between intimate partners, those which involved other domestic relationships (e.g., parent child), those in which the relationship was non-domestic but the partners were known to each other (e.g., acquaintances, friends, neighbours), and strangers. In general, pooling across all states and police departments, arrest was more common in intimate partner assaults (50%) than in the other three groups (Other domestic: 45%; Non-Domestic, known: 29%; Stranger: 35%)
The findings suggest that the highest rate of arrest for domestic assaults was to be found in the incidents that took place in a ‘preferred arrest’ location. But arrest was also considerably more likely in the ‘mandatory arrest’ locations than in the states in which the decision was left to the discretion of the police officer. The effect of mandatory and preferred arrest laws for domestic violence appears to have had a ‘spillover’ effect on other domestic, acquaintance, and stranger cases. In locations in which arrest was the preferred or ‘mandatory’ outcome for domestic violence cases, arrest was also more likely in other types of violence.
Dual arrests were rare. Only 1.9% of intimate partner incidents resulted in both partners being arrested, though this rate was slightly higher than in other domestic incidents (1.5%), acquaintance (1%) and stranger incidents (0.4%). There was, however, a good deal of variation across and within states. Mandatory arrest policies increased the likelihood of dual arrests compared to locations in which complete discretion was with the police officers. Mandatory arrest laws also, however, increased the likelihood of arrest for other domestic cases and cases involving acquaintances. Preferred arrest policies, however, did not increase the likelihood of dual arrest.
Conclusion. Mandatory and preferred arrest policies clearly have an impact. “Both higher overall arrest rates and higher dual arrest rates are associated with mandatory domestic violence arrest laws” (p. 297). However, preferred arrest laws only had an impact on arrest of the primary offender and not on the rate of dual arrests. None of these findings, however, relate to another important issue: victim safety and reduced re-offending.
Reference: Hirschel, David, Eve Buzawa, April Pattavina and Don Faggiani (2008). Domestic Violence and Mandatory Arrest Laws: To What Extent Do They Influence Police Arrest Decisions? Journal of Criminal Law & Criminology, 98 (1), 255-298.
The suggestion that the 1973 U.S. Supreme Court decision legalizing abortion was responsible for part of the crime drop in the U.S. in the 1990s is almost certainly wrong.
A 2001 study by Donohue and Levitt – popularized by Levitt in his book Freakonomics – suggests that the legalization of abortion in the U.S. in 1973 contributed to the 1990s crime reduction. They suggested that those women who are most ‘at risk’ of having children who are likely to offend – the poor, the young, the unmarried – are, through the availability of abortion, likely to avoid giving birth to future offenders.
One could infer from the “abortion hypothesis” that there should be a decline in the birthrate of so- called ‘at risk’ women after 1973 if the legalizing of abortion had a special effect on them. This study looks at the number of births to 15-19 year olds, the birth rate of unmarried women, and the percent of all births to unmarried women for the period 1940 to 2003. The Donohue-Levitt hypothesis would suggest that each of these should have declined immediately or soon after the 1973 U.S. Supreme Court decision in Roe v. Wade.
The data do not support the hypothesized mechanism for the crime decline. The number of births to 15-19 year olds peaked in the mid-to-late 1960s and was clearly on the decline prior to 1973. The birth rate for unmarried women increased more or less steadily from 1940 until the mid-1980s, though for the ten year period starting around 1968 the birth rate for unmarried women was fairly constant. Furthermore the percent of births to unmarried women increased steadily from the early 1950s until 2004. Hence there is no evidence supporting the hypothesized mechanism suggested by Donohue and Levitt.
Demographic changes in a country have not generally been shown to have much effect on crime (e.g., Criminological Highlights, V2N6#7; V5N4#4). The findings in this paper are similar to those reported by Franklin Zimring (2007) in his book The Great American Crime Decline. He points out that what actually occurred - increases in the number of children born to ‘at risk’ mothers - would be expected to increase, not decrease, crime. Indeed, as Zimring notes in his 2005 book American Juvenile Justice, three prominent criminologists in the early 1990s predicted that demographic changes would create thousands more muggers and killers (James Q. Wilson), hundreds of thousands more ‘super-predators’ (John Dilulio) and a ‘blood bath’ (James A. Fox). In reality, none of this happened: crime went down, and the number of ‘at risk’ youth also went up. Zimring (2007) also presents data from other countries, including Canada, that are not consistent with the ‘abortion effect’ hypothesis. It would appear that demographically-based hypotheses such as this one do not adequately explain actual changes in crime rates.
Conclusion. “The invalidation of the [American] anti-abortion laws had no discernable impact on the fertility levels of women at risk for giving birth to felonious children” (p. 144). Various models of possible impact of the court decision were assessed and none of them showed the hypothesized effect. Foote and Goetz in a paper first circulated in 2005 challenge the Donohue-Levitt findings on a different dimension. They demonstrate that Donohue and Levitt’s published findings did not contain the controls that they say they used. When these controls were included, the ‘abortion effect’ first presented by Donohue and Levitt effectively disappeared. The data presented in the current paper, then, support the conclusion that others have come to: liberalized abortion laws had little if any effect on crime.
Reference: Chamblin, Mitchell B., Andrew J. Myer, Beth A. Sanders, and John K. Cochran (2008). Abortion as Crime Control: A Cautionary Tale. Criminal Justice Policy Review, 19(2), 135-152.
Publication Type
- Regular Issue