Article 1

The language used to describe people returning to the community from prison is important if one wants to encourage public support for their reintegration into the community.

For people being released from prison into the community, there are sometimes legal restrictions placed on their ability to be full citizens. In addition, there are a variety of ways in which those transitioning from prison to the community find themselves stigmatized in the community. This paper examines the impact of the terms used to describe ‘returning citizens’ on the public’s level of support for reintegration measures.

Different terms are sometimes used to identify those moving from prison into the community. Some of this language focuses on ‘crime first’ terminology (e.g., “ex-convict”) and some focuses more on the person (e.g., “person who was formerly incarcerated” or “returning citizen”). In this study, respondents to an online survey were shown a picture of someone described as “struggling with homelessness” who was Black or White, or their race was obscured because they were not facing the camera. They also were described in one of three ways: (1) as an ex-convict, (2) as a person who was formerly incarcerated, or (3) as a returning citizen. For all three conditions, this description was followed by the identical phrase: “someone who has recently been released from jail or prison after serving their sentence and is returning to society.”

Respondents were asked 19 questions related to re-entry programs and policies on such issues as whether they should be allowed to receive government assistance, whether there should be special mental health services for them, and whether there should be special employment programs and educational programs for them. Because respondents were randomly assigned to one of the “language” and race conditions, we can assume that the different groups of respondents were equivalent.

More favourable views toward the person and toward policies that would help those returning to the community from prisons were found when the terminology “returning citizen” was used rather than when the person was described as an “ex-convict.” A “person formerly incarcerated” was in the middle, and was seen as significantly more deserving of services and supportive policies compared to those described as “ex-convicts”.  In addition, those who were asked to give their views of a “returning citizen” were significantly more likely to be willing to have contact with them (e.g., by having them live next door). Generally, respondents were more positive about the “returning citizen” than they were about “ex-convicts” even though the only difference was the terminology used to describe them. In other words, people seem to accept the idea that people can peacefully return to the community. But describing them in “crime first” language undercuts support for those returning to the community from prison.

Conclusion: Language is important. The Ontario Provincial Police, who move detained people between prison and courts before their guilt or innocence is decided, do so in vans identified as “Offender Transport” notwithstanding the fact that available data would suggest that each year thousands of those in these vans will not be found guilty. This study suggests that such language undercuts the willingness of ordinary citizens to interact with those who have had contact with the justice system. And it reduces the support of ordinary citizens for providing help and programs to aid in the reintegration of people who have had contact with the justice system. Language matters.

Reference: Jackl, Hilary M. (2023). The Effects of Language on the Stigmatization and Exclusion of Returning Citizens: Results from a Survey Experiment. Punishment & Society, 25(2), 471-499.

Article 2

Youths typically become increasingly likely to be involved in crime as they transition into high school. However, this increase is smaller if they end up in a physically separate school with a mix of students from both their own middle school and other schools in the school district. Disruption of established patterns can, sometimes, reduce the likelihood of involvement in crime. 

For most youths, crime is a social activity. But for many youths, the transition from Grade 8 to Grade 9 involves disruption of their social lives. This is most obvious when a youth’s high school is both physically separate from their previous school and includes youths from a variety of different previous schools. 

Where the high school is physically separate from the earlier school and the students in it are a mix of familiar and unfamiliar peers, students’ social lives change. This study takes advantage of a simple fact: the nature of the change with the move to high school varies in a manner not in the control of the student. This study followed students from Grade 6 to Grade 12 who lived in small towns and rural communities in two US states. Typically, therefore, youths had no choice as to where they went to high school. In some towns, there was no shift in location. In others there was a shift to a new school, some of which enrolled students from more than one middle school.

Prior to the transition from Grade 8 to Grade 9, students who subsequently transitioned to a new school did not differ in important ways from those who remained in the same school location when they moved to Grade 9. However, “when students experience a school transition from 8^th to 9^th grade, their delinquent behaviour decreases relative to students who do not make this structural change” (p. 53).  Not surprisingly, the change in the delinquency rate of a youth’s friends and the change in the amount of unstructured socializing each predicted the change in delinquency that took place as the youth transitioned into Grade 9. Said differently, it would appear that at least part of the reason that transitioning to a new school reduced a youth’s delinquency is that both their friends and the amount of unstructured socializing were more likely to change when the youth changed schools.  The largest effects of the transition to high school, not surprisingly, occurred in circumstances in which youths transitioned to a physically separate high school that had students from more than one middle school.

Conclusion: In this study, delinquency increased between Grade 8 and Grade 10, but declined slightly thereafter. Hence the transition to high school was happening when delinquency was increasing. But above and beyond individual characteristics of the students (e.g., gender, risk-taking, etc.) the environment in which the youths found themselves – in the same school or a different school – had an impact on their rates of delinquency. These effects suggest that a shift in schools may mean that youths form different social relationships and friendship patterns. These changes may affect friendships as well as academic interests and performance, but they also seem to result in lower levels of delinquency throughout high school.

Reference: Freelin, Brittany N., Cassie McMillan, Diane Felmlee, and D. Wayne Osgood (2023). Changing Contexts: A Quasi-Experiment Examining Adolescent Delinquency and the Transition to High School. Criminology, 61, 40-73.

Article 3

The increased diversity of Canadian police services in recent decades addresses only part of the issue of having a police service that reflects those it serves: Many white male police officers currently believe that their police services have gone too far in their efforts to become more diverse.

Diversity in the makeup of police services has been seen as one mechanism to increase public trust in the police. However, there is concern that the dominant group in Canadian policing – white men – are not content with the changes that have occurred. This paper examines the reactions of Canadian police to the attempts to diversify the organization in which they work.

Canadian police services are predominately white (88%) and male (78%), though obviously this varies considerably across police services. However, “organizational studies confirm that numerically increasing diversity without attention to cultural and structural change will not achieve acceptance and inclusion” (p. 352).

This study reports the results of 116 interviews and 727 online survey responses of Canadian police officers. It does not claim to have a representative sample, but the size and consistency of the findings are probably sufficient to identify a problem that needs to be addressed.

Many police officers (close to half of the survey respondents) agreed with the statements “I feel my police service has gone too far in its push for diversity” and “My police service promotes people to fill diversity spots, not based on merit.” Agreement with these two statements was, not surprisingly, highest within the group who described themselves as white males. Women and racialized police officers were less likely to agree. However, a non-trivial portion of women and racialized police officers also agreed with statements like these. 

This is not to suggest that all (white male) police officers opposed diversifying their own forces or the leadership of their force: “There were some white men in leadership who were supportive of diverse leadership, but they were aware of the undercurrent of resentment by others around them. As one ‘middle management’ white male suggested, ‘There is a recognition of… the general push to promote more diverse members. But that doesn’t mean that underneath the surface there’s not a major problem…. You won’t hear it in the station outright… but if you go out with those guys and they get drinking and talking, it all comes out and it’s ugly’ ” (p. 360).

Conclusion: “Increasing diversity in policing has become a popular reform narrative to improve public trust, police efficacy and marginalized officers’ workplace experiences. This study [concludes] that numerical representation alone has not, and will not, reform the institution in a timely and meaningful way” (p. 364). It demonstrates that even if reasonable representation of women and racialized minorities in a police service were to be achieved, there is likely to be resistance and hostility from within the existing force concerning the manner in which this representation was achieved and what it means for promotion within the institution. “The deeply embedded, often unconscious, everyday processes of institutionalized racism and sexism are larger than any one officer’s attitudes and behaviours” (p. 364).

Reference: Bikos, Lesley J. (2023). Taking the Temperature: An Intersectional Examination of Diversity Acceptance in Canadian Police Services. British Journal of Criminology, 63, 348-366.

Article 4

Conservative American politicians abandoned their opposition to financial support for victims of crime when it was proposed that the money come from those found guilty of crimes. Could it be that they had more interest in increasing the punitiveness of criminal sanctions than they did in the needs of victims?

This paper suggests that victim compensation in the US is an example of the proposition that support for some policies is “based on the political and moral meanings attached to particular policy financing tools.”  These tools are not neutral. Instead, they hinge “not only on the moral worthiness of who benefits from a policy, but also on the moral unworthiness of who pays” (p. 1159, emphasis in original).

The idea that there should be compensation for victims of crime first became an issue in the US in the 1960s. Traditional restitution didn’t work because offenders might not have been apprehended. In addition, even if they were apprehended, offenders typically had few resources. The original liberal idea was that the state had a responsibility to protect people from crime and had failed; hence compensation made sense. Alternatively, it was seen as a form of social insurance. Conservatives rejected both logics. But later, in the 1980s, when federal conservatives tied the financing of victim compensation entirely to fines paid by those found guilty of crime, compensation fit the conservative slogan “The criminal – not the taxpayer – will pay….” (p. 1160).

For those whose ideology is uncomfortable with benefits to anyone being paid from tax money, offender funding of victim compensation, like “tax expenditures” (where companies benefit from targeted tax reductions rather than direct transfer of money from the government to the company), has huge advantages: they are not seen as another government funded program.

States in the US (starting with high income, high crime, liberal states) created victim compensation programs in the 1960s. Some state laws had provisions where the state could try to recoup their costs from those found guilty of the offence. Others used fines (as a way of suggesting public money was not being spent). Later (in the 1970s) explicit “surcharges” were added so that compensation was seen as coming from criminals. And finally, later, some states charged those serving sentences in the community with fees that were to be used for the compensation of victims. Some states that had started with discretionary payments or surcharges moved to mandatory surcharges. Over time, the source of funds expanded such that payments by those found guilty were uncoupled from any harm that offenders might have done to anyone. [This, with a few exceptions, is the situation in Canada.] In other words, victims generally were to be compensated by making all those found guilty pay even if they had no victim who needed compensation, suggesting that “victim compensation” is more about “offender punishment” than concern about victims.

The Republican dominated US federal government favoured victim compensation in the early 1980s. A plan to divert funds from taxes on the sale of handguns went nowhere. In 1984, Republican politicians decided to describe victim compensation as a choice between “taxing criminals or taxing innocent taxpayers” (p. 1187). But behind this was another issue. In much of the discussion, the beneficiaries were seen as racially and ethnically diverse, but “criminals” were often seen or thought to be from “distinctly racialized social categories” (p. 1190).

Conclusion: In the US, the political meaning of legislation whose purpose was ostensibly to compensate victims of crime was linked to punishing offenders. Conservative federal politicians who had initially opposed victim compensation because they saw it as too expensive and not an appropriate role for government, supported it as soon as it became a way of punishing those who had offended. Hence compensation for victims became just one more way of increasing punishments in the criminal justice system. It logically follows that those found guilty of crimes where there was no victim (e.g., a theft where the property was recovered) should also pay into a fund for victims: They, too, were seen as deserving of (more) punishment.

Reference: Levine, Jeremy R. and Kelly L. Russell (2023). Crime Pays the Victim: Criminal Fines, the State, and Victim Compensation Law, 1964-1984. American Journal of Sociology, 128(4), 1158-1205.

Article 5

A meta-analysis demonstrates that the “classification” instrument Static-99R used by many correctional authorities (including Correctional Services Canada) is less accurate in assessing Indigenous prisoners than it is when used for white prisoners. Shouldn’t this be a source of concern?

There is now sufficient information to conclude that instruments meant to predict future offending or other misbehaviour can easily disadvantage groups such as Indigenous prisoners (see Criminological Highlights 17(2)#1, 17(6)#7, 20(3)#3).  But more generally, if these instruments are relied on for important decisions concerning those being imprisoned, they should have a high level of accuracy. A high level of accuracy means something more than just being better than flipping a coin.

The Static-99R is used by many correctional systems to assess sexual recidivism risk. This paper examines its accuracy for various groups. The fact that scales like this were developed using largely white samples does not necessarily mean that they don’t have predictive value for other groups. On the other hand, given that the social and cultural background of groups like Indigenous Canadians differs from that of white Canadians, careful analysis of the value of an instrument like the Static-99R is important.

There were 10 studies that could be found in which the accuracy of the Static-99R scale for predicting sexual recidivism for Indigenous people could be estimated and 9 studies where the accuracy for white people could be estimated. For neither group was the prediction very accurate, though, in general, the accuracy was higher for whites. One way of presenting the accuracy is a measure referred to as AUC which essentially is an index indicating, using this example, the likelihood that a randomly chosen person who recidivated had a score that was higher than a randomly chosen person who did not recidivate. If the scale was perfectly accurate, the AUC would be 1.0 (all recidivists would have higher scores than all of those who did not recidivate). If it didn’t predict at all then the AUC would be 0.5 (a “50-50” chance that the score of the reoffending person would be higher than that of the non-recidivist). Results from the different studies that made up this review can be combined in different ways, but using one standard approach, the AUC for whites was .682 and for Indigenous people was .635.

Neither of these assessments of the scale is very impressive even though they are “statistically significantly” better than chance.  In addition, Indigenous people scored higher overall on the scale, reflecting, no doubt, differences in their situations in society more generally. This finding suggests that, on average, if the same “cutoff” were to be used for white and Indigenous prisoners, the Indigenous prisoner would almost automatically “look worse.” But in addition, there is another problem this paper mentions, but could not examine. The scales were developed largely using white samples. Hence it is likely that each item on the scale has at least some predictive value for those in comparable white groups. However, for Indigenous prisoners, this may not be the case. That would mean that an Indigenous prisoner who was near an important threshold (e.g., the threshold of being recommended for parole) but who was scored on a question in a particular way, even if that had no predictive value for Indigenous prisoners, might lose out on a favourable recommendation for parole because of a characteristic that says nothing about their likelihood of reoffending.

Conclusion: The fact that predictive scales like this one have, at best, mediocre predictive value for Indigenous prisoners does not answer the question about whether they should be used or not. If the choice is using a mediocre scale or a completely subjective judgement from a correctional employee, it is likely that the scale will be better, or at a minimum, less arbitrary.  The more general question that might come from studies like this one is whether the amount of punishment we inflict on people should be determined by predictions that have been demonstrated to be inadequate.

Reference: Ahmed, Simran, Seung C. Lee and L. Maaike Helmus (2023). Predictive Accuracy of Static-99-R Across Different Racial/Ethnic Groups: A Meta-Analysis. Law and Human Behavior, 47(1), 275-291.

Article 6

Black men in Florida with stereotypically Black names get sentenced more harshly than comparable Black men with first names that are not associated with being Black. This holds even after controlling for legally relevant features of the case.

There are reasons to believe that first names are sometimes used as a way of identifying the strength of a person’s identification with a racial-ethnic group in society. Hence if people believe that Blacks are more likely to be involved in crime, and a stereotypically Black name is seen as an indicator of identification with that group, it may be that people with stereotypically Black first names will be punished more harshly than those with more “White” sounding first names.

This paper started by looking at sentencing records for 298 young Black men sentenced in Florida. Details of their background and offences were available and used as controls. The first names of these men were rated by 79 online American volunteers on how stereotypically Black and White each name was. A separate set of online volunteers rated each name on how likely they thought that “others” would see a person with this name as a violent criminal, as well as how likeable they thought a person with this name would be. Another sample of online volunteers rated an actual picture of each of the Black men who had been sentenced on how much the features of the person who had been sentenced looked like those of an African-American man.

Controlling for all the legally relevant variables, those men being sentenced with first names that were seen as being stereotypically Black (e.g. Jamal) got harsher sentences from the court than those who had White sounding first names (e.g., James). But in addition, when other rated characteristics of the defendant were controlled for (e.g., likeability, perceived social class of the first name, and Afrocentricity of the prisoner’s features), those with stereotypical Black sounding first names still got harsher sentences.

In a second study, an online sample of largely White volunteers were asked to make sentencing recommendations for 10 people.  All of those being “sentenced” were Black. Half of the people “being sentenced” were identified with first names that were stereotypically Black; the others were identified with first names that were stereotypically White. Participants recommended harsher sentences for those who were identified as having stereotypically Black names.

Conclusion: It appears that White Americans are much more likely to associate crime (and in particular violence) with African-Americans who have stereotypically Black names. Ordinary citizens recommend, and judges in Florida impose, harsher sentences for people with stereotypically Black first names than they did for people with White-sounding first names. “This bias was not explained by other qualities of the defendants (such as their physical appearance or criminal record, the specifics of the case… or even other aspects of their names…. Rather [the] results link stereotypical Black names to the representation of a violent criminal” (p. 177-8).

Reference: Kenthirarajah, Dushiyanthini, Nicholas P. Camp, Gregory M. Walton, Aaron C. Kay, and Geoffrey L. Cohen (2023). Does “Jamal” Receive a Harsher Sentence than “James”? First-Name Bias in the Criminal Sentencing of Black Men. Law and Human Behaviour, 47(1), 169-181.

Article 7

When judges oppose a change in the law, they may find ways of ignoring it. Finding ways to monitor, review, and ensure that judges comply with changes in the law may not be easy.

In 2020, the New York state legislature passed laws designed to reduce pretrial detention and also to provide timely access for defence counsel to material relevant to the case being tried that was being held by the prosecutor (discovery). After the law had been passed, the opposition to these new provisions became organized. In this summary of the full study, we will focus primarily on the first issue: the reform of bail laws. The issues related to the discovery law were similar.

The problems addressed by the changes in the bail laws were not new. They had been described in 2013 by the then Chief Judge of the NY Court of Appeals. Changes in the law were introduced and passed during an unusual political period when the Democrats controlled the two legislative bodies and the governorship. The cash bail system had been in place for many years and there was clear evidence that judges were setting cash bail amounts that were unnecessarily high. There was experimental evidence that most accused would show up for trial without having to post bail. But the law as administered was detaining people who didn’t need to be detained and pretrial detention populations were large.

Though there had not been much vocal opposition before the law was changed, immediately after its coming into effect substantial opposition was expressed by the police, prosecutors, and many politicians.  Prior to the change “judges had nearly unfettered power to set bail [amounts]” (p. 133). The changes that were made forbid the use of cash bail for certain minor offences and required judges to use the least restrictive means possible to ensure appearance in court. In other words, these reforms “achieve their objectives through the reduction of judicial discretion” (p. 131-2).

New York judges were not pleased to lose powers they once had, even if it had been shown that these powers were not serving the public interest. Referring to the new law, one judge, who apparently must have missed the classes in law school in which the importance of judges showing respect for the law was taught, described the new law in very simple terms in open court: “The law is stupid.” He then stated that to follow it “went against all common sense and wisdom” (p. 135). Even in some cases in which the prosecutors recommended supervised release, judges set cash bail even though release on cash bail was not allowed.   Some news articles about such cases described the judge who violated the law as being “heroic.” Breaking the law was supported by NYC’s mayor. Judges detained many more people than were recommended by assessment tools.

The reasons for judicial non-compliance with the law are complex. Data on the effect of the law were not made available. Support for the law was not provided by judicial officials. Compliance with the law was not encouraged by senior judicial leaders. One Chief Administrative judge publicly criticized the changes saying that its removal of judicial discretion for certain outcomes in certain cases was “breathtaking” (p. 146). An administrative procedure was established that allowed certain detentions of accused people to take place without a full hearing even though it was not contemplated by the law. Judges had an interest in being seen as tough: Judges in NY “have been removed, have lost elections or have been placed in undesirable positions for being too lenient” (p. 166). “Judges in NY are attuned to the political will for reform” (p. 167) and if the political will is against change, judges may refuse to follow the law.

Conclusion: Obviously the specific reasons for judicial non-compliance with the law are going to vary with the issue and the jurisdiction. However, what is important about the findings described in this paper is that non-compliance by judges can easily occur. “In court cultures like NY, reforms work when judges are either responding to the political will, like legislators, or when they feel that they are insulated from that will enough to be truly independent to do what is fair and right under the law” (p. 167).  Judicial opposition to legal change needs to be considered in many instances where a legislature has passed laws that judges may not like.

Petrigh, Angelo (2023). Judicial Resistance to New York’s 2020 Criminal Legal Reforms. The Journal of Criminal Law & Criminology, 113(1), 109-174.

Article 8

In rural communities, having friends who have been arrested increases the likelihood of being arrested by the police above and beyond standard predictors of criminal involvement.

It is well known that youths often commit crimes in groups, not so much as “criminal gangs” but just as a group of youths who hang out together and may commit crimes together. Given that we know that people who have been in trouble are, generally, seen in a negative light (see Criminological Highlights 20(2)#8), it would not be surprising if the arrest of a youth’s friends had negative impacts on that youth.

If a youth has a close friend who is arrested, that youth may be especially likely to be the subject of surveillance by the police and others. A separate question, however, is whether, after the friendship ends, the increased likelihood of police arrest is maintained. This paper uses data from 28 school districts in small and medium sized towns (up to about 45,000 residents) in Iowa and Pennsylvania in which youths were questioned each year from Grade 6 through Grade 12. 

Each youth, at each point in time, was asked about arrests in the previous year. The study looks only at the youth’s first arrest.  The primary independent variable is whether the youth had a friend arrested in the year immediately prior to the one that was the focus. Youths were asked to name up to two “best friends” and five “close friends” (p 363). Various control factors were part of the study including self-defined reported delinquency by the offender, their friends’ self-reported delinquency, and the youth’s substance use. The study examined these effects only for youths who had never been arrested. Hence a youth arrested in Year N, was dropped from analyses for Year N+1 and thereafter.

Not surprisingly, without any controls, if a youth had friends who had been arrested by the police, the youth was more likely to be arrested. When self-reported delinquency by the youth and their friends, and the youth’s substance use were controlled for, friends’ arrests still predicted the likelihood of the youth being arrested. Similarly, when the various control factors were included, those with friends who had been arrested were still more likely to be arrested.

The effect of once having friends who were arrested by the police on one’s own likelihood of arrest was still maintained if the youth dropped these people as friends. Interestingly, “compared with youth who have not had friends arrested, no significant difference in the likelihood of arrest occurs between youths who did and did not drop their arrested friends” (p. 371). The main results of the study were the same when the study examined separately the impact on likelihood of arrest by one’s “best friends” and “close friends.”

Conclusion: The results, then, were very consistent: “Even when controlling for other [factors] that predict one’s likelihood arrest including delinquency, substance use, demographic characteristics…, a friend’s arrest was still significantly and positively related to the likelihood of arrest the next year” (p 373). The effect of a friend’s arrest continued even if the youth dropped this person as a friend. High rates of policing of youths leading to high rates of arrests of these youths had, therefore, effects not only on these youths, but also on those who saw them as friends. If a youth has friends who have been arrested, the youth is more likely themselves to be arrested.

Reference: Tinney, Erin (2023) The “Stickiness” of Stigma: Guilt by Association After a Friend’s Arrest. Criminology, 61(2), 354-383.

Criminological Highlights is prepared by Anthony Doob, Rosemary Gartner, Olivia Berkovits, Sara Fruchtman, Maria Jung, Tyler King, Audrey Macklin, Roxy Shlapak, Jane Sprott, and Danielle Van Wagner.