View this issue as a PDF: CrimHighlightsV12N1.pdf
This issue of Criminological Highlights addresses the following questions:
- Why do people engage in violence over quite trivial matters?
- What determines whether a parolee will be returned to prison for violating a condition of parole?
- How can we reduce the number of wrongful convictions?
- Do ordinary members of the public believe strongly that offenders should be incarcerated?
- How was California’s imprisonment rate dramatically reduced under Republican Governor Ronald Reagan?
- Why do judges and juries sometimes disagree on whether an accused should be found guilty?
- Are members of minority groups who come into contact with the police more likely to be arrested than white suspects?
- Are serious delinquents likely to persist in offending after being placed in custody?
Fights that appear to be over “nothing” or fights over “trivial matters” are, in fact, fights over something: social rank.
The term “senseless violence” is interesting because it would appear to suggest that certain types of violence – typically violence in which there isn’t an obvious object that is the source of conflict – is irrational or inexplicable. “In effect, the issue of contention [in this type of violence] is so trifling or banal that it seems both noninstrumental and incapable of sparking the level of rage necessary to propel an actor toward the expressive use of violence against an adversary” (p. 62). This paper explains this “inexplicable” violence.
It is suggested that “senseless violence” occurs “in social relationships that are equal, or symmetrical in nature” (p. 63). “Violence over trivial issues of contention should be much more likely between [opponents of equal status] than between [opponents of different statuses] because trivial disputes symbolically represent an effort to change the status quo and establish dominance in the relationships” (p. 69).
This study examines accounts of 261 violent interactions described by 142 women who were interviewed while they were serving sentences in an Ontario prison. The women described, in their own words, up to 6 violent incidents that had taken place during the previous three years with people they knew but who were not intimate partners. These detailed accounts were transcribed, and the immediate issue over which violence erupted was identified. A violent altercation was classified as being the result of a substantive issue when the fight involved property, the protection of an individual, the termination of a valued relationship, or sexual rivalry or jealousy. Fights were classified as being the result of ‘trivial’ issues if the precipitating event involved insults, verbal slights, gestures, or other matters that could not be seen as ‘substantively’ important.
The relationship between the combatants was coded as being symmetrical (e.g., siblings, cousins, friends, fellow drug users, neighbours) or asymmetrical (e.g., parent-child, drug dealer – drug purchaser, prostitute-john, bartender-customer). Various controls were taken into account – presence of others at the altercation, age, age difference between the two, race, past arrests, whether the respondent frequented bars or clubs, number of major life stressors and the length of the account that the respondent gave of the incident.
The results were clear: 17% of the violent events involving asymmetrical relationships (people of different statuses) involved ‘trivial’ precipitating factors. In contrast, 45% of the violent events involving symmetrical relationships involved trivial precipitating events. What is notable, however, is that even though certain other factors (e.g., the presence of bystanders) predicted whether the violent altercation involved trivial issues of contention, these other factors were very clearly independent of the impact of whether it was a symmetrical relationship. Said differently, adding controls did not change the size or the nature of the relationship between symmetrical relationships and fights over trivial matters.
Conclusion: “Violence involving trivial or small matters is considerably more common among opponents embedded in identical social roles…. When women engage in violent altercations over ‘nothing,’ they do so largely against opponents in symmetrical social roles” (p. 88). When violence erupts over trivial issues, both parties to the altercation essentially are locked in a battle for social rank” (p. 62). In understanding interpersonal violence, therefore, it is important to go beyond individual characteristics of those involved and look at the relationships between those involved in the altercation.
Reference: Griffiths, Elizabeth, Carolyn Yule, and Rosemary Gartner (2011). Fighting Over Trivial Things: Explaining the Issue of Contention in Violent Altercations. Criminology, 49(1), 61-94.
The likelihood that a parolee will be found to have violated a condition of parole has at least as much to do with the parole enforcement system as it does the person who is being supervised.
In 1999, it was estimated that 42% of the growth in U.S. prison admissions was attributable to those whose parole was suspended or revoked. Returning a parolee to prison obviously can occur as a result of a new offence. But more often than not it is because of the breach of a condition of parole. Hence the decision to return a parolee to prison is one that is at least partially under the control of the parole officer. In Canada, for example, of the 700 unsuccessful day and full parole cases from federal penitentiaries in 2009/10, 77% were revoked for violations of conditions rather than for new offences. This study examines the role of supervision regimes on whether a parolee is deemed to have violated parole.
Traditionally, it is assumed that the violation of conditions of parole – like the committing of an offence – can be adequately understood by looking at characteristics of the offender who is being supervised. This paper, on the other hand, examines not only characteristics of the parolee, but also the nature of the supervisory regime and the characteristics of the officer who is doing the supervision. During 2003 and 2004 in California, of the 254,468 people on parole supervision, 49% were found to have violated their parole. Not surprisingly, personal characteristics did make a difference: those on parole for violent or sex offences were less likely to violate parole than property or drug offenders. Blacks, males, and those released when they were under 30 years old as well as those labelled as mentally ill were more likely to violate parole.
But the expression ‘violating parole’ ignores the fact that supervisory factors also determine whether someone ‘violates’ – or perhaps more properly ‘is found to have violated’ – parole. In California, three distinct types of supervisory regimes could be identified, based on differences in the number of drug tests (none to once per month) and reporting frequency (monthly by mail to a face-to-face meeting every 2 weeks). The caseload of parole officers also varied considerably, though in general caseloads were much higher than policy suggested they should be. Holding constant characteristics of the parolee, those subjected to high levels of supervision were more likely to be found to have violated parole. Caseload had an inconsequential impact on parole violations. But the introduction of a “New Parole Model” in the middle of the study period that mandated the use of drug treatment, electronic monitoring, and a residential community re-entry program as alternatives to returning the parolee to prison appeared to be responsible for a large increase in violations. In addition, it would appear that characteristics of the parole officers also made a difference: Parolees with Black parole officers or parolees assigned to parole officers with more than 3 years of experience were less likely to have their parole terminated. When the intensity of parole supervision is taken into account, the effect of the original offence for which they were imprisoned and then paroled is reduced. “This finding indicates that differences in supervision inflate the risks that offenders with serious, violent, and sexual prior offenses pose (which are already fairly low), and when supervision factors are controlled, the effects of offence history on risk of violation are reduced” (p. 388).
Conclusion: Whether or not a parolee is found to have violated parole has as much to do with the nature of the supervision as it does with who is being supervised. Intensive supervision – often focused on violent or sexual offenders – can increase the likelihood of the parolee being found to have violated parole. This in turn appears to make these categories of offenders appear more likely to violate, thus, in a circular way, justifying even more supervision. There are also local variations in the likelihood of parole suspensions or charges that appear to be independent of offender characteristics. To the extent that parole violations fuel incarceration rates and interfere with controlled re-entry into the community, it is clearly important to understand and develop policies that relate to the effective handling of parole violations.
Reference: Grattet, Ryken, Jeffrey Lin, and Joan Petersilia (2011). Supervision Regimes, Risk, and Official Reactions to Parole Deviance. Criminology, 49(2), 371-399.
We know a fair amount about the immediate causes of wrongful convictions. The question that has not been adequately addressed is why most jurisdictions don’t seem to be doing anything about them.
Much is known about the factors that seem to account for those cases in which the legal system has acknowledged that a convicted person is “factually innocent” (i.e., where someone else committed the crime or a crime did not take place). The first analysis of these cases was published in 1932 and since then the research has demonstrated that the factors leading to wrongful convictions have not changed very much.
The evidence that police can frequently be wrong in their investigations of criminal events is illustrated by the fact that in the U.S., between 1989 and 1996 about 25% of the ‘primary suspects’ identified by law enforcement officials in rape cases and whose DNA was sent to the FBI for testing were subsequently excluded by the results of the DNA test (p. 830). A flawed police investigation does not, however, lead to a wrongful conviction, but this finding does demonstrate that the raw materials for wrongful convictions are easy to find. Though the ‘true’ rate of wrongful convictions is unknowable (see Criminological Highlights 7(5)#3, 9(4)#5), most thoughtful estimates, based largely on cases involving DNA would suggest that about 2-5% of contested convictions for serious cases may involve factually innocent people.
The factors that lead to wrongful convictions are best thought of not as discrete causes, but as errors that might have been, but were not, corrected at some point in the path leading to conviction. The factor most commonly present (perhaps in as many as three quarters of known wrongful conviction cases) is mistaken eyewitness identification. Much is known about the circumstances leading to these errors. False confessions induced by psychologically coercive police interrogation methods appear to be the second most frequent contributor (see Criminological Highlights, 7(4)#7, 11(3)#4). Much is also known about how to minimize these problems. Tunnel vision – the ignoring of facts that do not fit an initial judgment of guilt – can occur at any stage and may be the result, in part, of other factors (e.g., a positive, but incorrect, eyewitness identification). False evidence from informants has, in recent years, surfaced as a common problem as has flawed forensic evidence (e.g., incorrect fingerprint or hair matches). Prosecutorial misconduct (e.g., coaching witnesses on what might be ‘helpful’ evidence and the suppression of evidence) along with inadequate or incompetent defense representation round out the top seven factors.
Obviously these factors may be more important in some cases than in others. Eyewitness errors are more common in cross-racial identifications. In addition, in particularly heinous crimes receiving high media coverage, police and prosecutors may feel pressure to complete their investigations quickly and conclusively, thus making them more susceptible to tunnel vision.
Conclusion. From studies of wrongful convictions, it is clear that a fair amount is known about what factors in a case (e.g., incorrect eyewitness identification, incorrect conclusions from a forensic test) go wrong and lead to wrongful convictions. In some cases – e.g., eyewitness identification procedures – it is known how those procedures can be dramatically improved. The most important barrier may not be knowledge, but the reluctance of key people to change. “It is… the professionals who staff our criminal justice system and the politicians and policymakers who employ them that may require the more significant improvement” (p. 866).
Reference: Gould, Jon B. and Richard A. Leo (2010). One Hundred Years later: Wrongful Convictions After a Century of Research. Journal of Criminal Law & Criminology, 100 (3), 825-868.
The use of custodial sentences for offenders is often justified by the assertion that ‘the public demands it.’ But public support for custodial sentences in many cases may be about as thin as the evidence that custodial sentences deter offenders.
Those responsible for sentencing policy – either as part of sentencing councils as in England & Wales or judges elsewhere – often talk about the need to promote public confidence in the justice system. This assumption is supported by simple surveys that suggest that in many countries (including Canada) the majority of the public responds to simple poll questions by saying that most sentences are too lenient. This study goes beyond these simple surveys to help understand better the circumstances when the public is content to use a sanction other than imprisonment.
A representative sample of 1023 adults in England & Wales read descriptions of one of three different cases: a serious assault, a serious household burglary, or a fraud involving a substantial loss of money. Only about a third of respondents indicated that all such offenders should be imprisoned. Most of the rest of the respondents thought that the decision maker should have discretion as to whether the offender was imprisoned. Respondents were given a list of 13 potentially mitigating factors and were asked, for each factor, whether it justified a more lenient sentence in all, most, some, or no cases. The majority of respondents thought that most factors (e.g., the offender has no criminal record, or the offender was a victim of abuse in childhood) would justify a more lenient sentence in at least some cases. Being a young person (defined as being 18 years old) was the only factor for which a majority thought that it should never result in a more lenient sentence. Clearly respondents wanted personal factors to have some weight in determining the sentence.
In another part of the survey, respondents were told that a judge had decided to impose a prison sentence on an offender (for either an assault or a fraud). They were then given a list of factors (e.g., the offender had no record, the victim did not want the offender to be imprisoned, the offender is caring for young children) and they were asked if the factor justified a community service order instead of prison. The majority of respondents thought that each of 6 mitigating factors would probably or definitely justify the imposition of a community service order instead of imprisonment for the assault. For the fraud, the fact that the offender was young was seen as probably justifying community service instead of prison by only 48% of respondents.
In another part of the survey, respondents had a relatively serious case described to them that would typically have resulted in a prison sentence. Not surprisingly about 4/5 of the respondents chose prison as the preferred alternative (over community service or a fine). However, about half of those who preferred prison found a detailed non-custodial order involving compensating the victim and doing a substantial number of community service hours to be acceptable instead of imprisonment.
The public is pragmatic about criminal penalties: Most respondents thought that the costs of administering sentences should be taken into account when imposing sentences. However, there was more support for considering costs in the case of a social security fraud than there was in the case of an assault.
Conclusion: The public clearly wants many or most mitigating factors to be considered in most cases. “While the public may ‘talk tough’ in response to opinion polls which ask whether sentencing is harsh enough, when considering specific criminal cases and individual circumstances, there is considerable support for mitigating punishments” (p. 194). When details of non-custodial sanctions are made salient to members of the public, they will tolerate them. Members of the British public appear pragmatic: they generally want costs to be considered when sentences are being imposed. It would appear that “members of the public react thoughtfully to questions relating to sentencing – and not simply with reflexive punitiveness” (p. 195). Those policy makers whose approach to sentencing does not go beyond ‘reflexive punitiveness’ may, therefore, not be representing public sentiment.
Reference: Roberts, Julian V. and Mike Hough (2011). Custody or Community? Exploring the boundaries of public punitiveness in England and Wales. Criminology & Criminal Justice, 11(2), 181-197.
California’s 34% reduction in its imprisonment rate between 1967 and 1971 during Ronald Reagan’s first term as governor occurred as a result of the implementation of a number of different policies designed by progressive policy makers with support from a fiscally conservative governor.
In his second inaugural address as governor of California in January 1971, Ronald Reagan proudly reported that during his first 4-year term violent crime in California cities had been reduced and “Our rehabilitation policies and improved parole system are attracting nationwide attention. Fewer parolees are being returned to prison at any time in our history and our prison population is lower than at any time since 1963” (p. 311). This paper examines how California’s imprisonment rate was reduced from 146 prisoners per 100,000 residents in 1968 to 96 in 1972.
It is generally agreed that imprisonment rates reflect policy decisions of governments and have little if any relationship to crime rates. Hence, the reductions in imprisonment rates that have taken place elsewhere (e.g., Finland, The Netherlands, Italy, Poland, France) need to be understood in the specific political and historical contexts in which they occurred.
Reagan took office with a pledge to “put our fiscal house in order… and cut… the cost of government” (p. 312). But even though there were a number of high profile murders during his first term, Reagan supported his officials’ recommendations on corrections policy and was aware of the fact that there were liberal changes that could be made under a conservative governor that would not be supported if they had come from a liberal. As one of his policy advisors noted “Reagan did not have to look like he’s tough” (p. 312).
California during this period was well known for its progressive criminal justice policies. Prior to Reagan’s election win in 1966, the state had set up a system that encouraged counties to place offenders on probation rather than sending them into the state prison system. Reductions in prison sentences from each county were rewarded with money transfers to the county to pay for county probation services. Not surprisingly, probation rates increased, and admissions to state prisons decreased. In addition, however, the rate of paroling inmates also increased. The effect was that fewer people were spending long periods of time in sentenced custody. By increasing the parole release rate, fewer inmates served their full sentences in prison. And once out on parole, fewer parolees had their parole suspended.
There was, then, no single factor responsible for California’s dramatic decrease in imprisonment. Instead decarceration was accomplished through a number of separate policy changes. Although crime and arrests were increasing during this period, admissions to state prisons declined. And, offenders were serving less of their prison sentences in prison.
Conclusion: California’s reduction of imprisonment did not happen by chance. The conservative Governor Reagan supported and applauded the decline in imprisonment in part because it reduced costs. But in addition, Reagan, the conservative, appeared to have supported the policies developed by his (correctionally liberal) experts. In the end, the reduction was seen as “a result of the work of professionals and real community support and good attitude in the community” (p. 301). The experts at the time realized that prior to Reagan’s ascent to power, at least a quarter of prison admissions at the time could be kept in the community instead. By focusing on good government, rather than the politics of hysteria, Reagan and his professional corrections service accomplished a goal that was applauded as being both fiscally and socially good policy.
Reference: Gartner, Rosemary, Anthony N. Doob, and Franklin E. Zimring (2011). The Past as Prologue? Decarceration in California Then and Now. Criminology & Public Policy, 10(2), 287-325.
Cases in which juries acquit but judges indicate that they would have convicted are likely to be described by either the judge or the jury as ‘close’ cases. But in addition, they are likely to be cases in which the defendant does not have a criminal record.
In the book The American Jury, authors Harry Kalven and Hans Zeisel presented data suggesting that cases in which the evidence was close, jurors were liberated from the facts and tended to decide cases according to their sentiments toward the defendant and their assessment of whether the law was being applied fairly in the case they were hearing. In the Kalven & Zeisel study, 88% of the disagreements between a judge and a jury involved the judge convicting and the jury acquitting.
This paper re-examines judge-jury disagreements in a study of 289 cases in four U.S. locations. Judges, jurors, and defense and prosecution lawyers answered questions about each of the cases. In this study, 73% of the judge-jury disagreements on verdicts involved jury acquittals of cases in which judges indicated that they would have convicted. These disagreements – jury acquits/judge would convict – occurred largely in cases in which either the judge or the jury (or both) thought that the case was ‘close,’ and in cases in which the accused person had no criminal history.
Jury sympathy toward the defendant does not appear to have the consistent impact on judge-jury disagreements that would be expected from the Kalven & Zeisel study. When judges view the evidence in the case as ‘close’, jury sentiment about the law – that applying the law strictly in this case would not be fair - predicted different outcomes between judges and juries. There was no such effect in cases in which the jury saw the case as close. This result “provides little support for a theory that suggests evidentiary uncertainty triggers the jury’s retreat to sentiment. Ultimately, the strongest and most consistent predictor of jury acquittal in opposition to a judge’s vote for conviction is a defendant without a criminal record. [In these data], the evidentiary factor of a criminal history has a much stronger influence than jury sentiment in predicting judge-jury disagreements” (p. 1583).
Conclusion: These data suggest that “in the face of evidentiary uncertainty, juries do not appear to retreat to sentiment; rather, a more complex process of weighting the value of evidence, including the criminal record of the defendant appears to explain [judge-jury] disagreements” (p. 1584). Clearly the justification for jury trials is that, in some cases, juries will arrive at different verdicts from judges. In this study, the judge and the jury agreed on 77% of the verdicts. Cases in which they disagreed on verdicts tended to be ones in which the judge or the jury thought the cases were close. But even in the ‘close’ cases, the judge and jury tended to agree on the outcome. The fact that juries tended to acquit more often – especially when the accused had no criminal record – may simply reflect different assessments by judges and juries of how certain one needs to be to convict someone with a clean record.
Reference: Frarrell, Amy and Daniel Givelber (2010). Liberation Reconsidered: Understanding Why Judges and Juries Disagree About Guilt. Journal of Criminal Law & Criminology, 100 (4), 1549-1586.
A meta-analysis of 27 independent findings demonstrates that minority suspects who come in contact with the police are more likely to be arrested than white suspects.
Researchers interested in the effect of race on the decision by police to arrest a suspect typically attempt to control for legal factors such as the strength of the evidence against the accused, the seriousness of the offence, the criminal record and any mandatory policies that might exist in the jurisdiction. Defining arrest as “taking a person into custody for the purpose of charging him/her with a criminal offence”, this study examines all available high quality studies carried out in the U.S. between 1966 and 2004.
A thorough search of published and unpublished sources located studies involving 27 independent data sets. Obviously these studies varied on a number of dimensions including whether the data were recorded by an observer, the police officer, or victims, whether the study focused on juveniles or people of all ages, and whether the study controlled for such factors as the amount of evidence, type of offence, the demeanour of the offender, the seriousness of the offence, the suspect’s prior record, and whether the victim made a request to the officer on whether to arrest the accused. The meta-analysis allows one to determine whether the inclusion of these variables affects the relationship between race and arrest rate.
Because some studies reported more than one estimate of the impact of race on arrest decisions, four different estimates were used: the average effect size, the largest, the smallest, and what was judged to be the methodologically best estimate of the effect. Nevertheless, the results are remarkably similar: between 19 and 24 of the 27 studies (depending on which effect size is included) show effects supporting the conclusion that minorities are more likely to be arrested than whites. Pooling across the 27 studies there was a significant effect of race. On average the arrest rate for whites was about 20%; for minorities it was about 26%. Studies varied, of course, on how adequately they controlled for legally relevant factors. However, the adequacy of the controls for legally relevant factors was not related to the race effect: even in the best studies, Blacks were more likely to be arrested than Whites. Similarly, those studies that attempted to control for the demeanour of the suspect showed effects as large as those that did not.
Conclusion. “The results are not mixed. Race matters [in police decisions on whether to arrest]. [The] finding is consistent with what most of the American public perceives, and that finding holds over time, research site, across data collection methods, and across publication types. Furthermore, controlling for demeanour, offense severity, presence of witnesses, quality of evidence at the scene, the occurrence or discovery of a new criminal offence during the encounter, the suspect being under the influence of drugs or alcohol, prior record of the suspects, or requests to arrest by victims does not significantly reduce the strength of the relationship between suspect race and arrest” (p. 498). Even though the overall average size of the effect might seem to be relatively small, “because of the interconnectedness of decisions made in the criminal justice system, even small racial differences that occur at many points in the criminal justice process will compound and produce profound effects further along in the system” (p. 498).
Reference: Kochel, Tammy Rinehart, David B. Wilson, and Stephen D. Mastrofski. Effect of Suspect Race on Officers’ Arrest Decisions. Criminology, 49(2), 473-512.
A study of serious delinquents demonstrates that most serious delinquents – even high rate offenders - did not persist in their delinquent careers after being found delinquent. Furthermore long stays in prison did not reduce reoffending and for some youths appeared to increase the likelihood of future offending.
Some political rhetoric would lead one to the conclusion that youth crime can effectively be addressed by identifying serious adolescent offenders, and then treating or incapacitating them. The difficulty, as many studies have shown, is that even defining who is a persistent or serious offender is problematic, and those who are labelled as serious or persistent do not necessarily persist (e.g., see Criminological Highlights 1(3)#7) 11(3)#1). These papers examine the offending patterns, over a 3-year period, of 1,354 serious young offenders, age 14-18, from two U.S. cities.
All of the youths in this study had been found guilty of a serious crime (mostly serious crimes against the person) and for most of the youths, this was not their first appearance in court. They (and a parent) were interviewed shortly after they were adjudicated as delinquent and roughly every 6 months thereafter and their self-reports of offending were recorded.
The youths were divided into 5 distinct groups on the basis of their 3-year offending patterns. 24% of these serious offenders were low rate to start with and almost never offended again. 34% of the youths had offended at a relatively low rate in the beginning of the period, but their offending rates declined over time. About 18% started with a moderate rate and continued offending at this rate throughout the 3-year follow-up. 15% started off with high rates of offending but declined to a very low rate over the 3 years. Finally, 9% started off with high rates of offending and remained relatively high.
For four of these five groups – all except those with initially low rates of offending - the more time the youth spent in the community rather than in custody, the higher the rate of offending, a result not surprising given that ‘time in the community’ equates with ‘opportunity to offend’. For the stable low rate offenders, however, (24% of the original sample) more time in institutional care was associated with higher rates of offending. Incarceration for them, it would seem, increased subsequent offending. In addition, ‘time in custody’ did not differentiate the two groups that started off with high rates of offending. The two high rate offending groups – those starting high and dropping off dramatically across the three year period (14% of the total sample) and those starting high and persisting with high rates of offending (9% of actual offending) -- spent almost exactly the same amount of time in custody.
Hence the data show that “even within a sample of juvenile offenders that is limited to those convicted of the most serious crimes, the percentage who continue to offend consistently at a high level is very small… [Moreover] our ability to predict which high-frequency offenders desist from crime and which do not is exceedingly limited…” (p. 469-470) even though the researchers had a total of 22 measures on the youth (including psychological assessments), the youth’s family background, and peers.
Conclusion. “The considerable heterogeneity in offending patterns in the immediate years after court involvement challenges the political rhetoric in juvenile justice and the popular and scientific fixation on identifying lifelong antisocial personality problems. These results do not support the view that serious offenders are headed toward a life of crime. Most, in fact, had very low levels of involvement during the entire 3-year follow-up period. Furthermore, for these youths, “incarceration may not be the most appropriate or effective option, even for many of the most serious adolescent offenders. Longer stays in juvenile facilities did not reduce reoffending; institutional placement even raised offending levels in those with the lowest level of offending” (Paper 2, p, 3).
References: Mulvey, Edward P., Laurence Steinberg, Alex R. Piquero, Michelle Besana, Jeffrey Fagan, Carol Schubert, and Elizabeth Cauffman (2010). Development and Psychopathology, 22, 453-475. Mulvey, Edward P. Highlights from Pathways to Desistance. OJJDP Juvenile Justice Fact Sheet, March 2011.