Criminological Highlights Vol. 6, No. 6 – October 2004

View this issue as a PDF: PDF iconCrimHighlightsV6N6.pdf

This issue of Criminological Highlights addresses the following questions:

  1. Who has benefited most from the U.S. crime drop?
  2. What is ‘real’ about the stories one hears about restorative justice?
  3. Why do so many people voluntarily allow themselves to be searched when they are in possession of illegal substances?
  4. Do Canadians really want murderers to stay in prison forever?
  5. Did ‘doing time’ in American prisons change when the ideology of corrections changed?
  6. Why did the U.K. import crime policies from the U.S.?
  7. Can police discretion with youths be controlled through legislation?
  8. How often do convicted sex offenders re-offend?

“Since the 1970s, U.S. criminal victimization has become markedly more concentrated in the poorest economic groups, as the overall drop in crime since 1974 has benefited upper-income households much more than it has benefited the poor” (p. 89-90).

Background.  The National Crime Victimization Survey in the U.S. has been collecting information about individual and household victimizations since 1974.  This study looks at those data (between 1974 and 2000), focusing primarily on violent crimes (assaults, robberies and rapes).  Generally speaking it has been well established in the U.S. that violent crimes tend to be concentrated among the poor, whereas household thefts and motor vehicle thefts tend to be crimes disproportionately experienced by the wealthy (in part, of course, because the poor are less likely to have goods worth stealing).  The study focuses on the ‘20-20’ ratio – the victimization rate of people in the poorest 20% of households in the U.S., divided by the victimization rate of people in the wealthiest 20% of households in the U.S. As this number increases, the concentration of crime among the poor increases. 

The findings demonstrate that violent crime has become more concentrated within the poorest part of American society.  “For violent crimes, the 20-20 ratio grew from 1.15 [the victimization rate for poor people was slightly higher than the rate for rich people] during the 1974-1980 period to 1.76 in the 1994-2000 period (a 53% increase)” (p. 95-6). This finding was consistent across various indices of the inequality of violent victimization rates. For burglary there were similarly consistent results, though the pattern for thefts (household and motor vehicle) was less consistent across measures. 

When one looks at serious violent crimes (rape, robbery, aggravated assault), the 20-20 ratio increased from 1.59 to 2.54 across this same 20 year period.   An analysis of the actual rates of victimization for rich and poor Americans shows that during the period of decline in rates of violence in the U.S., the increased inequality in rates of violent victimization is the result of “especially large drops in victimization among the wealthy” (p. 102). In addition, violent victimization inequality grew most dramatically for violent crimes where the victim knew the offender.

During the latter part of the 20th century, there were demographic changes amongst the poorest and richest groups of Americans that appear to account for part of this change. Two-thirds of the growth in inequality of victimization for violent crimes (p. 112) can be accounted for by demographic changes. During this period the richest group became relatively older (compared to changes in the poorest group), more likely to be married, and more likely to be working and living in a suburban setting.  Each of these characteristics of the wealthy is associated with being relatively less likely to be victimized. 

Conclusion. “The decline in crime that the U.S. has experienced over the past 3 decades has not been shared equally: The wealthy have enjoyed the benefits of this decline much more than the poor have” (p. 109).  “For the most serious violent crimes, the disparity between rich and poor has become quite large” (p. 110). 

Reference: Thacher, David (2003) The Rich Get Richer and the Poor Get Robbed: Inequality in U.S. Criminal Victimization, 1974-2000.  Journal of Quantitative Criminology, 20, 89-116. 

Back to top

Advocates of restorative justice (RJ) have created at least four myths about the origin and nature of RJ practices in the criminal justice system.   In the long run, RJ approaches may have a better chance of surviving if they are grounded in reality rather than myth even though the myths may, initially,  help sell the approach. 

strong mythological beliefs about the approach.   These myths include unsubstantiated claims that, in the long run, may well bring all restorative attempts in criminal justice into disrepute. (See also, Criminological Highlights, 5(3) #1).   These claims include the following: 

  • Myth 1: RJ is the opposite of retributive justice, implying that the former repairs harm, involves dialogue and negotiation, and involves the community. Traditional justice is seen as punishing, adversarial, and a procedure in which the community’s interests are made secondary to state interests.  Systematic observations of restorative (family group) conferences (See Criminological Highlights, 6(3) #7), however, suggest that participants have multiple aims, including censure or punishment for past offences, rehabilitation of the offender, as well as restorative interests. Reconciliation with the victim may or may not occur in this setting. Reconciliation and restorative outcomes occur much less frequently than is often claimed.
  • Myth 2: RJ uses indigenous justice practices.  Anthropologists have noted that many traditional forms of justice involve “harsh physical (bodily) punishments and banishment” (p. 62).  In addition, “the specific histories and practices of justice in pre-modern societies are smoothed over and lumped together as one justice form” (p. 63) in a form of nostalgic ethnocentric haze. The notion that modern conferencing in New Zealand had its origins in Maori culture is simplistic. It is more accurate to describe it as “spliced justice” – drawing principles from more than one justice tradition.
  • Myth 3: RJ is a  “ ‘care’ (or feminine) response to crime in comparison to a ‘justice’ (or masculine) response” (p. 64).  The suggestion is that RJ is caring, informal, superior and feminine, whereas retributive justice is tough (or nasty), masculine, formal, state based and inferior.  The difficulty is that “care responses to some offenders can re-victimize some victims; they may be helpful in some cases or for some offenders or for some victims or they may also be oppressive and unjust for other offenders and victims.  Likewise, with so-called ‘justice’ responses” (p. 66).  The difficulty is with the way in which RJ approaches have “been argued and sold to academic audiences and wider publics. There is a loss of credibility when analyses do not move beyond oppositional justice metaphors, when claims are imprecise and when extraordinary tales of repair and goodwill are assumed to be typical of the restorative justice experience” (p. 66).  “Attention needs to be given to the reality on the ground, to what is actually happening in, and resulting from practices that fall within the rubric of RJ” (p. 66). 
  • Myth 4: RJ often produces major changes in people.  There are many stories about dramatic successes in RJ, some of which are told by central RJ advocates and their friends.  There are at least two difficulties with these stories. First, some of the most commonly told stories in the RJ literature never happened and are best thought of as advocates’ dreams of what could or should happen.  Second, these stories are, like all stories, not necessarily representative of what happens more generally.  Hence the stories need not be given more weight than, for example, stories about how a good thrashing turned some young delinquent from crime. Systematic research about RJ practices shows that it is very rare that the “exceptional or ‘nirvana’ story of repair and goodwill” (p. 70) occurs – perhaps only 10% of the time.  Furthermore, the most methodologically sophisticated evaluation of RJ practices (the Reintegrative Shaming Experiment in Canberra, Australia) did not, for example, find any differences in recidivism (using traditional statistical standards) between those who went through RJ procedures and those who went to court.  In terms of creating long term changes in offending, then, RJ appears to be no better than court. 

Conclusion. Advocates of RJ often tell mythical stories about the origin and effects of this approach.  There are likely to be lessons that can be learned from RJ approaches, and presenting RJ as a solution to most justice problems may help, initially, to sell it to a wide audience.  In the long run, however, the true believers may create disillusionment with the whole RJ approach because their extravagant claims cannot be supported.

Reference: Daly, Kathleen. (2002)   Restorative Justice: The Real Story.  Punishment and Society, 4, 55-79.  

Back to top

U.S. courts have used different rules to interpret ‘requests’ by the police to carry out a ‘voluntary’ search of a suspect and ‘requests’ from suspects who want to speak to a lawyer.

Background.  How should the following be interpreted?  

  • Police officer to driver (whom he has just stopped): “Do you mind if I check [your luggage]?”   “Do you mind if I check your person?”  “May I see your driver’s licence?”  “May I look in your trunk?” (p. 230).  Are these requests that can be refused or are these polite forms of commands which, if refused, will escalate into clear commands? 
  • Police officer to driver: “Does the trunk open?”   Is this a question about whether the trunk opens (the literal interpretation) or a request, or is it a command? If there is an affirmative answer by the driver and he opens the trunk, was this done voluntarily or because he was ordered to do so by the police? The problem is that while “courts have little trouble using pragmatic information to determine that an officer’s informational question… can function as a request or command” (p. 231),  distinguishing a command from a request depends on other contextual information and cannot be determined by the words alone.  

Linguistic theory would suggest that courts have used “selective literalism” (p. 231) in interpreting the language of police and accused people. The difficulty in separating commands from requests is that for reasons of politeness, we often make requests and commands indirectly.  Nobody thinks that the words, “Could you pass the salt?” is a request for information about the ability of the listener to pass the salt.  Yet the meaning of “Don’t you think it would be a good idea to shine those shoes?” (p. 236) depends completely on whether the statement comes from a sergeant in the army who is inspecting his troops or a friend talking to someone who is about to go to a job interview.  The sergeant inspecting his troops is not asking a question; he is giving a command. “No” is almost certainly not an acceptable answer in this context.  The friend may be either asking a question or giving advice.   And when the police officer asks “Can I have a look in your trunk?” (p. 237) it seems likely that the officer isn’t asking the driver whether the driver thinks the officer has the ability to look in the trunk.  Similarly, when a police officer asks a suspect if he “minded” if the officer searched his person, it seems likely that few people would interpret this as a request for information about the suspect’s preferences.  Even the imperative form is often not a command:  While “Ready, Aim, Fire” is a command, “Shut the door, please” is usually seen as a request, and “Come in and sit down” is normally an invitation, “Watch out” is probably a warning, and the words “Enjoy your meal”, coming from a waiter in a restaurant,  is almost certainly more likely to be a wish than a command (p. 240). “The most common way to make a polite command is to phrase it as a request.”   A judge might issue a command to a law clerk by saying “Could you draft me a memo” (p. 241),  and few law clerks would try to turn an affirmative reply into a statement of the clerk’s memo-writing ability.

In U.S. courts, however, judges have been selective in the manner in which they interpret language.  “May I see your driver’s licence” is almost certainly, coming from a police officer, an order (p. 241).   “When a person in a position of power ‘asks’ or ‘requests’ us to do something, it will normally be interpreted as a command” (p. 241). In part, this is the case because many people will assume that it is a legitimate command.   

A similar problem occurs when police officers ask about illegal substances (e.g., drugs, guns). When a suspect indicates that he has none, and then the officers asks, “Would you mind if I search your vehicle and contents to be sure there is no contraband in the vehicle?” (p. 246) this statement, which literally is a request for information about the suspect’s feelings,  is not likely to be interpreted in those terms.  In such a case, the citizen is primed to agree to the search in part because “Refusing consent will only make them appear suspicious, so that if the officer did not have a legal reason to search them previously, he would certainly have one now.  Add to this the inherent coerciveness of the situation, and it ceases to be surprising that such large numbers of people consent to searches during traffic stops” (p. 246).  “It was only if the occupants of the car were aware that the police had no authority to order them to open the trunk that the ‘request’ to search the trunk could be interpreted as a true request with an option of refusal” (p. 247).  Courts in the U.S., however, have tended to interpret words that are almost certainly received as commands as being requests.  As if by magic, the context disappears and words which are heard as commands are turned into requests.

When a patron in a restaurant says “I’d like the salmon special” or “I feel like trying the salmon” it would be the waiter who was ideologically opposed to tips who would interpret this as being mere expressions of desire and reply, “That’s interesting. What would you like to order me to bring you?”.  Similarly, it would be our same tip-free waiter who would think that “Could you bring me a glass of water?” (p. 249) was a question about his ability to carry water across the room.  

When a person being interrogated for 6 hours without break suddenly says, “I need to use the toilet” (p. 250), the chances are that this is interpreted as a request, and perhaps even an urgent request.  But when a defendant says that he “might need to see a lawyer” or “would like to have a lawyer” courts have said that this is not the invoking of the right to counsel because (in the latter case) it is a statement of desire, not a request.  Thus when the question arises as to whether an accused person has invoked his or her right to counsel, U.S. courts suddenly force the speaker to use words whose meaning is not dependent on the context. The words, “I think I might need a lawyer” or “Maybe I need a lawyer” or “If I’m going to be charged with murder maybe I should talk to an attorney” (p. 251-2) are not seen, by U.S. courts,  as invoking the right to a lawyer because they are hedged just as “Didn’t you say I have the right to an attorney” was deemed by an appeals court not to be an invocation of the right because it merely asked a question.  

“Research by linguists over the past two or three decades has shown that an indirect speech style and greater use of hedging tends to be associated with people of lower socioeconomic status” (p. 253). Furthermore, while “men are more likely to make direct orders or requests, such as “close the door” or “Please close the door” women tend to use what are considered more polite formulations, “Will you close the door?” or “Won’t you close the door?” (p. 253).  This use of less powerful speech tends to be located primarily among those of lower social status.  Hence, the U.S. Supreme Court’s decision that “Maybe I should talk to a lawyer” was not an invocation of a right is likely to disadvantage certain groups within society.

Conclusion. “Judges are selective in [their decisions about] when they take pragmatic factors into consideration… Their interpretive practices tend either to ignore or to take into account pragmatic information when it benefits police and prosecutors.  The utterances that police officers make in seeking consent to a search are almost invariably interpreted as requests, even if the officer poses what is literally an informational question or if the circumstances are such that the suspect is likely to interpret the utterance as an order that should not be refused….  In contrast people subject to interrogation are held to a higher linguistic standard than are the police: they must be quite literal in invoking their right to counsel” (p. 256).  

Reference: Tiersma, Peter M. and Lawrence M. Solan.  (2004) Cops and Robbers: Selective Literalism in American Criminal Law.  Law and Society Review, 38, 229-265.  

Back to top

If ordinary Canadian citizens really think that sentences and the parole system are not harsh enough for the most serious cases, why don’t they act that way? 

Background. Public opinion surveys in Canada and elsewhere have suggested, for the past 35 years, that the public is dissatisfied with sentencing and with parole decisions, and would like both processes to be made harsher. In the area of parole, therefore, it would be easy to conclude that the public is “implacably opposed to granting parole [especially] to offenders convicted of those most serious crimes” (p. 104).  Most Canadians, in a public opinion poll conducted in 1987 said that murderers should not be eligible for parole (p. 108). Nevertheless, Canadian law allows all those serving life sentences to apply for parole.  Those convicted of murder and given parole ineligibility periods of more than 15 years can apply, after they have served 15 years, to go before a jury of 12 citizens and have the parole ineligibility period reduced – from an initial period of up to 25 years down to as little as 15 years. 

This study examines the results of the hearings held under S. 745 of the Criminal Code – best known in Canada as the “faint hope clause.”  Until 1 January 1997, the prisoner was successful in having the parole ineligibility period reduced if at least 8 of the 12 jury members were in favour of this outcome. Only about 25% of those eligible for a hearing applied. But of those cases (before 1 January 1997) in which hearings were held, 80% were successful in achieving at least some reduction in their parole ineligibility periods. 

After 1 January 1997, the rules were changed for those serving life sentences such that those convicted of multiple murders are ineligible to apply for an earlier parole hearing and a superior court judge must agree that there was a “reasonable prospect” for success.  Finally, the jury of 12 citizens has to be unanimous.  These changes were legislated immediately after one of Canada’s most notorious serial killers (Clifford Olson) had applied (unsuccessfully) to have his 25 year parole ineligibility period reduced.  The main effect of these changes appeared to have been to reduce the number of hearings: In the first three years after the rules changed, only about 12% of eligible prisoners had hearings.  But of those who did apply, 77% were successful.  

Conclusion.  Most Canadians indicate, in public opinion polls, that, in the abstract, they are in favour of keeping those convicted of murder in prison forever without allowing them to be eligible for parole.  Nevertheless, when given a chance to respond to individual cases, juries – by majority vote before 1997 and unanimously thereafter – are very likely to reduce the parole ineligibility period for those convicted of murder, even for those convicted of first degree murder.  When placed on juries, members of the public appear to be able to “discharge their duties to react as disinterested decision makers, even in cases involving prisoners serving life terms for the most heinous crimes” (p. 110).   “There may be an advantage in allowing jurors, rather than criminal justice professionals to make this decision.  If the decision to reduce the time served prior to parole eligibility is made by members of the public, the criticism that the parole system is too lax… loses much of its power” (p. 111).   Canadians, it seems, are not as tough as they sometimes sound.

Reference: Roberts, Julian V.  (2002). Determining Parole Eligibility Dates for Life Prisoners: Lessons from Jury Hearings in Canada.  Punishment and Society, 4, 103-113.

Back to top

The ideology, discourses, and policies having to do with imprisonment may have changed in the U.S., but for women in one of California’s women’s prisons, ‘doing time’ is pretty much the same as it was 40 years ago. 

Background.  Forty years ago, the U.S. federal and state imprisonment rate was approximately the same as Canada’s overall imprisonment rate.  It is now more than 4 times higher.  But the rate of imprisonment in the U.S. is not the only thing that has changed.  In the early 1960s, “the rehabilitative model dominated official penal discourse [in California]” (p. 268) but by the 1990s, there was, in the U.S.,  a “rejection of rehabilitative and normalizing goals [and] a growing emphasis on managerial goals and actuarial techniques to efficiently classify and contain what is seen as an essentially irredeemable population….” (p. 270).   But did this change in ideology have an impact on the day-to-day lives of those subject to it – the prisoners?

This study examines women’s experiences ‘doing time’ in the California Institution for Women (CIW) at two points in time: the early 1960s (the height of the rehabilitation era) and the mid-1990s (during the ‘get tough’ era).  The 1960s data came from a study by Ward and Kassebaum; the 1990s data were collected by the current authors, using interviews, informal discussions, and a survey of 887 inmates.

In many ways, CIW changed over this 35 year period. In the 1960s, the assumption was that, compared to male prisoners, women “had different problems and consequently, they needed different treatment” (p. 279). “CIW… relied on maternal and therapeutic methods in its efforts at rehabilitation” (p. 279).  Correctional workers (all women), most of whom had college degrees and training in social work, and prisoners were expected to develop relationships “of genuine professional interest, seasoned with warmth and friendliness” (p. 279, quoting the staff manual of the period). In 1995, there were twice as many women in CIW as there had been in 1963.  Half of the correctional officers were men.  Local rules had been replaced by state rules which demonstrated clear evidence of a “shift from the rehabilitative model to a managerial model” (p. 281).  Rehabilitation, though talked about as a goal of imprisonment, “had become an individual, not an institutional, responsibility” (p. 282). 

But how did all of this affect the experience of ‘doing time’ in CIW?  Prisoners in the two periods were interviewed and, on a survey, were asked a number of identical questions.  Generally speaking, in both periods, prisoners saw the best way to do one’s time was to mind one’s own business.  Detachment from and distrust of other prisoners was seen as the best strategy.  There were some differences in the results from the two time periods, but overwhelmingly, it was a matter of degree. As one woman who had been in CIW for 25 years noted (in 1995): “The faces have changed, the words have changed, the clothes have changed.  But the way women do time has not changed that much” (p. 298).

Conclusion: “Whether subject to the maternal, therapeutic regime of the 1960s that promoted rehabilitation through individualized treatment or to the neoliberal regime of the 1990s that shifted responsibility for rehabilitation onto prisoners, women at [the California Institution for Women] lived with and negotiated features of imprisonment  that shaped their experiences in comparable ways” (p. 299).

Reference: Gartner, Rosemary and Candace Kruttschnitt (2004). A Brief History of Doing Time: The California Institution for Women in the 1960s and the 1990s. Law and Society Review, 38, 267-303.

Back to top

Ordinary people are able to overcome their negative views of people with criminal records if they are given information about the person from a previous employer who indicates a willingness to rehire the former offender if given an opportunity to do so.

Background. Political scientists in the U.K. have noted that “domestic social policy [in Britain] has more and more involved the importation of ideas from abroad, particularly from the United States” (p. 165).  In the area of criminal justice policies, this is particularly interesting in light of assessments such as that of an American historian who remarked: “The least controversial observation about American criminal justice today is that it is remarkably ineffective, absurdly expensive, grossly inhumane, and riddled with discrimination” (p. 166).  Why, then, do British politicians find these policies so seductive when they talk about “evidenced based” policy making?  

The list of American criminal justice policy imports into the U.K. includes the following:

  • Zero tolerance policing which focuses “as much on minor infractions and incivilities as on major crimes” (p. 167). 
  • Curfews for young children.
  • Three strikes sentencing for certain crimes.
  • Private prisons. 
  • Electronic monitoring.
  • The creation of the office of a Drugs Czar.

Why have these changes occurred? Various reasons are offered, including the following:

  • Ideological proximity (Thatcher and the Elder Bush; Blair and Clinton).
  • Electoral success (e.g., the success of the Elder Bush’s ‘Willie Horton’ campaign and Clinton’s success in not allowing the Republicans to take over his Right flank on criminal justice policies.  Blair was apparently particularly impressed, during a 1993 visit to the U.S. with Clinton’s ability to avoid the “soft on crime” label, a traditional Democratic liability). 
  • The language of politics which “conceives of social problems as superficial and therefore as being easy to deal with” (p. 175) and which sees social problems as “simple, making intervention obvious and easily identifiable” (p. 175). 
  • Symbolic politics whereby being “soft on crime” has “become the crowning curse of political discourse” (p. 176) and “the need to appear tough has debased public discourse about criminal justice” (p. 177). 

Conclusion. The U.K. has imported some U.S. criminal justice policies in part because they work, politically.   But “there remain some quite important differences between the two systems of social control… It [is] abundantly clear that penal rhetoric was generally a more successful import than penal policy. [For example] In the UK there has been significantly more talk of ‘zero tolerance’ than there has of [actual] zero tolerance policing.  A drug Czar has been appointed but as yet the punitive practices of the ‘war on drugs’ have been held largely at bay” (p. 184-5) though in other areas serious toughening appears to have occurred. 

Reference: Newburn, Tim. (2002). Atlantic Crossings: ‘Policy Transfer’ and Crime Control in the USA and Britain.  Punishment and Society, 4, 165-194.

Back to top

New South Wales has managed to enact legislation that reduces the use of youth court proceedings. 

Background.  Australian states, like many jurisdictions (including Canada) have, in the past 40 years, moved away from a quasi-welfare model of responding to young offending toward a more justice orientation.  Along with this change has come a movement towards pre-court diversion and various forms of restorative conferencing.  New South Wales’ (NSW) Young Offenders Act 1997 (YOA) attempts to regulate police discretion and to use the courts only as a last resort.  The challenge faced by those drafting the YOA was that although  “the need to guide and regulate police discretion is… rarely disputed, … the appropriate mechanism for this regulation has been the subject of much debate and research” (p. 74).  Prior to 1997, NSW police instructions set out procedures and guidelines for the use of warnings and cautions by the police, but “these discretionary options were not well utilised” (p. 76).

The YOA provided “specific rules guiding police discretion in referral decisions” (p. 77). Warnings (in which admissions of guilt were not required) and cautions (where the youth had to admit guilt) were to be used for minor offences, whereas conferences were not supposed to be used for these offences.  If an investigating police officer thought a youth was not eligible for a warning/caution, a Specialist Youth Officer (SYO) would decide if the case should go to a conference or to court.  The decision of the SYO can be challenged by conference administrators.  Ultimately, the Director of Public Prosecutions (DPP) “makes the final decision on whether the young [person] should be cautioned, dealt with by way of conference or referred to the court” (p. 79).  Finally, if the “court considers that a less intrusive intervention should have occurred, the magistrate can either administer a caution or refer [the case for a conference]” (p. 79). 

The impact of the YOA on the use of youth court in NSW appeared to be quite dramatic.  In the first three years of the new law, there were approximately as many cases that came to the attention of the police as there had been in the previous three years.  Prior to the implementation of the new law, 82%-85% of these cases were sent to court.  In the three years after the implementation, 59%-68% went to court.  Some of this shift was due to the use of conferences, but about four times as many cases were diverted from court to cautions than were going to conferences.  The findings “provide no evidence that the introduction of the YOA has resulted in net widening of the overall juvenile justice system” (p. 84) – the increased use of cautions and conferences were offset by a reduction in the use of court. 

Conclusion. It would appear that having explicit rules about the exercise of discretion by police can make a difference.  Because the law governing discretion was an Act of Parliament (rather than a set of police directives) “the majority of officers took it seriously” (p. 90). In addition, the involvement of trained specialist youth officers in the decision making process along with various checks and balances for resolving disagreements meant that decisions were always being scrutinized. Finally, the legally mandated involvement of the prosecutor and, ultimately, the judge as gatekeepers to the court process meant that decisions by a single investigating police officer could not guarantee that a youth would end up in court. 

Reference: Chan, Janet, Jenny Bargen, Garth Luke and Garner Clancey. (2004) Regulating Police Discretion: An Assessment of the Impact of the NSW Young Offenders Act 1997. Criminal Law Journal, 28, 72-92. 

Back to top

An analysis of data from ten samples of sentenced sex offenders demonstrates that most sexual offenders who have been apprehended and sentenced do not commit further sexual offences.  

Background. For reasons that are not clear, many commentators seem to assume that once a person commits a sex offence, he will, if not imprisoned, continue committing these offences until he dies.  One difficulty in assessing the likelihood of recidivism is that recidivism is defined in different ways across studies.  Different follow-up periods and different criteria of offending (typically either charges or convictions) are used in different studies. In addition, different definitions of sex offences are used.  Different types of victims (children vs. adults, gender, incest vs. other types of relationships) are also studied. A previous study (Criminological Highlights, 6(3)#3) showed that recidivism rates for sexual offenders were generally no higher than recidivism rates for other types of offenders. This would suggest that a special focus on sex offenders (as opposed to other serious offenders) seems misplaced. 

Rather than looking at recidivism rates in a single sample, this study looked at 10 different sub-samples of adult male offenders (total N=4,724; 7 Canadian, 2 U.S. and 1 U.K. sample) with follow-up periods ranging from an average of 2 years to an average of 23 years. Most used re-conviction as the measure of recidivism, but there did not appear to be dramatic differences between the studies that reported only charges and those that reported convictions. 

The results demonstrate that pooling across all studies, 14% of the offenders had recidivated with another sexual offence after 5 years, 20% after 10 years, and 24% after 15 years.  When one looks at the recidivism rate for different types of victims, “boy victim child molesters” showed a 10 year recidivism rate of 28% as compared to 13% for “girl victim child molesters.”  Incest offenders had the lowest rate (9%) and rapists were about average (21%).  Not surprisingly, those with a previous conviction for a sexual offence were considerably more likely to re-offend sexually than those without a previous sexual conviction (32% vs. 15%).  Offenders over age 50 were considerably less likely to re-offend sexually within 10 years (11%) than were younger offenders (21%).  Generally speaking, the longer a person had been in the community, the less likely it was that he would re-offend.   

The recidivism results reported here are quite similar to those reported elsewhere.  A recent U.S. study, for example, found a sexual re-offending rate after 3 years of only 5.3% (See also Criminological Highlights, 3(3)#3;  5(1)#4; and 6(3)#3). 

Conclusion. It is suggested that in considering policy on how to deal with convicted sex offenders, there is a need to “differentiate between the high public concern about these offences and the relatively low probability of sexual re-offence” (p. 11). Other studies would suggest that incapacitation is an inefficient approach to crime prevention for these offences, in part because of the unpredictability of who will, if not incapacitated, re-offend (see Criminological Highlights, 3(1)#1).  Given the difficulty in predicting re-offending, it would appear that incapacitating sex offenders will not be a more effective crime control measure than incapacitating any other type of offender. 

Reference: Harris, Andrew J. R. and R. Karl Hanson (2004). Sex Offender Recidivism: A simple Question.  Ottawa: Public Safety and Emergency Preparedness Canada.

Back to top

Publication Type

Volume Number


Issue Number