Disproportionate Harm: Hate Crime in Canada

2. METHODOLOGICAL ISSUES

2. METHODOLOGICAL ISSUES

Before presenting statistics on the incidence and reporting rates of hate crimes, it is important to discuss the question of how to define a hate-motivated crime.

2.1 Hate Crimes: The Problem of Definition

Perhaps the central problem in the classification and recording of hate crimes is the issue of definition. The empirical data that follow in this report must be considered in light of the definitions which guided their collection. If definitions of what constitutes a hate crime are highly variable, this will generate inconsistency in statistics purporting to measure the activity. The focus in this section is on the definitions of hate crimes used by different police agencies across Canada and in other jurisdictions. First, however, it is worth noting a general definition provided by researchers working in the area. According to Garofalo and Martin (1991: 17):

A bias-motivated crime is a crime in which the offender is motivated by a characteristic of the victim that identifies the victim as a member of some group towards which the offender feels some animosity.

As will be seen, the definitions used by different police forces are more specific and more restricted in scope.

2.2 Hate Crime Definitions in Other Jurisdictions

2.2.1 United Kingdom

The definition used by the police in the United Kingdom is restricted to racially motivated crime incidents, and assumes the following form:

(a) Any incident in which it appears to the reporting or investigating officer that the complaint involves an element of racial motivation; or,

(b) Any incident which involves an allegation of racial motivation made by any person (Maung and Mirrlees-Black, 1994).

The British definition suffers from the deficiency that it excludes hate crimes directed at targets other than racial minorities. Thus, other forms of hate crime such as anti-semitism, or anti-gay attacks are not captured either by the official police statistics or by the periodic victimization survey (British Crime Survey). On the other hand, the British definition has the advantage of defining a hate crime by specific reference to the perception of the victim(s), even if this perception is at odds with the view of the investigating officer.

2.2.2 United States

Definitions of hate crimes vary across the United States. The following examples are representative of these:

Hate Crime: Any unlawful action designed to frighten, harm, injure, intimidate or harass an individual, in whole or in part, because of a bias motivation against the actual or perceived race, religion, ethnic background or sexual orientation of the victim (IACP National Law Enforcement Policy Centre, 1991).

An act which appears to be motivated or perceived to be motivated by the victim based on race, religion or ethnic background (Maryland: see Cook, 1991).

2.3 Hate/Bias Crime Definitions used in Canada

Some police forces provided a clear definition in response to the Department of Justice Canada request; for others, the definition quoted below comes from bias crime guidelines provided to officers.[3]

Metropolitan Toronto Police Force

A hate crime is a criminal offence committed against a person or property that is based solely upon the victim's race, religion, nationality, ethnic origin, sexual orientation, gender or disability.

Halifax Police Department

A hate crime is a criminal offence committed against a person or property, the motive for which is based in whole or in part upon the victim's race, religion, nationality, ethnic origin, gender, disability or sexual orientation.

Edmonton Police Service

Bias crime: A criminal offence committed against a person or property, that is based solely upon the victim's race, religion, nationality, ethnic origin or sexual characteristic.

Ottawa Police Service

A criminal offence committed against a person or property which is motivated by the suspect/offender's hate/bias against a racial, religious, ethnic, sexual orientation or disability group.

Winnipeg Police Department

Hate crimes are traditional offenses motivated by an offender's bias as a result of religion, race, nationality or sexual orientation.[4]

Montreal Police Force

The Montreal police force uses the same definition that is used in Toronto (see above).

Ministry of the Solicitor General / Correctional Service of Canada

Crime was motivated because of hate/bias toward the victim's racial, religious, ethnic or sexual orientation.

Policing Standards Manual, Province of Ontario

A criminal offence committed against a person or property which is motivated by the suspect/offender's hate/bias against a racial, religious, ethnic, sexual orientation or disability group.

Ontario Provincial Police

A criminal act against a person(s) or property that is based solely, or in part, upon the victim's race, religion, ethnicity, sexual orientation or disability.

Royal Canadian Mounted Police (RCMP)

The RCMP does not use the category "hate crime" in any formal way. However, some hate crimes are clearly addressed by the National Security Investigation Sections ofthe RCMP. Criminal, political or religious extremism, for example, can take a form that most people would recognize as a hate crime.[5] Most of the hate crimes described in this report fall within the ambit of the provincial or municipal police services, rather than within the jurisdiction of the RCMP in its federal role. Although the RCMP does gather information relating to ideologically-motivated serious crime, statistics are not routinely compiled on criminal incidents that were motivated by hatred.

One critical question springs out from this short list: how inclusive should the definition of a hate crime be? There are clear differences between the definition used in Toronto and Edmonton for example, in which the act must have been based solely on some victim characteristic, and the broader definition in which the words "in whole or in part" are used. Clearly, a uniform definition is necessary if the statistics are to be comparable across different provinces.

2.4 Towards a Uniform Definition

Before hate crime statistics can be collected in a systematic way, a standard definition of a hate or bias crime will have to be adopted. This will necessitate a critical examination of definitions currently used by Canadian police forces or organizations outside the criminal justice system, such as B'nai Brith. Some components of the definition will be critical. For example, the definition used by the Halifax Police Department describes a hate crime as an offence motivated "in whole or in part" by hatred of certain victim characteristics. The Metropolitan Toronto Police use a more restrictive definition of a crime that is based "solely" upon a victim's race, religion, nationality, ethnic origin, sexual orientation, gender or disability.

If the more general definition is used, there are likely to be far larger numbers of hate crimes recorded by the police across Canada. The Toronto version seems overly restrictive; almost no crime is committed solely for one reason or another. As with most human behaviour, hate crime may well be multiply determined. Once again the European experience is instructive. Hate crimes committed against racial minorities in England or Jewish citizens of France have not been random in the sense that any member of the particular minority group has been targeted. Whichever definition is chosen, it must be applied across the country, or else hate crime statistics will present a distorted picture.

As noted by the Metropolitan Toronto Police in their submission to the Department of Justice Canada:

It is essential, in our view, that the definition and criteria for determining the extent of hate crime be standardized so that when comparisons are made and programs developed, the supporting data is sound and will meet the tests of integrity...[otherwise] the Metropolitan Toronto Police may report 150 hate crimes, and another police force report only 70 hate crimes. These figures, representing different criteria and definitions, would limit their usefulness in further analysis and program development.

The experience with the issue of crime definition in the United States is also worth noting. By 1994, over 35 states had enacted statutes pertaining to hate-motivated crimes. These statutes take many different forms, but all states have had to grapple with the question of the extent of motivation required. There has been a fair amount of American case law on the issue, and the general result lies somewhere between the extremes of total and incidental motivation. That is, the state must show that the accused's hate motivation was a substantial motivation for the crime, and not just incidental to the act. However, the prosecution need not show that the crime would not have been committed in the absence of the hate motivation.

To rely on an exclusive motivation definition is likely to have the effect of seriously underestimating the incidence of hate crimes. Many hate crimes will not be classified as such if such an exclusive definition is used.

This point can readily be substantiated by comparing the hate crime statistics from two forces using different definitions, Toronto and Ottawa. As noted earlier, Toronto uses an exclusive motivation definition. In Ottawa, however, the Bias Crime Guidelines state the following:

Bias is to be reported only if the investigation reveals sufficient objective facts to conclude that the offender's actions were motivated, in whole or in part by bias (Ottawa Police Service, 1994).

Assuming that racial or ethnic hatred is a priori the same in the two urban centres, we would expect a much higher number of hate crimes in Toronto. There are approximately five times as many Criminal Code infractions in Toronto than Ottawa.[6] It would be reasonable, therefore, to expect the incidence of reported hate crimes to be approximately five times higher in Toronto than in Ottawa. In fact, a higher proportion of racial and ethnic minorities in Toronto as compared to Ottawa would lead one to expect an even higher rate of hate crimes in Canada's largest city since these minority communities are a principal target of hate crimes. The expectation of a much higher number of hate crimes in Toronto is also sustained by B'nai Brith data from the same year (1993). B'nai Brith records almost half the incidents of anti-semitism in Canada as occurring in Toronto, with only 16 percent of all incidents taking place in Ottawa.

What do the hate crime statistics recorded by the police show? In Toronto, there were 155 hate crime incidents recorded by the police in 1993. This compares to 176 for the same period in the city of Ottawa. The explanation for this would appear to be that one force is using a higher threshold to define a hate crime (a definition that requires exclusive motivation). The same point can be made using American data. Levin and McDevitt (1993: 171) note that:

New York City recorded 525 bias crimes in 1991, while Boston recorded 218 incidents with less than one-tenth of New York's population. This statistic surely reflects the Boston Police Department's more expansive bias crime definition, or, more accurately, the lower hurdle for bias intent which must be met for bias crime classification.

It is clear then, that a standard definition of hate crime as well as uniform criteria for application need to be developed. This recommendation has been made by several critical groups in the area, including the Metropolitan Toronto Police Force.[7] As well, the less restrictive definition by the Ottawa police service (among others) would appear to be the most appropriate one.

There is of course a danger to having a less stringent definition. If the motivation is not exclusive, interpretations of what constitutes a hate crime are likely to be variable, and this will undermine the comparability of statistics.

2.5 Relationship Between Hate Crime Definitions and the Sentencing Reform Bill

Since a critical provision in the recent sentencing reform statute (Bill C-41) calls for enhanced penalties for hate crimes, there should be some consistency between police definitions of such crimes and the statutory definition contained in the Bill. According to Bill C-41, a hate crime is one which, "was motivated by bias, prejudice or hate based on the race, nationality, colour, religion, sex, age, mental or physical disability or sexual orientation of the victim".[8] This Bill is consistent with legislation introduced in other jurisdictions. For example, proposed amendments to the Public Order Act 1986 in England and Wales provide for a doubling of the maximum penalty for any offence where the court is satisfied that the offence was committed on racial grounds.

Comparing the Sentencing Reform Bill's definition of a hate crime to the definitions used by different police forces reveals several inconsistencies. The sentencing statute's definition is clearly broader than several of the police definitions. For example, the physical disability of the victim is grounds for augmenting the penalty for an offence as a hate crime in the sentencing Bill, but does not figure in the definition used by the Winnipeg Police Department or the Edmonton Police Service. These inconsistencies will need to be resolved. And, since the sentencing Bill definition has received the endorsement of Parliament, a strong argument can be made that the police definitions should conform to the sentencing Bill definition, and not the reverse.

The final issue to be noted in this section concerns the nature of the incident reported to the police. In all the definitions provided above, the act which is the subject of an investigation has to be criminal in nature. However, some jurisdictions do not restrict their attention to potential crimes. Hate crime statistics have been collected and analyzed in Maryland since 1981. The definition for collection procedures is the following:

To report an act which appears to be motivated or perceived to be motivated by the victim based on race, religion or ethnic background (Cook, 1991: 107).

The act is not required to be a violation of any statute.

2.6 Limitations on Official Hate Crime Statistics

2.6.1 Unreported Crimes

A central deficiency of criminal justice statistics is that a proportion of incidents are never reported to, or recorded by the police. This statistic is known as the "Dark Figure" of crime, and it varies from offence to offence as a result of a multitude of factors, including public confidence in the criminal justice system, the seriousness of the crime, the magnitude of financial incentives to report, the potential for embarrassment to the victim and many other factors (see Hood and Sparks, 1978). For some crimes like sexual assault, the so-called dark figure may be in excess of 90 percent of incidents actually committed, while for other offences such as theft of a motor vehicle or break and enter (business residence) the dark figure may be as low as 5 percent of all incidents.

What is the dark figure likely to be for hate-motivated crimes? In the absence of reliable data, estimates of the proportion of unreported hate crimes must be rather speculative. However, there are several reasons to believe that of all forms of criminality, hate crimes are likely to be among the most underreported offences. First, the victims (direct or indirect) as well as witnesses may fear additional victimization if the police become involved. The experience in some other jurisdictions has been that witnesses are particularly reluctant to come forward in cases of racially-motivated violence, as these crimes are frequently committed by gangs or groups of offenders. For example, in England there have been several such incidents. In the most recent, a Bengali man was almost killed by a group of 20 white youths, only one of whom was charged. Although the assault took place in broad daylight, the judge noted in sentencing this offender that the police had met with "an almost blank wall of silence in their investigations", due to apprehension of retaliation (Manchester Guardian, 1994).

Second, a large number of hate crimes involve damage to property, and can only be effectively prevented by increased surveillance. Neither public nor private resources can be stretched to include continuous monitoring of property. In several European countries, reporting incidents of hate-motivated vandalism to the police has provoked further cases of hate-motivated mischief.

Another reason why hate crimes are not reported to the police concerns apprehension on the part of victims regarding the response of the criminal justice system. Racial minorities are a prime target of hate crimes. In fact, as police statistics presented later in this report reveal, they are the primary target of hate crimes involving violence. These same racial minorities may perceive racially motivated crimes as part of a larger problem of racism in some of Canada's major cities (see League for Human Rights, 1993). This perception of racism also includes the criminal justice system. Thus racial minorities are likely to be apprehensive of the criminal justice response to a report of a crime against someone in their community, and this may well deter them from reporting a crime. In one recent study, members of the Muslim community in Toronto, for example, expressed the view that they were not comfortable approaching the police to report a crime. This reluctance is likely to be even greater if the crime was motivated by hate or bias. In this sense, too, hate crimes are unlike other crimes.[9] It is also the case that some immigrants may not trust the criminal justice system (and specifically police officers) as a result of negative experiences in their countries of origin. For such people, there is an additional hurdle to overcome. Police services must be aware of this and make additional efforts to reach such individuals.

Victims of hate-motivated crime may be reluctant to report such crimes to the police on account of adverse reaction to such disclosures. Members of the gay and lesbian community may not wish to publicly disclose the fact that they were at certain locations, or may fear negative repercussions if they report a hate crime based on their sexual orientation (See Herek, 1994: 102-103 for a discussion of this form of "secondary victimization"). Research in other jurisdictions has clearly shown that the stigma associated with reporting an "anti-gay" crime has meant that the vast majority of such crimes remain unreported (see Herek, 1989). One major study in America found that members of the gay community were the most frequent targets of hate crimes, although these incidents were seldom reported to the criminal justice system (see Coldren, 1991; Herek, 1994). As well, the attitudes of some police officers towards members of the gay community may well inhibit the reporting of hate crimes of this kind. For these reasons alone it would be unwise indeed to conclude that the number of anti-gay crimes recorded by the police represents the total number of incidents actually committed. Testimony before the Parliamentary committee reviewing Bill C-41, the Bill which prescribes higher penalties for hate motivated crimes, is revealing in this respect. A number of Members of Parliament questioned the utility of including sexual orientation as a ground for hate motivation since in terms of numbers, "there is not an outrageous amount" (House of Commons, November 24, 1995, 65:28).

A final reason why hate crimes might fail to be recorded appropriately has to do with the special investigative problems. In order for a crime to be classified as a hate crime, the officer must record some evidence of hate motivation. In most cases involving crimes against the person this means the offender's use of language. Police officers must pay special attention to the circumstances surrounding the commission of the offence, and without special training, this may not be done (see Levin and McDevitt, 1993: Chapter 12).

There are sound reasons, then, to believe that the incidence of unreported hate crimes is much higher than crimes in general. It is also important to point out that variation exists in terms of the likelihood that different categories of victim will report to the police. An exploration of this issue is beyond the scope of the present report, however, some statistics from the United States suggest that hate crimes directed at the gay community are particularly likely to remain hidden from the criminal justice system. Berrill (1992: 115) for example, reports findings from a national survey of more than 2,000 people in eight cities. The survey found that 15 percent had been physically assaulted, 42 percent had been threatened with physical violence and fully 93 percent reported harassment of some kind as a result of their sexual orientation. Herek and Berrill (1992: 40a) summarize additional survey data and conclude that:

the sheer number of incidents reported in these studies is staggering…the quantitative and qualitative data gathered thus far are a frightening testament to the human cost of anti-gay bigotry.

2.6.1.1 Reporting Rates of Hate Crimes in Other Jurisdictions

Since this is the first document to provide information on hate crime incidents in Canada, it is obviously hard to make firm estimates about the percentage of such crimes that are reported to the police. Our best estimates are extrapolations from similar data in other jurisdictions. Home Office data from the United Kingdom in 1981 suggest that only one racial attack in ten was reported to the police (see Bowling, 1994). A more recent survey in London discovered an even lower reporting rate: only five percent of racial attacks were reported to the police (see London Borough of Newham, 1987).

The latest British Crime Survey showed that reporting rates were lower among Afro-Caribbean respondents when racial motivation was an element of the crime. Thus 30 percent of racially-motivated offences were reported to the police, compared to 43 percent of crime incidents without a racial element (for reasons that are unclear, this effect did not emerge from the Asian respondents). The authors of a recent report based upon the British Crime Survey (BCS) concluded:

The BCS estimate of the number of racially-motivated incidents [in England and Wales] is substantially higher than that recorded by the police. In England and Wales in 1991, 7,882 incidents were recorded by the police as having a racial element. This contrasts with the BCS 'best' estimate of 130,000 racially motivated crimes and threats over the same period (Maung and Mirrlees-Black, 1994: 20-21).

In fact, the dark figure is even greater than these figures suggest, on account of the fact that the police definition of a racially motivated incident is broader than that used in the BCS. There is no reason to suspect that reporting rates for racially-motivated attacks would be any higher than this in Canada. These data suggest that hate crimes involving personal injury are among the least likely forms of criminality to be reported to the criminal justice system.

2.6.2 Classification Problems

Another reason why a hate-motivated crime may fail to enter the official statistics is that it is dependent upon an accurate classification by a police officer. Police officers are understandably reluctant to undertake a decision about the motivation of the suspect. Or, failing to appreciate the importance of determining whether the crime was motivated by hate, officers may simply ignore the issue assuming that an assault is an assault, whatever the reason for the attack. Consider an argument which rapidly escalates into a fight. In the course of the fight, racial insults are exchanged and a criminal charge is eventually laid. Was this a hate crime? Was the assault racially motivated? And if so, was racial hatred the sole motivation or simply a precipitating factor? Only the accused knows the answer to this question[10], and he/she is unlikely to be forthcoming on the issue.Similar problems confront the investigation of property crimes against minorities. Unless the vandalism takes a specific form (such as spray painting racist symbols), it may pass unnoticed as a hate crime.

Another deficiency of hate crime statistics concerns the question of representativeness. As noted earlier, only a minority of sexual assaults are reported to the police. This means that a portrait of the nature of sexual assault based upon cases processed by the courts is unlikely to reflect the true nature of the crime. The same argument applies to hate-motivated crimes. The portrait of hate motivated crime that emerges from official statistics may be very different from the portrait that emerges from a victimization survey which includes incidents that were not processed by the criminal justice system.

For a variety of reasons, then, the hate crime statistics that have been collected to date probably underestimate the incidence of this form of crime, as well as presenting a distorted view of the kinds of hate crimes committed and the groups that are most likely to be targeted.

Conclusion

Extrapolating from research in other countries, it would appear that only a small percentage of hate motivated crimes -- perhaps one incident in ten -- are ever reported to the criminal justice system.

2.7 Data Sources

The data reviewed in this report are drawn from several sources. However, it should be clear from the outset that these sources represent only a fraction of the groups in Canada that have a stake in the issue of hate crime. These data were provided to the Department of Justice Canada in response to a request for information sent in 1994 to police forces across the country as well as to certain organizations such as B'nai Brith Canada that are known to collect statistics relating to hate crimes. The elusive nature of hate crimes means that perhaps more than any other form of criminality, a true picture of the prevalence of this crime can only be obtained by drawing upon diverse sources. These include criminal justice statistics (such as police data), victimization surveys, as well as sources outside the criminal justice system such as the periodic surveys conducted by B'nai Brith or the calls made to gay/lesbian assistance hotlines.