Mean Streets? - Policy Alternatives

Mean Streets? The Social Locations, Gender Dynamics, and Patterns of Violent Crime in Winnipeg 1

“The Mean Streets of Manitoba”(Winnipeg Free Press, July 23, 1998: A1).

Fuelled in part by newspaper headlineslike the above, public concerns over thelevels of violence in our communities haveintensified in recent years. Our perceptionand fear of violent crime has had a defi-nite impact on how many of us go aboutour daily lives. We warn our children notto talk to strangers. We avoid taking pub-lic transportation or venturing into certainparts of the city late at night. And we lockup our homes and install security devicesto keep out unwanted intruders. Eachtime the newspapers report yet anotherincident of violence on the “mean streets,”the calls for the criminal justice system to“get tougher” on violent crime becomelouder. These “tough” proposals usuallyinclude increased surveillance and moni-toring of public spaces, stricter law en-forcement practices and harsher prisonsentences for violent offenders.

Given our fears of violent crime,many of us take comfort in the notion thatour homes can provide a refuge from vio-lence. Yet, this image of the home as a ha-

ven has been shattered by the realizationthat assaults can occur in the privacy ofour houses as well as on the streets. Evenmore worrisome, the assailants are notstrangers, but people we know. The ideathat violence is a regular occurrence inmany homes has been made prominantby the women’s movement. Over the pastthree decades, feminists have organized,lobbied and advocated to put this issueon the political map. Key to their positionis that the victims of this violence arewomen and the offenders are the menwith whom they share an intimate rela-tionship; their husbands, boyfriends orcommon law partners. Estimates of theincidence of male violence against womenare alarming. The most recent and com-prehensive study found that one in fourCanadian women have experiencedphysical or sexual violence at the handsof a male partner (Statistics Canada, 1993).Such findings have produced a sea changein how we understand the problem of vio-lence. Once considered a private matter

Mean Streets?The Social Locations, Gender Dynamics, and

Patterns of Violent Crime in Winnipeg

by Elizabeth Comack, Vanessa Chopyk and Linda Wood

2 Canadian Centre for Policy Alternatives–Manitoba

between husband and wife and deemedto be “not the law’s business,” wife assaulthas been transformed into a public issuemeriting criminal justice response.

Feminists attribute the prevalence ofmale violence against women to the privi-lege, power and sense of entitlement thatmen are granted in a patriarchal society.Advocating for reforms in the criminaljustice system and the provision of serv-ices for abused women has been part of awider effort to realize women’s inequal-ity in both the public and private spheres.But this effort has been met with resist-ance by those who dispute the feministclaim that violence in the home is thehandiwork of men. In her book, When SheWas Bad: Violent Women and the Myth ofInnocence, journalist Patricia Pearson(1997) argues not only that “women areviolent, too” but that their violence canbe just as “nasty” as men’s: women beatup on their lovers, arrange for their hus-band’s murder, kill their babies and com-mit serial murders in hospitals andboardinghouses. Pearson draws supportfor her argument from studies which uti-lize the Conflict Tactics Scale (CTS) tomeasure abuse in intimate relationships.

Developed by American researcherMurray Straus (1979), the CTS is a quan-titative instrument which consists ofeighteen items and measures three differ-ent ways of handling interpersonal con-flict in intimate relationships: reasoning,verbal aggression and physical violence.The items are categorized on a continuumfrom least to most severe (for example,“discussed an issue calmly” and “cried”to “threw something,” “hit with a fist” and“used a knife or a gun”). Respondents ina survey are asked how frequently theyperpetrated each act in the course of con-

flicts or disagreements with their partnerswithin the past year and how frequentlythey had been on the receiving end. Theseself-reports of perpetration and victimi-zation are then used to construct estimatesof the rate of violence used by male andfemale partners. Most researchers whohave employed the CTS have foundequivalent rates for women and men onboth minor and severe types of violence(Straus and Gelles 1986; Steinmetz 1981;Brinkerhoff and Lupri 1988; Kennedy andDutton 1989). Such findings have led tothe conclusion that there is a sexual sym-metry in intimate violence; that is, thatwomen are just as violent as men.

Despite its popularity, the CTS hasnot been without its critics (see, for exam-ple: DeKeseredy and MacLean 1998;DeKeseredy and Hinch 1991; Dobash,Dobash, Wilson & Daly 1992; Schwartzand DeKeseredy 1993; Johnson 1996).Writers have noted that the CTS is an in-complete measure of intimate violence fora number of reasons: it measures only in-cidents of violence and thus ignores thesocial context of the violence (such aswhether a woman is acting in self-de-fence); it situates items only in the con-text of settling quarrels or disputes andthus misses assaults that “come out of theblue” or are motivated by the desire tocontrol another person; it relies on self-reports of violence and may thereby un-derestimate the incidence of violence bymales (who have been found to be morelikely to under-report); it fails to makeadequate distinctions between the sever-ity of different forms of violence (for ex-ample, “tried to hit with something” isdefined as “severe” while “slapped” isdefined as “minor”); and it does not cap-ture the outcome of the violence (for ex-


ample, the degree of injury incurred bythe participants).

Pearson (1997: 121) argues that thesemethodological concerns amount to un-warranted attacks by battered women’ssupporters who are invested in a genderdichotomy of “men as evil” and “womenas good.” In the same fashion, DonnaLaframboise (1999: A1), in a National Postarticle titled “Women are Men’s Equals inViolence,” tells us that “a good deal ofwhat we’ve been told about domesticabuse over the last 25 years is wrong” andthat studies conducted by “researcherswithout a political axe to grind” are moretrustworthy, valid and objective than threedecades of feminist-inspired work.

Anyone who follows the mediaclosely would be justified in their confu-sion over the mixed messages found incommentaries and reports on violentcrime. Does stranger violence pose thebiggest threat to our safety and security?Have feminists been leading us astray? Isviolence in the home the handiwork ofmen or are women “men’s equals” in vio-lence? How we resolve this confusion willdepend very much on the kinds of datawe bring to bear on the issue. One largelyuntapped data source is Police IncidentReports (PIRs) on violent crime.

Police Incident Reports

Criminologists have long recognizedthat official sources of information oncrime are limited by their nature andscope. For instance, the actual number ofcrimes that occur in a given area are largerthan those that are reported to police, andthose events that are subsequently re-corded in police reports are larger innumber than those that end up in court

and prison records. As well, given thepurpose of most official documents oncrime— to establish the presence of le-gally-relevant factors that would establishan act or event as a criminal matter—thepredominant standpoint or interpretationreflected in these accounts may well dif-fer from those of the participants in theevent.

Nevertheless, Police Incident Re-ports do offer a number of advantages asa source of information on violent crime.Since the police are the first point of entryinto the criminal justice system, policerecords will be more inclusive than thosemaintained by the courts or correctionalagencies. While many crimes (especiallyless serious ones like property offences)are not reported to police, events involv-ing violence have a greater likelihood ofpolice intervention and thus will be in-cluded in their records. Also significant isthat with the transformation of intimateviolence from a private trouble to a pub-lic issue, police policies and practices havebeen revised to focus more attention ondomestic violence. As a result, mandatorycharging and zero-tolerance policies haveopened a window into the private sphereof the home, making violence betweenpartners much more visible. With this in-creased police attention has come in-creased documentation as to what goes onwhen violence occurs between intimates.

To the extent that Police Incident Re-ports now include documentation on do-mestic violence, they also hold a numberof advantages over the Conflict TacticsScale for measuring women’s and men’sviolence. For one, while CTS research re-lies on respondents’ self-reports of perpe-tration and victimization, PIRs are basedon a variety of information sources, in-


cluding statements from complainants,accused and bystanders or witnesses.Since they draw from a number of ac-counts, PIRs offer a potentially richersource of information about the nature ofwomen’s and men’s violence than self-reports (Sacco and Kennedy 1998: 223).

Second, because PIRs record eventson a case-by-case basis and are normallycollected closer in time to the actual event,they do not encounter the problems ofrecall in CTS research, whereby respond-ents are asked to describe events whichoccurred over the previous twelvemonths.

Third, while CTS research ap-proaches the issue of intimate violencefrom the vantage point of “settling dis-putes” in marital relationships, PIRs aregoverned by the requirement to establishevidence (physical or otherwise) that acriminal act has occurred. Consequently,police officers include fairly detailed ac-counts in their reports of “who did whatto whom.” These descriptions provide arecord of the specific violence tactics usedby each of the participants in an event.

Fourth, criminal law distinguishesbetween violent offences on the basis oftheir “seriousness” (for example, Assault,Assault Causing Bodily Harm and Aggra-vated Assault). One important measureof seriousness is the harm or injury whichresults. This makes PIRs a good source ofinformation for measuring the degree ofinjury incurred by the participants in aviolent event—a variable missing in CTSresearch.

A final benefit of PIRs, especially forour purposes, is that they allow for a moreall-encompassing picture of the violenceengaged in by men and women. In addi-tion to capturing the violence that occurs

between intimates, we are able to broadenthe investigation to explore the other so-cial contexts in which violence occurs.

What, then, do Police Incident Re-ports tell us about the social locations andthe social relationships in which violenceoccurs? When men and women engage inviolence, do they differ in terms of the vio-lence tactics they employ or the harm theyinflict? To answer such questions, datawere gathered from City of Winnipeg Po-lice Incident Reports on violent crime.

Violent Crime in Winnipeg

Winnipeg offers a prime location forexploring men’s and women’s violence.According to official crime statistics, therate of violent crime in Winnipeg is oneof the highest of the nine largest Canadiancities. In 1995, for instance, Winnipeg hada rate of 1,198 incidents per 100,000 citi-zens for violent crime, second only to Van-couver (Canadian Centre for Justice Sta-tistics, 1995). Women make up a largerproportion of adults charged with violentoffences in Winnipeg than in the countryas a whole.1 It is also significant that Win-nipeg has been the site of a number ofcriminal justice initiatives pertaining todomestic violence.

In response to a national initiativein 1983, the Manitoba Attorney Generalissued a formal directive whereby policewere to charge all reported cases of assaultwhen there were reasonable and probablegrounds that an offence had occurred, re-gardless of the relationship between thevictim and the accused. Previously, it hadbeen left to the victim to initiate the com-plaint. The net effect of this mandatorycharging policy was to increase thenumber of arrests involving abusive part-


ners.2 A Family Violence Court was es-tablished in Winnipeg in September of1990 3 and, in response to a Domestic Vio-lence Review (Pedlar 1991), the WinnipegPolice Service introduced a zero-tolerancepolicy in July of 1993. Under this more rig-orous protocol, police are mandated to laycharges any time complaints have beenmade, regardless of the presence (or ab-sence) of corroborating evidence. The de-cision is then left up to the Crown attor-ney as to whether the case will proceed tocourt.

When we examine police records ofviolent crime, one factor rings clear: evenin an era of zero-tolerance, men are far morelikely to appear as accused persons thanwomen. Of the 23,090 charges laid for vio-lent crimes in the City of Winnipeg be-tween 1991 and 1995, women accountedfor only 15 percent of persons charged.While the quantitative difference betweenmen’s and women’s violence is obvious,what of the presence of qualitative differ-ences?

In order to explore qualitative dif-ferences between women’s and men’s vio-lence, data were gathered from WinnipegPolice Incident Reports on cases of womenand men charged with violence offences(assaults, robbery, sexual offences, andmurder/manslaughter) over a five yearperiod (1991 to 1995). A random sampleof women (stratified by offence type) gen-erated 501 cases or 15 percent of thosecharged. A sample of men (stratified byoffence type) charged over the same pe-riod was drawn for comparative pur-poses, yielding 501 cases or 2.5 percent ofthose charged. Information was collectedon: the social characteristics of the ac-cused; the social setting or location inwhich the violent event occurred; the re-lationship between the accused and thecomplainant or victim; and the outcomein terms of the injuries incurred and thecriminal justice response. Because wewere interested in the forms and severityof the violence used by women and men,each PIR was coded according to the vio-

Table 1: Nature of the Relationship Between the Accused and the Complainant

stranger 69 13.8% 90 18.0% 159 15.9%

friend /acquaintance 112 22.4% 95 19.0% 207 20.7%

family member 68 13.6% 21 4.2% 89 8.9%

(ex) partner 243 48.5% 285 56.9% 528 52.7%

police 9 1.8% 10 2.0% 19 1.9%

Total 501 100% 501 100% 1002 100%

Women Accused Men Accused Total


lence tactics employed during the courseof the event 4 as well as the resulting inju-ries incurred by the participants.

To present the findings of the study,the data have been organized in terms ofthe main social relationships in which vio-lence occurs: strangers; friends and/or ac-quaintances; family members; and inti-mate partners (see Table 1)5. Given thepublic’s pre-occupation with stranger vio-lence, we begin with an examination ofthose cases.

Stranger Violence: The “MeanStreets” of Manitoba?

What’s a guy gonna do? There’s a reces-sion, no jobs. I’m broke. I ain’t gonna starve.A guy needs cash to eat. (Case #661; Maleaccused charged with robbery)

I’ll be honest with you. I feel that I shouldbe able to walk the streets and back lanes andnot have to look at queers. Those queers triedto hit on a buddy of mine. (Case #856; Maleaccused charged with assault)

Okay, I did that one, too, but that’s it after[the video store robbery]. I didn’t do no more.The girl ripped off my balaclava. It’s nice tosee women take a stance against violence. Toobad it was on me. (Case #866; Male accusedcharged with robbery)

The public’s perception of crime hasbeen largely built upon fears of encoun-tering violence in the street late at night.These fears easily conjure up images of“dark strangers”—especially men—lurking in doorways and back alleys, wait-ing to do us harm.

While violence against strangers

does appear in the police reports, it is farless prevalent than newspaper headlinesmight have us believe. Of the 1,002 casesin our random sample, only 16 percent(159) involved violence between individu-als not known to one another. One-thirdof these events did conform to the pub-lic’s image of violence “on the streets,” as29 percent occurred in streets or lanewaysand another 4 percent happened in park-ing lots, parks or secluded areas. Yet, al-most half (47 percent) of the events involv-ing individuals not known to one anotheroccurred either at or near a drinking es-tablishment (15 percent of stranger events)or other commercial establishments suchas gas stations or stores (32 percent ofstranger events).6 The complainants inthese events were most often bouncers, se-curity guards and store employees.

Stranger violence was almost aslikely to happen during the daytime (29percent) as in the evening (32 percent) orlate at night (39 percent). As well, theprevalent image of the “dark malestranger” is one that does not hold upunder scrutiny of the data, as the major-ity (50 percent) of the men charged withviolence against strangers were Caucasian(like the 19-year-old homophobic manquoted above). While it is now widelyacknowledged, for example, that Aborigi-nal people are over-represented in thecrime statistics relative to their numbersin the general population, they are morelikely to be charged with violence againstmembers of their own communities, es-pecially intimate partners and other fam-ily members.

Of the different types of social rela-tionships in which violence can occur,stranger violence is the one most likely tobe a group-based activity (28 percent of


these events involved more than one com-plainant and half involved co-accused).Men accused who engaged in strangerviolence were most likely to do so ingroups with other men (93 percent of theirco-accused were male), and the vast ma-jority of their complainants (72 percent)were also male. While stranger violencemade up a smaller proportion of the casesinvolving women accused in the sample(14 percent for women versus 18 percentfor men), women who engage in strangerviolence are most likely to participatealongside men in mixed-sex groups andtheir victims are as likely to be female asmale. For both women and men accused,the violence used against strangers is mostlikely to involve tactics like “pushing,shoving, grabbing or pulling” (32 percentfor women versus 33 percent for men) and“property damage or theft” (49 percent forwomen versus 44 percent for men).

Even though stranger violence con-jures up the greatest fears for our physi-cal safety, it is actually the one form of vio-lent event of all the categories we exam-ined that has the least frequency of inju-ries to complainants. Complainants inevents involving strangers received noinjuries in one-third of the cases and, ofthose injured, the majority were of a“minor”nature (involving cuts, scratches,bruises and the like). Perhaps one of themain reasons why stranger violence re-sults in fewer and less serious injuries tothe victims is the motive behind many ofthese incidents: 37 percent of the chargeslaid by police in stranger events were forrobbery. This suggests that perpetratorsof stranger violence are not so much in-tent on inflicting injury or harm as theyare on stealing property (whether it bemoney, NIKE sports jackets or, in several

cases cited in the reports, cases of beer).Of the 1,217 complainants cited in

the 1,002 Police Incident Reports we stud-ied, the accused and the complainant wereknown to each other 78 percent of thetime. Rather than stranger violence com-mitted on the “mean streets,” the vastmajority of these events occurred in pri-vate dwellings between individuals whoare known to one another. Indeed, morefrequent than violence against strangerswere incidents of violence betweenfriends and/or acquaintances.

Violent Encounters BetweenFriends and/orAcquaintances

He was supposed to stay away from mydaughter and my house. He has her threat-ened at school; had his cousins steal her bike;call her a whore and then he’s standing acrossthe street making like he’s shooting her with apistol. He rapes my 10 year old daughter andhe’s free to go around doing this! (Case #431;Female accused charged with assault; themale complainant had been charged withsexual assault)

Contrary to the public’s perception ofviolent crime, we are more likely to en-counter violence from friends and/or ac-quaintances than we are from strangers.Violence against friends and/or acquaint-ances accounted for 21 percent (207) of thecases we studied in the Police Incident Re-ports. These events differed from strangerviolence in a number of significant ways.

While stranger violence is mostlikely to occur in public spaces, violentevents involving friends and/or acquaint-ances are characterized by their more pri-


vate nature: 57 percent of the incidents in-volving women accused and 72 percentof those involving men accused occurredin private dwellings. Friend/acquaint-ance events were also less likely to be agroup-based activity than stranger events(as only 22 percent involved more thanone complainant and 36 percent involvedco-accused). Like stranger violence, menaccused were most likely to aggressagainst other men, but with an importantexception: 23 percent of the chargesagainst men in the friend/acquaintancecategory were for sexual assaults againstfemales. (By comparison, sexual assaultsrepresented only 4 percent of the men’scharges in the stranger events.) As in theirviolence against strangers, women’s co-accused are as likely to be male as female.However, their violence is more often di-rected against female friends and/or ac-quaintances (64 percent of women’s com-plainants were female).

Whereas stranger violence was mostlikely to involve violence tactics like“pushing, shoving, grabbing or pulling”and “property damage or theft,” violencebetween friends and/or acquaintancestakes on a more serious tone. In additionto being “pushed, shoved, grabbed orpulled” (30 percent for women versus 36percent for men) complainants in eventsinvolving friends and/or acquaintancesare more likely to be “repeatedly punchedor beaten” by both women and men ac-cused (32 percent for women versus 33percent for men). Yet, there are also gen-der differences in the violence tactics usedagainst friends and/or acquaintances.Women accused were four times morelikely than men to “pinch, bite, scratch,or poke” (17 percent for women versus 4percent for men), eight times more likely

to engage in “hair pulling or cutting” (33percent for women versus 4 percent formen), and twice as likely as men to “kickor knee” (41 percent for women versus 21percent for men). Men, on the other hand,were twice as likely to “utter threats” (32percent for men versus 16 percent forwomen) than women accused. Men werealso far more likely to engage in “sexualassaults” (25 percent for men versus 3percent for women) and “sitting on or re-straining” (25 percent for men versus 6percent for women). These latter two vio-lence tactics are a reflection of men’s useof sexual violence when their victim isfemale.

That the violence against friendsand/or acquaintances is more seriousthan what we found in stranger violenceis also reflected in the resulting injuries.Whereas injuries were reported in 68 per-cent of the cases involving stranger vio-lence, injuries were sustained by com-plainants in 91 percent of the friends/ac-quaintances events. While the majority (60percent) of women’s complainants in-curred minor injuries, 60 percent of men’scomplainants received moderate andmajor injuries (such as broken bones, lac-erations, lost teeth or internal injuries).Corresponding with the harm inflictedduring the violent event, women and menwho engaged in violence against friendsand/or acquaintances were also likely toface more serious criminal charges: 29percent of women accused and 25 percentof men accused were charged with eitherAssault Causing Bodily Harm or Aggra-vated Assault. In contrast, only 16 percentof women and men involved in strangerviolence were charged with these offences.

That men and women are morelikely to be charged with violent crimes


against their friends and/or acquaint-ances runs counter to our perception ofviolence as stranger-related. This findingbecomes even more noteworthy when weconsider that acts of violence againststrangers are known to have a greater like-lihood of being reported to police, andthus being recorded in Police IncidentReports. Of special concern is the sever-ity of the violence that occurs in encoun-ters between friends and/or acquaint-ances—both in terms of the seriousnessof the violence tactics used and the inju-ries incurred by victims.

With regard to the gender dynam-ics involved when violence occurs, wom-en’s capacity for violence—especiallyagainst other women—is evident in thesedata. What is also apparent is that men’sviolence is not confined to their conflictswith each other: male violence againstwomen (in the form of sexual assaults) isalso a matter of concern. These genderpatterns become even more evident whenwe examine the violence that occurswithin the family setting.

Violent Encounters withFamily Members

While violence between intimate part-ners will be considered in the next section,our interest here is in understanding thenature of the violence that occurs betweenother family members (children, parents,siblings, extended family members andin-laws). These incidents represented only9 percent (89) of the cases we examined,although events involving other familymembers made up a greater proportionof women accused’s charges than they didfor men accused (14 percent for women

versus 4 percent for men).Violence against other family mem-

bers begins to look quite different fromwhat we saw with those events involvingstrangers or friends/acquaintances. Onedifference is the social location where theevent occurs, as family violence is mostlikely to occur within the private sphereof the home: almost all (91 percent) ofevents involving women accused and allof those involving men accused occurredin private dwellings. Another differenceis that family events primarily involve onecomplainant (77 percent for women ver-sus 90 percent for men) and only one ac-cused (88 percent for women versus 100percent for men). Children were the mostlikely victims for both women and menaccused (representing 41 percent and 48percent of women’s and men’s complain-ants, respectively). Siblings—especiallysisters—wer e the second most likely vic-tims for women accused. Extended fam-ily members or in-laws were the secondmost likely victims for men.

While men’s violence in the other re-lationships we have examined is mainlydirected at other men, their violenceagainst other family members is prima-rily directed at females. Despite the factthat men’s victims are most often femalechildren and relatives, they do not appearto shy away from using more severe vio-lence tactics: men accused not only alleg-edly “push, pull, shove or grab” theircomplainants (38 percent of the cases),they also “sit on or restrain” (29 percent),“beat or repeatedly punch” (24 percent)and “slam bodies or head into something”(24 percent). Also significant is the preva-lence of sexual violence: men accusedsexually assaulted their complainants in38 percent of the cases involving other


family members.Whereas men’s violence in the fam-

ily setting is primarily inter-gendered(male/female), women’s violence is morelikely to be intra-gendered, or directed atother females. While women accused arefive times more likely than men to engagein “hair pulling or cutting” (25 percent forwomen versus 5 percent for men) and to“pinch, bite or scratch” their complainants(21 percent for women versus 5 percentfor men), they are just as likely as men to“push, pull, shove or grab” (35 percent)and “beat or repeatedly punch” (27 per-cent) other family members. To this extent,women’s violence against other familymembers looks very similar to what wesaw with friends and/or acquaintances.

While there are some notable differ-ences in the social dynamics involved inmen’s and women’s violence, the fact thatwomen are engaging in violence tactics,sometimes with similar frequency as men,might be taken as evidence that they are“men’s equals” in violence. Nevertheless,we do not yet have the complete picture.Indeed, over half (53 percent) of the casesin the Police Incident Reports involvedviolence between intimate partners (hus-bands and wives, boyfriends and girl-friends, common-law partners and gayand lesbian partners). What remains to beseen, then, is whether women are “men’sequals” when it comes to partner violence.

Encounters Between IntimatePartners:7 Equals inViolence?

We both know that you deserve this . . .You’re my baby, you’re mine, all mine. You’remy property. (Case #672; Male accused

charged with sexual assault with aweapon)

You better lock me up, man, cause if youlet me out I’m gonna do something bad to her.(Case #727; Male accused charged withassault causing bodily harm)

Next time I’ll kill the bitch, before I callyou guys, I’ll just kill her. (Case #735; Maleaccused charged with assault causingbodily harm)

I did it because I’m tired of getting beatenup. (Case #156; Female accused chargedwith assault)

That violence between intimates hasattained the status of a criminal matter isclearly reflected in the Police IncidentReports. Partner violence made up 49 per-cent (243) of the cases involving womenaccused and 57 percent (285) of those in-volving men accused in our study. Savefor three cases involving gay or lesbianpartners, violence between partners wasinter-gendered (male/female; female/male). Even more so than family events,partner violence was likely to be a one-on-one event; 93 percent of cases involv-ing women accused and 89 percent ofthose involving men accused had onecomplainant and almost none ( 2 percentversus .4 percent) involved co-accused.

Studies that use the Conflict TacticsScale have concluded that a sexual sym-metry exists in intimate violence: men areas likely as women to be victims of abuseand women are as likely as men to be per-petrators of both minor and serious actsof violence. A different picture emerges,however, when we examine five indica-tors found in the police reports: 1. the so-


cial location of the event; 2. the violencetactics used by men and women accused;3. the use of violence by complainants; 4.the degree of injury or harm inflicted; and5. the question of “who called the police?”

1. The Social Location of the Event

Partner violence is most likely to oc-cur in private settings (see Table 2). In half(51 percent) of these cases, the violence oc-curred at a residence shared by the ac-cused and the complainant. Not all of thevictims were co-habiting with the accusedat the time of the incident. Some were girl-friends or boyfriends and some had endedtheir relationship with the accused. Inthese latter cases, there is often a restrain-ing or “no contact” order against the ex-partner.

There were 176 cases where violenceoccurred between men and women whowere not co-habiting. Of these cases, theviolent event occurred at the woman’sresidence 64 percent of the time. This find-ing would suggest that men may play agreater role in initiating the violence, in

that they are the ones journeying to theresidence of their (ex)partner, and vio-lence subsequently breaks out. Of the 48charges laid by police for Breach of Re-cognizance (eg. for violation of a no-con-tact or communication order) in eventsinvolving partner violence, 75 percent (36)were laid against men.

2. Violence Tactics Used by Men andWomen Accused

One of the ways to determine ifwomen are “men’s equals” in violence isto gauge whether there are statisticallysignificant differences in men’s and wom-en’s use of violence tactics. As we sawpreviously, men and women who en-gaged in stranger violence were mostlikely to push, pull or grab complainantsand to engage in property damage ortheft. There were no statistically signifi-cant differences in the violence tacticsused by men and women in strangerevents. 8 It was in events involving friendsand/or acquaintances where some gen-der differences in violence tactics became

Table 2: Social Setting or Location of Partner Events

private residence 221 91% 264 93% 485 92%

a. complainant’s 41 17% 82 29% 123 23%

b. accused’s 31 13% 22 8% 53 10%

c. both 127 52% 144 50% 271 51%

d. other 22 9% 16 6% 38 7%

Public Place 22 9% 21 7% 43 8%

Total 243 100% 285 100% 528 100%

Location Women Accused Men Accused Total


evident: of the violence tactics used bywomen and men accused, six showed sta-tistically significant differences.9 Only twoviolence tactics – “sitting on or restrain-ing” (6 percent for women versus 29 per-cent for men) and “sexual acts” (7 percentfor women versus 38 percent for men)—wer e statistically significant in events in-volving other family members. When weexamine the violence tactics used by menand women accused in events involvingpartners, however, gender differences be-come much more acute: there were nineitems where statistically significant differ-ences existed; seven of these were at the.001 level.

All of the violence tactics used bywomen and men accused in partnerevents are reported in Table 3. In terms offrequency, men accused utilize more vio-lence tactics than women accused (2.9 ver-sus 2.3 per person). With regard to thenature of the violence tactics used, menaccused were almost twice as likely aswomen to “push/pull/grab.” When com-bined with tactics such as “sitting on orrestraining,” “slamming body or headinto something,” and “strangling or chok-ing,” the picture which emerges is one ofmen using their physical strength or forceto overpower their female partners. Thegreater use of physical force by male ac-cused is reflected in the charges laid bypolice: men were twice as likely as womento be charged with Assault Causing Bod-ily Harm (19 percent for men versus 10percent for women).

On the other hand, women accusedare almost six times more likely than mento “pinch/bite/scratch/poke.” They arealso more likely to “hit with or throwsomething” and to “stab or slash.” Thepicture which emerges here is one of

women—lacking comparable physicalstrength or force to their male partners—r esorting to the use of objects or weap-ons during the course of a violent event.

Indeed, women accused were morelikely than men to use weapons in theirencounters with intimate partners (42 per-cent for women versus 28 percent formen). The weapon of choice for manywomen appears to be beer bottles andother “sharp objects” such as knives orscissors. These objects accounted for 66percent of the weapons used by womenin partner events (versus 51 percent formen) (Table 4). Men accused used firearmsin three of the cases, and were twice aslikely as women to use “blunt objects” likebats, wooden boards, hammers or metalpipes (19 percent for men versus 9 per-cent for women).

What is especially interesting inthese findings is the considerable rangeof “other” articles cited in the police re-ports (which accounted for 43 percentversus 40 percent of the weapons used bywomen and men respectively). “Otherobjects” included things like: bathroomplungers, telephones, wooden spoons,ashtrays, coffee pots, dishes, books, hairbrushes, TV remote controls, runningshoes, pizza boxes, clothes baskets andhair gel bottles. Taken together, this listsuggests that virtually any object whichis readily accessible in a domestic settingcould potentially be thrown or used to hitwith during the course of a violentevent—and thus subsequently come tobe defined as a “weapon” in Police Inci-dent Reports.

One consequence of the more fre-quent use of these “weapons” on the partof women accused is that when police arecalled to intervene, women are more likely


Table 3: Violence Tactics Used by the Accused in Partner Events

uttering threats 33 14% 63 22%

property damage/theft* 23 9.5% 51 18%

push/pull/grab/shake/ 68 28% 147 52%

pinch/bite/scratch/ 67 28% 14 5%

pull or cut hair 40 16.5% 47 16.5%

burn or spray 5 2.1% 4 1.4%

hit with/threw something ** 79 32.5% 51 18%

slapping 54 22% 67 23.5%

kicking/kneeing 47 19% 57 20%

sitting on/restraining** 4 2% 43 15%

punched 55 23% 72 25%

beat (repeated punching) 44 18% 62 22%

slamming head or body 6 2.5% 75 26%

sexual acts* 0 11 4%

strangling/choking** 6 2.5% 39 14%

stabbing/slashing** 21 9% 6 2%

shooting 0 1 0.4%

other# 0 12 4%

no. of violence tactics 552 822no. of accused 243 (2.3 per person) 285 (2.9 per person)*p<.01 **p<.001

+ percentages add up to more than 100 due to multiple responses# This category includes actions such as driving a vehicle at or unleashing a dog on the complainant.

Type of Violence Women Accused Men Accused

elbow/wrestle*

poke **

into something**


than men to be charged with the moreserious offence of Assault with a Weapon;32 percent of women’s violent crimecharges in the partner category involvedAssault with a Weapon compared withonly 11 percent of men’s charges. Anotherconsequence is that so long as ConflictTactics Scale researchers collapse all ofthese objects into one category, then “triedto hit with something” (which is definedas “severe” according to the scale) willinclude a woman who throws a laundrybasket or a TV remote control at her part-ner alongside a man who wields an ironbar or a hammer.

3. Use of Violence byComplainants

While women and men accused differin their use of violence tactics in their en-counters with intimate partners, what ofthe use of violence by their complainants?Conflict Tactics Scale researchers havedescribed the violence which occurs be-tween men and women in intimate rela-tionships as “mutual combat.” When we

examined the use of violence by complain-ants in partner events, however, we foundthat the female partners of men accusedused violence in only 23 percent of thecases. In contrast, male partners of womenaccused used violence in 65 percent of thecases.10

The violence tactics used by maleand female complainants tend to mirrorthe gendered differences we saw for maleand female accused. Male complainants,like male accused, utilize more violencetactics than female complainants (2.2 ver-sus 1.4 per person). Male complainants aremost likely to engage in “pushing, shov-ing or grabbing” (53 percent for men ver-sus 30 percent for women), “sitting on orrestraining” (17 percent for men versus 0percent for women), and “slamming bodyor head into something” (17 percent formen versus 3 percent for women). Womencomplainants, like women accused, aremore likely to “pinch, bite or scratch” (18percent for women versus 5 percent formen) and “hit with or throw something”(29 percent for women versus 10 percentfor men). 11 Such findings run counter to

Table 4: Weapons Used by Accused in Partner Events

firearm 0 0% 3 4%

beer bottle 28 27% 13 16%

sharp object 40 39% 28 35%

blunt object 9 9% 15 19%

other weapon 44 43% 32 40%

Total no. of weapons used 121 91

Total no. of accused using 103 (42% of accused) 80 (28% of accused)

Type of Weapon Women Accused+ Men Accused+


the depiction of violence in intimate rela-tionships as “mutual combat.”

4. The Degree of Injury or HarmInflicted

One of the factors which has beenmissing in Conflict Tactics Scale researchis that of “outcome,” in particular, thenature and extent of the injuries incurredby participants in a violent encounter. Inthe present study, 73 percent of women’scomplainants and 78 percent of men’scomplainants were injured during thecourse of the event. The majority of theinjuries (54 percent versus 64 percent)were classified as “minor” in nature andinvolved cuts, bruises, sprains, black eyes,bleeding noses and hair loss. However,when we investigated whether the ac-cused person incurred injury, we foundthat almost one-half (48 percent) of thewomen accused—as opposed to only 7

percent of the men accused—wer e in-jured during the course of the event.While the majority of these injuries wereof a “minor” nature, this finding addsweight to the conclusion that violentevents between women and men are notsymmetrical.

5. Who Called the Police?

Finally, one of the variables often in-cluded in Police Incident Reports is “whomade initial contact with the police?”Calls to the police can be interpreted as aform of “help-seeking behaviour” on thepart of someone in trouble. In incidentsinvolving partners, the complainants werethe ones to call the police in 77 percent ofthe cases involving a male accused (ver-sus 40 percent for women accused). Incontrast, however, it was the accusedwoman who called the police in 35 percentof those cases (versus only 5 percent for

Table 5: Who Made Initial Contact with the Police in Partner Events?

Complainant 89 39.7% 220 77.2%

Accused 68 30.4% 13 4.6%

Both 10 4.5% 1 0.4%

Witness/Bystander 42 18.8% 46 16.1%

Police Happened By 15 6.7% 5 1.8%

Total 224* 100% 285 100%

*There are 19 missing cases. p<.001

Women Accused+ Men Accused+


male accused) (Table 5). In more than one-third of the cases involving a woman ac-cused, then, she was the one who per-ceived the need for police intervention.Nevertheless, she also ended up beingcharged with a criminal offence.

More rigorous policies andprotocols that mandate police to laycharges in cases of domestic violence haveprovided us with a window into the do-mestic sphere that has previously beenunavailable. Through this window, we arenow able to investigate more fully thenature of the violence that occurs betweenintimate partners. When we do, we findthat—contrary to Conflict T actics Scaleresearch—while women ar e certainly ca-pable of violence, they are not “men’sequals.” In combination, the five indica-tors presented here—the social locationof the event, the violence tactics used bywomen and men accused, the use of vio-lence by complainants, the degree of harmor injury inflicted and who made the ini-tial call to the police—suggest that part-ner violence is asymmetrical.

In many respects, the violence tac-tics used by men and women reflect thegendered power differences that operatewhen violence occurs in intimate relation-ships. Male accused do not refrain fromusing brute force against their female part-ners: they push; they sit on; they strangle;and they slam bodies and heads. To thisextent, men’s violence in partner relation-ships conforms to the traditional “mascu-line script” premised on aggression anddominance (Kypers 1992; Stoltenberg1990; and Messerschmidt 1997). Com-ments made by women to the police sug-gest that they are very aware of their part-ner’s size and strength:

I don’t understand how you can assault a

guy that size. I don’t agree. I thought this wassupposed to protect the woman not the man.It was self-defence. (Case # 145; Female ac-cused charged with assault)

I don’t want to say anything, but I feel Ishould. I was defending myself. He has a blackbelt in judo and he uses it. He broke my cam-era. That’s how it started. (Case #146; Fe-male accused charged with assault)

That women are most likely to use so-called female tactics like pinching, bitingor scratching and that they resort to theuse of hitting with or throwing objects attheir male partners might be taken as evi-dence of self-defensive strategies in re-sponse to an aggressive partner. Thiswould add weight to the claims of femi-nist researchers (Dobash et. al. 1990) thatwomen’s violence in intimate relation-ships often occurs in the course of defend-ing themselves against their abusive part-ners. In any case, the net effect of thesegendered differences is that women arethe ones most likely to be injured whenviolence erupts in intimate relationships.

Responding to Violent Crime

Presenting the data on violent crimefound in Police Incident Reports—its so-cial locations, its social dynamics and thesocial relationships in which it occurs—has allowed us to appr eciate some of thepatterns of violence. These patterns canbe captured in the form of a chart (seeChart 1). On the left hand side of the chartfall those violent events that occur in pub-lic. These are most likely to be group-based, and to be acts of violence againststrangers. As we move from left to right,the social locations and social dynamics


shift along with the social relationshipsbetween participants. Events involvingfriends and/or acquaintances are lesslikely to occur in public spaces and to begroup-based, while those involving fam-ily members and intimate partners aremuch more likely to occur in privatedwellings and to be one-on-one incidents.

Our aim has not only been to mapthe patterns of violent crime, but also itsgendered nature. As the violence whichoccurs in the home has come under in-creasing scrutiny, the issue of whetherwomen are “men’s equals” in violence hasalso surfaced. In quantitative terms, menare far more likely to appear in Police In-cident Reports on charges of violent crimethan women. They represented 85 percentof those charged during the period of thestudy (1991 to 1995). However, when weexamined the qualitative differences be-

tween men’s and women’s violence somepatterns became evident (see Chart 2).Men’s violence against strangers andfriends/acquaintances tends to be intra-gendered—involving gr oups of men inaltercations with other men. As we movefrom the public to the private sphere,however, men’s victims are more likely tobe female—childr en, female relativesand intimate partners. Women’s violence,on the other hand, displays a differentgender dynamic. When women engage instranger violence, they are most likely todo so in the company of men and theirvictims are as likely to be male as female.Women’s violence against friends/ac-quaintances and family members, how-ever, is primarily directed toward otherfemales. Like their male counterparts,women accused are most often chargedwith violence against their intimate part-

Chart 1: Patterns of Violent Crime*

Social Location

Social Dynamics

Social Relationships

public (27%) ---------------------------------------------------------------------------- private (73%)

group-based (18%) ------------------------------------------------------------- one-on-one (82%)

stranger (16%) --------- friend/acquaintance (21%) ------ family (9%) -------intimate partner (53%)

* the 19 cases involving police officers as complainants have not been included here


strangers intra-gendered (male/male)

friends/acquaintances intra-gendered (male/male)

family members inter-gendered (male/female)

intimate partners inter-gendered (male/female)

Women’s Violence

strangers mixed sex-groups

friends/acquaintances intra-gendered (female/female)

family members intra-gendered (female/female)

intimate partners inter-gendered (female/male)

ners. What implications do these findingshold for our efforts to respond to violentcrime?

One of the key findings of this studyis that police reports reveal a contradic-tion in the public’s perception of violentcrime: while it is stranger violence that wefear most, we are much more likely to en-counter violence at the hands of peoplewe know, sometimes intimately. Thatstranger violence made up a relativelysmall proportion (16 percent) of violentcrime is, in large part, due to the impactof “get tough” policies like zero-tolerancewhich have resulted in an increasednumber of police reports involving do-mestic violence and a corresponding in-crease in the overall levels of violence re-ported in the media. In other words, oneof the main reasons why Winnipeg rankshigher than most other Canadian cities forviolent crime has to do not so much with in-

creased levels of violence in the “mean streetsof Manitoba,” but with the impact of a zero-tolerance protocol which mandates police tolay charges in cases of partner violence. 12

Stranger violence not only made upa smaller proportion of violent events. Bycomparison it was also less serious thanother violent events in terms of the vio-lence tactics used and the resulting inju-ries to victims. This is not to say that weshould ignore or downplay the incidenceof violence against strangers. But it doessuggest that we might re-think our strat-egies for responding to violent crime. Inparticular, the fears surrounding strangerviolence have produced some counter-productive strategies for reducing crimein the streets. So long as we “don’t talk tostrangers” or avoid venturing out intopublic spaces, we become dis-connectedfrom each other and leave our streetsempty and barren, thereby creating the

Chart 2: Qualitative Differences in Men’s and Women’s Violence

Men’s Violence Primary Gender Dynamics


conditions for stranger violence to occur.Installing lighting, cameras and emer-gency call-boxes will not fix this problem.It is only by actively revitalizing ourneighbourhoods and reclaiming the pub-lic spaces in our communities that thepotential for violence will be reduced.

What is more concerning when weexamine police reports on violent crimeis the prevalence and seriousness of vio-lence between friends and/or acquaint-ances. Both men and women were likelyto aggress against members of their samesex (with one important exception: menalso engaged in sexual violence againsttheir female friends and/or acquaint-ances). That individuals choose to resolvetheir conflicts with those they knowthrough violent means leads us to ques-tion the wider cultural messages in soci-ety that condone violence. And we do nothave to look very far. Whether it be TomCruise battling it out with his nemesis inthe latest Mission Impossible movie, “TheRock” going up against “The Undertaker”in a WWF bout or video games like Sol-dier of Fortune where the players can evis-cerate, decapitate, dismember and burntheir victims, the messages are prettyclear: violence has become normative. Tothe extent that these messages are directedat young men, we should not be surprisedwhen they begin to act out these scriptsin their own lives. Indeed, the police re-ports are replete with cases of young men“taking it outside” at the local bar to set-tle their differences with friends or ac-quaintances. There is also no reason tobelieve that women will be immune fromthe cultural messages that permeate oursociety, especially when those messagesherald violence as a means of “gettingyour way.” So long as violence is held up

as a resource to be used when confrontedwith problems and conflicts in our lives,then both men and women will be encour-aged to act out violently.

Our findings also indicate that thehome is certainly not a haven from vio-lence. Men commit sexual violenceagainst their children and other relatives,and women engage in violent acts againstfemale children and siblings. Neverthe-less, while both men and women have thecapacity for violence, the police reportsindicate that women are not “men’sequals,” especially when it comes to part-ner violence. Contrary to studies that usethe Conflict Tactics Scale, our findings tellus that men are more likely to use violencetactics that rely on the use of brute forceagainst their partners, whereas womenresort to using weapons or objects (likeTV remote controls or laundry baskets)during the course of an event and so-called “female tactics” (like pinching, bit-ing or scratching). Further, women’s com-plainants are much more likely than men’sto use violence, and women accused arealmost seven times more likely than mento receive injuries during the course of aviolent event. In these terms, our findingsconfirm the feminist claim that male vio-lence against women is a social problemdeserving of public attention and re-sponse. Nevertheless, some cautions arein order.

Changing police protocols repre-sented by the zero-tolerance policy of theWinnipeg Police Service have broughtmore men—and women—into thecriminal justice system on charges of vio-lent crime. Whereas partner violence rep-resented 43 percent of all men’s violentcrime charges in 1991, by 1995 it repre-sented 64 percent. An even more signifi-


cant change occurred for women. In 1991,23 percent of women accused’s chargeswere for partner violence. By 1995, it hadmore than doubled: 58 percent of all vio-lent crime charges against women in-volved partner violence. The vast major-ity – 80 percent of the charges laid againstwomen accused and 51 percent of thoselaid against men accused—wer e subse-quently stayed by the Crown.

While the underlying intent of thezero-tolerance policy was to assist victimsof domestic violence (who are predomi-nantly women), zero-tolerance hasopened the way for “double-charging” tooccur (whereby both partners are chargedwith a criminal offence when police arecalled in). Both the accused and the com-plainant were charged in 55 percent of thecases involving women accused and 10percent of those involving men. Stays ofproceedings were even higher in thosecases where double-charging occurred (88percent versus 70 percent respectively).

Findings such as these raise seriousquestions about the use of a zero-tolerancepolicy for responding to partner violence.At the very least, we need to examine themerits of a protocol that removes all dis-cretion from police in deciding whetheror not criminal charges are warranted ina particular case. At the same time, weneed to ensure that individuals are receiv-ing the help they require when they needit. One promising recommendation thatemerged from the public inquiry into themurder/suicide deaths of Rhonda andRoy Lavoie (Schulman 1997) is currentlyin the process of being implemented as apilot project by the Winnipeg Police Serv-ice. Police officers working in tandem withsocial workers will target high-risk do-mestic violence incidents and provide

support and direction to families in need(Winnipeg Free Press, November 7, 2000).There are a number of advantages to thisapproach.

Zero-tolerance resulted in signifi-cant increases in the workload of policeand Crown attorneys and a correspond-ing backlog of court cases which, moreoften than not, eventually ended with astay of proceedings. This meant that indi-viduals were less likely to get access tohelp at a time in their lives when theymost needed it. By instituting a crisis-in-tervention team to attend to the matter “asit is happening,” responders will be in abetter position to attend to the needs ofpeople in trouble by directing them toappropriate resources and supports in thecommunity. As well, whereas zero-toler-ance has led to an “over-criminalization”of families at risk, this new approach hasthe potential to direct cases away from thecriminal justice system and toward socialservice resources that might better ad-dress the root causes of domestic violence.In other words, such an approach has thepotential to ensure individuals’ physicalsafety and to connect them with resourcesin their community without over-extend-ing the reach of the criminal justice sys-tem into people’s lives.

The underlying causes of the violentcrime patterns described here are manyand complex. They connect to the urbandecay encountered by inner city commu-nities, the disappearing social safety net,the sense of alienation and anomie expe-rienced by marginalized groups, somemen’s sense of entitlement over thewomen in their lives, and broader culturalmessages that condone and normalizeviolence as a problem-solving tool. Giventhe depth and complexity of such causal


factors, is it reasonable to expect that oneset of institutions in society— the crimi-nal justice system—will be capable of al-leviating the violence? While “gettingtougher” on crime may seem to offer aquick and easy fix, the fact remains thatthe long-term sources for change lie in themore difficult and challenging task ofstrengthening our communities by pro-viding people with access to the ways andmeans for resolving their troubles in non-violent ways.


References Cited

Brinkerhoff, Merlin and Eugen Lupri (1988)“Interspousal Violence.” Canadian Journal of So-ciology. vol. 13 no. 4: 407-34.

Canadian Centre for Justice Statistics (1996)Uniform Crime Reporting Survey. Ottawa.

Canadian Centre for Justice Statistics (1995)“Canadian Crime Statistics.” Juristat vol. 16 no.10.

DeKeseredy, Walter and Ronald Hinch(1991) Woman Abuse: Sociological Perspectives.Toronto: Thompson Publishing.

DeKeseredy, Walter and Brian MacLean(1998) “‘But Women Do It Too’: The Contextsand Nature of Female-to-Male Violence in Ca-nadian, Heterosexual Dating Relationships.” inKevin Bonnycastle and George Rigakos (eds.)Unsettling Truths: Battered Women, Policy, Politics,and Contemporary Research in Canada. Vancouver:Collective Press.

Dobash, Russell P., R. Emerson Dobash,Margo Wilson and Martin Daly (1992) “TheMyth of Sexual Symmetry in Marital Violence.”Social Problems. vol. 39 no. 1 (February): 71-91.

Johnson, Holly (1996) Dangerous Domains:Violence Against Women in Canada. Toronto:Harcourt Brace.

Kennedy, Leslie and Donald Dutton (1989)“The Incidence of Wife Assault in Alberta.” Ca-nadian Journal of Behavioural Science. vol. 21: 40-54.

Kypers, Joseph A. (1992) Man’s Will to Hurt:Investigating the Causes, Supports and Varieties ofHis Violence. Halifax: Fernwood Publishing.

Laframboise, Donna (1999) “Men andWomen are Equals in Violence.” National Post.July 10, A1.

Messerschmidt, James (1993) Masculinitiesand Crime: Critique and Reconceptualization ofTheory. Maryland: Rowman and Littlefield Pub-

lishers.

Pearson, Patricia (1997) When She Was Bad:Violent Women and the Myth of Innocence. Toronto:Random House.

Pedlar, Dorothy (1991) The Domestic ViolenceReview into the Administration of Justice in Mani-toba. Winnipeg: Manitoba Justice.

Sacco, Vincent F. and Leslie W. Kennedy(1998) The Criminal Event: An Introduction toCriminology (2nd ed.).Toronto: Nelson Canada.

Schwartz, Martin and Walter DeKeseredy(1993) “The Return of the ‘Battered HusbandSyndrome’ Through the Typification of Womenas Violent.” Crime, Law and Social Change. vol.20: 249-265.

Schulman, Mr. Justice Perry W. (1997) AStudy of Domestic Violence and the Justice Systemin Manitoba. Report of the Commission of In-quiry into the Deaths of Rhonda and Roy Lavoie.

Statistics Canada (1993) “The ViolenceAgainst Women Survey.” The Daily. November18.

Steinmetz, Suzanne (1981) “A Cross-culturalComparison of Marital Abuse.” Journal of Soci-ology and Social Welfare. vol 8: 404-14.

Stoltenberg, John (1989) Refusing to be a Man:Essays on Sex and Justice. New York: Penguin.

Straus, Murray A. (1979) “MeasuringIntrafamily Conflict and Violence: The ConflictTactics (CT) Scales.” Journal of Marriage and theFamily. vol 41 no 1 (February): 75-88.

Straus, Murray A. and Richard J. Gelles(1986) “Societal Changes and Change in FamilyViolence from 1975 to 1985 as Revealed by TwoNational Surveys.” Journal of Marriage and theFamily vol. 48: 465-480.

Ursel, E. Jane (1998A) “Eliminating ViolenceAgainst Women: Reform or Co-optation in StateInstitutions.” in L. Samuelson and W. Antony(eds.) Power and Resistance: Critical Thinking


About Canadian Social Issues (2nd ed.). Halifax:Fernwood Publishing.

Ursel, E. Jane (1998B) “Mandatory Charg-ing: The Manitoba Model.” in Kevin Bonnycastleand George Rigakos (eds.) Unsettling Truths:Battered Women, Policy, Politics, and ContemporaryResearch in Canada. Vancouver: Collective Press.

Notes

1 In 1996, women made up 18 percent ofthose charged with violent offences in Winni-peg compared with 12.5 percent for Canada asa whole (Canadian Centre for Justice Statistics,1996).

2 For instance, the number of cases ofspousal assault where charges were laid in theCity of Winnipeg went from 629 in 1983 to 1137in 1989, an 80 percent increase in six years (Ursel,1998A).

3 Since that time, the number of casespassing through the court increased dramati-cally: from 1800 in 1990-91 to 2660 in 1991-92,3646 in 1992-93 and 4140 in 1993-94 (Ursel,1998B: 75).

4 This produced a list of eighteen differ-ent violence tactics which ranged in seriousnessfrom “uttering threats” to “shooting.”

5 Police were the main complainants in1.9 percent (19) of the cases studied. Because ofits small number, this social relationship is notexamined here.

6 Eleven percent of stranger events oc-curred in private dwellings, the remainder (9percent) occurred in cars, taxis or buses, at lei-sure centres, or in prisons and court.

7 Since violence does not always end oncea relationship is terminated (indeed, it some-times escalates), we include in this category in-dividuals who are currently in a relationship aswell as former or “ex” partners. For the sake ofbrevity, however, the term “partner” will be usedto refer to these relationships.

8 Statistical significance is defined hereas p < .01 (i.e. less than one percent of the timeor one time in 100 would the observed differ-ences be the result of chance).

9 Three of these tactics— “pulling or cut-ting hair” (33 percent for women versus 4 per-

cent for men), “sitting on or restraining” (6 per-cent for women versus 25 percent for men) and“sexual acts” (3 percent for women versus 25percent for men) were significant at the .001level.

10 This finding was significant at the.001 level.

11 The gender differences for each ofthese violence tactics were statistically signifi-cant at the .01 level or higher.

12 For example, in 1999 the WinnipegPolice Service made 3,500 arrests in cases in-volving domestic violence. That same year, po-lice in Edmonton – a city of comparable sizebut with no zero-tolerance policy – made 889domestic violence arrests (Winnipeg Free Press,July 7, 2000: A1).


Mean Streets?

The Social Locations, Gender Dynamics, and Patterns ofViolent Crime in Winnipeg

by Elizabeth Comack, Vanessa Chopyk and Linda Wood

December, 2000

ISBN 0-88627-252-1


Elizabeth Comack is a Professor of Sociology at the University of Manitoba. Her pre-vious publications include Locating Law: Race/Class/Gender Connections (1999; editor),Women in Trouble: Connecting Women’s Law Violations to Their Histories of Abuse (1996),The Feminist Engagement with the Law: Legal Recognition of the ‘Battered Woman Syn-drome’ (1993) and The Social Basis of Law: Critical Readings in the Sociology of Law (1991 &1986; co-editor with S. Brickey).

Vanessa Chopyk is a graduate student in the Department of Sociology at the Univer-sity of Manitoba. Her Master’s thesis research explores the issue of gender differen-tials in sentencing for violent crime.

Linda Wood is a graduate student in the Department of Sociology at the University ofManitoba. Her Master’s thesis research examines the impact of zero-tolerance policyon domestic violence.

This research was made possible by a grant from the Social Sciences and HumanitiesResearch Council (SSHRC). The authors would like to thank a number of individualsfor their generous support during the different phases of the project, especially WayneAntony, Ken Beiner, Norma Danylyshen, Murray Kull, Glen Lewis, Rick Linden, ToddScarth and Margaret Shaw. It goes without saying that any shortcomings of this workremain the responsibility of the authors.


Mean Streets?The Social Locations, Gender Dynamics, and

Patterns of Violent Crime in Winnipeg

H I G H L I G H T S

Main Findings

Stranger Violence

Contrary to the public’s perception,stranger violence represented only 16 per-cent of the charges laid by the WinnipegPolice Service for violent crime. Whileone-third of these events occurred “on thestreets,” almost half occurred either at ornear a drinking establishment or othercommercial enterprise. The complainantsin these latter events were most likely tobe bouncers, security guards and storeemployees. The event was as likely to oc-cur during the daytime as it was duringthe evening or nighttime. Rather than“dark strangers,” Caucasian males werethe ones most likely to be involved instranger violence. Compared with theother three types of social relationships westudied, stranger violence was most likelyto be a group-based activity. However, itinvolved less serious violence tactics andresulted in the least frequency of injuriesto complainants. Of those injuries, themajority were minor. The main motivebehind stranger violence was theft ofproperty.

Friend/Acquaintance Violence

Violence involving friends and/or ac-quaintances was not only more prevalentthan stranger violence (occurring in 21

Fuelled by newspaper headlines like “TheMean Streets of Manitoba,” public concernover levels of violence in our communi-ties has intensified in recent years. Thecommon perception of violent crime isthat it is the work of strangers out on thestreets at night who are waiting to do usharm. Meanwhile, feminists have drawnattention to the violence that women en-counter in their own homes, usually at thehands of their male partners. In response,several writers are now arguing that femi-nists have been leading us astray, and thatwomen are “men’s equals” in violence.Drawing on research which utilizes theConflict Tactics Scale, these writers claimthat a ‘sexual symmetry’ exists in maritalviolence.

In the effort to add some clarity to theconfusion of mixed messages found incommentaries on violent crime, we usedPolice Incident Reports to explore the pat-terns of violent crime in Winnipeg.

Our study is based on a random sam-ple of 1,002 cases (501 women and 501men) involving charges laid by the Win-nipeg Police Service for violent crime (as-saults, robberies and murders) over a fiveyear period (1991 to 1995).

The data were analysed according tothe main types of social relationship re-flected in the event: those involving stran-gers or people unknown to each other;friends and/or acquaintances; familymembers; and intimate partners.


percent of the events studied), it also in-volved more serious violence tactics anda greater frequency of injury to complain-ants. While men accused were most likelyto aggress against other men, 23 percentof their charges involved sexual violenceagainst females (compared with 4 percentfor stranger violence). Women accusedwere most likely to aggress against otherfemales.

Family Violence

Violence against other family mem-bers (children, parents, siblings and ex-tended family) represented only 9 percentof the events studied. These events weremost likely to involve only one accusedand one complainant.

Partner Violence

That women accused are charged withviolent offences involving their (mainlyfemale) friends/acquaintances and fam-ily members confirms that women havethe capacity for violence. However, ourfindings suggest that women are not“men’s equals” in violence, especially intheir encounters with partners.

Partner violence made up the major-ity (53 percent) of the cases where chargeswere laid by police over the period understudy.

We chart partner violence according tofive characteristics: social location; vio-lence tactics used; use of violence by com-plainants; injuries incurred by accused;and who called the police.

Responding to Violent Crime

Police reports reveal that, while it isstranger violence we may fear most, weare more likely to encounter violence at

the hands of people we know, sometimesintimately.

The home is not a haven from violence.Male violence against women they knowis a social problem deserving of seriouspublic attention and response. However,we need to re-examine our strategies.

The zero-tolerance policy imple-mented in 1993 has resulted in more men– and women – being charged with do-mestic violence. It accounts for Winni-peg’s high ranking for violent crime com-pared with other Canadian cities. For in-stance, in 1991, 23 percent of women ac-cused’s charges were for partner violence.By 1995, 58 percent of all violent crimecharges against women were for partnerviolence.

Yet 80 percent of the charges laidagainst women and 51 percent of thoselaid against men were subsequentlystayed by the Crown. Double-chargingoccurred in 55 percent of the cases involv-ing women accused and 10 percent ofthose involving men. Stays of proceedingswere even higher in these cases (88 per-cent versus 70 percent respectively).

These findings suggest the need to re-think a policy that removes all discretionfrom police in deciding whether or notcriminal charges are warranted in a par-ticular case. Rather, we need to ensure thatindividuals are receiving the help theyneed – when then need it.

One promising new developmentalong these lines in Winnipeg is the im-plementation of a pilot project that teamspolice officers with social workers to tar-get high risk domestic violence incidentsand provide support and direction tofamilies in need. These crisis-interventionteams offer the potential to attend betterto the needs of people in trouble, andavoid over-criminalizing families at risk.

Date post:	23-Nov-2021
Category:	Documents
Upload:	others
View:	4 times
Download:	0 times

Mean Streets? - Policy Alternatives

Documents