The text of this executive summary of Theresa Petkau’s study report was found at

Account and Accountability


Executive Summary

© June 1998 by Theresa Petkau

This study, undertaken in 1995/96, is a qualitative evaluation of in-service wife assault sensitivity training provided to front-line police officers. The stakeholder perspective sought is that of patrol constables. Data collection included the following: (1) open-ended one-on-one interviews with forty-three police officers including four administrative trainers, six sergeants, thirty-one patrol constables, and two detective constables; (2) attendance at two wife assault sensitivity training sessions, one at the local level and the other at the provincial level; (3) thirty hours participation in four police ride-alongs; and (4) engagement in numerous informal conversations with officers during frequent station visits over the six-month data collection stage. A symbolic interactionist approach was adopted in order to understand patrol constables' total environment (including social, cultural, and political conditions) that surrounds and impinges on them.

Findings demonstrate that patrol constables largely were unpersuaded by the training claims which for the most part were informed by the feminist perspective of wife assault, one of two major competing sociological perspectives on this issue. Application of a contextual constructionist approach to the training was useful in informing an under-standing of constables' lack of receptivity. When applied to the feminist account of wife assault, this approach--which assumes a subjective rather than objective definition of social problems--reveals that presentation of the following issues are central to the construction of wife assault as a social problem: wife assault as a category separate from other forms of intimate violence; a one-way victimization interpretive framework with its mutually exclusive categories of victim and offender; images of good (i.e., victim) and evil (i.e., offender); fostering of sympathy worthiness for the victim and condemnation worthiness for the offender; violence characterized by the dynamics of power and control as well as escalation; termination of the abusive relationship as a process; revictimization by a patriarchal social system; wife assault as not a lower class phenomenon; widespread occurrence of wife assault; and focus on the consequences of wife assault.

A contextual constructionist analysis further noted that to convince audiences of, obtain support for, and thwart challenges to this particular construction, wife assault claims-makers adopt several rhetorical strategies, some of which include the following:

1) Recounting atrocity stories or dramatic accounts based on first-person testimony;

2) Attributing reports of the victim's problematic behaviour (such as alcoholism, drug addiction, verbal aggression, inattentive mothering, and self-blame) to effects of victimization;

3) Dismissing accounts of the victim's violence as self-defense or retaliation;

4) Rejecting any explanation for the offender's violence (including alcoholism, drug addiction, psychological disorders, and so forth) except that of a desire to maintain power and control over the female partner (present or former);

5) Increasing women's perception of their stake in a solution to wife assault by suggesting that an abusive personality lurks behind even the most outwardly respectable man;

6) Constructing a continuum on which are found other forms of violence against women;

7) Subsuming the issue of wife assault under the more general category of male violence against women and children;

8) Disregarding the discrepancy between, on the one hand, an appeal for recognition of minor levels of violence as situated on a continuum and, on the other hand, a focus on extreme levels of violence in atrocity stories;

9) Overlooking the contradiction between the typification of extreme levels of violence in atrocity stories and minor levels of violence reflected in the majority of police reports;

10) Ignoring the inconsistency between the claim that all women are at risk and the typification of the middle class victim;

11) Discrediting measurements used in rival research;

12) Focusing on disconfirming cases in rival explanations only;

13) Criticizing rival explanations for perpetuating myths, proffering excuses, revictimizing the victim, and/or overlooking attention to root causes;

14) Appealing to nonfalsifiable claims but rejecting similar logic when appealed to by rival claims-makers;

15) Appealing to statistics from carefully selected studies to validate claims;

16) Selectively interpreting data in a manner consistent with the feminist account, and

17) Ignoring, discounting, or overlooking findings inconsistent with the feminist account. Regarded in this manner an argument can be made for wife assault claims-making as a social construction fraught with oversimplifications, dramatizations, and contradictions.

When this claims-making activity in the form of wife assault sensitivity training is directed to a patrol constable audience, acceptance by officers of the feminist account of wife assault is hindered further by police organizational culture in general and the patrol constable subculture in particular. Of the numerous factors related to the larger institutional reality, which appeared to shape constables' rejection of the training claims, some include the following:

1) Lack of dependence on images constructed by wife assault claims-makers because of constables' professional exposure to wife assault situations;

2) Professional exposure to wife assault situations that

  1. afforded constables with perceptions inconsistent with those constructed in the feminist account (such as the claim for escalation, the simplistic victim/offender dichotomy, and the merit of a priori status assignment of victim and offender),
  2. routinely confronted constables with numerous perceived unintended consequences of the wife assault mandatory charge and arrest policies, and
  3. gave rise to constables contesting ownership of the social problem of wife assault in order to make their own claims as holders of expert knowledge in this area;

3) A patrol constable subculture characterized by a general distrust of civilians and a larger police culture ideologically committed to neutrality, the latter two of which contributed to, first, constables' objections to the perceived biased wife assault mandatory charge and arrest policies and, second, constables' resistance to the training's a-priori status assignment of victim and offender;

4) Police procedure which requires constables to investigate in a disconnected manner each allegation of a specific violent act (rather than consider the impact of various acts in the context of the alleged abusive relationship as proposed in the feminist account);

5) Difficulty in establishing grounds for psychological harm in wife assault as well as delegitimation of psychological abuse vis-a-vis police administration's sanctioned non-enforcement of these types of charges;

6) Constables' structured inability to fully enforce police policies and procedures in a police organization deemed by constables as extraordinarily accountable, increasingly open to scrutiny, and devoid of administrative support;

7) Internal police procedures (such as undue administrative monitoring of wife assault/domestic violence reports as well as instituting sanctions concerning charging males only) which persistently exacerbated accountability concerns and gave rise to a perceived need for "protective posturing;" i.e., an approach adopted by constables whereby they assess situations based on the worst case scenario and organize their responses to organizational dictates based on those assessments;

8) Patrol constable subculture's lack of regard for not just wife assault statistics but statistical reporting in general, the latter of which is regarded by constables as a function of state policies, administrative sanctions, and the patrol constable subculture;

9) Legitimacy crisis on behalf of the trainers (both administrative police officers and shelter workers) caused by various internal procedures and protocols as well as the historical administration/front-line chasm and the long-standing hostility between shelter workers and patrol constables;

10) Patrol constable subculture's high value given to common sense street knowledge and low value given to standardized rules and procedures;

11) Structure of recruit training which reinforces at the outset of a constable's career the importance of street-level experience, and

12) Patrol constables' perception of training in general as an accountability weapon used by administration against the front line.

Theoretical Contributions

In noting the foregoing, the following theoretical contributions are made:

1) Support for the usefulness of applying a contextual constructionist perspective to training in general and sensitivity training in particular;

2) Support for the argument that policies often work better in theory than in practice inasmuch as policies, designed on the basis of social problem images, often are based on extreme examples and do not reflect the complexities of social life;

3) Support for the argument that the ‘street’ (or "shop floor") rather than the classroom not only reflects the final arena for learning but also structures experiences of workers in general and patrol constables in particular--independent of any formal training;

4) Support for the position that workplace training needs to be considered in the context of workplace learning, and

5) Challenge to the utility of workplace sensitivity training in substantive attitude change.

In summary, claims-making attempts in the training to replace constables' subjective definitions of wife assault with official (read feminist) definitions were unsuccessful. Patrol constables' rejection of the feminist account of wife assault largely stemmed from workplace accountability concerns as well as constables' claims to define their own account of the violence. This latter account, although rooted in constables' widespread professional exposure to intimate violence, is afforded no legitimacy in a politically charged environment where (at least in Ontario) power currently belongs to the feminist movement to define both what occurs in intimate violence as well as suggest solutions. This power extends to feminists' ability to control the content of wife assault sensitivity training and then label as insensitive any groups which do not support the feminist view--even though, as outlined earlier, this view can be criticized on several grounds. In revealing the political realities of the claims-making process involved in sensitivity training as well as the myriad factors which can impinge upon and frustrate that process, this study challenges the popular view of sensitivity training as a panacea for attitude change in the workplace.

Implications for Further Research

The study's discussions on possible unintended consequences of policies mandating charge and arrest in cases involving allegations of wife assault hopefully will give impetus to researching this issue with patrol constables from other police services. In the event similar findings are obtained, a province-wide evaluation of these policies appears to be in order.

Other areas for further research include examining the following:

1) Ways in which patrol constables resist mandatory directives;

2) Processes by which constables' perceptions of parties' accounts get reinterpreted into written accounts suitable for administrative monitoring;

3) Extent to which patrol constables encounter alcohol and female aggression when responding to domestic calls;

4) Victims' perception of police response, and

5) Effects of standardized rules in policing. As well, given that patrol constables routinely confront domestically violent situations in a manner unlike any other professional groups connected with this issue, further articulation of front-line officers' perceptions and understandings is recommended. In my opinion, these studies would be useful in contributing to the theoretical debate regarding domestic violence in general and wife assault in particular.

Overall, it is hoped that this undertaking provides an opportunity to better understand both what is involved in claims-making about wife assault as well as what accounts for the perspective of patrol constables who daily are faced with what they perceive to be inconsistencies between their lived experience and the foregoing claims. It also is hoped that this evaluation will lead to more effective training of patrol constables regarding their understanding of; and response to, domestic violence.


Based on this study's findings, several recommendations regarding wife assault sensitivity training are offered, some of which include:

1) Training needs to be adapted to a patrol constable audience. This adaptation requires consideration not only of the apparent contradiction between officer neutrality and a prior status assignments of victim and offender, but also administration's awareness of and responsibility for effects on front-line personnel of possible inconsistencies between written policies and implicit sanctions regarding constable violation.

2) Avoid making singular claims for truth. To this end, there needs to be an acknowledgment that wife assault sensitivity training is based on only one of several competing perspectives on intimate violence.

3) Clearly state the goals of the training as well as what is expected of patrol constables.

4) Avoid adopting the rhetoric of myths unless a direct link is made to the relevant perspective from which the myths originate.

5) Include speakers with other theoretical orientations about wife assault/domestic violence as well as experts on victimology; this approach should encourage a more free exchange of ideas.

6) Minimize reliance on the use of statistical information to support training claims.

7) Eliminate the use of wife assault training videos adapted for general audiences.

8) Provide constables with concise written guidelines on what constitutes reasonable grounds in cases involving allegations of domestic violence and wife assault.

9) Where available, have victims (female and male) discuss their perception of police response, both positive and negative.

10) Insofar as legitimacy of trainers is critical, two options suggest themselves:

  1. Replace existing trainers with those who currently hold legitimacy with patrol constables such as front-line supervisors; or
  2. Examine those areas where existing trainers' legitimacy is undermined with the intent of eliminating or reducing these areas. In connection with the first recommendation, a poll of constables' views might be useful.

With regard to the latter recommendation, the following is suggested:

  1. Police trainers no longer undertake investigations of wife assault/domestic violence complaints involving sworn members;
  2. Expand police trainers' role to include front-line work with constables;
  3. Amend protocol with shelters in order to establish front-line sergeants as liaisons with shelters when information from and/or access to shelters is required;
  4. Have shelter workers participate in police ride-alongs, and alternately, have constables spend time at shelters;
  5. Include patrol constables in police training of shelter workers.

11) Encourage and document discussion around perceived consequences of any mandatory policy (including the wife-assault charge policy) for the purpose of identifying and resolving problematic areas. If necessary, share concerns with other police services in order to work towards solutions, the latter of which may necessitate communication with state authorities.

12) Include patrol constables in police workshops about domestic/wife assault as well as related community discussions where police involvement is requested.

13) Given the move towards national consistency in police officer training, share these research findings with other police services across Canada.

For the most part, constables uniformly regarded rigid black-and-white policies dictating mandatory charge and arrest as an ineffective response to wife assault insofar as the policies purportedly thwarted a sensitive response to wife assault calls, first, by overlooking individual dynamics within each family, second. by ignoring wishes of the parties involved, and, third, by disallowing possibly more effective approaches. As well, officers criticized these policies on the basis that they,

(1) Deterred women non-supportive of charges from seeking help;

(2) In some cases, negatively impacted upon the family unit as well as the accused by burdening already stressed families and leading in many cases to their breakdown;

(3) Acted as a powerful tool in the hands of vindictive women;

(4) Gave rise to charges which officers considered unwarranted, and

(5) Resulted in unnecessary substantial costs to the criminal justice system in general and the police in particular. Officers also directed criticisms at the wife assault charge and arrest policies for negating their helping role, hindering their investigative role, violating their keen sense of justice, forcing a compromise of their sense of ethics and fairness, thwarting professional development of intuitiveness, contributing to the move towards rote policing, and serving as yet another example of administration's tacit sanctioning of policy violations (by virtue of staffing levels supposedly insufficient to allow for unabridged adherence to policies).