injusticebusters logo

Karla Homolka

"A moral vacuity in her which is difficult if not impossible to explain": law, psychiatry and the remaking of Karla Homolka

Anne McGillivray

Associate Professor of Law, University of Manitoba, Canada

Karla Homolka

Karla Homolka and her husband Paul Bernardo were convicted in 1993 and 1995 respectively of the abduction, torture, rape and murder of two teenage girls in the small lakeside town of St Catharine's, Ontario.

Police, lawyers and judges—indeed, the legal system itself—were profoundly challenged by the disjunction between available scripts for crimes of sexual violence and for women who 'go wrong'. The fragmenting of notions of justice and agency is seen in investigation and evidence-gathering, in the plea bargain made with Homolka and in Bernardo's trial, which starred Homolka as witness for the prosecution and publicly revealed for the first time the extent of her participation in the crimes.

Her sentence of 12 years concurrent was a compromise between contested notions of women's agency, a play-off between competing ideas about womanhood and subjectivity against male sadism and dominance. The public uproar following Canada's 'trial of the century' precipitated an investigation by the Ontario Attorney-General of the Homolka plea bargain. The resulting Galligan Report and its depiction of the relationship between Bernardo and Homolka is a central reference point of this article.

Fault lines in the cracking facade of Western law trace out an increasing pluralism in legal discourse and practice. Gaps between public and juridical perceptions of justice are widest in the high-profile murder trials foregrounding race, sex, gender and childhood which have consumed the last decade of the millenium. These most famously include O.J. Simpson criminal and civil trials in the deaths of his ex-wife and her friend, the torture and killing of baby James Bulger by two young boys and the plea bargaining of Karla Homolka who, like Rosemary West and Myra Hindly acted in consort with her husband in the rape and murder of children.1 The ethical standards of the legal profession show deep fracture as social fragmentation challenges notions of honourable lawyering. "No holds barred" adversarial games have come under Law Society scrutiny, yet such tactics are routinely viewed by criminal defence lawyers as central in their duty to clients. Homolka's plea bargain was necessitated by an extreme instance of such tactics—the concealing of real evidence. The fracturing of criminal law's once-seamless discourse is manifest in its inereasingly open reliance on the "soft" sciences of psychology, psychiatry, psychoanalysis, criminal profiling—the "psy" disciplines which play a central role in Homolka's ease. The case raises questions about the relationship of law and psychiatry, agency and demonisation, and the medicalising of evil at the close of the millennium.

Evil went out of the world at the end of the Middle Ages and madness came in (Foucault, 1965, p.223) but Western man is more than ever a confessing animal (Foucault, 1979). Although the confessional is now the law office and the clinic, confession remains central to the discipline of the soul and to the creation of the legal persona. Few have confessed more madness or more evil for more hours to more disciplines—police, lawyers, psychiatrists, psychologists, psychotherapists— than Karla Homolka, partner of serial rapist and killer Paul Bernardo. Her thousands of hours of confessions and the professional discourses flowing from them shaped her remaking from accused sex killer to a battered woman seduced into perversion, beaten into murder. Her persona, constructed from psy knowledges and the pragmatics of law, raises profound questions about law and lawyering in a fragmented world.

Karla Homolka entered into a plea bargain with the Ontario Crown on 14 May 1993 for her part in the murder of schoolgirls Leslie Mahaffy and Kristen French, 14 and 15, in consort with her partner Paul Bernardo. Homolka denied inflicting death but her participation in their abduction and assault constitutes first degree murder. Read into the facts at the sentencing hearing was her role in the drugging, sexual assault and death by aspiration of vomit of her sister Tammy, 15. Her sentence, 12 years concurrent on two counts of manslaughter, was denounced by lawyers as a "sweetheart deal", a "travesty of justice", a "deal with the devil". A fourth adolescent victim, Jane Doe, drugged, assaulted, forgetful but alive, was recalled to Homolka in a post-sentence dream. The nature and extent of Homolka's involvement in the Mahaffy and French cases was spectacularly revealed in six homemade videotapes later recovered. There was widespread belief that she had known where the videotapes were hidden, that she wilfully concealed the Jane Doe incidents and, most centrally, that her claims of being under Bernardo's control— a central tenet of the plea bargain—were spurious. Speculation was fed by a publicity ban on the plea bargain which stood until Bernardo's trial. Print and website sources imaged demonic duos, vampirism, Barbie and Ken perfect-couple perfectmurderers, sexy "Killer Karla", the comic "Karla's Web" featuring Homolka's psy confessions. The gaze centres, always, on Homolka.

Bernardo was convicted of first degree murder, kidnapping, aggravated sexual assault, forcible confinement and offering an indignity to a dead body and was sentenced to life imprisonment in September 1995. Homolka testified for the Crown. Responsible for the sadistic sexual assault of dozens of adolescent girls and young women (11 victims testified at the hearing, 23 charges were originally laid and 33 are attributed to him by police), Bernardo was declared a dangerous offender, having "shown a failure to control his sexual impulses and a likelihood of his causing injury, pain or other evil to other persons through failure in the future to control his sexual impulses" under §753 of the Canadian Criminal Code, RSC 1985,C-46. He was sentenced to indeterminate detention. That he would be incarcerated for his mortal lifespan seemed a foregone conclusion. Homolka, in the popular view, should have taken her seat beside him in the prisoner's box and seat of ultimate evil.

Underlying the disquiet was a fear that gender somehow won, that Homolka, being female, was held less culpable on that basis alone. This denied women's equality and moral autonomy. Conversely, it was a sort of witchery through pretended weakness, lies and manipulation, the successful use by an evil woman of obnoxious female traits.2 Feminists and anti-feminists could unite in their disapprobation of Homolka and her self-created images, first of normality—pretty teen, party girl, beautiful bride, dutiful daughter, supportive wife—and then of the controlled and battered woman with symptoms culled from the Lenore Walker classic on her gaol cell bookshelf. The criminal justice system needed Homolka's evidence to convict Bernardo, a far more dangerous offender in the view of the prosecution. As Homolka was herself vulnerable to prosecution for first degree murder, her evidence had to be bought by sentencing discount. This is not unusual. The vast majority of criminal cases in Canada are settled by way of guilty plea, usually involving a sentence bargain and often a charge bargain. The system depends on this "secret" of law. It is no secret to the courts, alerted to the nature of a bargain in various ways. Nor can it be said to be secret from the public, given high-profile cases of immunity given to informers in organised crime and the enduring popularity of courtroom fiction. Bargaining is often a hurried event over the telephone, in courthouse hallways, even the courtroom itself. These practices were criticised in Ontario in the Report of the Attorney-General's Advisory Committee on Charge Screening, Disclosure and Resolution Discussion (Martin, 1993, hereinafter the Martin Report), which recommended among other things full disclosure of plea bargain terms to the sentencing court. Even so, only in the rarest of cases does a bargain receive the elaborate attention given Homolka's bargain before and after the fact.

Plea bargaining is barter, with a buyer and seller and consideration for services in the sentencing discount. Where evidence is to be bought, the discount may extend to complete immunity, usually available only to organised crime informants. Homolka promised full disclosure and testimony against Bernardo in return for reduced charges—two counts of manslaughter—and a joint sentencing recommendation. In so doing, she escaped central blame for the deaths. In law, Homolka and Bernardo are equal partners in the crimes, x=y. Where x as an informer in a case like this receives a discount in charge and sentence to the value of a, the difference is passed on to A. The equation is now x - a <A + a. Homolka's discount was to be justified by the control exercised over her as battered woman in the medico-legal diagnosis. This requires a controller whose identity can absorb the excess blame, a. In the circumstances of Homolka's case, the controller must be of major proportions, a pervert, a sexual sadist, a psychopath. This is the medico-legal diagnosis of Bernardo.

Homolka's remaking is a subtle regendering to account for the problem of evil in women. Despite long association of women with evil and seduction into evil, actual evil in women is problematic. Women rarely kill and their killing is typically within the domestic relationship.3 Women who participate in sex killings are so demonised that femaleness itself becomes a perversion. Transforming Homolka from murderess to battered woman centres domesticity and enhances womanliness. In this therapeutic transformation from defendant to victim, from action to passion, from agency to compliance, evil becomes the familiar. Foregrounding the psychological webwork of relationships between Homolka, Bernardo and the dead girls through a battered woman identity focuses on power disparities and decentres partnership and agency. Homolka is now to be read as a victim standing with her child victims in Bernardo's sexual gaze, with them vulnerable to torture and death. The battered woman construct incorporates Homolka's domestic identity—her girlish love, marital aspirations, housekeeping, battering—in deliberate contrast to Bernardo's extramural stalking and rapes. This in turn foregrounds his dominance in their domestic murders. In her domesticity, Homolka aspired to recreate herself in the image of Bernardo's desire. Already an ageing child of 17 when they met, she play-acted the pubescent child-woman in sex with Bernardo during the five years they were together, to the extent of dressing in her dead sister's clothing and answering to her name. By helping him possess the objects of his desire, whom he had been secretly raping for years, she hoped to maintain their relationship. However familiar are Homolka's aspirations "to be" for Bernardo, it did not account in the public eye for the boundaries of womanhood she transgressed.

The reconstruction strategy was meant to bolster her credibility as prosecution witness and validate her plea bargain by discounting her moral fault and legal blameworthiness. It was a partial reconstruction conferring partial immunity. Homolka still had to go to gaol. Even so, her psy identity was contested by Bernardo and ridiculed in the media. The attenuation of justice, in the public perception, by psy sciences recalls the deeper disciplinary fracture and the history of hegemonic clashes between two powerful discursive fields with opposing constitutive missions— law and psychiatry.

Homolka's plea bargain was necessitated by system failures in investigation, autopsy, DNA analysis, FBI profiling and professional ethics (Campbell, 1996).4 Psy knowledges informed by feminist analysis of gender and victimization explained it. Was the bargain justified? Was the sentence fit? A petition to have the bargain annulled was signed by 300,000 Ontario residents and submitted to the Attorney-General of Ontario. To account for the cost of purchasing Homolka's evidence, Patrick Galligan, newly retired from the Ontario Court of Appeal, was appointed to inquire into the circumstances, legality and possible breach of the Homolka plea bargain. This essay follows the narrative and structuring of events as set out in his Report to the Attorney-General of Ontario on Certain Matters Relating to Karla Homolka (Galligan, 1996, hereinafter the Galligan Report). Where system failures arise in his narrative, Galligan defers to the Bernardo Investigation Review (Campbell, 1996).5 Galligan centres on the contested identity of Karla Homolka forged in the confessionals of the law office and the clinic and on the pragmatics of law which ultimately justify the bargain.

The Galligan Report considered three questions. Was the bargain appropriate? Should Homolka be charged with the sexual assault of Jane Doe? Should further proceedings be taken against her in the murders?6 Answers shade into one another. There is a "profound and widely felt sense of public disquiet" (p. 7) about her avoidance of full criminal responsibility in the murders and her late disclosure of the Jane Doe assaults was a prima facie breach of the bargain. The Report draws upon legal and psy discourses. Lawyers numerically play a major role. Galligan was a criminal lawyer prior to his appointment to the bench. He "sought out the advice of a relatively small group of lawyers, but one which I thought was sufficiently broad to give me a wide range of views within the profession" (p. 3) and he took as counsel a "top-flight member of the criminal defence bar" who was also an experienced Crown attorney (p. 6). Lawyers researched and co-wrote the report and interviews were held with prosecutors, counsel for Homolka and Bernardo, lawyers representing a variety of professional associations and lawyers representing the Mahaffy, French and Doe families. Also interviewed were a judge, three law professom and the Attorney-General of Ontario. The Report reflects the views and work of 46 lawyers.

The only lawyers not interviewed were Ken Murray, who resigned as Bernardo's counsel on the eve of trial, and Murray's legal counsel. Murray faces criminal charges and Law Society disciplinary proceedings for an act which, depending on one's view, is either a heroic or deeply misguided piece of defence work and potentially a breach of law and professional ethics. On 6 May 1993, when the 71-day search of the Bernardo-Homolka house had ended, over 100 videotapes had been reviewed on the premises (as the search warrant did not specify removal, Campbell, 1996, p. 211)and a single police officer was on guard, Murray entered the house together with two co-counsel, stood on the toilet of a second-floor bathroom, reached deep into the cavity above a recessed light fixture, removed the videotapes, put them in his briefcase and walked out (Campbell, 1996, p.213). The light fixture had been removed but police had not reached far enough into the attic cavity. Having no reason to suspect an officer of the court of interfering with a crime scene, the police officer did not search the briefcase. Homolka's plea bargain was sealed the following day. The lawyer locked the tapes in his office safe, where they remained hidden for another 16 months. The police search yielded only a blurred segment of tape showing a nude Homolka with an unidentifiable nude girl. Before letting her parents hospitalise her after a final beating, Homolka searched unsuccessfully for the tapes in the garage rafters where they had hidden Bernardo's rape paraphernalia. Without real evidence connecting Bernardo with the murders, the remaining option was the purchase of Homolka's testimony. Had the videotapes been found, had Murray turned them over, there would have been no plea bargain (Campbell, 1996, p. 220). "The tapes were crucial evidence and the failure to find them during the search led to the plea-bargain with Karla Homolka." Hidden in the attic cavity and the lawyer's briefcase were the needed secrets of Bernardo's complicity. Also hidden were images of Homolka, darker, lustier, more cruel and far more enigmatic than those drawn in her confessions. The disappearance of the tapes necessitated Homolka's reconstruction. Their reappearance profoundly challenged it.

Lawyers can say what the law is, how far it can be manipulated and whether a bargain is broken. There are things that lawyers cannot say which rest at the heart of the case. Chief Justice Brian remarked in 1977 that "The thought of man is not triable, for the devil himself knoweth not the heart of man" (YB (1977) P. 17 E. 4. 2a, pl. 2).7 The psy disciplines claim some knowledge here and theirs is the second professional discourse in the Galligan Report. Although vastly outnumbered by lawyers—five psychiatrists and three psychologists are quoted or cited—their contribution to the discourse of madness and badness, sickness and guilt, medicine and morality, is a tie-breaker. Where law cannot account for the modicum of difference in Homolka's culpability, the psy disciplines do so.

Galligan dramatically begins with the account of a victim of the mysterious Scarborough rapist. This unnamed woman, "not the first of Paul Bernardo's victims, nor would she be the last lives with terror, fear of being alone, and the inability to love, to trust or to lead a normal life ... swayed between living and non-living ... exhausted both physically, by inability to sleep, and emotionally ... indelibly scarred for the rest of her life" (p. 7). The rapist's modus operandi is detailed—night stalking, knife and ligatures to the neck, blows which broke one victim's arm, body scorings with a knife, vaginal insertion of twigs and other found objects, vaginal and anal rape, forced fellatio, death threats, female-specific insults, knife-point repetition of scripted praise for the rapist's body parts, bone-breaking, theft of jewelry, clothing and identification as mementos. This foreshadows the murders of Mahaffy and French—abduction, scripting, beatings and torture, patterned sexual practices, possessing the victim by humiliation and the threat of death, keeping mementos. Death becomes the ultimate possession of the object of desire and videotapes are the perfect memento. In foregrounding Bernardo's extramural evil, the narrative sets the stage for Homolka's domestic exculpation. When Galligan moves on to their relationship and her crimes, her abuse is at first a subnarrative hemmed about with interjections of doubt, where the judicial voice falters and must yield to psy explanations. By distancing her claims of abuse, this rhetorical tactic enhances them. It is deployed less to proclaim innocence than to subvert agency. The subplot of Homolka's abuse reflects and refracts Bernardo's rapes and with them anticipates his moral leadership in the murders. She is caught in a maelstrom of fate, Everywoman meeting the devil. Even so, her culpability is recalled to the reader when, at odd moments, Galligan foregrounds her moments of choice and sharply reproves her conduct. In his omnipotent paternal gaze, he may wish to forgive but must sternly judge.

Details of the first eight Scarborough rapes were sent to the FBI Behavioral Sciences Unit specialising in serial rape and sexual killings. An agent identified the assailant as a sexual sadist who "psychologically abused his victims and used more force than necessary to subdue them" (p. 13) and predicted that the "sadistic component" and the rate of offending would escalate due to "a life style change" (p. 14).8 Future victims would be kidnaped and this would guarantee their death. Assaults would be videotaped. The rapes would appear to end only because the rapist had moved away. He would be stopped only by external force—death or incarceration. This oracular prediction proved true. The predicted lifestyle change was Bernardo's involvement with the Homolka family, enabling his move from his parents' Toronto home to the small community of St Catherine's. The Scarborough rapes ended, Toronto police lost interest, and St Catharine's police failed to connect the rapes with a rape there and with the schoolgirl murders (Campbell, 1996, Ch.2). A police report naming Bernardo as a suspect in the Scarborough rapes, based on a statement from a former girlfriend, I.F., was filed 5 January 1988 (Campbell, 1996, p. 17). It was ignored.

Engagement, marriage and respectability granted Bernardo the privacy which enabled the predicted "total physical and psychological domination" of his victims. When Homolka met Bernardo in a hotel restaurant in October 1987, they had sexual intercourse within an hour or so and began dating. She was 17, in Grade 12, oldest of three sisters. Bernardo was 22, youngest of three children, a University of Toronto graduate. By this time he had raped at least twice and had practiced his domestic technique on an earlier girlfriend, J.M.G., whom he met in 1984 when she was 16. For three years, Bernardo "dominated and controlled" G. as "his little servant girl, playtoy" (p. 19), isolated her from her friends, forced painful sex on her, used handcuffs and ligatures and in November 1987 locked her in his car, punched, beat and raped her, pulled out clumps of head and pubic hair, raged and threatened to kill her. While he looked for his knife "in a frenzy", G. escaped and avoided him thereafter. She had little to worry about. Bernardo had found Homolka.

Bernardo became a Homolka family intimate. His sex with Homolka was vaginal, he treated her "like a princess", "the only girl in the world" (p. 21). He was considerate and attentive, he brought flowers, took her to movies and was respectful to her parents, he went out with Homolka and her friends and he continued his career as the Scarborough rapist. In his December rapes that year, he made one virginal victim kiss his penis and wish it a merry Christmas. He gave Homolka a gold chain and a dress and began to tell her how to do her hair, what to wear, where to go. Her friends were now stupid and immature. He introduced her to "large amounts" of alcohol and to fellatio and taught her such speeches as "My name is Karla, I am 17 years old, I am your little cocksucker ..." (p. 23).The Scarborough rapes continued and in late spring of 1988 he began anal sex with Homolka while she wore a dog choke collar. He took pictures of her with his new Polaroid and began beating her that summer. The Scarborough rapes continued. His names for Homolka alternated between princess and slut, bitch or cunt. Marriage plans went forward. They became engaged Christmas eve 1989 and by the spring of 1990, Homolka was calling herself his sex slave.

Her abuse, she claims, escalated. (Homolka was "not to argue with Paul, be a perfect, understanding girlfriend for Paul, remember you're fat and you're lazy and you're ugly", her note on her mirror reminded her).9 But here Galligan falters. "I must digress to say that I was very sceptical about her statements that she was subjected to violence and threats to the point where she was in such fear of him that she would do his bidding, no matter how monstrous" (p. 32). She lived with her parents and led her friends to believe that everything was wonderful. The study "Compliant victims of the sexual sadist" (Hazelwood et al.,1993, discussed below) "caused me to have an open mind on this issue because it documents similar phenomena occurring to women other than Karla Homolka".

He resumes his narrative. In the spring of 1990, Bernardo told Homolka of his desire for young sex slaves to be brought to her house and ordered her to pretend to be her 15-year-old sister Tammy when they had sex. By fall he was vaguely threatening violence if he couldn't have Tammy, who could be drugged and would never know. Homolka could get the drugs from the animal clinic where she worked. On 20 November 1990, Bernardo was interviewed by Toronto police on the basis of tips received in response to an unusually accurate police drawing of the Scarborough rapist. He confidently volunteered samples of saliva, hair and blood and told Homolka he was not the Scarborough rapist. She hoped that they would not "mess up" on the forensic evidence (p. 33). These "went into a black hole" and were not DNA-tested until January 1993, over two years later (Campbell, 1996, p. 355). "The tragic converse of these facts is that Bernado, during the 25-months his DNA was waiting to be tested, raped four young women and raped, tortured, and murdered two others. In hindsight, it is clear that these rapes and murders could have been prevented if Bernardo's DNA sample had been tested "... within 30 or even 90 days of the [threshold] serology test" (Campbell, 1996, p. 356). On 23 December 1990, as a Christmas present to Bernardo, Homolka drugged Tammy with Halcyion and Halothane, tranquilizers obtained through her connection with the veterinary clinic where she worked as an 'Animal Health Technician' (Campbell, 1996, p. 152). She and Bernardo took turns videotaping events and sexually assaulting her. At Bernardo's request, Homolka cunnilinguated her and was "repulsed" (p. 83). Tammy died, Bernardo attempted resuscitation, the drugs were flushed down the toilet, the camera hidden, the body washed and dressed and 911 called. In ruling the death accidental, the coroner failed to test blood and stomach contents, examine the genital area or query a large second-degree burn on the face (Campbell, 1996, p. 99).

Bernardo continued to live with the Homolkas until mid-January, when her parents asked him to leave so that they could mourn in private. He was furious and vowed never to return. He rented a pink clapboard house on Bayview Drive and supported himself smuggling cigarettes. Homolka said, the still-sceptical Galligan writes, that the verbal and physical abuse continued, heightened by threats to expose her role in the death of her sister. In another exercise of judicial finger-shaking, again abstracting himself from his narrative, he reports that Homolka "wrote a truly nauseating, reprehensible letter" (p. 37) to her friend in which she criticised her parents' selfish grieving and proposed limits on wedding costs due to funeral expenses. Bernardo raped again, this time in the St Catherine's area, and again ordered Homolka to bring young girls to his house. Homolka produced Jane Doe on 7 June 1991, a 15-year-old friend who worked at the veterinary clinic and who regarded Homolka (as did all her victims) as an older sister. Doe was drugged and sodomised, had no memory of the event and its repetition, and remained the object of Bernardo's daytime seductive attentions.

On 15 June 1991 in the early hours of the morning, Bernardo was prowling the residential area of Burlington. He found 14-year-old Leslie Mahaffy, gave her a cigarette, took her home, woke Homolka and told her he had a girl. Homolka went back to sleep. The next day, he blindfolded Mahaffy and invited Homolka in. Homolka made drinks, whispered questions for Bernardo to ask Mahaffy, cunnilinguated her and videotaped the rapes and beatings. She gave her sleeping pills and her own teddy bear to hold while Bernardo strangled her. He dismembered the body with a circular saw, encased body parts in concrete and threw them in a lake. On their wedding day two weeks later, the lake was lowered and the remains found. On their wedding night he told her that he was the Scarborough rapist.

Bernardo's abuse of Homolka escalated, Galligan writes, now sounding more sure. Bernardo began taking her on prowling trips by car while he watched, stalked and photographed women. He repeatedly demanded that she bring him young girls. In August, Jane Doe was again drugged and sodomised and eventually "succumbed" to Bernardo, fellating him "on a number of occasions" (p. 124). She ended the relationship in December 1992 and has no memory of the assaults. On 16 April 1992, 15-year-old Kristen French was walking home from school. Homolka enticed her into Bernardo's car. For the next three days French was bound, raped, sodomised, beaten and forced to perform sexual acts while feigning enjoyment for the videocamera. Homolka participated in scripting, direction and sexual assault. Although Bernardo twice left the house for food, Homolka did not free French. On the fourth day, French (declaring her support for Homolka and defying Bernardo to the end, receiving a last vicious beating for calling him a bastard) was strangled. Homolka cut off the hair, stripped, washed and douched the body and helped Bernardo deposit it in a country ditch. She scrubbed the house and car, vacuumed carpets and burned hair, clothes and the vacuum bag. Galligan dwells on these domestic details.

The FBI was called in to profile the murderers. This time the oracle was wrong, concluding that the killers were male blue-collar workers. Bernardo's abuse of Homolka again escalated. She was strangled, knives were thrown at her, she was hit with firewood, shoes and a flashlight, kicked, punched and raped, her hair was pulled out, she was pushed downstairs, her foot was punctured on a rusty nail. Galligan cautiously observes, "If her evidence is truthful, there can be no doubt that Paul Bernardo's treatment of her through the last half of 1992 is nothing short of a horror story. It is somewhat of a surprise that she is still alive and sane" (p. 41). On 5 January 1993, co-workers concerned about her absence from work called her parents, who took her to hospital. Galligan quotes from the report of the ER physician. Homolka had black-ringed "raccoon" eyes and a hemorrhage in the left eye from blows to the back of the head driving the brain forward into the skull, a contusion to the forehead, bruises down the side of her neck and along her arms, bruises and swelling to 75% of her legs from mid-thigh down, and a puncture wound from a screwdriver on her right thigh above the knee. She could not walk. She had been handcuffed, bound with the electrical cord used to strangle Mahaffy and French and left in the root cellar where the bodies had been stored. "Dr. Arndt reports at p. 7 that the emergency room physician at the hospital said, 'this was the worst case of wife assault that he had seen' " (p. 41). Bernardo was charged with assault and released on bail. On 9 January Homolka moved in with her aunt and uncle, consulted a legal aid lawyer and began divorce proceedings.

On 1 February 1993, Forensics informed Metro Toronto police of a preliminary DNA match between Bernardo's samples and samples from three rape victims. Bernardo was placed under surveillance. Toronto and St Catherine's police met together for the first time and Bernardo at last became a suspect in the Scarborough rapes and the St Catharine's murders. Police interviewed Homolka. She told them of her abuse by Bernardo. When they left, she told her relatives about the murders and retained a criminal lawyer. Bernardo was arrested 17 February 1993 on 23 rape charges. Murder charges were laid against him three months later.

The discretion not to charge must not be lightly used, Galligan writes, and public confidence is a serious factor in plea bargaining. The murder case against Homolka was strong, based on her own statements to family and co-workers. The case against Bernardo was weak to non-existent. Interjurisdictional committees of police forces and Crown attorneys were established to investigate, share evidence and decide on strategy. When Homolka's lawyer approached the Crown with an offer to bargain, it was clear that she was central to all strategies. The bargain, finalised 14 May 1993, required Homolka to give an induced statement testing her evidence and credibility, followed by a cautioned statement waiving her right against self-crimination to ensure that she came through. She would withhold nothing, assist police in their investigations, never enter a book or media contract for profit from her crimes and testify against Bernardo. There would be a publicity ban until Bernardo's trial. Charges would not be laid in Tammy's murder. In return, the Crown would recommend a sentence of 12 years concurrent on two counts of manslaughter. The sentencing judge agreed, writing, 'This accused has committed the worst crimes, however she is not the worst offender—for whom the maximum sentence [life imprisonment] is designed' (p. 107). Homolka had "something to sell", Galligan writes, and the Crown had "something to buy" but she "could not hold out for too high a price" (p. 63). Nor, it should be noted, could the Crown. The Mahaffy and French families "reluctantly acceded" to the bargain as "the lesser of two evils" (p. 111).

In accordance with the bargain, Homolka's part in all three murders was disclosed to the court. The Crown acknowledged that the evidence would found murder charges. Homolka's counsel made delicate reference to her assistance in the Bernardo case. 'Importantly, it is recognized by the Crown that she may help to ensure that those who may also be responsible for these odious crimes are brought to justice' (p. 104). Galligan details Homolka's contributions to the Bernardo murder investigation. In 400 hours of interviews with police and prosecutors, she made two disclosures leading to the first real evidence implicating Bernardo. Police had checked retail outlets for cement purchases for disposal of the Mahaffy remains without success. Homolka told them that Bernardo had returned unused bags for refund and was angry about having to leave his name. Records checked out. This was the first link between Bernardo and the Mahaffy murder. Police had exhaustively tested carpets for stains but Homolka pointed out an area she had taken special pains to clean. The underside bore traces of semen and vomit matching DNA from Bernardo and French. This was the first link between Bernardo and the French murder.

Homolka did not mention Jane Doe. Memories surfaced in a post-sentence dream. She recalled administering drugs, Doe's breath stopping, calling 911, canceling the ambulance, sitting up with her all night, Bernardo sodomising her, the videotaping. "Why didn't I remember all of this when they first questioned me?" she wrote to her lawyer (p. 130). "I have to tell them but what if they nail me for this too? Can you do something to make sure they don't?" She argues, "I never believed that the videotapes were destroyed. I believed the police were in possession of the videotapes when I was originally questioned. I ... admitted my involvement in things that are far more incriminating to myself than this". Why should she lie? The psy sciences could and did explain her memory loss in terms of returned memories, repressed memories, other sequelae of battering, drugs and alcohol. If she did not in fact remember the Jane Doe assaults, Galligan concludes, then a perjury charge would not succeed and the bargain could not be voided by late disclosure. "I hold no brief for Karla Homolka. Nevertheless ... she completely and fully disclosed her complicity" (p. 58). The administration of justice faces disrepute if the Crown reneges and, unless it can be proved that Homolka wilfully withheld knowledge of the Doe assaults, the Crown is bound.

The central concern, however, was the impact on Homolka's credibility if the bargain were withdrawn and if she were charged in the Jane Doe assaults. On her credibility alone would rest Bernardo's conviction. If charged with the Doe's assaults, Galligan argues, her defence would rest first on abuse of process in the Crown's violation of the bargain and second on necessity, duress, compulsion or "moral involuntariness" resulting from Bernardo's abuse. Acquittal would be unlikely but the trial would be long and complex and sentence would be token. In support of his theory, Galligan cites Bernardo jurors, one saying that the evidence of battering and duress was there "in huge, huge quantities" and a second who wrote to her parents, "I personally believe she was manipulated, controlled and battered" (p. 167). (Another juror, quoted in press accounts, said he did not believe a word of it.) Due to court congestion, Homolka's trial would not end before Bernardo's trial, violating a Canadian practice rule that charges against accomplices be dealt with before they can testify for the Crown.

Murder charges were laid against Bernardo four days after Homolka's sentencing. Preferred indictments were granted on all charges and trial began 4 May 1994. Homolka was cross-examined intermittently in prison by defence counsel over a three-month period. Jury selection began a year later, 1 May 1995. Then began a long series of motions, most evidentiary. Bernardo's rap Iyrics—"I'm a solo creep I make the girls weep acting out my crimes while others sleep", "I'm a deadly innocent guy", "I've got no remorse and I've got no shame"10 —his "teen model" catalogue cutouts11 and the "Compliant victims" study12 were among exhibits ruled inadmissable on the basis of prejudicial impact. On 12 September 1994, Bernardo's counsel Ken Murray approached the bench to withdraw from the case. His new counsel John Rosen took the videotapes, reviewed them and gave them to police 22 September (Campbell, 1996, p. 213). Rosen told the court, "I determined that it was my duty as an Officer of the Court and as a matter of law to turn all six videotapes over to the police". The videotapes spectacularly reveal Homolka's role in the assaults. They do not show the killings. That Bernardo's conviction for murder would rest solely on her credibility proved correct. His lawyers investigated the possibility of privately prosecuting Homolka for murder, in order to destroy credibility. Homolka began testifying 19 June 1995 and spent over three weeks on the stand. The defence conceded that Bernardo participated in the abduction and assault of Mahaffy and French but argued that he did not kill them, hypothesising that Homolka killed French while Bernardo was out of the house and Mahaffy strangled herself on the wire around her neck secured to Homolka's hope chest. To emphasise the point that Homolka's lack of credibility was Bernardo's sole defence, Galligan quotes from closing argument.

She is the only one who points the finger at him and says that they died by ligature strangulation at his hands, not hers.... that's what it comes down to. Do you believe her? ... It is a major lie. It is something that goes to the heart of her credibility. It is something that reveals her for what she really is. A person who would lie, cheat, steal and manipulate the evidence in an effort to protect herself.... Would you condemn a dog in a dogcatcher's court who is alleged to have bitten someone on evidence like that? ... Is her evidence believable and credible? Does it satisfy the intellect, soothe the conscience? (p. 119)

Bernardo's conviction is for Gallagher the ultimate justification of the plea bargain. "[D]istasteful as it always is to negotiate with an accomplice, the Crown had no alternative" (p. 111).13

Having preserved what it could of the credibility of its sole living eyewitness, the Crown had to make another tactical choice. Accomplices testifying for the Crown may be presented with the disdain accorded to participants in horrible crimes, or with sympathy. The Crown chose the latter. A battered and controlled Homolka would shore up credibility. This persona, although a justice necessity, also seemed less mask than reality. It was not contemplated that the videotapes, widely believed to have been destroyed, might reappear. Nor was it contemplated that these would make the nature and degree of her complicity spectacularly apparent at the moment of truth for Homolka and the justice system—the trial of Paul Bernardo. Homolka's assignment of a junior role in the killings had been widely derided outside Canada, fed by the media ban. Although the tapes were seen only by judge and jury (and by Bernardo, who kept asking that choice bits be replayed), they were heard by the entire courtroom.14 Seeing may be believing but hearing is more engaging, as media theorist Marshall McLuhan and violence theorist Lonnie Athens (1989) observe, and it was here that the image of Homolka-as-prey fully shifted to Homolka-as-predator in the public mind. In defending her medico-legal persona against public opprobrium, Galligan gives center place to the psy disciplines.

Galligan sets out the reconstruction process. Homolka had been seen by ER physicians in St. Catherine's on her admission to hospital and her lawyer had insisted on a psychiatric evaluation before going forward with the plea bargain. She was admitted to a Toronto hospital on 5 March 1993. Her stay of over six weeks for evaluation and treatment was much longer than expected. She was placed under the care of a consulting psychiatrist in prison and was evaluated by two psychiatrists and two psychologists chosen by the Crown. Psy reports stress anxiety and depression, learned helplessness, post-traumatic stress disorder, "pyschic numbing" (p. 152), lack of affect and other sequelae of psychological torture and abuse—in short, battered women's syndrome. All agreed that Homolka showed no evidence of masochistic tendencies, psychosis or sexual "deviance". Her abuse was chronicled by category—physical, sexual, 'psychological including evidence of domination" (p. 155)—in relation to the murders, assaults and Bernardo's extramural rapes. One psy evaluator writes that she was "hoodwinked and intimidated [and] believed that she was trapped in the same manner that an abused wife considers herself to be trapped and then having to fend for her life" (p. 80). Another explains that she is a "concentration camp survivor".

Indeed Karla's experience since her age 17, could be to some degree, compared to the experiences of concentration camp survivors who as well experienced horrendous tragedies and had to go through and perform actions in order to survive that under normal circumstances they would clearly have stayed away from, but in the interest of self-preservation or in the interest of preserving other people's lives, did see themselves [as] helpless and went through the actions as had been required of them.(p. 77)

"Indeed," one writes, "she was a victim herself" (p. 81).

It is "simplistic" to expect a battered woman to easily leave or refuse the demands of a spouse, Galligan observes (p. 105). Citing R. v. Lavallee (1990), 76 CR (3d)329, in which the Supreme Court of Canada recognized battered women's syndrome in the law of self-defence, he notes that the case casts light on her conduct but cannot excuse it. Psy findings meet lay definitions of compulsion but cannot support a legal defence of mental disorder, duress or necessity in Homolka's circumstances. Immunity from prosecution could not be granted. Psy evaluators helpfully agreed. Homolka was "technically of sound mind and free of disease of the mind of sufficient severity to cloud her awareness or cause her to be unable to appreciate the nature and quality of her acts" (p. 81), one report concludes in a deliberate echo of the criminal law definition of mental disorder. The report of psychologist Peter Jaffe, a Canadian authority on spousal violence, is quoted extensively. Jaffe spent ten hours interviewing Homolka, reviewed other psy evaluations and consulted Lenore Walker, author of The Battered Woman Syndrome, the book on Homolka's shelf Homolka reports a clear and consistent history of emotional, physical, and sexual violence that trapped her in an abusive relationship. In our opinion, she fits all the criteria for the battered woman's syndrome. In our views, she was groomed, by Mr. Bernardo, to become involved in increasingly bizarre and dangerous behaviour ... Mr. Bernardo involved her in three deaths that gradually increased her role and participation. According to Ms. Homolka, Mr. Bernardo's next instruction was for her to kidnap and kill someone by herself. (p. 156)

Pivotal to legal and psy evaluations of Karla Homolka is the study "Compliant victims of the sexual sadist" (Hazelwood et al., 1993). Where Galligan falters, unable to account for the gap between the degree of Homolka's abuse and her abuse of others—the gap between moral compulsion and moral agency—the study is the missing link. The study also closed gaps for psy evaluators. While conceding her abuse and duress, a psychiatrist retained by the Crown writes, "Karla Homolka remains something of a diagnostic mystery. Despite her ability to present herself very well, there is a moral vacuity in her which is difficult if not impossible to explain".15After reading the study, he found "a number of parallels with this case" which were "so striking as to be worthy of particular attention" (p. 155). Yet his conclusion, if any, is not recorded.

The study documents technologies of compliance used to control domestic partners by sexual sadists, defined as men sexually aroused by pain and degradation. Sexual sadists believe that women are by nature morally corrupt and corrupting, and "prove" this corruption by degrading and sexually torturing them. Their partners, having demonstrated their corruption by engaging in sexual perversity, "deserve" further punishment. This induced compliance confirms sadists' beliefs about women. Researchers interviewed seven partners of sexual sadists identified in the FBI database. All were sexually naive "nice" middle or upper-middle class women, professionally successful and of higher socioeconomic status than their partners. They were "non-aggressive" and "guilt-ridden" about the relationships, which lasted from three months to 13 years (Hazelwood et al., 1993, p. 475). As children, four women had been sexually abused, two physically abused and six psychologically abused. Compliance was induced in stages, beginning with romance and seduction and progressing to degradation and brutalization. sexual activities were introduced with psychological reinforcement, positive for compliance (compliments and attention) and negative for refusal (pouts and rejection). Regimes of increasingly brutal sex, social isolation and violence followed. Anal sex was the preference of all sadists in the study. Forced object penetration (usually anal, with such objects as a 12-cell flashlight), "scripting" women to say things pleasurable to their partners and recording in photographs, audiotapes and drawings were characteristic. All women reported painful prolonged bondage, all were beaten, all were bitten, six were strangled, five were whipped, four were hung, four were burned, three were forced into sex with others and four were forced into criminal acts, one pleading guilty as co-conspirator in the abduction and murder of victims acquired for her husband, for which she served five years. Tangled matrices of childhood experiences, low self-valuation and recent experiences of rejection, it seems, made these women vulnerable to manipulation. Sadists become expert at finding vulnerable women. At the centre of such relationships is not pain but control. The study quotes a sexual sadist.

Sadism: The wish to inflict pain is not the essence of sadism. One essential impulse: to have complete mastery over another person, to make him-her a helpless object of our will, to become her God, to do with her as one pleases. To humiliate her, to enslave her are means to his end, and the most important radical aid is to make her suffer since there is no greater power over another person than that of inflicting pain on her to force her to undergo suffering without her being able to defend herself. The pleasure in the complete domination over another person is the very essence of the sadistic drive.

(Hazelwood et al., 1993, p. 478; emphasis in original)

The shift in pronoun is telling, but the researchers do not comment. Another is quoted on his first murder. "I never thought it would be so easy to kill a person, or that I would enjoy it. But it was easy and I was enjoying the feeling of supremacy. A supremacy like I have never known before."

Psy evaluations are determinative in Galligan's conclusion that Homolka's amnesia respecting Jane Doe resulted from battering and psychological shock or, at least, that this could not be disproved. Evaluators agree on Homolka's symptomatology as a battered woman under duress sufficient to reduce her culpability, but not to exonerate her. She is neither dangerous nor sexually deviant, without Bernardo's influence she would not have offended and she will not re-offend, psy reports assure. Bernardo is not just an ordinary batterer. He is a sexual sadist doing to nice young middle-class women what other sexual sadists have done before. This is the clincher for Galligan and it almost convinces her most sceptical evaluator, although he continued to wonder at her moral vacuity. Homolka is not the ordinary victim of spousal assault and cannot be expected to act like one, because her batterer is not ordinary.

Qualitative case studies are important in evaluating the construction of female deviance and the disciplinary interface of law and psychiatry. Gender distinctions may be lost in large-scale quantitative studies where everyone who seems weird is sent for testing (cf. Menzies et al., 1992). When sex murderers are women and their victims are female children, values surrounding womanhood are doubly challenged. Consort with devils may be Hollywood entertainment or a preoccupation of the unreconstructed religious Right but centuries of such beliefs leave the residual metaphor of witch or vampire, the deadly seductive lamiaand succubi fictioned from the dark side of male sexuality. Beginning with Eve in Western tradition, women are imaged as corrupt and corrupting. The 1486 judicial text Malleus Maleficarum or "Witches' Hammer" was on the desk of every European magistrate for almost 300 years, in the fight against the great conspiracy of witchcraft against the Christian state. "All wickedness is but little to the wickedness of a woman. Wherefore S. John Chrysostom says on the text, It is not good to marry (Matthew 19: 10): 'What else is woman but a foe to friendship, an unescapable punishment, a necessary evil, a natural temptation, a desirable calamity, a domestic danger, a delectable detriment, an evil of nature, painted with fair colours!'—" (Kramer & Sprenger, 1971, p. 43). Reference is made to witches sacrificing infants to the devil and promising children to the devil for lustful uses. The Malleus reserves the extreme penalty of burning at the stake to advocating contraception and performing abortions (which deny paternity) and robbing men of virility (which denies the possibility of paternity). If men achieve the neutered state through prayer it is, however, a virtue. Consideration is given to whether the devil through incubi or succubi can engender children but the authors decide that only God can give life through women. "To conclude. All witchcraft comes from carnal lust, which is in women insatiable" (Kramer & Sprenger, 1971, p. 47).

Freud's reading of the Malleus confirmed his view that psychoanalytical theories of split consciousness recall medieval theories of demonic possession. Disclosure by judicially tortured women in witch hunts, he writes, parallels disclosures of the hysteric under psychoanalysis (Caudill, 1994, p. 46). Medieval witch hunts are a metaphor of late twentieth-century investigations of child sexual abuse. Freud's hysterics are victims of child sexual abuse. In Freud's initial analysis of his female hysterics, he concluded that they had been "seduced" by fathers or brothers in childhood (Wolff, 1988). He withdrew his Seduction Theory under threat of professional expulsion and instead submitted a paper which would dominate therapeutic and legal response to child sexual abuse until the 1970s. His new theory proposed a female Oedipal complex. In love with their fathers and desiring the death of their mothers, hysterics fantasise childhood sex with fathers. These complex turnings of victimisation and offending are now standard fare in "explaining" abuse and abusers and, conversely, in attacking victim credibility.16

Galligan, like Freud, leaves us with an irreconcilable duality. Homolka is an "equal partner" in "the grossest of evil" (p. 221). She may well have been, although we see little of this equality in Galligan's account. Alternately, she is just another battered partner of a sexual sadist, corrupted by his gaze, caught with her victims in a rare tragedy. But that troubling moral vacuity remains. Could Homolka not have lifted up her head for a moment in her five years with Bernardo, worked out the logic, saved a life besides her own? What so insulated her from these children's suffering that she would enable their torture and murder rather than risk her body or her marriage? Barely out of adolescence, she procured her sister for her boyfriend as a Christmas present, killed her by mistake and complained that her parents' mourning interfered with her Princess Di wedding. She procured more victims for Bernardo, twice repeated the deadly Tammy scenario with Jane Doe, videotaped and starred in all events, scrubbed up afterwards and give a father's day dinner while one child's body lay in the cellar of their Barbie-pink house. The victims were children, three are dead and Homolka could have saved them. Without devils, vampires and witches, without the confident assurance of the psy sciences that she was trapped in madness as surely as if she herself were legally insane, what is left? Perhaps we are just surprised at the banality of evil, Hannah Arendt's (1964) phrase which captures Homolka's ordinariness with that of Hitler's minions who fitted themselves to the ends of a charismatic leader. This did not absolve them of agency, as the Nuremberg Trials showed. A psy description of Homolka (above) as a "concentration camp" survivor brings home the terrible domesticity of the death camp and its trustees. More than any other metaphor of vampire or devil, or psy label of battered woman or compliant victim, it conveys the duality of this privileged child caught in iniquity and the inadequacy of metaphor when the camp walls are psychological.

The modicum of Homolka's guilt was calculated at the intersections of law, public opinion and the psy disciplines. That she would be demonised was clear to the prosecution and it took care to shape her to the Procrustean bed of the battered woman. Even so, the parts that could not be lopped off to fit resulted in 12 years in the Kingston Penitentiary for Women. Is she penitent? Does she accept her guilt? What is guilt, where discourses collide? Law is public, coercive. Law's past is objective, a matter of provable fact, not feeling. Guilt is about restoring the public equilibrium through conviction and the retributive satisfaction of punishment. The clinic is a private and invitational world in which the past is a matter of the patient's subjectivity, where guilt is exorcised by restoring internal equilibrium and self esteem, by "getting better".

The discord is evident in a series of leaked prison letters published in the Toronto Sun (Cairns & Burnside, 1994) in which Homolka comments on her law and pay moments. The only death overtly mourned is that of her relationship with Bernardo.

I started seeing one of the psychologists in addition to the psychiatrist last week. I felt like I was on the verge of another breakdown. Why exactly, I don't know. It could have something to do with the fact that an inmate got stabbed a while ago ... I'm going through some really difficult times dealing with the death of our relationship. That's how I'm trying to treat it—like a death—because I think it'll be easier to deal with that way. I am really having a hard time with this. I almost would rather have had him die, because then I'd at least be able to grieve properly. And visit his grave and say goodbye. (5 October 1993)

She moves on to therapy and to the future. "Paul" is an agent not of depravity and sorrow but of "insecurity". A new person is emergent, love is in the future and therapists are in the way.

I am looking so forward to getting out of here and rebuilding my life. I can't wait to see what the future holds for me—a new job, a new husband (... a loving one this time!) children.... Paul made me into a very insecure person. Hopefully that will all change in time. I am trying to really change myself back into a newer, better version of the person I was before I met Paul. I have a new therapist. She and I get along really well and she is really helping me. I absolutely HATE the psychiatrist here now. I only see him every other week now. He told me the last time I saw him that I would never heal until I told him all the details of what happened. Asshole. He is so mean and cruel. So now I don't really tell him anything. (28 October 1993).

Going to court interferes with her rebuilding. Paul is still mourned.

I'm really nervous about the future. When I'm finally released, that is. I know I really should be concentrating on things in the nearer future like court. But I just can't bear to think about such a painful thing anymore. You know it's really not fair. My trial is over and done with, but none of the pain is. Well, the pain will never be gone, but talking about it will just make it raw ... Our life wasn't supposed to end up this way. (24 November 1993)

Regrets about the past yield to hopes for the future, in a therapeutic transition from counseled to counselor.

I am constantly thinking about what my life is going to be like when I leave here. I am going to live so differently. Life with Paul ... as well as being in prison really opens your eyes to how life should be lived. I am going to do the things I always loved to do but never got to do. I'm going to start horseback riding again. I'm going to take lessons and even learn to jump! I also want to do volunteer work—I would love to work on the Kids Help Phone and I've decided what I want to do—work with abused women. I want to work to help prevent women from being abused and also to work with women who have already been abused. I want to go out on picnics with my friends and sister. Life is going to be so great when I'm out of here. I will never take anything for granted again in my life! (21 December 1993).

Life with Paul, life without Paul, life in prison, only one sister left with whom to pinic, her life as preparation for counseling children and battered women, for having children? Homolka will receive a sociology degree from the University of Toronto at about the same time as her parole eligibility, after four years of incarceration. At the time of writing, she has declined to apply for early release. She must be released in 2001. Her family is supportive, her dog has visited her in gaol, her cell is apart from others and decorated to taste, small details which continue to provoke public resentment. She was moved to an "open cottage" federal facility in Joliette, Quebec in the summer of 1997 where, according to press reports, she is popular with other prisoners.

'The truthful rendering into speech of who one is, to one's parents, one's teachers, one's doctor, one's lover, and oneself, is installed at the heart of contem porary procedures of individualization' (Rose, 1989: 240, citing Foucault, 1979). Confession for Homolka consisted of 400 hours with police and lawyers, 17 days on the witness stand, countless therapeutic hours—the truthful rendering into discourse of who she is. Does she know? Do we? The videotapes suggest something less or more than truth but the camera often lies. Her prison letters suggest something different, raising again that troubling moral vacuity, that fractured persona. Should she be left to her privacy, a curtain drawn over recovery, or is there something still to be learned from this 'very black case' (p. 220)?

Psychoanalysis reveals the irrationality of the body and its desires behind the mask of the intending subject of law, putting the lie to law's constructs of reasonableness. Love and desire, at the extremes represented by Homolka, are not reasonable. Without nuanced explanation of desire and its lack, law is left with simple dichotomies—innocence and guilt, good and evil, competence and mental disorder. There is a little room for shading even in such complex and endlessly refined medico-legal constructs as battered women's syndrome (cf. Shaffer, 1997). To fit law's Procrustean bed, too little love is racked to fit and excesses of desire are lopped off. Law is the robber of parts and the brutal restorer of lacks. Homolka could not be made to fit the battered woman bed, however her persona was stretched or shortened in medico-legal discourse. Her partial defence of duress, her partial agency, her partial prison term and the fragmenting of her meaning in law, in her self and in the public gaze are an invitation to see her meaning as constructed with and against that of Paul Bernardo and, ultimately, in terms of Western images of womanhood.

Bernardo employed all the control strategies described in the "Compliant victims" study, perfected over time. He learned from his mistakes with J.M.G. and I.F., earlier girlfriends. He explored the outer envelope of his attractive powers and his sexual preferences with the victims of his rapes. He found or made a devoted pupil in Homolka. He took legal advantage of the institution of marriage, with its privacy and conferral of spousal immunity, bearing traces of coverture. Homolka could not be compelled to testify against him. Bernardo was canny enough to save his confession of being the Scarborough rapist until his wedding night, and to involve her directly in procurement and murder. Coverture shuts doors. Battering, with its control of will, act, meaning and identity, shuts more (McGillivray, 1987; McGillivray & Comaskey, 1999). All shade into sadism, which is little more than the extremes to which control is taken. Psychopathic sexual sadists, one supposes, have no limits.

The "Compliant victims" study fails to account for a central finding. Respondents " 'fell' for the men relatively quickly, even though they recogniseda sinister side to them" (Hazelwood et al., 1993, p.477). Why? Is "low self-esteem" sufficient explanation for Homolka's pupilage? Was she seduced into submission, a slave without will or agency under Bernardo's Svengali magnetism? Was she attracted to his "sinister side" and an active, willing partner in some folie a deux? Was she the dominant partner, as Bernardo alleges, who led him into murder?

Patricia Pearson (1995,1997) offers a vigorous counter-argument to the battered woman compliant victim theory. She supports this with evidence of Homolka's robust participation, even leadership, in videotaped acts and dialogue, her maintenance of the sham of her marriage, her lack of visible injury until the end of the relationship (co-workers' evidence to the contrary is not cited by Pearson) and numerous slippages and inadequacies in explanatory aspects of her testimony. Homolka is a fiercely competitive narcissist, jealous of her virtually identical sisters and Bernardo's wandering eye. She made herself the centre of his attention by her direction and willing participation in the fulfilment of his desires. "The documentation of bruises [tracking Homolka's abuse with the rapes and murders] was a fool's arithmetic" (Pearson, 1995). The last beating was the only serious one, an announcement by Bernardo that the relationship was over. The blame, Pearson argues, lies with reductivist legal and academic feminist discourses which prefer to believe that women who kill are victims of the Svengali influence of men. Such discourses refuse to confront female aggression. Women can be evil, too.17

This is not news, even to feminist legal academics. But too much agency is problematic. Demonising offenders like Homolka creates a hyper-agency resulting in excessive prosecution of women—witchcraft, infanticide and concealment of birth, prostitution, abortion, contraception, the dismissal or demonisation of rape and child sexual assault complainants. Demonising women ignores the impact of intimate violence in childhood, adolescence and adulthood. It ignores women's social conditioning and the disordered lives of female offenders (Allen, 1987). Are women legally vested with too little agency? Basing her study on English cases of crimes against the person, Hilary Allen (1987) concludes that female offenders are disproportionately psychiatrised. Going beyond notions of judicial leniency and systemic bias, she asserts that, in the proper working of law and medicine, it is not a matter of preferring a psy disposition for women over a legal one but rather that women more often are seen to meet the theoretical and practical criteria18 for a medical disposition than are men. The discrepancy is produced by the interaction of the medico-legal structure—diminished responsibility in homicide and infanticide, for example—with structured and gendered ideas about human nature. Medicolegal discourses do not lie in words extraneous to what happens to women. These discourses are "ways of understanding, deciding and doing things; they are themselves the machinery of power in which both professionals and their subjects are equally enmeshed" (Allen, 1987, p. 111). The remoteness of legal and psychiatric discourses to the lay mind and everyday reason is amplified when these discourses interact and results may seem bizarre and shocking. Theorising about how men and women are treated, Allen asserts, may lead to a more balanced view of offending, bringing together understandings of women's painful and disordered lives with notions of evil and agency in men. However distorted one's inner life may be, she concludes, legal responsibility for lives lived in the material world should not be erased by psychiatry. Her argument recalls, as does Homolka's case, older debates about the relationship of law and pyschiatry.

Psychiatry forged in the eighteenth and nineteenth centuries in the specular and spectacular context of the asylum was entwined with legal ideas about criminal responsibility and the development of the criminal subspecialties of criminality and penology (Szasz, 1963; Pick, 1989; Cameron & Frazer, 1987). Psychoanalysis, the Freudian revolution, extended the clinical field of the abnormal into the normal and the unspectacular (Szasz, 1963; Stone, 1984; Rose, 1989). Although psychoanalysis implicates the abnormal in assessing the normal, Freud did not believe that psy insights should deny or even mitigate criminal responsibility, nor did he believe that such insights explain criminal behaviour. In his 1931 'Expert Opinion on the Halsmaan Case' in which a university student was charged with the murder of his father, Freud writes that "mention of the Oedipus complex has a misleading effect; it is at the least idle.... Precisely because it is always present, the Oedipus complex is not suited to provide a decision on the question of guilt" (qtd., Szasz 1963, p. 105).

Psychiatrist Thomas Szasz goes further. Law needs psychiatry to shore up its normative mission. "By declaring the criminal mentally ill, the society declares it unthinkable for any sane individual to be alienated from the wider community of Americans and uncommitted to its Puritan values" (Szasz 1963, p. 107). Psy findings deny moral choice. The morality of medicine is substituted for that of law and the coercion of the penal system is traded for that of the asylum. Psy findings relieve judges of the ultimate burden of apportioning guilt and judges rely on the psy disciplines to assuage their own guilt. Freud and Szasz assume that psychiatric discourse has the potential to subvert criminal justice by negating legal fault with notions of the unconscious. What was not contemplated by either law or psychiatry is the commonplace role now held by the psy disciplines in criminal discourse and disposition. If anything has been subverted, it is psychiatry by law, in its routine use of psy discourses to justify decisions at a micro level. As Freud asserts, to explain everything in psy terms of "complexes" and the unconscious is to explain nothing. Even so, women offenders continue to "present" as mysterious to the law, as weird cases explicable only by the psy disciplines.

Is the system biased toward the psychiatrising of women or is such data as Allen's distorted by its reliance on rare and often sensationalized crimes against the person committed by women? A Canadian study of Metropolitan Toronto Forensic Services (METFORS), which offers pretrial clinical evaluation of accused persons and processes 600 to 800 people each year, sought to demonstrate the relevance of gender to medico-legal outcome (Menzies et al., 1992). The study found no statistical differences in psyattribution—mental disorder, treatment history, dangerousness, criminality—that could be explained by gender. Rather, there was "an unexpected uniformity in the quality and scope of forensic encounters that appeared to transcend gender categories" (Menzies et al., 1992, p. 189). The different outcomes of Allen's study and the METFORS study may be explained by a shift in focus from crimes against the person in Allen's study to all crimes in the METFORS study, by the mass case processing of the Toronto clinic and by differences between UK and Canadian treatment of diminished responsibility in criminal law.

Culpable homicide is excused in circumstances of self-defence and mental disorder but not in circumstances of duress or necessity. In Canada, such circumstances can only mitigate sentence. Murder offers little sentencing discretion, pushing plea bargaining for manslaughter. English law recognises a defence of diminished responsibility, a subcategory of mental disorder reducing murder to manslaughter. Under English law, Homolka might well have taken her chances. At the centre of the debate, always, is agency and the capacity for choice. Freud and his followers demonstrate the fragility of this concept, yet stoutly defend its application in criminal law. Feminist analysis and family violence studies show how agency is subverted for women caught in violent intimate relationships. When a young Winnipeg woman, Lyn Lavallee, shot her abusive lover in the back of the head—he put the gun in her hand and said do it or she would be dead in the morning; she aimed out but missed—the Supreme Court of Canada cited psy findings and feminist theorists and found reasonable self-defence in the circumstances of battered women. The case acknowledged the need for psy evidence to explain the gap between social mythologies of wife-battering and the subjectivity of battered women (R. v. Lavallee (1990), 76 CR (3d) 329).

The defence of "battered women's syndrome" cuts two ways (Martinson et al., 1991). Moral will is eroded by intimate violence. This erosion lowers or erases agency and criminal responsibility. Alternatively, it syndromises women, reducing the social roots of battering and women's inequality to an individualized medical condition and thus relieving society of the burden of addressing those roots. Battered women's syndrome is a wedding of feminism, psychiatry and law, a new addition to the taxonomy of culpability in the old alliance of law and the psy disciplines. It is rarely a complete defence (Sheehy, 1994). Usually, as in Homolka's case, it figures in charge and sentence reduction. Had Homolka killed Bernardo, it is likely that she would have been exonerated. She killed others and this falls outside the parameters of all legal defences. Bernardo and Homolka were assigned medical identities—battered woman and sadistic psychopath—leading to different legal results. Battered women's syndrome diminishes legal fault. Psychopathy, the inability to feel another's pain, does not. It does not meet the definition of mental disorder in §16 of the Canadian Criminal Code (see R. v. Kjeldson [1981] 2 SCR 617), nor does it elicit the moderated compassion ("we all might have been there") which reduces sentence. If anything, the effect of a diagnosis of psychopathy is to heighten guilt. Psychopathy is untreatable, psychopaths shows no real remorse and are likely to reoffend and long-term incapacitation through incarceration seems inevitably appropriate.

Psychopathy is a widely cited explanation for sadistic sexual violence and especially for sex murderers. Yet explanations of sexual violence in lay mythology, criminology and medicine are contradictory and incoherent, as Cameron and Frazer (1987) argue. In their view, psychopathy is a convenient dustbin to which men who do unpleasant things are consigned. It has no meaning, its definition is tautological and it impedes feminist Reconstruction theories of male sexual violence. The aim of feminist Reconstruction, Cameron and Frazer assert, is to provide an alternate account of sexual violence which will enhance feminist politics and reconstruct desire. Sex murder will no longer be a metaphor for freedom and for the ultimate possession of the object of desire. What it will be, the authors do not say, but they are right to focus on psychopathy as a problematic medico-legal designation. Psychopathy has historically been overused, under-researched and incoherently defined. There is now some support for a theory of psychopathy based on brain function and childhood experiences of violence. Such a theory, when fully developed, might combine the work of Canadian researcher Robert Hare, based on CT scanning of prisoners identified as psychopathic and demonstrating significant differences in brain function, with the outcome of multiple experiences of violence in childhood (Miller, 1983; Athens, 1989).

We know little of Bernardo's childhood or his psychoanalysis. According to press accounts, he lived in his parents' working- or middle-class Scarborough, Toronto neighbourhood until age 26, in a room papered with upward-mobility positive-thinking mottos, a copy of Miss Manners on his beside table. There is mention of a mother who chased and terrified neighbourhood children, of perpetually curtained windows and much shouting, of an ordinary sister and brother and the parental nicknaming of Bernardo as "worthless bastard". He was in fact conceived in an extramarital affair. There is a quote from his father (Blackwell, 1995). "What family isn't dysfunctional? Everybody wants to be a victim, everybody wants a scapegoat. How about looking in the mirror and blaming yourself? ... Tumultuous? So what? There weren't beatings. You have three kids, you yell a lot." Psychologist Peter Jaffe's report on Bernardo for the Crown describes his parenting as harsh and inadequate. Bernardo legally changed his patronym to Teale in 1993, the year his father went to prison for sexual assaults committed two decades earlier.

Other than these few facts culled from news stories, we know little of Bernardo and his medical labeling. We have access to the results of Homolka's analysis through the Galligan Report. We know more about her than we do about Myra Hindlay, her English counterpart, also convicted of the sex-related homicide of children in concert with her "psychopathic" boyfriend Ian Brady (Cameron & Fraser, 1987). Hindlay is the only major counter-example offered by Cameron and Frazer to the presumption that sex murderers are men. Like Homolka, Hindlay is explained as either corrupted by her partner—the Svengali or Pygmalion effect—or as monstrously evil. No "battering" or "compliant victim" theory was offered for Hindlay, as none existed in 1966. With Hindlay, Homolka is accused of being the dominant partner. With Hindlay, Homolka stands with her victims in her partner's sexual gaze, and is so recorded in photographic mementos.19

In Homolka's case, legal discourse is determinative but psy discourse is of the essence. There are signs of struggle. Jaffe's psy evaluation of Homolka, for example, pushes for a finding of no legal agency. Other pay evaluators, more compliant with law's assertion of place, resist battering as a complete explanation. The case confirms the symbiotic relationship between law and psychiatry. Law and psychoanalysis, Freud (1906, p. 108) writes, have in common a concern "with a secret, with something hidden.... In the case of the criminal it is a secret which he knows and hides from you, whereas in the case of the hysteric it is a secret which he himself does not know either, which is hidden even from himself." Homolka's secrets are hidden from the justice system and must be discovered by the psy sciences. The need for these secrets is the ultimate justification of the plea bargain.

Homolka, Galligan concludes, should have stood with Bernardo in the prisoner's dock facing charges of first degree murder. Had the videotapes been found in time, there would have been no plea bargain. Had Homolka disclosed the Jane Doe assaults prior to the bargain, her total sentence under the plea bargain would be 14 or 15 years rather than 12 years, on the principle of totality of sentence. The difference did not justify voiding the bargain and risking Bernardo's acquittal and "there could be no room for error in the successful prosecution of Paul Bernardo" (p. 323). Further, "there is one immutable principle which stands out. It is that the word of the Crown is its bond. A case as horrible as this one, where public feelings run high against a particular individual, sorely tests one's principles. But if the principle is right, then it must prevail, even in this case. Because if that principle is abandoned in this case, then who can be sure that it will not be abandoned in another, and in another, until the time arrives when the word of the Crown is mere dross" (p. 209). The Crown made a tactical decision in full knowledge after weighty consideration of all known facts. "Distasteful as it is, the practice which has existed for over three hundred years of giving immunity or a 'discount' to an accomplice to obtain her evidence against a co-perpetrator is sometimes a necessary one and is a legal one" (p. 215). It was "a tactical decision which turned out to be right".

The dark side resurfaces in Galligan's epilogue. Homolka haunts the narrative but she herself disappears. It was a "very black case" (p. 220) for its gruesome evidence, its toll on victims and families and the damage caused to the health (and careers) of police officers and lawyers who investigated, prosecuted and defended it. There is a "bright side" in the coping of the victims' families with "tragedy and hurt of a degree which no human being should be asked to bear [proving that] the good in human nature can surmount even the grossest of evil, that evil which touched their lives" (p. 221). (Homolka's mother, at first sharing with other mothers the tragedy of all these lost daughters, was shunned when the videotapes were played at trial. She is also absent from Galligan's account.)

Homolka's demons are perhaps being exorcised in her thousands of hours of confession, a gift of the plea bargain. She remains caught in the maelstrom of public demonisation. Sympathy for this devil is constrained by her moral vacuity, by her desire for the Hollywood order of things—cute boyfriend, great looks, fancy wedding, perfect marriage, neat house—inverted, a wedding and three funerals, four if we count the only death she seems to mourn, that of her relationship with Bernardo."Our life wasn't supposed to end up this way", she writes (above). The gap between law and the psy disciplines in the face of the problem of evil in women is painfully apparent in Galligan's struggle to comprehend and contain the findings of law's junior discourse. The fracture between criminal justice and social notions of justice is wider than ever.

The Ontario justice system has spent the years since Bernardo's arrest exorcising its own demons. These are the system failures examined in the Campbell Report (1996). The result is a massive overhaul of the system, from policing and information-sharing to the processing of forensic evidence. The Mahaffy and French families are pursuing their own demons, taking their claim for the sealing of the videotapes to the Supreme Court of Canada20 but the videotapes may be needed in Bernardo's appeal. They also may be needed in the prosecution and discipline of his first lawyer, Ken Murray, a last demon of the justice system. Murray removed the videotapes from Bernardo's house on his client's instructions in May 1993. In the summer of 1994, he sought the advice of the Professional Conduct Committee of the Law Society of Upper Canada. He was advised to turn the tapes over to the court and to withdraw from the case. He did so in September of that year (above). In January 1997, he was charged with obstructing justice, possessing child pornography, conspiracy and making obscene material, in copying the tapes (Hess & Grange, 1997). All but the charge of obstruction were later dropped21. A complaint of professional misconduct was laid the following month, after two years of investiga tion by the Law Society of Upper Canada (Gorham, 1997). A paramount duty of the lawyer is the protection of confidential information. Real evidence—the knife, the gun, the robbery proceeds—is not subject to solicitor-client privilege. The trick is how to turn the evidence over to the authorities without compromising the client's identity.22 As the videotapes proclaim Bernardo's identity, this arguably was not possible, but as the tapes do not show the killings, were they evidence of murder? Did Murray intend to obstruct justice or did he only intend to serve his client? Or does client service only furnish the motive for obstruction of justice? These are the narrow legal issues.

The broader ethical issue emerges from the widening fracture between two views of moral lawyering. Does professional duty override the duty to obey the law? If so, should the lawyer be prepared to lose his or her licence in defending the client, even go to gaol? Should lawyering be civil disobedience? If so, should we require lawyers to pay the price? The adversarial model Murray followed is an extreme variation on the "neutral technician of law" (McGillivray, 1999). The model stresses the paramountcy of the lawyer's duty to the client to the exclusion of all other interests and duties.23 It is strongly defended in criminal defence work even by those who reject it in other legal arenas. Rule 10(10) of the Ontario Professional Conduct code states that 'the lawyer's duty is to protect the client as far as possible from being convicted' (Law Society, of Upper Canada, 1997, p.36). The lawyer must disregard "private opinion as to credibility or merits" and can rely on "all available evidence or defences including so-called technicalities not known to be false or fraudulent". Above all is the duty of confidentiality, wider than solicitor-client privilege at law. Although Canada abolished the death penalty in 1976, defence ethics are increasingly driven by cross-border US cultural influences. Canadian defence lawyers are absorbing a radical professional consciousness developed in the shadow of hanging, lethal injection and that useful invention by a US dentist, the electric chair.

The adversarial model of lawyering is set against the 'ethical model' which seeks to balance duty to the client with duty to the court and the law, to victims and their families, to the administration of justice and public service, through and beyond the immediate needs of the client.24 Variations on the model divide almost equally into calls for a return to some golden age of lawyering, in which lawyers are internally governed, and calls for stricter professional governance. Alternatively, a new lawyering is promoted, in which the lawyer is transformed by law by identifying through the client to a just cause. Concern for professional image, recognition of the impossibility of being at once neutral and ethical and a sense of a lost craft or art or moral or civic high ground run through the debates. In concealing the videotapes, did Murray break a lesser law in obeying a greater ethical duty to his client, or was he centrally mistaken as to what legal ethics require? Until this question is answered by the courts and the Law Society of Upper Canada, the Homolka case cannot be put to rest, safely staked out at the crossroads of law and legal professionalism.

What do we learn from Homolka's case? Criminal justice is pressed to the extreme, driven by the rarity of the female sex killer—what to do with her, how to see her.,—and the need for evidence against Bernardo, a far more serious threat to female children than his partner ever was. In the prosecution and professional discipline of Ken Murray, an ordinary court-appointed lawyer caught up in the bizarre, the limits of client advocacy will be tested. Homolka's bargain demonstrates the importance of the practical ethical standards set out by the Martin Report (1993) in everyday plea bargaining. Galligan delineates the bargaining process in cold prose standing in stark contrast to the horror to which it refers but he demonstrates a practical, ethical means for retaining important facts where plea bargaining has "ditched" the charges. Disclosing to the sentencing judge all aggravating factors even where there is no related charge is of central importance under Canada's new sentencing guidelines.25 Even so, the larger ethics of bargaining with a sex killer were so deeply challenging that legal actors withdrew from the case due to profound discomfort with Homolka's remaking and the dreadful facts which slowly emerged from her evidence and from forensic findings.

Centrally, Homolka's case is a morality tale about good and evil and the complexities introduced into this simple dichotomy by a multiplicity of discourses, each claiming truth. The psy disciplines are no longer engaged in hegemonic struggles over the field of criminal culpability. They are now law's shy handmaids, invitees on its premises, dismissible at will. As Josef K. is told in The Tnal, "It accepts you when you come and it dismisses you when you go".26 With Galligan's dismissal of the psy disciplines go the complex meanings of Karla Homolka.

In the course of writing this article, I have been challenged to give my opinion on the guilt of Karla Homolka. With every woman familiar with the case—younger women usually refuse to discuss it, sickened—I believe that I would at some point look up from my own condition, however brutally constrained or blindly self-construed, to save a life besides my own. I do not know if this is true. I hope it is. I have not been there. I am grateful that my daughter did not meet Paul Bernardo and I think every mother and daughter who did not also should be grateful. I hold no brief for her, I do not understand her, I have never seen her, but I do know, as a student of intimate violence, how far and how deeply women are conditioned by social institutions to distrust and dislike themselves and to turn to male authority and its reflection in female adolescent goals to be defined as worthy. If the definition of worthiness is thin—looks, lifestyle, social status, getting a cute guy with charismatic power to marry you to the envy of other women—it is less a problem peculiar to Homolka than a set of standards peculiar to our age. How do Homolka's goals differ, in essence, from those of Jane Austen's protagonists? Austen's characters search for moral probity in marital partners but this seems a lightly troubling matter, more concerned with the avoidance of cads. The gentleman cad of the eighteenth century novelic tradition becomes the vampire noble of the nineteenth century. Best known is Stoker's Dracula (1897), with its interrogation of lawyering and its depiction of the giddy young Lucy drawn easily into the seductions of the vampire to become vampire, taking the blood of children, achingly delivered of evil in her staking (McGillivray, 1999). Why don't we blame Bernardo, cad and vampiric seducer? Why do we blame Homolka for being seduced, when seduction is so necessary to our own romance?

Bernardo did not spring full-grown into evil from the frontal lobes of an avenging god with a bad headache, an apotheosis of Athenic wisdom in some post-edenic birth into evil. We do not know where he came from. We do not know whether his diagnosis of psychopathy is "real" or just a medico-legal label on the dustbin into which we toss cases of inexplicable motivation. We have some evidence that there are brain response patterns associated with psychopathy but we do not know if these are induced or congenital. We do know that the direction taken by those who do not feel the pain of others is conditioned by childhood. Bernardo might have succeeded as an ordinary sociopath working his way up the corporate ladder but he chose a different route, the humiliation, rape, torture and murder of adolescent women. He blames Homolka for the deaths but murder is not the least of what he did. His excuse may lie in bad parenting which, like low self-esteem, is another modern disease. Neither excuses criminal conduct but both, we now know, are generative of such conduct.

We ignore at our peril the lessons of childhood. Childhood is a subservient and outsider social status and individual children are picked out as special bearers of this message, as were Bernardo the bastard and Homolka of low self-esteem, somewhere in their childhoods. The status and treatment of children must be called into question in every case involving the injury of children. While law cannot answer to all subtleties of inequality and victimization, to child or spouse to spouse, law can take steps to improve the way we treat children and youth. However much we love to hate Bernardo and Homolka, neither of these once-children had to go this way and nor did their many victims—the victims of Bernardo's rapes in which Homolka acquiesced by not reporting, the children she procured, the children in whose torture and death she so actively participated. These children we refuse to forget. Somewhere we must believe in a redemption of childhood. We are left otherwise with a hopeless and recurrent evil.

I am grateful to W. Wesley Pue for his encouragement and helpful comments in the writing of this article and to research assistants Rekha Malawaya and Pierette Hebert for their insight into the 'Homolka problem' and their assistance in bringing this article to publication.

Notes

[1] On sexual connotations in the Bulger killing, see Freeman, 1997.

[2] The associations have a long legal pedigree. Women's weakness (imbecillitas), vulnerability (infirmitas), cunning (calliditas) and deception justified their legal disability under Roman law which defined women in the negative, as not being what men are. Women were also said to be prone to seduction and to deception in the senses both of deceiving and being deceived (and perhaps seducing as well as being seduced). Other recurrent themes are fragility (fragdatas) and frivolity (levitas). The images are taken up with vigor by seventeenth-century English legalists. On women, law and images in a psychoanalysis of law, see Peter Goodrich (1995) at 115 et seq. and Chs 5 and 6.

[3] In her Sisters in Crime (1975, cited in Pearson, 1997), criminologist Freda Adler hypothesized that female aggression will equal male aggression as women adopt male means of self-empowerment. Her "liberation hypothesis" has not been borne out. Statistics collected from United States studies of woman convicted of homicide suggest that slightly more than half killed persons outside the domestic circle. A study based on the population of Bedford Hills Prison in New York State in 1994 shows that of 254 female convicts, 12% killed in the course of a domestic dispute, 10% killed children through neglect or abuse and the remainder killed out of sexual jealousy (surely this should count on the domestic side?) or in drug-related and other business disputes. A review of prison records of homicidal women in an Alabama women's prison from 1929 to 1985 shows that 95% of stranger and business homicides committed by white women were committed after 1970, as were 60% of family and friend murders. These studies are cited in Patricia Pearson (1997). On problems with Pearson's work, see note 17.

[4] "The Bernardo case, like every similar investigation, had its share of human error. But this is not a story of human error or lack of dedication or investigative skill. It is a story of systemic failure. It is easy, knowing now that Bernardo was the rapist and the killer, to ask why he was not identified eadier for what he was. But the same question and the same problems have arisen in so many other similar tragedies in other countries. Virtually every inter jurisdictional serial killer case ... demonstrate] the same problems and raise the same questions. And always the answer turns out to be the same systemic failure. Always the problems turn out to be the same, the mistakes the same, and the systemic failures the same." (Campbell, 1996, p. 1).

[5] I surveyed reports filed by news services (six disks, on file with the author), evidentiary motions in the Bernardo trial and the Homolka sentencing transcript. The facts as set out here as far as possible remain close to the Galligan Report. A few inconsistencies have been corrected and I have tried to resist the temptation to include "extras". Materials not in the Galligan Report are placed in brackets. Detail illuminates yet sickens. Avoiding the pulp literature flowing from the case has been a pleasure.

[6] The victims' parents were represented by independent counsel in the investigation and prosecution of the case and the plea bargain. Doe and her mother agreed that no charges should be laid with respect to Doe's assaults. "Her unwholesome association with Paul Bernardo and Karla Homolka consumed a year and a half of her young life. The ensuing investigations and trial, with the terrible pressures which they exerted upon her, covered another two and a half years of her life. Now, almost five years after she first went into that infernal place which was 57 Bayview Drive, she and her family have asked simply that she be left alone." (Galligan, 1996, p. 213).

[7] Quoted in Williams (1961).

[8] The agent was John Douglas, whose work inspired Silence of the Lambs (Makin, 1995). "Criminal probing" is based on weapon, killing site, position of the body and minute information about the victim and gives an "experiential" rather than "scientific" portrait of the killer's habits, age, marital status, occupation and personality traits. Data was compiled from FBI files and interviews with, among others, Charles Manson, David (Son of Sam) Berkowitz, Sirhan Sirhan, John Wayne Gacy and Richard Speck. Serial killers may seek opportunities to dominate, manipulate and control because their lives have been dominated and controlled and "for the first time, they've got somebody. The thrill for them is to see that person begging for their life, tears streaming down their face. Then finally they throw the switch and kill them." The metaphor, "throwing the switch", is derived without comment from the electric chair. While videotaping is a recent trend, recording and scripting are not. "They usually script a woman, talk to them while having sex with her and use filthy language. They script them to say things that are pleasurable to them." Douglas's profiling was less accurate in the Canadian case of Guy Paul Morin, arrested in 1985 for the murder of Christine Jessop and exonerated by DNA testing after serving almost ten years in prison. Lies of witnesses continue to be acknowledged. As a result of the Morin disaster, Douglas no longer supports the death penalty: by the time he was exonerated in Canada, Morin would have been executed had he lived in almost any US state.

[9] R.v. Bernardo [voir dire, admissibility of evidence of spousal abuse of Karla Homolka] [1995] OJ No. 1380 DRS 95-14624 Court File No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 17 May 1995 (pare. 37).

[10] R.v. Bernardo [voir dire, admissibility of psychiatric evidence on Karla Homolka] [1995] OJ No. 2249 DRS 95-15614 Court File No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 27 July 1995 (pare. 8).

[11] R.v. Bernardo [voir dire, admissibility of evidence seized in Bernardo's home] [1995] OJ No. 1394 Court File No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 17 May 1995 (pare. 4).

[12] R.v. Bernardo [voir dire, admissibility of psychiatric evidence on Karla Homolka] [1995] OJ No. 2249 DRS 95-15614 Court File No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 27 July 1995.

[13] Galligan discusses plea bargaining or "resolution agreements" in some detail, concluding that it is entirely acceptable in principle and Homolka's bargain was appropriate and appropriately managed, meeting the standard set out in the Martin Report (1993). The Report recommends full disclosure of the agreement and of all aggravating factors, irrespective of charges laid, and stresses victim participation by way of impact statements and consultation on the prosecution of the case. Galligan gives as example the Ontario case of R. u Helmut Buxbaum (1989), 70 Criminal Reports (3d) 20. "We do not live in a Utopia", the judge observed, and "the so-called deal happens thousands of times a day in Canada" (p. 68). Buxbaum, a wealthy businessman, hired a hit man who hired hit men who hired helpers to kill his wife. In this surfeit of murderers, some were guiltier than others and Buxbaum as prime mover and his first accomplice were guiltiest of all. Galligan was counsel for one of the hired hit men and obtained for his client a large sentence discount for testifying for the Crown.

[14] R.v. Bernardo [1995] OJ No. 1472 DRS 95-16117 Court No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 29 May 1995 (pares 124, 125). "Although it is difficult to rationalize why the verbal, but not the pictorial, images may be publicly displayed, I can only try to analogize as I have in an eadierruling that traditionally we do not display, for public viewing, photographs of dead bodies, close-up photographs of wounds, photographs of autopsies, photographs of exhumations, and similar type evidence. Nevertheless, viva voce evidence of the witnesses who describe what is depicted in the photographs referred to is always heard in open Court. When Capital Punishment was in effect, in Canada, the executions were not carried out in public. This practice and policy was not because persons who voluntarily chose to attend would be shocked or horrified, but rather, in my view, because we, as a society, believed that it debases humanly."

[15] R.v. Bernardo [voir dire, admissibility of psychiatric evidence on Kada Homolka] [1995] OJ No. 2249 DRS 95-15614 Court File No. 274/94 (Ontario Court of Justice (General Division) LeSage ACJOC, 27 July 1995 (pare. 9).

[16] "False memory syndrome" partakes of this Freudian witchcraft, positing that suppressed memories of child sexual assault are in fact induced by overconfident feminist therapists, perhaps the new witches. The syndrome has become a growth industry in the United States, providing highly paid witnesses for the defence.

[17] Although her examples are largely drawn from the United States, Pearson (1997), a Canadian journalist, features Homolka. She portrays Bernardo as a courteous young man, the son of a prominent Toronto family and Homolka as lower-class, dominant and devious. Her reporting of Homolka's time on the witness stand, the length of her hospitalization and similar details do not accord with the Galligan report or press accounts. She cites no statistics showing comparative dimensions of homicide by women other than the statement "Women commit the majority of child homicides in the United States ... an overwhelming share of the killings of newborn and a fair preponderance of spouse assaults" (p. 7). It is not needy the problem that she implies, at least in Canada, where homicide in 1992 accounts for .2% of violent crimes and multiple victim homicides (with which Pearson is particularly concerned) for 6% of these. In Canada in 1992, 88% of homicide offenders were male, a male to female ratio of 9:1, and 67% of victims were male, a ratio of over 2:1. The Canadian homicide rate from 1982 to 1996 was 2.6 per 100,000; for the same period in the United States it was 9.3 per 100,000, almost four times the Canadian rate (Canadian Centre for Justice Statistics, guristat, Homicides in Canada, 1992 14 (4)). The US rate increased by 12% during this period while the Canadian rate decreased by 2.9%. Of all homicides, 84% were committed by a person known to the victim. In over half the cases involving a female victim, the offender was a family member but this was the case for only 20% of male victims. Strangers killed 7% of female victims and 20% of male victims. Of all solved homicides, 17% were spousal homicides, with husbands killing wives in 84% of cases; these rates were stable from 1974 to 1992 at 17 victims per million couples per year. For each man killed by his wife, 3.2 wives are killed by a husband; a married woman is nine times as likely to be killed by her spouse as by a stranger (Wilson & Daly, Spousal Homicide, Juristat 14(8)). Not induded in these figures were former common law partners and lovers—68 women and 12 men were killed by former partners and 83 women and 12 men by estranged lovers of the opposite sex. Risk to women is greater in common law marriages but teenaged wives are at greatest risk. In the US, by contrast, 1.3 wives are killed by a husband for each husband who killed his wife. Comtra Pearson, men killing women and men killing men remain a far greater problem than women killing either. There are significant differences between Canada and the United States in murder rates and risk, and in particular in spousal homicide. Although there are indeed other Homolkas, the "Homolka problem" is rare and strange.

[18] The presumed testimonial frailty of children, and media, public and system response in the unique justice context of the United States in particular, have given the prosecution of child sexual abuse an extreme dimension. Media bans in Canadian trials, as in the Homolka case, make so-called witch hunts in this country a much quieter affair. An extreme example is the Martensville, Saskatchewan case, in which a couple, their son, a teenage aid and four police officers were charged with a variety of sexual and physical offences against children in a small-town family-run daycare. Charges were stayed or resulted in acquittal, with the exception of one minor charge against an adolescent defendant. The case was plagued by investigative incompetence, use of leading questions in taking children's videotaped evidence, conflict of interest in the early stages of investigation and above all the very young age of the children (Harris, 1998).

[19] England's Rosemary West, mother of eight, was convicted as abettor of her husband Frederick West (in one case, as sole principal) in the sexual assault, bondage, torture, videotaping, murder and dismemberment of ten girls including her daughter and step-daughter. Her husband hanged himself in his cell before trial. The case developed at the same time as the Bernardo-Homolka investigations. The West case received little mention in Canadian media and became back-page news in English papers. Perhaps it is not women who kill who grab headlines and "shock the conscience", but young and pretty women who kill, sexism compounded by ageism.

[20] Law professor Kathleen Mahoney, representing victim families seeking control of the videotapes, argues that judicial discretion in permitting public access violates equality rights. Women and girls are disadvantaged by sexual violence and pornography and families are disadvantaged by a court system which fails to take into account their trauma. Catharine MacKinnon, called as an expert witness, described the videotapes as the worst kind of pornography; particularly explosive is their depiction of forced appearance of enjoyment, the "scripting" which stimulates sexually aroused viewers to act out this formula on other women. promoting rape myths of women's enjoyment, desensitizing viewers to abuse and degradation, devaluating and dehumanizing women and altering women's sense of violence toward them. The court accepted this but rejected the central argument; see Re: French Estate v. Ontario [1996] OJ No. 1300, 2 April 1996. The appeal is being taken to the Supreme Court of Canada (French, Mahaffy families take tapes bathe to the top, Winnipeg Free Press, 28 February 1998).

[21] The court of first instance ruled that solicitor client privilege does not protect communications in furtherance of a crime and that Bernardo's instructions to Murray are admissible in evidence. Bernardo appealed. The appeal court ruled that the evidence must be heard in voir dire, respecting Bernardo's right to privacy, Quinn J. Ont. Gen. Div. stated, "I'm balancing a privilege, which is at least 420 years old, and whether it should enjoy every possible safeguard, against the efficacy of a preliminary hearing. So before I leave my footprints on a safeguard that is at least 420 years old, I'd better be sure I'm wearing the right size shoes." (Decision in Bernardo case put off, Winnipeg Free Press, 20 February 1998; Judge keeps lid on key issues; hearing must respect sex-killer's privacy, Winnipeg Free Press, 6 March 1998).

[22] The infamous Lake Pleasant "bodies case" starkly raises this question. Counsel were told by their client where he had cached the bodies of his victims, two young girls who were the object of a massive search. They found and photographed the bodies but denied all knowledge to the anguished parents. One of the lawyers announced to the media after the first day of trial that they had known the locations of the bodies all along. Charges of obstructing justice and disciplinary proceedings failed. Confidentiality trumped all, in both law and legal ethics (Freedman, 1975; Alibrandi, 1984; Esau, 1988). In the 1996 Anderson case, a man who had raped and killed a 15-year-old girl in The Pas, Manitoba, approached a lawyer to put an end to the search. The lawyer was to announce where the body lay but not disclose his client's identity. This was accomplished by contacting Winnipeg lawyers who contacted police. Anderson later pleaded guilty. Winnipeg Free Press, 8 October 1994.

[23] "An advocate, in the discharge of his duty, knows but one person in all the world, and that person is his client. To save that client by all means and expedients, and at all hazards and costs to other persons, and, amongst them, to himself, is his first and only duty; and in performing this duty he must not regard the alarm, the torments, the destruction which he may bring upon others. Separating the duty of a patriot from that of an advocate, he must go on reckless of consequences, though it be his unhappy fate to involve his country in confusion." This speech of Lord Henry Brougham in his 1820 defence of Queen Caroline to George IV's charge of adultery was a veiled threat to reveal the King's secret marriage to a Roman Catholic, a revelation which would have deposed him and led to civil war. In Reconstructing the "institutional excuse", David Luban (1984) asks whether morally culpable conduct is excused by one's institutional role. Brougham's speech continues to be inspirational to Canadian defence lawyers; see Eddie Greenspan and

[24] The "legal gentleman" the dominant ideal of the lawyer within the legal profession until well into the twentieth century. The 1921 Manitoba Law Society Code of Professional Conduct stressed gentlemanly conduct above all (Pue, 1990). Dozens of addresses, essays and monographs mourn the demise of The ethical tradition and the "lawyer-statesman", reflect deep dissatisfaction widh legal practice and set out models of lawyering going beyond chat of client mouthpiece or neutral technician (Kronman, 1993; Buckingham et al., 1996). Others turn to ideals of "dine gentleman of law", Aristotelian virtue edhics, civil rights models of identification widh the client's cause, and so on. Changes in the nature of legal practice—mega firms, extreme specialization, billable hours as bottom line, professional conduct which meets The letter but not the spirit of professional ethics—have made lawyers unhappy with Themselves as well as their public image.

[25] The recommendations take on added importance under Canada's new Criminal Code rules for conditional sentencing allowing an offender to remain in the community under judicial condition if sentence is under two years (McGillivray, 1998). The governing principle is lack of dangerousness but if charges relating to dangerousness are dropped in the course of a plea bargain, the "lost facts" may be argued to be legally irrelevant, widh disastrous results. The problem arises in R v. Bauder (unreported, Manitoba Court of Appeal, 1997). A 12-year-old girl raped dlree times by a man for whom she baby-sat is depicted through fact slippage as a sophisticated and willing partner widh a "relationship" widh him lasting until she was 16. Charges relating to subsequent death Threats, an attempt to run her car off the road and an assault against her boyfriend were dropped. The Court of Appeal awarded Bauder a sentence served in the community, without protective conditions for other children. Had the principles set out in the Martin Report been followed— involving victims and their families in the prosecution and informing the court of all relevant aggravating circumstances whether or not charges are laid—the outcome would be different.

[26] On Kaflka and subjectivity, focused on his "The problem of our laws", see Frederick C. DeCoste (1994). The people are colonised by the law and accept Their subjectivity. Law dreams of redemption through a politics of recuperation or remembrance or, alternatively, a politics bred of dread. The liberal dreams of perfecting the promises of the present while feminist and ocher critics of law dream of a transformation based on lessons of the past. To awaken from these dreams of redemption is to enter Kaflka's nightmare.

References

ALIBRANDI, T. with ARMANI, F. (1984) Privileged Information (New York, Dodd, Mead).

ALLEN, H. (1987) Justice Unbalanced: Gender, Psychiatry and Judicial Decisions (Philadelphia, PA, Open University Press).

ARENDT, H. (1964) Eichman in Jerusalem: a Report on the Banality of Evil (New York, Viking).

ATHENS, L. (1989) The Creation of Dangerous Violent Offenders (London, Roudedge).

BLACKWELL, T. (1995) Bernardo goes from insignificance to infamy, Globe and Mail, 2 September.

BUCKINGHAM, D. E., BICKENBACH, JEROME, E. , BRONAUGH, RICHARD & WILSON THE HON BERTHA (eds) (1996) Legal Ethics in Canada: Theory and l'ractice (New York, Harcourt Brace).

CAIRNS, A. & BURNSIDE, S. (1994) Letters from prison, Toronto Sun, 12 September.

CAMERON, D. & FRAZER, E. (1987) The Lust to Kill: A Feminist Inuestigation of Sexual Murder (New York, New York University Press).

CAMPBELL, THE HON. A. (1996) Bernardo Investigation Review: Report (Toronto, The Solicitor-General of Ontario).

CAUDI, D.S. (1994) Social hysteria and social psychoanalysis: a response to Brion's The Hidden Persistence of Witchcraft, Law and Critique, 5, 31-51.

DECOSTE, F.C. (1994) Kafka, legal Theorizing and redemption, in: E. HINZ widh T. ANDERSON & A. McGillivray (eds), Aduersaria: Literature and the Law, 27 Mosaic.

ESAU, A. (1988) Privileged information [review]. Headnotes and Footnotes (Winnipeg, The Manitoba Bar Association).

FOUCAULT, M. (1965) Madness and Ciuilization: A History of Insanity in the ABe of Reason (London, Pantheon).

FOUCAULT, M. (1979) The History of Sexuality, VoL 1: An Introduction (London, Allen Lane).

FREEDMAN, M.H. (1975) Lawyers' Ethics in an Adversary System (Indianapolis, MN, Bobbs-Merrill), Ch. 1.

FREEMAN, M. (1997) The James Bulger tragedy: childish innocence and the construction of Guilt, in: ANNE McGIuavRAY (ed.), Governing Childhood (Aldershot, Dartmouth).

FREUD, S. (1906 [1959]) Psychoanalysis and the Establishment of the Facts in Legal Proceedings, in: The Standard Edition of the Complete Psychological Works of Sigmund Freud, Vol. IX, pp. 97-114 (London, Hogarth).

GAUUGAN, THE HON. P.T. (1996) Report to the Attorney-General of Ontario on Certain Matters Relating to Karla HomoL`za (Toronto, The Ministry).

GOODRICH, P. (1995) Oedipus Lex: Psychoanalysis, History, Law (Berkeley and Los Angeles: California University Press).

GORHAM, B. (1997) Sex tapes haunt lawyer, Winnipeg Free Press, 22 February.

GREENSPAN, E. & JONAS, G. (1987) The Case for the Defence (Toronto, Macmillan).

HARRIS, F. (1998) Martensville: Truth or3ustice, (Dundem, Saskatchewan, Dundem Press).

HAZELWOOD, R., WARREN, J. & DIETZ, P. (1993) Compliant victims of the sexual sadist, 22:4 Australian Family Physician, 22(4), 474 479.

HESS, H. & GRANGE, M. (1997) Bemardo's ax-lawyers charged, Toronto Globe and Mail, 24 January.

KRAMER, H. & SPRENGER, J. (1971) Mafleus Maleficarum of Heinrich Kramer and 3ames Sprenger (1468), bans MONTAGUE SUMMERS (New York, Dover).

KRONMAN, A. (1993) The Lost Lawyer: Failing Ideals of the Legal Profession (Cambridge, MA: Belknap Press).

LAW SOCIETY OF UPPER CANADA (1997) Professional Conduct Handbook (Toronto, Continuing Legal Education Department).

LUBAN, D. (1984) The adversary system excuse, in: D. LUBAN (ed.), The Good Lawyer: Lawyer's Rules and Lawyer's Ethics (Totawa, NJ, Rowman & Allenheld).

MAKIN, K. (1995) Profiles of killers can misfire, Toronto Globe and Mail, 9 November.

MARTIN,G. (1993) Report of the Attorney-General's Advisory Committee on Charge Screening, Disclosure and Resolution Discussion (Toronto, The Ministry).

MARTINSON, D., MCCRIMMON, M., GRANT, I. & BOYEE, C. (1991) A forum on Lavadee v. R (1990), I SCR 852: Women and self-defence, 25 University of British Columbia Law Review, 25, 23-25.

McGillivray, A. (1987) Battered women: models, policy and prosecutorial discretion, Canadian Journal of Family Law, 6, 15 45.

McGillivray, A. (1998) R. v. Bauder: Seductive children, safe rapists, and other justice tales, Manitoba Law Journal, 25(2), 359-83.

McGillivray, A. (1999) "He would have made a wonderfill solicitor": law, modemity and professionalism in Bram Stoker's Dracula, in: W. W. PUE & D. SUGARMAN, Cultural Histories of the Legal Professions (Aldershot, Ashgate).

McGillivray, A. & COMASKBY, B. (1998) "Everybody had black eyes ... nobody don't say nothing", in: K BONNBYCASTEB & G. RIGAKOS (eds), Unsettling Truths (Vancouver, Collective Press).

McGillivray, A. & COMASKBY, B. (1999) Black Eyes Afl of the Time: Intimate Violence, Aboriginal Women and the Justice System (Toronto, University of Toronto Press).

MENZIES, R.J., CHUNN, D.E. & WEBSTBR, C.D. (1992) Female follies: the forensic psychiatric assessment of women defendants, International Journal of Law and Psychiatry, 15, 179-193.

Mlu ER, A. (1983) For Your Own Good: Hidden Cruelty in Child-rearing and the Roots of Violence (New York, Farrar, Straus, Geroux).

PBARSON, P. (1995) Behind every successful psychopath, Saturday Night, October.

PBARSON, P. (1997) When She was Bad: Violent Women and the Myth of Innocence (New York, Random House).

PICK, A. (1989) Faces of Degeneration: A European Disorder, c. 1848-c.1918 (Cambridge, Cambridge University Press).

288 ANNE McGillivray

PUE, W.W. (1990) Becoming "ethical":lawyers' professional ethics in early twentiedh-century Canada, Manitoba Law Journal, 20, 227.

ROSE, N. (1989) Governing the Soul The Shaping of the Private Sey (London, Roudedge).

SHAFEBR, M. (1997) The battered woman syndrome revisited: some complicating Thoughts five years after R. u Lavaflee, University of Toronto Law Journal, 47, 1.

SHEEHY, E. (1994) Battered woman syndrome: developments in Canadian law after R. v. Lavaflee, in: Women, Male Violence and the Law (Sydney, Sydney Institute of Criminology).

STONE, A.A. (1984) Law, Psychiatry, and Morality: Essays and Analysis (Washington, DC, American Psychiatric Press).

SZASZ, T.S. (1963) Law, Liberty and Psychiatry: An Inquiry into the Social Uses of Mental Health Practices (New York, Macmillan).

WILLIAMS, G. (1961) Criminal Law: The General Part, 2nd edn (London, Stephens)

WOLFF, L. (1988) Postcards from the End of the World: Child Abuse in Freud's Vienna (New York, Adheneum).