
During a week filled with chemical attacks and cruise missile strikes in Syria, and a government’s cynical, Saul-like conversion to compassion on the way to an election closer to home, we must savour any small spark of common sense where we can find it.
I found mine reading “Court Watch,” law student Christina Macdonald’s always fascinating, inevitably insightful column about “what she finds notable or interesting” at the courts that week. (Macdonald’s column is yet another reason to subscribe to the Examiner… but I digress.)
Near the end of last Wednesday’s Court Watch, Macdonald offered an intriguing gem of a story under the heading: “29 years later, father who served seven years for manslaughter of son has name removed from Child Abuse Registry.”

The story focused on a 30-page decision by Justice Beryl MacDonald concerning an application by a man to have his name removed from the province’s child abuse registry. It is, Macdonald noted, a “complex decision” but one that is also “really interesting.”
It is. What was especially interesting to me were the times when Justice MacDonald stepped outside the usual boundaries of case law and legalese in order to offer a personal human reflection on a case that is all too human in a legal system that is all too rarely.
The back story begins in the 1980s when a then 21-year-old G.A. — as he is described in court documents — began a tempestuous relationship with a young woman. They were, the judge writes “both immature and were known to physically lash out when frustrated.”
After they had a son, they found themselves on the radar of Halifax Children’s Aid Society. They had “regular contact” with social workers and, on occasion, the child was removed from their care. The judge reports both parents hit their son from time to time and “were not often honest about what was happening in their life.” Writing later about G.A.’s occasional untruthfulness as he rationalized what happened in his life, the judge confesses: “I understand that phenomenon. Sometimes we want to shield ourselves from events we do not want to remember.”
In May of 1987, after an argument between the couple, the woman went into the bedroom where their 19-month old son was crying. “She lost her temper and forcibly struck the (child) with a karate-type chop to the midsection.”
The woman says she told her partner what she’d done. He claimed she didn’t; he thought the child was suffering from the flu. But he admitted he didn’t take the child to the hospital even as his condition worsened because he feared “the child might be taken from their care as he had been earlier.” The child eventually died.
Initially, both parents were charged with second degree murder, but he was convicted only on the lesser charge of manslaughter “because he did not get medical treatment for his son.”
G.A. served seven years in prison and his name was ordered placed on the child abuse register.

Fast forward to 2016.
G.A. — who is now 50, has had no further involvement with the criminal justice system and has now been in an apparently non-violent relationship with a woman for five years — applies to the courts to have his name removed from the registry. The ministry of community services objects, claiming he is still a danger but providing no evidence of that danger because, under the rules of this game, G.A. is the one who has to prove his name should be removed.
He must, as the judge notes, “prove a negative.”
This is where it gets interesting.
During his time in prison, the judge reports, “G.A. incurred no disciplinary charges… upgraded his educational level from grade 10 to grade 12, had his work performance described as excellent, attended a 12-week life skills course, and completed a cognitive skills course.”
Despite that — and despite a number of pre-parole assessments by psychiatrists and psychologists that, although “cautious” about potential anger management issues, made the point G.A. “seemed intent on making a life for himself in the community and doing so without crime.” Noted one psychologist: “I did not see him as a significant risk.”
But the author of his actual pre-parole report — who, the judge notes, was neither a psychiatrist nor a psychologist — concluded that “as a minimum, this offender should be prohibited from entering into relationships where young children are present… The emphasis must be on control rather than treatment.”
After his release from prison in 1996, that report would become the go-to basis for community services to jump in whenever G.A. attempted to begin a new relationship with any woman who had children. “The women involved were told their children would be taken out of their care by the Minister if they allowed G.A. to continue to live with them.”
Ironically, one of the reasons community services now clams G.A. cannot be trusted and why his name needs to be kept on the registry is that, as it told the court, he has not “provided information about any time he might have spent with young children.”
Catch-22. But it gets worse.
“Because G.A will not admit the role he played in his child’s death in the way assessors and the minister require (my emphasis),” the judge writes, “his efforts to explain himself are viewed as rationalizations. His failure to accept the labels others have used is accepted as evidence about his lack of emotional control, of his impulsiveness, of his untrustworthiness, of a likelihood he will be involved in episodes of domestic violence and harm or fail to protect children… He could not admit what they wanted him to admit.”
Of course the reality is that G.A. was found “not guilty of ‘unlawfully causing the death’ of his son. He was found guilty of manslaughter because he did not get medical treatment for his son. But those who have dealt with him do not make this distinction,” MacDonald writes. “To them he just made excuses and a jury believed him. They want him to admit ‘his guilt’ but this is not a guilt he can admit. I understand that.”
Those last three words — “I understand that” — are especially intriguing. It is not possible to parse what the judge intended, of course, but consider this haunting backdrop to the last more than two-and-a-half decades of criminal justice in Nova Scotia: Donald Marshall, Jr., a young Native man, spent 11 years in prison for a murder he did not commit, in part because he refused to acknowledge his “guilt” in a way assessors and the minister require.
Justice MacDonald doesn’t sugar coat. G.A. “does not present with a kind, compassionate, loving personality. Life is ‘all about him and his needs.’ But this does not automatically lead to the conclusion that there is a ‘real chance of danger’ that he will cause harm to others or fail to protect a child.”
The last line of her decision: “His name is to be removed from the Child Abuse Registry.”
Thanks for your kind comments and the deeper dive, Stephen! It’s a complex situation for sure.
Thank you, Pam Rubin, for parsing this decision, and the reporting of the decision so well.
I’m curious what happened to the woman who actually killed the child. What happened to her?
I wonder that too. Another column?
Justice Beryl MacDonald’s record should be evaluated in its entirety, before praising her “common sense” when it comes to abuse and the protection of children. I wonder if anyone at the Examiner has the resources to take a look at her remarks and decisions overall when it comes to protecting women and children? I would be interested in the conclusions drawn at that point.
A close read of the decision brings out many troubling points for those who care about preventing child abuse:
1. First and foremost: this man has admitted to multiple incidents of assaulting his infant son prior to his death. He has steadfastly refused to engage in counselling directed at reducing his risk to repeat such behaviours. This alone is enough in most universes to say “nope” to requests for removal from a child abuse registry. Let’s not obscure these simple facts.
2. The man was originally placed on the child abuse register because he had “hit his infant son in the face with his hand”and several other assaults, not because of his failure to get medical help.
3. The child abuse register functions as a screening tool for employers and non profits’ volunteer rosters. Don’t places where he would be in contact with children deserve to know this history?
4. The Justice admits that the man had a history of being “less than truthful.” When it comes to abuse, this means there is more likelihood of hiding, minimizing and rationalizing abuse, exactly as seen here as described in the assessors’ reports.
5. The Justice’s remarks betray her feelings about emotional abuse, or even its existence as a thing. She decides to wave away multiple previous courts’ findings that the man is a danger to children also based on emotional abuse. As many abuse victims will tell you, emotional abuse can be worse than physical abuse, and can take a lifetime for children to recover from. Before praising this decision, the author might have looked into the previous decisions concerning this man’s potential for emotional abuse, and why previous judges made that determination. In my experience emotional abuse is in fact very hard to get courts to recognize as a risk, and if multiple courts have done so that is worth taking note of, instead of waving away.
6. Justice MacDonald remarks: “G.A. faces a difficult burden of proof. He must prove a negative. He must prove on a balance of probabilities that he “does not pose a risk to children”.” I would not agree that a balance of probabilities (50%+) is a difficult or inappropriate burden of proof for this man to show he is safe to be around children in a family, employment or volunteer setting.
7. Justice MacDonald takes a gratuitous and somewhat incomprehensible dig in this case at the memories of abuse victims.
8. Risks to partners and children are not necessarily reduced because a man is “a pillar of the community” whether that be in prison or elsewhere. The following should not surprise Justice MacDonald or the author: “The fact that G.A. incurred no disciplinary charges while in prison, upgraded his educational level from grade 10 to grade 12, had his work performance described as excellent, attended a 12-week life skills course, and completed a cognitive skills course were of no assistance in convincing those who prepared assessments that contact between him and children would be risk free.”
9. What does this mean from the Justice? : “The assessment of G.A.’s risk has always been heavily influenced by the circumstances leading to the death of his son. This potentially created a bias against considering alternate explanations for his behavior.” Is the death and prior physical abuse of his son (by him) not supposed to influence assessments? Many assessments are done in part on a strict evidence-based actuarial basis, in which the existence of prior events of abuse or neglect are crucial factors in risk level determination.
10. At one point Justice MacDonald notes: “P.H. confirmed she remained “on cordial terms” with G.A. and confirmed G.A. exercised some control in the situation. He “put his hand to her throat in anger although he removed it at her request.” In the next paragraphs she completely dismissed this episode as the basis of any concerns, without explanation. In reality, strangulation is one of the most dangerous of assaults, and any attempts or tendency to employ this form of violence should be taken with the utmost weight in deciding safety risks. At a minimum it is a plain assault even without getting into “risk for lethality” predictions, and should not be dismissed from consideration. The fact that a victim “remained on cordial terms” with their abuser is irrelevant. As we know from Ghomeshi and elsewhere, users are woven into the social fabric and many victims for a variety of good reasons “remain on cordial terms” with them.
11. Why does Justice MacDonald emphasize this as a reason for assuming safety around children? “Dr. Litvin told his supervisor that subject didn’t feel that he had problems that need to be dealt with although the doctor was of the opinion the G.A. had a lot of anger bottled up inside of him and might be volatile. The Doctor did “not see subject as a threat to the community at large since he felt that he could handle things on his own. ” Not being a threat to the community at large does not mean someone is not a threat to intimate partners and children.
12. Why exactly is this unreasonable to expect given the history? ” [Eakins’]recommendation was that if he wanted to be permitted contact
with children he had to “involve himself seriously in therapy and otherwise meet the criteria for his removal from the Child Abuse Register…There is a specific need for him to work on his interpersonal relationships with women as the children of these women will be at risk of emotional harm if exposed to a coercive controlling relationship.” If Child Protection is informed by multiple past partners about this man’s coercive control, including attempted strangulation, at a minimum this man should do serious counselling before he is considered safe to be taken off the register. Justice MacDonald decries the fact that ex-partners may have informed CPS anonymously about GA’s coercive controlling behaviours. I suggest that many women must do so for safety reasons, that many other women never do so for safety reasons, and that the collection of their evidence by CPS as relevant should never be dismissed out of hand because their information is anonymous. We are not talking about sending this man back to prison; we are talking about a precautionary civil determination about his safety to be around children in multiple settings.
13. This statement raises Justice MacDonald’s understanding of the basic principles of assessment: “The Minister argues that the passage of time does not matter in these proceedings. But yet the Minister often relies on past behaviour as a predictor of likely future behaviour. ” Past violent behaviour is predictive of future risk regardless of passage of time, although it can moderate that risk in certain circumstances. That is simply a fact. The reverse is not necessarily true, however, actuarially. Everyone making these decisions should be acquainted with some of the science, and if they are not, they should be willing to defer to multiple assessments all saying the same thing.
14. “The Minister was convinced and remains convinced that G.A. has a propensity for violence not only toward children but also toward women with whom he may have a relationship. This conviction is primarily based upon reports conducted by psychologists. The experts repeatedly informed the Minister that
without completion of an anger management program and intensive psychological exploration through prolonged psychotherapy or counselling children in his care would be at risk.” There is in fact an actuarial connection between refusal to undergo counselling for partner and child abuse, and risk level of future violence. This is not simply a matter of refusing to state things in exactly the way demanded by experts; the refusal to engage in treatment at all despite multiple recommendations is an important risk indicator.
15. Justice MacDonald goes on to describe the admitted facts of this man’s abuse of an infant: There is no question that G.A. has in various conversations admitted to
slapping his son over a diaper on his buttocks, to hit him on the face, to throw him into his bed and to hold him tightly enough to leave marks around his stomach area. These admissions are sufficient to confirm that when he was 22 years of age he could harm a child and he did harm his son through these actions.”
She then however goes off in an unexpected direction of dismissing this as relevant because of her great sympathy for men who may theoretically be victims of female violence. Huh?
In the end no reference in her decision to remove him from the list is made to his admitted violence to an infant, and she only looks at the failure to report as relevant.
I know that for non-violent stable men it may sound scary that experts will have control over such determinations of risk, and access to children, and to feel personally a shudder of resistance against Kafkaesque systems interfering in our lives. But in my experience it is quite unusual for assessments to consistently and strongly recommend the way they have in this case. When the competing interests of children’s safety vs. a convicted admitted abuser’s freedoms to access children of new partners, or in employment, compete, as a community, we need to prioritize children’s safety because the stakes for children are extremely high. The onus is legally on the person who wants to be removed from the register. It is a balance of probabilities onus, so not in fact a very high hurdle. The Justice may take issue with this onus but that is the law, and I think most would say a good one.
In my view, this decision could have been appealed and may very well have been overturned. The dismissal of evidence about coercive control, of partner abuse, of indeed even admitted direct multiple incidents of violence against an infant, is problematic. It is in fact, how children die. Then we are all left scratching our heads about “what went wrong?” Well, this minimization of actual risk is the sort of stuff that “goes wrong.”
Thank you Pam Rubin.
Thank you for this. I know of two decisions from this judge that show massive errors in judgement and caused extensive trauma for parents and children. Here she is again, trying to rationalize this person’s lies and make excuses for it because of her personal emotional bent. I would echo the call for the records of this judge’s decisions to be evaluated before holding her up as any sort of paragon.
Your reply to Kimber’s article shows much more insight than any contained in the article itself. Thank you for taking the time to analyze the judgement so thoroughly.
Great column, thank you. Thanks to Christina Macdonald for noticing the decision, and especially to Mdm. Justice MacDonald for rendering it.
If you follow Community Services cases for any time at all, you will come up against the bureaucratic attitudes she describes again and again. It’s an intractable problem.