This story is part of an ongoing Globe and Mail investigation into the hundreds of missing and murdered indigenous women in Canada
What's this project about?
Indigenous women suffer elevated rates of violence. Statistics Canada has found that they are more than three times as likely as non-aboriginals to be victims of violent crime. They are far more likely to report encountering the most extreme forms of violence, such as sexual assault or having a gun used against them. And they are far more likely to go missing or die violently.
What causes this? Recent comments from the former federal Conservative government and RCMP emphasized the central role of family violence. While it is undeniable that aboriginal women endure higher rates of domestic violence, there is more to the story. Previous research has linked elevated violence to poverty and marginalization, intergenerational trauma and family breakdowns resulting from the colonial legacy and residential schools, addiction and other issues.
Hoping to better understand the over-representation of indigenous people among Canada's missing and murdered women, The Globe and Mail is compiling a database of relevant cases.
When we first began studying homicides in our database, one thing that jumped out was the large number of cases attributed to a handful of serial offenders, such as Robert Pickton, John Crawford and Cody Legebokoff. Upon deeper analysis of serial-homicide data, we found that indigenous women were overrepresented among the victims of serial killers in Canada.
We decided to create an immersive multimedia feature that explores some of the common themes that emerged from our analysis, by recounting the lives of five women who died at the hands of different Canadian serial killers. We believe their stories illuminate some of the social forces that contributed to their vulnerability to some of Canada's most dangerous men.
How did The Globe acquire this data?
In late 2014, the Native Women's Association of Canada provided The Globe with data collected during its Sisters in Spirit initiative. We obtained further data from Maryanne Pearce, who extensively researched this issue while completing her doctorate in laws degree at the University of Ottawa. We acknowledge their work and are grateful for their assistance. The Globe merged these two data sources and is dedicating considerable effort to verifying, expanding and understanding them. This process is continuing, and often involves acquiring further information from families of victims. We recognize this can be difficult for families, and deeply appreciate their co-operation.
We also acquired data about victims of Canadian serial killers from Mike Aamodt, a professor emeritus in the psychology department of Radford University in Virginia. Over more than two decades, he has compiled what he describes as the largest non-governmental serial-killer database in the world. He provided us with a subset containing data on serial homicide victims of both genders and all ethnicities in Canada. We corrected some entries to reflect information in our possession, but left this data mostly unmodified. We thank him for his generosity.
What are the limitations of The Globe's data?
There are many, but several are particularly significant. We do not have a complete census of all homicides or disappearances involving aboriginal women in Canada. One reason is that most of the cases we know about have come to our attention through news reports. (Our attempts to obtain official data from government and law enforcement, which would likely be more comprehensive, have been unsuccessful to date.) There are many reasons why a homicide may not receive news coverage, and thus be excluded in these databases. Moreover, electronic news databases are creatures of the past several decades; we suspect our database work is reasonably comprehensive for the years since 2000, but likely misses more cases as we move further back in time. We therefore cannot draw reliable conclusions about how the situation has evolved over time.
Our database contains many fields. It is not uncommon to find conflicting information on basic details such as the age or ethnicity of a victim or offender, for instance, and sorting all this out can sometimes be challenging. We have made a considerable effort to verify and correct entries, and this work continues.
We focused on vetting our database's cases involving confirmed and speculative serial killers to try to ensure the women were indigenous and also that they were either killed unlawfully or still missing. The Globe conducted many dozens of interviews with victims' families, lawyers, law-enforcement officials, authors, researchers and indigenous organizations. We obtained court transcripts and exhibits, including some requiring a judge's permission. We pored over inquiry reports and news stories. We conducted reporting on the ground in Manitoba and British Columbia.
What's the basis for The Globe's claim that aboriginal women are overrepresented among the victims of serial killers?
In our ongoing database work, we discovered that of the more than 700 homicide cases we identified dating from 1980, 18 (2.6 per cent) were slain by serial killers who were convicted of the offence. Since many serial homicides are likely never identified as such, the true proportion is probably far higher.
This struck us as unusual, because serial homicide is extremely rare. The FBI estimates it comprises less than 1 per cent of all murders. Full disclosure: Two researchers we spoke with doubted the validity of the FBI estimate. "This figure has been out there for decades," Enzo Yaksic, co-founder of the Atypical Homicide Research Group, told us, "and nobody can figure out where the source of the original information came from." When we asked the FBI, Special Agent Ann Todd of the Office of Public Affairs responded that the FBI does not maintain statistics for serial killers. We acknowledge there are concerns with this estimate, but it does seem to square approximately with the situation in Canada. When the total number of serial homicides in Canada from Aamodt's data since 1980 is compared with Statistics Canada's total number of homicides for the same period, the average Canadian rate seems to be slightly higher, at 1.14 per cent. We used the FBI's figure as an informal benchmark.
Our confidence grew once we acquired Aamodt's data. More striking conclusions can be drawn from it, for the simple reason that it includes victims of all ethnicities. Of the 87 homicides of females for which convictions were obtained against Canadian serial killers, 18 (21 per cent) involved aboriginal victims. Indigenous women represent just 4 per cent of Canada's female population. Once we incorporated population figures into our calculations, we determined that aboriginal women are roughly seven times more likely to be victims of serial killers than non-aboriginal women.
The true extent of serial predation on aboriginal females in Canada is likely higher. To begin with, consider the cases for which Pickton was never tried but where DNA and women's belongings, for example, were found on his farm. These cases were not brought to trial, so we do not consider them "confirmed" cases. Aboriginal women and girls also appear frequently among clusters of unsolved disappearances and homicides, such as those in the Edmonton area or along B.C.'s "highway of tears." When such cases are considered, Aamodt's database contains 189 female victims of all ethnicities dating from 1980 whose murders he attributes to serial killers. Of those, 53 victims (28 per cent) were identified as aboriginal.
What does The Globe mean by "aboriginal" and "indigenous"?
We use the terms interchangeably. We classify victims as aboriginal when they are identified as First Nations, Métis or Inuit. It should be noted, however, that ethnic identity is not always straightforward. Where we encounter individuals of mixed ancestry including indigenous, for example, we classify them as aboriginal – unless we happen to know how they self-identified.
Who qualifies as a serial killer?
The Globe adapted a definition created by a multidisciplinary group of experts convened by the FBI in 2005. It defined "serial murder" as "the unlawful killing of two or more victims by the same offender(s), in separate events." The RCMP uses the same definition. But if this seems straightforward, the process of selecting it was anything but.
It is generally accepted that the term "serial murder" originated in 1976 from FBI agent Robert Ressler, who coined it in connection with New York's "Son of Sam" homicides. There is no generally accepted definition, however, and seemingly subtle changes to parameters have radical implications for who is considered a serial killer, and who is not. "Attempts to find clearly delineated boundaries for a definition of serial murder for which there is no 'gray space' are unlikely to be successful," observed a paper by Christopher Ferguson and several colleagues from the University of Central Florida's psychology department.
Policing agencies often use the terms "serial murder" and "serial homicide" interchangeably. We do not. Under Canada's Criminal Code, culpable homicide is murder, manslaughter or infanticide, so we chose to use the term "serial homicide." Consider Shawn Lamb. He pleaded guilty to two counts of manslaughter, in connection with the deaths of Lorna Blacksmith and Carolyn Sinclair, and therefore committed two unlawful killings. And John Crawford, who killed four indigenous women, was convicted of three counts of murder and one count of manslaughter.
Quantification also presented a challenge. A series obviously involves more than one incident, but how many? Early definitions decreed serial homicide must involve at least three, or perhaps four, individual slayings. Later definitions, including the 2005 definition by the FBI multidisciplinary group, suggested two would suffice.
Simply defining serial homicide as multiple homicide is insufficient. To distinguish between serial and mass murder, some definitions incorporated a requirement that serial homicides be separated by an "emotional cooling-off period" lasting days, months or years. What if it's mere hours? Edmonton-area serial killer Joseph Laboucan killed Ellie May Meyer on April 1, 2005. He murdered Nina Courtepatte barely a day later. According to trial testimony, Mr. Laboucan sought another victim in downtown Edmonton immediately after Ms. Courtepatte's murder, but was unsuccessful. Some might classify Mr. Laboucan as a spree killer rather than serial murderer. We have classified him as the latter.
The question we struggled with most was: Who decides what constitutes homicide? Such judgments properly belong to the courts, not the news media. Yet owing to the large numbers of unsolved cases involving aboriginal victims, relying on criminal convictions alone likely understates – perhaps significantly – the true extent of serial predation against aboriginal women in Canada. After much internal debate, we adopted a restrictive caveat: the offender must have been convicted of these killings by a court of law. This requirement eliminated many cases that initially seemed relevant to us.
The most striking examples concern the first-degree murder charges against Pickton in the cases of 20 missing women, both aboriginal and not. Those charges were ultimately stayed, but none of the women have surfaced since they abruptly went missing from Vancouver years ago. Although convicted of six counts of second-degree murder, Pickton told an undercover police officer he had killed 49 women. A reasonable person might strongly suspect these other women were also victims of serial homicide, but we excluded them from our list of confirmed cases.
Gilbert Jordan presents another troubling example. For years he prowled Vancouver's beer halls in search of women to drink with; by his own estimate, he drank with 200 a year throughout the 1980s. As many as nine women died during such encounters; eight were indigenous women whose deaths were attributed to alcohol poisoning. He ultimately received a single manslaughter conviction in connection with the death of a non-aboriginal woman, Vanessa Buckner. Because he had a single conviction, we had to exclude his victims from our list of confirmed cases.
In light of Jordan's predatory conduct toward indigenous women, this outcome might seem unsatisfactory. Conducting surveillance after Buckner's death, investigators listened as Jordan induced other women to drink beyond the point of insensibility. (He continued such behaviour until his death in 2006, aged 76 years.) Justice John Bouck ventured into unfamiliar legal territory when convicting Jordan: "While he did not use a weapon, such as a gun or a knife to kill Vanessa Buckner, alcohol was his deadly instrument of choice." Although the same might have been said of his indigenous victims, it was not; Jordan was never convicted of any crime in connection with those deaths.
Any other cases?
There's also Thomas Svekla. He's a serial sex offender who abused girls as young as five years old. Although police considered him a suspect in as many as a dozen homicides in the Edmonton area, he received a single murder conviction for the death of Theresa Innes. (Svekla was acquitted for the murder of Rachel Quinney. He had directed RCMP to her body in a forest near Fort Saskatchewan. Although the judge acknowledged in his decision that the circumstances were "highly suspicious," he added that it "would be a travesty to act upon evidence of that sort.") We therefore do not consider him a serial killer.
There are several other individuals charged in connection with the deaths of multiple women, some of whom were indigenous. Because these charges have yet to be tested in court, we classify these cases as "speculative."
Which cases remain?
Our total of 18 confirmed aboriginal serial homicide victims includes the five aboriginal victims for which Pickton was convicted: Sereena Abotsway, Marnie Frey, Georgina Papin, Mona Wilson and Brenda Wolfe. To that we add Crawford's four victims: Shelley Napope, Eva Taysup, Calinda Waterhen and Mary Jane Serloin. Also included are Myrna Letandre and Jennifer McPherson, both murdered by Traigo Andretti, and Nina Joseph, killed by Edward Isaac. There is also Cynthia Maas and Natasha Montgomery, among the victims of Prince George's Legebokoff, who has filed an appeal of four first-degree murder counts. Lorna Blacksmith and Carolyn Sinclair, killed by Winnipeg's Shawn Lamb, are also included because these killings were consistent with the FBI definition, which is also used by the RCMP. (The same goes for the death of Ms. Serloin, for which Crawford was convicted of manslaughter).
Finally, there is Nina Courtepatte, murdered at 13 years of age by a group of people that included serial killer Joseph Laboucan, and Theresa Umphrey, a victim of Brian Arp.
What do you do with the other cases?
We do not ignore them. Even where convictions are absent, common sense dictates that some cases likely involved serial predation. Examples include Vancouver indigenous women who have been missing for more than a decade and whose DNA was found on the Pickton farm, and the women who died in Gilbert Jordan's presence. We classify such victims as "probable" victims of serial homicide. There are 17 in total.
We have classified many more cases as "speculative." Unsolved cases from the "highway of tears" and Edmonton area fall into this category, as do cases attributed to offenders charged with two or more homicides. We have identified 42 such cases. Some unknown number of these cases presumably involved serial homicide; the true proportion is not known, and likely never will be. Given that some of these matters are before the courts, and in others no charges have been laid, we do not identify the victims or any accused persons connected to these cases.
Why are there so many unsolved cases?
We cannot say for sure. One often-cited possibility is that serial killers are indeed responsible for some of them. Serial homicides are uniquely challenging to investigate and prosecute. Most homicides involve intimates, but serial killers are often not well known to their victims. Furthermore, some serial killers conceal the bodies of their victims; the lack of a crime scene or body can undermine investigations. Some serial predators commit crimes over multiple jurisdictions, but police agencies have had difficulty co-ordinating investigations. Also, investigators have sometimes resisted acknowledging that a serial killer may be active in their jurisdiction. It is worth recalling that one of history's most notorious serial killers, dubbed Jack the Ripper, was never identified or tried for the Whitechapel murders of 1888.
Racism's role is hotly debated. Some have alleged that homicides and disappearances involving aboriginal victims are investigated with less rigour than others. "The underlying reason why [Gilbert Jordan] went on so long is that cases involving native women are taken very lightly," Irene Richards, a First Nations family counsellor, told the Vancouver Sun in 1988. Echoes of her complaint are frequently heard in connection with other cases involving aboriginal victims. In his encyclopedic book studying Canadian serial killers, Cold North Killers, author Lee Mellor mused over the large number of sex-worker homicides in Canada per capita relative to that observed in his native Britain. "This is undoubtedly owing to the vast numbers of Aboriginal streetwalkers in western Canada whose disappearances usually go uninvestigated by police," he wrote.
Not everyone agrees. Convened to study the institutional failures that allowed Pickton to murder undetected in the Vancouver area for so many years, the Missing Women Commission of Inquiry heard this same accusation from groups representing interests in the aboriginal and Vancouver Downtown Eastside communities, as well as counsel representing the families of victims. Police agencies vehemently denied it. Commissioner Wally Oppal concluded "that systemic bias against the women who went missing from [Vancouver's Downtown Eastside] contributed to the critical police failures in the missing women investigations." He accepted, however, that investigating officers were on the whole "conscientious and fair-minded people" who did not intentionally discriminate against particular groups while conducting investigations.
Some of The Globe's data and information is inaccurate or incomplete. How do I let you know?
Please e-mail The Globe's MMIW team at MMIW@globeandmail.com. Any insights and information you can provide will be much appreciated.