Like The Hub?
Join our community.

‘Policing is failing Canadians’: Law professor Kent Roach on how and why it must change

Podcast & Video

Today’s episode of Hub Dialogues features host Sean Speer in conversation with University of Toronto law professor Kent Roach about his book, Canadian Policing: How and Why it Must Change.

They discuss why Canadians must demand better governance from police, the failure of “Defund the police” messaging, and the issue of wrongful convictions in Canada.

You can listen to this episode of Hub Dialogues on Acast, Amazon, Apple, Google, Spotify, or YouTube. The episodes are generously supported by The Ira Gluskin And Maxine Granovsky Gluskin Charitable Foundation.

SEAN SPEER: Welcome to Hub Dialogues I’m your host, Sean Speer, editor-at-large at The Hub. I’m honoured to be joined today by Kent Roach, a law professor at the University of Toronto, and an award-winning scholar on a wide range of legal subjects including law enforcement, counterterrorism, and national security. His most recent book, Canadian Policing: How and Why It Must Change has received considerable critical acclaim including being shortlisted in 2022 for the Balsillie Book Prize. I’m grateful to speak with him about the book and its case for reconceptualizing how policing is carried out in Canada. Kent, thanks for joining us at Hub Dialogues, and congratulations on the book.

KENT ROACH: Well, thank you very much for inviting me, Sean.

SEAN SPEER: You open the book with the following passage, “Canadian policing needs to change. It needs to become less violent and discriminatory, better governed, and more effective. All of this needs to be done for the sake of both the public and the police.” What’s your main argument? What are the chief problems requiring a radical overhaul of policing in Canada?

KENT ROACH: Well, policing is failing Canadians both by over-policing a wide variety of groups, especially Indigenous people, in the sense of street stops, violence that could be avoided, and so on. This is, obviously, Indigenous people, but also Black people, people with mental health challenges, and so on. At the same time, Canada is under protecting some of the very same groups, and so part of the book is to say that these concepts are actually related. They may seem counterintuitive but they are related.

The second main reason why I’m writing the book—and I started it really around the time of George Floyd’s murder, and of course as we’re talking today we’re dealing with Tyre Nichols’ murder as well—is I really want this to be a call to Canadians to demand better governance of the police. Ultimately, with the kind of police we have we should be able to assert democratic responsibility and control. I’m sorry to say that since the book has been completed, I think we’re moving backward.

The book includes material on the policing of the so-called Freedom Convoy but one of the things that really bothers me is whether it’s with discourse over that, or in relation to the Mass Casualty Commission, we have this sense, this growing sense, that the police have operational independence from those who we elect to government. I’m very concerned that this, effectively, could make the police a law unto themselves.

There are really two main problems that the book addresses.One is how do we understand and move away from this conundrum of both over policing and under protecting a range of disadvantaged and vulnerable groups? And second, how do we, as a society, claim a right to govern the police without interfering with the operation of the rule of law?

SEAN SPEER: We’ll come back to both of those main lines of analysis over the course of this conversation. Before we get there, I want to talk a bit about methodology. One of the most interesting parts of your research is that this isn’t a typical academic book based solely on data and literature and so on. You’ve actually joined police officers on their shifts in different parts of the province. Do you want to talk about those experiences? How have they come to shape your views on these subjects?

KENT ROACH: Yes, thanks. Some of my early research with my colleague, Marty Friedland, was spending a lot of weekends riding around in the back of squad cars in Niagara Falls, New York, and Niagara Falls, Ontario. Although you could say that that research is data, in Niagara Falls in New York, I saw something that can explain I think what happened with the awful death of Tyre Nichols. There were so many police officers that whenever we responded to a call, inevitably, within a minute or two of the first cruiser arriving, there were three or other cruisers arriving.

There was also in Niagara Falls, New York, not something like the scorpions kind of a street crime unit but there was a swing shift that came on during the busiest times, I believe it was 7:00 PM to 3:00 AM. The officers on those shifts were self-selected. They were I think a little bit more aggressive. They wanted to be busy and make arrests and so that has stuck with me. Where, in Niagara Falls, Ontario, there was a sense that it was only on a Saturday or Friday night at that time six or seven cruisers, often one-person cruisers, to police a much larger geographic and demographic area in Niagara Falls, Ontario.

That led to a much more reactive and cautious, policing style. That was some thinking that has influenced me, that having more police is not always the answer, and having police respond to mental health calls is also not the answer. Now, since that time, I’ve also been able to work with the Ipperwash Inquiry and the Air India Inquiry, and more recently Justice Epstein’s review of the Toronto Police missing persons’ investigations. That gave me a sense of some of the challenges because the missing persons’ investigations are dealing with sexual minorities who went missing from Toronto’s gay village.

Again, you have this idea that they were under-protected, in part, because of this historical legacy of over-policing gay men, in particular going back in Toronto to the infamous bathhouse raids. Then finally, I was able to spend seven years in total being a member of an expert panel assembled by the Canadian Council of Academies, first on the future of policing and second on policing of Indigenous communities. In both of these, I was able to work with scholars, non-legal scholars who are experts on policing often from sociology.

Even more importantly, I was able to work with retired police chiefs from Canada, the U.K., and the United States, and retired police officers including retired Indigenous police officers. This also gave me, I think, a perspective. And of course, after George Floyd, there was a sense of defunding the police, and although I think the police should be defunded to a certain extent, I really wanted to write a book that both police officers and activists would read even if it was at the risk of annoying both of those sides.

Then the last thing—and this may just reflect the fact that I’m getting older and I’m probably not going to write too many more books—is I also thought I should be honest about my own personal history. I’m, obviously, a white man privileged as a law professor. I reveal in the book that my life was saved when I was an infant and had a subdural hematoma by a Montreal police detective who rushed me to the hospital, likely saving my life. In early years before I wanted to be the catcher for the Montreal Expos, I very much wanted to be an RCMP officer and would ask my parents if I would reach the six-foot height that was required.

I’ve forgiven my parents for lying to me, because I’m quite short, although I’m taller than my parents. Now that I have children of my own, including one who’s a mental health nurse, I have a sense of why they didn’t want to break that temporary dream that I had as a child.

SEAN SPEER: As you say, Kent, these are all of the various factors and perspectives that go into the book and its analysis. I want to talk a bit about its origins. You mentioned earlier that you started working on the book following George Floyd’s killing by police officers in Minnesota. What motivated you about his tragic murder and what are the implications for Canada? Maybe to put it differently, are we wrong to assume that Canada is immune to these types of incidents?

KENT ROACH: Yes. One of my targets in a lot of my recent writing is this knee-jerk attitude that I think some Canadians still have, perhaps less than there used to be, that problems whether it’s policing or wrongful conviction racism are an American problem. Although it is different in Canada, and perhaps of a slightly lesser magnitude, the first substantive chapter of the book simply goes through what happened in Canada in June 2020 which, of course, is the month after George Floyd died.

I didn’t set out doing this. Yes, I knew I was going to write a book so I was paying close attention to what was happening. In that chapter, I go through what happened in that one month. It starts with Chantel Moore’s death in Edmundston. Rodney Levi’s death also in New Brunswick at the hands of an RCMP officer, both having to do with mental health. Then revealing Chief Allan Adam being subjected to a flying tackle and grounding by a second RCMP officer who arrives in a parking lot in Fort McMurray, the Regis Korchinski-Paquet incident and Ejaz Choudry death, and Peel in Ontario.

That chapter I hope serves as a wake-up call to that readers that, although it may not be as dramatic as George Floyd—in large part because there was very little video, except in the case of Chief Allan Adam—we do have similar problems in Canada.

SEAN SPEER: I want to take up your observations about the dichotomy between over and under-policing, and its connection to the problem of systemic discrimination. You cite, for instance, that 30 percent of our prison population is Indigenous. Help us understand the role of police discrimination in producing these outcomes relative to other possible contributing factors.

KENT ROACH: There, of course, are debates about the issue of indigenous over-representation. Is it all there because of intentional discrimination, whether it’s by the police or prosecutors or judges? I don’t think that that’s the case but I do think that it’s a huge problem when over 30 percent of the prison population is Indigenous compared to 5 percent of the population. It is, in comparative terms, a grosser over-representation than we see in United States prison populations with the African American or Black population, or in New Zealand or Australia with respect to the Indigenous population.

I think one of the things that I’ve learned over the years is that within these communities, this then becomes a personal issue for individuals, families, and friends. This means that if someone in your family, you think they have not been treated fairly by the police when they’ve been targeted or arrested or prosecuted, how are you going to feel when either that same person or that person’s sister then goes missing?

Are you going to feel confident in the police response? Are you going to think that the police are there genuinely to help you? You have to overcome this legacy of distrust. Certainly, with Indigenous murdered and missing women and girls inquiry, they spoke an awful lot about how the legacy of over-policing creates a legacy of distrust that then may hinder the ability of police to solve crimes involving those very same vulnerable groups and may hinder that willingness to cooperate.

Again, to go back to Justice Epstein’s report missing in this, she quite clearly found that the Toronto police didn’t have the awareness of the LGBT community that would have perhaps led to an earlier arrest of the person, the serial murderer who was eventually convicted with regards to the missing and murdered men. 

Again, this creates a distrust in the community if the police are seen as those who take sons and sometimes daughters away. This is something that I think that it’s a challenge that the police face, but more importantly it’s a challenge that all of Canada faces because too often we think of things as either you’re the accused or you’re the victim.

Of course, I’ve been involved with initiatives with respect to Indigenous sentencing, and one of the lessons of that is often even if the accused is guilty of committing a crime, that is related to the way that the person has been treated and the challenges that the person has experienced over the past. This dichotomy between the bad guys and the good victims just is an unrealistic and counterproductive dichotomy.

SEAN SPEER: The book covers various sources and causes for the growing problems with Canadian policing, including culture, governance, human resources, and even its underlying philosophy. Do you want to talk a bit about these underlying causes and how they may actually interact with one another?

KENT ROACH: Yes. Thank you for that. That’s an excellent question. The Canadian police have their historical origins with the Irish Constabulary. The Irish Constabulary was always a much more militaristic operation than the London Police, Sir Robert Peel’s bobbies. People may know that in London, there was a lot of skepticism about having public police. One of the ways that Sir Robert Peel overcame was to say that this isn’t going to be a military occupying force like we have in Ireland, or as is more the norm on the continent. That the police is going to be the public, that they’re just going to be one of us. Also, subject to governance by the public.

In Canada, in part, because we were a colony, we really started off with the Irish model with what has now become the RCMP created by Sir John A. Macdonald to open up the west. Obviously, connected with the displacement of Indigenous people, but also just the colonial militaristic model. If you look at provincial police forces, whether they be in Quebec, Ontario, or Newfoundland, and for a time in Alberta, in British Columbia, they were very much in that model. Like the military, the RCMP always wore red, whereas in London, Peel’s bobbies wore blue in order to distinguish them from the military.

I found an the interesting historical antidote when Alberta for a time got rid of the RCMP, and of course, they’re thinking about doing it again, they gave the Alberta Provincial Police blue uniforms and they protested until they got red uniforms back because they wanted to be militaristic. I think that this explains why we train police the way we do. There’s a real emphasis on physical fitness almost combat training. I’ve talked to RCMP officers who said, why do I have to ride a horse? The RCMP training camp is called Depot. Sold by the RCMP is a scary boot camp and called Depot because that’s what the Royal Irish Constabulary called its headquarters in Dublin, of course, when Ireland was all a British colony.

I think there’s a link between historical origins, paramilitary image, paramilitary training, and lack of civilian governance or lack of effective civilian governance. When you think about the RCMP or indeed the OPP, or Quebec police, we have no police service board. We simply have the minister who is responsible, the minister of public safety, and the commissioner. Right now there is a fight between the minister who has told the commissioner to eliminate a certain whole type of neck hold and he commissioner, who is apparently dragging her feet about implementing that.

It seems to me that on matters like that, in a democracy, ultimately, it’s the responsible minister who should win that battle. You may say, well, you’re only saying that because you like the minister of public safety. It seems to me that—if it is a matter for the minister of public safety, if we change government and another government had another idea and wanted to bring back that sort of neck hold, then I would say I might not like it as a citizen, but that is what we need to do in a democracy, at least subject to a Charter challenge or some other ruling.

I think that there is really a direct line between this paramilitary Irish Constabulary origins and this governance deficit which I think is one of the big themes in the book is this idea that the police are under-governed. The last thing I’ll say is and I’ve learned this, I guess, from some of my law and economics colleagues at the University of Toronto, is that policing is real expensive.

The average police officer earns over $100,000. I don’t begrudge them that but when we talk about, as we do now, adding more police officers we have to realize that with the one exception of Indigenous police forces, the cost of policing in Canada has been expanding since about the ’80s. At some point, we’re going to have to rethink because adding more police is certainly legitimate if that’s what we want to do as a democracy but they’re very expensive. At some point, it’s going to be a choice between police, nurses, and doctors.

Perhaps because I’m a father of a nurse, my bet is that a lot of people want to make sure that doctors and nurses are there. It’s not clear to me that there is a linear relationship between police strength and public safety. Again, I go back to my initial experiences in Niagara Falls, New York, and Niagara Falls, Ontario. Niagara Falls, New York, had lots more cops but I didn’t necessarily feel safer indeed, I felt probably a little bit more at risk when I was walking late at night in Niagara Falls, New York, as compared to Niagara Falls, Ontario.

SEAN SPEER: That’s a comprehensive answer. Thanks very much for that. I should just say in parentheses for listeners, there’ll be parts of Professor Roach’s analysis and recommendations that I’m afraid we won’t be able to get to in this conversation, including, for instance, his recommendation to effectively abolish the Depot training model and replace it with a more decentralized model with training facilities and capacity across the country but that only reinforces the need for you to ultimately read the book.

If it’s okay with you, I’d like to shift the conversation to solutions. Like most policy areas, policing in Canada touches on federalism questions, including the RCMP contract model with various provinces, which you’ve written elsewhere, “may be numbered.” What’s the role of federalism in your story and what ought to be done to improve the functioning of police services in light of these intergovernmental considerations?

KENT ROACH: There are two dimensions. One is what is the respective role of municipalities or regional government and the province with respect to local or regional police? Of course, in Ottawa, we saw the dysfunction of the Ottawa Police Service Board during the February convoy. I think one of the things that we need to think about is what is the legitimate provincial role with respect to policing and what is the legitimate local role. I think that some of this is related to the fact that our Constitution, the 1867 Constitution, really doesn’t conceive of local government as an entity, although it affects our quality of life greatly.

Secondly, federally, the RCMP provides local or regional policing in 8 of 10 provinces, and very importantly, in all three territories. This means that an RCMP officer who is policing Nunavut really has two bosses. One boss would be the responsible minister of the Nunavut government and another boss would be the commissioner, and through the commissioner, the minister of public safety. Sometimes when you have two bosses, you have no bosses.

One of the things that I look at is that in the Yukon, they have come up with an innovative governance structure, where, at least at the territorial end, they have a police council which has Indigenous representation, people appointed by Indigenous groups, the deputy minister of public safety. If they wanted to take more community safety approach, there’s nothing that prevents them from putting the deputy minister of health or the minister of health on there, to the extent that police deal with issues of health or housing or so on.

I don’t think that contract policing can disappear overnight. The city of Surrey has gone back and forth largely because of concerns about how expensive it would be to have their own police force. I do think that we need to have more effective local and provincial and territorial governance if the RCMP is to continue to be the local and regional police in most of Canada.

Then at the same time, maybe the RCMP has to be broken into two different entities because the RCMP is now not doing a particularly good job of contract policing which takes up most of its resources but is also not doing a particularly good job of federal policing, whether it be national security or money laundering, which takes up about a quarter of its resources. Again, if you try to do too much, you often end up doing everything poorly.

SEAN SPEER: You argue in the book that “Defund the Police” is a divisive and ultimately unsuccessful political slogan but that its underlying premise of shifting police resources to social services is broadly correct. Let me ask you a two-part question. First, what areas or types of social services do you think could produce a return on investment in terms of reduced crime and criminality? Second, how limited could policing be to have us still able to maintain basic order and stability in our society? How should we think about that, Professor Roach? Is there an optimal level of policing resources?

KENT ROACH: Yes. I see that as much as a research question. One of the things that, certainly, since I’ve written the book, Defunding the Police seems, in many ways, even more dead, that police budgets, including in Toronto, are increasing. I talk in the book about a report that was issued by the Toronto neighborhood centres called “Rethinking Community Safety” in 2021. Interestingly enough, it was signed on by the Gerstein Center, which does a lot of mental health issues, and it was signed on by Black Lives Matter, and it proposed reallocating 10 percent of the metro Toronto’s police budget, which is over $1 billion, to community programs.

The community programs can deal with mental health, can deal with housing, can deal with frequent customers of the system, can deal with the homeless, and so on. Part of its plan was that some of that money should also be spent on evaluation, what works and what doesn’t work. Canada doesn’t spend nearly as much work on, enough money on, researching what works and what doesn’t work. We shouldn’t think that one-size-fits-all. As a starting point, I think 10 percent of the police budget going to creative concerns and subject to evaluation would have been a defensible approach. Now, that still would have left Toronto with a 900-million-dollar budget.

I certainly realize that, like everyone else, the police are the one 24/7 response. Even doing things like triaging 911 calls, what is best to send the fire department and the paramedics? When do they need the police? These are all things that I think we need to do. I fear now, especially in Toronto with very high-profile public crimes on the TTC and elsewhere, that we will simply go back almost like Pavlov’s dog to the idea that we need to throw more money at the police.

Again, the comparison becomes down to what is the marginal cost of adding one more police officer not only at his or her 100,000 plus salary but also the attendant court costs and other court costs, versus funding a community organization which has the trust of a particular community and can employ volunteers in a much more economically efficient way than the police.

SEAN SPEER: To that point, I’m speaking to you from New York City, which, in the 1990s, essentially adopted a policy of over-policing to deal with concerns about criminality and public safety. These efforts were subsequently criticized for some of the issues that we’ve been discussing, and in turn, led to, for a lack of a better term, a more humanistic approach. In the past five years or so, however, we’ve witnessed renewed concerns about crime and public safety, and it’s led to something of a ratcheting back in the city’s policing philosophy. I guess my question for you is, is there a trade-off here? If so, how does your policing model aim to strike that balance?

KENT ROACH: Well, I think that that’s where governance comes in. Certainly, the Broken Windows policy was supported by New Yorkers at the time. I think that part of the cost of democratic policing is that the public has a right to make mistakes. At the same time, I think if we invest more in evaluating whether things like Broken Windows work or comparing it to other alternatives, so in Ontario, we have community health and well-being plans where we’re trying to take a more multidisciplinary approach. Again, it seems to me that this is largely a governance problem, in that we have the health silo, we have the housing silo, and we have the police silo.

We know in the real world these things are all connected and so I see these pilot projects that were proposed in the Rethinking Community Safety Report, which is a terrific report, well researched, only 20 pages, as a start. I don’t try and I don’t think it would be right to present one blueprint that would apply from coast to coast to coast, because that would be at odds with my sense that policing no less than health or any other government service, should be subject to democratic direction through elected governments.

SEAN SPEER: This has just been a great conversation. I appreciate your time. May I wrap up with a final question about another line of your research on the issue of wrongful convictions? Let me ask a two-part question. First, do we have a sense of how common they are in Canada? Second, what are some of the main factors or patterns that you’ve come to identify in your research?

KENT ROACH: Well, thanks for asking. With Amanda Carling and some students at the University of Toronto, we are going to launch on the 20th of February Canada’s first registry of wrongful convictions, and it will be at We have recorded 83 remedied wrongful convictions in Canada. Now, what are the number of unremedied wrongful convictions? Nobody knows and nobody ever will know, but the police do play a role in wrongful convictions.

One of the things that is striking from our research is that 1/3 of the wrongful convictions are cases where no crimes were ever committed. Cases like the Charles Smith case, where someone had suspicions and thought that babies had died on purpose, whereas when others looked at it, unfortunately, years later, they found out that the cause of death was natural, accidental, or undetermined.

Again, this goes into the fact that to the extent the police are responsible for these imagined crimes, wrongful convictions, which are 1/3 of our remedy wrongful convictions, this underlines another danger of stereotype. The stereotype is not only the proactive profiling stop, but it can actually, in some cases, go as far as a wrongful conviction. Now, having said that, one of the messages in my next book, which is called Wrongfully Convicted, and will be out in April with Simon & Schuster, is that the police shouldn’t always be blamed for what is sometimes called tunnel vision.

We think of David Milgaard, G. Paul Morin, the famous wrongful convictions, as cases of police tunnel vision, and they may well have been, but one of the things that I argue is tunnel vision is simply confirmation bias. It is something that we all do when we have a mass of conflicting data. That if police and this goes back to Justice Epstein’s report, one of her findings is, the police didn’t use case management systems, computerized case management systems, computerized victim linkage systems.

By using computers, this is a way that we can counter the limits of our own minds, which is we all form tentative hypotheses. I explained this to my students by saying, when I mark your exam, which has three questions, I’m aware that if you do really poorly on the first question, I have to fight the tendency to think you’re going to do poorly on the next two questions.

I mean, confirmation bias or tunnel vision, unfortunately, is just the way our minds work but good policing can also use things like case management and victim analysis, which can reveal both the true perpetrator in cases where there is real crime but can also help reveal when the police have the wrong perpetrator and when they may have wrongfully convicted someone or been close to wrongfully convicting someone.

SEAN SPEER: Oh, well, that’s great. I encourage listeners to check out the site and of course, Professor Roach’s forthcoming book. We’ll have to have you back on the podcast then but for now, I’m grateful for you to have joined us today. The book is Canadian Policing: How And Why It Must Change. Kent Roach, thanks for joining us at Hub Dialogues.

KENT ROACH: Thank you, Sean.