The Harper Government is not serious about fighting crime

Prime Minister Stephen Harper said the damnest thing yesterday. Asked whether he would reconsider calling a federal inquiry into murdered and missing aboriginal women, in the wake of the discovery of Tina Fontaine’s body in Winnipeg, he again refused. The reason he gave, however, was so strange. He said that such cases should be viewed as “crimes,” rather than as a “sociological phenomenon.”

Now I happen to agree with Harper that a federal inquiry would be a bad idea. But my reason for thinking that is the exact opposite of Harper’s. It’s precisely because the problem of violence against aboriginal women is primarily “sociological,” and not primarily a law-enforcement matter, that I don’t think a federal inquiry would be very productive.

To see why, just stop for a moment and reflect upon the statistic that is constantly being repeated in the press, that there are “1,200 murdered and missing aboriginal women in Canada.”* This brings up images of Robert Pickton, preying on women in Vancouver’s downtown eastside, accompanied by police indifference to the case of “yet another missing aboriginal woman.” And yet if we stop for a moment and think about what we all know about violence against women, it is easy to see that this is not the typical case. The usual rule of thumb for thinking about violence against women, including homicide, is that roughly 90% of it is “domestic” (mostly by family members and partners, some by acquaintances), with only 10% involving strangers.

As a result, if you called a federal inquiry into the case of the 1200 murdered and missing aboriginal women, what you would probably be calling would be an inquiry into around 1000 cases of (largely intra-aboriginal) domestic violence. Domestic violence against aboriginal women is, of course, a serious social problem. But I do not think it would be a good topic for a federal inquiry, for a variety of reasons. Most of these centre around the fact that it is a deeply “sociological” problem – in other words, that it is a product of a very complex set of social conditions. Of course, it also has a law-enforcement dimension, but these issues are extremely complicated as well, particularly on Indian reserves. People might disagree about this, but my own judgment is that a federal inquiry would not do much to clarify these issues, and the attention that it would bring to the problem would not really help to reduce it. (On the contrary, I could see it becoming deeply divisive, not just for Canadians but for aboriginal communities as well.) .

On the other hand, if the overrepresentation of aboriginal women among victims of homicide were just a “law and order” issue, as Harper maintains, then that would be a good subject for a federal inquiry. Such an inquiry might be productive, because the federal government could then figure out why law-enforcement is failing this community, and then do something to fix the problem. The policy levers are right there, since the government clearly controls the criminal justice system (well, indirectly – through its power over the provinces, which have power over municipalities, where most policing is done…), in a way that it does not control social conditions in aboriginal communities.

Okay, so then why would Harper say such a strange thing? I mean, if the murder of Tina Fontaine does not cry out for “sociological” investigation, I don’t know what does. It’s doubtful that she was randomly plucked off the street, like Holly Jones. Fontaine came from a deeply troubled family situation, had run away from a foster home, and was in a dangerous environment in Winnipeg. We don’t know who murdered her, but it seems obvious that the “social conditions” that she found herself in made her extremely vulnerable to exploitation and violence. And if you wanted to reduce the chances of this sort of crime occurring again, you would want to do something about these “social conditions” (in addition, of course, to prosecuting the person responsible for committing the crime).

The problem, for Harper, is that like many conservatives, he associates the “sociological” perspective on crime with the idea of diminished personal responsibility. In other words, he thinks that by trying to figure out what social conditions are more or less criminogenic, we are somehow shifting blame away from the perpetrators of crime, and putting it instead on “society.” This is why he considers it essential to stick to the “crime” framework, and not “commit sociology” (as he once put it).

Underlying this is a basic confusion about causation and responsibility (which lots of people share, not just Harper). It’s the idea that if we find out that exposure to condition x makes y more likely to do z, that this somehow diminishes y’s responsibility for doing z, and shifts it over to x. This is a common confusion, which arises from the thought that responsibility somehow “follows” the chain of causation, so if you trace back the cause to something beyond the individual, you are also shifting responsibility to something or someone beyond the individual. This is a philosophically incoherent view, which becomes apparent as soon as one thinks about how many different causes contribute to a particular event (see section 10 and 11 of this argument, which I always found persuasive).

Anyhow, the upshot is that just because a particular social condition that may obtain, e.g. coming from a broken home, increases the chances that one will go on to commit a crime, this does not mean that the perpetrator is any more or less responsible for committing the crime. The same thing goes for victims. To take an obvious example, the fact that Fontaine ran away from her foster home probably increased the chances that she would become victimized. But that obviously doesn’t make her responsible for having been murdered. It’s best to think of causation and responsibility as distinct notions, which do not track one another in any useful way.

This is a point that is familiar to philosophers, but difficult to communicate to the more general public. But setting aside the metaphysical issue, the obvious political problem caused by the confusion with respect to causation is that, if you want to try to control crime, you can’t just look at all the y’s who are doing it, you have to look at the x’s that are encouraging them. In other words, you can’t actually reduce crime just by whingeing about how evil the perpetrators are, or by threatening them with longer jail terms. This is because, as mountains of research shows, people are hugely influenced by their social environment, so that if you limit your crime control strategy just to the individual, while ignoring the environment, you’re tying both hands behind your back (all for reasons of bad ideology/metaphysics).

This is why I say that the Harper government is not serious about fighting crime. If they were, they would be deeply immersed in the sociological literature, trying to figure out what causes it, and how they can intervene to control those variables. What they are doing instead is trying to intervene in a “culture war” that exists largely in their own imagination.

The way they see it, liberals, sometime during the 1960s, basically became sympathetic to criminals. Ever since then, all the major trends in the justice system, all of the academic research, have basically been aimed at making criminals seem like victims, while displacing attention away from the real victims of crime. This is why most of the crime bills put forward by the Conservative government, like the recent “victims’ bill of rights,” are strange – because they don’t really have serious legal implications, they are aimed rather at bringing about a cultural change (e.g. by redefining what it means to be a “victim” of crime).

In other words, the government is trying to win some kind of a strange argument with ghosts from the 1960s. Which is fine, I suppose, it’s just that the Prime Minister should not allow this sort of cultural politics to interfere with the need to deal with serious issues, like figuring out an effective response to the tragic life and death of Tina Fontaine.


* At the risk of upsetting a lot of people, I just have to say a word about the “1200 murdered and missing aboriginal women in Canada” statistic, because I have serious reservations about the way that it has been used in public debate. This is because most of the time that it is mentioned, the person does not say what time period is involved. In fact, you have to search through the press pretty carefully to find out that the number is the sum of murdered and missing aborginal women over the past 30 years – which is to say, since 1984. If you ask “is that a lot?” most people would be hard-pressed to say. You would need to know how many Canadian women in total have been murdered or gone missing in that time period – and I’ve never met anyone who knows that off the top of their head.

By contrast, if you were to say that on average 40 aboriginal women have been murdered or gone missing per year in Canada, over the past 30 years, that makes it easier to assess, because most people know that there are not very many women murdered in any given year in Canada – more than 100, but certainly less than 200. So there is no question that aboriginal women are overrepresented. At this point, however, the question that naturally springs to mind is what the current rate is, and what the trend is. After all, 1984 is a long time ago, the murder rate has come down significantly. Has the murder rate of aboriginal women gone down or up, relative to the population generally? If it were going up, that would help to make the case for an inquiry. But I’ve never seen anyone present that information, people just use the brute number.



The Harper Government is not serious about fighting crime — 3 Comments

  1. The same kind of bad metaphysics is behind the “guns don’t kill people; people kill people” line.

    Those in favour of gun control are asking for the (sociological) causes of gun violence. The line quoted above changes the subject to the question of who is responsible for gun violence and answers – quite sensibly – that it’s the shooters. But, of course, we can hold on to that answer while still making the point that lack of gun control is a cause of gun violence.

  2. Ok, I’m going to avoid all the controversial parts of this post, which are tricky indeed, and just focus on one strand which I know is of interest to Joe in this piece and generally. Joe suggests that the peculiar Conservative crime policy might be understood as fighting a historical straw man who is sympathetic to criminals, ie “some kind of a strange argument with ghosts from the 1960s.” Maybe the arc is even longer than that, and here I’m influenced by a fascinating piece I just read by Joachim Savelsberg that uses a Weberian framework to explain the U.S. Federal Sentencing Guidelines. (!) The claim is that, prior to the late 19th century, classic liberal ideas like formal equality and individual responsibility were prominent in the western criminal law. This mode of “formal rationality” (which was problematic in both theory and practice for obvious reasons) then came to be attenuated by strands of substantive justice in the criminal law. Basically, the idea emerged that crime prevention required the complex social and psychological treatment of the individual rather than simply a state response to a straightforward notion of guilt. This project of “penal welfarism” (sociology!) then took hold, across North America in the early 20th century. Parole, penal rehabilitation, sentencing discretion, etc, are all part of that substantive, individualized project. But there are follow-up problems to this project, including sentencing disparity, lenience (violating to victims), and the fact that penal rehabilitation can be coercive, deceiving, futile (see Foucault). So we might see the Conservatives as driven by a neoclassical desire to “reformalize” the criminal law and clean up these excesses of substantive justice (best example: the subtly named Truth in Sentencing Act). Ok I am really just stealing from Savelsberg and applying it to Canada. So, see: Joachim Savelsberg, “Law That Does not Fit Society: Sentencing Guidelines as a Neoclassical Reaction to the Dilemmas of Substantivized Law,” American Journal of Sociology, Vol. 97, No. 5 (March 1992) 1346–81 at 1353.

  3. From the footnote:

    >the question that naturally springs to mind is what the current rate is, and what the trend is … I’ve never seen anyone present that information, people just use the brute number.

    For anyone who’s curious about this, it looks like the answer can be found in section 3.II here (ctrl+F “II. Trend in proportion of victims”):

    The absolute number of aboriginal female victims of homicide has been about constant since 1980, but because the absolute number of homicides of non-aboriginal women decreased over the same period, the *proportion* of total homicide victims who are aboriginal women has increased.