Getting Past the Bull: Evidentiary Problems and Breed Bans

Michael Robichaud

If you’re a Canadian with an internet connection and one or more functioning eyeballs, you’ve heard about Montreal’s pit bull ban (which has been suspended by a Superior Court judge pending a final ruling on its legality).

If I were to gauge the public’s reaction to the ban based on the tenor of my newsfeed alone, I’d say it’s been about as well-received as a Tommy Wiseau helmed musical adaptation of Mein Kampf.

Now, as far as sample populations go, my Facebook friends probably aren’t the most representative bunch. They’re overwhelmingly urban, young, small-l liberal, and a handful of them think the government is crop dusting us with nerve agents (or something). However, if I were to gauge the public’s reaction to the ban based on the tenor of my newsfeed alone, I’d say it’s been about as well-received as a Tommy Wiseau helmed musical adaptation of Mein Kampf. Not since Rob Ford’s days as a city councillor have municipal politics stirred this level of social media fracas: as of last Thursday, even Cindy Lauper (who has not yet accepted my friend request) had jumped into the fray. “It’s not the dog,” declared Lauper in a recent Facebook post, “it’s who is training and often mistreating them who should be banned.”

pit-bull-in-jailAlthough I’m not sure municipal governments have it in their power to ban people who wear Tapout shirts and like the Boondock Saints, Lauper’s words capture the sentiment of the “anti-ban” side of this debate nicely. Basically, any large dog that hasn’t been socialized properly can be a danger to the public, and singling out pit bulls—who, because of their reputation are stuck in a feedback loop of negligent ownership, violent behaviour, media demonization, and worsening reputation—is tantamount to a kind of breed discrimination.

The “pro-ban” side argues that, as a consequence of their long pedigree of dogfighting and bull baiting, pit bulls (a generic term that includes American Pit Bull Terriers, American Staffordshire Terriers, American Bullies, and Staffordshire Bull Terriers) are inherently more dangerous than other dogs. The fact that they account for approximately 32 per cent of dog bite-related fatalities in the U.S. despite making up only 6 per cent of the total dog population, they say, is a testament to that.

In Montreal, as elsewhere, the anti-ban side has been more organized and vocal, but the size of the opposing camps remains difficult to determine. A public opinion poll held shortly after the fatal June 8th pit bullattack that led to the ban suggested that a large majority of Quebecers (and especially Montrealers) favoured some form of Breed Specific Legislation (BSL)–but a lot has happened since then. Tempers, like the weather, have cooled, and reports that the dog involved may not have been a pit bull have further muddied the water. The most recent poll (taken on the weekend of October 15-16) shows the city of Montreal at what’s nearing an even split, and suggesting that the efforts of anti-banners are starting to pay off.

Wherever public opinion stands, it’s worth examining—as impartially as possible—the evidence deployed by supporters and detractors of this controversial policy, as both claim to be squarely on the side of science, expert opinion, and reason.

B.S.L.?  More like B.S. Data (swoosh!)

What struck me when I started researching this issue was just how awful most of the data is. Spoiler: it is extremely difficult to know whether or not bans are effective using the information most municipalities collect. In addition, both pro- and anti-banners alike are guilty of number-fudging, cherry-picking, and otherwise acting like the shameless lobby groups that they both clearly are.

When asked about the effectiveness of a ban, for example, pro-banners often quote the total number of pit bull attacks (which, unsurprisingly, tends to drop like a stone when a city wipes out its pit bull population).

What if, as pit bull attacks went down, Rottweiler attacks went up? Does that leave us any better off?

Whether or not this is a valid measure of public safety hinges on whether we accept the premise that pit bulls are uniquely dangerous dogs. Since the data on this is somewhat unclear (see part 2 of this piece), saying that pit bull attacks have gone down when a ban is instituted doesn’t tell us much about overall public safety. What if, as pit bull attacks went down, Rottweiler attacks went up? Does that leave us any better off? Unfortunately, many cities don’t collect detailed enough data on which breeds are biting and injuring people (some only distinguish between pit bulls and other dogs, and none I’ve seen have a reliable way of reporting the severity of bites), so they’re forced to justify implementing and repealing bans on the basis of conjecture.

That said, virtually all municipalities do operate a reporting service for those bitten by dogs. It’s from these very crude and easily obfuscated statistics that anti-banners draw most of their “scientific” legitimacy.

Ontario brought in a pit bull ban in 2005, for example, then saw total dog bites in the City of Toronto rise to historic highs in 2014. A slew of media dispatches from the anti-ban trenches touted these numbers as evidence that pit bull bans are ineffective, leaving out the teensy weensy little detail that the total number of both dogs and humans in the City of Toronto has also risen to historic highs, and that the two are living in increasingly close quarters as the condo boom densifies downtown.

When one compares the 2004 list of dog breeds who bite the most in Toronto to the same list in 2014, there doesn’t appear to be one single breed stepping up to fill the pit bulls’ shoes (who rounded out the list at number two in 2004, and dropped off entirely by 2014). Boxers are up there now, but so are Shih Tzus and Malteses. Overall, dogs on the top ten list account for a much smaller portion of bites in 2014, suggesting that many of the reports were isolated incidents involving varieties of breed that don’t often bite.

dog_with_muzzle_j1The only real way to gauge public safety before and after a ban would be to look at whether total deaths or serious injuries per capita related to dog attacks have gone down. This would address pro-banners’ claims that, because of the strength of pit bulls’ bite, their “grab and hold” instinct, and their ability to withstand significant amounts of pain before letting go, they tend to do a lot more damage than Malteses when they do attack. In short, we should be concerned with severity, not frequency, and we don’t really have much credible information on that. Deaths from dog attacks are generally too rare to be used in the service of statistical inference, and cities don’t maintain “bite severity ranking” systems.

To get around this, pro-banners will point to research by Merrit Clifton, who dutifully tracked media reports of dog attacks in the US and Canada from 1982 onwards, and argued that 70 per cent of “fatal and disfiguring” attacks on human beings have involved pit bulls. Anti-banners have spilt gallons of ink pointing to the methodological flaws in Clifton’s reporting, and I won’t spend much time on it here. Suffice to say that when one only looks at media reports of dog attacks, one picks up media bias, and given the public fear and hysteria surrounding the issue, news outlets over-reporting on pit bull attacks seems par for the course.

So where does that leave us, fellow lovers of evidence-based policy? Anti-banners equate themselves with expert opinion, but in reality most experts they cite are connected with groups actively lobbying against BSL. Pro-banners don’t seem to see a problem with measuring the success of a policy based on its output (lower levels of pit bulls) rather than its outcome (whether or not it makes the public any safer). And neither pro, nor anti-banners seem above citing a bunch of specious, nonsensical statistics, then mic-dropping and strutting off stage like they just won the debate.

Writing on the “marijuana scare” of the late 60s, Erich Goode pointed out that when it comes to issues that scientists (or policy makers) either don’t understand well, or can’t agree on, people simply choose an expert opinion that jibes with their own personal values. I would argue that this is the phenomenon we’re seeing play out in the dispute over breed bans.

Puzzlingly—given how heated and drawn out the furor over BSL has been—this isn’t a policy area that’s attracted much attention from impartial scholars (maybe because of the infuriating paucity of data). But there are a few fairly legitimate studies from which we might draw inferences—both on the alleged dangerousness of pit bulls, and the (in)effectiveness of bans—and this will be the topic of part two of this piece.

Stay tuned.

Mike is a graduate student at the University of Toronto (previously McGill). He grew up on Vancouver Island, but now sticks mostly to the ten square blocks surrounding his Toronto apartment. He’s worked as a policy researcher, a jug-hound, and just about everything in between. Mike’s interests include politics, post-punk, and 80s slasher flicks. He can’t believe he’s thirty-one.

Advertisements

2 responses to “Getting Past the Bull: Evidentiary Problems and Breed Bans

  1. Hey there,

    There actually is data showing BSL is effective at reducing the severity of injuries caused by dogs. It’s the Winnipeg study.
    What it shows is that post-BLS, the number of hospital visits for those requiring surgical intervention goes down.
    What goes up is the number of doctor visits for those wanting a rabies shot, such as those being licked by an unknown dog.
    It can be a bit confusing to understand this, since the study does use the catchall term “hospitalization” for all events.

    In the end, it does make sense. All dogs can be enticed to bite, but if you get rid of large dogs that are known to be in the care of irresponsible owners, the number of life-threatening or life-altering bites will drop.

    • Hi Kevin,
      Thanks for the comment, I appreciate your interest in this. The Winnipeg study is definitely on my radar, and is one of the pieces of evidence I’m going to consider in part 2 of this piece. Have a look back at the blog in a couple weeks if you like. Thanks for reading.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s