Back in April of 2008, I was planning the launch of a Toronto Star blog called Map of the Week.
I filed a stack of access-to-information requests with various provincial and federal departments, as well as the City of Toronto, looking for postal code-based geodata. I asked everyone I could think of for anything I could think of – school vaccination rates, fishing licences, on and on.
One request that went in the mail was to Ontario’s Ministry of Community Safety and Correctional Services: I asked for the first three characters of the postal codes of people on the Ontario sex offender registry.
Sex offender registries are both controversial as a concept and potentially dangerous in practice, as vigilante violence in the United States has shown.
But a map showing (for example) that there were six sex offenders in a postal area of 28,000 people (I reasoned) couldn’t possibly be used to identify an individual. At this level, as one of the judges said later, it is “generic statistical information.”
One of the judges: I’m getting ahead of myself.
The ministry used a variety of delaying tactics to avoid dealing with the request – which included waiting 30 days to return a cheque made out to the “Province of Ontario” rather than “Minister of Finance,” and then just ignoring the request for months.
Eventually, I filed a complaint with Ontario’s Information and Privacy Commissioner. This at least forced the ministry to come up with legal reasons to withhold the information.
In August of 2009, 15 months after I’d filed the request, the IPC adjudicator dismissed the Ministry’s arguments and ordered the data released, writing that
” … I am not persuaded that disclosure could reasonably be expected to allow offenders to be identified, and for the same reason, disclosure could not reasonably be expected to facilitate the commission of a crime or to hamper crime control.”
In theory, the Ministry had the right to appeal the decision in the courts, but I didn’t take that possibility very seriously until, well, until they did.
Two years (and two jobs) later, I found myself in an impressive Georgian courtroom at Osgoode Hall on a hot June morning, with a lawyer from the provincial Attorney-General making arguments to a panel of three judges, and the IPC’s lawyer, neatly organized files in front of him, waiting to respond.
In the end, he didn’t have to – the judges left for lunch, and as court’s stated resumption time came and went, the IPC’s legal team brightened up and the AG’s lawyer visibly drooped. When the judges came back, half an hour or so late, they said what we had already worked out for ourselves – that the arguments they had heard in the morning were so weak that there was no need to take court time to respond to them. Their decision, published a week or so later, is here.
Well, that settles that, I thought. The province won’t keep appealing (oh yes, they did). Well, I thought, the Court of Appeal will never take the case (oh yes, they did).
So there we were, in June of 2012, back at Osgoode Hall, in an equally impressive courtroom with a different set of three judges, hearing the same arguments from the same lawyer – with, as it turned out, exactly the same result. The judges heard from the AG’s lawyer, broke for lunch, came back, and said they didn’t need to hear from the other side. Their decision, published a week or so later, is here:
“The Ministry offered little, if any, evidence that the partial postal code information contained in the record, alone or together with other information, may be used to locate convicted sex offenders within communities or may engender an offender’s subjective heightened perception of this possibility, thereby adversely affecting compliance with the Sex Offender Registry.”
Well, I thought, that settles that. After this, there’s only the Supreme Court (will they really appeal to the Supreme Court? ), and even if that happens, the Supreme Court won’t want to hear an appeal of a case with two unanimous decisions going the same way, will they?
The answer came in December of 2012, as official Ottawa wound down for Christmas: Oh yes, they did.
Which brings us to this morning, when the court will hear arguments from the two parties, as well as the federal attorney-general and the federal information commissioner, who each have ten-minute slots.
You can read the factums here.
A decision will probably come at some point in the spring. (I’m in Toronto, working on Rob Ford-related files.)
Here are some thoughts:
1) If the information is released, there is no chance that it could be used to identify a convicted sex offender in the community.
This was my starting assumption, but it was reassuring to have it confirmed by two appeal courts and the Information and Privacy Commissioner, which is a major advocate for personal privacy.
We don’t know what the story is without seeing the data. But in principle, information that gives people an insight into their communities ought to be released unless there’s a compelling reason to withhold it. I don’t see the compelling reason (neither did six senior judges) but obviously the province of Ontario does.
A good comparator is gun licence data: Identifying the homes of gun owners carries potential harm by facilitating gun theft and related violence. But the RCMP has released FSA-based data on gun licence holders for years. Why? Because it’s impossible to target a specific house for theft based on knowing that (for example) in N4K, Owen Sound, there are 29,000 people and 1,851 firearms licence holders.
Example: There exists in the data a handgun, a Smith and Wesson .357 Magnum revolver, legally registered and owned by someone who bought it on December 12, 2002. Would many people want to steal that gun and do harm with it if they knew where it was? Of course. What do we know about where it is? It’s somewhere in Scarborough (an M1-series postal code), along with 427 other legally owned revolvers chambered in .357 Magnum. Could someone steal a gun based on that information? Of course not.
Which brings us to
As the lawyers like to say, you can never predict what the courts will do, and, as mentioned above, I haven’t seen the data. But if it does end up being released, I predict 1) a map giving us some insights into the relationship between geography and crime in Ontario, in general, and 2) questions about why it took an excruciating five-and-a-half-year process ending at the Supreme Court of Canada to produce it.
Which brings us to
3) In contrast to similar departments in other provinces, Ontario’s Ministry of Community Safety and Correctional Services is remarkably hostile to the access-to-information process.
It’s unavoidable or at least unsurprising, to a point, that a branch of government that deals with incarceration and policing will have a more closed and rigid approach to the FOI process than others.
Unintentionally, however, I ran a comparative experiment last year, asking several provinces for postal area-based data for jail inmates. (The Star obtained this data – and also six-character postal code data – years ago for Ontario, has reported on it several times, and nothing bad has happened; the bigger problem is that jails don’t consistently collect postal code data on inmates, which makes the data that does exist shaky).
BC doesn’t collect inmate address data, Alberta couldn’t provide it because few inmates provide postyabut went to some effort to give us alternative data based on community names, and Nova Scotia produced it twice in its entirety.
Ontario, on the other hand, although they have lost an IPC adjudicator decision in the past over this exact data set, denied access on six different grounds:
– 14(1)(l): “the disclosure could reasonably be expected to … facilitate the commission of an unlawful act or hamper the control of crime.”
– 14(2)(d): “A head may refuse to disclose a record … that contains information about the history, supervision or release of a person under the control or supervision of a correctional authority.”
– 20: “A head may refuse to disclose a record where the disclosure could reasonably be expected to seriously threaten the safety or health of an individual.”
– 21(1): “A head shall refuse to disclose personal information to any person other than the individual to whom the information relates.”
– 21(2)(e): “A head, in determining whether a disclosure of personal information constitutes an unjustified invasion of personal privacy, shall consider all the relevant circumstances, including whether … the individual to whom the information relates will be exposed unfairly to pecuniary or other harm.”
– 21(2)(f): “A head, in determining whether a disclosure of personal information constitutes an unjustified invasion of personal privacy, shall consider all the relevant circumstances, including whether … the personal information is highly sensitive.”
The appeal is on hold pending the outcome of the sex offender case, since it involves a lot of the same issues.
4) This case shows the strengths of the Ontario access-to-information system.
The main advantage of the Ontario system is that the IPC can issue binding orders. (Call me a bad citizen, but I often don’t go to the effort of filing federal complaints because they don’t solve the problem. In principle, I probably should.)
The IPC is also excellent in its willingness to put legal muscle behind its adjudicators’ decisions if they’re appealed to the courts. This case is as good an example as you’re likely to find – few individual requesters have the resources to go toe-to-toe in the courts with a branch of government for years.
5) Beyond the issues raised by a particular story, the case has to do with law enforcement exemptions in Canadian access-to-information law.
The court is dealing with the Ontario access-to-information law, but the decision has implications for how other provincial FOI laws and the federal Access to Information Act are interpreted.
Section 14 of Ontario’s law provides for a law enforcement exemption “where the disclosure could reasonably be expected to” result in various forms of harm. The issue is how high a standard a government agency wanting to apply the exemption has to meet: Ontario argues for a “reasonable apprehension of harm,” while the federal Information Commissioner argues that the expectation of harm “must be clear, direct and convincing”:
” … the important objective of access to information would be thwarted by a mere possibility of harm standard. Exemption from disclosure should not be granted on the basis of fear of harm that is fanciful, imaginary or contrived. Such fears of harm are not reasonable because they are not based on reason. The words “could reasonably be expected” refer to an expectation for which real and substantial grounds exist when looked at objectively.
” … Proof of a clear and direct connection between the disclosure of requested information and the injury that is alleged by the party resisting disclosure was required by this Court in Merck This standard does not require that the party resisting disclosure prove that the harm will in fact occur but it must demonstrate something well beyond a mere possibility of harm.”