Another analysis of Bill S-210 has emerged. Emily Laidlaw, a law professor at the University of Calgary, says it’s likely unconstitutional.
Bill S-210 is an unconstitutional mess. This is one of the many findings I had last year when I wrote my analysis of Bill S-210. In response to the growing concerns that have mounted since, staff working for Senator Julie Miville-Dechene, the official author of the legislation, published a so-called “FAQ” to push back against the concerns. Unsurprisingly, the “FAQ” was just as badly written as the legislation itself. As a result, it didn’t really change anyone’s minds in this debate. The senator, herself, started to grow frustrated that she was losing the debate and started accusing critics of being “ridiculous”. Unfortunately for the senator, however, her counterpoints ironically proved the points critics were making in the first place.
Simply put, the legislation, known as Canada’s age verification bill, would subject Canadians to highly intrusive surveillance. It would also implement mass internet censorship at the ISP level at the behest of the government – something that the Canadian Charter was designed to prevent. Additionally, it forces all Canadian websites to choose between risking a fine of $250,000 or implementing technology that doesn’t really exist that would put people’s sensitive personal information at risk. Privacy provisions are toothless as it simply asks nicely to destroy personal information afterwards – this with absolutely no provision in the bill or existing law that contains penalties for failure to do so. Most distressingly, it does nothing to keep children safe and, ironically, would probably put them at greater risk on the internet.
The pushback against this law is certainly growing at this point. Digital rights organization, Open Media, released their own FAQ, pointing to the critical flaws of the legislation. The FAQ basically agreed with our analysis and the concerns of others. At the same time, the organization also created a petition against the legislation for anyone to sign. In addition to this, the organization signed an open letter with 9 other civil rights organizations and experts to urge lawmakers to not pass this bill on the basis that it’s unconstitutional and beyond repair.
Yesterday, we wrote about another analysis of the legislation by David Fraser (AKA Privacy Lawyer). His analysis shed additional light on why the legislation is so bad. Essentially, the defences of content being for educational purposes among other things is toothless because the law says that a court must issue a judgment against any defendant based only on mere suspicion of wrong-doing. This as opposed to other legal standards such as the criminal standard of “beyond a reasonable doubt” or even the civil standard that is much lower than that. In short, if there is a suspicion, that equates to automatic guilt. Like many others, Fraser concluded that this bill is poorly written and very likely unconstitutional.
Now, another legal expert has weighed in. That is Emily Laidlaw, a law professor at the University of Calgary. She wrote a lengthy analysis of the legislation. What is interesting is that Laidlaw wrote in her analysis that she supports age gating content. So, she’s a supporter of the concept in general. However, she notes that Bill S-210 is not the way to do so. She did note that it was surprising to see this private members bill get so much traction despite the forthcoming online harms bill, expecting instead that it would wither on the legislative vine in the process. Instead, it is looking increasingly likely that it could be passed anyway.
She writes that there are four fundamental problems with Bill S-210. The first problem, for her, is that this sort of concept belongs in the online harms bill:
Age Verification Belongs in Online Harms Legislation
Age verification fits in online harms legislation, because true child protection requires a holistic approach. I expect that the Federal Government will introduce legislation modelled on the UK or EU. While different from each other, the core approach is the same, namely requiring online platforms to be responsible corporate actors by imposing a duty of care/due diligence requirement to manage the systemic risks of harm of their services.
Recommendation: Age verification should be addressed in online safety legislation as part of a broader package of child protection measures, including algorithmic accountability, content moderation, platform design, and commissioner oversight.
Personally, unless the Online Harms bill is drastically different from the so-called “consultation” version, I think the online harms legislation is also a terrible piece of legislation that would essentially carpet bomb the internet while solving nothing. So, for me, the only practically benefit of roping this legislation into the online harms bill is a delay in the proceedings. It’s not much, but it’s not nothing, either.
The second major problem is the fact that the scope is far too broad:
The Scope is far too Broad
Bill S-210 captures any and all internet services in its scope. Section 5 provides that “any organization that, for commercial purposes, makes available sexually explicit material on the Internet to a young person is guilty of an offence”.
As OpenMedia commented in their letter, it would put much of the internet “behind an age gate”. ‘Makes available’ is broad. Senator Miville-Dechêne confirms that the target is pornography providers. However, as drafted s 5 includes any internet service that enables content to be accessed, which would include social media, search engines and internet access providers. All of these services ‘make available’ sexually explicit material, because users post it even if it is against their terms and conditions. And all of these organizations make such material available for commercial purposes, because their profit derives from advertising or subscription services and similar. The provision should be re-drafted to narrowly target commercial pornography providers.
Further, sexually explicit material refers to the definition in s 171.1(1) of the Criminal Code, RSC 1985, c C-46 (Bill S-210 at s 2), which includes visual, audio, and written material. This is too broad for the implementation of age verification and would include sexually explicit written material.
Recommendation: Narrow the scope to commercial pornography providers or platforms whose dominant purpose is to make available sexually explicit material. Narrow the definition of sexually explicit material to visual material.
This is something I absolutely agree with and matches my concerns when I did my analysis. Despite claims to the contrary, this legislation applies to every website. This is thanks partly to the term “any organization”. What’s more, there are no provisions that fleshes out a specific scope. An example might be that this legislation only applies to websites that offer at least a certain percentage of explicit material. Just about any interpretation of pornography could apply. Heck, there are some people out there that consider pictures of luxury vehicles to be pornography (yes, that was a Clarkson reference if you saw the clip at 3:45 in that video).
In fact, when Senator Miville-Dechene asked if a provision like the website must have 33% of material be adult content to be scoped into the bill, I actually directly responded to her by saying that, yes, that would be an improvement. It wouldn’t solve all of the problems of the bill, but it would be an improvement. Like what she did with the Online News Act and the Online Streaming Act, she did what she did best and ignored me since I was a critic of the legislation in the first place despite my comment being constructive criticism.
At this point, this is a massive problem with the legislation. What’s more, making available is, indeed, a far too broad of a term. For instance, say a Canadian website maintains a web forum. Someone nitwit goes against the terms and conditions of the website and posts adult content. Under this bill, that would be a problem. Even if the moderators of the forum take down that material, someone else could also make a complaint that the website is making available adult content in violation of this bill. As a result, that opens up legal liability for the website. Some might argue that the website is acting in good faith and using due diligence, but the problem is, as per Fraser, that’s not a defence to a complaint in the first place.
If supporters insist that this is meant to be a bill to only go after adult website, maybe, I don’t know, put provisions in the bill that confirm that? Has that crossed any supporters minds at any point? Obviously, it is not in the bill at the moment.
The third problem Laidlaw points out is that the site blocking provisions are likely unconstitutional:
The Website Blocking Provisions are Likely Unconstitutional
The enforcement authority may apply to the Federal Court to order website blocking if an internet service makes available sexually explicit material and fails to take remedial steps ordered by the authority (at ss 8-9). The Bill appropriately leaves website blocking as a last resort mechanism and requires a court order. However, in a surprising twist, the Bill explicitly contemplates overbroad website blocking and appears to condone it (at s 9(5)).
The role and limits of website blocking under human rights law has been in front of the courts for many years in the UK and the EU, and the Federal Government can draw from their experience to amend s 9. The general principles are as follows:
- Website blocking should be a last resort and necessary to achieve an important objective;
- It should only be ordered by a court;
- It should be proportionate as in narrowly tailored, such as time limits, blocking of specific content or pages and not an entire website;
- There should be stringent safeguards in place to ensure procedural fairness and to guard against collateral effects, such as targeting content beyond that which is illegal. See commentary on recent EU cases here.
Recommendation: Delete s 9(5) and insert criteria for website blocking.
It’s these provisions that have compelled me to dub Bill S-210 the “Great Firewall of Canada”. We’re talking about a government who is ordering the blocking of websites. The constitutional problems with that are simply inherent. As Laidlaw points out, the blocking of websites should be an absolute last resort in extreme circumstances, not a knee jerk response and the first tool you reach for when talking about “offshore” websites. The website blocking provisions only compound the problems of this legislation even further.
The fourth problem for Laidlaw is the oddly worded comments that the verification criteria provisions are not stringent enough. On the surface, this made me think, “Um, it envisions technology that doesn’t exist. how is it helpful that we should strengthen the criteria?” That, however, is not what she means by this:
The Verification Criteria are not Stringent Enough
Finally, under s 11 the Bill leaves it to regulations to prescribe the age verification method, but identifies the key features that must be present in whatever method is prescribed. This is an appropriate approach to ensuring that the approved age verification method evolves with technology and society, which is best done through regulations. However, the list of key features is under-inclusive. Currently under s 11(2), the Governor in Council would consider if the method:
(a) is reliable;
(b) maintains user privacy and protects user personal information;
(c) collects and uses personal information solely for age-verification purposes, except to the extent required by law;
(d) destroys any personal information collected for age-verification purposes once the verification is completed; and
(e) generally complies with best practices in the fields of age verification and privacy protection.
I recommend two things are added to the list. First, cybersecurity is core to age verification and not currently addressed. It is implicit to the above provisions, but it should be spelled out as regulations might focus on data governance rather than the necessary security safeguards that underpin it. Second, freedom of expression is implicated in any age verification method and therefore must be central to consideration of what is adopted.
Recommendation: Add cybersecurity and freedom of expression as necessary considerations to the adoption of an age verification method.
In other words, when Laidlaw says that these provisions are not stringent enough, she means that there should be better privacy safeguards to be put in place. This is something I completely agree with. If we are to demand that highly sensitive personal information be added to massive databases, there should be laws in place that actively protects against this rather than a law that effectively says “pretty please” with no teeth. What’s more, Laidlaw says that freedom of expression should be considered within these provisions. While I found the title of this section to be weirdly worded, I ultimately agree with the recommendations offered here.
With all of these analysis being published these days, there appears to be a growing consensus. That consensus is that Bill S-210 is an unconstitutional mess. It’s poorly worded, is a direct attack on privacy and freedom of expression, and does little to “save the children” in the first place. The main criticisms of this legislation – whether they revolve around the censorship provisions, the lack of privacy safeguards, the overly broad definitions, or a host of other issues – it seems that there is a growing consensus that this is a fundamentally flawed bill. All of the above mentioned sources were all independent, yet generally came to the same or similar conclusions. If you don’t believe my analysis, check out the other sources. They all say the same thing. The only sources that agree with the approach of this legislation are generally widely discredited for not understanding what the legislation actually does and, instead, takes supporters comments at face value.
It’ll be interesting to see if any other analysis comes out of the woodwork at this point. Otherwise, I think the attention will end up focusing in on how government reacts to the consensus of this being a bad bill. Hopefully, it’s not a repeat of history where experts are just generally ignored and lawmakers push this bill through anyway like they did with the Online Streaming Act and the Online News Act, but the history doesn’t make that picture particularly bright.