CC By JP Low.
CJFE submitted the following response to a call for comments from the Office of the Privacy Commissioner of Canada (OPC) regarding its Draft Online Reputation Position Paper.
Although CJFE recognizes the importance for individuals to have a measure of control over their online reputation, the OPC’s Draft Position on Online Reputation fails to strike an appropriate balance between individuals’ right to privacy and the right to free expression.
The implementation of a “Right to be Forgotten” (RTBF) in Canadian law, modeled on the recently agreed European Union General Data Protection Regulation (GDPR), would lead to the creation of a complex - and mostly unchecked- system of private censorship. Such a system would also likely conflict with the Charter of Rights and Freedoms, section 2(b): the right to “freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication.” The “Right to Be Forgotten”, in the proposal being put forward by the OPC, poses a direct danger to press freedom and freedom of expression in Canada.
CJFE calls on the Canadian government to consult broadly with Canadians, civil society policy experts, and industry representatives before implementing any “Right to be Forgotten” in Canada. Any resulting policy must reject the European model of large-scale private censorship. The government instead should prioritize digital literacy education at an early age, as an alternative to the implementation of an overbroad internet censorship regime.
The OPC set out its original draft position in January 2018, which was broadly endorsed and built on in the February 2018 ETHI report. The concept of RTBF encompasses both the right to ask search engines to de-index web pages, as well as ‘source takedown’, or the right of erasure.
The OPC defines de-indexing as “the process by which a webpage, image or other online resource is removed from search engine results when an individual’s name is entered as the search term. Source takedown refers to the removal of the content from the internet."
Recent attempts to put this idea into action in Europe have had far-reaching, unintended consequences that undermine freedom of expression and the principles of the open internet. In the 3.5 years since the Google Spain (2014) case, which recognised the right to data de-indexing within the EU, there have been roughly 650,000 requests to review 2.4 million URLs. Of these 650,000 requests, 43% were granted. This means around 1 million URLs were de-indexed. That constitutes a significant amount of information that is now obscured from the general public within the EU. Worryingly for free speech advocates, much of what is now de-indexed is lawfully produced journalism. Google’s privacy counsel Peter Fleischer said at a conference in Toronto that about 20% of what has been requested for de-indexing is mainstream news articles.
Proponents of de-indexing argue that it is less intrusive than erasure, as it does not require absolute removal of the data. Yet information is only useful to the public in so far as it is available. De-indexing has been likened to removing a serial number from a library book, meaning that it is still available, but almost impossible to find. On the internet, there is no physical location and no material book to find, making it tantamount to erasure in practice.
Beyond these objections, the development and implementation of a right to erasure and de-indexing raise a number of other, perhaps insurmountable, problems.
Adjudication of disputes
ETHI argues that Article 17 of the GDPR should serve as the model for Canada’s implementation of the right of erasure. This would make data controllers (usually search engines) the initial adjudicators of requests. This will thrust search engines into the role of both court and publisher, deciding what is in the public interest and what ought to be displayed. This is particularly problematic for search engines as they do not own the content that the individual is asking to have removed. Editorial decisions must rest with publishers — not tech companies.
There are serious concerns with making search engines responsible for decisions to de-list or remove content. Corporations are ultimately accountable to their shareholders, whose primary concern is profitability, not values of accountability or transparency of decision making. Companies are incentivized to act in the easiest, most risk averse, and least costly manner. This will favour deference to delisting requests.
The sheer volume of requests, combined with the inherent vagueness of the ‘right to be forgotten’, means there is no simple way to determine whether or not an individual request is legitimate, or truly in the public interest. Search engines could understandably decide that it is safer to grant requests to remove content, thereby avoiding possible fines, than to spend the time needed to ensure the public interest and free speech rights are protected. In this way efficiency and liability concerns are prioritised above the public interest in protecting free expression and public access to information.
The ETHI report also discusses the possibility of a judicial branch or quasi-judicial branch making the decision. While there are benefits to this approach, pressing practical concerns remain. Since the Google Spain case in 2014 there have been roughly 650,000 requests made by individuals to have over 2.4 million URLs delisted. This number will rise further when the right of erasure is enshrined in law within the EU in May. One can question whether Canadian courts have the capacity to be able to effectively handle such an influx of cases. Additionally, judicial or quasi-judicial systems tend to progress slowly, casting doubt on Commissioner Therrien’s claim that the RTBF offers a faster and more efficient remedy for the removal of content from the web than under defamation laws. A system strained to capacity may not be able to assure sufficient quality of decision making to provide for protection of free expression interests.
What information is in the public interest
Determining whether or not information is in the public interest is an unavoidably subjective undertaking. Information which may not be relevant or of public interest may later prove to be so, say if a previously private figure decides to run for public office. Information thus de-indexed becomes nearly impossible to find, save for the very persistent or very skilled researcher, leaving the public less informed and more open to harm.
The right to be forgotten mandate can and has been used as a tool for wealthy and powerful individuals to clean Google searches of negative, truthful information linked to their names, restricting the public’s ability to fairly access legal and accurate information. This makes it harder for dissidents and journalists to reach the public, and leaves citizens less well-informed.
Removing material from search engine rankings impairs access to information and the right to know. Crucially, it risks altering the historic record, and similar to Strategic Lawsuits Against Public Participation, delisting requests could be used to chill advocacy, dissent or journalism on issues of critical public interest. In the day and age of #MeToo, it’s not difficult to see how powerful individuals might use the RTBF to scrub Google searches of negative, truthful information linked to their names.
The European Right to be Forgotten has already proven its inadequacy and should not be implemented in Canada. It poses unacceptable risks to Charter-protected freedom of expression rights.
The federal government should work closely with the Information Commissioner, the Privacy Commissioner, and representatives from civil society to develop means of protecting online reputation that do not pose such a serious risk to freedom of expression. Information and stories published online should not be made to disappear from the purview of inquisitive citizens.
CJFE supports proposals to improve general public education around online reputation and to provide “cyber hygiene” education for youth. There is a clear need to provide digital literacy education at an early age. Creating usable information for parents, young people and adults is likely to mitigate many of the problems people face online without undermining Canadians most fundamental freedom of expression rights.