We are very pleased to have the following guest post by Gerald Chan of Stockwoods in Toronto. Gerald Chan is a leading litigator in Canada, practicing in regulatory, criminal, and administrative law. He is co-author of Digital Evidence: A Practitioner’s Handbook by Emonds Publishing. His expertise in digital evidence makes the following commentary on the recent developments in text message privacy particularly valuable. We are very thankful for his contributions and encourage our readers to visit his biography to learn more about him, as well as purchase his text which, in our view, is essential reading to criminal and civil litigators alike.
Gerald Chan (Stockwoods LLP)
It’s no exaggeration to say that many people, especially young people, now prefer texting to talking on the phone. A phone conversation is time consuming and limits our ability to multi-task. Text messaging is far more convenient, allowing us to stretch a social conversation out over the course of a 16-hour day without taking too much time away from our daily responsibilities.
Of course, the law inevitably lags behind technology. As far back as 2008, 44 percent of Canadians said that SMS text messaging was the most common activity they performed on their cell phone aside from voice calls. Yet the Supreme Court of Canada would not address the issue of text message privacy until nearly a decade later. On December 8, 2017, the Court released two landmark judgments on this subject: R v Marakah and R v Jones. (Full disclosure: I was counsel to the British Columbia Civil Liberties Association (BCCLA), which intervened in both appeals.)
What the Court Held in Marakah
In Marakah, the question was whether a sender can retain a reasonable expectation of privacy in his/her text messages once they are sent to and received on another person’s device. This was relevant to the question of standing. Put another way, if A sends a text message to B, and the police discover that text message during a search of B’s phone, does A have standing to challenge the constitutionality of the search? Writing on behalf of a five-judge majority, McLachlin CJ said “yes.” (Moldaver J dissented, with Côté J joining him.)
McLachlin CJ reached this conclusion by applying the well-established “totality of the circumstances” test. She did so, however, in a manner that was sensitive to context. The context here was that of an “electronic conversation.” Therefore, the traditional factors under the totality of the circumstances test, which were developed to deal with claims of territorial privacy, had to be adapted. The “place of the search,” for instance, could be viewed as being the private electronic space that text messaging creates for the two parties to the conversation. And the factor of “control” should be understood as the freedom of individuals to choose how, when, and to whom they disclose their information. In the context of text messaging, individuals are choosing to disclose their private information to the recipient of the text message. The fact that doing so runs the risk of the recipient disclosing that information to a third party does not negate their reasonable expectation of privacy. To hold otherwise is to revive the long-discredited assumption of risk doctrine: the notion that we abandon all reasonable expectation of privacy whenever we run the risk that others might disseminate our private information. This doctrine would be destructive of privacy in the digital world where there is always the risk that information will be leaked beyond our intended audience.
Most importantly, McLachlin CJ considered the private nature of the information that individuals often communicate over text messaging. This appears to have been the driving factor in her analysis. She wrote:
A wife has no way of knowing that, when her husband appears to be catching up on emails, he is in fact conversing by text message with a paramour. A father does not know whom or what his daughter is texting at the dinner table. Electronic conversations can allow people to communicate details about their activities, their relationships, and even their identities that they would never reveal to the world at large, and to enjoy portable privacy in doing so.
In light of all this, McLachlin CJ concluded that senders of text messages can retain a reasonable expectation of privacy in the messages stored on the recipient’s device. Therefore, they may have standing to challenge unconstitutional searches of the recipient’s device where those searches reveal their text messages.
This decision will have significant implications for how police and regulatory authorities conduct their investigations. While the facts of Marakah concerned SMS text messages, McLachlin CJ made it clear that her reasoning would apply equally to other types of person-to-person communications tools, such as “Apple iMessage, Google Hangouts, and Blackberry Messenger.” In addition, while McLachlin CJ said at the beginning of her opinion that the exchange of electronic messages will not always attract a reasonable expectation of privacy, she clarified what she meant at the end of her opinion:
This is not to say, however, that every communication occurring through an electronic medium will attract a reasonable expectation of privacy and hence grant an accused standing to make arguments regarding s. 8 protection. This case does not concern, for example, messages posted on social media, conversations occurring in crowded Internet chat rooms, or comments posted on online message boards.
In other words, where electronic communications are being exchanged in the electronic equivalent of the public square, there may be no reasonable expectation of privacy. But it is safe to conclude that one-to-one text messages (whether exchanged through SMS messaging, Apple iMessaging, or some other electronic platform) will always attract a reasonable expectation of privacy in the post-Marakah world.
What the Court Held in Jones
The companion case of Jones raised a more technical question: where the police are obtaining historical text messages from the servers of a third party service provider such as Telus (as opposed to the recipient’s phone), what type of court order do they need? An ordinary production order (for which the standard is the default test for reasonable searches and seizures under section 8 of the Charter: reasonable and probable grounds) or the more rigorous Part VI authorization (otherwise known as a “wiretap” authorization, for which the police must also demonstrate “investigative necessity”)?
In the 2013 case of R v TELUS Communications Co, a plurality of the Supreme Court held that the police must obtain Part VI authorizations in order to acquire text messages from service providers on a prospective basis—that is, to obtain the production of future text messages. Should the standard be any different for historical text messages? Mr. Jones argued “no.” The BCCLA (for which I was counsel) supported this argument. Why should it matter whether the police seek judicial permission to acquire text messages the day before they come into existence, or the day after?
A majority of the Court disagreed with this argument. Writing on behalf of five justices, Côté J explained that the distinction between historical and future communications is a meaningful one under Part VI of the Criminal Code. The threat posed by the latter is unique because “when equipped with sophisticated surveillance technologies, the state may be tempted to embark on forward-looking, ‘fishing expedition[s] in the hope of uncovering evidence of crime.’” Therefore, a Part VI authorization is required for the latter but not the former.
Interestingly, Rowe J wrote a separate concurring opinion in which he agreed with Côté J’s interpretation of Part VI, but expressed concern that this enables the police to “in effect sidestep the requirements of Part VI by obtaining a production order immediately after the messages are sent.” He then explicitly stated that he was expressing “no settled view” on whether the ability of the police to obtain historical text messages with a production order (and not a Part VI authorization) was constitutional under section 8 of the Charter. I don’t imagine it will take long for enterprising defence counsel to see if they can get Rowe J (or at least a judge of a lower court) to answer this question.
Jones also raised a second, more general issue concerning the litigation of section 8 Charter claims: in seeking to establish a subjective reasonable expectation of privacy (which is one of the requirements for standing under section 8), does the defence have to lead evidence? Or can the defence rely on the Crown’s theory of the case? This has been a long-debated issue that often arises in drug possession cases. If the police find cocaine in a home, for instance, does A have to testify that he lives in the home in order to obtain standing to challenge the search of the home? Or can A simply rely on the Crown theory that he lives there, which is why he is being prosecuted in the first place?
Côté J’s majority opinion endorsed the latter approach. This is a very significant decision for the criminal defence bar, who will no longer have to risk calling their client in a Charter voir dire to establish standing. While evidence in a voir dire is not automatically admissible in the trial proper, an admission at the voir dire can restrict the permissible scope of defence evidence and submissions at trial. This is a gamble that defence counsel will no longer have to contemplate.
Next Steps: Reeves and Mills
Marakah and Jones are two of the biggest section 8 Charter cases in a long time. Nonetheless, they do not answer all of the pressing digital privacy questions related to text communications. In fact, less than one week after the Supreme Court released its judgments, the Court granted leave to appeal in two cases that will answer further questions about text message privacy: R v Reevesand R v Mills.
Reeves will require the Court to confront the thorny issue of third-party consent. In that case, the police searched a family computer that was co-owned by the accused and his spouse. The question was whether his spouse alone could consent to the search. The Ontario Court of Appeal said that she could. It would have been within the accused’s “reasonable expectations” that his spouse might have “a legitimate interest in consenting to police access to the shared space and property.” In order to reach this conclusion, the Court of Appeal had to distinguish R v Cole, in which the Supreme Court rejected the doctrine of third-party consent in Canada. The Court of Appeal distinguished Cole on the basis that the consenting party in that case (the school board employer) did not have an “equal and overlapping privacy interest” in the subject of the search (the school board-owned laptop) with the accused (a schoolteacher). The school board’s privacy interest was in the hardware (the physical device) rather than the software (the informational contents of the device).
Interestingly, however, the Crown made the same argument in Marakah to explain why the recipient of a text message should be able to consent to its search or seizure by the police: the sender and recipient of a text message have equal and overlapping privacy interests. But McLachlin CJ appeared to reject the argument. In para 50 of her opinion, she suggested that the police may have to obtain a warrant where the recipient of a text message alerts the police to its existence. Reeves will give the Supreme Court an opportunity to clarify this still-murky area of search and seizure law.
Mills takes the analysis one step further. What if the recipient of a text communication is an undercover police officer? Can that police officer capture the text communication as he is receiving it and, by doing so, seize the communication for investigative purposes? If this were a phone call, the undercover police officer would have to obtain prior judicial authorization before recording the call. The Supreme Court decided that in R v Duarte nearly thirty years ago. Should the law be any different for a text communication? The Newfoundland and Labrador Court of Appeal said “yes.” But it did so on the basis that “as the sender of such communications, Mr. Mills must have known that he lost control over any expectation of confidentiality” and “took a risk when he voluntarily communicated with someone he did not know.” This sort of risk analysis is similar to the line of reasoning that McLachlin CJ rejected in Marakah. Whether the Supreme Court will similarly reject the Newfoundland and Labrador Court of Appeal’s reasoning in Mills is something we will soon find out.
 Canadian Wireless Telecommunications Association, “2008 Wireless Attitudes Study Conducted on behalf of the Canadian Wireless Telecommunications Association” (12 September 2008) at 13, online: <http://www.cwta.ca/CWTASite/english/pdf/DecimaStudy_2008.pdf>.
 2017 SCC 59.
 2017 SCC 60.
 R v Marakah, supra note 2 at para 17.
 Ibid at para 28.
 Ibid at para 39.
 Ibid at para 40.
 Ibid at para 36.
 Ibid at para 18.
 Ibid at para 5.
 Ibid at para 55.
 Supra note 3.
 2013 SCC 16.
 R v Jones, supra note 3 at para 74.
 Ibid at para 85 (emphasis in original).
 Ibid at para 87.
 Ibid at para 19.
 Ibid at para 24.
 SCC File No 37676.
 SCC File No 37518.
 R v Reeves, 2017 ONCA 365 at para 62.
 2012 SCC 53.
 R v Marakah, supra note 2 at para 50.
 R v Mills, 2017 NLCA 12,  NJ No 55 (QL).
  1 SCR 30.
 R v Mills, supra note 24 at para 23.
 R v Marakah, supra note 2 at paras 45, 68.