On September 14, 2014, the Alberta Court of Appeal released its split 2-1 opinion in R. v. Fedossenko, 2014 ABCA 314, 316 C.C.C. (3d) 223 holding that the standard of “reasonable grounds to believe that an offence…has been or is suspected to have been committed”, contained in s. 487.012(3)(a) of the Criminal Code governing production orders, fulfills the minimum constitutional requirements for a reasonable seizure under s. 8 of the Charter of Rights. This test is certainly not one of probable cause which, in Canadian law, is most often phrased as reasonable and probable grounds. And it does not appear to be a test of reasonable suspicion as Parliament usually employs phrases such as “reasonable grounds to suspect” when using the lower standard. Or is it? And does it matter?

The majority, Justices Ellen I. Picard and Jack Watson, held at para. 7 that the lesser standard of reasonable grounds to suspect is constitutionally permissible based upon R. v. Chehil, [2013] 3 S.C.R. 220 where the Supreme Court of Canada concluded that the lower threshold respects the balance under s. 8 by permitting law enforcement to employ “legitimate but limited investigative techniques” such as the use of a sniffer dog. Justice B.K. O’Ferrall, in dissent, found at para. 48 that the reasonable suspicion standard violates the probable cause threshold in Hunter v. Southam, [1984] 2 S.C.R. 145 and that this narrow exception is confined to cases such as sniffer dogs that are minimally intrusive and narrowly targeted.

The respondent had been acquitted of impaired driving arising out of a car accident in which he was injured. He did not show signs of impairment and there was no perceptible odour of alcohol from him unless one leaned close to his mouth. He admitted to the police that he “had a few beers earlier”. His blood was taken at the hospital as part of his treatment. The police at the hospital told him he was under arrest and obtained his consent to have another blood sample taken for analysis. The arresting officer applied for a search warrant to seize the blood taken as part of the treatment. The application was denied by a justice of the peace who found that there were insufficient grounds to believe an offence had been committed. However, testing on the sample already in the possession of the police indicated a prohibited blood alcohol level. The arresting officer then made a second application seeking production of the hospital records. The second application was granted. The Crown conceded, on an evidence exclusion motion, that the police sample was obtained in breach of Fedossenko’s right to be secure against unreasonable search and seizure. The trial judge concluded that the police had failed to establish reasonable and probable grounds to believe Fedossenko had committed an offence and the order for production of the medical records should not have issued. The Crown’s appeal to the Court of Queen’s Bench was dismissed. The summary conviction appeal judge held that the standard for obtaining a search warrant and a production order was the same – reasonable grounds to believe that the Code requirements were met.

What is surprising about the majority opinion is their cavalier attitude about the respondent’s privacy interests. The blood sample contained core biographical information attracting a high privacy interest as contemplated by R. v. Plant, [1993] 3 S.C.R. 281. Yet, the majority did not undertake a Plant analysis. Nor was there any consideration of how that interest is enhanced by provincial legislation protecting hospital records. The bodily substance, containing DNA, was obtained incidental to the medical treatment. The defendant’s consent was given for that restricted purpose based on the reasonable expectation that the sample would be kept by the hospital (third party) in accordance with its legislatively imposed duties regarding confidentiality of patient records. The giving of the blood sample did not constitute waiver nor result in a reduced expectation of privacy. The only comments made by the majority, conclusory in nature, are found in three sentences in paras. 4 and 8: “This provision, s. 487.012(3) has two other requirements, namely (b) and (c). Altogether they provide for judicial authority to seize evidence which has already been obtained from the subject and is lawfully possessed by third parties. Nothing in this case had the effect of placing the accused’s medical interests in direct tension with his constitutional rights.”

The dissent by Justice O’Ferrall, however, analyzed and compared how the significance of privacy interests may vary depending upon the degree of intrusion irrespective of whether the object of the search or seizure is possessed by a third party. In my opinion, Justice O’Ferrall got it right where he stated the following, at paras. 51-52 and 61:

“…[T]he Supreme Court has not adopted the lower ‘reasonable suspicion’ standard for more intrusive searches, which I suggest the order to produce the respondent’s blood analysis constitutes. The principle that reasonable suspicion is not sufficient for intrusive searches is confirmed in R. v. Morelli, 2010 SCC 8 at para 91, [2010] 1 S.C.R. 253. The Court held that in the case of section 487 police searches of a suspect’s computer ‘as a matter of law, suspicion is no substitute for reasonable and probable grounds to believe either that the appellant committed the alleged offence’.

The Supreme Court has also held that a hospital patient undergoing treatment for injuries sustained in an automobile accident has a relatively big expectation of privacy; in other words, a reasonable expectation that the results of a blood analysis undertaken for medical reasons will not be shared with non-medical personnel without his or her consent: R. v. Pohoretsky, [1987] 1 S.C.R. 945; R. v. Dyment, [1988] 2 S.C.R. 417; R. v. Dersch, [1993] S.C.R. 768; R. v. Borden, [1994] 3 S.C.R. 145. In summary, the reasonable expectation of privacy differs significantly from searches conducted under circumstances when there is little or no expectation of privacy (e.g., airports and other transportation hubs) or when the search is minimally intrusive.

I agree with the majority when they say that production orders under section 487.012(1) may involve different privacy interests than search warrants under section 487(1) because search warrants permit police to enter into and search places, whereas production orders require only that a third party produce documents or data. But generalizations about the relative intrusiveness of search warrants versus production orders may not withstand scrutiny. As mentioned above, how intrusive a search warrant or production order might be will depend upon the facts. In this case, where the data sought to be produced is the analysis of blood taken from a person for medical purposes following a motor vehicle accident, the level of intrusiveness is high as this Court held in R. v. Taylor, 2013 ABCA 342, 561 A.R. 103 (which decision was upheld by the Supreme Court, [2014] S.C.J. No. 50). Indeed, the enforcement authorities recognized the intrusiveness of the search they were proposing to undertake when they initially applied for the search warrant which was refused. Where the document sought to be produced is the analysis of an accused’s blood extracted for medical purposes following an injury accident, the privacy interests engaged are more like those engaged in a body search. The intrusiveness of searches for data or documents in the hands of third parties may not always be as intrusive as a bodily search of the suspect, but they can be.”

The majority also did not consider the contrary conclusion reached by Justice Brian R. Burrows in R. v. Croft, 2013 ABQB, 304 C.C.C. (3d) 279 where, incidentally, the federal Crown conceded that suspicion should be read out of s. 487.012. Justice Burrows stated, at paras. 11-12:

“The accused first challenged s. 487.012 on the basis that the standard of belief that a crime has been committed is too low. The wording of s. 487.012(3)(a) on this point is, at best, clumsy. It says that the justice or judge must be satisfied, ‘that there are reasonable grounds to believe that (a) an offence…has been or is suspected to have been committed”. Arguably the reasonable belief standard established early in the section is lowered by the use shortly thereafter of the words ‘or is suspected to have been’. Arguably the wording creates a third and very murky level of standard between reasonable belief and reasonable suspicion – reasonable belief of the existence of a suspicion.

In argument, however, the Crown conceded that the confusion created by this clumsy wording should be resolved by interpreting s. 487.012(3)(a) as requiring that a reasonable belief standard be met notwithstanding the use of the word ‘suspected’ towards the end of the provision.”

If the majority intended to overrule or distinguish Croft, they should have said so. In the absence of that, the conclusion has to be that Fedossenko implicitly overruled Croft. I might add that Justice O’Ferrall did not consider Croft either. But, based on his conclusion in the privacy analysis, he probably didn’t have to. The majority’s failure to consider important authorities, however, was not limited to Croft. The majority also ignored the leading decision in R. v. Dyment, [1988] 2 S.C.R. 417, cited by Justice O’Ferrall, where the Supreme Court of Canada held that the use of an individual’s blood or other bodily substances, confided to others for medical purposes, for uses other than such purposes violates the personal autonomy of the individual. Such a seizure infringes upon all spheres of privacy – spatial, physical and informational. These zones of privacy were affirmed in R. v. Tessling, [2004] 3 S.C.R. 432 where Justice Ian Binnie described them with a slight variation in language, at paras. 21-23: personal privacy (bodily integrity), territorial privacy (privacy in the home) and informational privacy (information about ourselves and our activities). But that is not the end of the story.

The respondent filed an application for leave to appeal in the Supreme Court of Canada. While the application was pending, Parliament entered the arena and amended the section by the Protecting Canadians from Online Crime Act, S.C. 2014, c. 31. Section 20 of the Act replaced s. 487.012 of the Criminal Code, in its entirety, with the preservation demand regime. Although the preservation demand regime raises other questions, the reasonable suspicion terminology in Fedossenko was no longer contained in the Code and the question of national importance underpinning the leave application evaporated. The reasonable grounds standard in s. 487.014(2) is now the governing test. Leave to appeal was denied by the Supreme Court of Canada on March 26, 2015, without reasons. (I previously discussed the Croft decision on a related issue that a production order may not be used to obtain historic text messages. See: Police Must Get Wiretap Order to Seize Text Messages dated March 2, 2014)

The difficulty with Fedossenko is that the reasonable suspicion issue blurred what is really at stake in the digital age. In answer to my first question – it makes no difference whether the text of the former s. 487.012(3)(a) of the Code contained a reasonable suspicion standard or whether the phrase “reasonable grounds to believe that an offence…has been or is suspected to have been committed” was a variation of it. Both failed to comply with the minimum constitutional requirement of probable cause. In the Memorandum filed in support of the leave application in the Supreme Court of Canada, the issue that really mattered was put this way, at para. 32:

“The Court of Appeal’s interpretation of s. 487.012 has significant implications for vast amounts of Canadians’ personal information that is held by third parties, including many types of personal information that strike to our biographical core and reveal the most intimate details of our lifestyle and personal choices. A few examples of personal information frequently held by third parties include medical, psychological, and counselling records, financial records, emails and electronic appointment calendars that are stored on third-party computer servers, and web server and ISP logs that reveal the web sites we have visited. Especially in the age of cloud computing, Canadians place an enormous amount of personal information in the trust of third parties. For example, with the advent of services such as Facebook and Apple’s iCloud and the movement of our social lives online, our most personal forms of information – over which we have some of the highest expectations of privacy – are now commonly held by third parties, outside our direct control. Given these concerns, and in light of well-established s. 8 Charter doctrine, the Applicant respectfully submits that interpreting s. 487.012(3)(a) to require merely a reasonable suspicion that an offence has been committed would almost certainly offend s. 8 of the Charter.”

So, to return to my second question – yes, it does matter. In the era of Big Data, the cyber security interests of the individual are dependent upon the protection of personal information often held by third parties such as corporations, governments and institutions. As the Supreme Court of Canada held in Dyment, while Canadians may want, or be compelled to reveal personal information, there will be many circumstances where the individual will have a reasonable expectation that the information will remain “confidential and restricted to the purpose for which it is divulged”. In Canadian law, this is called the restricted purpose rule. The restricted purpose in Fedosssenko was the rendering of medical treatment. In the context of s. 8 of the Charter in the post-Snowden era, the protection of privacy is best achieved by the probable cause standard. Canadian Charter jurisprudence has never adopted the third party rule that has plagued Fourth Amendment doctrine. Let’s leave it that way.