Article body

Introduction

With more of our communication taking place online, courts are confronting a common scenario with greater frequency: an accused makes an admission in a text or an email to a complainant and she turns it over to police. It raises an old problem in a new guise. Do police conduct a seizure when they receive a piece of evidence from a civilian that engages a suspect’s privacy? Do they carry out a search when they proceed to review the item, even briefly? Must the police refuse to look, as an appeal court has recently asked, and first obtain some form of authorization?[1] If so, why? What measure of protection would calling this a search under section 8 of the Charter afford Canadians?[2]

Remarkably, the Supreme Court of Canada has not yet weighed in on the core questions underlying this scenario: whether police carry out a seizure when they receive evidence from a civilian third party over which a suspect retains a privacy interest against the state, whether a cursory analysis constitutes a search, and whether police require a warrant to do so. The questions have provoked widely diverging answers among trial and appellate courts and among commentators on privacy law.[3]

In offering differing responses to third party disclosure of evidence, courts and commentators alike have lost sight of what might be called the Supreme Court of Canada’s principled approach to deciding what constitutes a search or seizure and when it will be reasonable. The Court has applied this approach consistently, from its earliest cases on police receipt of evidence in Dyment and Colarusso, through a series of important cases on a reasonable expectation of privacy, including Buhay, Cole, and Spencer, and more recently when dealing specifically with digital evidence in Marakah and Reeves.[4] The thrust of this approach is to focus not on how police came into possession of the item — who gave what to whom and when — but to ask first whether the accused retained a reasonable privacy interest in the item against the state, and if so, whether police had authority to search or seize it. Time and again, taking this approach, the Court does not factor into its assessment of whether a search took place, or what would render the search reasonable (a warrant, consent, etc.), the status, identity, or intentions of the party providing the evidence to police. Lower courts and commentators, however, are beginning to do just this in the case of texts, emails, and screenshots: treat the matter differently on the basis that a civilian has brought police the item, suggesting that police receipt on voluntary disclosure is not a search or that something less than a warrant would make it reasonable.

Courts and commentators fail to follow the Supreme Court’s principled approach due in part to the fact that none of the Court’s cases on police receipt of evidence from third parties is directly on point. They deal with state actors transferring evidence to police, or police asking for or directly seizing evidence from a civilian third party. The closest the Supreme Court has come to addressing whether a third party giving police evidence engages section 8 is in Marakah and Reeves.[5] In Reeves, the Court held that a spouse could not consent to police seizure of a shared computer from the couple’s home, on the broader principle that “[w]e are not required to accept that our friends and family can unilaterally authorize police to take things that we share.”[6] This would presumably apply to a text exchange, an item in which the complainant has a shared privacy interest with an accused. In Marakah, the majority held that the accused retained a privacy interest in text messages found on a phone police seized from the recipient—despite the accused no longer having control of the messages.[7] In obiter, McLachlin CJ addressed the case of a complainant turning over a text exchange to police voluntarily. The accused might still retain a privacy interest in such messages, she suggested, even ones alleged to contain a threat against the complainant, and if so, police would need a warrant to read them; i.e., that police should indeed ‘just look away.’[8] What she did not explain is why. Her view is entirely consistent with the Court’s principled approach, and the missing rationale for applying it here can be found in those cases.

Without a Supreme Court decision directly on point, lower courts, losing sight of the Supreme Court’s principled approach to police receipt of evidence, have diverged in three directions. In some cases, the court recognizes a claimant’s privacy interest in the item but holds that a third party’s disclosure to police meant that police had done nothing on their part amounting to a taking or interfering with the claimant’s interest.[9] In some cases, the recipient’s consent authorizes police action (contrary to holdings on third party consent in Cole and Reeves).[10] A third body of cases is consistent with the Supreme Court’s approach in asking, first, whether the accused retained a reasonable expectation of privacy in the messages, and where they are found to be private, holding that reading the messages without a warrant was unreasonable.

Commentators are also in dispute over how to approach police receipt of digital evidence from third parties. Steven Penney suggests that the ruling in Cole and Reeves barring one party consent over the seizure and search of a shared computer should not apply to cases involving a chat participant turning over evidence to police and that one party consent should suffice.[11] Simon Stern proposes a reasonable suspicion standard authorizing a limited search.[12] Hamish Stewart contends that whether section 8 is engaged in a third party disclosure may depend on the content of the message.[13] Each of these positions departs from the Supreme Court’s principled approach in treating civilian disclosure as qualitatively different from other forms of search or seizure, calling for different standards. Yet none of these positions offers a compelling justification for departing from the Court’s approach in this context.

This article aims to foreground the Court’s principled approach, explain its general scope and, most crucially, articulate its rationale — its implied response to the pressing question in these cases of why police should ‘just look away’ when a third party brings them the printout of a chat, an email, or a photo, alleging it to contain evidence of a crime. The article also argues that McLachlin CJ’s dicta in Marakah and the Court’s reasoning in Reeves should form the basis of the Court’s approach to a future case directly on point and the approach of lower courts dealing with these facts. As McLachlin CJ observed in Marakah, a third party’s report of the content of a device or a private chat will “typically permit the police to obtain a warrant.”[14] Why, then, insist on the officer taking that step? What extra measure of protection would this provide to potentially innocent suspects? Put another way, what mischief on the part of the state would this avoid? Drawing on the Supreme Court’s earlier cases, I argue that the rationale here is that even if in most instances, police will easily obtain a warrant or have probable grounds for a search based on a civilian report, the requirement to establish reasonable grounds holds the state to a standard that helps to avoid casual or unnecessary intrusions into spaces or things that engage a high privacy interest. This affords all of us important assurance that the state will not be reviewing our texts, emails, or photos, on the basis of a civilian report about possible illicit content in them, unless the report is credible, the officer’s belief in the probability of an offence having occurred is reasonable, and an independent third party confirms this. This has formed the standard for a reasonable search for over two centuries and, as the Supreme Court has affirmed in several closely analogous cases, nothing in the third-party disclosure scenario justifies departing from it.[15]

The article proceeds in two parts. Part I sets out the Supreme Court’s principled approach by briefly canvassing decisions in a series of cases from Dyment to Reeves, to highlight the continuity and coherence of the Court’s approach and to distinguish its factual underpinnings. Part II considers the emerging conflict of opinion among courts and commentators, beginning with McLachlin CJ’s comments in Marakah on civilian disclosure of texts and recent trial and appellate decisions diverging from the principled approach. It then canvasses and critiques diverging views in recent commentary, before addressing unresolved questions about civilian disclosure to police.

The issues raised here are pressing. Civilian disclosures of text messages and screenshots are appearing in the case law more frequently. A diversity of approaches is emerging with no principled framework to unite them. Few areas involving section 8 of the Charter are more contested at present and, given how important digital communication is to us all, few issues in constitutional law have as wide a reach in terms of their potential impact.

I. The Supreme Court’s Principled Approach

The Supreme Court’s approach to whether police receipt of evidence constitutes a search or seizure under section 8 unfolds within a context of other decisions that define what actions on the part of police amount to a search or seizure. These are important holdings for understanding where and why the Court has drawn boundaries around the protections in section 8. After noting these holdings briefly, I canvas the Supreme Court’s principled approach by considering the cases in two groups into which they fall: those involving transfers of evidence to police from other state actors and those involving third party invitations to seize the evidence.[16]

A. When Section 8 is Engaged

Since the Charter applies only to state actors, the person conducting a search or seizure must be a state agent to engage section 8.[17] In Evans, Sopinka J held that state actors conduct a ‘search’ for the purposes of section 8 where “a person’s reasonable expectations of privacy are somehow diminished by an investigatory technique” or where “state examinations constitute an intrusion upon some reasonable privacy interest.”[18] In Dyment, La Forest J held that “the essence of a seizure under s. 8 is the taking of a thing from a person by a public authority without that person’s consent.”[19]

Police conduct a search when they inspect something;[20] when they carry out a wiretap or surreptitiously record a conversation;[21] and when they ask a third party for a claimant’s private information.[22] Police conduct a seizure when they exercise regulatory powers to demand documents,[23] and, in the course of a regulatory inspection, when they make copies of documents.[24] The connecting thread in each of these cases is an action on the part of a state actor that involves an interference with a privacy interest the accused retains against the state.

By contrast, the Court has made various findings on what does not constitute a search or seizure, and here too the connecting thread is a lack of interference with a privacy interest. Police did not conduct a search when they merely conversed with (rather than also recording a conversation with) an informer about a target;[25] when they obtained electricity records that were already public;[26] or when they demanded a driver’s license and insurance at roadside, given a lack of privacy in the information.[27] Similarly, police did not carry out a seizure when they collected garbage left at the curb, because it was abandoned.[28] A requirement in a collective bargaining agreement (of a federally regulated workplace) for the employer to provide the union periodically the names and addresses of members of a bargaining unit was not a seizure since members could reasonably expect the information would be used in this way.[29]

As the cases in the next section illustrate, when deciding whether a transfer of evidence to police or an acquisition amounted to a search or seizure, the Supreme Court focuses primarily on the question of privacy. The Court’s readiness to draw the inference that asking, glancing, taking, or receiving can amount to a search or seizure suggests that the standard is low—an interference with a privacy interest to any degree suffices to meet the test. In ways to be seen, this forms an important part of the Court’s principled approach and becomes especially contentious in the case law and commentary on civilian disclosure.

B. State Actors Giving Police an Item

The Supreme Court laid the foundation for its approach to police receipt of evidence from a third party in Dyment and Colarusso.[30] It then applied the approach in a series of cases (considered below), but this section canvasses Dyment, Colarusso, and Cole to show that a state actor coming into possession of evidence for one purpose can give rise to a search or seizure under section 8 by turning it over to police without being asked. The turnover in these cases is relevant to civilian disclosure to police in three ways. The Court in each case holds that the transfer of an item to police amounted to a search or seizure not because a state agent initially obtained the evidence or gave the item to police, but rather, because the accused retained a privacy interest against law enforcement when the officer received it. The cases also hold that police conduct a seizure when police merely receive the item or a search when they proceed to analyze or make investigative use of it—despite knowing or suspecting what it contains. And finally, the Court holds that the search or seizure required a warrant on probable grounds to be reasonable. The standard for a reasonable search turns not on how police obtained the item but on what they obtained.

In Dyment, a doctor took a vial of blood from an accused for medical purposes, after a car accident and without his knowledge. Discovering the accused had been drinking, the doctor decided to turn over the sample to police.[31] In concurring opinions, the majority held that the officer’s receipt of the samples constituted a ‘seizure,’ for the reasons La Forest J set out. Whether the officer’s receipt of the sample amounted to a seizure did not turn, for La Forest J, on whether the doctor volunteered or the officer asked for the sample. Evidence of the conversation between the two was unclear.[32] The matter turned on whether a privacy interest persisted. “If I were to draw the line between a seizure and a mere finding of evidence,” La Forest J wrote, “I would draw it logically and purposefully at the point at which it can reasonably be said that the individual had ceased to have a privacy interest in the subject‑matter allegedly seized.”[33]

The officer’s receipt of the sample became a seizure in Dyment not because the doctor chose to make a different use of it, but because the accused retained a privacy interest in the sample when the officer received it.[34] A further element assumed in La Forest J’s discussion but not made overt was the officer’s intention to use the sample for an investigative purpose.[35] Given the high privacy interest in the body, a seizure without a warrant was unreasonable.[36] McIntyre J, in dissent, declined to construe the officer’s receipt of the sample as a seizure, employing an approach that has reappeared in recent case law (canvassed in Part II, below). Any breach of privacy here was due to the doctor and impliedly—McIntyre J is silent on this point—the privacy interest ended there.[37] The officer was under a duty to make use of ‘real evidence’ and there was nothing improper in his doing so.[38]

In Colarusso, police took the accused to hospital following a motor vehicle accident involving a fatality in which alcohol was suspected.[39] The accused consented to staff taking a blood sample and, with an officer’s assistance, a urine sample “for medical purposes.”[40] Pursuant to powers in the Coroners Act, the coroner obtained the samples from a lab technician and turned them over to a police officer at the hospital for storage at a forensic lab for later analysis.[41] The Crown called a forensic toxicologist who had analyzed the samples at the lab at the coroner’s request, to assist in the latter’s investigation under the Act. The toxicologist’s blood-alcohol readings from the samples formed a basis for the accused’s conviction. In a 5-4 decision, a majority of the Supreme Court affirmed the trial judge’s finding that when the officer assisted in obtaining the urine sample, he was not conducting a seizure, but merely assisting hospital staff. However, after the coroner had arrived, because the accused retained a privacy interest in the sample, what followed amounted to an unreasonable seizure in two ways. The officer carried out a seizure when he received the samples from the coroner for storage at a forensic lab because they were then available for a prosecutorial purpose, the officer intended “from the outset” to use them for that purpose, and did so without a warrant.[42] Alternatively, the coroner’s seizure from hospital staff became unreasonable by making the samples available to police and failing to confine his use of them to the (non-criminal) purposes set out in the Coroners Act.[43]

The minority in Colarusso held that the coroner’s initial seizure from hospital staff was the only authority the Crown needed to make lawful use of the sample for the purpose of prosecution. Police delivering the samples to the lab did not amount to a seizure, since they were “essentially acting as a courier.”[44] There was no taking by police. The coroner’s involvement of police as a courier, the officer’s storage of the samples at the lab, and the toxicologist’s analysis were all authorized under the Coroners Act and fell within the scope of the coroner’s inquiry. The fact that the inquiry happened to produce evidence relevant to the criminal case is unfortunate for the accused but did not make any part of it a prosecutorial seizure.[45] Dissenting judges saw “no difference, as far as the police actions are concerned, between this case and one in which the Crown subpoenas a hospital lab technician to give evidence that came into existence during bona fide medical procedures and is relevant to a criminal proceeding.”[46]

The dispute between the majority and minority in Colarusso is framed as one turning on the limits of the accused’s consent. It might be framed another way: the accused’s consent was not fully informed. Had he known that taking a sample ‘for medical purposes’ might include some use by the coroner, he might not have consented. The larger point, however, is that both the majority and minority agree on the central issue: was the blood sample private? Did the accused’s initial consent permit the state to do what it did here? The two sides disagree on the answer, but they agree that whether there was a seizure under section 8 turns not on who gave what to whom, but on the validity and scope of the initial consent—which is another way of asking: did he retain some measure of privacy in the sample?[47]

In Cole, a technician in a high school discovered nude pictures of a student on a teacher’s board-issued laptop and reported them to the principal, who gave the laptop to police.[48] Without obtaining a warrant, police examined the content of the device and made a mirror image of its hard drive. Cole was charged with possession of child pornography. The Supreme Court held that in taking custody of the laptop for more than a brief period (to obtain a warrant) and examining its contents without a warrant, police had carried out an unreasonable search and seizure.[49] In the Court’s view, Cole retained a privacy interest in the content of the laptop given to police due in part to the understanding that he could make some personal use of it.[50] School officials may have been authorized to seize the device under a statutory duty to “maintain a safe school environment”, but this did not provide police authority to seize or search for a criminal investigative purpose.[51]

Relevant to the present discussion is how Fish J describes the handover and initial conduct of police upon receipt of the items. He makes clear that school officials were “entitled to inform” police of the illicit content on the device and police might have used this information to obtain a warrant.[52] But “receipt of the computer from the school board did not afford the police warrantless access to the personal information contained within it. This information remained subject, at all relevant times, to Mr. Cole’s reasonable and subsisting expectation of privacy.”[53] This raised the question of whether merely taking custody of the laptop constituted a seizure under section 8. Focusing on the receipt itself, Fish J writes:

The police may well have been authorized to take physical control of the laptop and CD temporarily, and for the limited purpose of safeguarding potential evidence of a crime until a search warrant could be obtained. However, that is not what occurred here. Quite the contrary: The police seized the laptop and CD in order to search their contents for evidence of a crime without the consent of Mr. Cole, and without prior judicial authorization.[54] [Emphasis in the original]

The need to posit temporary authority to hold the computer “until a search warrant could be obtained” implies that merely receiving the device constituted a seizure under section 8. This was premised on the finding that Cole had a “reasonable and subsisting expectation of privacy”, along with the officer’s investigative purpose in taking custody of the item. It did not turn on the identity of the party bringing police the item.

The Crown had also argued that police could rely on the school board’s consent to search the laptop, since the board owned it. The “underlying premise of this submission” Fish J held was that “a third party may waive another person’s privacy interest — thereby disengaging that person’s guarantee under s. 8 of the Charter.”[55] Rejecting this claim, the Court drew a bright line: a third-party may not waive another person’s privacy interest.[56] The rationale is directly relevant to civilian disclosure of digital evidence. As Fish J notes, the American version of the third party consent doctrine is premised on a risk that a party to whom a person makes a disclosure may turn it over to police.[57] The Supreme Court rejected the risk analysis in Duarte and Wong.[58] Moreover, Fish J held, third party consent fails to satisfy Canadian law on valid consent, which requires that it be “both voluntary and informed.”[59] Third party consent would entail police interference with a person’s privacy “on the basis of a consent that is not voluntarily given by the rights holder, and not necessarily based on sufficient information in his or her hands to make a meaningful choice.”[60]

C. Police Asking for or Taking an Item from a Third Party

The Supreme Court has also applied its principled approach in a series of cases where police ask a third party for information or a civilian reports a crime and invites police to seize the evidence. As it had done in Dyment, Colarusso, and Cole, the Court finds a search or seizure and decides whether it was reasonable not by focusing on details surrounding the taking or receiving of the item but, rather, on the principled basis of asking, first, whether police actions involved an interference with a persisting privacy interest and, if so, whether police had authority.

Buhay offers the clearest example of the Court’s principled approach, by drawing out its logic most explicitly. The accused rented a locker in a bus depot and security guards detected an odor of marijuana emanating from it.[61] An agent for Greyhound opened the locker and guards found a duffle bag inside containing marijuana. The guards placed it back, locked the locker, and called police. They directed police to the locker; officers smelled marijuana emanating from it; a Greyhound agent opened the locker, and an officer seized the bag. Arbour J, for a unanimous Court, affirmed the findings of the courts below that in their initial search, guards were not state agents, having acted independently from police.[62] But when a guard opened the locker and the officer took the bag, police conducted a seizure under section 8. Encapsulating holding, Arbour J wrote:

In this case, it cannot reasonably be said that the appellant had ceased to have a privacy interest in the contents of his locker. The subsequent conduct of the police should be considered a seizure within the meaning of s. 8. I see no basis for holding that a person’s reasonable expectation of privacy as to the contents of a rented and locked bus depot locker is destroyed merely because a private individual (such as a security guard) invades that privacy by investigating the contents of the locker. The intervention of the security guards does not relieve the police from the Hunter requirement of prior judicial authorization before seizing contraband uncovered by security guards. To conclude otherwise would amount to a “circumvention of the warrant requirement”.[63]

Despite a third-party directly leading police to the locker and police knowing its contents, the accused had retained a privacy interest against the state. Once again, the assessment turned not on the identity of the actors, their knowledge, or the choreography of their encounter, but on the question of a subsisting privacy interest.

In Spencer, the Court held that when police asked a third party for information pertaining to the accused, police carried out a search because the accused had a privacy interest in the information. Police had asked Shaw for the subscriber information attached to an internet protocol address implicated in a child pornography offence.[64] The Court held that a person retains a privacy interest in this information against the police, despite the allowance for voluntary disclosure of it in the Personal Information Protection and Electronic Documents Act.[65] Cromwell J, for the Court, framed the entire decision around the assertion that whether Spencer’s rights under section 8 were engaged depended on whether he had a privacy expectation in the information at issue. “If he did,” Cromwell J asserted without elaboration, “then obtaining that information was a search.”[66] Since police had no authority for requesting the private information at issue, their request, receipt, and use of it violated section 8. The Court did not decide what would authorize a reasonable search here, aside from “exigent circumstances or a reasonable law”.[67] Yet the need for authority, where a service provider volunteers the information, demonstrates yet again that whether police conduct a search and when it would be reasonable does not turn on the fact of voluntarily disclosure.

In Marakah, police investigating firearms offences obtained text messages authored by the accused when they searched the recipient’s phone without authority.[68] The majority of the Court held that a sender can, in some cases, retain a reasonable privacy interest in messages they send despite not having control over them in the recipient’s hands. The decision turned on two core ideas: the high degree of privacy accorded to text messages,[69] and the Court’s rejection of the risk analysis in Duarte.[70] McLachlin CJ held that “[t]o accept the risk that a co-conversationalist could disclose an electronic conversation is not to accept the risk of a different order that the state will intrude upon an electronic conversation absent such disclosure.”[71] A sender does not have full control over a message they have sent, but they do exercise “meaningful control over the information they send by text message by making choices about how, when, and to whom they disclose the information.”[72] For Moldaver and Côté JJ in dissent, a sender’s “total absence of control” over a message in a recipient’s hands renders any expectation of privacy on the part of a sender unreasonable.[73]

For both the majority and the dissent in Marakah, whether police carried out a search or seizure against the accused in obtaining his texts from a recipient’s phone turned—as it has in all of the Court’s earlier cases on police acquisition of evidence—on whether he retained a privacy interest in the item at issue. Finding that he did retain an interest, the majority assumed, without discussion, that nothing about the acquisition of the texts from the recipient’s phone called for a lower standard for searching it. They held it to be unreasonable without a warrant or other authority.[74]

The Supreme Court took a similar approach of looking first at the question of privacy in Reeves, a case involving police taking an item with a civilian third-party’s consent.[75] Reeves’ spouse told police she found child pornography on a home computer she shared with her spouse. An officer came to the home and, with her consent, took the computer and held it for four months without a warrant and without reporting the seizure (as is required under section 489.1 of the Criminal Code). All members of the Court agreed that the spouse’s consent did not serve to waive Reeves’ privacy interest in the computer against law enforcement, and that police removal of it constituted an unreasonable seizure.[76] The holding turns on the majority’s affirmation of three propositions: from Marakah, the notion that shared control over an item is not fatal to the finding that Reeves retained a privacy interest;[77] the point from Cole that a co-possessor of a privacy interest cannot waive the accused’s privacy interest, because the risk of third party disclosure does not vitiate the need for first party consent;[78] and that the heightened privacy interest that computers attract calls for heightened protections—either a warrant or valid consent from each privacy holder.[79] The majority’s reasoning on the second of these propositions speaks directly to the problem of third party disclosure of digital evidence to police.

The Crown sought to distinguish Cole by arguing that “Reeves’ spouse had an equal and overlapping privacy interest in the computer.”[80] Whereas in Cole, the school board owned the computer but Cole was its only user, here there were two users. Karakatsanis J, on behalf of the majority in Reeves, held “I cannot accept that, by choosing to share our computers with friends and family, we are required to give up our Charter protection from state interference in our private lives. We are not required to accept that our friends and family can unilaterally authorize police to take things that we share.”[81] The Crown raised the concern that this reasoning would “prevent victims of crime who have received threatening or harassing text messages from showing them to the police.”[82] Karakatsanis J declined to address the issue, noting it was not the question before the Court.[83] Yet, given the holding in this case that police need each party’s consent to seize a shared computer, it would make sense to take a different approach to texts only if they entailed a different kind of shared interest. A holding to this effect, however, would run contrary to the Court’s approach to text messages in Marakah as capable of retaining privacy without control.

Reeves is also instructive on a further point: whether mere receipt of digital evidence would be a seizure. Karakatsanis J suggests that it would be a seizure by virtue of its consequences:

...while the privacy interests engaged by a seizure may be different from those engaged by a search, Reeves’ informational privacy interests in the computer data were still implicated by the seizure of the computer. When police seize a computer, they not only deprive individuals of control over intimate data in which they have a reasonable expectation of privacy, they also ensure that such data remains preserved and thus subject to potential future state inspection.[84]

If the mere taking of a computer is held to be a seizure in this passage by depriving a person of a different order of control entailed in sharing with a friend (state preservation, future inspection)—the same should apply to all forms of digital evidence, including texts or emails.

This survey of Supreme Court holdings was meant to show the consistency and coherence of the Court’s approach to deciding whether a search or seizure has taken place under section 8 and whether it is reasonable. Amidst a variety of circumstances touching on police intentions, degrees of knowledge, and particulars around the physical handling of the evidence, the Court has continuously applied a principled approach, asking not ‘who gave what to whom?’ but rather, ‘did the accused retain a reasonable privacy interest in the item police acquired?’ In every case where the answer is yes, the Court holds there was a search or seizure. The Court also assesses the appropriate standard for whether it was reasonable (i.e., did it require a warrant, only probable grounds, first party consent, etc.) strictly in relation to the ideas of (a) a state intrusion into (b) a reasonable privacy interest. In none of these cases does the third party’s identity or intentions matter to whether police engaged section 8 or justify a lesser standard for a search than would otherwise apply. So long as in asking for, taking, or receiving the item, police had an investigative purpose (Dyment, Colarusso, Cole) or took investigative action (Buhay, Marakah, Reeves) and a privacy interest persisted, the search or seizure required a warrant on probable grounds.[85]

II. Conflicting Views on Third Party Disclosure of Digital Evidence

In the cases above, a state actor gives police evidence or police request it from a third party. The Court has yet to decide a case where a civilian volunteers an item to police, leaving open two questions. Does an officer conduct a seizure merely by receiving the item or a search by perusing it, even briefly?

The closest the Supreme Court has come to addressing these questions is in obiter in Marakah.[86] The obiter is directly on point in dealing with digital evidence—civilian disclosure of text messages. In ways to be seen in what follows, McLachlin CJ’s position is consistent with the Court’s principled approach to search and seizure, but it runs contrary to lower court holdings and recent commentary. I look first at the obiter in Marakah and draw on the Court’s earlier holdings to trace out the rationale supporting it, before contrasting it with conflicting case law and commentary.

A. McLachlin CJ’s Obiter in Marakah

In Marakah, as noted, the majority held that a recipient can retain a privacy interest in a text message police obtain from a recipient’s device.[87] For Moldaver and Côté JJ in dissent, this raised a host of practical concerns for police and courts.[88] The Chief Justice makes her comments on third party disclosure in response to these concerns.

Notably, Moldaver J, who authored the dissenting opinion, recognized that if the Court finds that a sender can retain a reasonable privacy interest in messages found on a recipient’s phone, police receipt of them on voluntary disclosure by the recipient would entail a search or seizure.[89] Given the Court’s dismissal of third party consent in Cole, “the police would never be able to obtain information about an accused through electronic communications offered by victims and witnesses on consent.”[90] As a consequence, the “overall number of instances where the police will be required to obtain judicial authorizations to gather evidence could increase dramatically.”[91] The increase in warrant applications would slow investigations, prolong trials, and “strain police and judicial resources” in a justice system “stressed to the breaking point.”[92] Most concerning still: “police may require a warrant even where a victim or his or her parents voluntarily provide police with threatening or offensive text messages”—a fact the majority conceded.[93]

McLachlin CJ offered two arguments in response, the first going to the heart of the Court’s framework for reasonable search in section 8:

Moldaver J. rejects any interpretation of s. 8 that would allow sexual predators or abusive partners to retain a reasonable expectation of privacy in text messages that they may send to their victims (para. 169). However, since Hunter, prior judicial authorization has been relied on to preserve our privacy rights under s. 8. In consequence, the fruits of a search cannot be used to justify an unreasonable privacy violation. To be meaningful, the s. 8 analysis must be content neutral.[94]

McLachlin CJ might also have added here that whether one has a reasonable expectation of privacy in something is a normative question and does not turn on whether the use one makes of a space or a thing is criminal in nature.[95] Her second argument is that the majority’s position might make police work more onerous, but there is “nothing in the record to suggest that the justice system cannot adapt to the challenges”, nor is it the case that “text messages sent by sexual predators to children or sent by abusive partners to their spouses will not be allowed into evidence.”[96] Where obtained in violation of section 8, messages might still be admitted under section 24(2) of the Charter.[97]

The Chief Justice then ventured an opinion on how the holding in Marakah would apply to a civilian who brings police evidence:

Assuming that s. 8 is engaged when police access text messages volunteered by a third party [...], a breach can be avoided if the police obtain a warrant prior to accessing the text messages. As stated in Cole, “[t]he school board was . . . legally entitled to inform the police of its discovery of contraband on the laptop” and “[t]his would doubtless have permitted the police to obtain a warrant to search the computer for the contraband” (para. 73). Similarly, victims of cyber abuse are legally entitled to inform the police, which will typically permit the police to obtain a warrant. The police officers will be aware that they should not look at the text messages in question prior to obtaining a warrant.[98] [emphasis added]

An officer would need a warrant to look at the messages because even looking would constitute a search. In keeping with the Court’s principled approach, receiving a copy of the message from a victim, rather than police asking for or taking it, does not change the analysis. It turns on whether the accused retained a privacy interest in the message. Nor does the situation call for a different standard. Put another way, for Moldaver and Côté JJ, what would make for a reasonable search turns on how police obtained the item (i.e., the fact of receiving the item from a civilian). For McLachlin CJ, what would make for a reasonable search turns on what police obtained.

The Chief Justice explained why, in her view, the dissent’s concerns do not pose a significant impediment to recognizing this as a search or seizure and insisting on a warrant. But she did not explain why police should require a warrant or other authority. In other words, insisting that police should obtain a warrant may be consistent with the view from Hunter onward that the “fruits of a search cannot be used to justify an unreasonable privacy violation” and the privacy analysis being content neutral. But she does not explain why, in the specific case of a police officer meeting with a civilian, the officer must avoid looking at a printed copy of a chat she had with a potential suspect without first obtaining a warrant (or having other authority). If, as McLachlin CJ pointed out, a civilian’s report will “typically permit the police to obtain a warrant”, why insist on the officer taking that step? What extra measure of protection does this provide to potentially innocent suspects? What mischief on the part of the state does it avoid?

The Court has confronted this issue in various cases where obtaining a warrant may seem like a formality. For example, in Duarte, the Court held that even where police have one party’s consent to conduct a wiretap (i.e., an informant), they still need to obtain a warrant.[99] In Spencer, police needed authority to carry out a search before asking Shaw for a user’s subscriber information, despite the fact that privacy law permits a service provider to make disclosures to law enforcement.[100] The Court in these case offers two main reasons for insisting on a warrant even if, in most instances, police will easily obtain one or have grounds for a search. The requirement to establish grounds holds the state to a certain standard that helps to avoid casual or unnecessary intrusions into our privacy. And the hurdle or hindrance is worth the trouble it imposes upon police because the privacy interest in the item or space they would search or seize is high and the hurdle makes unnecessary intrusions less likely.

The Court in Reeves applies this reasoning explicitly in relation to police seizure of a computer. As Karakatsanis J writes:

The unique and heightened privacy interests in personal computer data clearly warrant strong protection, such that specific, prior judicial authorization is presumptively required to seize a personal computer from a home. This presumptive rule fosters respect for the underlying purpose of s. 8 of the Charter by encouraging the police to seek lawful authority, more accurately accords with the expectations of privacy Canadians attach to their use of personal home computers and encourages more predictable policing.[101]

As noted above, the Court in Marakah accorded a similar high degree of privacy to text messages (“it is difficult to think of a type of conversation or communication that is capable of promising more privacy than text messaging”).[102] If both computers and texts engage a high degree of privacy, the Court’s insistence on a warrant or other authority for a search or seizure where police lack the accused’s consent is meant not simply to burden police, but to better protect privacy through more predictable policing. As the Court held in Hunter, predictable policing and protecting privacy are both supported in turn by clear rules and objective standards that help to avoid unreasonable searches before they occur.[103]

The unspoken rationale for applying these principles to the case of an officer about to scan even only a page-long print out of a text exchange a civilian has brought them is that it affords everyone an important measure of assurance. We need not be concerned that the state may be reading our texts or email on the basis of a civilian report about possible illicit content in them, unless the report is credible, the officer’s belief in the probability of an offence having occurred is reasonable, and an independent third party confirms this. Insisting upon these requirements is consistent with a wider societal expectation that our conversations, photos, or emails will not easily end up in the hands of the state—for preservation or investigation. Conversely, given the ease with which digital evidence can be shared, stored, or turned over, the failure to provide this safeguard imperils everyone’s sense of privacy, dignity, and autonomy.[104]

When would police not have reasonable grounds for a warrant, resulting in unnecessary intrusions on private messages or emails? In a few cases. One would be upon initial or early contact with a complainant, when the complainant has not yet finished conveying the information they have to convey about the allegations and the content of the document at issue. One can readily imagine a scenario in which a complainant provides an officer a document early in the course of a meeting to give a statement and the officer perusing it before obtaining all the facts. Another scenario would be when what a complainant tells police about the content of an exchange, photo, or email does not give rise to a reasonable probability that the document at issue will contain evidence of an offence (as is required under section 487 of the Criminal Code). For example, a person might allege that they have been assaulted but only vaguely describe the content of an exchange or email in a manner that leaves unclear whether the document is likely to contain evidence relevant to the assault. Further scenarios where police might lack grounds would include a complainant making an allegation that does not amount to an offence (i.e., the suspect did something offensive or obnoxious, but not illegal) or frailties in the complainant’s account cast into doubt the likelihood than offence has been committed and evidence of it is to be found in the document to be searched. Requiring the officer to hold off on looking at the evidence until gathering all the facts—and obtaining independent confirmation of their assessment—would avoid unnecessary intrusions into private emails or exchanges in each of these cases.

B. Conflicting Approaches Among Lower Courts

In the absence of a Supreme Court decision directly on point, uncertainty remains. Yet the scenario of a suspect making inculpatory admissions in an email or text to a complainant who turns it over to police is becoming all too common. Trial and appeal courts have arrived at different conclusions about whether police seized or searched an item on receipt or review of it, many of them by neglecting to apply the Supreme Court’s principled approach.

The reasoning in recent cases takes one of three forms. In some cases, the court recognizes a claimant’s privacy interest in the item but holds that a third-party’s voluntary disclosure of it to police (or a screenshot of it) meant that police had done nothing on their part amounting to a taking or interfering with the claimant’s interest.[105] In some cases, the recipient’s consent authorizes police action.[106] A third body of cases is consistent with the Supreme Court’s principled approach, in taking the primary question to be whether the accused retained a reasonable expectation of privacy in the messages at issue or what is depicted in a screenshot or found on a device.[107] A brief look at some of these cases shows how peculiarities in the evidence, or how police obtain it, can lead judges to overlook an underlying simplicity in the issues.

In King, the accused’s wife, suspecting him of infidelity, obtained his password, accessed his devices, and discovered what she believed was child pornography.[108] She took screenshots of some of the images on his devices and saved the screenshots to a USB drive. She described to police the images she saw and gave the USB drive to the officer. He in turn examined the screenshots on the disk and used the information to obtain a warrant to seize and search King’s devices. The trial judge held the accused had a privacy interest in the screenshots the officer reviewed on the USB drive since the subject matter of the search was not the drive or the screenshots themselves, but, in effect, the glimpse they afforded inside King’s devices. Looking at the screenshots, the officer could see “how [the accused] organized his desktop computer, including the backdrop or ‘wallpaper’, the placement and titles of folders, the titles of images and videos, and the presence of other non-illicit information.”[109] All of this revealed “personal information about Mr. King”, rendering the officer’s act of viewing the screenshots a form of search. Following Reeves, the court held that King’s wife could not consent to the search on his behalf and that it was unreasonable under section 8.[110]

The Alberta Court of Appeal treated the pictures the wife had taken on the USB drive as a further form of her describing or reporting the alleged offence.[111] Slater J held that “[t]he examination of the USB flash drive by the police was not an examination of any thing or place that belonged to or that was under the control of the appellant.”[112] The trial judge’s assessment was more plausible: i.e., the officer did more than merely look at a picture of a picture. He looked at the content of the picture within a picture, giving him a substantial view of the content of the accused’s computer. The difference this makes can be easily illustrated by considering screenshots given to police that depict not pornographic images but a private email, text exchange, or journal entry. Clearly in that case, an officer proceeding to read the texts would be doing more than ‘receiving a report.’

King has since been cited in a later appellate decision for the proposition that “where information is brought to the police by an independent third party acting on its own initiative, and where the police merely receive and review it, there has been no search or seizure by the state which engages s. 8.”[113] Trial courts have applied similar logic. In Bear-Knight, the court found the accused did not have a privacy expectation in the screenshot of a Facebook message the complainant brought to police.[114] Regardless of this, police receipt of the item from a third-party presenting it to them voluntarily meant that this could not be a search or seizure or, if it was, the party’s consent sufficed to authorize police to conduct it.[115] In Lowrey, a mother and her 14-year-old daughter brought police the print-out of a conversation the daughter had with the accused on Facebook.[116] The court held that despite the accused’s “marginal expectation of privacy” in it, the daughter’s consent was “sufficient to authorize the search [the officer had] undertaken.”[117] In Admurski, the court held: “[t]here was no seizure by the police because the text messages were voluntarily turned over to the police by a concerned parent.”[118] In Phagura, the complainant consented to police taking photos of a text exchange on her phone that she had with the accused.[119] The court found the accused lacked a reasonable privacy interest in the messages, but proceeded to address whether the complainant’s consent provided sufficient authority for the search.[120] Since she and the accused had “an overlapping interest” in the exchange, it was not reasonable for the accused “to think or expect that Ms. M.K. would not be able to consent to provide such information to the police.”[121] Her consent was “valid and sufficient for the police to obtain copies” of the exchange.[122]

There is certainly an appeal to the reasoning in these cases, arising from a sense of not wanting to needlessly impede civilians reporting a crime or police investigating it in good faith. An often-cited expression of this reasoning is found in a passage in Orlandis-Habsburgo.[123] In this case, an electricity company, Horizon, volunteered suspicious consumption data to police. The court held that police did not seize or search the data on receipt of it from Horizon because the company had a working arrangement with police from the outset, which rendered Horizon a state-agent. Horizon’s initial investigative steps constituted an unreasonable search.[124] In obiter, however, Doherty JA considered what would have happened if Horizon were independent. Company officials could describe the suspicious data to police, which police could then include in an affidavit for a warrant. Why, then, should things be different if an independent Horizon were to give police a copy of the data itself (as occurred here)?[125] Doherty J asserted:

I have considerable difficulty with the submission that s. 8 is engaged if the police look at information in which an accused has a legitimate privacy interest, even if that information is brought to the police by an independent third party acting on its own initiative. On that approach, s. 8 would be engaged if a “whistleblower” took confidential documents belonging to her employer to the police to demonstrate the employer’s criminal activity. Must the police refuse to look at the documents to avoid violating the employer’s s. 8 rights?[126]

It may indeed be formalistic to require a warrant to permit an officer to read a document when he or she already knows what it contains. Yet this is essentially what happened in Buhay, Cole, and Reeves, where police already knew what the locker, laptop, or home computer contained.[127] Despite this, in each case the Court held that a third party’s disclosure of the content did not negate the accused’s privacy in it against the state. The answer in these cases—along with McLachlin CJ’s obiter in Marakah—to Doherty JA’s question ‘must the police refuse to look?’ is simply: yes. The broader explanation for why (canvased in the previous section and found in a host of Supreme Court decisions from Hunter to Reeves) is, in short, the high privacy interest in digital evidence can best be protected and societal expectations best met by holding police to certain standards before searching.[128]

A third group of cases centers the analysis on a persisting privacy interest, consistent with the Supreme Court’s approach. A good example can be found in Morgan, where the accused was held not to have a privacy interest in messages he sent a woman he met only the night before and in texts he had sent to her on her mother’s phone.[129] In KA, the court held that messages sent by two alleged pimps and human traffickers to their alleged sex worker and slave were not messages in which the accused had an objectively reasonable privacy interest, given the exploitive and tenuous nature of the relationships involved.[130]

Applying similar reasoning, other courts have come to the opposite conclusion—the accused did retain a privacy interest in the evidence turned over, resulting in a finding that a warrantless search of the evidence was unreasonable. One example can be found in the trial decision in King, noted earlier, involving police examination of screenshots.[131] Another can be found in R v CM, where the court held that the accused retained a privacy interest in messages sent to his former stepdaughter, which she gave to police.[132] The finding was based in part on the accused having known the recipient for a long time and standing in “a quasi-parental role towards her.”[133] In Devic, a person responding to a Craigslist ad posted by a member of ‘Creep Catchers’ had a privacy interest despite not knowing their anonymous interlocutor.[134] In Rafferty, the court found a privacy interest in messages found on a phone a deceased son’s father brought to police, incriminating the accused who was believed to be a dealer.[135]

The survey in this section was meant to show that lower courts are divided in their approach to assessing police receipt of digital evidence. The reasoning in support of diverging approaches is not compelling. It offers no response to the holdings from Hunter to Reeves that see the need for a warrant or other authority (aside from one-party consent) as an effective means of protecting the high privacy interest in the items at issue. By contrast, lower courts taking a principled approach demonstrate the viability of treating police receipt of texts and photos as a form of search, requiring a warrant, and the range of outcomes for which this still allows.

C. Conflicting Approaches Among Commentators

Commentators are also divided on questions raised by civilian disclosure of digital evidence. The scholars considered here agree that police engage section 8 on receipt of evidence from a civilian. But each of them offers a position that diverges from the Court’s principled approach in suggesting that civilian disclosure renders the situation different from other searches and seizures police conduct. In their view, police are justified to search a text exchange or an email on one party consent, reasonable suspicion, or no grounds or consent at all. These views diverge from McLachlin CJ’s position in Marakah that police should need a warrant before conducting a search, and also from the Court’s decisions (surveyed in Part I) holding a warrant to be necessary to render a search reasonable in a variety of scenarios. As the survey in Part I showed, the Court consistently found a warrantless search or seizure unreasonable without factoring in how police obtained the item. I argue that none of the alternative views canvassed here offers a compelling argument for diverging from McLachlin CJ’s approach to text messages or from the Court’s principled approach more broadly.

Steven Penney weighed in on the debate after the Supreme Court had decided Marakah but before it had decided Reeves.[136] Professor Penney lauded the Court of Appeal’s approach in Reeves to shared privacy interests and its application of this approach to police entry into the Reeves’ home (with spousal consent).[137] Penney supports LaForme JA’s argument that a co-resident cannot reasonably expect their housemate to never invite an agent of the state into the home, since “the other might have a legitimate interest in consenting to entry by law enforcement into common spaces from time to time.”[138] The rule, in Penney’s view, should apply to text messages because “while it is sometimes difficult to decide whether residential spaces are shared or exclusive, received text communications are always shared between the parties.”[139] On this basis, Penney proposes a rule permitting third parties “including victims of and witnesses to criminal activity, to consent to searches of their devices for incriminating messages sent to them” on the basis that it would “further public safety and crime control without unduly diminishing the privacy of electronic communications.”[140]

Penney offers a cogent alternative to the approach suggested in Marakah and Reeves. Yet he provides no support for his assertion that permitting police to rely on the consent of the recipient of a text to read the messages would further public safety without unduly diminishing privacy in text messages generally. This is a value judgement—a reasonable but different view of where to strike the balance between law enforcement and individual privacy from the one McLachlin CJ offers in obiter in Marakah canvassed earlier. Penney shares the view of Moldaver and Côté JJ in Marakah that when it comes texts messages the balance is best struck in favour of one-party consent. The Chief Justice’s preference for a warrant in this case need not be taken to mean that nothing short of a warrant requirement would strike an appropriate balance of interests here. Yet, the majority’s position in Marakah and Reeves, holding that police need a warrant to search text messages or a computer suggests that police need something more than third party consent to render state interference reasonable, given the high degree of privacy in text messages and the goal of avoiding unnecessary intrusions before they occur.[141] Short of a warrant, the reasoning in these two cases implies, privacy in digital communications would indeed be unduly diminished.

Put another way, a rule permitting police to rely on one-party consent to a text-exchange would have the effect of permitting state intrusion into immensely private spaces on the whim of a disgruntled recipient. Penney might counter that obtaining reasonable grounds will often involve an officer hearing so much about the content of an exchange that it would hardly “differ substantially from police looking at the messages themselves”.[142] That may indeed be the case in some instances. But this is really a form of the ‘why bother getting a warrant?’ argument addressed above.

Simon Stern offers a subtle analysis of Marakah, informed by a background in media theory and search law on other forms of written communication.[143] He takes issue with Supreme Court’s holding in Marakah that if a sender retains a reasonable privacy interest in text messages, police need a warrant on probable grounds to seize them. He sees a better alternative in the rule crafted in R v Fearon.[144] The Court in Fearon held that under certain conditions, police may conduct a warrantless but limited search of a phone incident to arrest. In Stern’s view, if in this context the “showing of a heightened privacy interest as to a certain item makes no difference” and “any evidence [police] collect is admissible, whether it is a bus transfer or a personal diary”,[145] a police power to seize and search text messages volunteered to them on reasonable suspicion should be reasonable.[146]

Stern misreads Fearon. He correctly notes that the majority assumed a warrantless search would “serve important law enforcement objectives” including “identifying accomplices or locating and preserving evidence that might otherwise be lost or destroyed.”[147] But the key point for the majority was that police on arrest need to conduct a search for these purposes promptly—and their need to do so is pressing.[148] In the view of the dissent, it was not.[149] Nor did it seem pressing to the US Supreme Court in Riley v California, decided earlier that year.[150] For the dissent in Fearon and the USSC in Riley, police could wait to search for evidence, and concerns about nefarious uses of a phone following an arrest were no more than speculative and did not outweigh the high privacy interest in a device.[151] Stern entirely overlooks this debate and offers no analogue to the balancing involved in police receiving a text exchange from a recipient. What pressing need does law enforcement have in reading a text or email a civilian has brought them promptly, rather than waiting to obtain a warrant? McLachlin CJ’s treatment of the scenario in Marakah assumes no such need. Short of a text alleged to contain an imminent threat of serious harm, there is no case for urgency on the part of police.[152]

Hamish Stewart commented on McLachlin CJ’s obiter in Marakah in the context of an essay on a reasonable expectation of privacy as a right.[153] Stewart’s comments are prefaced by a consideration of Moldaver J’s observations on the consequences of the majority’s holding in Marakah. Finding a continuing privacy interest in a message by a sexual predator who threatens to kill a child leads, in Moldaver J’s view, to the absurd result of the predator’s section 8 rights being engaged when a parent turns over the messages to police.[154] Stewart agrees with Moldaver J that this result is “untenable,” but contends that Moldaver J is mistaken in thinking it flows from the majority’s holding. The majority held that the predator may have a reasonable expectation of privacy against the state but not against the victim. If the victim or her parents “voluntarily provide the text messages to the police”, Stewart contends, “section 8 of the Charter would not be engaged because, being private actors, they have no Charter duties toward the suspect, just as section 8 is not engaged where a complainant shows a threatening letter or email to the police.”[155] Stewart suggests, however, that police might need a warrant or other authority to “obtain … a complete record of the exchanges between the suspect and the complainant.”[156] He acknowledges McLachlin CJ’s differing view in obiter in Marakah, but assumes she was merely framing police receipt of threatening texts on civilian disclosure as a search in that passage “for the sake of argument.”[157] He offers the “tentative view” that police would not need a warrant to read threatening messages, since disclosure would not constitute a search.[158]

Stewart’s view on civilian disclosure of private messages runs contrary to the Supreme Court’s holdings on receipt and transfer. It also runs contrary to the Court’s holding on third party consent to search items over which parties share an interest, noted earlier. Stewart offers no argument for why police do not conduct a search when a parent shows an officer the threatening text itself but do conduct a search when the officer proceeds to read the entire exchange. An approach consistent with Supreme Court jurisprudence on privacy and receipt would be to suggest that a threatening text or email could still engage a privacy interest, but an imminent threat would provide police authority to carry out a warrantless search in exigent circumstances.[159] Alternatively, as McLachlin CJ suggests in Marakah, a court might find that police intruded on a privacy interest in reading a text containing an alleged threat, and doing so without authority violated section 8. But given the technical and limited nature of the breach, the evidence would be admitted under section 24(2).[160]

A final criticism to note here pertains to McLachlin CJ’s assumption that police could in most cases readily obtain a warrant to read a private exchange based on a complainant’s description of its contents. Michelle Biddulph calls this into question.[161] Under section 487(1) of the Criminal Code, an affiant seeking a warrant must establish probable grounds that an offence has been committed and evidence of it will be found in the place to be searched.[162] An affiant would need to be somewhat detailed, and an officer’s description of what the complainant said about the texts would be hearsay. In some cases, this may not suffice for probable grounds.[163] She thinks a warrant requirement places “needless and impossible burdens for the police... to know something that they already knew.”[164]

Biddulph raises important concerns. In many cases, police may face a real burden in obtaining a warrant. But the burden is not a needless one. Missing in Biddulph’s analysis is a consideration of the rationale for putting police to the trouble of obtaining a warrant before proceeding to search presumptively private messages. As canvassed earlier, imposing a standard affords a measure of assurance to everyone that the state will not intrude on private communications without certain steps being taken. This accords with a wider societal expectation against state interference in private spaces generally. Far from being pointless, we impose a burden on police in these cases precisely in order to make it more difficult for them to intrude.

D. Remaining Questions and How They Should be Resolved

The challenge for commentators and judges is to understand why a brief text exchange that a civilian hands over to police—even only a single page or two—is not different from a closed container, a locked cabin, or a book titled ‘secret journal’ that police might encounter in the course of an investigation. In the latter cases, it is easy to see how police proceeding to open or analyze the object would clearly constitute an intrusion of privacy. It is easy to overlook how the same can be said to happen the instant an officer begins to review, peruse, or analyze a short exchange, email, or photo. Courts and commentators alike fail to distinguish between the officer having knowledge of the content of an item and an officer reading or reviewing the item itself. They assume that because in some cases this may amount to a distinction without a difference, no distinction should be drawn in any case. Yet, as the majority held in Marakah, a text exchange can engage a high degree of privacy.[165] So might a photo or an email. The reasoning about texts in Marakah suggests that if a suspect retains a privacy interest in a communication against the state, police conduct a search if they review it—even if police obtain it from a third party who volunteers it, even if it is brief, and even if police have a good sense of what it contains.

However, in the absence of a Supreme Court decision directly on point, the Court has yet to decide two discrete issues raised here: do police conduct a seizure when they take custody of a text, email, or photo when a civilian volunteers it, and do they conduct a search if they proceed to review it, even in a cursory fashion?

Turning to the first of these questions, McLachlin CJ’s dicta in Marakah and the Court’s holding in Cole assume that, when a civilian has described the incriminating content of an item, police are authorized to take custody of it briefly, to obtain a warrant.[166] In Reeves, the officer testified to not believing he had reasonable grounds to seize the computer under the ‘plain view’ provisions of the Criminal Code,[167] and the Court held the officer could not rely on the spouse’s consent. A reading of the three cases suggests that an officer carries out a seizure under section 8 when they receive an item a civilian volunteers in which the accused retains a privacy interest, even a text, email, or photo in which the civilian has a shared interest. A possible source of authority for the seizure can be found in the plain view powers of the Criminal Code, which permits an officer in the execution of her duties (i.e., receiving a report) to seize an item they reasonably believe will afford evidence of an offence.[168] Another source would be the Code power to seize in exigent circumstances, though this would apply in a narrower set of circumstances.[169]

Turning to the second question, whether police conduct a search if they proceed to review a text exchange a recipient discloses—even briefly—should turn on the question of privacy and not consent. It should turn on whether the accused retained a reasonable privacy interest in the message against the state, and thus not on the identity of the party disclosing it, or the fact that a civilian with a shared interest in it voluntarily disclosed it rather than police having asked for or taken it. McLachlin CJ’s obiter in Marakah offers a blueprint for a future holding consistent with the Court’s earlier jurisprudence. If, applying the four-part test for a privacy interest in Marakah,[170] a suspect is held to retain a reasonable privacy interest in a text-exchange turned over by a complainant, in keeping with Cole and Reeves, the complainant’s consent should not serve as authority for a search.[171] Even a victim of ‘cyber abuse’ turning over a chat to police would not, in McLachlin CJ’s view, authorize police to read the texts without a warrant.[172]

Taking this approach, whether civilian disclosure of digital communications engage section 8 will likely be determined by deciding whether the accused retains an objectively reasonable expectation of privacy in the message. This should involve asking whether a reasonable person in the position of the accused might have accepted as reasonable the form of disclosure made in that case to the state if it were presented to them as a hypothetical. Lower courts have posed the question in precisely these terms. For example, can a male in an exploitive relationship with an adolescent female reasonably expect her not to reveal to parents or police communications that urge her to engage in criminal activity?[173] There have been ample criticisms of the Court’s decision in Marakah to not declare that all personal digital communication gives rise to a reasonable expectation of privacy (in contrast to all private conversations surreptitiously recorded by a state agent).[174] Taking a case-by-case approach to whether an exchange is private may appear to run contrary to the Court’s earlier jurisprudence affirming a normative and content-neutral approach to assessing reasonable expectation of privacy.[175] (In this author’s view, Marakah’s case-by-case approach is a normative inquiry, one that looks not at the content of a communication but at its form or nature.) Setting that debate aside, however, the point here is that concerns about the absurdity of police needing a warrant or civilian disclosure engaging section 8 belong in the assessment of whether the accused retained a reasonable privacy interest in the communication—and not, if he did have an interest, whether receipt constituted a search or seizure.

Conclusion

Courts are increasingly relying on evidence disclosed by civilian recipients of texts, email, and other digital evidence volunteered to police. Courts and commentators have found it challenging to apply to this situation the Supreme Court’s long-standing, clear, and consistent jurisprudence on what constitutes a search or seizure and when it will be reasonable. In many cases, judges and scholars have been quick to assume that since a piece of evidence turned over is brief or slight (a one-page exchange, a screenshot) and since police have been told of its contents, they do nothing to interfere with any possible continuing privacy interest a suspect may have if an officer proceeds to peruse it. They do not see the item as engaging a significant privacy interest, if any, or the perusal as an investigative technique. And more to the point, if it were put to them that even a brief text possibly containing a threat still engages a privacy interest and perusing it is a search, they would ask ‘why should it matter’? Why should police need authority to conduct this search when they already know what a text contains and the exchange or email is so brief?

This article has sought to demonstrate that the Court has consistently approached the question of whether a search and seizure has taken place not by asking who gave what to whom and why. In every case, the Court has asked instead whether an accused retained a reasonable privacy interest against the state in the item at issue and whether in taking, receiving, or perusing the item, the police had an investigative purpose. The Court’s holdings on digital evidence, including Cole, Reeves, and Marakah suggest that we may retain a high degree of privacy in even a brief or seemingly slight piece of evidence such as a short exchange or a screenshot of some portion of the content of a device, and that police review of it entails a search.

Judges and commentators have also overlooked the rationale for insisting that police obtain a warrant or other authority to render a search in these circumstances reasonable. In many cases, this will seem redundant or pointlessly formalistic, given the knowledge the officer already possesses at the time she proceeds to peruse a brief text. The point is not to make her work unnecessarily difficult. As the Supreme Court has held in numerous cases canvassed here where police already the know what a search will likely reveal, putting police to the trouble of obtaining a warrant or other authority affords everyone in Canada the assurance that the state will not interfere with a reasonable expectation of privacy unless certain standards are met. Digital communications are ubiquitous and a central part of our lives. The assurance of privacy from state interference may seem abstract and remote to many, but its absence would affect us all.