Can’t Take It

Can the police seize a computer (without searching it) if only one of its co-owners consents?

In R v Reeves, 2018 SCC 56, delivered last week, the Supreme Court held that section 8 of the Canadian Charter of Rights and Freedoms, which protects the right not to be subject to unreasonable search and seizure, prevents the police from seizing (even without searching) a computer located in the common area of a home with the consent of only one, but not the other of the home’s occupiers. As in a number of other search-and-seizure cases, it is Justice Karakatsanis who takes the lead in articulating a narrow view the police powers. Unlike in R v Fearon, 2014 SCC 77, [2014] SCR 621 and R v Saeed, 2016 SCC 24, [2016] 1 SCR 518 she carries a strong majority of her colleagues ― all but one, in fact, on this issue ― with her. And unlike in those two cases, I suspect that Justice Karakatsanis’ pro-privacy disposition has not served her well.

The facts of the case are somewhat quirky. Mr. Reeves had shared a home ― and a computer ― with his common-law spouse, Ms. Gravelle. So far, so ordinary. However, following a family violence incident, Mr. Reeves was barred from being at the home without Ms. Gravelle’s consent, which she eventually withdrew. Still, as a matter of property rights, both the home and the computer were still shared between Mr. Reeves and Ms. Gravelle. Ms. Gravelle also informed the authorities that she had previously found child pornography on the computer. A police officer came, and, with Ms. Gravelle’s consent, he took the computer back to the police station ― where it sat, seemingly of no interest to anyone, for four months, despite the Criminal Code requiring such seizures to be reported to a justice of the peace. Eventually, the police finally concocted a warrant application ― which the trial judge later found to be tendentious and deficient to the point of invalidating the warrant ― and searched the computer, duly finding the child pornography, leading to charges against Mr. Reeves, who argued that the evidence was obtained in violation of his Charter rights and should be excluded.

 There were two possible violations of Mr. Reeves’ rights for the Supreme Court to look into. First, the police officer’s entry into and search of the shared home; second, the seizure of the shared computer. (There was no dispute that the lengthy detention of the computer in violation of the Criminal Code and its search pursuant to a warrant that was subsequently invalidated were constitutionally problematic.) However, for the majority, Justice Karakatsanis does not pronounce on the requirements of the Charter with respect to police entry into a shared home with the consent of only one of its occupiers. She finds that the matter is best left for another time, when it will be more fully argued. Justice Moldaver, in a concurring opinion, agrees that now is not the time to dispose of the question ― and proceeds to lay out a detailed case for why the police have a common law power to enter to speak with one co-occupier of a shared home, while insisting that this argument is only tentative.

For the majority, the case turns on the question of the seizure of the computer. This, in turn, divides into two sub-issues: first, whether Mr. Reeves had a “reasonable expectation of privacy” in the machine; and second, whether Ms. Gravelle could obviate or waive his rights by consenting to the seizure anyway.

As Justice Karakatsanis explains, “[t]he reasonable expectation of privacy standard is normative, rather than descriptive” [28] ― it is not really about what the person concerned expected his or her privacy rights to be in the circumstances, but about where the balance between privacy and societal interests (in particular, in the investigation and punishment of offences) ought to be struck. In deciding this question, Justice Karakatsanis insists that “the subject matter of the seizure was the computer, and ultimately the data it contained about Reeves’ usage, including the files he accessed, saved and deleted”. [30] Even though, as the Supreme Court previously held, a separate warrant would be required to actually search the data contained in the computer, “Reeves’ informational privacy interests in the computer data were still implicated” [30] because he lost control of it, including the ability to destroy it. The data computers contain can be “highly private” [34], and thus not only the search, but also the mere seizure of “a personal computer from a home” “presumptively require[s]” “specific, prior judicial authorization”. [35] This is so even when the computer is shared and no one individual can expect absolute privacy when using it.

As for a co-occupier of a home consenting to the police seizing a shared computer, it “cannot nullify” [41] an existing reasonable expectation of privacy: “[t]he decision to share with others does not come at such a high price in a free and democratic society”. [44] Those others can report suspicions to the police, but it does not follow that the police can do anything they (the others, that is) consent to. It is not their (the others’) rights that are at stake, after all, and the fact that they too may have rights or privacy expectations over the same object or space is beside the point. As for what the police can do if a person actually brings an object in which another has a reasonable expectation of privacy to them, that “remains for another day”. [46] 

In the event, the majority, as well as Justice Moldaver, conclude that the Charter breaches in this case are serious enough to warrant excluding the evidence found Mr. Reeves’ computer. Justice Côté, in a concurring opinion, agrees with this outcome ― even though, as I am about to explain, she does not think that the seizure of the computer amounted to a Charter breach at all. (The Supreme Court, which only considers outcomes in its statistics, will triumphantly count Reeves as yet another unanimous decision ― yet as Peter McCormick recently explained here, it is a mistake to do so.)

On the key issue of the case ― the application of section 8 of the Charter to shared spaces and objects ― Justice Côté takes an approach that is the opposite of the majority’s. Unlike Justice Karakatsanis, Justice Côté directly addresses the question of whether police can enter shared spaces with the consent of a single occupier ― and answers in the affirmative, albeit with a possible qualification. Justice Côté writes that “it is not objectively reasonable for a cohabitant … to expect to be able to veto another cohabitant’s decision to allow the police to enter any areas of that home that they share equally”. [112] Such a veto would amount to a negation of “consenting cohabitant’s liberty and autonomy interests with respect to those spaces”. [112] It would also “require the police to identify, locate and obtain the consent of every person who lives in the home, or has any expectation of privacy with respect to common areas of the home”, [114] which is likely to be unduly burdensome at best, if not quite impossible. And applying this approach to shared virtual spaces or objects ― say, a text messaging chain ― would produce similarly perverse consequences. Meanwhile, the search of common areas of a shared home is unlikely to produce intrusions into a person’s deepest secrets.

Justice Côté takes a similar approach to the seizure of a shared computer. While acknowledging that searching such a computer would require prior authorization, she argues that the mere seizure consented by one of the computer’s co-users is not a violation of the other co-users’ rights ― and thus disagrees with the majority on this key point. Again, it is not reasonable to expect that a co-user will not allow the authorities to seize a shared computer, and concluding otherwise would deny the co-user’s  autonomy. The context of co-ownership and joint control influences the scope of one’s reasonable expectation of privacy. Furthermore, Justice Côté stresses the fact that only the seizure of the physical object, not the information it contains, is at issue, and reproaches the majority for conflating the two. She also points out that the majority’s approach may well amount to recognizing an expectation of privacy not just in favour of a co-owner of a computer, but also of, say, a guest who used it at some point in the past. Justice Côté adds that of course a co-owner should be able to take a shared computer to the police ― and letting the police take it is no different.

As noted at the outset, I am inclined to think that Justice Karakatsanis has indeed gone too far in protecting privacy here. Justice Côté is right that the majority conflate interests in maintaining control of a physical object and those in ensuring the privacy of the data that this object contains. And it is true that, by effectively granting a veto to each co-user (and perhaps even a past user ― perhaps the majority would distinguish that case, but it is indeed unclear how, and it’s unfortunate that Justice Karakatsanis doesn’t address the point), the majority compromise the autonomy ― and interfere with legitimate interests ― of other co-users. It would, as Justice Côté says, be odd if such people couldn’t take things of which they have legitimate control to the police ― and no less odd if they could not invite the police to take such things.

At the same time, a couple of points bother me about Justice Côté’s reasons. First, she might be too sanguine about the prospect of police not gaining access to particularly private information in common areas of shared dwellings. This may indeed be a reasonable assumption if the people living together are what in North America are misleadingly called) room-mates, and elsewhere, more accurately, flat-mates, who each have a private room. But if they are spouses, or otherwise family members, the distinction between common and private spaces within the shared home may not be drawn with any clarity. Perhaps this does not matter after all, but Justice Côté would have done well to address this issue.

I also am somewhat puzzled by Justice Côté’s references to the odd circumstance that Mr. Reeves had lost access to the theoretically-shared home and computer that were the objects of the police’s interest here. It’s not quite clear how much this fact matters to Justice Côté’s conclusions. I think it’s not particularly significant in her reasoning regarding police entry into a shared dwelling, but on the seizure issue, Justice Côté explicitly says that it is “relevant” that Mr. Reeves “lacked control [of the computer] as a result of his own actions”. [130] Yet not only was this “result” an indirect and unintended, albeit foreseeable one, but, more importantly, one is left in some doubt about how Justice Côté thinks more ordinary cases, where this “relevant” factor will not be present, ought to be decided.

Ultimately, Reeves might not be a very important case. The one issue it actually decides, whether police can seize shared computers with the consent of one but not all of their users, may not recur all that often, insofar as people increasingly use personal laptops, tablets, or smartphones. I don’t actually know if they do, but I suspect that they might. Perhaps its chief interest is in the trends that it confirms: Justice Karakatsanis’ role as the Supreme Court’s leading pro-privacy voice, and Justice Côté’s as its leading independent thinker. On the whole, the Court needs both, even when they disagree, and even in cases where neither is quite right. 

Its Own Place

“The mind is its own place,” says Milton’s Satan. And since computers have, for all practical purposes, replaced our brains, so are those, right? The Supreme Court of Canada, at any rate, agrees. In a case decided last week, R. v. Vu, 2013 SCC 60, it held that police cannot search a computer on the basis of a warrant to search the place in which the computer is found. The search of the computer requires a separate warrant.

As Justice Cromwell, writing for the unanimous court, explains,

The traditional legal framework holds that once police obtain a warrant to search a place for certain things, they can look for those things anywhere in the place where they might reasonably be; the police do not require specific, prior authorization to search in receptacles such as cupboards and filing cabinets. The question before us is whether this framework is appropriate for computer searches; in short, should our law of search and seizure treat a computer as if it were a filing cabinet or a cupboard? (Par. 1)

The answer is that it should not, because “computer searches give rise to particular privacy concerns that are not sufficiently addressed by that approach” (par. 2).

The search warrant which the police had obtained authorized a search “documentation” that would prove the ownership and occupation of the appellant’s residence. The main question for the Court was whether this was enough to authorize the police to look for such documents on the computers (or cell phones) that they might find in the residence, in addition to looking for any physical copies.

Searching a computer says Justice Cromwell, is not the same thing as searching the physical space of a residence. The amount of data to be found on a computer is huge, and that data includes things “that users cannot control, that they may not even be aware of or may have chosen to discard and which may not be, in any meaningful sense, located in the place of the search” (par. 24). From the standpoint of the privacy interests they engage, computers are unprecedented technology, unlike anything the law has had to deal with in the past. The same thing goes for cell phones: “present day phones have capacities that are, for our purposes, equivalent to those of computers” (par. 38). Not only can computers hold information that would fill up whole libraries, but they record the websites they are used to visit, the terms they are used to research, and the history of the editing of the documents they are used to create. And theirs, rather like ours in fact, is a memory that does not really forget even told to do so: a computer does not actually delete a file, but merely makes the disk space it occupied available for reuse should the need arise. Finally, a computer ― unlike, say a filing cabinet ― gives one access to information that is not physically located on it, perhaps, “in the cloud,” on servers half a world away. In short, for the police to find something in a physical place, someone needs to have put it there, and no one must have thrown it out; neither of these conditions holds on a computer.

Because computers contain information that is thus so different from information that can be found by searching a physical place, a distinct judicial authorization is required to search a computer ― an authorization to search the place where the computer is found is not enough. Nor is after-the-fact review of the reasonableness of the search of a computer. Justice Cromwell concludes that “the privacy interests at stake when computers are searched require that those devices be treated, to a certain extent, as a separate place” (par. 51).

The way it will work in practice, he explains, is that

[i]f police come across a computer in the course of a search and their warrant does not provide specific authorization to search computers, they may seize the computer (assuming it may reasonably be thought to contain the sort of things that the warrant authorizes to be seized), and do what is necessary to ensure the integrity of the data. If they wish to search the data, however, they must obtain a separate warrant (par. 49).

Such warrant need not, however, set out a detailed protocol for the search, or exclude in advance some “areas” of the computer to be searched. Not only are such protocols likely to be difficult to draw up in advance, but after-the-fact review will also allow courts to develop better rules for governing future searches of computers. A warrant can, in appropriate cases, impose conditions on how a computer is to be searched, but it need not always do so.

Sounds right to me. And it is worth pointing out, I think, that this case illustrates the development of new privacy expectations that come with new technologies. As I suggested here, the influence of technology on privacy rights is unlikely to be a one-way ratchet leading to the eventual disintegration of the idea of privacy. Some ideas of what is and what is not private will disappear in the digital world. Others will not. And others still will appear which did not exist before. Although I have regularly expressed my skepticism about courts’ ability to deal with new technologies, this case provides grounds for at least some optimism.