Oregon Supreme Court Adopts Use Restrictions on Nonresponsive Data for Computer Warrants

A ruling under the state constitution, but still really interesting.

|The Volokh Conspiracy |

As regular readers know, I have argued in my academic writing that the Fourth Amendment should be interpreted to impose use restrictions on nonresponsive data seized pursuant to a computer search warrant. In a new decision, State v. Mansor, the Oregon Supreme Court appears to have adopted my approach under Oregon's state equivalent of the Fourth Amendment.

Let me start with some context. Computer warrant searches require the government to find a needle in an enormous electronic haystack. When the police execute a warrant to search for and find the needle of evidence, they usually need to seize the haystack first to search it. I have argued that a warrant to seize the needle should allow the police to seize the haystack to search for the needle. But there's a catch: The government should ordinarily not be allowed to use whatever else they find in the haystack. If the warrant is only to seize a needle, the police can only take away and use the needle, unless there are exigent circumstances exposed by the discovery of other evidence. The nonresponsive data—other evidence that may exist in the haystack but is not described in the warrant—ordinarily can't be used. For the details of my view, see this article.

In today's decision, the Oregon Supreme Court adopted my approach under its state constitution. Here's the key part of the Court's opinion:

[The search for the evidence described in the warrant] was supported by probable cause, was sufficiently specific, and was not overbroad. The nature of a computer search, however, means that, in searching for that [evidence], that the forensic examiners were likely to come across or discover additional information. And, in this case, the forensic examination searched for and uncovered information, later used at trial, that went far beyond the scope of the warrant.

To ensure the protection of Article I, section 9, rights, we must consider what restrictions, if any, should be imposed on the use of information police obtain through reasonably executed warranted computer searches when those searches uncover evidence beyond that authorized in the warrant, and when no exception to the warrant requirement supports the collection or use of that evidence.

In our view, the privacy interests underlying Article I, section 9, are best protected by recognizing a necessary trade-off when the state searches a computer that has been lawfully seized. Even a reasonable search authorized by a valid warrant necessarily may require examination of at least some information that is beyond the scope of the warrant. Such state searches raise the possibility of computer search warrants becoming the digital equivalent of general warrants and of sanctioning the "undue rummaging that the particularity requirement was enacted to preclude." Mansor, 279 Or App at 803 (internal quotation marks omitted).

Although such searches are lawful and appropriate, individual privacy interests preclude the state from benefiting from that necessity by being permitted to use that evidence at trial. We thus conclude that the state should not be permitted to use information obtained in a computer search if the warrant did not authorize the search for that information, unless some other warrant exception applies. See Kerr, 48 Tex Tech L Rev at 24 (suggesting use restrictions for data "nonresponsive" to the warrant). Put differently, when the state conducts a reasonably targeted search of a person's computer for information pursuant to a warrant that properly identifies the information being sought, the state has not unreasonably invaded the person's privacy interest, and the state may use the information identified in the warrant in a prosecution or any other lawful manner. But when the state looks for other information or uncovers information that was not authorized by the warrant, Article I, section 9, prohibits the state from using that information at trial, unless it comes within an exception to the warrant requirement.

I'm pleased to see my approach adopted, although unfortunately the case involves horrific facts—and a terribly drafted warrant that caused the problem.

Here's a very simplified version of the facts (enough to explain the legal context, no more). The police suspected that Mansor had abused his infant son in a way that led to his tragic death. The police obtained a warrant to search Mansor's computer for evidence of abuse. But the warrant they drafted was incredibly narrow. As construed by the Oregon Supreme Court, the warrant only authorized a search for the Internet search terms that Mansor had entered in to his browser for a specific 15-minute window before Mansor had called 911 to report his son's injury. As the Court construed the warrant, at least, that was all the warrant authorized: All other evidence of child abuse on the computer was outside the warrant.

You can guess what happened. During the investigation, agents believed there was a lot more relevant evidence on the computer. And I'm guessing they didn't expect courts to read the warrant as narrowly as the Oregon Supreme Court did. In any event, the agents ended up looking through several years' worth of Internet browsing records for evidence related to child abuse. And their hunch was right: The broader search revealed evidence relevant to the charged crime that was then used at trial to convict the defendant.

This is was a horrific crime, and it's always hard to see a conviction overturned in a case like this. It's especially hard when the scope of the warrant was itself unclear, so the ruling seems like a bit of a technicality. But note that the result in this case would have been different if the police had merely drafted the warrant properly. As far as I can tell, the officers had probable cause that would have allowed them to obtain a much broader warrant allowing a search through the computer for any evidence of child abuse. All of the evidence would have been admissible at trial if the warrant had not been written in such a remarkably narrow way. It's also important to note that the Oregon Constitution's Article I, section 9—the state equivalent to the Fourth Amendment—does not have a good-faith exception that would have been an alternative ground for affirmance in the federal system. See State v. Johnson, 120 Or App 151, 156 (1993) (noting lack of a good faith exception in the state constitution).

Mansor is not a Fourth Amendment decision. As I mentioned at the top, it was decided under the Oregon Constitution's search and seizure provision that goes beyond the federal Fourth Amendment. Given that, the decision is interesting and relevant more by way of illustration than precedent. But I though the opinion was worth flagging nonetheless for those interested in computer search and seizure law.