The Ninth Circuit recently issued a remarkable decision on the search and seizure of computers in United States v. Payton. I suspect this case may go en banc, as the issue is pretty important, so it may not be the last we’ve heard about it. Still, I wanted to flag the many errors in this opinion for readers in case the case stays on the books. You can read the appellant’s opening brief here, the government’s brief here, and the reply brief here; you can listen to the oral argument here. (I would have blogged about it earlier, but I was on my blogging hiatus when it came down on July 21.)
The facts of the case are simple. The police had reason to believe that Payton was selling drugs from his home, and they obtained a warrant to search his home for drugs, sales ledgers relating to drugs, and financial records for the person who lived in the home. The affidavit of probable cause specifically requested permission to search any computer located in the house. The affidavit explained that based on the officer’s experience, financial records were likely to be found on a computer, and so it requested permission to search any computers on site and seize any machines if evidence was found on them.
During the search, an officer saw a computer in the bedroom that was “on” but had its screensaver up. The officer moved the mouse, which removed the screen saver. He then clicked open a file, and saw that it contained child pornography. Further investigation confirmed that Payton had other images of child pornography on his computer, leading to child pornography charges. No drugs or records about drugs were discovered.
In a decision by Judge Canby, joined by Judge Wardlaw and District Judge Mills, the Ninth Circuit suppressed the child pornography discovered in the computer. According to Judge Canby, the high storage capacity of computers made it constitutionally unreasonable to search a computer in those circumstances. Although the affidavit requested permission to search computers discovered when the warrant was executed, the affidavit was not incorporated: The warrant itself did not explicitly authorize the search of computers.
Further, there was no specific evidence found in the home near the computers that suggested that the evidence such as financial records were stored in the computer. The officers did not find any other evidence within the scope of the warrant elsewhere in the home, and that made it unlikely that there would be evidence described in the warrant inside the computer. The court concluded: “In the absence of any circumstances supporting a reasonable belief that items specified in the warrant would be found on the computer, the search did not meet the Fourth Amendment standard of reasonableness.”
Judge Canby also commented on the policy consequences of a contrary rule:
Our confidence in our conclusion is buttressed by contemplating the effect of a contrary decision. In order to uphold the search in this case, we would have to rule that, whenever a computer is found in a search for other items, if any of those items were capable of being stored in a computer, a search of the computer would be permissible. Such a ruling would eliminate any incentive for officers to seek explicit judicial authorization for searches of computers. But the nature of computers makes such searches so intrusive that affidavits seeking warrants for the search of computers often include a limiting search protocol, and judges issuing warrants may place conditions on the manner and extent of such searches, to protect privacy and other important constitutional interests. We believe that it is important to preserve the option of imposing such conditions when they are deemed warranted by judicial officers authorizing the search of computers. If unwarranted searches of computers are automatically authorized by upholding the search in Payton’s case, that option will be lost.
It’s hard to know where to begin with the number of problems with this opinion. I think I’ll start with the big-picture conceptual issues; then turn to Supreme Court precedent; and then go to Ninth Circuit precedent.
1) To begin with the most basic conceptual error, the Fourth Amendment simply does not require warrants to list the items to be searched. As the text of the Fourth Amendment makes clear, warrants must “particularly describ[e] the place to be searched and the persons or things to be seized” (emphasis added). The Fourth Amendment requires a description of the things to be seized, not a description of containers that are searched during the hunt for the things to be seized. I don’t think I have ever seen a warrant that specifically listed the items to be searched: There isn’t even a place for that on the standard warrant form.
The text of Rule 41, the rule that governs search warrants, makes this point plain: It requires warrants to list the property that the agents want to “search for and seize.” Note search “for,” not search. (This was a state warrant, to be clear, so Rule 41 isn’t directly applicable; I point it out only to emphasize that there is no requirement that warrants list the items to be searched.) When the officer moved the computer mouse and saw the image, he didn’t seize a computer. He didn’t even seize the image. Rather, he searched the computer without seizing anything. It was unneceesary to get prior authorization to seize a computer because no computers were actually seized.
2) Now I’ll move on to Supreme Court precedent. There is no specific Supreme Court precedent on how the Fourth Amendment applies to the search of computers. But it’s worth noting that the Supreme Court has been clear that a warrant to search a place for specific evidence permits the search of anything in that place where the described evidence could be located. Here’s Justice Stevens making the point pretty forcefully in United States v. Ross, 456 U.S. 798 (1982):
A lawful search of fixed premises generally extends to the entire area in which the object of the search may be found, and is not limited by the possibility that separate acts of entry or opening may be required to complete the search. Thus, a warrant that authorizes an officer to search a home for illegal weapons also provides authority to open closets, chests, drawers, and containers in which the weapon might be found. A warrant to open a footlocker to search for marihuana would also authorize the opening of packages found inside. A warrant to search a vehicle would support a search of every part of the vehicle that might contain the object of the search. When a legitimate search is under way, and when its purpose and its limits have been precisely defined, nice distinctions between closets, drawers, and containers, in the case of a home, or between glove compartments, upholstered seats, trunks, and wrapped packages, in the case of a vehicle, must give way to the interest in the prompt and efficient completion of the task at hand.
It seems to me that Judge Canby’s decision in Payton tries to create just such a “nice distinction” for computers. But I think it’s hard to square with the Supreme Court’s view in Ross.
3) Finally, the Payton opinion purports to root its approach in a recent Ninth Circuit decision, United States v. Giberson, 527 F.3d 882 (9th Cir. 2008). This doesn’t work, however, because the legal issue in Giberson was very different. The question in Giberson was whether the police could seize computers onsite and take them away for later search pursuant to a second warrant when the initial warrant was for physical documents. The warrant didn’t ask for permission to seize computers, but the agents did seize them, figuring that the computers might contain those documents in physical form. They then obtained a second warrant to search the computers. Thus the question in Giberson was whether the seizure was permissible when the first warrant did not explicitly authorize it: The Giberson court concluded that, in the context of that case, it was.
The Payton court reasons that if Giberson was fact-specific, then there is a general reasonableness test applicable here: Accordingly, there must be facts according to which the search and seizure of a computer with a warrant was unreasonable. It then announces that this is such a case. But whether the Fourth Amendment allows a seizure of an item not named in the warrant is very different from whether the warrant allows a search for evidence described in the warrant that happens to be on a computer. As I explained above, the Fourth Amendment requires a description of the items to be seized, and limits the permitted seizure to those items, but the Fourth Amendment does not require a description of the items to be searched. Rather, so long as the police have a valid warrant, they can search anywhere in the place to be searched that could store the items to be seized.
I should add that I am quite sympathetic to the concern motivating the panel: As the oral argument makes clear, the judges were concerned with the core problem of computer search and seizure of how to limit computer searches. I’m concerned about that, too, and have written an article on how best to approach the problem. But the Payton court’s effort to address the problem just doesn’t work: In addition to its many practical problems, which I’ll leave out for now, it’s pretty clearly inconsistent with the text of the Fourth Amendment and Supreme Court precedent. The panel instead should have dealt with this problem as a plain view issue, and analyzed whether the opening of the files was justified as within the scope of the warrant based either on the subjective approach of United States v. Carey (10th Cir. 1999), the objective approach used outside the computer setting, or using some other approach to plain view.