In People v. Herrera, decided earlier this week, the Supreme Court of Colorado ruled that a warrant to search a seized cell phone did not justify opening a folder that was unlikely to contain the evidence described in the warrant. Because the government couldn’t open the folder, the evidence found inside it was suppressed.

I think the reasoning of the court is pretty puzzling, but that the court probably reached the right result. Here’s a run-down.

I. The Facts

First, the facts. The mother of a young girl named Faith W. contacted the police and reported that Herrera had sexual interactions with Faith W. that were facilitated by text messages between her and Herrera. An officer posing as a young girl named “Stazi” began texting with Herrera to see if Herrera would want to set up a sexual tryst with “Stazi,” too. (I guess Stasi was taken as a police undercover name? But I digress.) Herrera sent back sexual responses to “Stazi” that led to Herrera’s arrest.

At the time of Herrera’s arrest, the police seized but did not search his cell phone. The police later obtained a narrow warrant to search the cell phone and seize just two kinds of evidence. First, the warrant authorized the seizure of evidence showing that the phone belonged to Herrera. Second, the warrant authorized the seizure of the messages and photos sent between Herrera and “Stazi.” Importantly, the government did not get a warrant to seize any records of interactions between Herrera and Faith W.

An officer executed the warrant by manually searching the phone. During the search, he came across an app for the Kik Messenger, which is a Microsoft Internet-based text messaging app. The officer opened the app and found that the messages were organized by name. While scrolling down to look for messages between Herrera and “Stazi,” the officer found a text message folder identified by the name ‘Faith Fallout.’ The officer later testified that he believed the messages in the folder were the messages between Herrera and Faith W. So he clicked on the name for the folder and indeed found the set of communications between Herrera and Faith W.

The government is now prosecuting Herrera, and prosecutors want to use the records found in the ‘Faith Fallout’ folder. Herrera moved to suppress those records.

II. The Colorado Supreme Court’s Reasoning

In an opinion by Justice Eid, the Colorado Supreme Court agreed 5-2 that the contents of the folder should be suppressed. The Court focused on the government’s authority to open the ‘Faith Fallout’ folder. According to the Court, neither the warrant nor the plain view exception allowed the government to open that folder. Because the folder couldn’t be opened, the evidence inside couldn’t be used.

First, opening the folder was not permitted by the warrant itself because allowing the search of the folder would violate the particularity requirement:

The warrant in this case authorized a search of Herrera’s cellphone for text messages between Herrera and “Stazi” as well as for “indicia of ownership.” The People contend that the warrant thus permitted a search of the text messages contained in the ”Faith Fallout” folder because any message found there would reveal Herrera as the owner of the phone. We believe this argument proves too much, as it would authorize a general search of the entire contents of the phone. Indeed, the People argue that any piece of data on the phone, including any text message on the phone, would have the possibility of revealing Herrera’s ownership of the phone. This rationale transforms the warrant into a general warrant that fails to comply with the Fourth Amendment’s particularity requirement.

Second, the Court ruled that the plain view exception did not allow opening the folder because it was not reasonable to think that it contained evidence described in the warrant:

In executing a search warrant, police officers may search areas in which the items identified in the warrant might reasonably be found, including closed containers. People in Interest of D.F.L., 931 P.2d 448, 452 (Colo. 1997); see also People v. Koehn, 178 P.3d 536, 537 (Colo. 2008) (where warrant authorized search of defendant’s residence for firearms and ammunition, officers were justified in searching kitchen cabinet and pants and seizing incriminating items found there in plain view). We analogize the “Faith Fallout” text message folder to a closed container, which Detective Slattery opened to discover its contents—namely, the text messages between Faith W. and Herrera. Here, the warrant authorized Detective Slattery to search for messages between “Stazi” and Herrera. The question, then, is whether the “Faith Fallout” folder was a container in which messages from “Stazi” could reasonably be found. We agree with the trial court that it was not.

The trial court concluded that in searching Herrera’s cell phone, Detective Slattery was authorized to search for messages from “Stazi’s,” or Detective Dodd’s, number. As noted above, however, the police had an objective basis to believe that the “Faith Fallout” folder was associated with a different number—that is, one that belonged to Faith W., not Detective Dodd. Furthermore, there was no evidence before the trial court that a specific folder in the Kik application could contain messages from multiple numbers. Instead, the evidence indicated that each folder could only be associated with a single number. And the trial court specifically found that there was no link between the “Stazi” number and the “Faith Fallout” folder. Thus, because the evidence objectively indicated that the “Faith Fallout” folder contained messages from Faith W. and only Faith W., the police had no objective basis to conclude that the folder would contain messages from “Stazi.”

. . . If we were to hold that any text message folder could be searched because of the abstract possibility that it might have been deceptively labeled, we would again be faced with a limitless search, as with the People’s first argument. We instead proceed cautiously in applying the plain view doctrine to searches involving digital data.

III. My Analysis — Probably the Right Result, But A Puzzling Way to Get There

Herrera is a puzzling opinion. I think the result is probably right, but it takes some unusual twists and turns to get there. Let me start with what I see as the missteps, and then I’ll say why I think the outcome was probably correct.

First, I don’t think it works to say that the folder couldn’t be opened because allowing it to be opened would violate the particularity requirement. The particularity requirement is about the facial validity of the warrant, not how it is executed. The Fourth Amendment requires the warrant to state with particularity the place to be searched and the items to be seized. Here the warrant was very particular. The place to be searched was the phone, and the items to be seized were the ID information and the messages with “Stazi.” That’s about as particular as you can get, I think. If the folder couldn’t be opened, the reason for that isn’t the particularity requirement.

Second, I’m puzzled by what the court calls its “plain view” analysis. As an initial matter, it doesn’t have much to do with plain view. The court rests its analysis on whether the folder could be opened under the warrant. But that’s not a plain view question, which is about warrantless seizures (and sometimes warrantless searches). Instead, it’s a question of what kind of search under the warrant is constitutionally reasonable.

Turning to whether the warrant authorized the opening of the folder, the court frames the question in what strikes me as a surprising way. According to the court, the key question is whether it was reasonable to think that the evidence described in the warrant would be in the folder. But that’s not the question courts generally ask in this setting. Instead, the cases ask whether it’s reasonable to think that the evidence could be in the place searched. See, e.g., United States v. Rutkowski, 877 F.2d 139, 141 (1st Cir. 1989)(“Any container situated within residential premises which are the subject of a validly-issued warrant may be searched if it is reasonable to believe that the container could conceal items of the kind portrayed in the warrant.”); United States v. Ross, 456 U.S. 798 (1982) (“A lawful search . . . 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.”). As I have understood it, the doctrine is mostly about size: Could the item fit in the place to be searched? See, e.g., United States v. Evans, 92 F.3d 540, 543 (7th Cir. 1996) (holding that a container in the place to be searched can be opened and searched if it “is large enough to contain the contraband or evidence that [the officers] are looking for.”).

As I see it, Herrera quietly changes could to would. The evidence described in the warrant could be in the folder. The data might have been there. Whether it was reasonable to think it would be in the folder — the question the Court asks — doesn’t strike me as the right question.

Having criticized the opinion, I’ll now turn to why I think it probably reached the correct result. There were two paths the court could have taken to get there.

First, the court could have followed the approach of the Tenth Circuit and ruled that the officer’s subjective intent to go outside the warrant and look for other evidence rendered the subsequent search unreasonable. See United States v. Carey, 172 F.3d 1268, 1273 (10th Cir. 1999). In Herrera, the officer took the stand and admitted that he opened the file because he thought it contained evidence outside the scope of the warrant that he nonetheless wanted. That’s exactly what happened in Carey, leading the Tenth Circuit to suppress the evidence. It would have been easy for the Colorado court to just follow the Tenth Circuit’s approach and reach the same result.

Second, and more aggressively, the court could have followed an approach endorsed by my scholarship and held that evidence outside the warrant was inadmissible because courts must impose a use restriction on non-responsive data in digital warrant cases. Because the messages with Faith W. were outside the scope of the narrowly-drawn warrant, they could not be used and would be suppressed.

If that seems a harsh result, note that the government could have easily solved that problem by just drafting the warrant more broadly. The warrant in this case should have asked for communications between Herrera and Faith W., too, or more broadly of communications of a sexual nature between Herrera and any person who is or purports to be under the age of consent. Better drafting would have solved the problem under my proposed rule, the Tenth Circuit’s approach, and (I think) the Colorado Supreme Court’s view.

Finally, I want to flag Adam Gershowitz’s article, forthcoming in the Vanderbilt Law Review, that makes some arguments similar to that made by the Herrera court. Check it out here: The Post-Riley Search Warrant: Search Protocols and Particularity in Cell Phone Searches. In some ways, the Colorado court echoes what Adam suggests but without the ex ante search protocols. So while I disagree with some of the reasoning in Herrera — and with Adam’s somewhat-related position — the Colorado opinion certainly gives some fuel to the arguments Adam makes.