The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Court invalidates cell phone warrant as overbroad


Here's an interesting new case on the scope of cell phone warrants: United States v. Winn, __ F.Supp.3d __, 2015 WL 553286 (S.D.Ill. February 09, 2015) (Rosenstengel, J.). In the case, the state applied for a search warrant to search the cell phone of a man named Winn. Winn had been seen using his cell phone to take pictures or videos of young girls in public places while he was rubbing his genitals. The government asked for and obtained a warrant to search Winn's phone for all evidence of public indecency stored inside it.

Officers searched the phone pursuant to the warrant using a mobile phone forensic tool known as a Cellebrite UFED Touch machine. The device "extract[ed] data from Winn's cell phone and to generate a report detailing what was extracted." The report revealed "two calendar events, a forty-four item call log, twenty-three contacts, eighty text messages, 312 images, and twenty-five videos." Among the images were child pornography, leading to child pornography charges.

Judge Rosenstengel, a relatively new Obama appointee, suppressed the child pornography on Fourth Amendment grounds. According to Judge Rosenstengel, the warrant was defective for two basic reasons. First, the warrant did not limit the search by data type. The warrant asked for all evidence on the phone that was evidence of public indecency, but there was only reason to think that the evidence would be in the specific form of pictures or videos:

The warrant authorized the seizure of "any or all files" contained on the cell phone and its memory card that "constitute[d] evidence of the offense of [Public Indecency 720 ILCS 5/11-30]," including, but not limited to, the calendar, phonebook, contacts, SMS messages, MMS messages, emails, pictures, videos, images, ringtones, audio files, all call logs, installed application data, GPS information, WIFI information, internet history and usage, any system files, and any delated data (Docs.22-2, 22-3).

The major, overriding problem with the description of the object of the search-"any or all files"-is that the police did not have probable cause to believe that everything on the phone was evidence of the crime of public indecency. The description was a template used by the St. Clair County State's Attorney's Office for all cell phone searches. Templates are, of course, fine to use as a starting point. But they must be tailored to the facts of each case. This particular template authorized the seizure of virtually every piece of data that could conceivably be found on the phone. The Supreme Court put the scope of such a wholesale seizure in perspective by explaining that it "would typically expose the government to far more than the most exhaustive search of a house." Riley v. California, 134 S.Ct. 2473, 2491 (2014) (emphasis in original). Obviously, the police will not have probable cause to search through and seize such an expansive array of data every time they search a cell phone. And, in the case of a misdemeanor crime, it is difficult to fathom why the police would ever need, or have probable cause, to do so. That certainly holds true in the instant case.

Based on the complaint supporting the search warrant, there was probable cause to believe that only two categories of data could possibly be evidence of the crime: photos and videos (see Doc. 22-2). The complaint did not offer any basis-such as facts learned during the investigation or Detective Lambert's training and expertise-to believe that the calendar, phonebook, contacts, SMS messages, MMS messages, emails, ringtones, audio files, all call logs, installed application data, GPS information, WIFI information, internet history and usage, or system files were connected with Winn's act of public indecency. In fact, the narrative portion of the complaint did not even mention those categories of data. Furthermore, Detective Lambert admitted at the hearing that he had no reason to believe much of that data contained evidence of the crime of public indecency.

. . . The bottom line is that if Detective Lambert wanted to seize every type of data from the cell phone, then it was incumbent upon him to explain in the complaint how and why each type of data was connected to Winn's criminal activity, and he did not do so. Consequently, the warrant was overbroad, because it allowed the police to search for and seize broad swaths of data without probable cause to believe it constituted evidence of public indecency.

The judge then ruled that even if the warrant had been limited to appropriate data types—here, video and photos—the warrant would have been unconstitutional:

With respect to the categories of data for which the police did have probable cause-the photos and the videos-the warrant was not as particular as could be reasonably expected given the nature of the crime and the information the police possessed. The warrant merely described the category of data, rather than specific items, which allowed the police to seize all of the photos and videos on Winn's phone, as opposed to specific photos and videos. As the Supreme Court explained, a sixteen-gigabyte smart phone like Winn's can hold thousands of pictures and hundreds of videos dating back "to the purchase of the phone, or even earlier." Riley, 134 S.Ct. at 2489, 2490. This is not a case where the police needed to browse through hundreds of photos and videos to find what they were looking for because Detective Lambert knew the precise identity and content of the photos/videos sought. However, he failed to mention the characteristics that made those photos/videos easily identifiable and distinguished them from others that were irrelevant. For example, the warrant could have described the location of the incident as well as the subjects of the images-children at a swimming pool, or more specifically young girls in swimsuits at the Mascoutah Public Pool. See, e.g., Mann, 592 F.3d at 780-81 (where warrant authorized police to search for "images of women in locker rooms or other private areas" for evidence of voyeurism).

Most importantly, the warrant should have specified the relevant time frame. The alleged criminal activity took place on one day only-June 18, 2014-and the police were looking for photos or videos taken that same day. There was nothing in the complaint indicating that any data created prior to that date was connected to the suspected public indecency.

I gather the key to this case is the very limited purpose of the search. The witnesses saw the defendant take pictures while rubbing himself, and the warrant was for evidence of public indecency. The only likely evidence of this specific crime on the defendant's phone was really indirect—obtaining photos or videos of girls taken in public on the day in question that would help to corroborate the claims that he was in public and being indecent while taking the photographs or videos. The narrow role of the photos in the factual context of the case explains the judge's view that the warrants had to be for those specific photos and not just general evidence of "public indecency" in whatever form it came.

It would be a different case if the government had probable cause to believe that Winn had child pornography on his phone. In the child pornography context, probable cause to believe there is one image of child pornography on a computer justifies a search through the entirety of the computer for that and other images of child pornography. The search for digital contraband can be very broad, just like a search for illegal drugs doesn't need to be limited to a specific package of drugs.