The Volokh Conspiracy

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


Peffer v. Stephens, on Probable Cause and Home Computer Searches

A computer search decision from a new Sixth Circuit judge, John Bush, has been generating some controversy. Let's take a look.


On Thursday, the Sixth Circuit decided a Fourth Amendment case that is drawing a lot of criticism online. The case, Peffer v. Stephens, is authored by a new and somewhat controversial Trump appointee, John K. Bush. A few people have asked me to take a look at the case, wondering if it's as crazy and extreme as some (okay, Slate's Mark Stern) say.

My tentative take: I think the court reached out to answer a big question it didn't have to answer. I'm skeptical that the court was right to paint with such a broad brush. With that said, I don't think the decision is as far-reaching or harmful as others seem to think, in part because I suspect future courts will limit it to its facts.

I. The Facts

This case is a civil suit challenging whether there was probable cause to issue a search warrant. The police suspected that evidence of impersonating an officer and witness intimidation would be found in Peffer's house, and they obtained a warrant to search Peffer's house for that evidence. Specifically, the police suspected that Peffer had authored a letter and a flier that was evidence of the suspected crimes, and they obtained a warrant to search for and seize records relating to the crimes including in electronic form. The warrant then says that in the course of searching for those items, the government may seize and then search any computers that might store the records in electronic form.

The officers executed the warrant and took away a bunch of computers. In the end, though, prosecutors declined to prosecute Peffer. Peffer and his wife filed a civil suit, claiming (among other things not relevant here) that the warrant lacked probable cause. In particular, the Peffers mainly argued that the affidavit did not establish probable cause that evidence of the crime would be in the house for two reasons. First, the affidavit did not provide reason to beileve that the evidence described was evidence of a crime based on then-existing state law. And second, the affidavit did not make the case that the evidence described would be found in the home.

II. The Sixth Circuit's Opinion

Judge Bush's opinion treats those two arguments separately. First, the opinion rules that it was not clearly established that that the letters were not evidence of a crime. As a result, the officer "would be protected by qualified immunity from liability for executing an otherwise valid search warrant seeking evidence that Mr. Peffer violated those criminal statutes."

Next, the opinion rules that there was probable cause to believe that the evidence described would be in the house. Much (but not all) of the argument, Judge Bush says, was based on the claim "that no assertion was made that Mr. Peffer owned either a computer or a printer or, if he did, that he kept those items at the . . . residence" that was searched. The affidavit argued that the letters and fliers were computer-generated, and likely written by Peffer, and therefore that there was probable cause to find evidence about them in the Peffer house. According to the Peffers, though, there was no reason to think that Mr. Peffer had a computer at home that stored the evidence.

Judge Bush rejected the Peffers' argument. This is the passage that has generated a lot of controversy, so I will present it in full with the footnoted material in brackets:

It appears to be a question of first impression in this circuit whether the nature of a computer is such that its use in a crime is alone sufficient to justify an inference that, because of "the nature of the things to be seized," ibid., evidence of the crime is likely to be found in the alleged criminal's residence. But this question is not a difficult one to answer based on basic principles.

As a general rule, it is reasonable, ceteris paribus, to assume that a person keeps his possessions where he resides. [FN9: See, e.g., United States v. Aljabari, 626 F.3d 940, 946 (7th Cir. 2010) ("When probable cause exists to believe an individual has committed a crime involving physical evidence, and when there is no articulable, non-speculative reason to believe that evidence of that crime was not or could not have been hidden in that individual's home, a magistrate will generally be justified in finding probable cause to search that individual's home.")].This presumption is of course rebuttable and cannot always be relied upon by a magistrate in finding a nexus between the object used in a crime and the alleged criminal's residence, because the "totality of circumstances presented" in the affidavit may suggest that the object is more likely to be found elsewhere or nowhere at all. Brown, 828 F.3d at 382. The affidavit may, for example, include evidence suggesting that the object was not owned by the alleged criminal; that it was discarded, sold, or was otherwise disposed of; that the alleged criminal, while retaining possession of the object, stores it elsewhere than his residence; or that the object no longer exists.

If an affidavit presents probable cause to believe that a crime has been committed by means of an object, however, a magistrate may presume that there is a nexus between that object and the suspect's current residence, unless the affidavit contains facts that may rebut that presumption. [FN10: As with all findings of probable cause, this presumption is subject to a staleness analysis. The Peffers did not argue that any probable cause established in the affidavit had gone stale.]

And although we have not articulated this presumption in precisely this manner, it underlies our previous decisions in cases analyzing the connection between the objects used in a crime and the alleged criminal's residence. Our jurisprudence in this area has not always been as clear as one might hope, but an analysis of several of our nexus-jurisprudence tributaries shows that it is not as muddled as one might fear.

When it comes to guns, because we "have acknowledged that individuals who own guns keep them at their homes," United States v. Smith, 182 F.3d 473, 480 (6th Cir. 1999), a suspect's use of a gun in the commission of a crime is sufficient to find a nexus between the gun that was used and the suspect's residence. For example, in United States v. Vanderweele, an informant told an ATF agent that he had seen Vanderweele in possession of a silencer at a clubhouse. 545 Fed.Appx. 465, 467 (6th Cir. 2013). The agent sought a search warrant for Vanderweele's residence based on nothing more than the informant's statement that Vanderweele had been in possession of a silencer and his awareness, "based on his training and experience, 'that firearms, ammunition, and related items are commonly stored within the owner or possessor's dwelling.' " Id. at 469. We held that based on these alleged facts, "[t]he magistrate judge had reason to believe that the silencer would be found at Vanderweele's house," and we upheld the warrant. Ibid.

This is consistent with our holdings in similar cases. [FN 11: It is also consistent with the approach taken by many, but not all, of our sister circuits. United States v. Anderson, 851 F.2d 727, 729 (4th Cir. 1988) ("It was reasonable for the magistrate to believe that the defendant's gun and the silencer would be found in his residence … even though the affidavit contained no facts that the weapons were located in defendant's trailer."); United States v. Steeves, 525 F.2d 33, 38 (8th Cir. 1975) (upholding a warrant on the basis that "people who own pistols generally keep them at home or on their persons"); United States v. Rahn, 511 F.2d 290, 293–94 (10th Cir. 1975) (upholding a warrant because "it is reasonable to assume that [defendant's] house was where he kept things and it is pretty normal … for individuals to keep weapons in their homes"); Bastida v. Henderson, 487 F.2d 860, 863 (5th Cir. 1973) (upholding a warrant because a "reliable informant stated that on the day after the robbery [Defendants] were still armed with automatic pistols" and "[a] very likely place to find them thereafter would either be on the persons of the assailants or about the premises where they lived"). But see United States v. Charest, 602 F.2d 1015, 1017 (1st Cir. 1979) (invalidating a warrant because there was "nothing in the affidavit from which a factual finding could be made that the gun used in the shooting was probably located at defendant's premises" and "[c]ommon sense tells us that it is unlikely that a murderer would hide in his own home a gun used to shoot someone").] See, e.g., United States v. Goodwin, 552 Fed.Appx. 541, 546 (6th Cir. 2014)(upholding a warrant to search illegal-gun purchaser's residence where an affidavit established that the gun sought was valuable and that "owners usually keep machine guns in their homes"); United States v. Cobb, 397 Fed.Appx. 128, 133 (6th Cir. 2010) (upholding a warrant to search bank robber's residence for the clothing and gun he had used during a robbery because the "reasonable inference is that this clothing and gun likely would have been in [defendant's] possession six weeks following the final robbery"). [FN12: This area of our jurisprudence is admittedly murky. In United States v. Bethal, the panel declined to find a nexus between the gun the defendant used in a shooting and his residence, despite the affidavit's asserting that the defendant had been "identified as one of the drive-by shooters." 245 Fed.Appx. 460, 469 (6th Cir. 2007). Over a dissent, the panel reasoned that because "persons accused of murders often dispose of the guns utilized in the crime soon afterward" and "the affidavit … provided no indication that at the time of the search, [Defendant] was still participating in gang-related shootings, or was seen carrying a gun," the affidavit failed "to establish any relationship between [Defendant]'s residence and the fair probability that weapons and drugs would be found there." Id. at 468–69. We question Bethal's emphasis on ongoing criminal conduct and are not bound by this unreported decision. See United States v. Ennenga, 263 F.3d 499, 504 (6th Cir. 2001) ("We need not concern ourselves with any perceived inconsistency, however, because [inconsistent case] is an unpublished case and therefore not a controlling precedent.").]

It is clear that the use of a gun in the commission of a crime is sufficient to establish a nexus between the suspected criminal's gun and his residence. Computers are dissimilar to guns in many ways, including the nature of the crimes in which they are used and the relative ease with which guns can be transported and discarded. Computers are similar to guns, however, in that they are both personal possessions often kept in their owner's residence and therefore subject to the presumption that a nexus exists between an object used in a crime and the suspect's current residence. This is borne out by our cases involving the consumption of child pornography via computer.

Although we have never been asked to pass judgment on a magistrate's finding of a nexus between the computer used to consume child pornography and the alleged consumer's residence based on nothing more than the use of the computer, we have placed our imprimatur on a number of search warrants issued based on affidavits with scant evidence supporting a nexus beyond the use of a computer. See, e.g., United States v. Elbe, 774 F.3d 885, 890 (6th Cir. 2014) (finding that affidavit established nexus because the suspect's residence had high-speed internet and the suspect had been observed using a laptop on his front porch); United States v. Lapsins, 570 F.3d 758, 766 (6th Cir. 2009) (finding that affidavit established nexus because the IP address used to distribute prohibited material was accessed by a residential modem located in the general vicinity of suspect's residence and that suspect had participated in an online chat regarding prohibited material between the hours of 6:30 and 8:30 a.m., when suspect would be at home); United States v. Terry, 522 F.3d 645, 648 (6th Cir. 2008) (finding nexus because affidavit established that the suspect had a computer at his residence and had sent an email containing prohibited material at approximately 2:30 a.m.).

Just as guns, and other possessions, are generally kept in the home, so too are computers, and so we readily find a nexus between computers used in the consumption of child pornography and the suspected consumer's residence.

The principle that we are now articulating also explains why we are more reticent to find a nexus between drugs and their distributor's residences. In Brown, the defendant was apprehended leaving the location of a sale of more than 500 grams of heroin, and was found to be in possession of $4,813 in currency. 828 F.3d at 378–80. Responding to the Government's argument that "the magistrate judge was entitled to infer that evidence of drug trafficking would be found at Brown's residence because he was a known drug dealer," we pointed out that "we have never held … that a suspect's status as a drug dealer, standing alone, gives rise to a fair probability that drugs will be found in his home." Brown, 828 F.3d at 383. Recognizing that "[i]n the case of drug dealers, evidence is likely to be found where the dealers live," we nevertheless held that "if the affidavit fails to include facts that directly connect the residence with the suspected drug dealing activity, or the evidence of this connection is unreliable, it cannot be inferred that drugs will be found in the defendant's home—even if the defendant is a known drug dealer." Id. at 383–84.

This is because, unlike guns and computers that are used in the commission of a crime, when drugs are used in the commission of a distribution offense, the distributed drugs are no longer in the possession of the suspected distributor. The affidavit therefore must establish some other reason to believe that drugs or other evidence of crime would be found in the suspect's residence if searched. See, e.g., United States v. Raglin, 663 Fed.Appx. 409, 411–12 (6th Cir. 2016) (finding nexus because affidavit established that a suspect drove to the defendant's house directly after trafficking drugs, at which point $38,000 appeared in a purse on the roof and the defendant's girlfriend told officers that guns were in the house); United States v. Kenny, 505 F.3d 458, 461 (6th Cir. 2007) (finding probable cause because affidavit established that an informant identified the defendant as the "cook" of a large ongoing trafficking operation that was taking place on his property, which was corroborated by independent evidence); United States v. Newton, 389 F.3d 631, 636 (6th Cir. 2004) (explaining that affidavit established that defendant was engaged in a "continual and ongoing" drug distribution scheme), rev'd on other grounds, 546 U.S. 803, 126 S.Ct. 280, 163 L.Ed.2d 35 (2005); United States v. Jones, 159 F.3d 969, 974 (6th Cir. 1998) (finding that affidavit established that defendant was seen distributing narcotics directly outside of his domicile). Unlike drugs, guns and computers are objects that generally remain in the suspect's possession after commission of the crime, and therefore it is reasonable to believe those possessions to be stored at the suspect's residence, absent evidence to the contrary.

Here, the affidavit included allegations that Mr. Peffer had used a computer in the commission of his crime, that evidence of the crime would likely be found on that computer, and that Mr. Peffer resided at the Bierri Road residence, thereby establishing a presumption that evidence of the crime would be found at the Bierri Road residence. That the affidavit did not allege that Mr. Peffer owned a computer or that he kept one at the Bierri Road residence is immaterial, because the averment that he used one in the commission of a crime is sufficient to create the presumption that it would be found at his residence. See e.g., Vanderweele, 545 Fed.Appx. at 469. The affidavit did not suggest any reason to believe that Mr. Peffer had used a computer that did not belong to him, that he had thrown out or otherwise disposed of the computer, or that he kept the computer elsewhere. Indeed, the affidavit did not suggest any reason to believe that the computer used in the commission of the crime would not be found at the Bierri Road residence, and therefore the only reasonable conclusion that a jury could draw is that a nexus existed between the evidence sought and the Bierri Road residence.

Because Sergeant Stephens executed a valid warrant supported by probable cause as to a connection between the mailings and the Bierri Road residence, when he searched that house, there was no Fourth Amendment violation on this ground.

III. My Analysis

Here are some tentative thoughts.

First, I'm not sure it works to treat the Peffers' two probable cause arguments separately. The Peffers claimed that the warrant was defective because the conduct wasn't evidence of a crime and because that evidence wasn't going to be in the home. Judge Bush treats those as two separate claims. He first disposes of the scope-of-crime claim on qualified immunity grounds, and he then takes on the claim that the evidence wasn't sufficiently likely to be there. But I'm not sure they can be separated. Probable cause has to be assessed all at once, I would think. The question is, overall, whether evidence of crime is likely to be in the place to be searched. If it's not clear that the conduct alleged is a crime, then I would think that discounts the odds that there are evidence of crime in the place to be searched. Off the top of my head I can't think of cases where that issue has come up. But that's my instinct, at least.

Of course, the bigger controversy over the Peffer case is about the court's apparent conclusion that "the nature of a computer is such that its use in a crime is alone sufficient to justify an inference that, because of the nature of the things to be seized, evidence of the crime is likely to be found in the alleged criminal's residence." What about that?

Here are three thoughts about that in particular. First, skimming over the briefs, I am somewhat perplexed as to why the court tried to paint with such a broad brush. Here's the opening Peffer brief; here's the officer's brief; and here's the reply. I only skimmed them, so maybe I'm missing something. If so, I apologize. But based on my quick read, I don't see how they asked the court to take a position on whether computer-generated evidence is likely to be at a suspect's home. So one way this passage is odd is that it's not clear the parties really briefed it and the court had to reach it.

Second, I'm skeptical that it works to paint with such a broad brush as the court did. The category of "computers" and "computer-stored evidence" seems just too broad. In a world of global computer networks and the cloud, with many (most?) people having different electronic devices in different places used in different ways, it's hard to generalize about where computer-stored evidence is likely to be.

In this case, for example, the court was dealing with letters and fliers that presumably were generated with some kind of word processing software. I think it's probably the case these days that if someone is creating fake letters and fliers as part of criminal activity, and they're just regular computer users with no obvious sophistication, they're probably just doing it on a home computer. The home computer probably generated copies of the documents, either as automatic backups or maybe as a version saved by the user, that are there in some form on the machines. Probably.

But I think the point is about the specific kind of document, perhaps generated by someone with no computer sophistication. It's not about the fact that it was generated on a computer generally. If you change the nature of the document—make it, say, a threatening text message that is more likely to be on a phone the suspect carries around than on a computer always at home—then the odds change about where it could be stored. If you change what we know about the user—say, make him very technically sophisticated—the odds change again.

Given that, I wouldn't be surprised if future courts tend to limit Peffer to its facts. The likelihood that using a computer to commit a crime means that evidence of the crime will be found in a computer inside the person's home at some future point just seems too fact-dependent to justify a broad generalization of the answer. The nature of the probable cause determination calls for a more fact-specific holding, it seems to me.

Third, I think there's some interesting tension between Judge Bush's ruling and Judge Srinivasan's ruling for the D.C. Circuit in United States v. Griffith, 867 F.3d 1265 (D.C. Cir. 2017). Griffith opened:

Most of us nowadays carry a cell phone. And our phones frequently contain information chronicling our daily lives—where we go, whom we see, what we say to our friends, and the like. When a person is suspected of a crime, his phone thus can serve as a fruitful source of evidence, especially if he committed the offense in concert with others with whom he might communicate about it. Does this mean that, whenever officers have reason to suspect a person of involvement in a crime, they have probable cause to search his home for cell phones because he might own one and it might contain relevant evidence? That, in essence, is the central issue raised by this case.

Judge Srinivasan's opinion isn't in direct contrast with Judge Bush's, but it is skeptical about presuming that a computer is in a home. Srinivasan's opinion demands more evidence, both that a person has a device and that the device still is likely to have evidence on it. It's an interesting contrast with Judge Bush's opinion. I wouldn't be surpised if future lower court cases on this issue grapple with both precedents.