EX ANTE REGULATION OF COMPUTER SEARCH AND SEIZURE
Orin S. Kerr
96 Va. L. Rev. 1241
IMAGINE you are a federal magistrate judge. The FBI comes to you with a warrant application to search a suspect's home and seize his computers. You review the application and confirm that it satisfies the Fourth Amendment and the Federal Rules of Criminal Procedure. The affidavit establishes probable cause, the warrant particularly describes the items to be seized, and the application is made by a federal agent to search property located in your district.
But there's a catch. The Fourth Amendment regulates both the issuance of warrants and their execution. Although the application satisfies the legal standards for issuing a warrant, you believe that the agents likely will violate the Fourth Amendment when they search the suspect's home, seize his computers, and then search the computers for evidence. You worry that the agents will grab more computers than they need and then look through the seized computers in an overly invasive way. You want the FBI to execute the warrant lawfully, so you consider imposing restrictions on how the warrant is executed to ensure it will be executed in a reasonable and therefore constitutional way.
The law of computer search and seizure remains undeveloped, so the ideal restrictions are uncertain. But perhaps you will sign the warrant only if the FBI agents agree to use a particular search protocol. Or perhaps you will sign the warrant only if the government agrees to minimize the seizure of the suspect's hardware or if the government agrees to waive any rights to seize any items discovered in plain view outside the warrant. Whatever restriction you choose, the goal is to protect the Fourth Amendment ex ante: you can best protect the Fourth Amendment by imposing conditions when you issue the warrant on how it later will be executed.
In the last decade, many federal magistrate judges have embraced this practice. Restrictions have varied widely from circuit to circuit and judge to judge, but magistrate judges generally have tried four kinds of limitations in computer warrant cases: first, conditions limiting the seizure of computer hardware from the physical place where the warrant is executed; second, conditions restricting the time period before seized computers are electronically searched; third, restrictions on how the computers are searched to limit access to evidence outside the warrant; and fourth, conditions on when the seized hardware must be returned.
Ex ante limitations on computer warrants recently received an enthusiastic endorsement by the en banc Ninth Circuit in United States v. Comprehensive Drug Testing, a case about searching a computer file for records of steroid use in professional baseball. After announcing a series of restrictions for magistrate judges in the Ninth Circuit to impose on the execution of all computer warrants, Chief Judge Kozinski celebrated the importance of ex ante restrictions on warrants to safeguard Fourth Amendment protections:
[W]e must rely on the good sense and vigilance of our magistrate judges, who are in the front line of preserving the constitutional freedoms of our citizens while assisting the government in its legitimate efforts to prosecute criminal activity. Nothing we could say would substitute for the sound judgment that judicial officers must exercise in striking this delicate balance.
The practice of conditioning computer warrants on how they are executed is a significant development in Fourth Amendment law. Many magistrate judges have embraced the practice as the best way to deal with the new dynamics of computer searches and seizures. Some scholars have agreed, envisioning such practices as important ways to limit computer searches.
But is this new practice legal? And is it wise? That is, do magistrate judges have the constitutional authority to condition the issuance of warrants on how the warrants will be executed? And to the extent the legality of the practice remains open, are such restrictions wise as a matter of policy? Are such restrictions helpful tools for magistrate judges on “the front line of preserving [our] constitutional freedoms,” as Judge Kozinski claims, or are they misguided limitations that may backfire in practice? The scholarly literature has not yet considered these questions closely. Despite their importance, these restrictions are so new and, as magistrate judge practices, so hidden from view that legal scholarship has yet to focus on them.
This Article shines a light on the new practices, and it then argues that ex ante regulation of computer warrants is both constitutionally unauthorized and unwise. The restrictions are unauthorized because the Fourth Amendment contemplates a narrow role for magistrate judges. Magistrate judges have no inherent power to limit how warrants are executed beyond establishing the particularity of the place to be searched and the property to be seized. When magistrate judges do impose restrictions on how a warrant is executed, those restrictions have no legal effect. The constitutionality of the search must hinge on whether the search was reasonable as judged ex post, not on whether the government complied with restrictions imposed by the issuing magistrate judge ex ante.
Ex ante regulation is also unworkable and counterproductive. Predictions of reasonableness are highly error-prone, as they are made in brief ex parte proceedings with few facts. In this setting, ex ante restrictions prohibit reasonable steps ruled out by judicial error. The scope of ex ante error will be a function of constitutional uncertainty: the more unclear the relevant legal rules, the more uncertain will be the restrictions needed to ensure reasonableness. Conversely, as the law becomes clear, predictive errors will decrease. As the law of reasonableness develops for computer searches, however, ex ante restrictions also become useless. The police will follow the rules because they know they will be imposed ex post, without a need for ex ante restrictions. From this perspective, the perceived need for ex ante restrictions is simply a response to the present legal uncertainty of computer search and seizure law.
* * *
I. Four Types of Ex Ante Conditions on the Execution of Computer Warrants
* * *
II. The Supreme Court and The Role of Magistrate Judges in Issuing Warrants
Does the Fourth Amendment law allow judges to impose restrictions on the execution of computer warrants to ensure that they are executed in a reasonable way? This Part reviews the Supreme Court case law on the role of magistrate judges in the execution of search warrants. These cases are rarely addressed in Fourth Amendment scholarship. That scholarship tends to focus on sexy questions like what is a search, or how specific exceptions should or should not apply. In contrast, there is very little scholarship on the role of magistrate judges in issuing warrants.
A review of that case law indicates that existing Fourth Amendment doctrine contemplates a surprisingly narrow role for magistrate judges. Magistrate judges are required to assess probable cause and particularity, and to comply with other rules imposed under state or federal statutory law. But magistrate judges appear to have no inherent power to limit how warrants are executed in the name of constitutional reasonableness. The reasonableness of executing warrants must be determined by judicial review ex post rather than ex ante. When a magistrate does impose an ex ante restriction on a warrant, the restriction has no legal effect. The constitutionality of the search hinges on whether the search was reasonable as judged ex post, not whether the government complied with restrictions imposed by the magistrate ex ante.
Finally, the narrow role suggested in the Supreme Court's decisions is reflected in statutory warrant authorities. Warrants are regulated by statutory law in addition to the Fourth Amendment. The federal search warrant statute and most analogous state statutes use language that denies judges the power to reject warrant applications based on how they are executed. The statutory authorities state that judges “must” issue warrants when probable cause has been satisfied. The combination of Fourth Amendment case law and statutory text strongly suggests that magistrate judges are acting outside their proper authority in imposing ex ante restrictions, and that such restrictions are unenforceable and have no legal effect.
A. Lo-Ji Sales v. New York and the Requirement Not to Act as an “Adjunct Law Enforcement Officer”
The Supreme Court's leading case on the limited role of magistrate judges in the execution of a warrant is probably Lo-Ji Sales v. New York. Lo-Ji Sales presents the extreme case of a magistrate judge controlling the execution of the warrant by participating in the search. Although it involves an extreme case, the Supreme Court's harsh rejection of the judge's creative role in executing the warrant suggests a narrow role for magistrate judges.
Lo-Ji Sales involved a search of a bookstore for obscene materials. A police officer had purchased two obscene films from an adult bookstore, and he approached a local magistrate for a warrant to search the store. The officer wanted to seize obscene films beyond the two that he already purchased and viewed, but he did not know which of the other films satisfied the constitutional test for obscenity. He therefore asked the magistrate to help him execute the search. Under the officer's proposal, the magistrate would accompany the officers to the store with a warrant that initially listed only the two known films as items that that the government could seize. The magistrate judge would review the additional materials himself onsite, and then tell the officers which films they could seize based on the magistrate's judgment about what was obscene. The magistrate agreed, and he came to the search site and pre-approved which films counted as obscene and therefore were seizable. After the magistrate approved the seizure of a particular film, he would add the name of the film to the warrant and the government would seize the film.
The Supreme Court unanimously rejected this unusual arrangement. The open-ended warrant did not specify what could be seized, and the magistrate's on-site control hurt rather than helped matters. Specifically, the magistrate's participation in the execution of the warrant violated the Fourth Amendment's requirement of a neutral and detached magistrate. According to the Court, the magistrate had “telescope[d] the processes of the application for a warrant, the issuance of the warrant, and its execution.” As a result, the magistrate “allowed himself to become a member, if not the leader, of the search party which was essentially a police operation.” There was no bad faith on the magistrate's part, the Court made clear: the magistrate was simply trying to make sure that a neutral officer made the judgment as to how to execute the warrant. But by combining the issuance and execution of the warrant, the magistrate “was not acting as a judicial officer but as an adjunct law enforcement officer.”
Lo-Ji Sales significantly undercuts the rationale of ex ante limitations on computer warrants. Like the judge in Lo-Ji Sales, a magistrate judge who tries to control the execution of the warrant to ensure it is reasonable “telescope[s] the processes of the application for a warrant, the issuance of the warrant, and its execution.” Ex ante restrictions turn the warrant application process into something much more than a check on probable cause and particularity. The restrictions try to regulate the entire process all at once. It's true that ex ante restrictions do not require the judge to be present when the warrant is executed, unlike in Lo-Ji Sales. But they do impose a virtual presence: the judge ensures that his own judgment as to how to execute the warrant will control the execution of the search much like the magistrate controlled the execution of the search in Lo-Ji Sales.
To be clear, the procedure rejected in Lo-Ji Sales differs from the imposition of ex ante search restrictions on computer warrants in a critical respect. The judge in Lo-Ji Sales was making impromptu judgments on what to seize, not how to search. But that difference cuts against the lawfulness of ex ante search restrictions, not in its favor. Determining what items the police have probable cause to seize is a core traditional function of magistrate judges reviewing warrant applications. A warrant is a judicial order allowing the government to enter the place to be searched to seize the items the judge has allowed to be seized. In contrast, how a warrant should be executed is traditionally a question for the police rather than the judge issuing the warrant.
B. Dalia v. United States and the Standard of Ex Post Review
The second case on the role of judicial review in the execution of warrants is Dalia v. United States. In Dalia, the FBI obtained a warrant under the Wiretap Act to use a surveillance device to listen in on the office conversations of a suspect named Dalia, who was believed to be engaged in a conspiracy to steal property. The warrant permitted agents to use the bug, but it did not say anything about how the government was supposed to install it. In the case of a traditional physical search, of course, such an extra step would be unnecessary. Government agents execute a search and seizure all at once by entering the property and removing the information described. Bugging is different. Like a computer search, bugging a private space occurs in two stages instead of one. The Government first enters the property to install the device, and it later turns on the device to listen in on the conversations. The FBI wiretap order in Dalia did not mention the first stage, however. By its terms, it only authorized the second stage.
The FBI executed the warrant by covertly entering Dalia's office at midnight and spending three hours installing a listening device in the ceiling. They later turned on the listening device pursuant to the warrant. When the warrant expired, the agents re-entered the office covertly and removed the bug. The government tried to prove Dalia's crimes using the recordings of Dalia's private conversations. Dalia responded by filing a motion to suppress based on the warrant's failure to authorize the physical search. Dalia made a range of arguments, two of which are important here. First, Dalia argued that that the Fourth Amendment required the warrant to say that it would be executed by means of a covert entry. Second, Dalia argued that that electronic surveillance warrants were “unique” and required special treatment because they involved two different sets of interests: the invasion of the physical space to install the device and the invasion of privacy to listen in on the conversations.
The Supreme Court disagreed. First, the Court rejected the notion that the warrant was defective because it did not explain how it could be executed. “Nothing in the language of the Constitution or in this Court's decisions interpreting that language suggests that . . . warrants . . . must include a specification of the precise manner in which they are to be executed,” the Court explained. “On the contrary, it is generally left to the discretion of the executing officers to determine the details of how best to proceed with the performance of a search authorized by warrant--subject of course to the general Fourth Amendment protection ‘against unreasonable searches and seizures.”’
Second, the Court rejected the notion that warrants for electronic surveillance were unique because their execution required two distinct stages, the entry and the listening, each of which raised different interests. According to the Court, “This view of the Warrant Clause parses too finely the interests protected by the Fourth Amendment.”
Often in executing a warrant the police may find it necessary to interfere with privacy rights not explicitly considered by the judge who issued the warrant. For example, police executing an arrest warrant commonly find it necessary to enter the suspect's home in order to take him into custody, and they thereby impinge on both privacy and freedom of movement. . . . Similarly, officers executing search warrants on occasion must damage property in order to perform their duty. . . .
It would extend the Warrant Clause to the extreme to require that, whenever it is reasonably likely that Fourth Amendment rights may be affected in more than one way, the court must set forth precisely the procedures to be followed by the executing officers. Such an interpretation is unnecessary, as we have held--and the Government concedes--that the manner in which a warrant is executed is subject to later judicial review as to its reasonableness.
Dalia has direct relevance for computer searches because the case for ex ante restrictions on computer warrants resembles the claim for special treatment of bugging warrants. In both cases, the warrants are executed in two stages. Physical entry comes first, followed by a search of the space second. In both cases, the breakdown of the warrant process into two stages raises special privacy issues. And yet Dalia rejected the argument that the two-stage warrant requires ex ante approval of how the warrant was to be executed. Instead, the Court concluded that the execution of bugging warrants should be judged using the same ex post review for reasonableness that occurs with traditional warrants.
To be fair, Dalia's usefulness is limited by the nature of the defendant's claim in that case: Dalia argued that a restriction on the method of executing the warrant was required, not that it was permitted. Further, Dalia was seeking only a very bare-bones statement as to how the physical search would be conducted: as Justice Brennan explained in his dissent, a blanket statement that covert entry was permitted would have been sufficient. In contrast, the question here is whether ex ante restrictions as to the method of execution are permitted, and the restrictions at issue can be quite detailed. At the same time, the Court's rejection of the idea that two-stage warrants raise special concerns, and its emphasis on the general practice of having ex post rather than ex ante review, seems to undercut the rationale of ex ante search restrictions on computer warrants.
C. United States v. Grubbs and the Plain Text of the Fourth Amendment
The third important case on the role of magistrate judges in issuing warrants is United States v. Grubbs. Grubbs reviewed a Ninth Circuit decision on the requirements of anticipatory warrants. Anticipatory warrants are warrants based on probable cause to believe that evidence will be in a particular place in the future even though the evidence is not there when the warrant is signed. Anticipatory warrants are premised on the idea that at a future time, some event will happen that will bring the evidence to the place to be searched; at that time the warrant can be executed and the place searched. Such warrants are used mostly in narcotics cases to allow searches when drug deliveries are accepted.
The Ninth Circuit decision under review in Grubbs invalidated an anticipatory warrant because it did not contain a particular description of the triggering condition--that is, the event that would happen to bring the evidence to the place to be searched. Under Ninth Circuit precedent, anticipatory warrants were required to have a particular description of the triggering condition in order to limit the government's discretion on when the warrant could be executed. That case law instructed magistrate judges to be “particularly vigilant in ensuring that the opportunities for exercising unfettered discretion are eliminated . . . [by] set[ting] conditions governing an anticipatory warrant that are ‘explicit, clear, and narrowly drawn so as to avoid misunderstanding or manipulation by government agents.”’
The Supreme Court reversed in an opinion by Justice Scalia. Justice Scalia reasoned that the plain text of the Fourth Amendment did not require anything beyond a particular description of the place to be searched and the property to be seized: “The language of the Fourth Amendment . . . does not include the conditions precedent to execution of the warrant.” Scalia also rejected the argument that a particular description of the triggering condition was needed to limit the government's discretion in executing the warrant. “That principle is not to be found in the Constitution,” Scalia intoned. “The Fourth Amendment does not require that the warrant set forth the magistrate's basis for finding probable cause, even though probable cause is the quintessential ‘precondition to the valid exercise of executive power.’ Much less does it require description of a triggering condition.”
The relevance of Grubbs lies in the similarity between the rationale for ex ante restrictions on anticipatory warrants and the rationale for ex ante restrictions on computer warrants. The Ninth Circuit had required its magistrate judges to pre-approve the triggering condition to limit the government's discretion in how it executed the warrant: the purpose was to “ensur[e] that the opportunities for exercising unfettered discretion [we]re eliminated.” This rationale had no takers on the Supreme Court. The Court adhered instead to the plain text of the Fourth Amendment, which requires only a particular description of the place to be searched and the items to be seized. According to the Court, the extra requirement of preapproval of the triggering condition in the warrant itself was not permitted. The requirement simply was “not to be found in the Constitution.”
D. Richards v. Wisconsin and the Legal Effect of Ex Ante Restrictions
The combination of Lo-Ji Sales, Dalia, and Grubbs suggests that magistrate judges should not impose ex ante restrictions on warrants to ensure they are executed in a reasonable way. Some judges appear to be doing it anyway, however, which raises an important question: What are the legal consequences if the government breaches a magistrate's ex ante restriction?
This issue arose in Richards v. Wisconsin, and the Court's decision indicates that the answer is “none.” Richards concerned the Fourth Amendment's knock-and-announce rule, a rule that agents executing a warrant normally must knock and announce their presence as police officers before entering a home. Under an exception to the rule, agents can dispense with the rule if they have a reasonable suspicion to believe that knocking and announcing their presence would be dangerous, futile, or inhibit the investigation.
In Richards, agents sought a warrant to search a hotel room for drugs based on suspicion that Richards was selling drugs from inside it. The agents asked the judge to sign a warrant permitting them to execute it without first knocking and announcing their presence. The magistrate judge found probable cause and signed the warrant, but he expressly rejected the request to dispense with the knock-and-announce requirement by crossing out that part of the warrant. When the agents executed the warrant, however, they did not announce their presence. The discovery of cocaine and cash in the hotel room led to charges, and the defendant moved to suppress on the ground that the agents had violated the knock-and-announce rule. Further, the agents had expressly violated the magistrate judge's will: the judge had declined to let the officers dispense with the requirement but they had done so anyway.
The Supreme Court disagreed in a unanimous opinion by Justice Stevens. According to Justice Stevens, the magistrate's refusal to allow a no-knock warrant had no effect. The refusal did “not alter the reasonableness of the officers' decision” to execute the warrant without knocking and announcing. The reasonableness of entering without announcing their presence “must be evaluated as of the time they entered the motel room,” Justice Stevens reasoned, based on the “actual circumstances” the officers confronted. The issuing magistrate could not know these circumstances ex ante, as he “could not have anticipated in every particular the circumstances that would confront the officers when they arrived at Richards' motel room.” Although the officers did not have evidence to execute a no-knock warrant when the warrant was obtained, at least “in the judgment of the Magistrate,” what mattered was whether they had that evidence at the moment they entered. “[A] magistrate's decision not to authorize a no-knock entry should not be interpreted to remove the officers' authority to exercise independent judgment concerning the wisdom of a no-knock entry at the time the warrant is being executed.”
Viewed in isolation, individual cases like Lo-Ji Sales, Dalia, Grubbs, and Richards do not definitively rule out the lawfulness of ex ante restrictions on the execution of computer warrants. Taken together, however, these four cases undercut every aspect of the lawfulness of such restrictions. Lo-Ji Sales requires magistrates to take a hands-off approach to executing warrants; Dalia rejects the notion that two-stage warrants raise special concerns that justify a deviation from the normal rule; Grubbs rejects limitations designed to ensure the reasonableness of searches; and Richards indicates that such restrictions, where imposed, have no legal effect. All four cases emphasize that the reasonableness of a search pursuant to a warrant must be assessed ex post rather than ex ante. Taken together, these four cases point to the conclusion that the Fourth Amendment does not permit ex ante restrictions on the execution of computer warrants. Where such restrictions are imposed, they have no legal effect.
* * *
III. The Normative Case Against Ex Ante Restrictions for Computer Warrants
Even assuming that ex ante restrictions are lawful, are they good policy? Proponents of ex ante restrictions, both in the judiciary and the academy, reason that ex ante restrictions are a powerful tool for protecting Fourth Amendment interests. As Judge Kozinski argued in United States v. Comprehensive Drug Testing, magistrate judges are on the front lines of the Fourth Amendment:
[W]e must rely on the good sense and vigilance of our magistrate judges, who are in the front line of preserving the constitutional freedoms of our citizens while assisting the government in its legitimate efforts to prosecute criminal activity. Nothing we could say would substitute for the sound judgment that judicial officers must exercise in striking this delicate balance.
Is Judge Kozinski right? Are ex ante restrictions on the execution of computer warrants a sensible way to balance privacy and security interests in the new world of computer searches and seizures?
The proper answer is “no.” Ex ante restrictions are unworkable and unwise for two core reasons. First, the combination of error-prone ex ante judicial review and more accurate ex post judicial review will result in systematic constitutional error. Instead of requiring reasonableness, ex ante review will result in reasonable steps being prohibited by judicial error. The likelihood of error will be a function of constitutional uncertainty. The more unclear the relevant legal rules, the more uncertain will be the restrictions needed to ensure reasonableness. However, as the law of reasonableness becomes clear, ex ante restrictions also become useless: the police will follow the rules because they know they will be imposed ex post, without a need for ex ante restrictions. From this perspective, the perceived need for ex ante restrictions is merely a response to present legal uncertainty.
Of course, it is better to prohibit unreasonable searches ex ante than invalidate them ex post while the law remains uncertain. Perhaps this carves out a role for ex ante restrictions, just as a placeholder until the law becomes settled? Again, the answer is “no.” The difficulty is that ex ante restrictions impair the ability of appellate courts and the Supreme Court to develop the law of unreasonable searches and seizures in the usual case-by-case fashion. Assuming ex ante restrictions are not null and void, they transform Fourth Amendment litigation away from an inquiry into reasonableness and towards an inquiry into compliance with the magistrate's commands. Search and seizure law cannot develop in this environment. For that reason, ex ante restrictions cannot be temporary measures used until the law becomes settled. Ironically, those measures will actually prevent the law from being settled.
* * *
B. Why Ex Ante Restrictions Introduce Constitutional Error
Ex ante restrictions tend to introduce constitutional errors in this environment. To be sure, such restrictions stem from the best of intentions: they reflect a good-faith effort to identify what will be constitutionally reasonable. However, ex ante predictions of reasonableness will be more error prone than ex post assessments for two major reasons. First, ex ante restrictions require courts to “slosh [their] way through the factbound morass of reasonableness” without actual facts. Second, ex ante restrictions are imposed in ex parte hearings without legal briefing or a hearing. Both reasons suggest that ex ante restrictions often will inaccurately gauge the reasonableness of how warrants are executed.
The major difficulty with ex ante restrictions is that the reasonableness of executing a warrant is highly factbound, and judges trying to impose ex ante restrictions generally will not know the facts needed to make an accurate judgment of reasonableness. Granted, magistrate judges might have a ballpark sense of the facts, from which they might derive a sense of what practices are ideal. For example, they might think that it is unreasonable to seize all of a suspect's home computers if on-site review is possible. Alternatively, they might think it is unreasonable to conduct a search for image files if the warrant only seeks data not likely to be stored as an image. They might think it is unreasonable to keep a suspect's computer for a very long period of time without searching it. All of these senses will be based on a rough concept of how the competing interests of law enforcement and privacy play out in typical computer searches and seizures.
At the same time, these ballpark senses of reasonableness can never improve past very rough approximation. A magistrate judge cannot get a sense of the exigencies that will unfold at each stage of the search process. The reasonableness of searching on-site will not be known until the agents arrive and determine how many computers are present, what operating systems they use, and how much memory they store. The needed time window before the government searches the seized computer depends on how much the government can prioritize that case over other cases, given existing forensic expertise and resources, as well as which agency happens to be working that case. The reasonableness of different search protocols depends on the operating systems, an analyst's expertise in forensics, which forensics programs the government has in its possession, what kind of evidence the government is searching for, and whether the suspect has taken any steps to hide it. Finally, the reasonableness of retaining seized computers that have already been searched depends on whether the government might need the original computer as evidence or whether it ends up containing contraband that should not be returned and is subject to civil forfeiture.
The magistrate presented with an application for a warrant simply cannot know these things. Judges are smart people, but they do not have crystal balls that let them predict the number and type of computers a suspect may have, the law enforcement priority of that particular case, the forensic expertise and toolkit of the examiner who will work on that case, whether the suspect has tried to hide evidence, and if so, how well, and what evidence or contraband the seized computers may contain. Magistrate judges can make ballpark guesses about these questions based on vague senses of what happens in typical cases. But even assuming they take the time to learn about the latest in law enforcement resources and the computer forensics process--enough to know about typical cases--they cannot do more than come up with general rules that they think are useful for those typical cases.
The errors of ex ante restrictions are particularly likely to occur because warrant applications are ex parte. The investigators go to the judge with an affidavit and a proposed warrant. The judge reads over the materials submitted. The judge can modify the warrant, but his primary decision is whether to sign or reject it. The entire process takes a matter of minutes from start to finish. No hearing occurs. There is no testimony beyond the affidavit in most cases, and the affidavit usually contains only standard language about computer searches. A prosecutor may be present, but need not be. Obviously, no representative of the suspect is present to offer witnesses or argument.
In that setting, judges are particularly poorly equipped to assess reasonableness. The most they can develop is a standard set of ex ante restrictions that they use in all computer warrants, perhaps one shared with other magistrate judges in their district. More careful scrutiny is both impractical and unlikely. The ability of a magistrate judge to assess reasonableness in that setting is a far cry from her ability to rule on reasonableness in an ex post hearing, in which agents and experts can take the stand and counsel for the defendant can cross-examine the agent, offer his own witnesses, submit written briefs, and present oral argument.
Some magistrate judges have implicitly recognized their factual and legal disadvantage by allowing decisions to be made later or allowing the government to petition for an amendment of restrictions. For example, in United States v. Brunette, the government was given thirty days to search seized computers but was then given a thirty-day extension upon request. In In the Matter of the Search of: 3817 W. West End, the warrant allowed the government to execute the physical search and then submit a proposed search protocol for the electronic search. In United States v. Olander, the warrant contained a condition formally delegating the task of reasonableness to the officers. The warrant formally required the government to examine the computer equipment during the physical search stage and then make a reasonableness judgment as to whether the computers needed to be seized. All of these conditions implicitly recognized that the magistrate judge could not accurately gauge reasonableness ex ante.
At the same time, all of the ex ante restrictions will necessarily be poor proxies for an ex post review of reasonableness. Instead of substituting for ex post review of reasonableness, ex ante restrictions supplement those restrictions. Ex ante limitations force the government to follow two sources of law: the reasonableness of executing the warrant imposed by reviewing courts ex post, and the restrictions imposed by the magistrate judge ex ante. If the ex ante restrictions happen to be modest, or are drafted in a way that ensures that they are always less than or equal to the restrictions of reasonableness ex post, then such restrictions will merely replicate the ex post reasonableness determinations. But every time an ex ante restriction goes beyond ex post reasonableness, the restrictions will end up prohibiting the government from doing that which is constitutionally reasonable. The limitations will be unreasonable limitations caused by judicial error.
* * *
D. Ex Ante Restrictions Prevent the Development of Ex Post Reasonableness
Ex ante restrictions only have an arguable role when appellate courts have not yet determined the rules of reasonableness. Does this mean that magistrate judges should continue to impose ex ante restrictions until the appellate courts and Supreme Court settle the rules? After all, it is much better for an unconstitutional search not to occur than for the Constitution to be violated and a court to then announce the violation. Should magistrate judges continue to impose such restrictions until that law of computer search and seizure becomes clear?
The proper answer is “no.” The reason is that ex ante restrictions themselves impair the ability of appellate courts and the Supreme Court to develop the law of reasonableness. Ex ante restrictions effectively delegate the Fourth Amendment to magistrate judges, transforming Fourth Amendment litigation away from an inquiry into reasonableness and towards an inquiry into compliance with the magistrate's commands. Search and seizure law cannot develop in this environment. Ex ante restrictions effectively deny courts an opportunity to announce the law in a de novo fashion. For that reason, ex ante restrictions cannot be temporary measures used until the law becomes settled. Ironically, those measures designed to further constitutional reasonableness will actually prevent the law of reasonableness from developing.
To understand why ex ante restrictions inhibit the development of legal standards of reasonableness, we need to return to how Fourth Amendment law develops. The trial judge holds a hearing, makes factual findings, and hands down a ruling. That ruling can then be appealed, which can then lead to an appellate decision. The appellate decision reviews the facts under a clearly erroneous standard and the law de novo, meaning that the appellate court reviews the lawfulness of the search and seizure without any deference to the magistrate judge who signed the warrant or the trial judge who made an initial legal ruling. When an appellate court reaches the merits of the lawfulness of the government conduct, the court renders a de novo ruling on reasonableness based on factual claims as asserted by the plaintiff or, more rarely, from a jury verdict. Rulings based on de novo legal conclusions can then be appealed up to the U.S. Supreme Court, which can hand down a decision articulating how the reasonableness of the Fourth Amendment applies to a particular set of facts.
This process does not work well with ex ante restrictions. When a magistrate imposes ex ante restrictions on a search warrant, and those restrictions are understood to be binding, the ex ante restrictions naturally become the focal point of litigation on the lawfulness of the warrant's execution. The restrictions provide a clear standard. Defense challenges to the lawfulness of the warrant's execution will point first to violations of that standard. Challenges will focus not on the reasonableness of the warrant execution, but rather the compliance or lack of compliance with the magistrate judge's restrictions. The cases on ex ante restrictions confirm the dynamic. In case after case, the litigation concerns whether the government complied with the magistrate's limitation, not whether the government's conduct was constitutionally reasonable.
This focus interferes with the usual process of Fourth Amendment rulemaking by effectively delegating the governing legal standard to individual magistrate judges. It denies appellate judges their usual means of establishing reasonableness by ensuring that appellate judges are not asked to review the reasonableness of the government's conduct. On appeal, the defendant will argue that the government should lose because law enforcement violated the ex ante restrictions: the focus becomes the ex ante restrictions, not constitutional reasonableness. Appellate courts will be in the position of deciding whether the government complied with the restrictions and the significance of those violations. The constitutional reasonableness of the government's conduct not only won't be decided, it may not even be briefed.
This dynamic arguably explains the extremely unusual Ninth Circuit en banc decision in Comprehensive Drug Testing. The Ninth Circuit's decision is a true blockbuster. If it stays on the books, it will revolutionize computer search and seizure law. But it is also one of the oddest decisions in the Federal Reporter. That oddness results in part from the fact that the litigation over the warrant executing in Comprehensive Drug Testing was almost exclusively about whether the FBI violated the ex ante restrictions, not whether its conduct was reasonable. After agreeing that the government had violated the restrictions, and that the violations justified a ruling for the plaintiffs, Chief Judge Kozinski then announced a comprehensive set of new ex ante restrictions for magistrate judges to impose.
The Comprehensive Drug Testing court did not say whether the rules handed down were based on the Fourth Amendment, the federal supervisory power, or something else. This likely resulted from the dynamic of the lower court litigation: because the litigation focused entirely on compliance with the ex ante restrictions, the Ninth Circuit's task was largely addressed to assessing compliance. The ex ante restrictions took on a life of their own. Instead of using the facts of the case to analyze the reasonableness of the government's searches and seizures, resulting in a rule or standard of reasonableness imposed ex post, the en banc Ninth Circuit simply announced new restrictions for magistrate judges to impose ex ante.
Ex ante limitations on the execution of computer warrants have arisen from the best of intentions. The magistrate judges who have devised such restrictions have acted out of a commendable effort to protect Fourth Amendment rights in light of a new world of computer search and seizure. The new facts of computer search and seizure change the basic facts of criminal investigations, and those changes trigger the need for new law. As the Ninth Circuit recognized in United States v. Adjani, “[a]s society grows ever more reliant on computers as a means of storing data and communicating, courts will be called upon to analyze novel legal issues and develop new rules within our well established Fourth Amendment jurisprudence.” Magistrate judges are on the front lines of the new world: they are seeing the changes before the rest of the judiciary.
Although such restrictions reflect the best of intentions, magistrate judges do not have the constitutional authority to impose limits on how warrants are executed to ensure that the resulting searches are reasonable. Where magistrate judges do impose restrictions on how warrants are executed, reviewing courts should recognize that these restrictions have no effect: while the executive branch often will follow such restrictions, it need not do so. Ex ante restrictions on the execution of warrants are also unwise. The factual vacuum of ex ante and ex parte decisionmaking leads such restrictions to introduce constitutional errors that inadvertently prohibit reasonable search and seizure practices. Further, ex ante restrictions prevent the development of ex post rules of reasonableness that appellate courts must create to account for the new environment of computer search and seizure.
In short, magistrate judges should stand down. They should cease placing conditions on the execution of computer warrants
, and they should instead let reviewing trial and appellate courts review the reasonableness of warrant execution ex post.