Wednesday, July 15, 2009

Copying as a Seizure (Again)

I’m going to revisit an issue I addressed in a post I did several years ago. The issue is whether copying data files is a seizure under the 4th Amendment.

As I’ve noted in earlier posts, the 4th Amendment prohibits unreasonable searches and seizures. As I’ve also noted, a “search” violates a reasonable expectation of privacy under the test the Supreme Court announced in Katz v. United States, 389 U.S. 347 (1967); and as the Supreme Court held in Soldal v. Cook County, 506 U.S. 56 (1992), a “seizure” interferes with our possession and use of our property.

I think the issue of whether or not copying data is a 4th Amendment seizure is an important one because if copying is neither a search (which I don’t think it is) nor a seizure, then it’s completely outside the scope of the 4th Amendment. If copying is completely outside the scope of the 4th Amendment, then officers can copy data without getting a warrant authorizing them to do so and/or relying on an exception to the warrant requirement as their authorization for doing so.

Why don’t I think copying is a search? As I noted, searches violate – intrude on – a reasonable expectation of privacy under the 4th Amendment. Let’s assume, for the sake of analysis, that someone has a legitimate 4th Amendment expectation of privacy in the data stored on their computer. To really reinforce that assumption, we’ll also assume that this person lives alone and so doesn’t share the computer with anyone else and doesn’t give anyone else access to it, whether in person or remotely. The contents of that hard drive are, therefore, protected by the 4th Amendment’s guarantee of privacy.

Assume a police officer equipped with the appropriate forensic software makes a copy of the hard drive. We’ll also assume the officer’s being in the home to make the copy didn’t itself violate the 4th Amendment because I want to focus on the specific act of copying the data. If, as I believe is usually true, the officer doesn’t observe the contents of the data during the copying process, then I do not see how we can characterize the copying as a search.

He hasn’t looked at the data; no human being has looked at it. The computer and software he’s using have, in a literal sense, “looked at it” because both have had some level of access to the data. I, however, do not see that as a true 4th Amendment search, if only because the 4th Amendment was clearly intended to protect the privacy of our places and things from observation by people (law enforcement officers specifically).

We could construe the act of copying the data as a search under at least two theories: One theory is the one I noted above, i.e., that the implements have in a sense “looked at” the data and we’ll impute their “observations” to the law enforcement officer. The other theory is that the copying by the equipment is essentially the first step toward this officer’s viewing the contents of the hard drive, so it is the beginning of a search. We could also have a third theory if and when the programs officers use to copy data have attained a level of artificial intelligence; at that point, we still wouldn’t have a human being observing the data but an entity with a level of intelligence would be doing so. We could then, I suppose, impute the artificial intelligence’s viewing the data to the officer.

I concede that copying data COULD be construed as a search under these, and perhaps other, theories. I really don’t think that’s the way to go, though, because I think we really have to torture the notion of “search” to apply it to the non-observational copying of data.

I think it makes much more sense to treat copying data as a seizure. Copying data is, of course, not a traditional, zero-sum seizure. A traditional, zero-sum seizure is analogous to traditional, zero-sum theft: In both, the possession and use of property passes entirely from one person (the rightful owner) to another (the officer seizing the property or the thief stealing it). Zero-sum seizures are the only kind of seizures that are possible with tangible property, i.e., property that exists only in the physical world.

Zero-sum seizures are therefore the only kind of seizures the drafters of the Bill of Rights were thinking about when they wrote the 4th Amendment. That, though, does not mean we have to limit the applicability of the 4th Amendment to zero-sum seizures. After a few false starts, the Supreme Court recognized that unless it construed the 4th Amendment broadly -- to encompass changing technologies -- the 4th Amendment would become increasingly irrelevant to modern life. Since the 4th Amendment is the closest thing the Constitution has to a guarantee of privacy and security in the possession of property, we do not want it to become a pretty-much-dead letter.

Expanding the traditional, zero-sum conception of seizure to encompass non-zero-sum seizures is consistent with the approach the Supreme Court took in holding that tapping phone conversations is a 4th Amendment “search.” In 1928, in Olmstead v. U.S., 277 U.S. 438, the Supreme Court held it was not a search for federal agents to use wiretaps on the phone lines outside Olmstead’s home to listen to what he said when he making calls from his home phone. (Olmstead had argued it was a search because the officers were able to hear what he said when he was in his home, the home being the most private of all 4th Amendment enclaves.)

Because many members of that Court were conceptually mired in the nineteenth century, they said the eavesdropping didn’t violate the 4th Amendment because the officers never physically entered Olmstead’s home. They were construing the 4th Amendment to reach only the evil it was originally designed to address: officers kicking down someone’s door, forcing themselves into the home and rummaging through the contents of the house. In his dissent, Justice Brandeis pointed out that

[s]ubtler and more far-reaching means of invading privacy have become available to the government. . . .The progress of science . . . is not likely to stop with wire tapping. Ways may . . . be developed by which the government, without removing papers from secret drawers, can reproduce them in court, and . . . expose to a jury the most intimate occurrences of the home. . . . Can it be that the Constitution affords no protection against such invasions of individual security?

Olmstead v. U.S., supra (Brandeis dissenting). In 1967, in the Katz case, the Supreme Court reversed its Olmstead decision and said wiretapping is a search. In so doing, the Court moved beyond a literal interpretation of the 4th Amendment and into one that can encompass advances in technology. I think we should do essentially the same thing with how we define 4th Amendment seizures.

At the moment, the only case I can find in which a judge specifically rules on the copying-as-seizure issue is U.S. v. Gorshkov, 2001 WL 1024026 (U.S. District Court for the Western District of Washington 2001). As you may know, in the Gorshkov case FBI agents copied data from a computer Gorshkov used in Russia, without first obtaining a search and seizure warrant. Gorshkov argued that copying the data constituted a 4th Amendment search, but the federal district judge disagreed:

[C]opying the data on the Russian computer was not a seizure under the Fourth Amendment because it did not interfere with Defendant's . . . possessory interest in the data. The data remained intact and unaltered. It remained accessible to Defendant. . . . The copying of the data had absolutely no impact on his possessory rights.

U.S. v. Gorshkov, supra. I vehemently disagree.

When officers copy data, a transfer takes place. Before officers copy the data on John Doe’s hard drive, Doe is the only person who had possession of it. After they copy it, both the officers and Doe have a copy of the data. Doe has not, as the Gorshkov judge correctly noted, entirely lost possession of the data. He has, I argue, lost a quantum of his possessory interest in the data.

In an earlier post, I wrote about an Oregon case in which the defendant was charged with using a computer to commit theft after he copied a password file belonging to his employer. The defendant claimed he didn’t commit theft because the employer still had the password file; he just had a copy of it. If we apply the Gorshkov judge’s approach to defining a seizure of property to this defendant’s argument, then he’d win; the Oregon court would have had to have dismissed the theft charge against him because the employer had not “entirely lost possession of the data.”

That, though, isn’t what the Oregon court did. It noted that “theft” is defined as taking someone’s property without being authorized to do so and that “property” is defined as “anything of value.” The court found that the evidence showed the passwords had “value,” which meant they were “property.” It then held that the defendant committed theft because he deprived his employer of exclusive possession of the passwords, which deprived the employer of property because much of the value of passwords lies in the fact that no one else knows what they are.

If copying data is theft, I think it also has to be a seizure, a non-zero sum seizure. Any time someone copies my data without permission, I “lose” something; more precisely, I lose part of my exclusive possession and control of the data. Even if my data doesn’t consist of passwords, much of its value lies in the fact that it’s mine and I, alone, control it. I think, therefore, that the loss of the ability to exercise sole control over one’s data justifies defining copying as a non-zero sum 4th Amendment seizure . . . which would not mean law enforcement officers couldn’t copy data. It would mean they’d need to have a warrant or an exception to the warrant requirement (e.g., consent or the existence of exigent circumstances) to be able to copy the data without violating the 4th Amendment.


1 comment:

Anonymous said...

Copying data is both search and seizure! Under your scenario, an officer could copy an entire hard drive and it would not be considered a search. Yet, what do you think the officer is going to do with the copy? If they truely intend to simply make a copy, but never look at its contents, this would not be a search, but an extremely bizarre seizure. What is the cop going to say in court? "See, this guy had data on his hard drive, he's clearly guilty. Who else keeps data on their hard drive but the guilty?" Sure, the cop has seized the data, but having data is not a criminal offense. It's the contents that can make it criminal. And the only way to establish the contents is look at, i.e. search them.

Too the same degree, how does an officer search, but not seize data later entered into evidence in court? She searches for specific files and makes note of what she finds, without actually copying the files. Yet, this would result in an absurd situation in which an officer was authorized to search for, but not seize, narrowly defined evidence of criminal activity. For example, the officer would be forced to testify that she found child pornography on the defendant's computer, but when asked to produce it, would have to say that the court must simply take her at her word without presenting any evidence. It's like saying "Your honor, we found 1000 kilos of crack, but we only looked at it. We didn't photograph it. We didn't seize it. We just made a note of it, arrested the defendant, and left the house never to return."

You write, "We could construe the act of copying the data as a search under at least two theories: One theory... is that the copying by the equipment is essentially the first step toward this officer’s viewing the contents of the hard drive, so it is the beginning of a search."

Isn't this the only logical way to construe copying data? Otherwise, the situation is the absurd ones that I described above. I am not a lawyer, so maybe I am missing the distinction you are making between the legal meaning of search and the legal meaning of seizure, but it seems to me that barring the absurd scenarios I described above, data cannot be either searched or seized, but must be both searched and seized if it is to be entered into evidence in court.

You argue that data can in fact be seized, but not searched, so I challenge you to give an example in which data that was seized, but not searched, could be used in a court case. How do you enter data, which you have never looked at, into evidence at a trial?