Commentary on DOJ’s Brief in United States v. Jones

By Orin Kerr
on Aug 12, 2011 at 10:27 am

The Department of Justice has recently filed its brief in United States v. Jones, the pending case on whether installing and using a GPS device to detect the location of a suspect’s car without a warrant violates the Fourth Amendment. DOJ brief’s provides an interesting example of how Fourth Amendment arguments are constructed, which gives us clues about how the Justices might approach the Jones case.

I. The Four Models of Fourth Amendment Protection

The Supreme Court has said that whether government conduct is a Fourth Amendment “search” depends on whether it violates a person’s “reasonable expectation of privacy.”  But the Supreme Court has struggled to determine what makes an expectation of privacy “reasonable.”  A few years ago, I wrote an article trying to explain why the Court’s cases have such a hard time explaining this issue.  See Orin S. Kerr, Four Models of Fourth Amendment Protection, 60 Stan. L. Rev. 503 (2007). My argument was that the “reasonable expectation of privacy” test was really four different tests — that there were four different kinds of arguments that the Supreme Court made about when the government violates a reasonable expectation of privacy and therefore is a Fourth Amendment “search.” Here are the four models I identified:

(1) the probabilistic model, by which an expectation of privacy is reasonable or not based on how common or expected the government’s conduct appears to be;
(2) the private facts model, by which an expectation of privacy is reasonable or not based on the nature of the information disclosed regardless of how it is obtained;
(3) the positive law model, by which an expectation of privacy is reasonable or not based on whether the government’s conduct violated some source of law outside the Fourth Amendment; and
(4) the policy model, by which an expectation of privacy is reasonable or not based whether it is desirable to have that outcome as a matter of policy.

To be clear, the Justices don’t label these arguments as I have. Rather, the opinions use these different arguments at different times when arguing that the Fourth Amendment either applies or does not apply to government conduct.

My article argues that the Supreme Court uses all four models instead of just one model because no one model accurately distinguishes less intrusive from more intrusive law enforcement steps, which is what the reasonable expectation of privacy test needs to do. Briefs and opinions therefore tend to mix and match the models, making arguments using the model or models that support the argued-for result in that particular context.

The result is rather messy, and gives all sorts of room to the Justices to justify different results.  But it also enables a workable regulatory system that gives substantial guidance to the lower courts. The law ends up relying on some models in some contexts and other models in other contexts, ideally in ways that pick models based on what models best track the invasiveness of the government’s conduct in that factual setting.

II. The Government’s Merits Brief in United States v. Jones

Now let’s turn to United States v. Jones. In Jones, the police attached a GPS tracking device to the suspect’s vehicle and tracked the vehicle’s location on public streets over time. The first question raised by the case is whether the use of the GPS device violated the suspect’s “reasonable expectation of privacy.” DOJ’s argument that the answer is “no” provides a helpful case study of how arguments based on the four models tend to work.

The government’s main argument is based on the private facts model. In prior precedents, the brief explains, the Court has said that location information about a car on public streets is not protected. In other words, the fact that a car is in a particular place on a public street at a particular time is simply not a private fact. The same approach should govern here, the brief contends. What matters is the information obtained, and here the information obtained was not private for Fourth Amendment purposes. See pages 17-22.

The government’s brief next argues that the Supreme Court should reject the probabilistic model. The most relevant cases have rejected the probabilistic model, the brief explains (citing Knotts and Smith), while the cases that embraced the probabilistic model (such as Greenwood and Bond) are very different on their facts. The Court should therefore ignore what the DOJ brief calls “the likelihood approach” to the Fourth Amendment in this case. See pages 22-27.

After a foray into why the mosaic theory won’t work, see pages 27-33, the DOJ brief then argues using the policy model. There has been no evidence of abuses of GPS surveillance, the brief argues. The legislature is ready to regulate this area in the absence of Fourth Amendment protection. As a result, the brief argues, there’s no need to subject GPS surveillance to a warrant requirement as a matter of policy. See pages 33-37.

In a long footnote, the DOJ brief then argues against using the positive law model here. Although some state laws make it unlawful to install a tracking device without a warrant, such laws should not influence whether an expectation of privacy is reasonable. See pages 36-37 n.4.

The brief then concludes this part of the argument by returning to the private facts model. The “type of information” that the GPS device revealed was not private, and therefore there was no violation of a reasonable expectation of privacy. See pages 37-39.

III. Guessing the Arguments of the Respondent’s Brief

The Four Models make it possible to offer some educated guesses about what the Respondent’s brief might look like.

First, the Respondent’s brief probably will press the models that the DOJ brief rejected. The Respondent presumably will press the probabilistic model and argue that reasonable people don’t expect this sort of surveillance. Second, the brief may advocate the positive law model: It will detail the laws requiring warrants for tracking devices and explain how such an interference with property is a traditional trespass. For both reasons, the brief will argue, the use of a GPS device violates a reasonable expectation of privacy.

The Respondent’s brief may also explain how the private facts and policy models cut in its favor, not the government’s. The brief could take a macro view of the information obtained in order to use the private facts model. That is, the brief will look at the information obtainable in the aggregate and argue that the aggregate facts are private under a private facts approach.) Finally, the Respondent’s brief will almost certainly use the policy model to offer a very different picture of the policy implications of allowing GPS to go unregulated.

In short, the two sides will end up offering dueling sets of models. Each side will press some of the models and de-emphasize or reject others, leaving it for the Justices to figure out which models should govern and how they should apply in this particular case.