Diaz-Castaneda and the Possibility of Fourth Amendment Use Restrictions:
Yesterday, in United States v. Diaz-Castaneda, the Ninth Circuit had some interesting dicta in an otherwise-routine Fourth Amendment decision. The dicta touches on a topic that law professors have long been very interested in, so I thought I would flag it.

  The facts of the case are simple. A police officer saw a truck driving on the highway, observed its license plate number, and then ran a computer check of the plates and learned that the registered owner of the truck had a suspended license. The officer saw that the driver of the truck matched the description of the owner of the truck, so he pulled over the truck. The issue in the case was whether the officer's viewing the plates and running the computer check was a Fourth Amendment "search" of the driver that occurred without probable cause and was therefore unlawful.

  In an opinion by Judge Fisher, the court concluded that no search occurred: the license plates were in plain view, so the officer did not search or seize anything by seeing the tags. This is clearly right in light of cases like New York v. Class. But then the court did something unusual; it adopted parts of a framework from a 6th Circuit dissent suggesting that license checks could be searches in some cases because they might retrieve particularly sensitive information.

  In Diaz-Castaneda, the Ninth Circuit concluded that the license check in this particular case was not a search, but that other checks might be searches in some cases:
[T]here is no indication that license plate checks in Oregon result in the retrieval of information that "may not otherwise be public or accessible by the police without heightened suspicion." The information that [the officer] accessed about [the driver], namely that he was the registered owner of the truck and that his license was suspended, was already present in the police database and presumably available to any inquiring police officer. Were this not the case — for example, had [the officer] used [the driver]'s license plate to obtain information that [the driver] reasonably expected would be unavailable to the police, or had [the officer] violated police guidelines regarding the proper searching of databases — our conclusion might very well be different.
  This suggestion is just a suggestion, and obviously doesn't settle anything. But it's hinting at a pretty revolutionary idea. The Fourth Amendment has always been understood to regulate government collection of information but not subsequent use of information. Once information is collected, the government is free to use it however it wants. But here the Court seems to be suggesting a potential use restriction. First, license check "might very well" be a search if the officers do not comply with "police guidelines regarding the proper searching of databases." Second, it might be a search if the information retrieved by the officer is information that a driver wouldn't reasonably expect an officer to be able to retrieve (which I gather refers to information in government databases held outside the police department, although it' not clear).

   I'm understanding the language correctly, the panel seems open to the idea that that the government might collect information without violating a reasonable expectation of privacy -- leading to its inclusion in some kind of government database -- but then could retrieve the information in a way that does violate a reasonable expectation of privacy. That is, the court seems to be suggesting the possibility of a Fourth Amendment use restriction beyond the rules regulating the initial government collection of information. This is a pretty revolutionary suggestion, and I'll be very interested to see if a future Ninth Circuit panel takes that ball and runs with it.

  Thanks to Howard for the link.
My first thought on an example of information obtained by the government in which a person has a reasonable expectation of privacy would be use of information acquired by Medicaid to further a police investigation.

The reason I jump to this conclusion is that as a litigator, I often find myself arguing with the other side about the extent to which my client's physician-patient privilege is waived by putting an injury in issue. For example, I once had an attorney tell me that he was entitled to my clients opthamological records because she'd put her eyesight into issue by testifying about what she'd seen. Likewise, we routinely refuse to disclose our female clients' ob/gyn records.

Although the billings submitted to Medicaid probably aren't complete medical records, they do have ICD-9 codes associated with them. It is possible that by using that diagnostic information, the police could identify a particular suspect in a particular town.

Clearly, that's information that has been collected by the government (Medicaid/Medicare), but that most people would not expect the police to have access to.

That's all I got.

7.19.2007 9:32am
John C (mail):
Isn't use restriction inevitable - and eminently desirable? In the Information Age, where information can be collected once and held forever, where multiple collections of data from disparate entities can be collected by one central entity (the government), where information must be given over by individuals almost as a matter of course (to do everything from register for benefits to buying a book from Amazon), isn't it the case that eventually, some restrictions on the government's use of information it already has will have to be implemented, lest Big Brother have a complete database on us, accessible to every government officer always and forever, for any reason or (under current doctrine) for no reason at all? I don't want to live in that kind of country.
7.19.2007 10:14am
John C,

Use restrictions are indeed very desirable, and there are many statutes that impose them. Privacy law includes dozens of privacy statutes including such restrictions, and I take it that those statutes are uncontroversial.

The normative issue is whether the courts should reject the existing version of the Fourth Amendment and begin to create a new code of criminal procedure that imposes use restrictions as a matter of constitutional law -- and if so, what those use restrictions should look like. I suspect readers would disagree as to whether it's proper and desirable for judges to take on that task.
7.19.2007 10:23am
I agree with John C, and I disagree with how Orin is framing the issue. The "existing version" of the Fourth Amendment is, as it always has been, the text of the Fourth Amendment. The text of the Fourth Amendment simply provides a guarantee against "unreasonable searches", and again as always has been the case, what is reasonable depends on the context of the action. So, as the nature of the real world changes with respect to matters relevant to the Fourth Amendment, it is only to be expected that the proper application of the Fourth Amendment's reasonableness standard will also change. As for whether judges should participate in that process, of course they should--they have a duty to apply the Constitution to cases properly before them.
7.19.2007 10:41am

But the issue is not reasonableness; the issue is what is a "search." The reason the Fourth Amendment does not have a use restriction is that the notion of conducting a "search" has historically meant gaining access and collecting information, not using information that has already been collected. Your textual case would be very strong if the Fourth Amendment prohibited "unreasonable searches, seizures, and uses of information," but it only prohibits unreasonable searches and seizures. If you want to be a textualist and ignore all of the cases, you need to come up with a textualist reason why use of already obtained information is a "search."

Incidentally, if my memory serves me correctly, Harold Krent has made the best case for why a use restriction should be read into the concept of searches; I would check out his 1995 article in the Texas Law Review for more.
7.19.2007 10:55am
Just to elaborate a bit, it seems to me one of the issues arising out of the changes described by JohnC is the question of when we could apply any sort of law-enforcement intent test, which is going to be required for an application of the Fourth Amendment's reasonableness test. With something like a wiretap, we can talk about the law-enforcement intent of the people conducting the wiretap. But it is increasingly the case that information which we generally wish to keep private is going to accumulated in a wide variety of places with no law-enforcement intent behind that accumulation (these could be places in the hands of third parties but also often in the government's hands). Accordingly, increasingly the first point at which a law-enforcement intent will become relevant is not going to be when the information was accumulated, but rather when the accumulated information was pulled out by some sort of search run by a government official.

So, it seems quite clear to me that we will indeed have no choice about extending the Fourth Amendment to cases involving the search for information already collected, because increasingly that will be the first point at which we could identify any sort of law-enforcement intent.
7.19.2007 11:06am

I'm curious: Where in the text of the Fourth Amendment are you finding your intent test? The Supreme Court has unanimously rejected an intent test; I'm interested to hear your textualist case that this is wrong.
7.19.2007 11:16am

We cross-posted. On the textual point, I would first note that the Fourth Amendment refers to "searches and seizures", so it is certainly not the case that a search has to involve some sort of seizure. With that in mind, I think it makes perfect sense to view a search for specific information within a large body of information already accumulated for a different purpose (or no specific purpose at all) as in fact a search.

Indeed, I think that is implicit in the word search--that the person doing the searching is looking for something in particular. And as I noted, increasingly it is the case that the first time such a search will happen, at least for law enforcement purposes, is well after the information was already accumulated.

Or if you prefer, this would be a two-part search: first the large body of information was accumulated (for whatever purpose), and then the search was completed when the official searched through that accumulated body of information for the specific information in question. The fact that the first step of these two-step searches would be common to many possible searches shouldn't bother us, because that is simply a byproduct of the new information technology.

And obviously, what happened at both steps of this two-step search process may well be relevant to a Fourth Amendment review. Indeed, it seems quite odd to me to restrict one's Fourth Amendment review to just the first step of this two-step search process, which is what you seem to be advocating.
7.19.2007 11:22am

We cross-posted again. You wrote:

"Where in the text of the Fourth Amendment are you finding your intent test? The Supreme Court has unanimously rejected an intent test ...."

Perhaps I was unclear. In cases like Bell, TLO, Griffin, Skinner, and Vernonia, the Supreme Court has distinguished between searches where the purpose is to further law enforcement versus searches where the purpose is to further other "special needs". As the Court held in Vernonia:

"Where a search is undertaken by law enforcement officials to discover evidence of criminal wrongdoing, this Court has said that reasonableness generally requires the obtaining of a judicial warrant, Skinner, supra, at 619. Warrants cannot be issued, of course, without the showing of probable cause required by the Warrant Clause. But a warrant is not required to establish the reasonableness of all government searches; and when a warrant is not required (and the Warrant Clause therefore not applicable), probable cause is not invariably required either."

Textually, the Court has based this distinction in the word "unreasonable". Again, as the Court had noted in Vernonia:

"As the text of the Fourth Amendment indicates, the ultimate measure of the constitutionality of a governmental search is 'reasonableness.' At least in a case such as this, where there was no clear practice, either approving or disapproving the type of search at issue, at the time the constitutional provision was enacted, whether a particular search meets the reasonableness standard '"is judged by balancing its intrusion on the individual's Fourth Amendment interests against its promotion of legitimate governmental interests."' Skinner, supra, at 619 (quoting Delaware v. Prouse, 440 U.S. 648, 654 (1979))."

That is why whether the government had a law enforcement intent at the time of the search is relevant to an application of the Fourth Amendment.
7.19.2007 11:36am
I just did a brief skim of the opinion and, maybe I'm failing to see the depth of the analysis here, but I do think that the issue is not one of a distinction between "collection" and "use" but whether the second step of using a license plate to get personal information is itself a search.

I have to think all they are saying is that the police's "search" does not end once they see the publicly exposed license plate, and that there is a second step where they punch it into their computers. If, in some future case, this retrieval was unreasonable in some way, it would violate the 4th amendment because it would still be a "search." It may be parsing hairs but I don't think this counts as "use."

I will readily admit though that I am not a 4th amendment scholar and the terms may be more technical than I am assuming.
7.19.2007 11:50am
By the way, it may help to amend my previous comment to refer to a "law-enforcement purpose test", rather than a "law-enforcement intent test". I suspect (although he can correct me if I am wrong) that Orin has in mind the "subjective intent" and "pretext" cases, and I did not mean to get into that particular thicket. Rather, I have in mind cases in which everyone agrees the search for particular information within the accumulated body of information was for a law enforcement purpose.
7.19.2007 12:02pm
It seems like the court is saying that if the data had not been originally collected by the police, than it should not be accessible for use in further criminal matters.
7.19.2007 12:05pm
And as one last addendum, here is the Supreme Court in Brigham City recently making the same point:

"As respondents note, we have held in the context of programmatic searches conducted without individualized suspicion--such as checkpoints to combat drunk driving or drug trafficking--that 'an inquiry into programmatic purpose' is sometimes appropriate. Indianapolis v. Edmond, 531 U.S. 32, 46, 121 S.Ct. 447, 148 L.Ed.2d 333 (2000) (emphasis added); see also Florida v. Wells, 495 U.S. 1, 4, 110 S.Ct. 1632, 109 L.Ed.2d 1 (1990) (an inventory search must be regulated by 'standardized criteria' or 'established routine' so as not to 'be a ruse for a general rummaging in order to discover incriminating evidence'). But this inquiry is directed at ensuring that the purpose behind the program is not 'ultimately indistinguishable from the general interest in crime control.' Edmond, 531 U.S., at 44, 121 S.Ct. 447. It has nothing to do with discerning what is in the mind of the individual officer conducting the search. Id., at 48, 121 S.Ct. 447."

Again, I was definitely not intending to revisit the subjective intent issue.
7.19.2007 12:37pm
NYC has a scheme for congestion pricing that will have computers capture the tag numbers of all vehicles going in and out of the pay zone. What are the limits of what the city can do with the info?
7.19.2007 12:56pm
Frater Plotter:
The Founders, of course, did not have information retrieval systems that were any more complicated than a logbook or filing-cabinet. They certainly did not have the notion of data warehousing or data mining, wherein vast swaths of collected data (collected without a formal search) can be quickly searched for desired properties ....
7.19.2007 12:58pm
Hattio (mail):
I have a slightly off-topic question. The opinion (or at least Professor Kerr's summary) speak of the information readily available to a police officer. Is this really how a reasonable expectation of privacy is defined? I doubt it, but I don't really know because in most cases, what's obvious to a police officer would also be obvious to the public. (Admittedly a police officer may realize the significance of something more easily than the public).
7.19.2007 1:01pm
Hattio, I'm a little cynical about this, but ever since Prosser sold the world on the idea that taking a picture of Person A and showing it to Person B is identical to writing down a description of Person A and reading it to Person B is identical to Person B having been there to see Person A in the first place, a pretty good rule of thumb is "Reasonable expectation of privacy" means "Expect to have no privacy."
7.19.2007 1:29pm
Hattio (mail):
When did Professor Kerr say that, and secondly, from a Fourth Amendment Standard solely I have to agree with him. Assuming person A isn't a shut-in who only reveals himself to the pizza delivery guy.
7.19.2007 1:41pm
Hattio, sorry, I have a bit of an axe to grind with the idea that unless you're a shut-in who only reveals yourself to the pizza delivery guy, you have no privacy for 4th Amendment or tort purposes.
7.19.2007 1:45pm
Puts me in mind of the "plates for dates" phenomenon, where a cop allegedly runs the plates of a car with an attractive woman behind the wheel in order to get her address &phone number. Creepy anywhere, but perhaps would be held unconstitutional in the Ninth?

(Now my mind jumps to standing challenges to any such suit that might be brought. "You're not attractive enough to have standing, ma'am; you're flattering yourself to think getting your plate run is more than conjectural. And even if you were a bona fide hottie, where's the 'harm' in being asked out by one of our boys in blue? Is that really something you want us to remedy?")
7.19.2007 1:46pm
Anne (mail):
Suppose a local sheriff has posted to a state-wide database information on a CCW permit, which comes up on the license plate check as "person of interest." That results in a more intense stop, and has resulted in confiscation of legal firearms. Usually the firearm is returned after a thorough check of the database, but that may take a couple hours. I fear that the "person of interest" flag could make a traffic stop a felony stop... It's an ugly scenario.
7.19.2007 2:09pm
Hattio (mail):
You're missing the point. It's not that you have no reasonable expectation of privacy at all. Merely that you have no reasonable expectation of privacy in your image. After all, aren't you showing your image to the public everytime you walk out your front door?
7.19.2007 2:24pm
John C (mail):
You're missing the point. It's not that you have no reasonable expectation of privacy at all. Merely that you have no reasonable expectation of privacy in your image. After all, aren't you showing your image to the public everytime you walk out your front door?

And this is getting to the larger point. I have no reasonable expectation of privacy in showing my face to those I pass on the street, but what about those taking my picture? When I walk out the door, I don't consent to showing my face to the "public" (all 6 billion of them, and all their ancestors, forever). Isn't it reasonable to assume that I don't consent to taking that picture and, say, putting it up on the Internet for all time?

I think the point to all of this is that with the new-found ease of electronic searching of data and "data permanence," long-held conceptions of "reasonable expectations" are breaking down. And, from my point of view, should break down.
7.19.2007 2:52pm
John C,
Right in one.

It's similar to the flaw I see in the third-party doctrine (U.S. v. Miller, Smith v. Maryland, etc.). If you share a piece of information with a person, particularly if it's for a particular purpose, then I think you do have a reasonable expectation that the only person who will know that information is that person and the only use they'll make of it is the purpose you gave it to them for, and the law needs to recognize that.
7.19.2007 3:08pm
Bryan DB:
Orin at 9:55am
"The reason the Fourth Amendment does not have a use restriction is that the notion of conducting a "search" has historically meant gaining access and collecting information, not using information that has already been collected."
But Orin, the way you're talking about "using" and "gaining access" misses the point, I think. If a search is "gaining access," then isn't the officer's "gaining access" to information in which a person has a REP a "search?" As Phantom noted, I don't expect the police to be gaining access to information I provide on my medical history just because some other arm of the government collected the information, and I'd certainly consider that a "search" (gaining access) if they went to the information because someone linked up the databases in a way that I wouldn't reasonably expect.
How about this "search": Sometime in the future, all objects in the world are hyperlinked to all other objects in the world. The police walk into my house and see (among the evidence they're seeking) a medicine bottle. They touch the medicine bottle and learn something incriminating from my medical records. Is that a "search" of new evidence or just "use" of evidence already available? That seems a lot like a search for information in which they otherwise had no interest, not just a use of information that was somehow available, through further actio, once they walked into the room.
With the government databases, the government could eliminate all REP by announcing that every database in the government was linked to every other database, such that no "search" ever occurred. I think that a "use" restriction makes perfect sense in this context, and it doesn't seem it would be all that hard (famous last words) to define a "search" as going outside a standard definition of what's allowed to be in a police database.
7.19.2007 5:09pm
Kelvin McCabe:
Okay, change the facts a bit. Cops are bored or work in a location that doesnt have a particularly high crime rate. So they set up shop at a mall parking lot and systemetically run the license plates of every single car in the lot, hoping to find something to later bust the person for once they return to the car and drive away.

Now, here in IL, there used to be wide body of case law that prohibited police "fishing expeditions" for evidence of crime. Normally, this would be during a routine traffic case where the cops incessantly look for drugs/whatever that had no purpose or connection with the reason for the initial stop for speeding/whatever.

Ignoring the horrible Caballes decision for now, Why is it inherently WRONG (under the 4th amendment) for police to go on suspiconless "fishing expeditions" for evidence of some types of crimes (i.e, possession of drugs, guns,etc...) but Okay for them to go on suspicionless "fishing expeditions" for information/administrative crimes? A suspicionless fishing expedition that results in an arrest is the SAME to the arrestee.

I realize I am in the minority here, but even for a limited intrusion terry stop, there has to be individualized suspicion. Its the lack of individualized suspicion that irks me with this "i was just driving my police cruiser and the guy in front of me was doing nothing wrong, but i ran his plates anyway and then i noticed..." Who cares if the information would have been available to any other cop during a routine traffic stop (once that cop ran the plates or drivers license # after the stop). There was no initial stop, no individualized suspicion and hence, to me, no reason for the plates to be run in the first instance.

A cop merely looking at license plate numbers does nothing. A cop radioing in the license plate # to a dispatcher, who then pulls up a driving abstract, and then reports the results to the officer is something else. Isnt the cop then actively investigating (or as i would say "fishing")? If not, what he is doing? If so, shouldnt he have some or any reason to do so, other than the car was in public and he is a cop? Or are we all inherently suspect by virtue of being in public?
7.19.2007 5:32pm
Steve2 says:
If you share a piece of information with a person, particularly if it's for a particular purpose, then I think you do have a reasonable expectation that the only person who will know that information is that person and the only use they'll make of it is the purpose you gave it to them for, and the law needs to recognize that.
A tells B "in confidence" that A plans to commit a crime (a murder, say). Under your proposed rule, may B voluntarily go to the police and disclose A's statements? May prosecutors bring B before the grand jury and demand that he testify as to A's statements? (Assume that B is not a priest/confessor or an attorney.)

Same result if B turns out to be an undercover police officer?

I think maybe there are some problems with your proposal, Steve2.
7.19.2007 5:57pm
Apodaca, I'm willing to accept the hamstringing of law enforcement if that's what it takes to get rid of the proposition from U.S. v. Miller that there's no reasonable expectation of privacy in your bank records since you've shared the information with your bank. I'm not sure measures to protect individual privacy can ever be draconian enough.
7.19.2007 6:02pm
Hattio writes:
I have a slightly off-topic question. The opinion (or at least Professor Kerr's summary) speak of the information readily available to a police officer. Is this really how a reasonable expectation of privacy is defined?
No, it's not. That's part of what's so puzzling about the dicta: I can't tell if Judge Fisher incorrectly thinks that this is the test, or if he knows it isn't and wants to create a new set of principles.
7.19.2007 7:20pm

FWIW, I'm actually working right now on an article defending the third party doctrine, and explaining why so many of the criticisms of it are misguided (and at the same time, why it is a narrower doctrine than many people think). I hope to post a copy of the draft in a few weeks.

7.19.2007 7:22pm
...Were this not the case — for example, had [the officer] used [the driver]'s license plate to obtain information that [the driver] reasonably expected would be unavailable to the police, or had [the officer] violated police guidelines regarding the proper searching of databases — our conclusion might very well be different.

Because the court found there was no search in the instant case, and hence no Fourth Amendment violation, we don't know under what circumstances they might conclude that use of the license plate information to come up with more resulted in an improper search. Can anyone suggest a plausible fact pattern starting with the use of a readily observable license plate number and going on to the discovery of information that the driver could reasonably expect to be unavailable to the police, giving the Ninth Circuit reason to find that there had been a search and a Fourth Amendment violation?
7.19.2007 9:56pm
Can the IRS require people to provide information about illegal businesses for tax purposes and then prosecute people for the underlying businesses? Can the Census Bureau require people who live in a state with fornication laws to respond to a tell it about their living arrangements, and could the states use it?

It would seem that self-incrimination represents a reasonable analog of the situation. Perhaps the 9th Circuit would have better phrased matters in terms of the idea that law creating a duty to turn over information creates an implied use immunity for that information. There might also be a takings concept -- personal information is valuable, as anyone who creates marketing lists knows. I think the concept has some viability, although not necessarily under the 4th Amendment or a generalized "privacy" concept.
7.19.2007 11:38pm
Ah, I look forward to that, Professor Kerr. Most of the articles I've read have been critical of the doctrine - and really of court treatment of privacy in general, whether in 4th Amendment reasonable expectation terms or regarding invasion of privacy torts.
7.20.2007 1:28am
SethB (mail):
There's some information that I'm required to provide to an agency of the government (e.g. IRS, Census Bureau) that the government would otherwise need a warrant to get. I would argue that use of that information is a search.

In some states, the mapping from license plate to name/address is available to the public. It's clearly not a search for the police to use that information (any more than for them to use the information in the phone book).
7.20.2007 1:42pm
Parking enforcement personnel go about with handheld computers entering license plate numbers so they can come back later and see if anyone is still parked in the same space after the allowed 1,2, 4 or however many number of hours. (More modern approach than putting chalk marks, which can be wiped away easily enough, on tires.) Would the Ninth Circuit think it improper if those entries made for purposes of parking enforcement were used to find stolen vehicles, ones with unpaid traffic/parking violations, those that might have been involved in hit-and-runs, etc.? Is "intrusiveness" an irrelevancy, and it would not matter that no one was stopped and/or hassled, that none of the information involved was truly "new," that it all immediately pertains to the vehicle's registration and operation?
7.21.2007 2:14am