More on What FISA and the Fourth Amendment Requre -- And What They Don't:
Over at Balkinization, Marty Lederman looks at the current dispute over the scope of NSA wiretapping powers and offers the following take:
Second, there are two notable reasons to think that the Fourth Amendment is not implicated in this setting. First, there are some precedents indicating that if you call someone whose calls are being monitored without violating their reasonable expectation of privacy, then it does not violate your Fourth Amendment rights to have the call recorded. See, e.g., United States v. McNulty, 47 F.3d 100, 104-106 (4th Cir. 1995) (person speaking to a person on a cordless call being monitored by the police). I don't think the Supreme Court has addressed this theory, and I gather some VC readers will find it unpersuasive, but there are such cases on the books.
Second, the Fourth Amendment generally is only triggered by intentional conduct. See Brower v. Inyo County, 489 U.S. 593 (1989) ("Violation of the Fourth Amendment requires an intentional acquisition of physical control."). If the government collects a call and doesn't know that it is conducting a search or seizure under the Fourth Amendment -- because it doesn't know a party has Fourth Amendment rights -- it may be that under Inyo there is no Fourth Amendment violation. (This latter argument is subject to the uncertainty as to whether wiretapping is a search or seizure or both, as well as how the mens rea requirement applies to searches, but I think it is still worth noting.)
[I]t must be the case that the NSA's aim is not simply to surveil foreigners who it already suspects as being part of Al Qaeda. It can obtain a FISA order as to those folks. What it wants, instead, is to be able to intercept foreign communications coming over domestic wires where (i) it does not have probable cause to believe that any of the parties is a terrorist or agent of a foreign power; and (ii) there is a chance that some of the intercepted communications will be with persons in the U.S.As to the first point, I think that is clearly right. Federal law plainly allows the government to monitor foreign people with no connection to U.S. soil so long as the monitoring occurs outside the United States. Such persons have no Fourth Amendment rights under Verdugo-Urquidez, and FISA does not apply. So the government wants to do the same monitoring from the U.S. switches that they can do from monitoring posts outside the U.S. That is, they want to take out the relevance of where the monitoring site happens to be. The question is whether FISA allows particular measures to try to do that given the possibility of picking up calls and e-mails to the U.S.
FISA does not allow surveillance where both of those conditions obtain. (Indeed, insofar as the surveillance picks up U.S. persons in such cases, without proof that either party is the agent of a foreign power, it would probably violate the Fourth Amendment, too, at least according to the pre-FISA cases that considered the question.)
Second, there are two notable reasons to think that the Fourth Amendment is not implicated in this setting. First, there are some precedents indicating that if you call someone whose calls are being monitored without violating their reasonable expectation of privacy, then it does not violate your Fourth Amendment rights to have the call recorded. See, e.g., United States v. McNulty, 47 F.3d 100, 104-106 (4th Cir. 1995) (person speaking to a person on a cordless call being monitored by the police). I don't think the Supreme Court has addressed this theory, and I gather some VC readers will find it unpersuasive, but there are such cases on the books.
Second, the Fourth Amendment generally is only triggered by intentional conduct. See Brower v. Inyo County, 489 U.S. 593 (1989) ("Violation of the Fourth Amendment requires an intentional acquisition of physical control."). If the government collects a call and doesn't know that it is conducting a search or seizure under the Fourth Amendment -- because it doesn't know a party has Fourth Amendment rights -- it may be that under Inyo there is no Fourth Amendment violation. (This latter argument is subject to the uncertainty as to whether wiretapping is a search or seizure or both, as well as how the mens rea requirement applies to searches, but I think it is still worth noting.)
Related Posts (on one page):
- Are Telcos Still Liable for NSA Cooperation?
- My Take on the New FISA Amendment:
- More on What FISA and the Fourth Amendment Requre -- And What They Don't:
I think it's an open question, although I doubt it. (Otherwise it would take out the mens rea issue entirely, as all surveillance involves gathering information.)
As to the 4th Amdt., note that it doesn't prohibit all searches, but just unreasonable ones. I seriously doubt that in a time of war, being surveilled for national intelligence (and not criminal) reasons for communications with known or reasonably believed enemies of this country would be considered unreasonable. And, since the TSP is well known now, there is no reasonable expectation of privacy either in such a call.
Yes, if the FBI tried to use the fruits of such an intercept in a criminal prosecution, then that just might change the 4th Amdt. dynamic. And it would definitely change it if they made a habit of that. But in the time that the TSP has been in operation, the number of those prosecutions has been exactly zero.
I think it's more accurate to say that the number is zero as far as we know. After all, the FBI could probably do this without anyone knowing. At least for some cases.
I don't find this very persuasive, whether as the reason for the Administration's position or as justification to change the statute. I think Prof. Lederman is more likely correct:
"[I]t must be the case that the NSA's aim is not simply to surveille foreigners who it already suspects as being part of Al Qaeda. It can obtain a FISA order as to those folks. What it wants, instead, is to be able to intercept foreign communications coming over domestic wires where (i) it does not have probable cause to believe that any of the parties is a terrorist or agent of a foreign power; and (ii) there is a chance that some of the intercepted communications will be with persons in the U.S."
Likewise, whether purely extraterritorial communication is covered by the Fourth Amendment seems purely jurisdictional and intent is neither here nor there.
Nor do I understand your initial response in the comments, that such a reading would eviscerate Brower (which did not involve a jurisdictional question at all). One could both collect information as to purely foreign communications both in a way that would be and would not be a search. If one intended to collect information as to purely foreign information in a way that would not be a search, but it turned out to be a search (they thought a particular communication was public when it turned out to be private, and the mistake was one of fact rather than law, for instance), Brower would apply (and the question would be moot anyway, if it turned out indeed to be outside the jurisdictional reach of the Fourth Amendment).
And even if you took the mens rea part out of the equation, why would that be an absurd result? The text of Brower only refers to "means intentionally applied," which would, indeed, involve any aspect of the intentional gathering of information - even if one, whose means intending to gather purely lawful data, also accidentally gathered some unlawful data. Comparing the concept of the intentional interception of a communication between France and some unknown location, and a "parked and unoccupied police car slips its brake and pins a passerby against a wall," seems quite a stretch, IMHO.
The government is collecting data inside the United States; why should this be analogized to a jurisdictional element of a criminal statute? I don't see what's jurisdictional about it, as the question is who is being monitored not where the monitoring is occurring.
This creates a situation where all the traffic folows through servers in the US, even if the communication is between two people in Pakistan. Such a communication is not protected by either FISA or the Consititution, but there is no way to verify this without looking at it and the if one of the parties turns out to be in the US it is too late. There is no way to get a warrant because the IDs are anonymous and there is no other way to specify the target.
What the NSA might want to do is to intercept all IM traffic between US servers and clients in Waziristan. Then if a conversation is found to involve a US party, a warrant could individually be obtained or the traffic could be discarded. Unfortunately the law did not anticipate communications where the endpoints can constantly change identity.
Modern technology creates modes of communication that could not be anticipated by the authors of FISA. If Congress is going to write this type of legislation, then it needs to periodically update it.
I can grant this analysis only if the lack of probable cause listed in (i) includes the situation where an enemy agent's identity is not known or is not known before the intercept, but there is considerable evidence from the transmission itself and its origin that it is from a bad guy. The US does not only want to monitor top terrorists whose names and faces are known; we would also like to monitor signals from small fry who just joined the organization.
That explains the "basket warrant" mentioned in the previous post. It is quite possible that certain areas are being monitored because they are known operating bases of organizations. However, not every enemy agent's identity is known. A call made internationally would normally be monitored without a FISA warrant. The identity of the bad guy would not be known until after the intercept-- and the intercept itself would be the first evidence of probable cause for the individual.
There's also the further question, mentioned in the other post, of whether a FISA warrant is necessary in order to intercept a call being two non-US persons but routed through US soil or a US switch at some point. So (ii) does not necessarily have to apply as well, at least under some interpretations.
Under the most stringent interpretation, the traffic of an agent who only communicates through signals routed through the US could never be intercepted, because enough evidence against the individual specifically to get a FISA warrant could never be obtained without the warrant. Even if that person is calling from a known Al Qaeda base.
Justin, the question is both. Any US person requires a FISA warrant. That includes US citizens wherever they are, and lawful aliens when they are inside the US. (It also includes corporations incorporated in the US and organizations compromised substantially of US persons.) In practice, it also includes anyone inside the US, including illegal aliens, unless substantial evidence demonstrating their illegal status exists. In the latter case, presumably a FISA warrant could be obtained anyway.
Professor Kerr, you touch on the possibility of unintentionally picking up calls and e-mails to the US, but you should go further. International listening posts present the same kinds of issues. First off, purely international communications may be made by US persons abroad, and hence be prohibited without a FISA warrant. Secondly, signals intercepted by international listening posts may have ultimate destinations in the US. With modern digital signals it is often impossible to know the parties involved and the origin and destination until a signal is intercepted and decoded.
The NSA has various operational safeguards under Executive Orders to deal with such situations already, requiring among other things machine-only handling of communications and discarding of those with termination inside the US, the discarding of communications within a time limit not demonstrated to be by targeted non-US persons, and so on.
A US domestic listening post is still extremely useful for listening to purely international calls between non-US persons, but there are those who believe that the domestic portion of the call's routing would require a FISA warrant.
Then there is this:I don't even know how to make sense of that -- a communication by definition has at least two parties, and the gist of your claim is that wiretapping can only occur when the government has prior possession of probable cause against BOTH parties. That's certainly not how wiretapping works when it's not foreign intelligence -- they only have a warrant against one party to the communications, and in most communications that the target has the other party is completely innocent.
If you call your drug dealer to discuss a drug deal, and he happens to be sitting in a donut shop one table away from 2 cops, and he puts you on speakerphone so the cops can hear, then at your drug trial you don't get to claim that the testimony of the 2 cops should be thrown out because you are a US citizen and they didn't have a FISA warrant to listen in to your call.
I believe that is incorrect. The definition of "electronic surveillance" in the the FISA statute, 18 USC 1801(f), protects U.S. persons only within the United States.
However, the "minimization procedures" given in paragraph (h)
that "minimize the acquisition and retention, and prohibit the dissemination, of nonpublicly available information concerning unconsenting United States persons" apply to U.S. persons wherever they are, and must be followed unless there is a FISA warrant or with a few exceptions. (Imminent danger of death to another person, for example.)
It's somewhat unclear in the case of the material being monitored outside the US, but if the switch is within the US:
is considered electronic surveillance.
You misunderstood my point. If the call were purely foreign, no warrant would be required and none is generally sought. It is definitely NOT true that FISA warrants are routinely sought against foreign targets not visiting the US. They are not. However, if one end of the call if with a US person, then SOME FISA warrant is required, whether on the foreigner or the US person.
Therefore, the introduction of a US person transforms the situation from one where no warrant is required and generally none is sought to one where a warrant is required on one of the parties.
And while there may be two parties to the call, when it comes to intelligence surveillance particular information (including but not limited to identifying information) about one US person party to the conversation may not be analyzed, kept, or disseminated if the warrant is only obtained for the other party.
Is it so strange that FISA is written differently from non-intelligence surveillance?
The Troung line of cases holds that the 4th Amendment does not require warrants to conduct surveillance of agents of foreign groups so long as the primary intent is to obtain intelligence and not criminal evidence. While the Supreme Court has not weighed in on this question, I would not assume that that our current conservative Supreme Court is likely to overrule this nearly unanimous precedent of a far more liberal group of 1970s and early 80s circuit courts.
Furthermore, even if by some serendipity both sides of an international telecommunication involving al Qaeda telephone numbers were not agents of a foreign group, I am not aware of any cases holding that persons or property crossing our international border have an expectation of privacy from government search which would require a warrant under the 4th Amendment. Consequently, I am unsure why international telecommunications should be any different.