01/09/2014 06:59 pm ET Updated Mar 11, 2014

The NSA's Telephone Metadata Program Is Unconstitutional

In my last post, I concluded that the NSA's bulk telephony metadata program is a "search" within the meaning of the Fourth Amendment. But because the Fourth Amendment prohibits only unreasonable searches, the next question is whether the program is "unreasonable." This turns out to be a rather complicated question. So, bear with me as I try to work my way through it.

At the outset, it is important to recall exactly what the bulk telephony metadata program does. Under section 215 of the Foreign Intelligence Surveillance Act, as interpreted by the Foreign Intelligence Surveillance Court, the NSA is authorized to obtain from telephone service providers on a daily basis the calling records of millions of Americans. The calling records, or metadata, consist of the phone numbers called by a particular phone number and the phone numbers that have called that particular number. They do not include any information about the identities of the individuals or the contents of the calls. The NSA holds this metadata in a very large database for a period of five years, after which it is expunged.

Whenever NSA analysts have reasonable and articulable suspicion that a particular phone number is associated with a person involved in terrorist activity, they can "query" the database to determine what phone numbers the suspect phone number has been in touch with. This is the only purpose for which the NSA may access the database.

In 2012, the NSA queried a total of 288 phone numbers. Based on these queries, the NSA found 16 instances in which a suspect phone number was directly or indirectly in touch with another phone number that the NSA independently suspected of being associated with terrorist activity. In such cases, the NSA turns the information over to the FBI for further investigation.

In terms of the "connect the dots" metaphor, the purpose of the program is not so much to discover new "dots" but to determine if there are connections between two or more already suspect "dots." For example, if a phone number belonging to a terrorist suspect in Pakistan is found to have called a phone number in the United States that the government independently suspects belongs to a person involved in possible terrorist activity, alarm bells (figuratively) go off very loudly, alerting the government to the need for immediate attention.

It is important to note that the program does not allow the NSA or anyone else in the government to access the metadata for any purpose other than what I have just described. There are rigorous safeguards in place both internally through the NSA's Office of General Counsel and externally pursuant to oversight by the Department of Justice, the congressional intelligence committees, and the Foreign Intelligence Surveillance Court to ensure that no one accesses the database for any other purpose.

The question, then, is whether this "search" process is "unreasonable" within the meaning of the Fourth Amendment.

As a general rule, the Supreme Court has held that conventional "searches" -- for example, when the police rummage through a person's home, wiretap his phone calls, or read his mail -- are presumptively "unreasonable" and therefore unconstitutional unless the government first obtains a judicial warrant based on a finding that there is probable cause to believe that the search will turn up evidence of a crime.

The NSA's bulk telephony metadata program clearly cannot pass muster under that standard. It requires neither a warrant nor a finding of probable cause when it collects the call records of millions of individuals, and it requires neither a warrant nor a finding of probable cause when NSA analysts query the database to investigate the call records of specific individuals whom they suspect of terrorist activity.

That does not end the matter, however, because the Supreme Court has held that many forms of searches are "reasonable" even if they do not satisfy the warrant and/or probable cause requirements. For example, the court has held that the government can inspect all homes in a town for possible housing code violations -- without any showing of probable cause to believe that any specific home has such a violation. It has held that the government can require all persons boarding an airplane to pass through a magnetometer -- without any showing of probable cause to believe that any specific person is carrying a weapon. It has held that the police can use roadblocks in which they stop every car approaching a particular checkpoint in order to check for drunken drivers -- without any showing of probable cause to believe that any specific person is drunk. It has held that a police officer can frisk an individual he has stopped for questioning if the officer has reasonable and articulable grounds to suspect that the person is armed and presently dangerous -- without any showing of probable cause to believe that the individual has either committed a crime or possesses a weapon.

The question, then, is whether the bulk telephony metadata program is another example of a search that is "reasonable" even though it requires neither probable cause nor a warrant. In addressing this question, it is helpful to keep in mind that the metadata program, unlike most searches, has two distinct elements -- (a) the collection and storage of the metadata, and (b) the querying of the database. This complicates the analysis.

With respect to the first part of the program -- the collection and storage of the telephony metadata, it is worth noting at the outset that the government routinely collects all sorts of data about us. It collects information about such matters as taxes, employment, health, travel, families, census data, etc. One might argue that, if those types of data collection are not unconstitutional, then the same should be true of the government's collection of telephony metadata under section 215. An important difference, however, is that unlike most other data-collection activities, the section 215 program gathers the metadata precisely for the purpose of sniffing out possible wrongdoing. In that sense, it more directly implicates the concerns underlying the Fourth Amendment than most other information-gathering programs.

In my judgment, and the law here is quite unsettled, the reasonableness of any particular program of government information collection should turn on four primary factors: (1) How important is it to collect the information? (2) How private is the information collected? (3) How will the information be used? (4) How confident are we that the information will be used only for proper purposes?

How do these factors play out in the collection and storage of telephony metadata under section 215?

(1) The metadata is collected under this program for a clearly legitimate and, indeed, important purpose. Protecting the national security is surely a compelling government interest, and the President's Review Group found that the metadata program is useful, if not essential, to achieving this goal. Certainly, the government's interest in protecting the national security is more important than its interest in, say, preventing drunk driving and housing code violations.

(2) Defenders of the program maintain that it does not threaten significant privacy interests because it deals only with metadata and not with the contents of the phone calls. The collection of metadata, they argue, is analogous to the use of magnetometers at airports (as compared, say, to a requirement that all passengers must be strip-searched, which presumably would be deemed "unreasonable"). Metadata may in fact reveal less private information than the contents of the calls, but as I have explained in prior posts, and as the President's Review Group clearly found, comprehensive inquiries into an individual's telephone calling records can reveal a huge amount of highly personal information about an a individual's private life.

(3) For that reason, if the government were authorized to use the database to investigate the personal lives of ordinary Americans, the telephony metadata program would clearly be "unreasonable." But under the section 215 program, the government is authorized to access the metadata for only a very specific, very narrow, and carefully targeted purpose. It cannot legally access the metadata for any purpose other than to "connect the dots" between one suspected terrorist and another. It cannot legally access the database to learn anything about the private lives of ordinary Americans. It can legally use the database only to determine if suspected terrorist X is in contact with suspected terrorist Y. Period. Any other use of the data is forbidden. This is an essential feature of the program, and it sharply limits its potential threat to individual privacy.

(4) Despite such limitations, however, there is always the danger of abuse. When the government has access to such information, there is always a risk that it will exploit the information for a broad range of impermissible purposes. There is always the possibility that some future J. Edgar Hoover or Richard Nixon will turn this extraordinary pool of private information to ignoble ends. Moreover, even in the absence of actual abuse, individuals may fear that the government will use this information -- illegally -- to their detriment, and this in itself can affect their behavior in ways that seriously infringe on individual freedom. The safeguards in place to prevent such abuse are therefore critical to the "reasonableness" of the program. In the current system, the safeguards that now exist to prevent such abuse are quite extensive. But no system of safeguards is perfect, and in the wrong hands serious harm is always possible.

On this particular issue -- which in my view is central to the "reasonableness" of the program, the President's Review Group -- which addressed this question from the standpoint of public policy rather than constitutionality -- recommended that at least two additional safeguards are necessary. First, the metadata should be held for a period of only two rather than five years, thereby limiting the potential harm to privacy. Second, the metadata should be held, not by the government, but by a private entity -- either by the phone companies themselves or by a newly-created private agency. These additional safeguards are necessary to reduce both the risk of abuse and the fear of abuse. But with these additional safeguards in place, the collection of metadata, used solely for the purpose currently authorized, seems to me to meet the test of "reasonableness."

The second component of the section 215 program focuses on the circumstances in which the government can query the database. At present, the NSA is authorized to query the database if NSA analysts find that there are reasonable and articulable grounds to believe that a particular phone number is associated with terrorist activity.

There are two possible objections to this procedure. First, one might argue that, to meet with requirements of the Fourth Amendment, the NSA should not be able to query the database unless a judge -- rather than an NSA analyst -- makes the critical determination that there are reasonable and articulable grounds to believe that a particular phone number is connected to terrorist activity. There is no good reason to dispense with the requirement of a judicial determination, at least in the absence of an emergency. As in other searches, the decision to search should be made by a neutral and detached judge rather than by a government official engaged in the adversarial process of ferreting out the "bad guys." The absence of such a requirement, in my judgment, violates the Fourth Amendment.

Second, although reasonable people can certainly differ on this, I am inclined to think that the "reasonable and articulable suspicion" standard meets the test of "reasonableness" in this context because of the very narrow and targeted nature of the inquiry -- that is, the only information that can lawfully be disclosed as a result of a query is that a suspected terrorist has been in touch, directly or indirectly, with another suspected terrorist.

In conclusion, then, in my judgment the existing program is unconstitutional. As currently structured, it violates the Fourth Amendment's requirement of "reasonableness." On the other hand, it should be possible for the government to correct the deficiencies in the program in a manner that both preserves its legitimate value and substantially mitigates the risks to privacy that it currently poses.

There are those who maintain that this program is obviously constitutional and those who maintain that it is obviously unconstitutional. They are both wrong. There is nothing "obvious" about this. If this ever gets to the Supreme Court, it will be interesting.