We now have two district court rulings on the legality and constitutionality of the NSA meta-data surveillance program. One decision, from Judge Leon, concludes that the program is likely unconstitutional, and the other decision, from Judge Pauley, concludes that the program is lawful and is not unconstitutional. Thus the stage has been set for a debate at higher levels of the court system.

This debate has thus far focused primarily on the Fourth Amendment questions raised by this surveillance program, as well as the statutory questions. On the Fourth Amendment front, Judge Pauley relies on a traditional interpretation of Smith v. Maryland, holding that the government can obtain your phone records, without probable cause or a warrant. Judge Leon presents a newer perspective that the broad scope of this NSA meta-data surveillance program makes it an unreasonable search even though a search of a more limited scope might be reasonable. On the whole, Judge Pauley's approach is more in line with the way that courts have addressed these Fourth Amendment questions in the past, and there are strong arguments that the courts should stick to this traditional approach, such as those made by Orin Kerr.

Unfortunately, the First Amendment issues seem to be a secondary concern in this debate. Judge Leon didn't address them. Judge Pauley spent a few paragraphs on them in his decision. However, numerous commentators on Pauley's ruling have skipped over that section. For example, Benjamin Wittes didn't include the First Amendment as a "key issue" in his blog post about Judge Pauley's ruling.

I think that the First Amendment issues deserve closer consideration. A program that collects telecom metadata is a program that collects information about a person's associations. We have a right to freedom of association, and that right is obviously impacted on some level by government surveillance of it's exercise. Any desire or expectation that we have that our telecom meta-data should remain private must stem from a desire to be able to exercise our right to freedom of association without government scrutiny. (See Freedom of Association in a Networked World: First Amendment Regulation of Relational Surveillance by Katherine J. Strandburg, NYU)

Judge Pauley dismisses the First Amendment concerns with this program for two reasons, neither of which I find completely persuasive. The first reason is the principle that the First Amendment does not prohibit the government from performing searches that are considered reasonable under the Fourth Amendment. The idea that the framework of the Fourth Amendment provides adequate protection against the impact that searches can have on First Amendment rights is a decent rule of thumb, but even Pauley seems to concede that there may be circumstances in which a First Amendment violation has occurred due to a search absent a violation of the Fourth Amendment. I think this is most likely to occur in a circumstance where a search is considered "reasonable" without meeting the Fourth Amendment's warrant and probable cause requirements, as those requirements provide much of the teeth that help ensure that the subject's rights aren't being trampled. It is also distinctly possible in this context because of the expansive scope and indiscriminate nature of the data collection in the program at issue.

Pauley's second reason for dismissing this concern is a finding that the plaintiffs can only speculate that their meta-data might be accessed or inspected by a government agent and this speculative fear is not sufficient to establish that a violation of the First Amendment has occurred. While the plaintiffs can only speculate that their data will be accessed, they know for certain that it will be collected. The certain collection and storage of the data can be a deterrent to the exercise of First Amendment rights in ways that are categorically different from a set of circumstances in which the collection of the data is itself uncertain, and I do not think that Pauley provides this distinction it's due consideration. Furthermore, prior precedents such as Amnesty v. Clapper and Laird v. Tatum dealt with uncertain collection rather than certain collection and uncertain access.

The problems may be easier to understand if considered in a different context. Section 215 of the Patriot Act is not limited to telecom meta-data. If the NSA telecom meta-data program is considered lawful, similar programs that collect other types of meta-data may also be considered lawful. For example, the NSA might construct programs that collect and retain DNS server logs from Internet Service Providers (which provide the names of the servers accessed by that ISP's customers), or netflow records from the same ISPs, which provide a list of all the IP addresses those customers accessed.

Such a program could provide the government with a database of every website visited by every American, going back many years. This data could be useful in a terrorism investigation. For example, if a website was discovered that was known to be a coordinating point for terrorist attacks, it might be helpful to be able to go back into the past and see who accessed that website before it was discovered. However, Americans would also be put into a position where every time they clicked on a link on the Internet, that click was a part of their permanent record stored with the government, and might be accessed in the future. Could this certain collection but uncertain access deter people from visiting websites that they suspect the government might investigate in the future?

Take, for example, WikiLeaks. Wikileaks is a foreign website that has published information that the United States government considers classified. Although Wikileaks is certainly a newsworthy website and many Americans have accessed it in order to better understand news that has been reported about it, the United States government has warned its own employees that they could be sanctioned for accessing the site. The government surely must consider Wikileaks to be a valid foreign intelligence target given that it is a foreign website disseminating stolen, classified information, and it would not be surprising to learn that the government might query a database of Internet meta-data to see who had accessed it, as any of those people might be a source of one of the leaks. If such a database existed, and the government queried it, it might reveal the identities of Americans who accessed Wikileaks before Wikileaks first published information that the U.S. government considers classified.

What this means, is that if you were planning to access a website like Wikileaks, even if nothing on that website today would provoke the ire of the United States government, you have to ask yourself whether that website might provoke the government's ire in the future, and whether you are comfortable being associated with that website at that time. People who want to avoid being subject to a government investigation might rationally choose not to access such a site.

Hopefully you can see that this set of circumstances creates a real tension between the First Amendment right that people have to access and read publicly disseminated information that is relevant to current events, and the existence of a government database of historical Internet meta-data that might be accessed only for valid foreign intelligence or anti-terrorism reasons. Of course, the present case is about telecom meta-data and not Internet meta-data. However, I would assert that the same sort of deterrence effect might reasonably exist in the context of telecom meta-data. If you choose to place a phone call to anyone else who might become the subject of a national security related investigation at any time in the future, your association with them is in the record. Thus you have to ask yourself, when you call someone, not just who they are today and whether the government might take issue with them today, but who they might become in the future as well as how the rules governing access to the government's database might change in the future.

In my view, for what its worth, the fact that you have to ask yourself these kinds of questions before communicating with someone or potentially before visiting a website does implicate First Amendment rights, and any government program that raises these concerns should be subject to strict scrutiny. This is distinct from a circumstance in which you have to speculate about whether or not the data is being collected at all, as in such a situation you can rationalize that the government may not be collecting data, and you can adapt to a change in either the government's approach to data collection or the purposes to which data is being used by the government in real time, instead of having to speculate in advance about what might change in the future.