My good friend and fellow libertarian Professor Randy Barnett published a provocative column in the Wall Street Journal yesterday taking the position that the federal government's creation of massive databases should be regarded as unconstitutional. Much of his argument pertains to the operation of the NSA program, but he also notes the Consumer Financial Protection Bureau's efforts to create massive databases of vital credit information from consumers in the name of protecting them from financial institutions.
I strongly agree with Barnett that the activities of the CFPB should be viewed with great suspicion. This intrusive operation is inconsistent with any coherent libertarian view of the relationship between individuals and goverment. I think that I speak not only for myself but for millions of other individuals who do not want the CFPB messing around with my financial information under some wishy-washy, paternalist rationale. As best I can tell, most of the bureau's interventions are foolish or worse. There is no legitimate social interest that is served by this activity.
Also note that the collected data is not kept in a lockbox, but is directly available for inspection and use by government agents, without any safeguards for how it is used. I do not care whether the Congress approves of these actions -- I think that they count as unreasonable searches and seizures and should be stopped forthwith. That judgment does not, of course, carry over to the broad class of “required records” that are needed to run, for example, a system of revenue collection. But the view that "We are doing this for your benefit, oh citizen" leaves me completely and totally cold.
Barnett and I disagree, however, on the matter of the NSA, on which I give the national security interest much more weight than he does. In the case of the NSA, the internal controls are far stronger than those with the CFPB. Also, unlike the CFPB, the NSA does vital national work that we cannot possibly do for ourselves. I shall not repeat here the arguments that I made first with Roger Pilon in the Chicago Tribune, nor those that I made with Mario Loyola in the Weekly Standard, nor indeed those that I advanced on Ricochet (here and here), except to say that it makes all the difference in the world whether the information is kept in a credible lockbox once collected, so that it may only be used by government officials on a showing of probable cause.
Nonetheless, I hasten to add, as I said on the latest edition of Law Talk, that I was deeply disturbed to learn about some particulars of the FISA court that require urgent Congressional attention. I realize that it is highly dangerous to publish redacted versions of the court's opinions, lest they contain information of value to the nation’s enemies. But it is precisely because that information cannot be easily accessed that extra care needs to be taken to make sure that those aspects of the program that can be made public operate in ways that are above suspicion. Here are some concerns that track a recent New York Times story by Eric Lichtblau on the question.
-- First, the recent revelation that all appointees to the FISA Court are made by the Chief Justice of the Supreme Court is deeply troublesome because there is no reason to concentrate so much power in the hands of a single person on such a critical issue. Splitting the job so that two more justices also make appointments would be a simple reform that could go a long way.
-- Second, it leaves me very uneasy that 10 of the 11 FISA court judges are Republicans. This is an issue where appearances matter and Democratic judges should be part of the system.
-- Third, the expansion of the FISA court jurisdiction to such matters as cyberterrorism may well be warranted. But what is disturbing is that it is done by judicial opinions that are largely secret. There is no difficulty in having a public debate in Congress. I suspect this conversation could be taken up in a public forum without compromising any operational portion of the program.
-- Fourth, it is unnerving to see the very high (99+ percent) rate of approvals on surveillance requests to the FISA court. That could well be justified, but the optics are bad when there is no advocate for the other side that is brought into the process. There is no need to let private parties in, but government agencies should have a permanent staff whose charge is to argue against some of the warrants on questions of scope. It could be that all is well. But there is no harm in finding out for certain.
Where I take strong exception to Barnett's position on the NSA is in this proposition from his piece:
With the NSA's surveillance program, the Foreign Intelligence Surveillance Court has apparently secretly approved the blanket seizure of data on every American so this "metadata" can later provide the probable cause for a particular search. Such indiscriminate data seizures are the epitome of "unreasonable," akin to the "general warrants" issued by the Crown to authorize searches of Colonial Americans.
In principle, I think that the comparison to general warrants from the colonial era is way off base. As Barnett notes, the early cases allowed damage actions against unreasonable searches and seizures, for in cases like Entick v. Carrington in 1765, government agents did indeed trash private homes. But the reason why damages fell out of style was because most searches did not cause that kind of damage. Assuming that surveilled data is kept under lock and key, the greater danger would be for that data to be unavailable when some serious national security danger arises.
I think that extraordinary care is needed to make sure that these programs operate as advertised, which is why the recent information about the FISA Court raises concern. At its core, however, the NSA program is well designed. The real tragedy here is that serious mistakes on fixable matters could undermine the credibility of a program that should be reformed, not scrapped.