The Bush administration has repeatedly claimed that the lapse of the Protect America Act, which expired Feb 16 and has yet to be replaced, has significantly impeded the war on terrorism. I think this is probably complete nonsense, and not just because alternative legal tools are available: The Foreign Intelligence Surveillance Act, for example, is tried and true and has the undoubted advantage, for officials genuinely interested in catching suspected terrorists and putting them on trial, of having withstood many challenges under the Fourth Amendment. The same cannot be said for the PAA or the National Security Agency's warrantless wiretapping program, which was the forerunner of the PAA.
But it is unclear why the lapse of a mere law presents any obstacle to an administration that has argued that it has the right under the Constitution to wiretap suspects for national security purposes without first getting a warrant.
Meanwhile, I've been taken to task by Glenn Greenwald, of Salon.com, and Lisa Graves, deputy director of the Center for National Security Studies. I was wrong, they say, to suggest that the PAA did not really broaden wiretapping authority beyond what FISA had authorized since 1978. They both insist that the PAA authorized surveillance, inside the United States, of phone calls and emails into and out of the country, thus exposing millions of innocent Americans to the risk of being wiretapped when they made or received international calls.
I think I have to eat my own words, because, I agree, it looks as if I did misread the law. But in my defense I must say that the law is very confusing. The FISA was complex to start with, but the PAA seems to have been written intentionally to confuse. (For convenience's sake, I will use the present tense when discussing the PAA).
It's difficult to attempt to explain this without getting into mind-numbing technical detail, but there are two provisions in the PAA that appear almost contradictory. I had thought the second of the two provisions was the important one but I now see I was wrong -- I think.
The first of the two provisions, called section 105A, states that "electronic surveillance" as it is defined by FISA, must not be interpreted as including surveillance aimed at people "reasonably believed" to be outside of the United States. The second provision, called section 105B (a) (2), states that before either the attorney general or director of national intelligence authorizes the acquisition of foreign intelligence information under the authority of the PAA, he must ensure that the eavesdropping does not constitute electronic surveillance -- which is to say, once again, electronic surveillance as the term is defined by FISA.
And so 105B (a) (2), the second provision, which seems to be intended to guard against trespassing on FISA's authority to require warrants for electronic surveillance inside the United States, is limited by 105A, which has already modified what "electronic surveillance" means.
But the two provisions appear to be stuck in a loop of circular logic that seems problematic to me. But of course, it would seem problematic if I haven't understood it; that doesn't mean nobody else has understood it. I do understand that.
But...but...I wonder why 105B (a) (2) is even there in the first place. It seems redundant. If the intention was to make clear that FISA did not apply to suspects outside the country, then 105A, which states that the definition of electronic surveillance must not be interpreted as covering anybody reasonably believed to be overseas, seems to do that adequately. Why did those who drafted the bill insert 105B (a) (2)?
In a report, produced by the Congressional Research Service, that considers the impact of the PAA on FISA, legislative attorney Elizabeth B. Bazan raises an interesting question. She points out that by removing a certain category of surveillance from FISA's definition of electronic surveillance, which is what the PAA accomplished and what civil liberties advocates are concerned about, the PAA may have run afoul of a criminal law which makes intentional interception of communications inside the United States illegal. The law in question, 18 USC 2511, contains a special exception that permits electronic surveillance as defined by FISA.
The crucial question seems to concern where, exactly, the acquisition of foreign intelligence information is, or was, taking place. If the actual interception happens outside the United States, then it is beyond FISA's reach, and always has been. But if intelligence officials inside the United States direct their surveillance at a person reasonably believed to be in a foreign country, and that person calls somebody inside the United States, that would amount to the kind of surveillance Graves and Greenwald are concerned about, that formerly would have needed a warrant under FISA. But Bazan's analysis implies that because, under the PAA, that surveillance would no longer count as "electronic surveillance," then it might be in breach of 18 USC 2511's prohibition against the intentional interception of wire communications, an offense punishable by a fine or imprisonment for up to five years, or both. The disclosure or use of the contents of communications intercepted in breach of 18 USC 2511 would carry the same penalty.
Unless, that is, the administration were to assert that the president's power under Article II of the Constitution was sufficient legal authority for the surveillance (directed at persons reasonably believed to be outside of the United States) that is now removed by the PAA from FISA's definition of electronic surveillance. In that case, Bazan suggests, the constitutionality of the prohibition of such interception by 18 USC 2511 would be for a court to decide.
It's as clear as mud, as any fool can see.