“White House Agrees to Review of Surveillance Program,” says The New York Times, while The Washington Post‘s headline reads “Bush Compromises On Spying Program.” Something there strikes me as a bit surreal—as though a president’s deigning to permit constitutional review of a massive NSA program of warrantless eavesdropping were an act of noblesse oblige—but according to the Electronic Frontier Foundation, it’s not so much a “compromise” as “a rubber stamp for any future spying program dreamed up by the executive.”

That may be a bit too pessimisitic: The draft bill sponsored by Sen. Arlen Specter (R-Pa.) at least affirms the role of the coordinate branches in overseeing wartime surveillance activities, as against the president’s “Infinite Crisis” theory of executive power. But there are a couple worrying things in there on the basis of my very brief skim. EFF claims that the bill “creates a process for the executive branch to seek court review of its secret surveillance programs” but “doesn’t actually require the government to do so.” That’s not obvious to me on the basis of the bill’s language, but it’s certainly one reading—and doubtless one that will be asserted in the future whenever a president would prefer to forego review. But the draft bill also shifts any challenges to electronic surveillance programs to the secretive FISA court, with disclosure of information about such programs to opposing counsel explicitly at the discretion of the Attorney General or Director of National Intelligence. And then there’s this:

Notwithstanding any other law, the President, through the Attorney General, may authorize electronic surveillance without a court order under this title to acquire foreign intelligence information for periods of up to 1 year [if directed solely at]

(i) the acquisition of the contents of communications of a foreign power, as defined in section 101(a), or an agent of a foreign power as defined in section 101(b)(1); or

(ii) the acquisition of technical intelligence, other than the spoken communications of individuals, from property or premises under the open and exclusive control of a foreign power.

Since that gives a pretty broad window for surveillance without authorization, it seems like it’s fairly important what’s covered there. It seems as though it might include at least the foreign half of any communication to a U.S. person from a foreign one, and possibly also non-voice communications from U.S. persons that are stored abroad. More once I’ve had a chance to peruse the bill (and the statutes it modifies) more thoroughly.

Addendum: Regarding Jacob’s suggestion at Hit & Run that all surveillance be authorized only for specific targets named in advance, this is actually a problem I take up at some length in an article that should show up in a future issue of the magazine. And I find myself agreeing with some of the tech and legal scholars I talked to for that piece, that current technology may require a different way of thinking about oversight and privacy protection in searches. To be very brief, it might be that the optimal model for certain kinds of surveillance is a tiered model where, for example, you have systemic oversight of a broad information gathering program by some body of judges and technical experts, and then a second layer when intelligence officers want permission to attach personally identifying information to some communication flagged by an AI.

Second Addendum: Jack Balkin thinks it’s very bad indeed. One crucial bit that had slipped my notice is that surveillance programs may target “a person reasonably believed to have communication with or be associated with a foreign power that is engaged in international terrorism activities or in preparation therefore or an agent of a foreign power that is engaged in international terrorism activities or in preparation therefore.” That’s huge.