Hanging by a ThinThread [Stephen Spruiell]
The Baltimore Sun has a puzzling story up today. In it, reporter Siobhan Gorman reports that the National Security Agency had been working on some kind of phone records database since the late 1990s. Gorman's anonymous intelligence-community sources clearly preferred the original program — called ThinThread — stating three reasons why it was superior to the program described last week in USA Today:
- All the phone numbers in the database were encrypted to protect privacy;
- It had a component that monitored for abuse; and
- It just worked better for reasons that the article doesn't specify.
This is the truly remarkable passage:
Officials say that after the successful tests of ThinThread in 1998, Taylor argued that the NSA should implement the full program. He later told the 9/11 Commission that ThinThread could have identified the hijackers had it been in place before the attacks, according to an intelligence expert close to the commission.
But at the time, NSA lawyers viewed the program as too aggressive. At that point, the NSA's authority was limited strictly to overseas communications, with the FBI responsible for analyzing domestic calls. The lawyers feared that expanding NSA data collection to include communications in the United States could violate civil liberties, even with the encryption function.
Taylor had an intense meeting with Hayden and NSA lawyers. "It was a very emotional debate," recalled a former intelligence official. "Eventually it was rejected by [NSA] lawyers."
After the 2001 attacks, the NSA lawyers who had blocked the program reversed their position and approved the use of the program without the enhanced technology to sift out terrorist communications and without the encryption protections.
The NSA's new legal analysis was based on the commander in chief's powers during war, said former officials familiar with the program. The Bush administration's defense has rested largely on that argument since the warrantless surveillance program became public in December.
If Gorman's sources are correct, a program that could have identified the 9/11 hijackers was derailed because, "[NSA] lawyers feared that expanding NSA data collection to include communications in the United States could violate civil liberties, even with the encryption function."
Gorman also reports:
The strength of ThinThread's approach is that by encrypting information on Americans, it is legal regardless of whether the country is at war, according to one intelligence official.
This passage directly contradicts what the NSA lawyers said before 9/11, which is that the program would violate civil liberties even with encryption. Encryption was irrelevant to the NSA lawyers' legal analysis. Encryption didn't change their minds about the program's legality — the 9/11 attacks did. When Congress authorized the president to respond, it gave him the power to intercept enemy communications as part of his constitutional duty to repel foreign threats. We faced a new type of enemy. The rules changed.
What bothers me about finger-pointing stories like this one is that the focus is placed on internecine disputes and political attacks. But out of all the noise in this article, a signal comes through loud and clear:The NSA had developed this potentially life-saving program before 9/11, but privacy concerns prevented its implementation. The 9/11 attacks changed all that. Now that the program's successor has been leaked to the media, the privacy absolutists want to undermine it again.
ONE MORE THING: I'm not saying that the encryption and abuse-prevention components of ThinThread aren't desirable. But I'm also not knowledgeable enough about them to know what potential drawbacks led the NSA to scrap them (and while the sources for this article are knowledgeable, they obviously have an ax to grind). My point is that such details are irrelevant to the legal analysis. Either you think this type of intelligence-gathering falls within the powers of the executive to command the military, or you don't.
05/18 01:15 PMShare