The NSA and Compartmentalization

This week, the NSA has found itself in the middle of a couple of scandals. What people are not realizing is that the NSA exploits compartmentalization quite well.

On one hand, it’s used to restrict classified information to individuals or teams on a need-to-know basis. On the other hand, the NSA uses compartmentalization to counter oversight and limit the visible scope of a goal.

Take the call metadata program, for example. I’m not sure whether the program name has been leaked (PRISM is another program entirely), but let’s call it “HORSESHIT”.

While HORSESHIT’s scope is collecting call metadata (phone numbers, date & times, who they called, etc) you can virtually guarantee HORSESHIT has sister programs (let’s make up the name FROGBUTT) that DOES connect the information from the HORSESHIT program to the personal data collected in the FROGBUTT program. There may be an additional program called DONKEYSNOT that records the contents of all the phone calls in the US, this is, in turn, chained to the data from the other programs.

It’s compartmentalized. If information from one program is compromised, only one part of the overall picture becomes unveiled. The NSA can plausibly deny the actual scope of a program. Citizens think “Whew! They’re only collecting my call metadata, that’s not so bad!”. It lessens damage to “national security”. Jigsaw intelligence.