5.2 KiB
(U) The SIGINT Philosopher: In Praise of Not Knowing
FROM: (U//FOUO) [REDACTED] Run Date: 08/16/2012
(U) I have to begin with a complaint. Hopefully, this will not incite a war with our IAD comrades, but my beef is really with IT types. You've ruined a great word. If you Google the word "agnostic" on your NSANet browser right now, you will find the results rife with tech-y terms like these:
- The PW storage environment is agnostic regarding...
- Comcast's protocol-agnostic...
- Application layer is bearer agnostic...
(U) ...and so on. In all those uses, "agnostic" means something like "indifferent to." This tech-y use of the word has even snuck out into the common vernacular; a leader once called on SID to become "geographically agnostic."
(U) But that's not what the word means. As the classically trained analyst who sits across from me will attest, its Greek roots literally mean "without knowledge." An "agnostic" in religious terms is someone who is uncertain about God's existence, and who possibly believes the question is unknowable. But that doesn't mean agnostics don't care whether God exists.
(S//SI//REL) I ascend this arcane etymological soapbox not just because a word has been abused. Words change meaning in any language, and there's naught I can do about it. But the admission that one does not know something is really an excellent intellectual exercise. I recall one time, while stationed at SUSLAK, when I briefed the South Korean Ministry of National Defense on a North Korean military exercise. The South Koreans had just gotten done giving eleven hypotheses. I explained simply that we weren't sure. The General looked at his countrymen and said, "If you don't know, say you don't know like the Americans."
(C//REL) The IC is full of questions to which we do not know the answer. And yet we are smart, ambitious, conscientious people who were hired to answer those questions, dagnabit, and so we go about trying to answer them industriously. We can take one of two approaches. First, we can write our best guesses, and couch them with all sorts of qualifying language. This is the approach some take with North Korea. Although we may know less about North Korea than any country on earth, hardly a day goes by when my "highlights of IC reporting" e-mail doesn't have a North Korea report in it, in which some industrious thinker has just pontificated that Kim Jong-Un might be open to the West because Mickey Mouse appeared in a North Korean parade. Just by volume, you'd think we actually knew a lot about this country.
(U//FOUO) NSA often takes a second approach, where we attempt to address a very small subset of an important question. Say there is an intelligence need (IN) for Zendian economic information. People want to know how the main sectors of the Zendian economy are performing, whether the needs of the people are being met, and what the high-level plans are to direct the economy. Reasonable questions. But NSA doesn't have the access to answer them. Their reclusive leaders do not use electronic communications we can target. Their missions abroad use triple pig-Latin encryption, and underlying it all is the darn Zendian language that nobody here can even speak. So what do we do? Well, it turns out we can get access to a Zendian gum-chewing manufacturer that has its operations based in the North Pole. So our Zendian Econ team writes 132 reports on Zendian gum sales under the Zendian economic IN. Analysts in other agencies are so happy to get anything, they report that they are grateful for the reporting, which leads leadership here to conclude the team is doing its mission.
(U//FOUO) Now, I've been in positions before where I was working the equivalent of the Zendian gum factory. There wasn't much I saw of use in the collection I had to work with. And if that's you, and that's where you find yourself, you absolutely should try to do something rather than nothing. You're a civil servant, and you should do the best you can to give value for your paycheck.
(U//FOUO) But it would also be nice for those laboring on these kinds of missions to have their leaders occasionally recognize how little we really know.* Often, in their understandable desire to keep morale high and fight for resources, they tend to always focus on the positive: "We got 87 positive feedback reports from customers! Our customers love our stuff!" That's fine, but going overboard with this sort of thing can lead analysts to conclude they have gone crazy. Occasionally hearing "you know, we just aren't able to answer the questions our customers really want" can actually be a refreshing bit of honesty. A little (real) agnosticism would really help.
(U) This kind of honesty can help to "rip off the bandage" and reveal how difficult the task of good intelligence gathering really is. It can prevent throwing good money after bad when leadership believes its current efforts are working. It also helps maintain our ethos as honest, straight-shooting Americans.
(U) But, you know, whatever. I'm agnostic to the whole thing.
(U) Note:
*(U//FOUO) I realize some of you have great access and really can answer a lot of your customers' questions. For you, just know that I hate you, and click on over to a recent Signal v Noise.