We live in an amazing information age. For the first time in human civilization, anyone—well anyone with access to the internet—can find out just about anything about any topic. A few words typed into a search engine, hit enter, click on a link, and voila! Instant knowledge.
Of course it is not as easy as all that. Just because a query returns some data doesn’t mean that all the information is accurate. We also live in an era of disinformation and misinformation, some of it spread deliberately through malicious actors (hello Russian troll factories!), and some shared by people who do not take the time to research whether what they post is really true (no, a popular English swear word did not arise from the phrase for “unlawful carnal knowledge” or “fornication under consent of the King”).
As a consequence, it is still difficult to decide whether any given source—the worldwide web or your Aunt Tessie—is a good source. And that makes it a challenge to figure out which information to use, which to doubt, and which to reject. Maybe we will get to a point where we can reliably make that determination, but with the advent of chatbots—and 10,000 years of human obfuscation behind us—that is not likely.
Which brings me to intelligence. At its heart, intelligence is simply information (the French word “renseignements” in fact covers both ideas), albeit information that is collected “secretly.” You know, all that cloak and dagger stuff. And, like all information, it can be good or bad.
Those of us who worked in national security had a job to do: Collect intelligence, assess it for accuracy, and provide our findings to our paymasters, i.e., the government. If we were to regularly pass on inaccurate information we would not be in our positions very long. This is why we used a number of tests to our data, the chief ones being source verification and corroboration from multiple sources, to allow us to confidently state that what we have and share is the best information possible under the circumstances (time constraints, difficulty in processing data (e.g. encryption), sensitivity of what we are reporting on, etc.)
At the same time, we have to protect our sources and methods. The old military adage “loose lips sink ships” still applies. The disclosure of a source or method can lead to the disappearance of both. If a given piece is obtained through a specialized technique and that approach is compromised, you can kiss your source goodbye as the “target” makes changes to prevent future exploitation. Leak a human source’s identity and that source can end up dead, or turned to provide intelligence on you to your adversary.
This is why intelligence agencies never, or very, very rarely, disclose sources to their customers. We may provide generalized information on the origins of what we have passed on with phrases like “reliable,” adding qualifiers such as “confirmed,” “believed,” or “uncertain,” but we do not say, “Oh by the way, John Doe of 123 Main Street is our human asset.” We expect those who consume intelligence to understand this and trust that we would not willfully provide intelligence we knew to be inaccurate or for which we had significant doubts.
Intelligence officials know that their contribution to national security and decision-making is but one piece in a vast bucket of information and are not naïve in thinking it is the only data available. Furthermore, we accept that our job is to advise government, not to tell it what to do. At the same time, we expect those who read our products to trust that they have been prepared with the utmost care and that all steps have been taken to ensure accuracy and relevance.
The government does not want to use what we know? Fine, that is its prerogative. Just don’t tell us that we do not know what we are doing.
