The National Security Act of 1947 established the Central Intelligence Agency and the position of Director of Central Intelligence. In addition to setting the Director’s salary ($14,000 per annum) and giving him/her the power to fire CIA employees, the Act specified the various duties of the new CIA relating to national security and intelligence. It authorized the CIA “to correlate and evaluate intelligence relating to the national security” and to protect sources and methods.
But there is one word that never appears in the authorizing language—and that word is SECRETS. Yet over the decades, the CIA and other intelligence community organizations have defined their responsibilities primarily in terms of secrets.
The Merriam-Webster dictionary lists “information” and “news” as synonyms for intelligence and defines it as “information concerning an enemy or a possible enemy or an area.” Much of the useful information about our potential adversaries has always resided in the public domain; just take a look at any of the declassified National Intelligence Surveys available from the CIA’s electronic reading room—section after section is written at the unclassified level. And yet, we still debate today how important open source information should be for the Intelligence Community.
My contention is that information in the public domain is capable of meeting most of the knowledge needs we have about most issues that concern the United States. There are notable exceptions, of course. Our troops need to know exactly what opposing militaries are trying to hide. Many governments will try to conceal their weapons programs and their intentions. But I think we could make much better use of the taxpayer’s dollar if we only relied upon secret sources and methods for issues that warranted them, and based our sense-making efforts on most issues on unclassified information.
I still remember when I realized the true value of open source information. About 15 years ago, I was reading a book by the noted historian Ian Kershaw—The Hitler Myth. In it, he addressed many of the conventional arguments about Adolf Hitler—that he was a master politician, that he was a charismatic leader, etc., and deconstructed them as best he could to get closer to truth. One chapter had to do with whether Hitler maintained the support of German citizens until the bitter, bitter end. The conventional wisdom was that in fact Hitler remained popular even as the German people suffered greatly in the last year of the war.
Kershaw offered the following facts to argue that German popular support for Hitler did fade. He noted that in one Munich newspaper, obituaries for fallen German soldiers in 1939 always claimed the soldier had died for “Der Führer.” But by 1944, the percentage of soldiers who had died for “Der Führer” had dropped to single digits—at which time it became mandatory to say so. Clearly the views of German families had evolved.
This example of scholarly research evolved my thinking about open source information. Hitler’s popular support was obviously a critical intelligence issue. No doubt millions could have been spent to acquire unique information on the topic.
But a clever and much less expensive research approach was able to deliver important insight. The allies could have tracked soldier’s obituary notices in real time. I couldn’t help but wonder how many current intelligence issues could be similarly tackled if we only pursued promising unclassified approaches.
Ian Kershaw’s The Hitler Myth predates the rise of big data and social media. The world today is full of obvious and non-obvious indicators of social phenomena. We can identify flu epidemics and political shifts based on the nature of Google searches. I always pay attention to the amount of graffiti in cities I visit as an indicator of underlying popular resentment of authority. I just came back from two weeks in Spain; there was a lot of graffiti.
Fully harvesting open source information for national security insights requires some significant tradecraft changes. Much intelligence received by analysts is prepackaged for easy use. It specifically addresses major intelligence questions. Using open source information would require intelligence officers to design research approaches to uncover hidden insights.
Imagination would be an important tool, as would be patience; much trial and error would have to be tolerated. I often wonder how many research dead ends Kershaw pursued until he hit upon an approach that offered real insight. But I’m confident that if it turned its efforts in this direction, the Intelligence Community would be rewarded.
For too long, intelligence agencies have been like the man who loses his car keys late at night. When asked why he only searches for them under the streetlight, he answers that it’s the only place he can see. Secrets are the IC’s streetlight, but the information it needs can often lurk in unsuspected and uninspected corners. Taking better advantage of open source information is one way of turning on more lights.