“If I can’t use this information, then what good is it to have it?! Why even collect it in the first place?!”
It’s a cry of frustration, an angry rhetorical exclamation I heard many times during my 30-year career as an operations officer at the CIA. Usually it comes from ambassadors or senior members of the national security apparatus in Washington, and occasionally even from analysts in the intelligence community who have been provided with a truly stunning piece of information acquired clandestinely from human or technical sources. The sense of frustration among these consumers of intelligence is heightened when the topic is critical and timely, and when both the government and the American public are clamoring for answers to difficult questions.
This is precisely where we as a nation find ourselves when discussing the claim by the U.S. intelligence community that Russian hackers attempted to influence the U.S. presidential elections. The White House has ordered a report on what the hackers did, and to what extent they were trying to help Donald Trump, before Trump is sworn in as president Jan. 20. News reports say officials at the FBI and the CIA have agreed that hackers targeted the Democratic National Committee and Hillary Clinton’s campaign chairman, John Podesta, to boost Trump. President Obama suggested this month that Russian President Vladimir Putin knew about the hacks.
But proving that case to the public at large will get complicated. The discussion about the cyber intrusions and what U.S. intelligence agencies know about them goes directly to some of the most sensitive questions in the business: the best way to protect sources and methods, and how to use clandestinely acquired information to resolve a politically charged issue. Secrets intelligence agencies want to keep compete with an understandable desire to reveal them. Facts may help resolve the matter, but in revealing the facts, the government may also reveal how we got them. It is truly not an overstatement to say that technical capabilities we have spent years and millions to develop could be rendered useless in one news cycle if disclosure is not handled correctly. Worse — and I do not exaggerate — if it were human sources that provided the information, they could lose their lives.
This is the central tension between clandestine intelligence collection and an open society, the conundrum facing intelligence officials and politicians alike. Normally in a democracy, ideas such as transparency, direct attribution, fact-checking, research and the like are viewed as virtues. In fact, we often gauge how democratic a society is by such measures: How much of the internal workings of a government can we see? Are journalists and researchers hindered by a government when doing their work? The problem with clandestine intelligence is that while all of those virtues are alive and well in the intelligence community — really, they are — they exist inside a closed system. Operations officers routinely keep the identity of sensitive sources from analysts (with exceptions, such as in some counterterrorism work). Professional methodologies ensure that sources are validated, and their information is vetted: Multiple sources providing information on the same topic are compared and analyzed; clandestinely acquired information is layered onto known facts to test for reliability; human intelligence is compared to intelligence acquired from other intelligence disciplines such as signals intelligence or overhead. For the most part, that system works. Analysts and collectors work together closely on this, and it is in both their interests to ensure fabrication does not occur. There are oversight committees in the House and Senate that monitor clandestine collection and intelligence analysis — but they are select committees, whose members commit to keeping the secrets shared with them to themselves. And for the most part, they do.
There are systems in place, in other words, to ensure good intelligence is shared securely within the government. There are not many systems in place that are designed to read the public in.
Our intelligence agencies were designed to provide intelligence primarily to consumers in the executive branch — the president, White House aides, the Cabinet, the Pentagon and other national security officials. The idea was to ensure policymakers would have the best information so they could make the decisions for which American voters elected their administration. While the system has evolved, and members of the legislative branch and local and even tribal officials receive some sensitive information — especially on terrorism-related cases — public disclosure of intelligence was not really envisioned. Eventual declassification was considered, but only after a buffer of many years had passed, usually at least a decade. Secrets stolen at great risk were intended for a relatively small group of senior officials trusted by the American electorate to make crucial decisions.
None of that will be very satisfying to those outside the intelligence community. Cynics will understandably decry any call from intelligence agencies to “just trust us,” especially given that U.S. intelligence agencies have not always gotten it right. In this era of unprecedented sharing on the Internet, the idea of providing intelligence only to senior government policymakers can also seem dated. These concerns are a normal and healthy part of our democracy. Even though I worked hard in the past to protect CIA human sources and other methods, I do not begrudge those who call for investigations and transparency. It helps keep everyone honest.
But here is the reality: Human sources overseas understand viscerally the risks involved in stealing secrets and passing them clandestinely to the U.S. government. They are also highly sensitive to current events, given today’s interconnected world. Sources will tell their case officers, “I know you will want to use the information I have, but to do so will put me at great risk. Can you guarantee that you and your organization will protect me?” If they sense there is no such guarantee — or worse, if there are examples of when such guarantees were useless — they will self-edit. Sometimes they will simply refuse to report. Similarly, if technical collection methodologies are made public, the adversary — in this case, Russia — will take quick action to cut off the technical accesses gained after years of careful work.
In both cases, the flow of intelligence simply ends, and future collection can be seriously limited. New sources have to be recruited; new technical operations have to be undertaken. It takes years and is expensive, which is why the government uses intelligence collection as a last resort when attempting to obtain information abroad. Collecting information overtly, such as when the State Department uses its officers for official contact with foreign governments, carries less risk and less expense.
Is it not possible to find some way to use clandestinely acquired information in a more public way? Can’t we figure out how to declassify at least some of the intelligence we have on the Russian hacks without jeopardizing sources and methods? Sometimes, yes. While it still makes collectors and even analysts nervous to release it, finished intelligence — which can be worded in a way that does not reveal whether the source was human or technical — provides greater source protection than raw intelligence, which typically indicates how the information was collected. Some details in the reporting can also be omitted to further obscure sensitive collection details. In fact, “writing for source protection” happens daily when the intelligence community provides consumers in the government with information.
But when politics is injected into the equation, all bets are off. This is exponentially more true the more partisan and bitter the politics. Normally, when U.S. collectors obtain highly classified Russian information that has no bearing on domestic politics, government consumers get the intel and it contributes (hopefully) to better policy. But when the government is divided politically and many of the differences are playing out in public, the chances of that highly classified information leaking — from both sides of the political divide — become unbearably high.
So as Lenin once famously asked, what is to be done? On the one hand, it is important for people to understand that the Russian hacks were just one part of a much broader attempt to sway American opinion using propaganda methodologies they have perfected over decades, such as disseminating false news to favor one candidate over another. The intelligence we have collected would probably help in providing a better understanding. But revealing anything could quickly become a problem. If one piece of intelligence is revealed, one political side or the other will almost certainly feel the information favors their adversaries. They will demand additional information. Worse, questions like, “How exactly did you get that information?” or “Where did that come from?” and “When precisely did you know that?” will inevitably be asked — and the protection of sources and methods will begin to erode.
This slippery slope will not end well and will likely damage the intelligence community’s ability to collect in the future. We may simply be forced to make a Solomonic choice, trying to split the baby down the middle by releasing some of the intelligence but not so much as to compromise current and future collection. This path will be unlikely to please either intelligence officers arguing for source protection or those outside the intelligence community clamoring for a more complete release of information. Of course, it is not the Russians to blame for this last problem, but rather our own fractured political system.