Why there's no silver bullet for cyber deterrence

Ever since Russia’s efforts to meddle in the 2016 election came to light, U.S. officials -- particularly a number of senators -- have been calling for a comprehensive cyber deterrence strategy. While a number of former officials have testified that a silver-bullet deterrence strategy is unlikely, one former Pentagon executive said deterrence is possible, but complicated.

Unlike nuclear deterrence, which deals with relatively few actors and variables, cyber deterrence requires addressing a wide range of threats, actors, unknown capabilities and escalation potentials, said James Miller, former under secretary of defense for policy.

Speaking at a Brookings Institution panel on cyber threats, Miller, who also served as co-chair of the Defense Science Board’s Task Force on Cyber Deterrence, said deterrence must be tailored to individual actors, which he identified primarily as Russia, China, North Korea, Iran and terrorist groups.

"And in order to deter them -- rather than just respond to them -- you need to have a plan in advance, you need to communicate to some degree your capabilities and your intent to respond," he said. "Some of that is done through action, not just through speeches."

Miller said the U.S. needs to do more to release information about its capabilities, demonstrate them in military exercises and use them in the real world in response to adversaries' actions.

"The first tools that we pick off of that shelf are going to be diplomacy and economic sanctions and political steps to reinforce our alliances and partnerships," he said. "But without question, offensive cyber needs to be a part of that mix."

Deterrence must be forward looking and anticipate escalation and counter retaliation as much as possible, he said.

"Each one of those actors has the potential to escalate beyond an initial cyberattack ... and we need to think clearly through what that escalation may look like," Miller said. "And when we're responding, [we need] to aim to deter them, not just from future cyberattacks ... but to deter them from escalating against us or our partners and allies."

The Defense Science Board looked at three deterrence models to find elements that can be drawn into a cyber deterrence framework, Miller said.

The criminal justice model requires giving actors the sense they might get caught and punished in order to convince them not to take the risk of launching an attack, he explained.

The conventional deterrence model, he said, requires signaling -- for example, flying bombers over the Korean peninsula -- to make an adversary aware of your capabilities and think twice.

He said one lesson from nuclear deterrence is that a strategy for Russia will differ from a strategy for North Korea or for Iran, which reinforces his argument that cyber deterrence has to be heavily tailored to each adversary.

"One thing we shouldn't take from [nuclear deterrence], people who spent their careers in nuclear deterrence tend to think that defenses are irrelevant," Miller said. "In cyber deterrence that's just not the case. Defenses and resilience is a fundamental starting point for effective deterrence posture, both to raise the bar for even the most capable actors and to crowd out others, and for the credibility associated with any response."

In terms of actually formulating a strategic framework, Miller told FCW that it would take several months to develop basic campaign plans for each threat actor.

"It will take a period several months beyond that to do realistic table-top war gaming to play them through -- and that has to be an interagency process," he said. "That's complicated always at the beginning of an administration," especially one that is behind the curve in appointing senior officials, he added.

"The third step is to begin execution, and that can start in small bits," he said. "So the bottom line is six months should be enough to get the front-end execution of a comprehensive campaign plan for each of these potential adversaries."

Those potential adversaries are really individuals more so than states, he said. "You don't really deter states, you deter individuals and group decision-makers, and it's going to have to be changed over time and set in a broader context of our international relations with these countries."

Deterrence campaigns have to be extremely flexible, signaling to adversaries that the strategy could change on a weekly or even daily basis, he said.

"It means setting priorities -- you can't deter everything," Miller said. "Deterring Russia from hacking our mid-term election has got to be very high on the list with respect to them, and we have to communicate that to them in a way that doesn’t imply that everything else is OK."