When one nation launches a missile at another, it's easy to pinpoint the aggressor. But during a cyber attack, the aggressor may not be so identifiable, and the traditional rules of warfare don't quite fit.
As nations increasingly develop their cyber offenses and defenses, an international think tank in Estonia is researching a range of legal questions and concepts around clashes in cyberspace.
One of those questions is how to label these skirmishes and whether it's appropriate to call them "cyber warfare" or "cyber conflict," said Rain Ottis, a scientist with the Cooperative Cyber Defense Center of Excellence in Tallinn (CCDCOE).
The CCDCOE was launched in May 2008 to help NATO countries deal with ever-growing cyberthreats by focusing on defense tactics, training, protection of critical national infrastructure, and policy and legal issues.
Although several nations have experienced significant cyberattacks, "we don't have a single good instance of real cyber warfare," Ottis said. He believes that warfare occurs between states.
"We are trying to come up with a way to explain this in a more formal way so not everything by default is cyber warfare," Ottis said. "Personally, I don't want to devalue the word 'war.'"
How to define a cyber incident is one of the topics on the agenda for the CCDCOE's 2010 Conference on Cyber Conflict in June, which will include a new legal and policy track.
CCDCOE researchers are also part of a working group studying the laws of armed conflict to see how cyber attacks should be interpreted. The laws of war, encompassed in international treaties -- some of which are more than 100 years old -- deal with issues such as when a nation can go to war and what is considered legal when at war, Ottis said.
It's brand-new legal territory, but one with which nations will soon have to deal. "When the first cyber war kicks off, mostly likely in conjunction with a physical war, all of these questions will come up in a hurry," Ottis said.
The working group will eventually write a manual for how cyber conflict fits into the existing laws of war.
The CCDCOE is also looking into how Cold War-era concepts such as deterrence fit into cyberspace. Deterrence -- which is based on meeting aggression with greater aggression -- doesn't quite apply, said Kenneth Geers, a civilian with the U.S. Navy's Naval Criminal Investigative Services who is assigned to the CCDCOE.
Geers presented a paper last October in Moscow on deterrence in cyberspace. One of the problems with deterrence is attribution, or identifying the enemy.
"It's really easy to hide in cyberspace," Geers said. "You need much more than computer log files to know what happened."
The basic building blocks of deterrence are capability, communication and credibility. There's also the question of whether a physical response such as bombing is appropriate.
"You have to be able to get back at the aggressor, and in cyberspace, there's no guarantee of that," Geers said. "You may not know who is attacking you, and to get back at them, you have to hack back or do a kinetic response."
It is hard to deter an aggressor who can invest a small amount and cause the target 100-fold damage, Geers said.
Geers is also writing a paper exploring how the 1997 Chemical Weapons Convention (CWC) could be used as an arms control model for cyberspace, exploring concepts such as prohibitions and inspections. Again, cyberspace poses vexing questions.
"There's just not a way, given the fact there are gigs of data on something the size of a stick of gum, that you can possibly verify that no malicious code exists anywhere," Geers said.