After the attacks at Charlie Hebdo in Paris, January 2015, security was tightened all over Europe. It was announced that the security and intelligence agencies of the member states would intensify their collaboration, specifically in the domain of data compatibility and exchange, in an attempt to tighten security to such an extent that new terrorist attacks would become impossible - intelligence would ensure that plans would be known long ahead of the moment they were planned to be executed.

Evidently, the Paris attacks on Friday 13th of November 2015 demonstrated the opposite. A relatively large group of terrorists infiltrated Paris in a coordinated shoot-out, bombing and hostage operation causing death and injuries to a large random group of victims.

What is more, the surviving terrorists escaped. One member of the terrorist team - Salah Abdeslam - made his way back across a heavily patrolled border into Belgium; his car was stopped but he was not recognized by the patrolling police officers. In the days that followed, Belgian security and intelligence forces believed that he was hiding in the Brussels district of Molenbeek. The Belgian government declared the highest level of terror alert, and an enormous police action in Molenbeek was launched on 22 November. Armed soldiers patrolled the streets of Brussels (and were deployed in other cities as well), the Brussels Metro was closed, and schools would remain closed until further notice.

The massive raid of 22 November didn't yield any results. A large handful of people were arrested, but all but one of them were released the next day. No weapons, bombs or other terrorist equipment were found. And, humiliatingly, Abdeslam was not captured. He had escaped.

The surveillance structure now put in place, in which the privacy of all citizens is sacrificed to the priorities of a safe public space, and in which all communication traffic is stored as "data" enabling security forces to achieve an unprecedented level of accuracy in identifying possible and effective criminals, to anticipate their plans and to locate their whereabouts - this structure appears to have failed, both in Paris and in Molenbeek, and probably in every other search currently being conducted across Europe. It has failed systematically throughout the entire episode.

The reason for this is quite simple for those who are interested in human behavior: the focus on electronic systems of communication narrows human behavior to just a segment of what we do, believing that this segment suffices to capture the totality of behavior. Empirically, we now see, this assumption is wrong. Abdeslam and his colleagues, at several points in the past weeks, operated outside of the structures of electronic surveillance, and they did so in a way that kept them adequately informed and alert to the moves of the security forces. Talking face-to-face, or through a network of face-to-face relays, for instance, still exists as a mode of communication, and it cannot possibly be traced by surveillance systems.

Increasing the capacity of this surveillance system will not, for reasons too clear to warrant much explanation, improve its accuracy. It will still cover just a fraction of the social environment in which real people deploy real social behavior and leave the rest unchecked. While, in return, the privacy of an entire population has been surrendered, and might be invaded whenever - for reasons never disclosed to the target - one becomes a "suspect" in the eyes of the surveillance operators.

There is a tendency, when people become desperate, to do more of the same but do it harder, faster and more intensive. In the field of security, the failure of the present system of surveillance should result in a profound revision of its assumptions and methodology. Because while it is definitely totalitarian in scope and pervasiveness, it does not offer total security. In fact, whenever it has to live up to its promises, it appears to fail miserably. It is time to recognize this failure.