From the analysis of shrapnel fragments in a room struck by drones in Pakistan, to a recreation of a secret Syrian detention centre, forensic architects are using new research methods to expose state violence, war crimes and human rights violations.

Architecture is typically understood as a creative process, a discipline concerned with the process of construction, the ways in our material world is built.

Media player: "Space" to play, "M" to mute, "left" and "right" to seek. 

Eyal Weizman, however, is interested in architecture as destruction, that is, the often contested site at which violence and architecture intersect.

The emerging practice of forensic architecture can explain the recent history of sites of armed conflict.
The emerging practice of forensic architecture can explain the recent history of sites of armed conflict. © Getty Images: Anadolu Agency

....

When Professor Weizman and his team were investigating American drone strikes they found the missiles left a signature on buildings: a small hole in the roof. They had to rely on satellite images, but those images have very real limits and make the job of investigators almost impossible.

"The size of the pixel, that is the limit of visibility, in a satellite image is 50cm squared on the ground, while the size of the hole is 30cm squared. So you cannot see those holes from the satellite image, they simply become part of the pixel," he says.

Forensic architecture, then, takes place at the threshold of detectability, that is, the state of visibility at which an object hovers on the brink of being observable or not.

"We always have to be more creative than state and intelligence agencies," Professor Weizman says.

"State forensics is based on a simple principle: the investigators must know more than the criminal. In counter-forensics — civil society forensics that look at state crimes — you have less visibility, you look at things in lower resolution and know less than the perpetrators.

"Here, architecture takes another meaning; it becomes a creative practice of assembling, building models, thinking through solutions, operating under the threshold of detectability."