The Modular Advanced Armed Robotic System, or MAARS robot, kills human targets from almost two miles away. MAARS robots have motion detectors, acoustic sensors, siren and speaker systems, non-lethal laster dazzlers and encyption technology to make the robotic killer "extremely safe and tamper proof," says DARPA. (U.S. Army/U.S. Army)

Dina Temple-Raston is NPR’s counterterrorism correspondent and the author of four books. She recently finished a Nieman Fellowship at Harvard University, where she studied big data and the intelligence community.

Several years ago, the Defense Advanced Research Projects Agency (DARPA) began looking into a new medical procedure called transcranial direct-current stimulation, or tDCS. To the outsider, the experiments looked more Mary Shelley than cutting-edge neurology. Scientists taped sponges laced with metal to various parts of a patient’s scalp and then hooked them up to a small 9-volt battery — just like the ones that go in smoke detectors or old transistor radios. Then the scientists sent a small electrical pulse to the brain and watched to see what would happen.

What they discovered was extraordinary: Stroke patients improved their motor skills, chronic pain sufferers appeared to find relief, and the stimulation seemed to help some patients with learning. It was this last revelation that interested DARPA most. The agency gave scientists at the University of New Mexico a grant to see whether tDCS might have a military application — specifically, whether it could be used to help soldiers train faster.

Entertaining the idea of a better military through the judicious use of a 9-volt battery goes a long way toward explaining why the mere mention of DARPA often is accompanied by a full caseload of moral judgments. The agency has long had a habit of eschewing political correctness and shrugging off what people might say or think so it can put science first. It has never worried about how unconventional research might look on the front page of, say, The Washington Post. That attitude has led to notable successes but also some troubling outcomes. DARPA is responsible for stealth technology, tank simulators and the M-16 rifle on the one side of the ledger, but on the other side data-mining programs such as Total Information Awareness and the research that led to harsh interrogation techniques used on prisoners after 9/11.

Annie Jacobsen explores that tension in her fascinating new book, “The Pentagon’s Brain,” which she presents as the first comprehensive history of an agency many Americans may not even know exists. Jacobsen tracks DARPA’s beginnings as an informal gathering of scientists struggling with problems of the Cold War and allows readers to see its transformation into what it is today: a high-tech incubator that introduces the newest technologies, for good or ill, to soldiers on the battlefield.

DARPA’s Extreme Accuracy Tasked Ordnance (EXACTO) program, is developing a self-steering bullet to increase hit rates for difficult, long-distance shots. (DARPAtv via YouTube)

According to Jacobsen, many of DARPA’s original members were working in the Southern California offices of the Rand Corp. think tank in the 1950s. At the time, they were churning out reports about nuclear weaponry and doomsday scenarios. Jacobsen writes that “competition was valued and encouraged at RAND, with scientists and analysts always working to outdo one another.” And nowhere was that competitive spirit more apparent than at lunchtime, when the scientists began playing Kriegspiel, a chess variant once favored by the German military. With maps of the world spread across lunch tables, the great minds of the Cold War era would spend hours on the game. “Lunchtime war games included at least one person in the role of umpire,” Jacobsen writes, “which usually prevented competitions from getting out of hand. Still, tempers flared, and sometimes game pieces scattered.”

The master of these games, by Jacobsen’s account, was a mathematician and former child prodigy named John von Neumann. He was considered so bright that he was hired by Rand’s mathematics division on rather unusual terms: He was supposed to “write down his thoughts each morning while shaving, and for those ideas he would be paid $200 a month — the average salary of a full-time RAND analyst at the time.” Eventually, he was also charged with a rather delicate project. He performed the precise calculations that determined at what altitude over Hiroshima and Nagasaki the atomic bombs had to explode in order to kill the maximum number of civilians on the ground. He determined that height was 1,800 feet.

Von Neumann was one of the original members of DARPA and its precursor organizations. Other notables included John Wheeler, a Princeton University physicist who coined the term “black hole”; Herb York, the first director of Lawrence Livermore National Laboratory; theoretical physicist Edward Teller; and a former president of the California Institute of Technology, Marvin “Murph” Goldberger.

It wasn’t until Congress formally created DARPA in 1958 that its key advisers moved from the lunch tables at Rand and other think tanks and took on a more conventional shape. What were once brainstorming sessions at summer homes on the Cape became official meetings at the National War College at Fort McNair in Washington. The Defense Department gave the group a code name — Project 137 — and the scientists received security clearances so they could discuss sensitive information while they identified “programs not now receiving adequate attention” in the national security realm.

While the full array of those programs is still classified, Jacobsen does a good job giving readers a sense of what it was like to be in the room. “The scope of the national security threats facing the nation left many of the Project 137 scientists with a deep sense of foreboding,” she says. Quoting from Wheeler’s after-action report: “The group has developed a stronger feeling for and deep appreciation of the great crisis with which the nation is faced. The group senses the rapidly increasing danger into which we are inexorably heading.”

Wheeler’s report is just one of the many sources Jacobsen calls upon to uncover details about DARPA’s secretive government projects. She has some experience writing about these kinds of covert programs. Her previous book, “Operation Paperclip: The Secret Intelligence Program That Brought Nazi Scientists to America,” used newly declassified documents to reveal details about a U.S. effort to employ Nazi scientists as armament specialists during the Cold War. DARPA, she makes clear, was created in a similar atmosphere — a feverish post-Sputnik era when investing in arms and technology seemed like the only way to stay ahead of the Soviets.

Fast-forward to today: With its $3 billion annual budget and its advanced technologies and programs, DARPA is the force behind some of the world’s coolest gizmos. These run the gamut from the Global Positioning System to prosthetic hands that may give amputees a sense of touch. Jacobsen illustrates DARPA’s Zelig-like quality of appearing center stage at opportune moments when it is least expected. Case in point: the 2001 anthrax scare that shuttered the Senate’s Hart Building. “DARPA was asked to provide science advisors to help,” Jacobsen writes. The approach they took was based on technology that DARPA’s Defense Science Office had developed after an E. coli outbreak at Jack in the Box restaurants in 1993: self-sterilizing packages triggered by light or humidity. DARPA adapted that technology to help it quickly decontaminate the office building.

DARPA’s Towed Airborne Lift of Naval Systems (TALONS) research effort demonstrates a prototype of a low-cost, fully automated parafoil system designed to extend maritime vessels’ long-distance communications and improve their domain awareness. (DARPA)

Then there is my personal favorite: the Stealthy Insect Sensor Project, which used honeybees to locate bombs. “Bees have sensing capabilities that outperform the dog’s nose by a trillion parts per second,” Jacobsen writes. “Using Pavlovian techniques, scientists cooled down groups of bees in a refrigerator, then strapped them into tiny boxes using masking tape, leaving their heads, and most of the antennae, poking out the top. Using a sugar water reward system, the scientists trained the bees to use their tongues to ‘sniff out’ explosives, resulting in a reaction the scientists call a ‘purr.’ . . . The bees were tested with various explosives, including TNT and C4” and were working well.

“But when the Army learned that DARPA planned to send bees to Iraq as a countermeasure to the IED threat,” Jacobsen writes, “they rejected the idea. The reality of depending on insect performance in a war zone was implausible.” The decision took the sting out of the project; the bees never deployed.

To produce the book, Jacobsen conducted dozens of interviews with former DARPA members. She clearly has plenty of material to work with, but sometimes it is difficult to know how she feels about it all — she seems conflicted, and that uneasiness is apparent to the reader. An organization that works under the credo “anything imagined can be tried” clearly makes Jacobsen uncomfortable, and she suggests time and again that while the agency’s descent to the dark side isn’t inevitable, it is something to keep watch over. DARPA, Jacobsen writes, was responsible not only for early research into brainwashing and Agent Orange, but also the hearts and minds campaign in Vietnam and post-9/11 data-mining programs that seemed, at a minimum, to dance along the edge of our civil liberties.

Even in the area of brain science, there is reason to be both hopeful and worried. DARPA researchers have “developed and are testing implantable wireless ‘neuroprosthetics’ as a possible means of overcoming amnesia,” Jacobsen writes. The scientists are looking for treatments for post-traumatic stress disorder by adapting transcranial direct-current stimulation. In some cases, they are surgically implanting multiple electrodes in various parts of the brain. The electrodes transmit information to a control center, which can send electrical impulses to the brain to relieve such things as anxiety or fear. It’s the electroshock therapy of old — but by remote control.

Another scientist is trying to build what is essentially an artificial brain. The rubric describing this research is synthetic cognition. “Roboticists define artificial brains as manmade machines designed to be as intelligent, self-aware, and creative as humans,” Jacobsen writes. “No such machine yet exists, but DARPA scientists . . . believe that, given rapid advances in DARPA technologies, one day soon they will.”

For the more than 300,000 Americans who the Department of Veterans Affairs believes returned from Iraq and Afghanistan with brain injuries, DARPA’s search for new remedies sounds like a noble enterprise. But it also gives pause, and this happens often when reading “The Pentagon’s Brain.” One can’t help feeling there is something faintly creepy about it all, which may be exactly what Jacobsen intended.

The Pentagon’s Brain
An Uncensored History of DARPA, America’s Top-Secret Military Research Agency

By Annie Jacobsen
Little, Brown. 552 pp. $30