The Pentagon's Brain (26 page)

Sutherland and Taylor began asking DARPA contractors at various university research laboratories around the country what they thought about the networked computer idea. The feedback was unanimous in favor of it. In general, scientists and engineers were frustrated by how little access to computers they had. This got Sutherland and Taylor thinking. Why not try linking several of these university computers together so the DARPA contractors could share resources? To do so would require building a system of electronic links between different computers, located hundreds of miles apart. It was a radical undertaking, but Sutherland and Taylor believed it could be done.

Bob Taylor went to DARPA director Charles Herzfeld to request enough money to fund a networked connection linking four different university computers, or nodes. Herzfeld told Taylor he thought it sounded like a good idea but he was concerned about reliability. If all four computers were linked together, Herzfeld said, when there was a problem, it meant all four computers would be down at the same time. Thinking on his feet, Taylor said he intended to build a concept into the system called network redundancy. If one connection went down, the messages traveling between the computers would simply take another path. Herzfeld asked how much money Taylor thought be needed. Taylor said a million dollars.

Herzfeld asked, “Is it going to be hard to do?”

“Oh, no. We already know how to do it,” Taylor said, when really he was guessing.

“Great idea,” said Herzfeld. “You’ve got a million dollars more in your budget right now.” Then he told Taylor to get to work.

Taylor left Herzfeld’s office and headed back to his own. He later recalled the astonishment he felt when he looked at his watch. “Jesus Christ,” he thought. “That only took twenty minutes.” Even more consequential was the idea of network redundancy—making sure no single computer could take the system down—that emerged from that meeting. It is why in 2015, no one organization, corporation, or nation can own or completely control the global system of interconnected computer networks known as the Internet. To think it came out of that one meeting, on the fly.

The first four university sites chosen were Stanford Research Institute in northern California; the University of California, Los Angeles; the University of California, Santa Barbara; and the University of Utah in Salt Lake City. In 1969, ARPA contractor Bolt, Beranek and Newman became the first east coast node. By 1972 there were twenty-four nodes, including the Pentagon. The person largely responsible for connecting these nodes was an electrical engineer named Robert Kahn. At the time, Kahn called what he was working on an “internetwork.” Soon it would be shortened to Internet.

This network of ARPA nodes was growing, and Kahn wanted to devise a common language, or protocol, so that all new nodes could communicate with the existing nodes in the same language. To do this, Kahn teamed up with another DARPA program manager named Vint Cerf, and together the men invented the concept of Transmission Control Protocol (TCP) and Internet Protocol (IP), which would allow new nodes seamless access to the ARPANET. Today, TCP/IP remains the core communications protocol of the Internet. By 1973 there were thirty-six ARPANET nodes connected via telephone lines, and a thirty-seventh, in Hawaii, connected by a satellite link. That same year the Norwegian Seismic Array became connected to the ARPANET, and J. C. R.
Licklider’s vision for an “Intergalactic Computer Network” became an international reality.

In 1975 DARPA transferred its ARPANET system over to the Defense Communications Agency, and in 1982 standards for sending and receiving email were put in place. In 1983 the Pentagon split off a military-only network, called MILNET. Today the ARPANET is often referred to as “the most successful project ever undertaken by DARPA.”

Between the advances in computer technology, networking power, and the ARPANET, DARPA was primed for the development of an entirely new C3-based weapons system. Sometime in 1974, DARPA commissioned several classified studies on how the Pentagon could best prepare itself for a Soviet invasion of western Europe. The strategist leading one analysis was the former RAND mathematician Albert Wohlstetter, author of the nuclear second-strike doctrine, or NUTS. Wohlstetter, now a professor at the University of Chicago, sought “to identify and characterize” new military technologies that would give the president a variety of “alternatives to massive nuclear destruction.” Wohlstetter assembled a study group, called the Strategic Alternatives Group, to assist him in his analytic efforts. In February 1975 the group completed the generically titled “Summary Report of the Long Range Research and Development Planning Program.”

In the report, Wohlstetter concluded that several Vietnam-era DARPA projects merited renewed attention. Topping the list was the effectiveness of laser-guided bombs and missiles. In the last year of the Vietnam War, the U.S. Air Force sent 10,500 laser-guided bombs into North Vietnam. Roughly one-half of these bombs, 5,100 in total, achieved a “direct hit,” with another 4,000 achieving “a circular error probable (CEP) of 25 feet.” Compared to the success rate of unguided “dumb” bombs of previous wars, including World War II, Korea, and most of Vietnam, these statistics were
to be interpreted as “spectacularly good,” wrote Wohlstetter. The best example was the bombing of the Thanh Hoa Bridge, a 540-foot steel span across the Song Ma River, roughly seventy miles south of Hanoi. The bridge was an important supply route for the North Vietnamese during the war, and they kept it defended with garrison-like strength. The bridge was surrounded by a ring of three hundred antiaircraft systems and eighty-five surface-to-air missile systems. A wing of Soviet-supplied MiG fighter jets was stationed nearby. For years in the 1ate 1960s, the Air Force and the Navy tried to destroy the bridge but could not. By 1968, eleven U.S. aircraft had been shot down trying to bomb the bridge. Then, in May 1972, after a four-and-a-half-year bombing halt, fourteen F-4 fighter bombers equipped with newly developed laser-guided bombs were sent on a mission to bomb the bridge. With several direct hits, the bridge was destroyed. “It appears that non-nuclear weapons with near-zero miss may be feasibly and militarily effective,” Wohlstetter wrote in praise of these new “smart” weapons.

Also of interest to Wohlstetter were DARPA’s early efforts with mini-drones, which had played a major role in advancing laser-guided weapons technology—a fact largely underreported in military history books. DARPA’s Vietnam drone program had grown out of DDR&E John Foster’s love of model airplanes and remote control. Two of the mini-drones, called Praerie and Calere, caught Wohlstetter’s eye. Praerie and Calere were exceptionally small at the time, each weighing seventy-five pounds, including a twenty-eight-pound payload that could be a camera, a small bomb, or an “electronic warfare payload.” Each was powered by a lawnmower engine and could fly for up to two hours. Praerie carried a TV camera and used laser target technology. It was the first drone to direct a cannon-launched guided projectile to a direct hit on a tank, a milestone achieved at Fort Huachuca, Arizona, during an undated field test. The Calere drone was equally groundbreaking. It carried forward-looking infrared, or FLIR, another Vietnam-
era invention, which allowed the drone to “see” at low altitudes in the dark of night.

DARPA also developed another, much larger, “more complicated” drone that interested Wohlstetter, as revealed in an obscure 1974 internal DARPA review. Nite Panther and Nite Gazelle were helicopter drones, “equipped with a real time day-night battlefield reconnaissance capability including armor plate and self-sealing, extended-range fuel tanks.” The drone helicopters were deployed into the battlefield, starting in March 1968, in response to an urgent operational request from the Marine Corps. To create the Nite Panther drone, DARPA modified a Navy QH-50 DASH antisubmarine helicopter—originally designed to fire torpedoes at submerged submarines—and added a remotely controlled television system, called a “reconnaissance-observation system,” which could transmit real-time visuals back to a moving jeep, acting as a ground station. The jeep was loaded with racks of telemetry and television equipment, antennae, and a power supply. The drone operator sitting in the jeep was able to operate and monitor the drone helicopter from takeoff to touchdown. Images captured by the drone, flying over enemy territory, were recorded by the equipment on the jeep, then relayed back to a shipboard control station, where commanders could send high-performance strike aircraft to bomb targets identified by the drone. This was groundbreaking technology during the war. In 1974 Wohlstetter recognized its future potential. Conceivably, as computers got smaller and were able to process data faster, a drone could be sent deep behind enemy lines to photograph targets and send the images to commanders in real time.

Another significant DARPA technology that allowed these Vietnam-era systems to converge was a satellite-based navigation technology called Global Positioning Systems, or GPS. GPS began as a classified military program, the purpose of which was to direct weapons to precise targets. DARPA’s pioneering GPS program was called TRANSIT. It began in 1959, when ARPA contracted with
the Johns Hopkins Applied Physics Laboratory to create the first satellite positioning system, using six satellites, three for positioning and three as spares.

After several failed launches, TRANSIT finally took up residence in space in June 1963. To deny enemy access to this kind of precise targeting information, the system was originally designed with an offset feature built in, called selective availability (SA). If an individual were able to access the GPS system with a private receiver, the information would be offset by several hundred feet.

Over the next ten years, the Navy and the Air Force developed their own satellite-based navigational systems, but each system was incompatible with the other. In 1973 the Pentagon ordered DARPA to create a single system shared by all the military services, and a new DARPA program called NAVSTAR Global Positioning System emerged. It was a herculean effort filled with technical stumbling blocks and failed rocket launches. Finally, starting in 1989 a constellation of twenty-four satellites, each fitted with atomic clocks to keep them in sync, was sent aloft and began orbiting the earth. The U.S. military now had precise navigational coverage of the entire world, in all weather conditions, in real time.

During the 1990s, interest in satellite-based global positioning technology grew, and European companies began developing GPS-like systems for civilian use. In an effort to keep the United States at the forefront of the burgeoning new industry, in May 2000 President Clinton discontinued the selective availability feature on GPS, giving billions of people access to precise GPS technology, developed by DARPA.

To Albert Wohlstetter, working on the DARPA analysis in the mid-1970s, the fusion of various Vietnam-era technology systems—sensors, computers, laser-guided weapons, the ARPANET, drones—offered great promise and potential in the development of what he called a “system of systems.” The following year, on the basis of
suggestions made in the “Summary Report of the Long Range Research and Development Planning Program,” DARPA initiated a new weapons program called Assault Breaker. A series of once disparate technologies could come together to fulfill Lukasik’s vision to “command, control, and communicate.” Using technologies that also included radar tracking and camera confirmation, Assault Breaker would one day allow commanders to precisely strike targets—even moving targets—deep behind enemy lines. Imagining a system in which this kind of weaponry and technology could work together was unprecedented. All of it had emerged from the Vietnam War.

In the 1970s, the Soviets were notorious for maintaining a network of spies in and around Washington, D.C., and it did not take long for the Russians to learn about DARPA’s classified Assault Breaker plans. When they did, the Soviet military brass began studying the concept and planning countermeasures. In 1978 an article about Assault Breaker appeared in the classified Soviet military journal
Military Thought.
That the Soviets knew about DARPA’s “system of systems” might have gone unnoticed had it not been for the sharp eyes of Andrew W. Marshall, a former RAND analyst and Wohlstetter protégé who now had his own office inside the Pentagon. Marshall served as director of the Office of Net Assessment, created by the Nixon White House in 1973 and dedicated to forecasting future wars. At RAND, Marshall had secured his reputation as a master game theorist, and at the Pentagon, his wizardry in prognosis and prediction earned him the nom de guerre Yoda, or the Jedi Master. It also put him in regular contact with DARPA directors and program managers, as he continued to be for over forty years.

Part of Andrew Marshall’s job in the 1970s was to monitor what Soviet generals were writing in their classified journals. In reading
Military Thought,
Marshall learned that the Soviets felt so threatened by the prospects of an Assault Breaker–like system of
systems that they were running exercises to practice countermeasures against one. Soviet fears of DARPA’s Assault Breaker concept did not stop there but made their way to the top of the Soviet military chain of command. In 1984 Marshall Nikolai Ogarkov, chief of the general staff of the armed forces of the Soviet Union, worried in a classified memo that Assault Breaker gave the Americans the ability to conduct “automated reconnaissance-and-strike complexes,” a capability that must be regarded as a “military-technical revolution.” Marshall renamed the Russian pronouncement a “revolution in military affairs,” which had since become a celebrated Pentagon maxim. The saying defines what happens when one country or fighting force creates a technology or tactic that makes everything else subordinate to it and makes many of the other side’s earlier weapons systems obsolete.

Just a decade before, in the wake of the Vietnam War and with his agency’s budget slashed, Stephen Lukasik had appealed to Congress to allow DARPA to pursue “high-risk projects of revolutionary impact.” Lukasik told Congress that in the modern world, the country with the most powerful weapons would not necessarily have the leading edge. He argued that as the twenty-first century approached, the leading edge would belong to the country with the best information—with which it could quickly plan, coordinate, and attack. Eleven years later, his vision proved correct. The Soviets felt deeply threatened by DARPA’s C3-based revolution in military affairs.