HOUSE_OVERSIGHT_018310.jpg

there might be no response. The US, with its priceless collection of bombers and missiles and million-man army, could not strike back for the simple reason that the nation’s field officers would have no way to talk to each other, or to commanders in Washington. The military radio and telephone systems America depended on for her safety would not likely endure an initial Soviet strike. This was the problem Baran had been told to solve.“At the time we didn’t know how to build a communication system that could survive even collateral damage by enemy weapons,” he recalled later. RAND determined through computer simulations that the AT&T Long Lines telephone system, a copper web that carried essentially all the nation’s military communications, would be cut apart by relatively minor physical damage.!°> The military had spent, already, a fortune on the problem. (They had spent half a fortune, it turned out, trying to hide it.) The result was an expensively designed, gorgeously featured telephone network linking military bases to strategic command posts. But because the lines and their switching centers were laced out in a pattern with just a few big central nodes, like a bicycle wheel with spokes, it had almost no chance of surviving the very thing it was designed to help prevent, a Soviet strike. If you gazed at an inked-out map of this network, with its central hub staffed by senior commanders and then radiating lines out to bases and missile silos, it even looked, well, like a target. If the USSR could bullseye those hubs with a bomb or two, the rest of the network would fold. The Soviets could do whatever they wanted: Invade Berlin, roll into France, obliterate Los Angeles. America’s military would be deaf. And as Soviet missiles became more accurate, this seemed an inevitability. “We will soon be living in an era,” Baran wrote, “in which we cannot guarantee survivability of any single point.” The situation, as a carefully screened handful of scientists at RAND knew, was in fact even more perilous. Shortly before Baran arrived at RAND, scientists testing hydrogen bomb designs in the Pacific discovered that radiation from their explosions fuzzed communications for hundreds of miles. A Soviet attack, even if it 105 “At the time”: There is a fair amount of debate about this question of if the Internet design was intended for survivability or if some other systemic need - such as linking research institutions - accounted for the distributed architecture that emerged. See, for instance, Barry M. Leiner, Vinton G. Cerf, et al, “A Brief History of the Internet”, ACM SIGCOMM Computer Communication Review Volume 39 Issue 5, October 2009, 22-31. However, an examination of primary source documents shows the evolution of Baran’s thinking clearly and produces documentary evidence for the origins of the problem he and various figures at RAND were aiming to solve. Others arrived at the packet switching model, but it is clear Baran’s path to the design emerged from the security problems he was considering. For much of the information here see “Oral History: Paul Baran” Interview #378 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc. (Available online); Paul Baran, “On Distributed Communications I: Introduction to Distributed Communications Networks,” United States Air Force Project RAND (August, 1964); Baran “On Distributed Communications XI: Summary Overview,” United States Air Force Project RAND (August, 1964) 78 HOUSE_OVERSIGHT_018310