"The steps between my original suggestion of the chess playing machine, Mr. Shannon's move to realize it in the metal, the use of computing machines to plan the necessities of war, and the colossal state machine of Pere Dubarle, are in short clear and terrifying... The mechanical control of man cannot succeed unless we know man's built-in purposes, and why we want to control him." Norbert Wiener, The Human Use of Human Beings, 1950.
I THINK THEREFORE I AM
Norbert Wiener's popular legacy is the word "cybernetics", which he crafted in his 1948 book "Cybernetics," subtitled "Control and Communication in the Animal and Machine." Cybernetics, according to Wiener, described a new way of looking at life. Where once scientists imagined the universe as a giant clockwork where everything was set according to a pattern, Wiener postulated that the universe was a massively disorganized unpatterned place. Whatever order existed, Wiener thought, came from the exchange of information -- messages, coding, decoding -- between everything from the smallest atomic particle to galactic clusters. Information created order in a disorderly universe. The human brain was a message processor at its core.
Wiener, by analogy, postulated that the human brain and the mechanical brain of the newly invented digital computer were similar. For Wiener, the modern computer was the closest thing to a mechanical brain ever invented. When Wiener invoked this theory, in 1947, the Cold War had not yet fully unveiled itself. The Berlin Blockade of 1948, the Soviet detonation of a nuclear weapon in 1949 -- these had yet to become history -- and little mention in "Cybernetics" appears concerning the potential dangers and temptations Wiener's cybernetics introduced.
By 1950, when Wiener published a second book called "The Human Use of Human Beings", his work took an explicitly political and social tone. He wrote the "Human of Use Human Beings" for non-mathematicians; unlike "Cybernetics" there were no mathematical equations covering the pages. Instead, Wiener emphasized a fear. If information is the currency of life, controlling the shape of things, then wasn't it possible, in theory, to send out messages which would effectively control the way people perceive the world? Critics of Wiener's "Cybernetics" raised this issue, reading his work as the means to the creation of a theoretical machine, a "machine to govern." Wiener felt such an idea wasn't ludicrous, writing in the first edition of "The Human Use of Human Beings" that such a machine "is quite possibly being planned by a secret military project for the purposes of combat and domination." Then, mysteriously, that edition of "The Human Use of Human Beings" disappeared.
All later editions of the book, after the first edition from which these quotes come, use a vastly different text, so different that no two pages are alike. The later editions are much less concerned with secret projects to control human behavior. Instead, Wiener focuses on the theory of cybernetics in relation to Newtonian physics and game theory. The passage concerning the "machine to govern" loses its tone of imminent doom. "The [machine to govern] is not frightening because of any danger it may achieve of autonomous control over humanity. It is far too crude and imperfect to exhibit a one-thousandth part of the purposive independent behavior of the human being." (From the ninth printing, 1967). Why Wiener so totally changed the thrust of his book is hard to say. Wiener died in 1964.
THE INTERGALACTIC NETWORK
Wiener's work and concerns, however, had a deep influence on several crucial computer scientists. Every Tuesday night throughout the 1950's Wiener held a kind of seminar-salon in his home, near MIT in Cambridge, Massachusets. In attendance at various points were a who's-who of computer science history, including junior MIT faculty who would go on and have a tremendous impact on the future of computer science. One of these was a young faculty member named J.C.R. Licklider. He studied the field of "psycho-acoustics" -- how sound travels -- funded in part by the US Air Force. Around 1957 Licklider used his first digital computer and got hooked, displaying the now-familiar symptoms of late nights hunched over a monitor in a dark room. He became so obsessed with computers that he effectively switched fields. In the then-nascent field of computer science he wrote a seminal paper which has Wiener's concerns all over it, but where Wiener sees no hope to the monolithic machine, other than the delay of time, Licklider finds an alternative -- "interactive computing".
Titled "Man-Computer Symbiosis" (1960), Licklider postulates that the real use of the computer is not as a cold calculating engine, but as an intimate symbiotic partner in human activity. Licklider explicitly snubs the young A.I. community, dismissing the obsession with creating machines to replace people. His alternative is now dominant, and he shifted computer science towards the study of people interacting with computers as a partner. Licklider could have faded into obscurity were it not for the fact that in October 1962 he became Director of the Pentagon's Information Processing Techniques Office (IPTO). A division of ARPA (the Advanced Research Projects Agency), Licklider initiated the sequence of events which led to the ARPANET, Internet's parent. Had he not done so, there's a fair chance the Internet we're using today wouldn't exist.
Back then, when this computer network was still just an idea, Licklider called it his "Intergalactic Network", and sent memos to the universities receiving IPTO money titled "To the members of the Intergalactic Network." But before he could open the financial sluice gates, he had to kill another project. The Air Force, working hard on Wiener's theoretical "machine to govern" had spent much of the fifties trying to use computers to simulate human behavior. In practice this meant feeding a computer seemingly random information -- on Thursdays Kruschev read Pravda, not Isvestia, and the night before the Soviet Air Force General drank a whiskey, not vodka -- and somehow all these observations would produce an accurate scenario of what the Soviets were really up to. The computer would play Sherlock Holmes and conclude that the Soviets were building a new missile, or whatever. Licklider yanked away his budget from these "asinine projects" as he called them. It was Licklider who changed the name of the department from Command and Control Research to Information Processing Techniques.
In the years that followed, an estimated 70% of all funding for computer science research in the United States came from ARPA, and much of it followed the path Licklider set in 1962. Licklider funded research leading to the creation of the first computer mouse, "windows" and "icons." His control of the purse led to the funding of America's first graduate programs in computer science, along the philosophical lines he favored. And, on top of all this, Licklider explicitly funded California schools, wanting to transplant his ethic to the West Coast. Two prime recipients were Berkely and Stanford, and their students, steeped in visions of interactive computers, contributed to the creation of Silicon Valley, and the ultimate manifestation of Licklider's intimate machine: the personal computer. But the crown jewel of Licklider's crusade was his initiation of the events leading to the Internet. Had Licklider not altered course, odds are that there would be no Internet and the seemingly silly musings of Wiener would, in hindsight, appear prescient.
MEME and its contents copyright by David S. Bennahum. Duplication for non-commerical use is permitted. Contact me if you have questions. Direct comments, bugs and so on to me at davidsol@panix.com.
