Beelzebub’s ProtegE – Deconstructing the Myth of Whitfield Diffie
YORKTOWN HEIGHTS, NY – Bailey Whitfield “Whit” Diffie first heard about the applications of computational work-based cryptographic methods from Alan Tritter at IBM’s Thomas J. Watson laboratory, when he met with him and other members of the company’s cryptography research team over a weekend in 1974. Diffie was nearing the end of his sojourn across the United States, which semi-official crypto biographer Steven Levy paints as the romantic adventure of an eccentric genius, whose “libertarian ethic” drove him to scour the country in search of the best way to achieve cryptographic privacy.
Darting from coast to coast in a Datsun 510, the future “father of cryptocurrency” was on a mission to find answers to the puzzle of secure cryptographic communications in a civilian environment and electronic commerce, in particular. Diffie had come to Yorktown Heights looking for Tritter, specifically, since the latter’s work at Lincoln Laboratory with the SAGE air defense system’s Identification Friend or Foe (IFF) problem held important clues about how to overcome similar issues presented in non-military crypto settings.
IFF’s prototypical proof-of-work approach could be further explained to Whit by Tritter’s celebrated IBM colleague and fellow Lincoln Lab employee, Horst Feistel, who had recently patented the first block cipher called Lucifer. Unfortunately for Diffie, Feistel happened to be away on vacation at the time, and had to settle for a later meeting. In the meantime, lab director Alan Konheim suggested he look up a former Watson employee Martin Hellman when he returned to Stanford, where Diffie held a guest research appointment at John McCarthy’s Stanford Artificial Intelligence Laboratory (SAIL), assisting in the design of ARPANET’s cryptographic security features.
Diffie’s status as ‘éminence grise’ of the cypherpunk movement revolves around the paper that he co-wrote with Hellman in 1976, titled New Directions in Cryptography, which set forth the principles of public-key cryptography that lay at the heart of most Internet communication protocols. The popular myth promoted by Levy and others, who use words like “outlaws”, “rebels” and “hackers” to describe the generation of DoD-funded academics and Silicon Valley tech professionals responsible for building the cybernetic enclosure, promulgates disingenuous notions of their anti-establishment motivations for what are nothing more than naked capitalist pursuits, as the paper’s first paragraph makes abundantly clear:
“We stand today on the brink of a revolution in cryptography. The development of cheap digital hardware has freed it from the design limitations of mechanical computing and brought the cost of high grade cryptographic devices down to where they can be used in such commercial applications as remote cash dispensers and computer terminals.”
According to accepted lore, Whit had taken a leave of absence from SAIL after a troublesome interlude, in which he had begun to question his own abilities and career path. But rather than seek new horizons, Diffie made deeper inroads into the world of cryptography that had been re-introduced to him by his mentor at the MITRE corporation, Roland Silver.
Yet another product of Lincoln Laboratory, Silver was a close colleague of John McCarthy and had worked with himduring the early stages of the time-sharing project at Bolt, Beranek & Newman (BBN). Along with Feistel, Silver had been among the hundreds of Lincoln Lab staff members who made the transition into the MIT spinoff, and once the time-sharing project came under the auspices of the Defense Department as ARPANET, BBN would be awarded the main development contract, with MITRE and SAIL together forming the core triumvirate of the proto-Internet project.
Diffie was brought into these circles immediately upon his graduation from MIT with a B.S. in mathematics. Hired by the MITRE corporation in 1966, he would work under Silver out of Marvin Minsky’s AI lab in Cambridge, Massachusetts, on an early machine learning project for the defense contractor meant to streamline the design of electronic circuits. Simultaneously, Silver schooled Diffie on modern cryptographic methods, rekindling a waning interest he had harbored since childhood.
In 1969, McCarthy invited the young mathematician to join him on the West Coast and contribute to the efforts underway at SAIL. Located near the rolling hills around Felt Lake in Palo Alto, California, the sister institution of MIT’s AI lab – also co-founded by McCarthy –, was housed inside the D.C. Power Laboratory building. Over 150 people, from graduate students to research associates and faculty members, converged in this off-campus site of Stanford University’s Computer Sciences department to build the cybernetic universe that was to be inhabited by the scores of disaffected post-war youth waiting in the wings.
1600 Arastradero Rd.
On the same day John F. Kennedy was inaugurated as President, DoD radar defense specialist Jack Ruina became director of the Advanced Research Projects Agency (ARPA), a relatively new agency created by the landmark Defense Reorganization Act of 1958, that would spearhead the construction of the cybernetic enclosure through the coordination of private, public and academic interests.
Ruina recruited J.C.R. Licklider, then still supervising John McCarthy’s time-sharing PDP-1 experiments at BBN, to lead ARPA’s Information Processing Techniques Office (ITPO), which would spawn ARPANET several years later. Four months after Kennedy uttered his famous call for Americans to “ask not what” their country could do for them and scarcely a few weeks before he promised to put a man on the moon, McCarthy delivered his own seminal address on the occasion of MIT’s centennial celebrations, proposing the idea of a “computer utility” – a distributed network of time-sharing computers serving a function similar to electric utilities.
The concept found immediate support among top academics, leaders of industry and the banking sector, all of whom stood to reap enormous windfalls from the transformation of virtually all communications systems from analog to digital technology. McCarthy moved out West, where the bulk of the novel technology was being manufactured at Stanford Industrial Park by the likes of Hewlett Packard and others, to establish SAIL with funding from ARPA.
Preceding the existence of the university’s own computer science department, SAIL had only 15 people contributing to the project when it was finally created in 1965. At that time, SAIL’s scope of research was also expanded, and additional funds were procured from ARPA to upgrade the computer system. Moving soon after to the larger facilities at the foothills of the Santa Cruz Mountains, the project grew quickly as more financial support flowed in from the National Science Foundation (NSF), the National Aeronautics and Space Administration (NASA) and other federal agencies.
Faculty from other departments attached themselves to SAIL in the process, and continued to grow the project’s reach. The Department of Health, Education, and Welfare, now Health and Human Services (HHS), stepped in to underwrite the Heuristic Programming sub-project, which was initiated by Joshua Lederberg, chair of the school’s department of genetics, and Edward Feigenbaum, who was the first head of the computer science department and founder of the Knowledge Systems Laboratory (1968) – another artificial intelligence lab at Stanford.
Lederberg’s participation highlights the nature of SAIL as a linchpin of Cold War military research into state-of-the-art computer technology. While long-recognized within the academic milieu, Lederberg was also a known fear monger who availed himself of fabricated accounts from Soviet defectors about Russia’s bioweapons program to push for increased funding for America’s own biological weapons research. ARPA itself, was the result of the exaggerated claims about Soviet technological superiority and the fanciful “missile gap”, which led the U.S. Congress to appropriate tens of billions of dollars for the erection of the Military Industrial Complex.
By the time Whit Diffie arrived to perform relatively pedestrian tasks, like checking proofs and software logic structures, SAIL had ballooned into a huge undertaking with dozens of sub-projects, including robotics, speech recognition, hardware design, music, hand-eye research and many others. Perhaps the only thing not being researched directly at SAIL was digital money, though many of its alumni would eventually work their way into the thick of it.
Snow White’s Hippies
It was too early to tackle such a daunting enterprise, but the concept was hardly new. McCarthy’s MIT colleague, Martin Greenberger had laid out the basics in a 1964 Op Ed in The Atlantic, which envisioned “universal credit cards” that would one day “all but eliminate the need for currency, checks, cash registers, sales slips, and making change”. Greenberger had taken his cue from McCarthy’s “computer utility” speech years before and McCarthy returned the favor, when he elaborated on Greenberger’s ideas in 1971, calling for “new forms of digital commerce” that would be made possible by “home information terminals” during his Turing Award lecture.
Greenberger’s “money keys” included “computer-managed markets” and “personalized insurance”, which were already beginning to knock on Wall Street’s door in the mid-60s as the first electronic trading systems were put in place before the end of the decade. Even before Diffie published his academic tract in 1976, the CEO of Visa, Dee Hock, saw the panorama clearly, when he came up with his “Electronic Value Exchange” (EVE), a digital payment system he envisioned as “arranged energy […] flowing seamlessly through computer networks around the earth”.
The writing had been on the wall for nearly two decades by the time New Directions was penned. Digital currency was not a matter of if, but when, and also how. Deploying it required the construction of the infrastructure for a data economy, beginning with the creation of both ARPA and NASA through the Defense Reorganization Act and the execution of its first steps during the Kennedy administration.
IBM, “Snow White”, and the “seven dwarfs”, comprised of Burroughs, Control Data, General Electric, Honeywell, NCR, RCA, and Sperry Rand, were the principal private sector partners in the paradigm shifting exercise, with AT&T – the telecommunications monopoly –, transitioning into digital transmission technologies, starting in 1962 with the launch of the Telstar satellite, which carried the first microwave telephone and television signal ever from IBM’s facilities in Endicott, NY. A year later, AT&T installed the first commercial electronic switching system.
IBM’s Watson Laboratory’s first director was recruited from the Office of Naval Research (ONR) and signaled the new status quo of revolving doors between tech corporations and the Pentagon. When Diffie came looking for Alan Tritter in the mid-seventies, he knew the computer scientist’s work there was happening under the indirect auspices of the National Security Agency (NSA), not to mention Horst Feistel’s long career with the U.S. Air Force before joining Tritter at Lincoln Labs and later working for MITRE corporation, himself.
Before leaving SAIL for an extended sabbatical carrying a copy of David Kahn’s Codebreakers everywhere he went to telegraph the anti-establishment sentiments he’s lionized for, Diffie was intimately involved in erecting the scaffold of the data economy at Stanford. Margaret “Peggy” Waters, a programmer he was dating before embarking on the journey that would eventually lead him to get hitched to a British Petroleum geologist he knew, was part of the computer design team at SAIL that developed the first graphical interface, known as SUDS.
Short for Stanford University Design System, SUDS was the original CAD system and was run on the SUN (Stanford University Network) workstations at SAIL. Andy Bechtolsheim, a German electrical engineer who designed the SUN network in 1977 while earning his PhD at the Institution, co-founded Sun Microsystems in 1980 by essentially stealingStanford’s system, including SUDS. The original SUN network was based on the Xerox Alto developed at the nearby Xerox Palo Alto Research Center (PARC), spun off from UC Berkeley’s ARPA-backed time-sharing research project.
Whitfield Diffie would join Sun Microsystems as it Chief Security Officer in 1991, just as the so-called cypherpunk mailing list was about to start gathering steam. Started by another Sun Microsystems engineer, the cypherpunks’ listserv would be run by a new generation of computer scientists and technophiles emerging out of the same SAIL networks, who would find a worldwide audience descended from the phone phreak community of the late 1960s and early 1970s for the dubious concept that learning how to interface with the world through a computer screen was some kind of revolution and that trading in digital tokens tied to a computer hard drive translated to economic freedom.
