Burdens of Proof: Cryptographic Culture and Evidence Law in the Age of Electronic Documents

Cryptography and National Security

On the Brink of a Revolution 47 performance anxiety based on key size would become a recurrent theme in cryptographic culture. Cryptography and National Security The successive publication of David Kahn’s The Codebreakers in 1967, Feistel ’s overview of Lucifer in Scientific American in 1973, Diffie and Hellman’s “New Directions” in 1976, the DES standard in 1977, and the RSA paper in 1978 fostered the rapid development of an academic cryptographic research community. It gathered in August 1981 at the University of California , Santa Barbara, for the first academic conference devoted to cryptographic research, CRYPTO ’81. The following year, a group of attendees, including David Chaum, Whitfield Diffie, and David Kahn, met to establish the International Association for Cryptologic Research (IACR).19 The conferences were the theater of some dramatic exchanges between this burgeoning community and the NSA, which viewed with increasing anxiety public dissemination and discussion of technologies with direct relevance to national security.20 The NSA warned the community that public presentations of cryptographic research might be in possible violation of the 1954 Munitions Control Act. It followed up with attempts to control researchers’ NSF funding.21 In 1982, then director of the NSA Admiral Bobby Imann proposed a truce to the community in the form of a mandatory review agreement, arguing that If the scientists did not agree to the voluntary review of their work by the intelligence agencies, they would face a “tidal wave” of public outrage that will lead to laws restricting the publication of scientific work that the government might consider “sensitive” on national security grounds. . . . Imann warned a meeting of the AAAS that “the tides are moving, and moving fast, toward legislated solutions that in fact are likely to be much more restrictive, not less restrictive, than the voluntary censorship system.”22 Such warnings stood in stark contrast to the cryptographic community ’s embrace of the scientific ethos of open communication: David Kahn’s report on CRYPTO ’82 features a picture of a broadly smiling Adi Shamir shaking hands with Len Adleman (respectively, the S and A of RSA), congratulating him over the break of the Graham-Shamir knapsack public-key algorithm.23 The handshake powerfully symbolized the new ethics of codemaking and codebreaking, activities now openly performed 48 Chapter 3 in public forums by scientists allied in the common pursuit of objective knowledge. In 1986, further restrictions on public dissemination were attempted through the use of the patent secrecy system, which enables the Patent and Trademark Office to prevent the disclosure or publication of any subject matter deemed detrimental to national security.24 By the 1990s, however, cryptography enjoyed amounts of public attention rarely experienced by an esoteric mathematical field, emerging as the “first political movement in the digital era.”25 Cryptographers appeared on the cover of the New York Times Magazine and Wired, and articles on the use of cryptography for privacy protection appeared in Scientific American and Communications of the ACM. It was thus with a renewed sense of purpose that the community embarked in the mid-1990s in another round of confrontation with the intelligence establishment. This time, the battle concerned the social and technical adequacy of a system known as the Clipper chip and a set of rules restricting the export of cryptographic products outside of the United States. The issues, collectively known as “the crypto debate,” spawned a small policy industry, sharply divided along pro–law enforcement and cyberlibertarian lines. Because widespread access to encryption technologies would result in diminish wiretapping capabilities for law enforcement, the Clipper proposal sought to ensure that every communication device in the United States be equipped with capabilities for key escrow, to be accessed by authorities as required by circumstances. Export controls sought to diminish the dissemination of technologies felt to be an essential element of national security, by limiting the export of so-called strong cryptography, that is, products with key lengths thought sufficient to defeat the capabilities of intelligence agencies. The cryptographic research community largely espoused the cyberlibertarian position. It argued that access to cryptography was essential to a democratic society increasingly structured around computer networks and provided the only effective tool to escape the looming onslaught of Big Brotherism.26 Some saw even greater potential for public-key encryption. Timothy May, an Intel engineer and founder of the “Cypherpunk” mailing list, prophesied that free cryptography would act as “the wire clippers which dismantle the barbed wire around intellectual...