The NSA turned it down in 1955 because the thinking around encryption was based on mathematics, and this method is digital. If you look at the very last page of the Nash letters, there is an administrative note that a ‘Dr. Campaigne’ signed off on the rejection letter.

If you look at this internal NSA article, it discusses the thinking around unbreakable ciphers, and interestingly it is written by a ‘H.Campaigne’ - I’m not sure if it’s the same person, but nevertheless it displays the official thinking around crypto internally at NSA at the time (and even now). It is undated but it would be interesting to find out if it’s the same Mr. Campaigne that rejected the Nash crypto and if it was written before or after he saw Nash’s letter.

http://www.nsa.gov/public_info/_files/tech_journals/The_Unbreakable_Cipher.pdfIn this article, Mr. Campaigne talks about unbreakable encryption only being possible if one has enough randomness to make a ‘one-time pad’ - and discusses some primitive methods for gathering such randomness. Another indicator for primitive thinking around cryptography is the text-oriented view of the message to be encrypted. These cryptographers began with the concept of ‘cleartext’ as in alphabetical symbols that needed to be rearranged in some way to produce another string of alphabetical symbols.

The Nash crypto is digital, not text-based, so its lowest unit of information is correctly the bit, not the letter. As you know, ASCII characters are represented by 8 bits, so the character ‘A’ would be 01000001, and the Nash protocol would essentially split the symbol in half and randomize each half, then fuse the two halves back together.

Download the NashAlgorithm hereUnlike mathematical schemes, there is no letter ‘A’ floating through the process, the information is broken down into its informational ‘atoms’, and a sophisticated (yet digitally simple) combination of deterministic randomization and phase shifting takes place, introducing a complexity of another order, while being computationally extremely efficient.

In a physical analogy, a mathematical cipher is similar to taking a photograph and shredding it, rearranging the pieces into a puzzle that it is hoped will be time-consuming for the adversary to reassemble. The Nash protocol works on a much higher principle - the cipher data is like a hologram, and attempting to reconstruct the clear data by looking at the cipher stream is as futile as examining a holographic image with a microscope, all one sees is a speckle pattern - surface features that are in a completely different physical and informational domain as what they encode.