Computer Builders Meet Cryptanalysts

Mauchly,_John_and_Eckert,_J._PresperComputer Builders Meet Cryptanalysts

John Mauchly and J. Presper Eckert in this February 1946 photo are reading the print-out from the computer they built at the Moore School in Philadelphia.  Come the summer, a quiet Navy officer cryptanalyst will attend their seminars and link computers to the forerunners of NSA and open a door for the construction of cyberspace.

Here’s how it happened.  Right after V-J day, code-breakers at the Navy Communications Annex and their Army counterparts over at Arlington Hall quit in droves.  Since it was August, it was too late for Howard Campaigne to return to teaching math at the University of Minnesota that fall.[i]  He stayed in Washington.  The daily deluge of communications traffic to decrypt was over.  With breathing room, “we shifted to a longer-range view and started looking for improved ways of doing things,” he told NSA historians later.

The cryptanalysts heard of a summer program at the University of Pennsylvania in summer 1946. Eckert and Mauchly would talk about the ENIAC computer.  We should send somebody, Campaigne’s boss told him.  Find someone.  Campaigne picked Lieutenant Commander James T. Pendergrass.

When Pendergrass returned, he and Campaigne quickly drafted a report on why computers should take over cryptanalysis.

Today it seems ironic, but convincing spymasters to adopt computers was no easy task. Decryption had gotten very fast and efficient during the war.  The ENIAC was an awkward collection of machines in large rooms.  Why switch?  Pendergrass and Campaigne constructed proofs for how a fast, digital, computer could become good enough to solve the classic problems of cryptography. Pendergrass attacked the problem by starting with the top-level logic behind how to tackle Enigma, the Grenade and a very different system known as the Hamelin.  Appendices to the report postulated that a computer could process the work so much more quickly that it would deliver results fast enough to keep up with routine changes in enemy ciphers.[ii]

“By 1947 both the Army and Navy cryptologic organizations were committed to

acquiring general-purpose computers,” wrote historian Colin Burke. They had, however, no clear idea which among several competing concepts might work — if, indeed, any of them would.

“We saw an opportunity to apply it to cryptography and we were off after computers,” Campaigne summed up later.[iii]

[i] Oral History with Howard Campaigne, NSA OH-14-83. DOC ID 3067796.

[ii] Colin Burke, An Introduction to a Historic Computer Document: Betting on the Future –

The 1946 Pendergrass Report Cryptanalysis and the Digital Computer, p. 70.

[iii] Campaigne ,Oral History, p. 73.

Thank you for posting the 1946 photo of Eckert and Mauchly.