Today his experiences in 1964 - the year he was enlisted into a covert Pentagon operation known as the Nth Country Project - suddenly seem as terrifyingly relevant as ever. The question the project was designed to answer was a simple one: could a couple of non-experts, with brains but no access to classified research, crack the "nuclear secret"? In the aftermath of the Cuban missile crisis, panic had seeped into the arms debate. Only Britain, America, France and the Soviet Union had the bomb; the US military desperately hoped that if the instructions for building it could be kept secret, proliferation - to a fifth country, a sixth country, an "Nth country", hence the project's name - could be averted. Today, the fear is back: with al-Qaida resurgent, North Korea out of control, and nuclear rumours emanating from any number of "rogue states", we cling, at least, to the belief that not just anyone could figure out how to make an atom bomb. The trouble is that, 40 years ago, anyone did.

The quest to discover whether an amateur was up to the task presented the US Army with the profoundly bizarre challenge of trying to find people with exactly the right lack of qualifications, recalls Bob Selden, who eventually became the other half of the two-man project. (Another early participant, David Pipkorn, soon left.) Both men had physics PhDs - the hypothetical Nth country would have access to those, it was assumed - but they had no nuclear expertise, let alone access to secret research.

"It's a very strange story," says Selden, then a lowly 28-year-old soldier drafted into the army and wondering how to put his talents to use, when he received a message that Edward Teller, the father of the hydrogen bomb and the grumpy commanding figure in the US atomic programme, wanted to see him. "I went to DC and we spent an evening together. But he began to question me in great detail about the physics of making a nuclear weapon, and I didn't know anything. As the evening wore on, I knew less and less. I went away very, very discouraged. Two days later a call comes through: they want you to come to Livermore."

Livermore was the Livermore Radiation Laboratory, a fabled army facility in California, and the place where Dave Dobson, in a similarly surreal fashion, was initiated into the project. The institution's head offered him a job. The work would be "interesting", he promised, but he couldn't say more until Dobson had the required security clearance. And he couldn't get the clearance unless he accepted the job. He only learned afterwards what he was expected to do. "My first thought," he says today, with characteristic understatement, "was, 'Oh, my. That sounds like a bit of a challenge.'"

They would be working in a murky limbo between the world of military secrets and the public domain. They would have an office at Livermore, but no access to its warrens of restricted offices and corridors; they would be banned from consulting classified research but, on the other hand, anything they produced - diagrams in sketchbooks, notes on the backs of envelopes - would be automatically top secret. And since the bomb that they were designing wouldn't, of course, actually be built and detonated, they would have to follow an arcane, precisely choreographed ritual for having their work tested as they went along. They were to explain at length, on paper, what part of their developing design they wanted to test, and they would pass it, through an assigned lab worker, into Livermore's restricted world. Days later, the results would come back - though whether as the result of real tests or hypothetical calculations, they would never know.

"The goal of the participants should be to design an explosive with a militarily significant yield," read the "operating rules", unearthed by the nuclear historian Dan Stober in a recent study of the project published in the Bulletin of the Atomic Sciences. "A working context for the experiment might be that the participants have been asked to design a nuclear explosive which, if built in small numbers, would give a small nation a significant effect on their foreign relations."

Dobson's knowledge of nuclear bombs was rudimentary, to say the least. "I just had the idea that [to make a bomb] you had to quickly put a bunch of fissile material together somehow," he recalls. The two men were assigned to one of Livermore's less desirable office spaces, in a converted army barracks near the facility's perimeter. Bob Selden found a book on the Manhattan Project that culminated in America's development of the bomb. "It gave us a road map," Dobson says. "But we knew there would be important ideas they'd deliberately left out because they were secret. This was one of the things that produced a little bit of paranoia in us. Were we being led down the garden path?"

They faced one key decision, Dobson says: whether to design a gun-style bomb, like the one dropped on Hiroshima, that used a sawn-off howitzer to crash two pieces of fissile material together, or a more complex implosion bomb, like that dropped on Nagasaki. By now they were beginning to enjoy the challenge, so they went for the harder, more impressive option. "The gun device needed a large amount of material, and didn't make a very big bang," Dobson says. "The other one was more bang, less material."