COPElAND | 203
became the ‘person who really . . . pushed the whole field ahead’, Bigelow explained, was that he
understood what was implied by Turing’s universal machine idea:^22
Turing’s machine does not sound much like a modern computer today, but nevertheless it was. It
was the germinal idea.
Although Turing is not mentioned explicitly in von Neumann’s papers developing the
design for the Princeton computer, his collaborator and co-author Art Burks told me that their
key 1946 design paper (co-authored also by Goldstine) did contain a reference to Turing’s
1936 work.^23 There they emphasized that ‘formal-logical’ work—in particular, Turing’s 1936
investigation—had shown ‘in abstracto’ that stored programs can ‘control and cause the execu-
tion’ of any sequence (no matter how complex) of mechanical operations that is ‘conceivable
by the problem planner’.^24 More information about Turing’s influence on von Neumann can be
found in Chapter 6.
So Turing’s theoretical ideas of 1936 were the ultimate inspiration for both the Manchester
and Princeton computer projects. In 1947, thanks to Newman and Good, an early version of
the Princeton design ended up as the logical blueprint for the Manchester Baby, as we shall see.
But the Princeton computer worked only because Williams baled out von Neumann’s group,
tossing them the memory design he had invented for Baby (in return for substantial royalties,
of course). It was a wonderfully tangled loop, with Turing at its centre.
Prequel: from Bletchley Park to manchester
It was, of course, no coincidence that, as soon as the war ended, Turing and Newman both
embarked on projects to create a universal Turing machine in hardware. But before the 1970s,
when the British government declassified captioned photographs of Colossus,^25 few people
had any idea that electronic computers had been operational at Bletchley Park from 1944 (see
Chapter 14). It was not until 2000 that the British government finally declassified a complete
account of Colossus (as explained in Chapter 17).^26 Having no inkling of Colossus, early his-
torians of computing formed the view that the British pioneers inherited their vision of large-
scale electronic computing machinery from the ENIAC group in the United States.^27
ENIAC (Chapter 14) and its successor EDVAC were certainly the fountainhead of inspira-
tion for some British computer pioneers (Douglas Hartree and Maurice Wilkes, for instance).^28
However, it was Colossus, not ENIAC, that formed the link between Turing’s pre-war exposi-
tion of the universal machine and Newman’s post-war project to build an electronic stored-
program computer. Colossus was a pivotal influence on Turing also. Flowers told me that once
Turing had seen Colossus, it was just a matter of his waiting for an opportunity to put the idea of
his universal computing machine into practice.^29 Other mathematicians and engineers work-
ing in Newman’s section at Bletchley Park also made the connection between Flowers’ racks of
electronic equipment and the idea of an all-purpose stored-program computer. Members of
the Newmanry were (so Newmanry codebreaker Donald Michie reported) ‘fully aware of the
prospects for implementing physical embodiments of the UTM [universal Turing machine]
using vacuum-tube technology’.^30
Newman laid plans for his Computing Machine Laboratory as soon as Manchester University
told him that they were going to appoint him as professor. His formidable talent as an organizer,