262 | 24 TURING, lOVElACE, AND BABBAGE
7000 manuscript sheets had been published, and any significant knowledge would necessarily
have involved physical examination in situ. The detail of this archival material was not stud-
ied in earnest until the late 1960s, and scarcely anything was published on the specifics of the
designs until the 1970s.^38 These 19th-century sources were accessible in libraries and archives.
In Turing’s own time there was a paper on Babbage’s Difference Engine by Dudley Buxton,^39
grandson of a junior colleague of Babbage into whose hands Babbage had entrusted a major
cache of manuscripts.^40 The paper, presented at the Science Museum in December 1933 and
published by the Newcomen Society, describes the Difference Engine in fair detail, touches
on the transition to the Analytical Engine and outlines techniques for direct multiplication
and division. Several practitioners allied to Turing’s subsequent interests participated in the
discussion following Buxton’s paper. Included were D. Baxandall, Keeper of the Mathematics
Section at the Science Museum, L. J. Comrie, Superintendent of the Nautical Almanac Office
who implemented on commercially available calculators Babbage’s method of tabulation by
differences, and A. J. Thompson from the General Register Office.^41 Buxton’s paper signalled
that Babbage still had a visible profile in the period of Turing’s famous 1936 paper. Turing does
not mention Babbage’s archival drawings or the other published sources.
In August 1951 Turing, with four Cambridge friends, visited the Science Museum in South
Kensington where the science exhibits for the Festival of Britain were displayed.^42 On public
view would have been Babbage’s small experimental model of a part of the Analytical Engine
under construction at the time of his death in 1871 and the ‘beautiful fragment’ of Difference
Engine No. 1, but it is not known whether he saw them.^43 In any event, neither of these physical
relics illuminated the design principles or their mechanical operation.
In the context of his reference to ‘Lady Lovelace’s objection’ Turing comments that Lovelace’s
‘memoir’ was the source of ‘our most detailed information of Babbage’s Analytical Engine’. This
was the earliest known reference he made.^44 In quoting Lovelace’s ‘objection’ Turing says that
Lovelace’s statement was quoted by Douglas Hartree. This opens up the possibility that Turing’s
source was not Lovelace’s Sketch itself, in all its extensive detail, but Hartree’s brief overview.^45
It is reasonable to conclude that Turing did not consult 19th-century sources by H. P. Babbage,
by Lardner, or the technical plans by Babbage himself, and that his knowledge of Lovelace
and the Analytical Engine was most likely no more than Hartree’s summary account. If this
is so then, given that Hartree’s book was not published until 1949, Babbage’s and Lovelace’s
ideas could not have influenced Turing’s 1936 paper. Robin Gandy supports this view. He com-
mented that Menabrea’s account and Lovelace’s notes appear not to have been known even to
some of those familiar with Babbage:^46
Thus there is no reason to suppose that the mathematicians involved in the discoveries of 1936
(including Hilbert, Bernays, Gödel, and Herbrand) were familiar with Babbage’s theoretical ideas:
certainly none of them refer to his work.
In overall terms there does not appear to be a direct overt line of influence between Babbage–
Lovelace in the 19th century and the pioneers of electronic digital computing in the 20th cen-
tury. Most leading figures knew of Babbage and the legend of his mission but none appear to
have had any detailed knowledge of his designs.
In the light of a strong congruence of ideas between two pioneering eras separated in time we
are left with the suggestion, both reassuring and confining, that core ideas, articulated a century
apart, embody something fundamental about the nature of computing.