Robotics Reference
In-Depth Information
In 1869 Jevons produced a more advanced logic machine, with a
keyboard added, a machine that was able to properly analyse syllogisms.
This machine used an ingenious system of rods, wires, levers, springs and
latches to enable the wooden rectangles to be moved automatically, elim-
inating the cumbersome and complicated method used for the abacus,
and the machine was limited to expressions with four or fewer terms and
their negatives: A, not A, B, not B, etc.
Jevons' construction was the first machine that could solve logical
problems at an acceptable speed. The machine was about three feet tall
and, because of its appearance, it was described as a “logic piano”.
The other major figure in the development of nineteenth-century ma-
chines designed specifically to solve problems in logic, was Allen Mar-
quand (1853-1924), the son of a New York banker. He studied at Johns
Hopkins University with Charles Peirce, who has been described by Max
Fisch as “the most original and the most versatile intellect that the Amer-
icas have so far produced.” [4] Peirce well understood Babbage's plan for
the Analytical Engine and was also interested in logic machines, so it was
probably from Peirce that Marquand acquired his own interest in logic
machines, which was further developed after Marquand was appointed
tutor of logic at the College of New Jersey, later renamed Princeton
University.
Marquand first constructed a crude but technically improved version
of Jevons' machine in 1881 and, a few months later, with a Princeton
colleague, Charles Rockwood, he created a second more advanced design,
using a mechanical action with rods and levers connected by pins and
catgut strings, this design requiring the use of only ten keys rather than
the 20 employed in Jevons' machine.
Marquand's machines were limited in the complexity of argument
they could handle and both could only produce logical combinations
consistent with
the concluding proposition, rather than producing the
proposition itself. He proposed a third version of his machine that would
have changed the action from mechanical to electro-mechanical (i.e., us-
ing components that were partly electrical and partly mechanical), but
difficulties with the new electrical technology prevented him from ad-
vancing beyond building a prototype out of a hotel annunciator, a device
that indicated which of the hotel guests has rung a bell from their room.
Although his invention worked well, Marquand abandoned his mechan-
ical logic machine project when he was offered a position as a professor
of art and archaeology at Princeton.
