Notes for the Lifebox, the Seashell, and the Soul - Rudy Rucker

Notes for The Lifebox, the Seashell, and the Soul, by Rudy Rucker
Organism Many Many Network
Human Society Many Many Network
The Mind Many One Global
Table 1.4: Some computer architectures
Von Neumann Self-Reproduction
In the late 1940s, von Neumann gave some ground-breaking lectures on the topic of
whether or not it would ever be possible for a machine, or “automaton,” to reproduce itself.
Usually a machine makes something much simpler than itself — consider a huge
milling machine turning out bolts. Could a machine possibly fabricate machines as
complicated as itself? Or is there some extra-mechanical magic to self-reproduction? To
simplify the problem, von Neumann suggested that we suppose that our robots or automata
are made up of a small number of standardized parts:
I will introduce as elementary units neurons, a “muscle,”
entities which make and cut fixed contacts, and entities which supply
energy, all defined with about that degree of superficiality with which
[the theory of neural networks] describes an actual neuron. If you
describe muscles, connective tissues, “disconnecting tissues,” and
means of providing metabolic energy . . . you probably wind up with
something like ten or twelve or fifteen elementary parts. 1
Using the idea of machines made up of multiple copies of a small number of
standardized elements, von Neumann posed his question about robot self-reproduction as
follows.
Can one build an aggregate out of such elements in such a
manner that if it is put into a reservoir, in which there float all these
elements in large numbers, it will then begin to construct other
aggregates, each of which will at the end turn out to be another
automaton exactly like the original one? 2
Using techniques of mathematical logic, von Neumann was then able to deduce that
such self-reproduction should in fact be possible. His proof hinged on the idea that an
automaton could have a blueprint for building itself, and that in self-reproduction, two steps
would be necessary: (1) to make an exact copy of the blueprint, and (2) to use the blueprint
as instructions for making a copy of the automaton. The role of the blueprint is entirely
analogous to the way DNA is used in biological self-reproduction, for here the DNA is both
copied and used as instructions for building new proteins.
The complexity of a reservoir full of floating machine parts hindered von Neumann
from making his proof convincing. The next step came from Stanislaw Ulam, who was
1 From “Theory and Organization of Complicated Automata,” 1949, reprinted in John von Neumann,
Theory of Self-Reproducing Automata, University of Illinois Press, Urbana 1966, p. 77.
2 From “The General and Logical Theory of Automata,” 1948, reprinted in: John von Neumann,
Collected Works, Vol. 5, Macmillan, New York 1963, p. 315. The weird scenario described in this quote is
reminiscent of a scene in Kurt Vonnegut, Jr.’s Sirens of Titan where an unhappy robot tears himself apart and
floats the pieces in a lake.
p. 67