254 Are the Androids Dreaming Yet?
Epstein Suite indexes the text; the original document lives at its official source. We don't host the original file — view it on the official release to read it in full.
View the original on the official releaseDocument text
Text is machine OCR and may contain errors. Confirm against the original source above.
254 Are the Androids Dreaming Yet?
chaotic equation so complex that the only way to see it is to run the
program and watch the output: no form of program analysis will give
you any clue as to what it produces. There is only one stipulation. The
answer to problem Y MUST be held within the program as a computable
algorithm. Put another way, the computer must already be ‘programmed’
to answer the question.
Could a human mathematician be pre-programmed from birth?
Yes, there is no fundamental objection to this. Mathematicians could be
born to solve the problems they solve. But this would present a couple of
issues. Where is this program stored? And who, or what, programmed
the mathematician? Could we perhaps find an experiment to determine
whether mathematicians are pre-programmed?
One view held by philosophers is that the Universe programmed
the mathematician. They believe we live in an entirely determined
Universe with no free will. There is then no mystery as to how Andrew
Wiles came up with his proof. He was destined to do it from the dawn of
time. The ink that fell from his pen to the paper was always going to fall
in just that way. We live in a clockwork Universe and although we might
feel we have free will, this is an illusion. I simply don’t believe this. If I
am right and humans do exercise free will, Andrew Wiles cannot be a
computer. And because Andrew is not alone in discovering proofs, those
mathematicians cannot be computers either. Humans are, therefore, not
computers.
The Chance Objection
I said there was no automatic way to solve any problem above the
logic limit, but this is not quite true. There is one automatic method
you could deploy to generate a non-computable proof, the infamous
‘monkeys and typewriters’ idea where we use random chance to generate
information. Many people have suggested it is possible to write a play
such as Shakespeare's Hamlet by simply typing random characters until
we happened upon the play. The argument is flawed.
The first flaw is the process would take a super-astronomically
long time. Even if every atom in the Universe were a monkey with a
typewriter, it would take orders of magnitude longer than the age of the
known Universe to come up with the script to a play or a mathematical
proof.
The probability of finding a solution to Fermat’s Last Theorem
by chance is about 1 in 10°°. That’s 1 with 50,000 zeros after it. For a
comparison, there are only 10’ atoms in the known Universe. To be, or
HOUSE_OVERSIGHT_015944
Have a question about what this document contains?
Ask the documents