[ExI] How not to make a thought experiment

[John Clark]

Since my last post  Gordon Swobe has posted 5 times.

> No sir. I have never written a program that had consciousness.

And I will tell you exactly how Swobe knows this, because he as never written a program that acted intelligently.

> it seems to me totally absurd to assign consciousness to a pile of beer cans and toilet paper, or to a billion men pulling levers, or to a nation of Chinese people talking to one another on the telephone (Block, 1978) or to an arrangement of cats chasing mice, or to pigeons trained to peck as Turing machines (Pylyshn, 1985) or to any number of other such theoretically possible "Turing equivalent" implementations.


We are asked to accept that certain ideas are untrue entirely because of the incredulity of a person who goes by the name of Gordon Swobe. Never mind that Turing proved that theoretically you could get any behavior you wanted out of one of his machines, never mind that Darwin showed that if it were not linked with intelligence consciousness would never exist on planet Earth, never mind that the fossil record shows that strong emotion existed hundreds of millions of years before advanced intelligence; none of that matters, Gordon Swobe thinks a beer can computer is kinda goofy and if Swobe thinks something is odd that proves it could never exist. 

> Functionalists of the computationalist persuasion need to defend those kinds of bizarre notions to support their philosophy.


But Swobe believes a thinking machine made out of grey goo is not bizarre because goo is inherently more logical than beer cans or toilet paper.

> I think most people find it intuitively implausible that we would create a new conscious entity if we trained a large group of pigeons to type in a Turing equivalent pattern. 

What Swobe says above is correct but it tells us nothing about nature. It's intuitively implausible that an object in motion will stay in motion unless acted on by another force, it's intuitively implausible that simultaneity is not an absolute property, it's intuitively implausible that one object can be in two places at the same time; and yet all these things are true. There is a reason our intuition stinks in these areas, it's because the conditions where these facts become important were unlikely to be encountered by our ape-like ancestors, so Evolution had no reason to make our intuition good in these areas.

 John K Clark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20100224/bf1fd3bc/attachment.html>