From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!psych.toronto.edu!michael Tue Mar 24 09:54:27 EST 1992
Article 4359 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!psych.toronto.edu!michael
>From: michael@psych.toronto.edu (Michael Gemar)
Subject: Re: Monkey Room
Organization: Department of Psychology, University of Toronto
References: <1992Mar6.004252.1593@ccu.umanitoba.ca> <1992Mar6.213755.17977@psych.toronto.edu> <1992Mar7.180909.10713@ccu.umanitoba.ca>
Message-ID: <1992Mar9.175416.8708@psych.toronto.edu>
Date: Mon, 9 Mar 1992 17:54:16 GMT

In article <1992Mar7.180909.10713@ccu.umanitoba.ca> zirdum@ccu.umanitoba.ca (Antun Zirdum) writes:
>In article <1992Mar6.213755.17977@psych.toronto.edu> michael@psych.toronto.edu (Michael Gemar) writes:
>>>1) how likely is this to happen.
>>>2) It can be argued that up until the room was opened, the 
>>>monkey & teletype formed a system that (just by the random
>>>ordering of atoms) was intelligent! When the room was
>>>opened the system died!
>>
>>This must be a meaning of the word "intelligent" of which I was previously
>>unfamiliar.
>>
>>The "Monkey Room" example is merely meant to show that seemingly intelligent
>>behaviour can arise by random chance.  Certainly you don't *really*
>>believe that a system that produces random responses is intelligent?
>
>It all depends on what you mean by random, is the 'random' person 
>in the following example producing random responses? And YES
>I do believe that if the system produces intelligent behavior
>it IS intelligent!

If you genuinely believe that you can describe the outputs of a randomly
behaving system as "intelligent," then we simply have no grounds for
discussion.

>>
>>> What is there that precludes that
>>>a bunch atoms will not come together (randomly) and
>>>form a complete human being, intact with memories and
>>>everything! Let us assume that he was your copy in
>>>every detail, now is he intelligent?
>>
>>Sure.
>>
>>>Same as the monkey argument!
>>
>>Not at all, and I am amazed that you can't see the difference.
>>Take the case of someone with a multi-sided die that has all the characters
>>in English.  If the die just happened to produce a string of characters that
>>seemed like a conversation, would you be willing to say that the die was
>>actually "intelligent"?  I would hope not...
>
>I am amazed that you cannot see the simularity!
>Re: the die, if you happened to ask it questions and it 'just
>happened' to produce the correct responses every time, what
>it your objection to calling it intelligent?
>Note: this is not the same thing as having the die reproduce
>Shakespeare, this can be considered truly random chance
>behaviour.
>Also, note that the die is not the THING that I am calling
>intelligent, it is the system that includes the dice!

But the system is *driven* by randomness, its behaviour is
*determined* purely by chance.  The rest of the "system" is
irrelevant to this point.

>(I am befuddled when people speak of randomness as something
>out of this world, the universe is not random in any way!
>It is uncomputable, but this does not mean that it is
>undeterministic, it simply means that to compute the universe
>you need a reference point somewhere outside the univerese,
>therefore something that we could never have!)
>From this point of view, when does something become random
>behaviour? This is the problem that plagues me, since a
>person can be created by random atoms, is that person
>henceforth random?

You are confusing at what level the randomness occurs.

>Another objection that I have heard recently, computers
>cannot possibly be entities like humans because they are
>tools of humans! Well, I have news for you, humans are
>also used as tools of other humans, does that make them
>any less intelligent, understanding, alive?

This *is* a silly argument, and hasn't been made by me or, as
far as I can recall, anyone else in this newsgroup.

>I do not see the differences that people like you purport
>there are between physical systems we call humans, and
>physical systems we call machines!

The crucial differences under discussion are between humans
and computationally equivalent programs.  If you do not see
the distinction between these things, then you have not been
following the discussion...


- michael




