Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!ittatc!dcdwest!sdcsvax!sdcrdcf!hplabs!hpda!hpisoa2!hpitg!spar!ellis@spar
From: ellis@spar
Newsgroups: net.bio
Subject: Re: Orphaned Response
Message-ID: <238@spar>
Date: Wed, 30-Apr-86 16:35:00 EDT
Article-I.D.: spar.238
Posted: Wed Apr 30 16:35:00 1986
Date-Received: Sun, 11-May-86 16:50:41 EDT
References: <311@dg_rtp>
Lines: 59

>>...  As far as I can tell, evolution is the implicit rationale whereby
>>     teleological arguments (`the purpose of my heart is to
>>     pump blood') are supposedly reduced to `proper scientific causal'
>>     explanations (`hearts gradually evolved because pumping blood
>>     led to species more fit to survive').
>>     
>>     What's so awful about `purpose', anyway? Goals imply that `information'
>>     exists which can refer to and cause potential future real world states
>>     of existence. ...
>
>Perhaps I'm misreading what you've been saying here, but what most
>people think about when they hear "goal" is the notion of a force
>directing a process toward an ideal future state. Saying "I'm going to
>direct everything in my life toward becoming a lighthouse keeper" is
>different than saying "everything that's happened in my life makes it
>possible that I'll be a lighthouse keeper some day." And this seems to
>be implicit in your concept of information that somehow refers to
>future real states. Suppose we have a dammed lake at the top of a
>hill, and a valley below. If the dam is removed, does the lake's
>flowing to a stable state in which all water resides in the valley
>involve somehow the system's referring to future real states in which
>the water either flows or doesn't flow downhill to the valley? And how
>does information 'refer' to a system's potential future states? 

    I do not see how the dam in your example possesses the required internal
    organization to contain a hypothetical model of the future inside
    itself, such as possessed by highly goal-oriented computer programs
    and biological systems.

[This]
>approach strikes me like the 'anthropic principles' certain cosmologists 
>are enamored of: a lot of fun to think about but ultimately not that
>useful as models that can generate falsifiable hypotheses.
>- Cheers, Bill Ingogly

    Cosmologists? Like who? Most of the ideas I'm concerned with here are
    from biology, cybernetics, artificial intelligence, cognitive science,
    and philosophy. I apologize for my stupidity and general inability to
    make a convincing case for notions that are as new to me as they
    apparently are to most readers...
    
    An example swiped from Dennett's "Brainstorms" is relevant here.

    If I were to try to understand a how a chess program worked, the first
    thing I'd assume is that its intention was to win at chess. As I
    understood it better, I might learn that it was not aware of 
    various goal-oriented concepts in chess, such as the desirability
    of advancing pawns with the intention of forcing queening situations.
    
    Intentional statements differ in verifiability from kosher causal
    statements in degree only. In both cases, we infer statements that are
    not logically deducible from observations; in both cases, we need be
    prepared to alter our hypotheses in the face of new evidence.
    
    I agree it is one goal of science to reduce intentional statements to
    `rigorous' causal statements. What I do not see is why such reduction
    necessarily invalidates teleological explanation. 

-michael