From newshub.ccs.yorku.ca!torn!cs.utexas.edu!sun-barr!west.West.Sun.COM!cronkite.Central.Sun.COM!texsun!exucom.exu.ericsson.se!pc254185.exu.ericsson.se!exukjb Wed Sep 23 16:54:36 EDT 1992
Article 6993 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!torn!cs.utexas.edu!sun-barr!west.West.Sun.COM!cronkite.Central.Sun.COM!texsun!exucom.exu.ericsson.se!pc254185.exu.ericsson.se!exukjb
>From: exukjb@exu.ericsson.se (ken bell)
Newsgroups: comp.ai.philosophy
Subject: Re: Defining intelligence
Message-ID: <exukjb.196.717031725@exu.ericsson.se>
Date: 20 Sep 92 23:28:45 GMT
References: <BILL.92Aug9124642@ca3.nsma.arizona.edu>   <1992Sep11.00857.16070@ms.uky.edu>
Sender: news@exu.ericsson.se
Organization: Ericsson Network Systems, Inc.
Lines: 158
Nntp-Posting-Host: pc254185.exu.ericsson.se
X-Disclaimer: This article was posted by a user at Ericsson Network Systems
              The opinions expressed are strictly those of the user and
              not necessarily those of Ericsson Network Systems.

In article <1992Sep11.00857.16070@ms.uky.edu> oldham@ms.uky.edu (Joseph Oldham) writes:
>From: oldham@ms.uky.edu (Joseph Oldham)
>Subject: Re: Defining intelligence
>Date: 11 Sep 92 04:08:57 GMT

>exukjb@exu.ericsson.se (ken bell) writes:

>>In article <1992Sep7.14929.16475@ms.uky.edu> oldham@ms.uky.edu (Joseph Oldham) writes:
>>>From: oldham@ms.uky.edu (Joseph Oldham)
>>>Subject: Re: Defining intelligence
>>>Date: 7 Sep 92 05:49:29 GMT

>>>exukjb@exu.ericsson.se (ken bell) writes:

>>>>In article <1992Aug13.044325.16707@zip.eecs.umich.edu> marky@dip.eecs.umich.edu (Mark Anthony Young) writes:
>>>>>From: marky@dip.eecs.umich.edu (Mark Anthony Young)
>>>>>Subject: Re: Defining intelligence
>>>>>Date: 13 Aug 92 04:43:25 GMT

>>>... lots and lots deleted ...


>>>>>OK, I've seen enough!  A new characterization of "intelligence": the 
>>>>>capacity to adapt means to ends.  Differential abilities in respect of 
>>>>>adapting means to ends by different orders of living being indicates 
>>>>>why intelligence is both a type-word and a straightforwardly adjectival 
>>>>>one. The more intelligent something is, the better able it is to find the 
>>>>>best means to attain its ends. 

>>>>>Theoretical Advantages:

>>>>>      1. Accounts for our use of the word to cover any kind of living 
>>>>>           entity capable of acting purposefully (this does not mean nor 
>>>>>           entail that entities to which we may properly ascribe 
>>>>>         intelligence select their own goals, only that they have them 
>>>>>         and go about attaining them, and that their behavior must be 
>>>>>         given a purposeful explanation.)             	

>>>>>      2. Accounts for our gut feeling that intelligence is closely 
>>>>>         tied to consciousness, because only conscious beings            
>>>>>         know values (rooted in needs & desires) and thus can have 
>>>>>         purposes. And, since consciousness is a matter of    
>>>>>         degree, so is intelligence.  	

>>>>>Anticipated Objection1: Doesn't intelligence enter in to the choice of 
>>>>>                      ends as well as means? 

>>>>>Anticipated Objection2: Must the ends be pre-visioned (requiring 
>>>>>                      consciousness)?

>>>>>------------------------------------------------------------------------------
>>>>>Kenny Bell                          *        Welcome to Mind Wars
>>>>>Ericsson Network Systems, Inc       *        Abstract Arts BBS 386-7907
>>>>>P.O. Box 833875                     *        Severity with oneself is heroism.
>>>>>Richardson, TX 75083-3875           *        --A.G.Sertillanges (France, 1943)
>>>>>------------------------------------------------------------------------------

>>>Why not just say intelligence is the ability to solve problems.  Then if you
>>>think of "What problem to solve" as a problem you make goal selection
>>>a problem to solve.  The solution to the the "problem problem" may come from
>>>physical needs --  to take on the problem of satisfying hunger is not
>>>necessarily a conscious choice, but an "intelligent" selection even if
>>>automatic.  When we choose a more abstract problem to solve -- like "What
>>>is intelligence" -- not always an intelligent choice of problem :-), we
>>>have a conscious solution to the "problem problem."  So maybe our "intellignce
>>>metric" has to include whether the problem we're solving is situtaion
>>>appropriate, whether we're working on it as a matter of ncessity or
>>>choice, how tough it is, and what our solution is.  Note we get high
>>>marks for choosing to take on a tough problem like 'What is Inteligence"
>>>even if we haven't solved it, and so long as we're not starving in the
>>>process :-).

>>>Seems to be pretty much what you were saying but perhaps fares better wrt your
>>>anticipated objections.

>>>J.O.

>>>-- 
>>>Joseph D. Oldham
>>>oldham@ms.uky.edu
>>>oldham@UKMA.BITNET
>>>home: 233 7614

>>Well, it would fare well in terms of the listed anticipated objections, 
>>except that it may have the rather queer result that we could not 
>>ascribe intelligence to a being in a completely unproblematic environment -- 
>>unproblematic either because of structural features of the environment 
>>itself or because of structural features of the being in question, e.g. this 
>>being had no needs to satisfy, no wants, no central nervous system so no 
>>pain-pleasure modality, etc.  If a being in these circumstances would have 
>>no "problems", then such a being could not, on your analysis of 
>>"intelligence", *be* intelligent at all!  And, come to think of it, if 
>>you're right about this, intelligence could *not* properly be ascribed to 
>>what people have meant by the word "God", since on the at least 
>>normative western theistic view obviously this being doesn't have any
>>"problems" to solve, nor, being omnipotent, would he have to adapt means 
>>to ends to accomplish any purposes he might have.

>>  Uhmm. Perhaps if given enough time I could think of a few more odd 
>>consequences along these lines, although I am now beginning to feel like I 
>>am mischieviously misinterpreting your views.  This is not my intent.  

>>------------------------------------------------------------------------------
>>Kenny Bell                          *        Welcome to Mind Wars
>>Ericsson Network Systems, Inc       *        ANTHEM BBS 386-7907
>>P.O. Box 833875                     *        Severity with oneself is heroism.
>>Richardson, TX 75083-3875           *        --A.G.Sertillanges (France, 1943)
>>------------------------------------------------------------------------------


>Gee. (or, "Don't tell God he doesn't have problems..." :-) ).
>Ok, a being in an unproblematic environment. (That'll be the day.)
>Can such a being have "ends"?  Say it can.  Then it has a problem of how to
>achieve those ends.  Oh, is it omnipotent too!  (That *always* creates
>problems :-).  Well, then it has a trivial solution to all of its problems.
>If I had a trivial solution to all my problems I'd feel pretty darned
>intelligent. (i.e. the "zero-adaptation" -- choosing just the right means,
>is intelligent behavior.)

>I surely intended my "problem solving" formulation to be quite close to your
>"means adapting" formulation.  As you point out, at least as I phrased it
>some odd results might be found.  But it seems useful.  And I think your
>earlier "anitipcated objections" need an answer far more than any objection
>grounded in a totally unproblematic environment -- which I can't really
>conceive.  If you warp the environment too badly you may well find an
>environment in which "intelligence" really isn't worth talking about. Then
>you'll be out of job.

>J.O.

>-- 
>Joseph D. Oldham
>oldham@ms.uky.edu
>oldham@UKMA.BITNET
>home: (606) 233 7614

 
More interesting, perhaps, might be whether in my warped environments 
intelligence *could* still be talked about?  Maybe we need to set up some 
some possible worlds semantics and ask whether we would still be able 
to talk about something being intelligent in those worlds?  Then, by 
a process of elimination, we could pick out what aspects of what possible 
worlds would have to be present in order to allow for the presence of 
intelligence in those worlds?  I guess you would start with a minimal 
possible world and build from there, starting with a world in which 
intelligence would be possible until you get a situation in which the 
concept of intelligence could get no purchase at all.  





------------------------------------------------------------------------------
Kenny Bell                          *        Welcome to Mind Wars
Ericsson Network Systems, Inc       *        ANTHEM BBS 386-7907
P.O. Box 833875                     *        Severity with oneself is heroism.
Richardson, TX 75083-3875           *        --A.G.Sertillanges (France, 1943)
------------------------------------------------------------------------------


