From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!usc!orion.oac.uci.edu!cerritos.edu!arizona.edu!arizona!amethyst!organpipe.uug.arizona.edu!NSMA.AriZonA.EdU!bill Tue Mar 24 09:55:54 EST 1992
Article 4470 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca comp.ai.philosophy:4470 sci.philosophy.tech:2291
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!usc!orion.oac.uci.edu!cerritos.edu!arizona.edu!arizona!amethyst!organpipe.uug.arizona.edu!NSMA.AriZonA.EdU!bill
Newsgroups: comp.ai.philosophy,sci.philosophy.tech
Subject: Re: Causes and Goals
Message-ID: <1992Mar16.031843.14299@organpipe.uug.arizona.edu>
>From: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Date: 16 Mar 92 03:18:43 GMT
Reply-To: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Sender: news@organpipe.uug.arizona.edu
References: <6374@skye.ed.ac.uk> <1992Mar11.201637.21875@psych.toronto.edu> 
 <1992Mar14.010014.552@organpipe.uug.arizona.edu> <1992Mar15.170938.9882@husc3.harvard.edu>
Organization: Center for Neural Systems, Memory, and Aging
Lines: 101

In article <1992Mar15.170938.9882@husc3.harvard.edu> 
zeleny@zariski.harvard.edu (Mikhail Zeleny) writes:
>
>Bill Skaggs:
>>  What, then, is a "goal", and how does something come to have
>>one?
>>
>>  Answer:  Any system (such as the human genome) that has evolved
>>by natural selection can be assigned the primitive goal of 
>>surviving.  Other goals and functions are inherited from this
>>primitive goal.  Machines inherit goals from the fact that they
>>are designed for specific purposes by beings that have evolved
>>by natural selection, and the components of machines perform
>>functions in helping them achieve their goals.  

MZ:
>Now you have a problem.  

I have a lot of problems, but let's not dwell on it, please.

>Setting aside the commonly acknowledged problems
>inherent in the sort of vitalism you seem to be assuming (most evolutionary
>biologists these days disclaim the theoretical indispensability of any
>notion of final cause), . . . 

Virtually all evolutionary biologists recognize the necessity of
functional (and thus teleological) explanations.  Some, for example
Gould, feel it is important to stress that not *all* features of
organisms need have functional explanations.  Even Gould, though,
agrees that many aspects of phenotype can be explained by the fact
that they increase the fitness of organisms.

> . . . what you have is yet another quasi-materialist
>vulgarization of the Hegelian view of universal history, the main
>difference being that the primitive goal you postulate is outward-, rather
>than inward-directed.  

No!  Please!  Not Hegel!  Anything but Hegel!

>Now, the goal-inheritance scheme you are advocating
>must be either wholly transitive, . . .

It is.

> . . . i.e. reducing the teleology of every
>natural system to the one primitive goal of survival, . . .

I prefer to say "deriving the teleogy of everything that
has teleology from the one primitive goal of survival".
There is a trap here that needs to be avoided.  It is
easiest to explain with an analogy from the game of chess.
The goal in a game of chess is to win, but that goal is usually
not directly useful for selecting moves, because it is
usually impossible to calculate directly whether a given
move will lead to victory.  Moves tend to be made on the basis
of secondary goals, such as, for example, maximizing the
value of one's pieces, or maintaining a good pawn structure,
or getting a rook on an open file.  These secondary goals
are derived from the observation that they are correlated
with winning, but the correlation need not be perfect.  On
some occasions moving a rook to an open file gets one
quickly checkmated; nevertheless, it is still true that
the secondary goal of having a rook on an open file is
derived from the primary goal of winning.

But actually, the goal of winning is not "primary".  It is
derived from some kind of social goal.  There may be several
more levels of derivation, but at the end of the chain is 
the goal of survival of the genome.

Because secondary goals are only indirectly related to the
primary goals they are derived from, the actions they lead
to may sometimes be irrelevant or even counterproductive
with respect to the primary goals.  Thus the human value
system, derived from the primary goal of survival of the
human genome, could conceivably lead to the destruction of
the human genome in a nuclear holocaust.

> . . . or compositional,
>i.e. capable of combining this goal with other primitive (irreducible)
>goals . . .

But how would such goals come into being?

> . . . on any level of organizational structure; in other words, either all
>of your goals are wholly reducible to the single primitive goal of survival
>of the species, in which case you don't have any personal goals that would
>enable *you* to mean *anything* by your own words, . . .

As I have explained above, personal goals can be derived from
the primitive goal of survival without being identical to it.

> . . . or you have some private
>goals, in which case your evolutionary story is bereft of any explanatory
>force in determining the private semantics of your utterances.  Either way,
>you lose.
>

I think I have blunted one of the horns of this dilemma.

	-- Bill


