Re: Singularity-worship

From: Eliezer Yudkowsky (sentience@pobox.com)
Date: Thu Dec 05 1996 - 20:33:28 MST


> >The SAT - not IQ tests, just the SAT - measures the ability to trace
> >single causal links.
> This is a remarkably strong assertion, backed up by exactly nothing.
> Philosophers are still debating what the phrase "causal link" means,
> so I'm surprised to find that you seem to think that we have a test
> that measures the ability to trace these things.

Of course philosophers are still debating over what causality is. This
is probably because the last time I checked, there were no less than
eight separate and distinct phenomena that could arguably be called
"causality", and philosophers still haven't untangled the four types
they know about - computational, platonic, thermodynamic and cognitive
causality.

You appear to be interested in "cognitive causality", that is, the way
in which the human mind assigns causal links. This is a matter solely
for cognitive science/artificial intelligence and doesn't affect
philosophy in the least, except that philosophers need to know what's in
their heads so they don't get it mixed up with the rest of the Universe.

As for my assertion about the SAT, it's backed up in Algernon's Law.
The short version is that I scored astronomically high on the SAT (age
eleven, 670V/740M, came in second for the 7th grade of the entire
Midwest) but have a relatively normal IQ of 143, and I do causal
analysis, so the SAT apparently tests just that. Don't ask me why.

> I'm not asking for a general definition of
> "meaning", just for a definition of the phrase "meaning of life" as you
> use it.

Didn't I just do that? The meaning of life may be defined as a
self-justifying goal. It's the ethical equivalent of the First Cause,
which has no causal precedents. The Meaning of Life is simply something
meaningful in and of itself, without being part of a greater goal or
being urged on us by evolution.

> The word "self-justifying" is empty of content. Justification is a
> relationship between two distinct information structures. If *anything*
> is self-justifying, then everything is self-justifying; hence, the
> phrase "self-justifying" does not distinguish a set to which it applies
> from a set to which it does not apply. Terms that do not distinguish are
> empty of content. Or perhaps you are prepared to explain to me what is
> the precise difference between self-justifying and non-self-justifying
> entities?

You've been taking too many philosophy courses. The above diatribe is
not distinct from: "Reference is a relationship between two sentences.
If one sentence is self-referential, they all are." And yet, not all
sentences are self-referential, nor is the term "self-referential" empty
of content.

An example of a self-justifying entity would be the reason why anything
exists at all. An example of a non-self-justifying entity would be the
goal of crossing the room.

> If that's your definition of smartness, then I doubt that there is
> any such entity. There is no such thing as "generalized problem
> solving skill"; rather, there are different skills which are useful
> for solving different kinds of problems. We have exactly no reason
> to believe that all skills are reducible to some hypothetical
> "uberskill" called "intelligence" or "smartness"... these latter
> words refer to ill-defined bundles of skills, and you simply
> cannot measure ill-defined bundles.

What a philosopher! Suppose I say: "Smartness is an abstraction,
existing solely in the human mind, from the observation - also existing
in the human mind - that System X - also in the mind - can solve problem
Y - also in the mind. To give this term, "smartness", a useful
definition, we say that if System X can solve more problems than System
Y, or solve them more elegantly - where elegance is in the mind - then
System X is smarter than System Y. The terms used may not precisely
reflect the truth, not having definitions down to the level of such
definitively real items as quarks or whatever the fundamental particles
may be, but if we attempted so foolish an endeavor as to make all mental
assertions correspond precisely to reality, we wouldn't be able to think
or walk across the room. The most we can hope for is terms which are
useful, experimentally testable, and precise. I believe that
"smartness", or - so I don't get another lecture on there being multiple
types of smartness - that type of smartness which has to do with the
rotation of mental pictures - is experimentally testable, precise, and
useful. Other types of smartness are more vaguely defined due to our
primitive grasp of cognitive science, but are still useful.

> The primary thing I object to about your "diatribes", as you call
> them, is that you are treating a large number of important open
> questions as if they were closed questions. As for your tagline,
> that's why I took this private.

You came a bit close with that question about causality. I won't say
what the other four types are. This letter at least is being redirected
towards the list, since apparently there are still people out there who
have trouble with my working definition of the Meaning of Life.

Yours,
Eliezer S. Yudkowsky.

-- 
         sentience@pobox.com      Eliezer S. Yudkowsky
          http://tezcat.com/~eliezer/singularity.html
           http://tezcat.com/~eliezer/algernon.html
Disclaimer:  Unless otherwise specified, I'm not telling you
everything I know.


This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 14:35:52 MST