RE: Metarationality (was: JOIN: Alden Streeter)

From: Ben Goertzel (
Date: Sun Aug 25 2002 - 10:20:05 MDT

> > It evolved from and arises from a complex of
> > other processes. These other processes are arational or non-rational
> > (language consistently used by Goertzel and others here), not
> > _irrational_ (language incorrectly implicitly attributed to them).
> They are neither arational nor nonrational; they are *rational*.
> Or rather they are arational processes which, insofar as they work, work
> because they sometimes happen to implement something which roughly
> resembles rationality - we *are* thinly modded chimps hacked to support
> general intelligence, after all. But the point is that, insofar as these
> processes work, they work because they are rational and are rational
> because they work.
> --
> Eliezer S. Yudkowsky
> Research Fellow, Singularity Institute for Artificial Intelligence

Eliezer, you're using the word "rational" to mean "anything that works".

So if you have an organism with a goal, then any process that it uses that
significantly helps it to achieve that goal is "rational" by your
definition. (Correct me if I've paraphrased your idea incorrectly, please.)

This is a meaningful sense of the word "rational", but it's a very broad

In this broad sense, it's true, even processes that have nothing to do with
*reasoning* as commonly conceived may be viewed as "rational."

For instance, a bird, acting purely on instinct, is acting rationally when
it mates with another bird, because it's acting in a way that "works" in
terms of achieving its goal of reproducing itself.

I generally have been using the word "rational" to mean something narrower.
Basically, to mean "following a process of logical inference", meaning a
process inside some mind in which premises are represented, conclusions are
represented, and processes for getting from the premises to the conclusions
are carried out. Note that this sense of "following a process of logical
inference" is implementation-independent; it applies to neural nets and so
forth as easily as to minds embodying some sort of explicit
symbol-manipulation component.

I guess that your broad definition, and my narrow definition, both capture
different aspects of the ambiguous natural language term "rationality."

In my sense, there are nonrational cognitive processes, because there are
processes that don't proceed step by step but LEAP to conclusions based on
nonlinear-dynamical integration of a vast amount of weak pieces of evidence,
sometimes using emotion as a guide. In your sense, if these processes
"work" in terms of achieving system goals, then they're rational.

Not all of our disagreement is wiped away by clarifying our different uses
of the term, but I have a feeling about 75% of it is.

I would suggest that, in other forums, if you wish to make sweeping
statements about rationality, you clarify first your extremely broad
definition of the term. Certainly, in future, before talking about
rationality I will take care to clarify my own meaning of the term, so as to
avoid confusion....

-- Ben G

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:40 MDT