--- In ai-philosophy@yahoogroups.com, Marvin Minsky <minsky@...>
wrote:
>
>
> On Aug 1, 2007, at 9:38 AM, anssihyytiainen wrote:
>
> > That is, if we just simulate the behaviour of each neuron of a
human
> > brain and use the results of that simulation as an AI-system, we
> > still don't know if there is a subjective experience to the system
> > (albeit it functions like any person when viewed from the
outside).
> > That is true because we do not know what are the requirements for
a
> > subjective experience. "Intelligence" is different
from "subjective
> > experience".
> >
> > I agree that Searle is making a (fatal) mistake when he just talks
> > about "human properties", for example when he says semantics is an
> > intrinsic property of humans.
> ----[snip]
> >
> > However, "semantical understanding" does not automatically mean
there
> > is subjective experience!
> >
...
> And
> if you still are wondering about
> the nature of subjective experience, I'd like to see your comments
on
> the ideas in section 9-5 of
> http://web.media.mit.edu/~minsky/E9/eb9.html

I remember making a comment couple months ago about the
of "directness" and about the higher cognitive levels having a better
access to what is going on at lower ones.

I don't think it is the directness (or "effortlessness") that
Chalmers finds so mysterious. I.e. I think he could well agree with
your assertion; "the apparent `directness of experience' is an
illusion that comes because our higher mental levels have such
limited access to [the lower level systems]..."

But how I think he would reply is that if we did have subjective
experiences about the (simpler) lower level processes, those
experiences would be equally mysterious, because they would also be a
result of "many little things doing their dance" (performing pattern
recognition to the patterns at the lower levels). (A naive question
would be "why don't each of those little things have an experience of
their own")

So what I'm saying is, I would make a different reply to Chalmers;

CHALMERS: "Why should a physical system, no matter how complex and
well-organized, give rise to experience at all?"

...on the other hand, why should it not? There are certain (tacit)
assumptions in our worldview that can make it seem like subjective
experience would not rise, but then apparently some physical systems
DO have a subjective experience. So we can safely say it is possible
for a "physical system to have a subjective experience". (I'm not
going to discuss solipsism)

I'm not even going to get into "pain" or "pleasure". I'm just talking
about how do you "see a box", when it's recognition has in fact been
performed by many little things doing their dance.

The error in my previous sentence is when I said "many little
things". It is once again up to our very own interpretation of
reality to break it into "things" in some specific way. That's the
critical clue. We do it, because it allows us to understand, among
other things, how brains work.

There is absolutely no reason in the world to assume that "things"
(that we have defined!) by themselves have "experiences" in some
naive realistic sense. Perhaps there is no "correct" way to classify
reality, but then there is also no reason why one could not adopt
some sort of "process-centric" view where reality is broken into
(semantically defined) "physical processes" instead of "things", and
consequently say "physical processes" have experiences. (or
rather "are" experiences)

My "self" can be seen as a physical process. I have an experience.
Seems to work so far!

The interesting thing is that it is our inclination to break reality
into (semantical) things that first of all allows us to understand
reality, but then again block us from understanding what it means
to "experience" (in so far that we don't accept some naive realistic
view of "any given thing experiencing reality as it hits its surface")

btw, when I say "processes are experiences", I don't mean to say
processes are ontologically fundamental "entities" (that have
temporal identity to themselves), or that "all processes are or have
experiences". It would be silly (incoherent) to claim that. Clearly
there are some specific requirements before a subjective experience
exists. Evidently the processes in the brain need to be very specific
for the subjective experience to occur (=we can become unconscious)

Likewise one should not take it as an assertion of ourself being
somehow "the same process all the time" (identify that process
as "the same process" ontologically from one moment to the next).
That is, again, because when we say something is "the same" from one
moment to the next, that is entirely up to a specific semantics. I.e.
completely up to how one wants to understand the situation. There's
just no ontology behind it.

-Anssi