Re: Sentience

From: Steve Nichols (steve@multisell.com)
Date: Wed Dec 20 2000 - 01:44:01 MST


Date: Wed, 20 Dec 2000 20:48:45 -0500 (EST)
From: Dan Fabulich <daniel.fabulich@yale.edu>
Subject: Re: Sentience

> >Don't be silly. I commit no Athena fallacy: I have no account
> >WHATSOEVER about how mind comes about, say nothing of one like you
> >describe.
>
S> Exactly, the Athena Fallacy IS that you have no account of how the
S> mind came about. The only human-been philosophical resource is
S> the method introspection ... which is an inadequate tool.

><quibble>To say that someone has committed the Athena fallacy is to
>say that they have some *incorrect* view as to how the mind comes about.
>I am in *ignorance* as to how the mind comes about. I don't *know*.
>I can't possibly be committing the Athena fallacy on account of
>this.</quibble>

No, the point about the Athena Fallacy (my invention, so you can't say
it means something different) is that consciousness & intellect did not
emerge fully formed .... but arose over millennia, with intermediate phases.
Philosophers IGNORE the evolution of consciousness and try to deal with
what find through introspection (or verbal examination). However, you cannot
fully explain the presenting phenomena without an understanding of the
formative processes. For example, your analogy about the Turing machine
has no precedents in evolution, or any basis in fact about the architecture
of the E-1 brain. You are not alone, most philosophers make this error.

S> I have pointed out that I don't care if you phrase your description of
S> consciousness as " a phantom median eye" or as "the experience as
S> if a median eye was present" ...... but the vernacular does not impact
S> on the phenomena, your whole posting is an attempt to confuse the
S> issue ... clarification requires elimination of "epiphenomenalism"
S> gobbledegook and all the other philosophical pseudo-language.

>But if the phantom eye is a feeling, a mental object, then you're no
>better off than when you started. We began by saying "I wonder how
>the physical world is related to these mental objects!" if you reply:
>"they're linked through the phantom eye!" and it turns out that the
>phantom eye is, itself, a mental object, then you haven't answered the
>question.

I think that there is a recursive, or additional self-referential property
owned by the illusion of the median eye that is not shared by other
aspects/objects of consciousness, which is that the neurosignature
associates it as being PART of the self, the physical body, rather than as
externally originating. Muller's principal &c. links particular sensory
signals to the nerve or special sense organ that carries the signal.

The feeling of the phantom median eye, unlike visual sensations that go
away when we close our eye, is pervasive and continuous (the archetype
physical "mind/ pineal eye " had no eyelids or muscles of direction). So
I have always said that the phantom "eye" is not a separate entity from the
"phantom median vision" ..... my theory started as being called Phantom
Eye hypothesis, but I changed it to MVT precisely to emphasise that
what is important is that "consciousness" or mind is median-vision shaped,
which tells more that median-eye shaped.

S> How is a hypnotic suggestion or verbal cue a "chemical stimuli?"
S> The brain of one person will react in response to suggestions
S> given by another person (or TV &co). Change of state, both physical
S> and conscious.

>My error. I was interpreting "chemical" to include physical stimuli
>as well, without telling you. Anyway, the point stands that the brain
>cannot act independently from its external physical stimuli; that it
>is entirely determined by them.

What about dreaming when the body is paralysed and there is no
external physical stimuli?

S> If we were, then we could not experience internal light with no
S> outside reference or causal source.

>You're relying on the philosophical parts of MVT to defend this claim;
>these are the very points which I reject.

OK, I've only sold you on part of the package so far, but think that
the conceptual stuff likely follows, even if I haven't yet developed a
conscious machine on the MVT model, or a telepathy device to check
for consciousness. There seems no reason in principle why a device
could not me made that could sit in a shielded room, away from a
subject looking at playing cards or visual stimuli, and arriving at correct
predictions of what card the person is thinking of ... quantum physics
allows this possibility, as does logic. Then your points about subjectivity
might be rejected.

S> I am rather insulted that you compare me (and everybody else) with
S> lumps of rock.

>People were offended when they were compared with apes. But the
>analogy is apt.

On the contrary, I identify with my primate brethren, and do not think I
have any brain parts that E-1 animals such as cats have. But what value
does your rock analogy have, we are nothing like rocks? You are really
stretching this doubtful "all made of atoms" thing to absurd lengths.
How is a rock intelligent, let alone conscious?

S> How do you deal with holes (distinct from GAPS, which depend on
S> the fact that there was ONCE something present but no longer, whereas
S> a hole might be integral)?

>Yes, I cut them out because I agreed with them.

So you accept that there are non-physical objects, and further accept the
difference between "hole" and "gap" although both may be indistinguishable
at the time of examination? Do you accept that holes are completely
PSYCHOLOGICAL (contents of consciousness) and not at all physical?

>"NO!" you say. "It solves the Leibniz objection!" No, it doesn't.
>See, non-physical objects can't move physical objects around, and they
>CAN'T MOVE HOLES AROUND EITHER. They can't in any way affect the size
>or shape of holes; to do so, they'd have to move the *stuff* near the
>holes, which, you concede, is impossible. Neither does it help us to
>say that the *holes* cause consciousness. Because, while you can show
>that the holes are necessary for conscious behaviour, you can't show me
>how the *holes* can affect the non-physical world, either. You just
>say that they DO affect the non-physical world, but you *don't* say
>how *holes* can do that.

I raised the hole point to illustrate that non-physical entities can exist.
There is an analogy between holes and the abstract median eye.
But what is additional regards the median eye is that its FUNCTIONS
as well as its form have been lost. There is an abstraction activity in
place of the original median eye input activity .. nature abhors a
vacuum!

>You can say: "Listen, not-having a median eye causes consciousness
>reports, and other intelligent behavior." And I say, great! This is
>fantastic news. That's MVT'. But then you go on to say: "What's
>more, not-having a median eye causes consciousness itself!" and I say:
>How can you possibly prove that?

OK, but questioning how I can prove it is not questioning that it is true.
I can triangulate, and use evidence from several independent fields
(physiology, evolutionary, psychological, conceptual/ philosophical &c.
that all point towards MVT as having solved consciousness, but as you
have realised, the empirical data alone only makes a probabilistic case
since measurability of subjective experience is (not yet, and maybe never)
possible.

This does not make MVT any less true. We must evaluate competing theories
and go with our best judgement .... name ANYTHING that can be proven to
absolute certainty. Deductive logic and mathematical proofs rely on the
identity statement (1 + 1 or a = a) and so can be denied.

S> >Brains are like Turing machines in that they are composed entirely of
S> >simple physical parts that do simple physical things. These physical
S> >parts move deterministically in lock-step. (Though, to qualify this
S> >view, I'm a fan of many-worlds.)
>
S> the dualist problem has been described
S> precisely that the pineal gland (or any brain part) cannot be "moved" as,
S> say, a hand can be moved.
S> What parts of the brain "move?" ...

>Neurotransmitters flushing about in your synapses, along with ions
>pumping through your membranes, allows electro-chemical signals to be
>sent throughout the brain and down your axons to your muscles, glands,
>and other parts of your body. These are the brain's "moving parts."

Out of context ... the Descartes' discussion is about wilful moving of the
brain, like the hand can be moved. We cannot move any bits of our brain.
Sure, blood moves through it, and nerve impulses, like all of the rest of
our
body, but these are autonomic processes ... not consciously controlled.

S> Also, the point with parallel computing is that things happen in
S> PARALLEL, not in serial lockstep.

>True, but this can be emulated by a Turing machine: in a given quantum
>of time, it updates the first part, then the second part, then the
>third part, ... and finally the Nth part. Then it updates the next
>quantum of time, etc. Thus a Turing machine is functionally
>equivalent to any parallel machine.

Are you appealing to the reduction to statistical mechanics when you make
this claim? Even if you are, you are still wrong because the parallel brain
works in REAL TIME, not in simulation. Furthermore, the neural computers
exhibit graceful degradation and can re-route and self repair. The
grandmother
cell claims have been widely rejected, memory and information is distributed
and associated with weight-states, so what experts are you citing to back
your
claim that the mammalian brain resembles a serial computer?

S> >Moreover, the physical world is causally closed. Nothing non-physical
S> >interacts with the brain.
>
S> See the verbal cue example above.

>Ah, but verbal cues, while they aren't *chemical*, are entirely
>*physical*.

I think I have you on the run here. Say the verbal cues are written on a
page ... together with a load of meaningless scribble .... and further
say that you are dreaming of seeing the words. Maybe they are shocking,
something like "The Writing is on the Wall" and you wake up, shocked.

Which part of the verbal cues are physical here?

S> Yes, nothing non-physical interacts with the PHYSICAL bits of the
S> brain, but thoughts (dreams, and other non-physical stuff) can
S> interact with the NON-physical gap in the brain. Consciousness is
S> the infinite-state feedback loop between the brain and the
S> environment made possible because the brain expects to receive data
S> from a peripheral sensor that has gone missing.

>Right. So: a red-wavelength beam of light hits my eye. It hits
>red-receptors in my eye, which send electrochemical signals up to the
>brain. How does this signal make me have the conscious sensation of
>red? All I can see it doing is causing more physical activity in the
>brain. The fact that there's no medial eye there doesn't seem to help
>this explanation out.

Yes, a key question .... we can see that the original red wavelength is
translated into action potential signals by the rods and cones so that
the information can be carried along the optic nerve. There is no actual red
light anymore, but the end perception is of a red light, not of an
electrochemical
buzz. The abstract/ illusory sensor (locus of awareness) is the part of the
brain
made of the same type of information, and allows the new signals to be
reintegrated as meaningful experience ... (1) It is generic, the pineal eye
predates special senses and of course the electrochemical medium is the
same (generic) whether the original information is auditory, visual or
olfactory.
(2) I don't like Dennett's Cartesian theatre case, but do think that the
conceptual
space vacated by the old physical sensor(gan) does give a structure for the
mass of generic neural information to be interpreted against. The TV screen
decodes signals (via the aerial/ nerve channel) into a moving picture and
sound
that can be observed by the viewer ... the phantom median eye provides a
virtual
screen (shape and place) to reintegrate experience.
(3) There is a neurocomputational account based on the deep pathways of the
limbic brain associated with median eye data processing in early information
that underlies more recent pathways, for instance in the visual cortex.
Gibson
gives a gestalt account of your red light example ... and would say it
actually is
seen IN the eyeball .. but the organism reacts as a whole entity (eg. if the
light
is a car about to hit you, your whole body instinctively takes evasive
action.

I do think that E-2 animals (with pineal eye still working) actually see red
light,
but I claim specifically that they cannot IMAGINE red light without it
actually
being present in their retina. It is abstract and self-referential
consciousness
that MVT seeks to explain .... including dreams. Insects and E-2 animals
can't think about the red light or create the signals for it internally, and
so are
rather more deterministic animals than we (or monkeys) are ... though I
would
say even in the case of insects they display teleological, ends directed
behaviour,
and would not want to rule out awareness entirely.

>Similarly, suppose my consciousness has made a decision. It wants to
>send a message down to my hand to type the word "red". How does this
>signal get to my brain? Once again, the fact that there's no pineal
>eye doesn't answer the question. And, not coincidentally, no signal
>from my consciousness is necessary: my brain seems to be doing all the
>work *already*: that signal from my eye triggered a purely physical
>complex system, which sent the message back down to my hand to write
>the word "red." How did the consciousness help this process? Once
>again, the non-existence of the eye doesn't help.

Sure, much background processing is going only unaware to you ... but
we have an ability to "focus our mind" or concentrate, and to make
decisions. Our ancestors' attention (during reptilian phase) was the
unavoidable spotlight of the pineal eye/ median spectral field, which
reacted hormonally in response to light, and the hormones governed
muscle action and behaviour directly. We now act indirectly, but the
attentional "space" or pervasive illusion of self, is where the pineal eye/
median spectral field would have been, the same centres in the brain are
activated (I think the reptilian, limbic structures and brain stem,
hippocampus
and amygdale and still fundamental in formulating our emotional, reactive
responses to information received).

So I think we can focus in on particular information of importance ....
rather
like the reading head of your Turing machine perhaps (?) because of this
singularity (in Leibnitz' monad sense, the undivided "structure" in the
brain)
and make Descartes' "sens communs" out of the chaos, the diverse sense-
information ..... an illusion or simulation of the outside world represented
on
the walls of our internal cave (Plato) seen by a sense-organ that itself is
an illusion or hallucination of a structure that no longer juts out of our
skull.

>Let me predict an answer: "the holes are in just the right places and
>in just the right shape such that they were intimately involved in
>both physical systems! The non-existence of the pineal eye explains
>why you wrote the word 'red'; what's more, that 'red' signal passed
>right through where the pineal eye would have been!' You're sure
>right about that. But all this is part of MVT', the part of MVT
>that makes only scientifically verifiable statements.

Now you can compare your prediction against my somewhat fuller
answer, I hope it is a bit clearer to you, and that you might see where
your Turing notion comes from?

>What you didn't explain is what the holes have to do with
>*consciousness*. Why does the fact that the pineal eye
>would-have-been there imply that your conscious mind had a
>*sensation*? How did your conscious decision get the holes into the
>right places?

Don't get too carried away with the holes analogy (which comes
originally from Descartes talking about memory being a process
rather like punching holes in a cloth .... or card-reader .... it is
the functions and neuronal pathways associated with trace pineal vision
that interests me more.

>Conscious behavior IS a trick of nature. But what about consciousness?

I am not sure that I agree with your behaviourist distinction. Consciousness
is a clumsy, blanket term, I agree. "Mentation" or "mental events" are
not "behaviour" in the same way that having a pee is behaviour. I happen
to think that we are in "trance" to some degree or other all of the time,
after all, even you seem to accept that we only experience red light
indirectly
via the medium of electrochemical signals, and thus can never know the
world "directly."

In this analysis, there isn't any "behaviour", just awareness of behaviour.
So the question is not 'what about consciousness' which is a given, but
"what about BEHAVIOUR?"

S>> I find it hard to agree with your Epiphenomenalist assumption that
S> everything that has happened in the universe to date would have
S> happened in exactly the same way if no animal had ever been even
S> remotely aware of anything, and that we are robots following out a
S> behavioural script. This is not my experience.

>This is like replying to the claim that the earth spins and revolves
>around the sun with the claim "this is not my experience." From here,
>it looks the same either way! It looks the same whether the sun goes
>around the earth or whether the earth goes around the sun. Only when
>we attempt to do some scientific inquiry does the correct view come to
>the foreground.

My experience is shared by most people, and even most philosophers,
in that I can find not the slightest shred of evidence to support the wild
epiphenomenalist claim that everything would be EXACTLY the same
if no consciousness or feeling had ever happened. You accept this
logical extension of your position presumably ... it is standard philosophy.

>Similarly, from the inside, it looks the same whether you're making
>decisions or whether you're acting on externally determined desires.
>Indeed, I argue that you've been externally determined to believe that
>your desires are internally determined. But when we do some science
>we see that only physical elements are at work, and that non-physical
>elements play no role in the way we behave.

You persist with this belief-desire hypothesis ... there is plenty of turgid
response to this in the literature. Suffice to say (actually replying to
your
earlier point that I can't dropMVT because of my desire to have that belief
... which I reject because my desire is to find the TRUE or most true
possible
answer ... MVT isn't a particularly desirable scenario in some ways, I would
prefer that there is a benign God who will see me safely through life to a
blissful eternal afterlife .... but unfortunately the evidence persuades me
of MVT and not areligious belief, (sob, weep).

S> So you seem to agree with McGinn ... a very defeatist stance, and very
S> unextropian! You haven't managed to answer my point against this
S> that, and I repeat since you avoid answering here:

>It's only "defeatism" if you're giving up on the possible. Is Godel's
>Incompleteness Proof pessimistic about the possibility of designing a
>formally complete arithmetic? No. It shows that it's impossible.
>Every optimist should accept "defeatist" arguments of that kind.

You must ditch the possible if it contradicts the probable!!

Free-will is a much more fruitful stance than determinism, and does allow
that many aspects (genetics &c) are determined. Utility must decide here.

> S> Perhaps I should develop new and irresistible hypnotic applications
> S> from MVT and enforce belief in it ... would this satisfy you?
>
S> This example was deliberately crafted because it underlines the
S> pre-eminence of consciousness ... and goes against epiphenomenalism.
>
S> How do you answer this method of "proving" MVT?

>MVT' may well have some useful applications. You would not prove MVT,
>but only MVT'. You still would not show how holes affect
>consciousness, and would not have shown that consciousness can affect
>holes.

For the sake of argument, I am using a powerful new hypnosis developed
from MVT that enforces amnesia if necessary of your qualms about
consciousness (forget holes ... they are for illustrative purposes only to
defeat your physicalist/atomist claim, it is the functional abstraction that
explains
self-referential consciousness). So assume that my hypnotic suggestion
works,
just as a thought-experiment, and answer whether or not I have proven MVT to
your satisfaction ... after all, you now believe what I tell you, and I tell
you it is
proved!

S> I fully intend to pull the turf away from under your feet! "Philosophy"
S> (love of argument) cannot be roped off and protected in the way
S> you would like ... weasel words cannot stop the march of progress.

>Progress can march on. Why, if we never talked about consciousness
>ever again, we'd lose nothing at all. Intelligence is all that
>matters. Consciousness is a joke.

And sentience, awareness, mentation, mind ....? I agree with you to
a point, but if you think that there is only conscious behaviour, and
further
think that MVT solves it (animal intelligence or whatever you want to call
it) then why are you persisting this dispute?

>Why progress towards an understanding of a stupid word that
>philosophers made up? Why not grant that they made it up so as to
>keep you from leaving them out of a job?

But I want them out of a job. They just serve to confuse .....

S> So I don't accept your notion of MVT' ... because knowledge is seamless.

>No, claims can be analysed into parts. MVT is a compound claim about
>observable facts and unobservable facts. I accept MVT's observable
>claims and reject MVT's unobservable claims.

OK, I know that there is a weak and strong version of MVT, but if you
accept all the empirical premises of MVT is is difficult to reject the
conclusions or come up with different ones. Proof positive may be
forthcoming, but will it be in your lifetime, and what are you going to
believe in the meantime?

>You'd say: "No, the Fairy theory is stupid. I accept all the bits
>about observable stuff but reject the bit about fairies dancing on the
>particles."

>What would you think of me if I said: "I don't accept that notion
>... because knowledge is seamless!"

>That's what I think of your rejection of MVT' on those grounds.

I accept this except that I think the conclusions about unobservable
parts of MVT do fit the facts well and are not absurd like your fairy
example.

S> The mind-body problem ... only like can interact with like, therefore how
S> does the (observable) brain manifest (non-observable) mentation .... can
S> be dissolved by the MVT explanation in terms of a illusionary sense-organ
S> which is nevertheless continuous (even supervenient, in Jaegwon Kim's
S> mereological sense of this silly word, perhaps) with the brain matrix.

>No. You don't explain how consciousness can interact with holes.

The above paragraph talks about how only an illusionary sense-organ can
sense illusionary sense-data ... and mentions Kim's "whole-part fusion"
or merological supervenience notion ..... we have left holes behind with
discussion of atomism ...

>The illusionary body part isn't there *at all*.

It persists as a trace, archetypal (embryonic even) memory or engram
and as a phantom or "felt" part of the body matrix.

> You have "dealt with"
>the world of consciousness only if you argue that "like affects like:"

I do argue that, following Leibnitz.

>that the world of consciousness can interact with non-existent objects
>because consciousness doesn't exist in the first place.

We have an overloading of the word "existence" here. Let us stamp
out mere verbal triviality and stick to the phenomena where possible,
please.

> >It uses a different algorithm from the one Kasparov uses. But
> >Kasparov uses an algorithm too. (His is certainly better, but it's
> >hard to read off from the neurons, and ever trickier to program.)
>
Ss> The human player tends to use entirely different approaches, Casablanca
S> for example said that he only ever looked one move ahead, "but the best
S> move!"

>Yes, but his nerves did the rest, unconsciously. I rarely think very
>hard about breathing, but it's rather complex.

But admitting that your Deep Blue doesn't have nerves or unconscious
processes undermines your case that human and computer are the same!

S> Anyway, you haven't explained away the programmers needed for Deep Blue
...
S> Kasparov can modify whatever heuristics he uses as a result of feedback
from
S> outcomes, but bugs in Deep Blue needs programmers .... homunculi.

>I'm saying that it's a coincidence that programmers were needed to
>program Deep Blue while none were needed for Kasparov. Deep Blue was
>designed by people. Kasparov looks designed because he evolved. Deep
>Blue could have evolved, if the evolutionary pressures went the right
>way, but they didn't, so we got us instead.

Exactly "they didn't" ... so the two things are entirely different types,
and you
cannot use deep blue as even an analogy for animal intelligence, let alone
consciousness!

S> Will get round to dealing with your other points when I have more
S> time. But to date, nothing you have said convinces me much. Why do
S> you bother to plow a trough which you think is sterile and
S> pointless? Science can throw up new solutions and information, but
S> it seems your type of tautologous speculation can never get
S> anywhere.

>No, it won't. In fact, I think something quite radical about
>consciousness, that gets us quite far indeed, but we're talking about
>something else at the moment.

But you have said that consciousness is stupid, and doesn't exist
because there is only "conscious behaviour" .. so you cannot put
forward a "theory of consciousness" without contradiction.

>Yes, determinism vs. free will is a pointless debate. Except for when
>people start saying that computers can't be like us because one of us
>is determined and the other is not. That's when we have to have the
>argument.

I think that computers can't be like us only because even if they had
self-referential consciousness it would be DIFFERENT from ours,
because the self they would be aware of would be a (silicon) entity
and not a human or posthuman entity. I also happen to think that
self or self-organisation, independent decision-making, is not
a feasible concept in a world that is wholly determined. Even you
backtracked and allowed that in NLP psychotherapy an individual
could modify their metaprogramming (degree doesn't really matter,
you have breached your deterministic principle in that previous mailing).

Level Up
www.steve-nichols.com
Posthuman bodhisattva

- -Dan

      -unless you love someone-
    -nothing else makes any sense-
           e.e. cummings



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:32:30 MST