Dear Scott,
Your 7/9 1:27 +0000 answer to my question if it is really a problem if
Pirsig ignored consciousness doesn't really convince me. You wrote:
'I think we need it, but that may be that it was in trying to imagine how a
computer could be conscious, I realized that it logically could not, because
every event in a computer is separated spatially and/or
temporally from every [other] event, so there is no way that two or more or
a million events could be grasped as a whole. For the same reason, a brain,
considered as a spatio-temporal object could not be conscious either. From
this I realized that the basic problem was SOM, the belief that reality
consists of subjects being aware of objects ... My answer to the question of
consciousness is ... that consciousness must be presupposed, and everything
else explained as products of consciousness.'
If you cannot explain how something can be conscious, you explain it as a
product of consciousness...
It feels like a language game to me, as logic pushed to the absurd.
Couldn't we say that if 'consciousness' IS something worth worrying about
(if it exists), it must refer to an 'experience'? It is an experience that
we, humans, obviously have (occasionally). Before we start wondering if
others (animals, computers etc.) have it or can have it too, we must somehow
define it. As we, also obviously, don't share their experience (don't feel
intimately connected to an extent that makes their experience of
'consciousness' obvious to us), we must define 'consciousness' by observable
behavior resulting from it. What can others observe about us when we
experience 'consciousness' which they cannot observe when we do not?
You seem to use 'grasping separate events as a whole' as a criterion. That
reminds me of your/Nishida's 'logic of contradictory identity'. Must
something/someone experience/grasp events/things as being 'separate' and yet
'one' to be conscious? Interesting possibility, but how do we know (from its
behavior)?
We can describe behavior that might indicate 'consciousness'/'awareness',
e.g. fixing one's eyes on something. This can be observed in a predator
spotting a prey. (The predator may even be supposed to grasp simultaneously
its separateness from its prey and their potential/desirable one-ness ....
when the prey will satisfy its hunger.)
Even to an earthworm could be ascribed some level of 'consciousness' when it
retracts itself from our touch: It must somehow be 'aware' of us as a
potential danger.
'Self-consciousness' might be gauged (apart from self-description in some
understandable language) by behavior that is neither a set of comparable
reactions in comparable circumstances, nor completely rigid behavior
regardless of circumstances, but flexible enough in comparable circumstances
and stable enough in different circumstances to suppose some sort of
'reasons' or 'sense of identity' behind it.
Such 'behavioral definitions' of consciousness can be made more 'fool-proof'
by including more types of behavior in a definition and requiring several of
them. (This prevents mistakes due to single 'behaviors' included in such a
definition being simulated by supposedly unconscious machines or happening
to be found -as 'exception that proves the rule'- in a much lower life form
than were we are willing to recognize as having the defined level of
consciousness.)
Only when we have decided on a behavioral definition of consciousness can we
start to figure out how a computer could be created displaying that behavior
and therefore being -probably- conscious. The problem is, starting from our
own experience of 'consciousness', that we can be 'conscious' at a lot of
different 'levels'. We may still retract from touch when asleep. We can be
observed fixing our eyes on something/someone and even following its/his/her
movements while daydreaming and not remember having done so afterwards. Even
when we know what we are doing and why, we can still 'only' do so because we
identify and act in unison with a group (e.g. a group of hooligans), be very
disappointed with ourselves afterwards and say something like 'I didn't know
what I was doing' (while having been very vocal about what you did and why
at the time itself).
The question how 'consciousness' expresses itself in observable behavior is
comparable to the question how inorganic, biological, social and
intellectual patterns of values express themselves in observable behavior.
The question 'how to create a conscious computer' can be translated into
'how to create a computer through which patterns of values of a specific
Q-level can express themselves'. This has two advantages over using the
concept of 'consciousness'. Unlike conscious/(subconscious)/unconscious,
there are sharp dividing lines between Q-levels (even if some of the
contributors to this list draw them in the wrong places (-;). Secondly, even
if we can't always make out for sure if a single animal/computer/human
expresses patterns of values of a specific Q-level, we should at
least be able to tell if a lot of them together could form/create/perpetuate
the necessary type of patterns of values.
Could we create robots that express (together) or program software that
simulates biological patterns of values? I think that has already been done.
Could we do so for social patterns of values? I.e. could we create different
'societies' of robots/e-bots creating, passing on and thereby perpetuating
and developing different 'cultures' (cumulative sets of ways of doing
things) that interact with their (un)natural environment? Maybe, but I am
not aware of that having been tried.
Only after we have done that can we hope to do so for intellectual patterns
of values. Intellectual patterns of values need as a basis that the means of
communication between social robots/e-bots (necessary to pass on 'culture')
reaches a sufficient level of complexity AND 'slack' (there must be several
ways to communicate the same 'message', there must be need for
interpretation and there must be the possibility of misinterpretation and
'strong misreading' -in Rorty/Matt's words- that creates intellectual
'content').
Isn't 'self-consciousness' the level of consciousness you aim for when
trying to create a 'conscious computer'? It doesn't seem impossible to me to
simulate 'animal consciousness' in a computer, at least at the earthworm
level but probably also at the level of say a cat. But I know very little
about AI, so I may be wrong.
If I am right your problem is (in my words) how to create a computer (or
rather a group of computers) that can make the jump from (expressing only)
social patterns of values to (expressing also) intellectual patterns of
values. To me the 'consciousness' concept seems unnecessary and even
unhelpful (because of the many different ways in which it can be defined) to
solve that problem.
By the way: you didn't answer my question what you meant 5/9 22:59 +0000
with 'Pirsig ignoring the many/one problem'. It's (for me) obviously not
related to the Q-intellectual/Q-social distinction (as Maggie suggests 7/9
12:23 -0400). Both intellectual patterns of values and social patterns of
values need many people to unmistakably express themselves and to perpetuate
themselves. Both come into being because of the sense of 'truth'
respectively 'status' of (many) individuals.
Again I don't see a problem, nor a hole in the MoQ.
Interesting that you refer to Rudolf Steiner. I have a smattering of
antroposophy too, having an antroposophical family doctor, having taught
economics at the local Free School for a short period and eating a lot of
bio-dynamical food. You don't need to translate from German for me, if you
have original quotes and are not sure about your interpretation. I have more
than a smattering of German.
What does Rudolf Steiner mean with a 'pure intellectual pattern' (which is
in your interpretation needed to motivate our actions to make us free)?
The term 'driving forces' reminds me of my favorite way of interpreting
astrology (as character analysis): I use to interpret planets in a horoscope
as 'driving forces' that can co-operate more of less harmoniously, that can
express themselves more or less easily and that have a preference for
expressing themselves in specific aspects of our lives. I see a character,
that can be analyzed by studying a horoscope, indeed as a set of patterns
that limit our freedom, without determining our behavior. If we are
conscious of these patterns, we can use the implied strengths and limits the
implied weaknesses. If we are unconscious of them, if we don't use our
freedom, they determine our fate.
Thanks for your explanation of 'the ego is basically a social construct'. I
see your point, but it bears hardly any relation to my definitions of the
social and intellectual levels in the MoQ.
Ego-creation is for me a process which is part of the intellectual level. It
implies becoming aware of some of one's 'driving forces'. Fear of others
(the drive to protect your self), desire of others (the drive to connect)
and reflecting on others (a combination of the drive to communicate/exchange
and the drive to independence and critical appraisal of orthodoxy and
conventions) can be among them. Driving forces become aware because of
'mirroring' by others. In that sense ego is indeed a social construct.
I have Rene Girard's 'Des choses cachées depuis la fondation du monde.' in
Dutch translation on my bookshelf. His ideas about 'mimetic rivalry' (we
want things because others want them) considerably influenced my ideas. It
describes for me the origin of the intellectual level in a way that is
complementary to the way in which Pirsig describes it in chapter 30 of
'Lila' (with 'ritual' as core concept).
I agree with your objection against 'concentrating on immediate experience'.
I just borrowed the phrase from your description of John B.'s ideas. Your
'We can only let go, and so rediscover that the true "I" that the ego masks
is DQ.' exactly expresses how I see things.
With friendly greetings,
Wim
MOQ.ORG - http://www.moq.org
Mail Archive - http://alt.venus.co.uk/hypermail/moq_discuss/
MD Queries - horse@darkstar.uk.net
To unsubscribe from moq_discuss follow the instructions at:
http://www.moq.org/md/subscribe.html
This archive was generated by hypermail 2b30 : Fri Oct 25 2002 - 16:06:31 BST