Nick:
I find your distinction between metaphysical questions and factual questions
helpful because it clarifies the vague feeling I expressed about making "some
sort of error" when I said that consciousness is "having an inner subjective
life". I no longer feel it is an error but I should categorize it as a
metaphysical position rather than a scientific fact. (I prefer the term
``scientific fact`` to your term ``fact``.) It still seems like a good argument
("I know consciousness exists because I experience it") even though this cannot
be a scientific argument.
Eric, Steve, et al:
Thanks for your very interesting comments. I would like to add some further
comments about Dennett. I both enjoyed and was frustrated by his book
"Consciousness Explained". I recommend it highly but with the following
caveats:
(1) I wish the book were organized differently. I think it should have started
with
"The Challenge" (section 5 of chapter 2, p.39-42). I paraphrase this challenge
as:
First, Dennett says he wants to explain Consciousness in
scientific terms, without invoking anything beyond contemporary science. I take
this to mean that he wants to show that we can analyze and explain human
behavior entirely in scientific, materialistic terms without appealing to any
'mysterious' forces. (Therefore, to focus on the behavior rather than the
motives, of conscious people, Dennett starts by telling speculative stories
about the phenomenology of consciousness.)
Secondly, he doesn't want to be like behaviorists who "pretend they
don't have the experiences we know darn well they share they share with us. If
I [Dennett] wish to deny the existence of some controversial feature of
consciousness, the burden falls on me to that it is somehow illusory." (p.40 of
the book).
Thirdly he wants to do an honest job of explaining the empirical
evidence.
This challenge intrigued me. The first and second goals seem almost
contradictory. I wondered how he could possibly pull it off.
(2) As far as I remember, Dennett never summarizes how he met this challenge.
(I read this book over 15 years ago and I might have forgotten the summary. At
any rate, as I go over the book now, I can't find the kind of summary I would
like to see.)
So here is my summary of how Dennett did:
(a) After having read the book, I feel there is no theoretical barrier to
explaining all of the behavior of apparently conscious beings in purely
materialistic terms.
(b) My memory is that Dennett explains the feeling of being conscious in terms
of the strong AI hypothesis, which says that any system that carries out a
sufficiently complex task will automatically be conscious. I am not certain if
I believe this, but it or something like it seems necessary if we take the
first two goals seriously. Dennett apparently believes that the emergence of
consciousness depends only on the behavior exhibited. By contrast, Chalmers
argues that a conscious systems and a non-conscious system could exhibit the
same type of behavior. I don't see any reason to favor either position, but I
prefer Chalmers.
(3) On Dennett's style: This is what I find both frustrating and intriguing. He
seems to discuss various ideas without fully arranging them into an argument,
as I would tend to do. Dennett relies on this tendency of the reader to
complete the argument. So Dennett spends less time on argumentation and more on
telling stories. Sometimes it works, sometimes it doesn't. As mentioned above,
I came away with a strong feeling about the first part of the challenge. I also
had a strong feeling that our consciousness often fools us into thinking it is
in control when it isn't. I liked Dennett's presentation of the Pandemonium
model of language (based on work of Selfridge, Dawkins and others) and I feel
it explains a lot of things that would otherwise be murky. On the other hand, I
was dissatisfied with the chapter on "Qualia Disqualified". I even found myself
agreeing with his students (and others) that he hasn't really explained
consciousness --but I think he gave us a great tour of the issu
es. (If I had written the book, and arranged it more logically, the thread of
the arguments might have been clearer, but it would have been a much less
effective book.)
--John
________________________________________
From: Friam [friam-boun...@redfish.com] on behalf of Eric Smith
[desm...@santafe.edu]
Sent: Saturday, August 16, 2014 12:31 PM
To: The Friday Morning Applied Complexity Coffee Group
Subject: Re: [FRIAM] BBC News - Ant colony 'personalities' shaped
by environment
Hi Steve,
I am neither knowledgeable, nor do I have time to report even my own
experiences, without making a mess of things. But perhaps I can give some
titles of things people have pointed out to me.
There seem to be several schools of approach (meaning, groups of people who
criticize each other a lot). I't hard even to know how to break them down into
clusters, because there are several axes of variation.
There is a school who are mechanistic, and who think of themselves as
mechanistic.
At one end within that school, one has Dan Dennett. Much of what he says seems
to me like a lot of effort to beat the dead horse of mysticism, and I have no
patience for that, because I find it tedious and uninteresting. Beyond that,
it is not clear to me how much he has contributed in real ideas. One that
seems okay, if I understand it from informal conversations that have involved
him, is that it involves a kind of recursive self-reference of thought.
Meaning, that thought is a process for handling responses to events (or, in a
very broad use of the noun, "things"), and part of what consciousness does is
render the state of thought as a "thing" in its own right, having the same
symbolic kind of representation as the mind gives to other "things", so that
thought can then process a representation formed about its own state. This
seems like part of the common lore, expressed already in this thread, and not
novel. Dennett seems to want to associate this ability specifical
ly with language, and seems almost to want to treat it as an _application_ of
linguistic faculty. I don't think that is a well-motivated position, but I am
glad Dennett does it because it makes an important point. Language, in having
syntax, can manipulate words within the syntactic system, much as it uses words
to manipulate ideas within semantic systems. That is hard to understand in
language, and making us aware of the fact that it is hard, even though it has
been before our eyes for centuries, seems helpful in expressing part of what
makes assigning clear meaning to statements about consciousness hard.
On another extreme from Dennett but still materialist, we have Giuglio Tononi
and his "Phi" measure. Basically, Tononi adopts information theory as a
language, and within that language introduces a concrete notion of what it
means for an information system to be irreducible, in a way that I think is
analogous to the notion of irreducibility of representations of groups, in the
theory of representations. The details are different because information
theory is a different structure from algebra, but the basic notion of
something's not being splittable into factors is the same. I am now a couple
of years out of date wrt Tononi's publications, but I think it is fair to say
that Tononi asserts that having a very large irreducible component of
information is the _essence_ of consciousness, and that all the other things
like self-reference (which I would argue are also essential, even if
irreducibility is too) are merely other phenomena of mind but not the thing
that distinguishe
s conscious states. The Tononi development has the virtue of being an actual
idea that is formalized and thus unambiguously exchangeable among people. It
may also have a kernel of something important. Many people who work in
consciousness seem to think it does. For my taste, it is too non-embodied to
likely be a very comprehensive part of the right answer. I think both the
embodied dimensions of the things that contribute to conscious states, and some
kind of recursion, are primitives that are essential. Tononi has a large book
about this, and I think several shorter papers that are on the arXiv.
Somewhere in here is Christof Koch, who is also considered one of the important
contributors, but I don't know what his ideas are. I include him because if
you are asking who the thought leaders at the moment seem to be, my
understanding is that he is one of them.
There is also Max Tegmark, who has a recent paper "Consciousness as a state of
matter", available from the arxiv. This (which I have read) seems to me to be
a smart mathematician's discussion of a generally nice point, which adds
nothing of substance to where we are stuck. Tegmark is making an argument with
which I agree, that most-everything we see in nature that is robust is a "state
of matter", understood as modern physics uses the term. Hence, the distinctive
and characteristic nature of consciousness too. But the only thing about
consciousness in what Tegmark builds is what he gets from Tononi. The rest of
it is more about the theory of measurement in quantum mechanics, than it is
anything that distinguishes consciousness from other patterns of order to which
we have given names and phenomenologies.
Now, if I understand it at a distant second hand, Chalmers has a criticism of
all of these kinds of positions, notwithstanding their technical differences,
which is that he would claim they fail to recognize what he calls "the hard
problem". I do not know exactly how Chalmers uses language, and I cannot speak
for him, but to try to use my own language to express what I think he says, I
would say he asserts that these mere characterizations of mechanism are not
"accounting for" what we mean when we report "the experience of" this or that.
Here, the word "qualia" is often introduced, to refer to the antecedent of such
reports.
I think Dennett thinks of (and perhaps calls) Chalmers the worst sort of
Cartesian dualist, whereas Chalmers would say that Dennett is claiming that
consciousness "doesn't really exist", or something morally equivalent. I
believe both of them think of the axis on which they hold opposite ends as
different and bigger than any of the axes that separate the technical people
from one another. Chalmers seems (for good or ill) to attract people who do
want to be dualists or mystics (or mysterians), so without putting in a lot of
time with original material, it is hard to get a clear picture of him through
the people who claim to render him.
Ih the middle of all this, helping us sort it all out, is John Searl, who has a
short little book "The problem of consciousness". Searl is at his best when
using pellucid common language to explain why everyone else is being silly. He
is much less impressive when asked to introduce an actual new idea that moves
the discussion forward. However, in saying that, I do not mean to diminish the
value (or the enjoyment) of his criticisms. He has some language in there
about various kinds of dualists, which I find mystifying, because it all exists
within such self-referential
circles of language that I wouldn't know how to link it to anything in the rest
of the world. But, if you want to know about dualists, this is a good place to
find them categorized.
I find reporting on a lot of this like I think I would feel if sent to the
middle east to report on exactly why it is necessary for some factions to fight
other factions. There seems to be a long way between being humans, and so
exercising the individual and social behaviors that constitute bringing ourself
to share or coordinate various internal states that we refer to with names for
awareness or states of mind or whatever, and finding a language that, in
symbolic form, makes a faithful representation of what it is that distinctively
allows us to be what we are and do what we do. Each of these guys seems to
bring attention to the absence of such language in one or another way. What I
can't understand is why they think there is anything more than "a hard problem"
of inventing a valid language to faithfully reflect the structure of a natural
phenomenon, and their main difference is in how much each thinks he has
captured and the others have not. But I think they would argu
e there is more to their positions than that.
Of course, I have no expert knowledge, and haven't put that much time even into
reading their literatures as an outsider and tourist. So it is to be expected
that a lot of it will pass over me.
Several of these guys have either TED talks, or lectures that stream on the
web, which are shorter than reading their papers, but even more unsatisfying.
Oops. Too much text.
All best,
Eric
On Aug 16, 2014, at 11:04 AM, Steve Smith wrote:
> Gentlemen,
>
> I am also interested in both the nature of consciousness and the nature of
> knowledge regarding what appear to be entirely subjective phenonomena
> (arising from the fact of consciousness?).
>
> The last time I attended a Cognitive Neuroscience conference (6 years ago?) I
> was impressed with how far things had come with regard to correlating brain
> imaging and *reported* subjective experiences. I realize that sometimes
> more data and even higher quality data doesn't necessarily improve a model
> qualitatively, but I have been hoping that there would be some conceptual
> breakthroughs from this work.
>
> Unfortunately, as the popular media and the population in general (which is
> chicken, which is egg?) have taken a stronger interest in science (or has
> come to fetishize the artifacts of science?) there is a lot more "noise" to
> sort through to find signal. The number of articles or even entire issues
> of magazines and the number of books on the topic has risen dramatically in
> the past 10 years or so, but I rarely see what looks like new insight into
> the nature of consciousness.
>
> I'm hoping someone here with more direct experience or more patience with the
> literature (BTW, the "hard literature" on the topic is generally too opaque
> for me, so I'm lost in a middle-ground limbo between the popular accounts and
> the actual work-product of scientists) knows of new insights or new twists on
> the old models to share.
>
> Does anyone have a short list of recent publications which reframe the
> question in a new way?
>
> - Steve
>> Hi Nick,
>>
>> One of the problems in discussing consciousness is that it seems very hard
>> to break it down into simpler concepts. There are what might be called
>> "high-level" words such as "inner life", "awareness", "apprehension", which
>> suggest consciousness but only to someone who already ha a sense of what
>> consciousness is. Whereas low level words, which refer to things that can
>> be readily measured do not seem adequate to get at the real meaning of
>> consciousness. So we are left with metaphors. When I use words such as
>> "access" and "inner life" they suggest a container but they are not
>> necessarily used to denote an actual container but to describe a situation
>> which has some of the properties of a container.
>>
>> However, there does seem to be a real container that describes the
>> information I have access to. I get raw information from my body. This is
>> not to say that my consciousness is located in my body, but that what I know
>> about the outside world starts with how my body senses the outside world.
>> These senses are then processed or contemplated somehow and this results in
>> what I think I know about the world. There is no way that "I can see exactly
>> what you see" because what you see comes from your body and what I see comes
>> from my body. If we literally mean "see" then what you see is what enters
>> your eyes and what I see is what enters my eyes. You might tell me about
>> what you see, but that is not the same as seeing what you see because what
>> you have seen has been processed by you then reformulated in terms of
>> speech, which is then processed by me. Even if we witnessed the same event,
>> we would have slightly different viewpoints, and our eyes are different,
>> and, in any case, we w
ou!
>> ld start interpreting the incoming rays of light as soon as they started to
>> enter our respective eyes.
>>
>> You also gave examples in which I might infer what you saw. This seems to
>> presuppose I have a theory of what Nick is all about or some means of making
>> inferences. (I don't have a well-articulated theory of Nick, but I do arrive
>> at conclusions about what to make of you. I'm not certain how I do this, but
>> I am certain that I do it all the time, quite effortlessly and almost
>> automatically.) At any rate this drawing of inferences does not seem to be
>> seeing exactly what you see, but a way (not necessarily very accurate) of
>> getting a rough approximation of what you saw.
>>
>> --John
>>
>
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
