Significance (contd)
someone2
1. Definitions
If something in this paper is referred to as 'consciously
experiencing' it will mean that it is like something to be that thing.
This definition was first put forward by Thomas Nagel in his article
'What is it like to be a bat' [1], and has subsequently been used as
the definition of choice by numerous other philosophy professors.
According to this definition, if, when considering what it might be
like to be a certain object, your answer was that it wouldn't be like
anything, then you wouldn't be considering the object to be
consciously experiencing. Whereas if you thought it would be like
something, then you would be considering the object to be consciously
experiencing. To illustrate, if when considering what it would be like
to be a glass or a mobile phone, you thought that it wouldn't be like
anything, then you would be considering neither the glass or mobile
phone to be consciously experiencing. Whereas if you were considering
that it would be like something to be another human (perhaps similar
to what it was like to be you), then you would be considering them to
be consciously experiencing. A popular science fiction film called
'The Terminator' used the cinematic device of depicting a first person
perspective for a robot from the future to suggest what it would be
like to be the robot, thus suggesting that the robot was consciously
experiencing.
The assumption that 'something that consciously experiences can have
its behaviour described in the same terms of physics as something that
doesn't' will be referred to during the paper as the 'universality
assumption'
The term 'empirical evidence' refers in this paper to evidence based
upon what it is like to be a human.
2. Aim
What this paper aims to do, is highlight the contradictory position in
any materialist or physicalist theory which makes the universality
assumption and claims to be based on any empirical evidence.
3. The Thought Experiment Robot
Daniel Dennett in a paper 'What RoboMary Knows' [2] made the claim
that:
'If materialism is true, it should be possible (“in principle!”) to
build a material thing–call it a robot brain–that does what a brain
does, and hence instantiates the same theory of experience that we
do.'
Dennett futher points out; 'thinking in terms of robots is a useful
exercise, since it removes the excuse that we don't yet know enough
about brains to say just what is going on that might be relevant,
permitting a sort of woolly romanticism about the mysterious powers of
brains to cloud our judgement'.
So to avoid any sort of woolly romanticism about what we know, and
about the mysterious powers of any mechanism to cloud our judgement
let us avoid asserting that materialism is true, and that the robot
brain does what a brain does.
Let us consider the operation of the robot and the robot brain to be
explainable given our current understanding of physics, and that it
can pass the Turing Test as outlined by Alan Turing in the paper
'Computing, Machinery and Intelligence' [3], which would mean that if
one were to be corresponding via email with individuals in a group,
which was made up of humans and the robot, that you wouldn't be able
to distinguish which correspondent was the robot.
4. Some theories of the conscious status of the robot
Certainly there can be many hypotheses about whether the robot of our
thought experiment were consciously experiencing or not, or what the
conscious experience might be like. A couple will be constructed here
for illustration purposes only. Both theories of consciously
experiencing examined can be considered to be making the universality
assumption. It is this assumption of the theories which the paper aims
to highlight a problem with, as such the point isn't restricted to the
two theories put forward here, but can be applied to any theory which
makes this assumption.
Firstly let us consider a school of thought based broadly upon a
suggestion put forward by Place in the paper 'Is Consciousness a Brain
Process' {4]. Which is while we may be able to identify our conscious
experiences separately from our brain processes, for example in an out
of body experience, actually our conscious experiences are organic
brain processes. So what we identify as our conscious experiences is
really an identity of the processing of our organic brain. Since what
we identify as conscious experiences are identical to physical
activity of the organic brain, being one and the same, and the robot
hasn't an organic brain; this school of thought hypothesises that the
robot isn't consciously experiencing. Let us refer to this as Theory
A. While not dealt with directly here, it should be noted that another
branch of thought, could hypothesise that it would be like something
to be the non-organic brain, but that it would be different from being
an organic brain.
The second school of thought is based broadly upon a suggestion by
Putnam in the paper 'Philosophy and Our Mental Life' [5]. Which is
that conscious experiences cannot just be an identity of neural state
as suggested by Theory A, for if they were, then we would have to
consider alien creatures to not be consciously experiencing simply
because their brains are organised differently, perhaps being based on
different chemistry. Instead it suggests that part of the brain
performs a certain function, and can be identified as consciously
experiencing, and that any physical activity which performed the same
function, regardless of the chemistry or architecture used to perform
it, can be identified with the same conscious experiences. Let us
consider them to argue that the robot brain is performing the same
function as the human brain in order to produce equivalent behaviour.
As such their hypothesis is that the robot is consciously
experiencing. Let us refer to this as Theory B.
5. Scientific experimentation to distinguish between the theories
Two theories regarding a subject matter must differ in their
implications of behaviour for the subject matter for them to be
scientifically distinguishable.
Since theories A and B make the universality assumption, they assume
that something could have its behaviour described in the same terms of
physics regardless of whether it is consciously experiencing or not.
So the expected behaviour for the robot, is for it to follow those
laws of physics, for both the hypothesis that it is consciously
experiencing and for the hypothesis that it isn't. Thus competing
theories which make such an assumption, will never suggest a
difference in expected behaviour for the robot, simply because differ
with regards to whether the robot would be consciously experiencing or
not. So while how the robot would react would be open to scientific
investigation, what it would be like to be it would not.
That does not mean however that all theories which make such an
assumption could never be distinguished from one another. For they may
make other assumptions which could be investigated. Though if these
additional assumptions were shown to be correct, then there would
never be any behavioural difference to distinguish whether they, or
theory A or B were correct, given the universality assumption.
To illustrate, consider a third theory, Theory C, which is based upon
a suggestion made by Koch and Crick in their joint article 'Neural
basis of consciousness' [6]. The suggestion being that there would be
a physical difference between neurons that contribute towards us
consciously experiencing and those that don't. Going further, let us
suppose that the proponents of theory C suggest that there would be a
physical difference between neurons that contribute towards what it is
like to hear, and those which contribute towards what it is like to
see. They suggest that due to lack of such physical differentiation
within the robot, that if it did consciously experience, the
experience would likely consist of variations of a single property,
for example, fluctuations of the colour green.
Now certainly theory C could be distinguished from theories A and B if
it's primary hypothesis of a behavioural difference between neurons
that contribute towards our conscious experiences and those that
don't, were found to be false. Though if this primary hypothesis was
substantiated, there would still remain the issue of scientifically
distinguishing whether the hypothesis provided by theory A or B or C,
regarding the robot's conscious status, was correct.
6. Conclusion
If physicalists which make the universality assumption were asked to
to consider how a 'zombie' would be expected to behave if it were
physically identical to a human, and yet didn't consciously
experience, they might object. They might argue that it involves
conceiving of a conceptual framework or ontology where there are two
entities, firstly the physical human, and secondly its consciousness
which can be separated from former to produce the zombie. So to
stipulate in a thought experiment that this should be done, would be
to stipulate the adoption of such an ontology, and to beg the question
that implications for such an ontology has implications for the
physicalist ontology in which there is only the single entity, the
physical human.
There is an alternative approach, which is asking if there is any
difference in expected behaviour between the hypothesis that the
entity which is the physical human can correctly be identified as
consciously experiencing, and the null hypothesis that the entity
which is the physical human can be correctly identified as not
consciously experiencing. One hypothesis clearly won't correspond to
reality, and nor is it supposed to, since the two hypotheses are
mutually exclusive. In this approach there is no suggestion of an
ontology in which there there are two entities; firstly the physical
human, and secondly its consciousness which can be separated from
former. Both hypotheses are suggestions about how the entity should be
correctly identified. In the same way that if all that could be
measured of an object falling through a vacuum was its rate of
descent, how the object should be identified with regards to shape may
be hypothesised about. Likewise if the object were instead descending
through the earth's atmosphere. Another example would be if a robot
factory produced a range of robots to perform different functions, and
the colour of the plastic used internally in the robots varied
dependent on the function it was built for. If the various function
categories weren't known, and neither was it known the colour range of
plastics used, people could still nevertheless hypothesise as to
whether a robot built for a particular task could be identified as
containing yellow plastic. Expected behaviours can be suggested for
the various hypotheses of how the entity should be identified.
It might be questioned how considering what we know not to be reality,
such as the null-hypothesis that the entity which is the physical
human would be correctly identified as not consciously experiencing,
can help us in our understanding of reality. The thought experiment of
whether the robot is consciously experiencing helps illustrate the
answer.
The thought experiment of the robot illustrates that if the
universality assumption were true, there would be no scientific way of
distinguishing between theories A and B. This isn't an issue of
sensitivity of behavioural measurement, or an inability to construct a
scenario in which the difference between the hypotheses would become
significant to the expected behaviour. It is that the significant
difference between the theories has no significance to the expected
behaviour. To dispute one would only have to state conceptually how to
discern whether the robot was consciously experiencing or not, and
thus have distinguished between whether theory A and B was correct, to
illustrate that the significant difference between the theories is
significant to an expectation of behaviour for the robot. If whether
the robot was consciously experiencing or not isn't significant to how
it is expected to behave, then it follows that it wouldn't be
significant to how humans would be expected to behave either. So no
theory which makes the universality assumption can be considered to be
based on empirical evidence, since that would contradict the
implications of the universality assumption, which is that what it was
like to be something, which is the basis of empirical evidence, would
not be significant to the behaviour of the person formulating the
theory.
So returning to the question of how can the consideration of what we
know not to be reality, such as the null-hypothesis that the entity
which is the physical human would be correctly identified as not
consciously experiencing, help us understand reality, it does so in
the same way as the robot thought experiment, in that it helps us be
clear about the expected behavioural significance of something
consciously experiencing if the universality assumption were true, and
the implications of it.
7. References
[1] Nagel, T. (1974), "What Is it Like to Be a Bat?", Philosophical
Review, pp. 435-50
[2] Dennett, D.C. (2006), "What RoboMary Knows", Phenomenal Concepts
and Phenomenal Knowledge: New Essays on Consciousness and Physicalism
by Torin Alter and Sven Walter, Oxford Universtity Press ISBN-10:
0195377044 ISBN-13: 9780195377040,
[3] Turing, A.M. (1950). "Computing machinery and intelligence". Mind,
59, 433-460.
[4] Place, U.T (1956:Feb), "Is consciousness a brain process", British
Journal of Psychology, 47p44
[5] Putnam, H (1975 ), "Philosophy and Our Mental Life" , Mind,
Language and Reality. Philosophical Papers, vol. 2, Cabridge
University Press, ISBN-10: 0521295513 ISBN-13: 9780521295512
[6] C. Koch, and F. Crick, (2001), "Neural basis of consciousness",
International Encyclopedia of the Social & Behavioral Sciences.
Elsevier Science Ltd., pp. 2600-2604, ISBN-10: 0080430767 ISBN-13:
9780080430768
someone2
Since there would be no way of distinguishing scientifically between
theories A and B, it could even be argued that if the universality
assumption were true, and theory A was true, then perhaps all your
brain cells could be replaced by electronic chips, such that when
total replacement had occured, it wouldn't be like anything to be you,
the human you experience being would no longer consciously experience,
and yet it would behave the same. Which is just a thought experiment
to illustrate that if the universality assumption were true, there can
be no evolutionary advantage to consciously experiencing. Since if the
universality assumption was true, it wouldn't be significant to the
way things evolved whether theory A was correct or theory B. Nothing
could be considered an evolutionary advantage only if a particular one
of the theories was correct.
Kermit
<snip>
>. Since if the
> universality assumption was true, it wouldn't be significant to the
> way things evolved whether theory A was correct or theory B. Nothing
> could be considered an evolutionary advantage only if a particular one
> of the theories was correct.
If they are scientific, and indistinguishable, they are the same,
however differently worded they may be.
Kermit
Garamond Lethe
> Significance of what it is like to be human
>
Just a few writing notes. I will not be addressing your argument.
>
> 1. Definitions
>
> If something in this paper is referred to as 'consciously
> experiencing' it will mean that it is like something to be that thing.
> This definition was first put forward by Thomas Nagel in his article
> 'What is it like to be a bat' [1], and has subsequently been used as
> the definition of choice by numerous other philosophy professors.
I define 'consciously experiencing', following Nagel[1], as "that it is
like something to be that thing".
> According to this definition, if, when considering what it might be
> like to be a certain object, your answer was that it wouldn't be like
> anything, then you wouldn't be considering the object to be
> consciously experiencing. Whereas if you thought it would be like
> something, then you would be considering the object to be consciously
> experiencing.
[The second person (i.e. "you") is rarely used in formal writing.
I think the above can be removed.]
> To illustrate, if when considering what it would be like
> to be a glass or a mobile phone, you thought that it wouldn't be like
> anything, then you would be considering neither the glass or mobile
> phone to be consciously experiencing.
> Whereas if you were considering
> that it would be like something to be another human (perhaps similar
> to what it was like to be you), then you would be considering them to
> be consciously experiencing.
For example, one does not consider "that it is like something to be"
a drinking glass or a mobile phone, but we have no difficulty
apprehendng "that it is like something to be" a human.
> A popular science fiction film called
> 'The Terminator' used the cinematic device of depicting a first person
> perspective for a robot from the future to suggest what it would be
> like to be the robot, thus suggesting that the robot was consciously
> experiencing.
[That's fine.]
> The assumption that 'something that consciously experiences can have
> its behaviour described in the same terms of physics as something that
> doesn't' will be referred to during the paper as the 'universality
> assumption'
>
[Here things begin to get foggy.
1. We don't use physics to describe "that it is like something to be"
or "conscious experiences".
2. "the same" is left undefined. The same what? Coefficients?
Theories?
3. How is this any different than physicalism?]
I use the accepted definition of "physicalism"[?] as the position
that the behavior of both entities that experience consciously
and those that do not can be reduced to physical causes.
> The term 'empirical evidence' refers in this paper to evidence based
> upon what it is like to be a human.
[Thus consciously experiencing is empirical evidence? That's probably
not what you meant.]
<snip>
someone2
The theories aren't the same though. In one it is posited that it
wouldn't be like anything to be the robot, in the other it is posited
that it would be like something to be the robot. So they suggest a
difference in reality, it isn't a case of they were suggesting the
same thing but simply worded it differently. Certainly one wouldn't be
correct, but there would be no scientific experiment to distinguish
which one, if the universality assumption was correct.
someone2
Thanks for the advise on the re-wording, I'll look at changing the
sections.
Regarding where you think things are getting foggy.
1). I agree, we don't use physics to describe what it would be like to
be something though we do use it to describe the behaviour of physical
causes that physicalism suggests it can be reduced to.
2). Yes I will reword this to laws/terms of physics, rather than just
terms of physics to save confusion.
3). The difference is that there are theories which would be counted
as physicalist theories, given the definition you supplied, which
don't make the universality assumption. For example, property duality
is a physicalist theory, two types of properties, but only one
substance, thus both types of property are physical causes. Such that
property dualism would be a physicalist theory which doesn't make the
universality assumption though.
Garamond Lethe
As to #3:
I commend you for being aware of property duality. I don't recall
having heard about it before. When I checked google, I found exactly
two entries for it (ANDing it with "philosophy of mind"). You should
absolutely include this nuggest of information somewhere in your paper.
But the introduction isn't the place for it.
In the introduction, you're letting the reader get the 10,000-foot
overview of what you're going to be talking about. It's ok to be
imprecise and incomplete --- you'll be making up for that later in
the paper.
Perhaps the following would work better if you're concerned that
the intro might be misunderstood:
I use the accepted definition of "monistic physicalism"....
someone2
Currently I don't mention property dualism in the paper so as not to
confuse people. Even "monistic physicalism" can be confusing as with
property dualism there is only the physical, there aren't two
substances as in the dualism put forward by Descarte, so it is
monistic in the sense that there exists only the physical. Because of
this, I might just not mention it, as such theories don't make the
universality assumption, and therefore are outside of the scope of
this article. My next paper will deal with physicalist arguments which
reject the universality assumption such as property dualism, and show
why they are implausible. Though it is maybe slightly more complicated
than this paper, and I need to work on it being as easy to understand
as possible.
If you are interested in property dualism btw, David Chalmers mentions
it in section 9 of this paper:
http://www.klab.caltech.edu/~koch/Elsevier-NCC.html
Thanks for this feedback on the paper btw.
someone2
Sorry I gave the wrong link, it should have been:
http://consc.net/papers/nature.html
You should read section 9 at least, as the links that come up on the
search you did, don't cover the issue in as much depth as Chalmers.
Jim Lovejoy
bd3f-e5f...@18g2000yqa.googlegroups.com:
That's what he said dummy. If there's no scientific experement to
distinguish them, they're the same.
If that's too compley for you to understand, tell me, and I'll type slower
next time.
someone2
> someone2 <glenn.spig...@btinternet.com> wrote in news:12ffcee5-e069-4d8b-
> bd3f-e5f7589bb...@18g2000yqa.googlegroups.com:
I understood what he was saying, it was just that he was wrong, as are
you. Tou didn't seem to understand my response. The theories aren't
the same.
Theory A is a version of the Identity Theory of Mind:
http://plato.stanford.edu/entries/mind-identity/
Whereas Theory B is a version of functionalism:
http://plato.stanford.edu/entries/functionalism/
There is a difference in reality posited by each theory.
If aliens had two theories, one that you consciously experienced, and
could feel pain, the other that you weren't consciously experiencing,
and thus didn't feel pain any more than a mobile phone did, you can
presumably understand the difference between what they are
suggesting.
An analogy might be if only the rate of descent of an object falling
through a vacuum could be measured. One group might posit the object
was a cube, another that it was a sphere. Though there would be no
expected difference in what could in this scenario be measured, i.e.
the rate of descent. That doesn't mean that there was no difference in
what the groups were suggesting. Remember this last one was just an
analogy, don't get all confused about what it is trying to illustrate,
and then bother pointing out that if it wasn't just the rate of
descent that could be measured, that the analogy doesn't work.
Ymir
<cf8e7a37-c639-4592...@e11g2000yqo.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
The problem here is that if the only behaviour which the theories
predict is how the object falls, then neither theory is a theory of
shape -- if one theorist claims the object is a cube and another claims
it is a sphere, but their theories only address how object falls, then
the claims regarding shape aren't claims made by the theory -- they are
simply assertions tacked on by the theorist, and the theories themselves
must be viewed as equivalent.
Getting back to consciousness, it is important to understand that before
one develops a theory of *anything*, there must exist something which
one is actually trying to account for. Theories of consciousness try to
address the fact that we tend to divide the universe into two distinct
(albeit fuzzy) classes of entities -- those which are conscious and
those which are not. We do this because we recognise that certain
objects (e.g. people) behave very differently from other classes of
objects (e.g. rocks).
Anyone who tries to develop a theory of consciousness is going to have
to begin by clarifying exactly what sorts of behaviours they consider to
be indicative of consciousness.
In your posts, you talk about two different (hypothetical) physicalist
theories which make identical predictions regarding the behaviour of a
given entity, but which differ with respect to whether they attribute
consciousness to it. The only coherent interpretation of this is that we
are dealing with two different theories which operationalise
consciousness in different ways -- that is which differ with respect to
what kinds of behaviour they take as being indicative of consiousness.
Now clearly it is possible that we might choose to operationalise
consciousness in a way which turns out not to accurately reflect the
intuitive notion we're trying to get at. One of the above theories might
be taking certain classes of behaviours as providing evidence for the
existence of mental states when in fact no such states actually exist.
What I don't understand, however, is why you think that this pitfall is
a problem only for *physicalist* theories. If you reject what you refer
to as the 'universality assumption', the exact same potential problem
exists. There's no reason to think a dualist would be any less prone to
misattributing consciousness to a non conscious entity or failing to
recognise consciousness in a conscious entity.
Andr�
someone2
> In article
> <cf8e7a37-c639-4592-8f55-b1b7f4efd...@e11g2000yqo.googlegroups.com>,
--------------
The problem here is that if the only behaviour which the theories
predict is how the object falls, then neither theory is a theory of
shape -- if one theorist claims the object is a cube and another
claims it is a sphere, but their theories only address how object
falls, then the claims regarding shape aren't claims made by the
theory -- they are simply assertions tacked on by the theorist, and
the theories themselves must be viewed as equivalent.
--------------
The tacked on assertions are different though. With the consciously
experiencing example, both theories are attempts at explaining
conscious experiences, so while the theories could be thought of as
only assertions, what they are asserting is different.
The problem the paper discusses disappears if the universality
assumption is done away with, and there are physicalist theories that
do this, don't be mistaken and think that all physicalist theories
make the universality assumption. If the universality assumption is
done away with, then with regards to something they posit to be
consciously experiencing, they would be expecting a difference in
behaviour between their hypothesis that it was consciously
experiencing, and the null hypothesis that it wasn't, and thus whether
something is consciously experiencing or not can be said to be
significant to how it is expected to behave. With the universality
assumption though, it couldn't.
Ymir
<27480388-d843-4c93...@24g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
In the case of the falling objects example, they are quite clearly just
tacked on assertions. The fact that they are different has no bearing on
the adequacy of the two theories since these assertions do not follow
from those theories.
It seems to me that you are positing various theories which you are
claiming to be theories about conscousness, but then treating their
actual claims regarding the consciousness of the robot as if they are
merely tacked on assertions. You talk about how the theories accurately
predict the behaviour of the robot, but you overlook the fact that for
any given theory its behavioural predictions would be intimately
intertwined with its predictions regarding the conscious state of the
entity in question.
> The problem the paper discusses disappears if the universality
> assumption is done away with, and there are physicalist theories that
> do this, don't be mistaken and think that all physicalist theories
> make the universality assumption. If the universality assumption is
> done away with, then with regards to something they posit to be
> consciously experiencing, they would be expecting a difference in
> behaviour between their hypothesis that it was consciously
> experiencing, and the null hypothesis that it wasn't, and thus whether
> something is consciously experiencing or not can be said to be
> significant to how it is expected to behave. With the universality
> assumption though, it couldn't.
>
Would you mind providing a clear example of how a physicalist theory
which doesn't adopt the universality assumption avoids the problem which
you claim exists for those which do. The example which you give in your
earlier post is property dualism, but you don't explain how this theory
avoids the issue.
So, let's consider your hypothetical robot -- how would a property
dualist expect the behaviour of the robot to differ depending on whether
it is conscious or not, and how exactly would these differences follow
from property dualism?
Also, if property dualism�s predictions follow from the general idea
that there exist mental properties which are not purely reducible to
physical properties, and that these mental properties play a causal role
in determining the behaviour of the robot, please be sure to explain how
the presence or absence of these mental properties is established.
Andr�
someone2
> In article
------------
It seems to me that you are positing various theories which you are
claiming to be theories about conscousness, but then treating their
actual claims regarding the consciousness of the robot as if they are
merely tacked on assertions. You talk about how the theories
accurately predict the behaviour of the robot, but you overlook the
fact that for any given theory its behavioural predictions would be
intimately intertwined with its predictions regarding the conscious
state of the entity in question.
------------
I'm not treating the claims as tacked on assertions. The claims are
theories attempting to explain why it is like it is to be human. The
theories simply make the universality assumption, so any prediction
they make will simply be that the behaviour would be the result of the
thing following the same laws of physics that something that doesn't
consciously experience does. As I have mentioned earlier:
Theory A is a version of the Identity Theory of Mind:
http://plato.stanford.edu/entries/mind-identity/
Whereas Theory B is a version of functionalism:
http://plato.stanford.edu/entries/functionalism/
Perhaps after reading, you will understand that while they are
different theories, they never make different predictions of
behaviour, nor is there ever any expected difference in behaviour
between a hypothesis that something was consciously experiencing, and
the null hypothesis that it wasn't.
--------------
Would you mind providing a clear example of how a physicalist theory
which doesn't adopt the universality assumption avoids the problem
which you claim exists for those which do. The example which you give
in your earlier post is property dualism, but you don't explain how
this theory avoids the issue.
So, let's consider your hypothetical robot -- how would a property
dualist expect the behaviour of the robot to differ depending on
whether it is conscious or not, and how exactly would these
differences follow from property dualism?
--------------
If it was consciously experiencing, it's behaviour would differ from
the expected behaviour for if it was assumed that it wasn't. They
might not know exactly the behaviour they'd expect if it was, but they
know the behaviour they'd expect if it wasn't. In other words they
would expect to find interactions that you don't find in things that
aren't consciously experiencing. The way to test would be if the
behaviour could be predicted without such interactions. If it could
then according to property dualism it wouldn't be consciously
experiencing.
[This is ignoring any arguments that the interactions might just
always happened to have caused the same behavioural effect as would
have been expected if they were absent, for as far as I know, property
dualists don't make this claim]
Ymir
<c8487fff-b2a4-400f...@k19g2000yqn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I understand what functionalism and mind-brain identity are. It's your
claim that no theory which assumes universality can make different
predictions for conscious/non-conscious things which I don't follow.
The problem, I believe, is that you are tacitly assuming that
consciousness is some mystical, non-physical thing which is exactly the
sort of position which physicalism rejects. For physicalists, conscious
experience *is* a type of behaviour, albeit one that normally can only
be observed indirectly.
Let me digress from consciousness momentarily and consider only entities
which I think we would both agree are not conscious - a rock and a
river. These things both exhibit very different behaviours, yet those
behaviours follow from the exact same laws of physics. The differences
in their behaviours follow from differences in their compositions. It
wouldn't make sense to say that a theory which assumes the same laws are
in effect for both would predict exactly the same behaviours regardless
of whether we claimed the rock or river were flowing or not.
But this is exactly what you seem to be suggesting with respect to
consciousness. Physicalists argue that nonconscious entities exhibit
different behaviours from conscious one due to the fact that they have
different compositions. To claim that a robot is conscious is to claim
that it exhibits certain behaviours, behaviours which a nonconscious
robot would not exhibit. These differences would stem from differences
in composition, though both would follow the exact same physical laws.
As far as I can see, saying that the physicalist would make the same
behavioural predictions regarding the robot whether it was conscious or
not is no different from claiming that they would make the same
predictions regarding the behaviour of a body of water regardless of
whether it was flowing or not.
Given your claims, it is clear that you do not see these two cases as
analogous, but it is entirely unclear to me *why* you don't see them as
analogous. If you could explain the relevant difference to me, it might
help me to understand the basis of your claim.
> --------------
> Would you mind providing a clear example of how a physicalist theory
> which doesn't adopt the universality assumption avoids the problem
> which you claim exists for those which do. The example which you give
> in your earlier post is property dualism, but you don't explain how
> this theory avoids the issue.
>
> So, let's consider your hypothetical robot -- how would a property
> dualist expect the behaviour of the robot to differ depending on
> whether it is conscious or not, and how exactly would these
> differences follow from property dualism?
> --------------
>
> If it was consciously experiencing, it's behaviour would differ from
> the expected behaviour for if it was assumed that it wasn't. They
> might not know exactly the behaviour they'd expect if it was, but they
> know the behaviour they'd expect if it wasn't.
What specific behaviour would they expect if it wasn't?
someone2
> In article
You show where you aren't understanding when you say:
---------
To claim that a robot is conscious is to claim that it exhibits
certain behaviours, behaviours which a nonconscious robot would not
exhibit. To claim that a robot is conscious is to claim
that it exhibits certain behaviours, behaviours which a nonconscious
robot would not exhibit
---------
The article has defined consciously experiencing, and I supposed you
were responding to the article. If you had of been, then you would
have understood, that to claim the robot is conscious, is to claim it
would be like something to be the robot. To claim the robot isn't
conscious, is to claim that it wouldn't be like anything to be the
robot. Theories A and B make different claims about whether the robot
is conscious or not, but they both expect it to behave the same given
the build.
You on the other hand had, without mentioning it, decided to use a
different definition from the article, where the robot would be
defined as 'consciously experiencing' whether it was like something to
be it or not, as long as it performed certain behaviour. Which goes
some way to explaining why you continued responding as though you
understood the basic point that was made. Though I'm still not clear
why you were unable to understand that theories A and B would never
expect a difference in behaviour, unless perhaps you're claiming not
to have read the article.
> > --------------
> > Would you mind providing a clear example of how a physicalist theory
> > which doesn't adopt the universality assumption avoids the problem
> > which you claim exists for those which do. The example which you give
> > in your earlier post is property dualism, but you don't explain how
> > this theory avoids the issue.
>
> > So, let's consider your hypothetical robot -- how would a property
> > dualist expect the behaviour of the robot to differ depending on
> > whether it is conscious or not, and how exactly would these
> > differences follow from property dualism?
> > --------------
>
> > If it was consciously experiencing, it's behaviour would differ from
> > the expected behaviour for if it was assumed that it wasn't. They
> > might not know exactly the behaviour they'd expect if it was, but they
> > know the behaviour they'd expect if it wasn't.
>
> What specific behaviour would they expect if it wasn't?
They'd expect it to follow exactly the same behavioural laws of
physics that other non-conscious things are seen to follow
Ymir
<417d31bc-946e-47de...@r2g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> On 28 July, 21:14, Ymir <inva...@invalid.invalid> wrote:
> > In article
> > <c8487fff-b2a4-400f-a5e5-02ee7a4a5...@k19g2000yqn.googlegroups.com>,
<snip>
I think that you are using a much narrower definition of behaviour than
I am. To claim that it is like something to be something implies having
a certain class of experiences, and having an experience is a behaviour
-- its a behaviour of the brain which isn't necessarily outwardly
manifested, but it is still a behaviour which can be investigated using
tools such as BOLD-fMRI etc.
By way of analogy, consider the following computer program:
void main(void) {
register int i = 0;
for (;;)
i += 1;
}
(if your're not familiar with C, all this program does is sets up a
variable with the value of zero and then increments that variable until
the universe ends or someone shuts off the computer).
If I compile this program and run it, the computer will not appear to be
doing anything since the program doesn't generate any output or respond
to inputs. However, running this program still constitutes a behaviour
on the part of the processor -- just one which might be difficult to
observe without appropriate tools.
If you want to exclude such internal activities from your definition of
'behaviour', that's fine, but when interpreting my argument please bear
in mind that I am using a broader definition.
If you want to view consciousness as some more ephemeral, Jacksonian
sort of thing, it still needs to be operationalised in terms of
behaviours which can be observed. That's true for *any* theory, not just
ones which adopt the universalist assumption.
Consciousness isn't special in this regards. The same requirement exists
for even the most basic physical properties. For example, we ascribe to
particles the abstract property of 'mass', but mass is not something
which we can actually observe. What we observe is how the particle
behaves -- how it responds to different types of forces, etc. and we
infer that a particle has mass from these observations. If a given
particle responds to various forces in a way which is consistent with it
having a particular mass, it would seem somewhat silly to argue that we
don't really know this simply because we have only looked at behaviour
rather than at mass more directly. But this seems to be what you are
doing with respect to consciousness.
> > > --------------
> > > Would you mind providing a clear example of how a physicalist theory
> > > which doesn't adopt the universality assumption avoids the problem
> > > which you claim exists for those which do. The example which you give
> > > in your earlier post is property dualism, but you don't explain how
> > > this theory avoids the issue.
> >
> > > So, let's consider your hypothetical robot -- how would a property
> > > dualist expect the behaviour of the robot to differ depending on
> > > whether it is conscious or not, and how exactly would these
> > > differences follow from property dualism?
> > > --------------
> >
> > > If it was consciously experiencing, it's behaviour would differ from
> > > the expected behaviour for if it was assumed that it wasn't. They
> > > might not know exactly the behaviour they'd expect if it was, but they
> > > know the behaviour they'd expect if it wasn't.
> >
> > What specific behaviour would they expect if it wasn't?
>
> They'd expect it to follow exactly the same behavioural laws of
> physics that other non-conscious things are seen to follow
So given this, exactly what would lead you to accept the fact that other
people are conscious? What behaviours do people exhibit which don't
follow the same physical laws as non-conscious things?
Also, I'd like to note that the above appears (at least to me) to alter
some of the claims made in your paper and in previous discussions.
You give the example of two researchers who *accurately* predict the
behaviour of a given robot but who differ with respect to whether they
claim it is conscious or not. You then go on to suggest that
universalist theories would be incapable of making different predictions
regarding the behaviour of conscious and non-conscious entities.
However, from the above it seems that you are really trying to claim
that universalists would be unable to make accurate predictions
regarding the behaviour of conscious entities *at all*. Is that correct?
If so, your argument would be greatly bolstered by giving a specific
example of some behaviour which a conscious entity exhibits which (e.g.)
property dualists are able to accurately predict, but which
universalists cannot.
Andr�
someone2
> In article
----------
I think that you are using a much narrower definition of behaviour
than I am. To claim that it is like something to be something implies
having a certain class of experiences, and having an experience is a
behaviour -- its a behaviour of the brain which isn't necessarily
outwardly manifested, but it is still a behaviour which can be
investigated using tools such as BOLD-fMRI etc.
----------
Well theory A would be saying that the conscious experiences are
behaviours of the organic brain. As such the robot doesn't have the
behaviours. Theory B is saying that it isn't the specific behaviours
of the organic brain, but more general, such that if the robot brain
performed a similar function it would consciously experience. Though
theory A will never predict a difference of behaviour from theory B,
thus no scientific way to distinguish between them.
Regarding the property dualisty, and why they'd expect other humans to
be consciously experiencing is because they are, and what reason have
they to believe they are a special case? And yes, the property
dualists would be saying that the people that made the universality
assumption would be unable to correctly predict the behaviour of
something that is consciously experiencing, at least for as long as
they held their assumption. That the universality assumption was
wrong, and that whether something is consciously experiencing or not,
is significant to how you'd expect it to behave, in the sense of their
being a difference between the expected behaviour for the hypothesis
that it was consciously experiencing, and the null hypothesis that it
wasn't.
Kermit
In those matters which are not testable, they are not scientific. I
have all sorts of models which explain the universe, but which are not
testable.
That's why your tacked-on descriptions of consciousness or not are
irrelevant. If they can't be tested, what do they even mean?
>
> Theory A is a version of the Identity Theory of Mind:http://plato.stanford.edu/entries/mind-identity/
>
> Whereas Theory B is a version of functionalism:http://plato.stanford.edu/entries/functionalism/
>
> There is a difference in reality posited by each theory.
What difference? What would be perceived, counted, weighed, or
otherwise measured?
>
> If aliens had two theories, one that you consciously experienced, and
> could feel pain, the other that you weren't consciously experiencing,
> and thus didn't feel pain any more than a mobile phone did, you can
> presumably understand the difference between what they are
> suggesting.
>
Yes. I *act* differently if I am in pain or not. Descartes said that a
dog which is kicked doesn't feel pain, because it has no soul. It only
*acts like it's in pain because its body is constructed to react that
way for mechanical reasons. How on Earth would we test this?
> An analogy might be if only the rate of descent of an object falling
> through a vacuum could be measured. One group might posit the object
> was a cube, another that it was a sphere. Though there would be no
> expected difference in what could in this scenario be measured, i.e.
> the rate of descent. That doesn't mean that there was no difference in
> what the groups were suggesting.
It's all empty fluff if we can't determine the shape of the objects.
And if both groups of scientists agree that the shape wouldn't affect
the rate of fall, they wouldn't bring it up; it wouldn't be part of
their theories.
> Remember this last one was just an
> analogy, don't get all confused about what it is trying to illustrate,
> and then bother pointing out that if it wasn't just the rate of
> descent that could be measured, that the analogy doesn't work.
It's a fine analogy. The shape would not be part of the theory. But
normally the shape *could be determined. How would you determine if a
robot was *really conscious or not? What observations would be
different if theory A were wrong? Theory B?
And arguing about imaginary, currently unbuildable robots tell us
nothing about human minds and the possible existence of undetectable
souls, which is your real goal.
Kermit
Kermit
But physicalists would not expect this.
You keep trying to establish your argument by claiming that
physicalists would predict the conclusion you are trying to prove.
They would not.
You have been trying to do this for many months (years?), and it won't
work. Your argument is circular, and you claim a position from a
philosophical stance that wouldn't be held. You have also indicated
that you think this will shed light on the nature of the human mind,
and the existence of souls. It will not, I think.
>
> You on the other hand had, without mentioning it, decided to use a
> different definition from the article, where the robot would be
> defined as 'consciously experiencing' whether it was like something to
> be it or not, as long as it performed certain behaviour.
Yes. That is the physicalist stance.
> Which goes
> some way to explaining why you continued responding as though you
> understood the basic point that was made. Though I'm still not clear
> why you were unable to understand that theories A and B would never
> expect a difference in behaviour, unless perhaps you're claiming not
> to have read the article.
If they do not predict any discernable difference, they do not differ
scientifically.
If they do not describe differences in behavior based on consciousness
of not, they are not physicalists.
>
> > > --------------
> > > Would you mind providing a clear example of how a physicalist theory
> > > which doesn't adopt the universality assumption avoids the problem
> > > which you claim exists for those which do. The example which you give
> > > in your earlier post is property dualism, but you don't explain how
> > > this theory avoids the issue.
>
> > > So, let's consider your hypothetical robot -- how would a property
> > > dualist expect the behaviour of the robot to differ depending on
> > > whether it is conscious or not, and how exactly would these
> > > differences follow from property dualism?
> > > --------------
>
> > > If it was consciously experiencing, it's behaviour would differ from
> > > the expected behaviour for if it was assumed that it wasn't. They
> > > might not know exactly the behaviour they'd expect if it was, but they
> > > know the behaviour they'd expect if it wasn't.
>
> > What specific behaviour would they expect if it wasn't?
>
> They'd expect it to follow exactly the same behavioural laws of
> physics that other non-conscious things are seen to follow> > In other words they
> > > would expect to find interactions that you don't find in things that
> > > aren't consciously experiencing. The way to test would be if the
> > > behaviour could be predicted without such interactions.
Why on Earth would you expect the same behavior without the same
interactions, whether the object is conscious or not?
All objects and processes follow all the laws of physics at all times.
Right now, always. People, airplanes, dogs, comets, computers, stars,
tornadoes, rocks, and rock bands.
Do you have any evidence to the contrary?
> > > If it could
> > > then according to property dualism it wouldn't be consciously
> > > experiencing.
Why would physicalism be different in this respect?
Kermit
Ymir
<74c3d63e-dace-4652...@o32g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
If theory A and B are adopting the same operational definition of
consciousness then they *must* make different predictions regarding its
behaviour. Those who claim that consciousness can only be realised by
organic brains (and I'm not clear that there are such people -- identity
theories of mind don't normally go quite this far) are claiming that
there are behaviours which brains exhibit which cannot be reproduced in
the robot, therefore they would not predict the same behaviours from the
robot as anyone who argues that it is in fact conscious.
The only way that theories A and B could make identical predictions
regarding the behaviour of the robot but differ with respect to whether
they assign consciousness to it would be if they were assuming different
behavioural criteria for identifying consciousness. If this is the case,
though, then they are presumably also making use of different
definitions of consciousness. There would be no contradiction in saying
that something is conscious under definition A but not conscious under
definition B.
> Regarding the property dualisty, and why they'd expect other humans to
> be consciously experiencing is because they are,
I agree that they are, but I'm asking what observable criteria you are
using for drawing this conclusion. Unless you have some criterion, you
would have no basis for denying consciousness in a computer, a robot, or
a toaster.
> and what reason have
> they to believe they are a special case? And yes, the property
> dualists would be saying that the people that made the universality
> assumption would be unable to correctly predict the behaviour of
> something that is consciously experiencing, at least for as long as
> they held their assumption.
So can you provide some example of an aspect of human behaviour which
violates the physical laws which govern the behaviour of inanimate
objects? Unless you can do this I see no basis whatsoever for this
claim. Your argument seems to presuppose this, but nowhere do you
provide actual evidence for it.
> That the universality assumption was
> wrong, and that whether something is consciously experiencing or not,
> is significant to how you'd expect it to behave, in the sense of their
> being a difference between the expected behaviour for the hypothesis
> that it was consciously experiencing, and the null hypothesis that it
> wasn't.
The problem is I am unaware of any physicalist who would claim that
consciousness doesn't lead to behavioural differences. You keep
attributing this position to physicalists, but unless you can provide an
example of a physicalist who actually makes such a claim, I'd consider
the possibility that you are misinterpreting the physicalist position --
in effect you are arguing against a strawman of your own creation.
Andr�
someone2
> In article
-----------
If theory A and B are adopting the same operational definition of
consciousness then they *must* make different predictions regarding
its behaviour. Those who claim that consciousness can only be realised
by organic brains (and I'm not clear that there are such people --
identity theories of mind don't normally go quite this far) are
claiming that there are behaviours which brains exhibit which cannot
be reproduced in the robot, therefore they would not predict the same
behaviours from the robot as anyone who argues that it is in fact
conscious.
-----------
They definition of consciously experiencing that they are using, is
that it would be like something to be the thing. Both theory A and
theory B would be agreeing that at the molecular level there are
differences in behaviour between the robot and the human, by virtue of
a different chemical composition.
Consider a build of robot, and consider the expected behaviour of that
build if it simply followed the same laws of physics as things which
don't consciously experience. Can you see that neither those that
believe in theory A nor those that believe in theory B aren't going to
expect that behaviour. It seems to be getting a bit silly now, you
repeatedly avoiding facing it, and trying to reformulate the issue in
such a way as to disguise the fact of the matter.
Ymir
<338a46d3-8ead-41b3...@r38g2000yqn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
That's a conceptual definition, not an operational one. Theories require
that concepts be operationalised in terms of observable behaviour.
You yourself claim in previous posts that consciousness will have an
effect on behaviour. Given this, those differences in behaviour will be
observable to anyone regardless of their theoretical persuasion, and a
theory which purports to be one of consciousness is going to try to
account for those differences. You seem, however, to be suggesting that
your hypothetical physicalists are entirely oblivious to those
differences and can simply assert that something is conscious even if
their theory predicts behaviours which don't resemble the behaviour of
conscious entities.
Moreover, you seem to be insisting that the only behaviours which
physicalist theories can account for are those associated with
non-conscious entities (an entirely unsupported assertion). But if this
were the case, why would either physicalist assert that the robot (or
whatever other entity was under consideration) was conscious when their
theory (according to you) can only predict nonconscious behaviour?
Also, you seem to have glossed over one of the questions which I asked
above, one which is rather critical to providing justification for your
point of view. Can you provide an example of some aspect of human
behaviour which contradicts the laws of physics? If not, what basis do
you have for asserting that conscious entities do not follow the same
laws of physics as nonconscious ones? The argument in your paper strikes
me as a petitio principii, and I do not appear to be alone this regards,
so please don't simply refer me back to the paper -- I'm interested in
an actual example.
> Both theory A and
> theory B would be agreeing that at the molecular level there are
> differences in behaviour between the robot and the human, by virtue of
> a different chemical composition.
>
> Consider a build of robot, and consider the expected behaviour of that
> build if it simply followed the same laws of physics as things which
> don't consciously experience. Can you see that neither those that
> believe in theory A nor those that believe in theory B aren't going to
> expect that behaviour.
I can't parse the above. Can you please rephrase.
Andr�
someone2
> In article
Consider a robot which would have a certain expected behaviour if its
build were to simply follow the same laws of physics as things which
don't consciously experience.
Can you understand that both those that believe in theory A and those
that believe in theory B are going to expect that behaviour, for both
the hypothesis that it is like something to be the robot, and for the
hypothesis that it won't be.
Kermit
All things, conscious or not, follow the laws of physics, all the
time. Please provide an example of something which you believe does
not.
>
> Can you understand that both those that believe in theory A and those
> that believe in theory B are going to expect that behaviour, for both
> the hypothesis that it is like something to be the robot, and for the
> hypothesis that it won't be.
Philosophers may commonly define consciousness that way. I do not know
of any brain researcher who does. I have not read them all; could you
quote one? Every scientist I have read who is specializes in the brain
and behavior defines consciousness operationally. Consciousness
partially determines behavior. Consciousness *Is behavior (although
not all of it). If two scientific theories do not make different
predictions of observable events, they do not differ scientifically.
Please justify or clarify your strange assertion that conscious
entities do not always obey the laws of physics. You may as well try
to discuss nutrition while making claims that humans are powered by
nuclear reactors.
Scientific theories only consider observable events or their
consequences.
All things obey all laws of physics at all times.
Unless you address or clarify these two claims, you will get nowhere.
Kermit
someone2
Well if the neuroscientists define consciousness as something
different, then they aren't claiming it would be like anything to be
the thing they are labelling as 'conscious' and so are talking about
something different. The claims/assertions you feel I have to justify
aren't in the article, though your demands that I justify them did
clumsily distract away from you mentioning whether you understood and
agreed with the two paragraphs in the post you replied to.
[Though for your general information here is a paper from Francis
Crick and Koch http://www.klab.caltech.edu/~koch/Elsevier-NCC.html
where you'll notice consciousness is not defined 'operationally',
since the whole point is they don't know the neural correlate to it. I
think Francis Crick counts as a scientist who has some specialisation
in the brain. http://en.wikipedia.org/wiki/Francis_Crick
Further more your assertions that the universality assumption is true,
isn't fully accepted, for example if you read property dualism in
section 9 of http://consc.net/papers/nature.html , you'll notice
scientists like Roger Penrose are proponents of it.
http://en.wikipedia.org/wiki/Roger_Penrose . You understand presumably
that with property dualism, though things which consciously experience
and things that don't both would follow laws of physics, they wouldn't
be following the same ones. You couldn't describe the behaviour of
something that consciously experiencing soley using the laws of
physics which explain the behaviour of non-conscious things.]
Ymir
<9d0d5207-9ab0-4923...@w6g2000yqw.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I understand your claim. I simply don't consider it to be true (or even
meaningful). If theory A asserts the robot is conscious and theory B
asserts it isn't it it has to be because one theory predicts that the
robot will behave in a conscious manner whereas the other one predicts
it won't. Otherwise, any claims regarding consciousness don't actually
follow from the theories. They are simply assertions tacked on by the
theorists rather than claims of the theories.
Now can we get back to the question I asked above (which you have now
glossed over twice in a row) regarding your actual evidence that
conscious entities exhibit behaviours which don't follow the laws of
physics. Since you seem so certain of this providing a single example
shouldn't be all that difficult.
Andr�
someone2
> In article
Let's keep on the topic we are on, since it is the one in the article.
You have stated that you didn't consider it to be true that both
theories would have been expecting the robot build to give the
behaviour expected if it had simply followed the same laws of physics
as things which don't consciously experience. Though the only way that
could be true, is if one expected the behaviour to have shown a
diversion following the same laws of physics as things which don't
consciously experience. But both theories make the universality
assumption, so they would never expect that. So you were wrong.
Jim Lovejoy
news:cf8e7a37-c639-4592...@e11g2000yqo.googlegroups.com:
After reading your responses, I have to go back on my promise.
It's not possible to type *that* slowly.
I'd use simpler words, but words of less than one syllable are rather
scarce.
Instead I'll put it into words that the audience can understand, even if
you can't.
If two theories *always* make the same predictions, then they are really
the same theory. In the case you make, they are the same theory with
different untestable assumptions tacked on.
Untestable assumptions are not part of science. I'll repeat, in the hope
that eventually even you will understand. UNTESTABLE ASSUMPTIONS ARE NOT
PART OF SCIENCE.
Feel free to reply, to make clear just which part of my statements you
don't understand.
someone2
Firstly, the theories are philosophical theories, and it is plainly
wrong to suggest that a theory that it would be like something to be
the robot is positing the same reality as a theory that said it
wouldn't be like something to be the robot. The clue to it being
wrong, was that such a conception entails a contradiction.
If you wished to say that from a scientific perspective, that only the
universality assumption itself would be a scientific theory, since
there would be no scientific way to distinguish between philosophical
theories A and B, and that as such they should simply be regarded as
interpretations of the universality assumption, then that's fine. It
doesn't change the issue, nor would it be to correct to suggest that
the interpretations are just different ways of positing the same thing
(as explained).
The universality assumption itself is what implies that whether
something is consciously experiencing or not isn't significant to
expected behaviour, and would imply that us consciously experiencing,
wasn't significant to any attempts at trying to explain it.
Furthermore, if it were claimed that there was empirical evidence
(evidence based upon what it was like to be human) to support the the
universality assumption, then they would be in a contradictory
position. This is what the article was showing, the clue was in
section 2, labelled Aim.
Ymir
<dce6f12d-a11f-4a16...@j32g2000yqh.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> On 30 July, 17:13, Ymir <inva...@invalid.invalid> wrote:
<snippage>
> > I understand your claim. I simply don't consider it to be true (or even
> > meaningful). If theory A asserts the robot is conscious and theory B
> > asserts it isn't it it has to be because one theory predicts that the
> > robot will behave in a conscious manner whereas the other one predicts
> > it won't. Otherwise, any claims regarding consciousness don't actually
> > follow from the theories. They are simply assertions tacked on by the
> > theorists rather than claims of the theories.
> >
> > Now can we get back to the question I asked above (which you have now
> > glossed over twice in a row) regarding your actual evidence that
> > conscious entities exhibit behaviours which don't follow the laws of
> > physics. Since you seem so certain of this providing a single example
> > shouldn't be all that difficult.
> >
>
> Let's keep on the topic we are on, since it is the one in the article.
The question very much is on topic -- you may not directly mention this
issue in the paper, but it it an implied assumption of the article which
you have made more explicit in subsequent discussion.
You mentioned in a previous post that you had submmitted this to a
conference. It was rejected, but I assume that you are likely planning
to submit it to additional fora. I'd take the opportunity to use this
discussion as practice for any forum which does choose to accept your
submission. I can assure you that this is a question which will come up,
and if you don't at least attempt to address it, you will not come
across very well to your audience.
One of the reasons people present their ideas at conferences is to
explore objections to or consequences of their argument that they hadn't
necessarily anticipated, and one cannot in any forum expect the
discussion to focus only on those issues which you had hoped to discuss.
It's perfectly legitimate at a conference to answer a question with 'I
don't know', or 'I haven't explored that issue sufficiently to provide a
complete answer at this time', but simply dismissing a question as
irrelevant generally doesn't go over very well.
> You have stated that you didn't consider it to be true that both
> theories would have been expecting the robot build to give the
> behaviour expected if it had simply followed the same laws of physics
> as things which don't consciously experience.
All things follow the same laws of physics, but that doesn't necessarily
entail that they have identical behaviours. I suggested that the two
theories couldn't have identical expectations regarding the robot's
behaviour if (i) they were making different claims regarding its
conscious status, and (ii) they were in agreement in how they defined
consciousness operationally. Obviously, at most one of their sets of
predictions could actually be correct.
Let's imagine that we were to discover an alien technological artifact
whose function, initially, is unknown because we do not have an
appropriate power supply for it. Two groups of scientists investigate
it. The first concludes that it generates a destructive beam of energy
intended for use as a weapon, whereas the second concludes that it
generates a nondestructive beam of energy which is intended as a
scientific tool. Clearly they are predicting different things about its
behaviour, but both groups are operating under the assumption that it
acts according to the same laws of physics.
Your paper seems to treat consciousness our consciousness theorists in a
way which would be analogous to saying that both of our alien-artifact
theorists make identical predictions regarding *every* aspect of the
artifacts behaviour, but who still differ with respect to whether they
claim the artifact fires a destructive or non-destructive energy beam.
You then conclude from this that theories which assume that the same
laws of physics govern both destructive and non-destructive beams are
fundamentally flawed. This, however, is a nonsensical position. The
claim that the theories could make such contradictory claims while
predicting identical behaviours is non-sensical. Hence any conclusion
which follows from this is equally nonsensical.
> Though the only way that
> could be true, is if one expected the behaviour to have shown a
> diversion following the same laws of physics as things which don't
> consciously experience. But both theories make the universality
> assumption, so they would never expect that. So you were wrong.
But neither I, nor any physicalist, actually holds to the view that the
physical laws governing conscious entities are any different from those
governing nonconscious ones. You are the one making this claim, and this
is why I and others have pressed you to provide some evidence to
actually back this up. This is why I ask above for an example of some
aspect of human behaviour which violates physical laws.
When people question this assumption, you tend to either point them back
to your paper, or merely repeat the assertions which you make there, but
the argument in the paper simply isn't convincing. This is why people
are pressing you for additional evidence, and asking you to answer a
variety of questions regarding your position which may go beyond what
you intended to cover in the paper, but which try to get you to flesh
out some of your underlying assumptions.
When you present an argument and find that a significant number of
people do not find it compelling, or claim not to understand your
position, you really need to acknowledge the possibility that either
their is a fundamental flaw in your reasoning (in which case repeating
the same argument won't make it more convincing), or that you have
failed to accurately communicate the substance of your argument (in
which case making the same case in the same manner is unlikely to
convince).
Andr�
someone2
> In article
You know how consciously experiencing is defined in the article. The
difference between theory A and B is that theory A theorises that it
wouldn't be like anything to be the robot, whereas theory B suggests
it would be. Since both make the universality assumption, neither will
expect a difference in behaviour given the robot build, even though
one theorises it would be like something to be the robot and the other
one theorises it wouldn't. You are unable to dispute this is the case,
so you keep changing the scenario, in order to avoid admitting it.
Which seems like intellectual cowardice.
There are physicalists that don't claim the universality assumption,
see property duality section 9 for example http://consc.net/papers/nature.html
.
Ymir
<92539186-3a5c-47ac...@j21g2000yqe.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I know which definition is being used in the paper, but I don't consider
that definition to be useful, and I certainly am not accept conclusions
which rely solely on this particular definition. I am aware of very few
people in any discipline who rely on this definition (including people
such as Dennett who you erroneously claimed makes use of this based on
the fact that he referred to it at one point when arguing against it).
> Since both make the universality assumption, neither will
> expect a difference in behaviour given the robot build,
If you adopt your definition without operationalising it in some way,
then its not clear how any theory -- universalist or not -- would be
able to make any predictions.
To illustrate this, consider the following: While flipping through
channels the other night I caught a brief portion of some rather crappy
comedy in which two not-very-academically inclined college students were
contemplating *what it would be like* to be the bathtowel wrapped around
the otherwise unclad female lead.
Now, clearly they're weren't contemplating this the same way Nagel
contemplated his bats, but let's imagine that someone actually posed
this question more seriously.
Suppose you have two dualists (or, if you don't want them to be dualist,
pick any other univerality-denying position). One of them claims that it
actually would be like something to be a towel, whereas the other denies
this. How exactly would these non-universalists resolve the question of
whether the towel had experiences or not?
If you come up with a criterion which in any way refers to behaviour (or
lack thereof), then you've completely destroyed your own argument
against the physicalists since they would have equal access to those
behavioural criteria and their conclusion regarding whether a
robot/towel/whatever was conscious would follow from whether their
behavioural predictions correspond to conscious behaviour or not; in
which case they would not make different predictions about consciousness
if they made the same predictions regarding behaviour.
If, on the other hand, you provide no criterion at all (which is the
impressions that I get from your posts), then the dualists are in
exactly the same position which you allege the physicalists are in --
they can repeat 'It is/isn't like something to be a towel' as many times
as they want but this doesn't provide any evidence which might convince
the other one of their position. Moreover, they make no predictions
whatsoever regarding anything about the towel *which is actually
accessible to us*. Neither science nor philosophy get anywhere unless
their claims actually make a difference which we actually have access
to. As Ayer would argue, unless you can point out some observable
consequence of your claim, your claim is for all intents and purposes
meaningless.
> even though
> one theorises it would be like something to be the robot and the other
> one theorises it wouldn't. You are unable to dispute this is the case,
> so you keep changing the scenario, in order to avoid admitting it.
> Which seems like intellectual cowardice.
I'm trying to take your scenario and turn it into something which
actually makes sense and which involves physicalists who actually hold
to the tenets of physicalism. I'm also making use of alternate scenarios
to try to draw your attention to the petitio inherent in your argument.
You claimed at one point that you had discussed your paper with your
philosophy professor. I'd suggest that you forward this entire thread to
him. Regardless of what his position is on the subject of consciousness
I am quite sure that he will tell you that the participants in this are
not being evasive, intellectually cowardly, or any of the other things
which you have accused us of. We are simply raising the exact sorts of
questions which anyone presenting a paper (either formally at a
conference, or informally on the internet) should expect to be asked,
and should be prepared to at least attempt to answer.
> There are physicalists that don't claim the universality assumption,
> see property duality section 9 for example http://consc.net/papers/nature.html
> .
Property dualism is not, IMO, a form of physicalism. It purports to be a
form of monism, but I've never found this claim to be particularly
compelling. That aside, though, property dualists would face the same
problems regarding the robot as they would face with the towel above if
they adopt your particular definition of consciousness. The problem you
claim exists follows from your definition, not from the universality
assumption.
Andr�
someone2
> In article
> > .
>
> Property dualism is not, IMO, a form of physicalism. It purports to be a
> form of monism, but I've never found this claim to be particularly
> compelling. That aside, though, property dualists would face the same
> problems regarding the robot as they would face with the towel above if
> they adopt your particular definition of consciousness. The problem you
> claim exists follows from your definition, not from the universality
> assumption.
>
Dennet didn't just refer to the definition at one point in his
article, he used it throughout his article, and quoted several other
philosophers that also used it. He doesn't dispute that it is like
something to be a human, since he considers what it would be like to
be the robot, his issue is with the ontological significance it being
like something to be human has.
Property dualism is a form of physicalism, because they hold that only
the physical exists. That they hold this makes them physicalists.
------------
If you adopt your definition without operationalising it in some way,
then its not clear how any theory -- universalist or not -- would be
able to make any predictions.
To illustrate this, consider the following: While flipping through
channels the other night I caught a brief portion of some rather
crappy comedy in which two not-very-academically inclined college
students were contemplating *what it would be like* to be the
bathtowel wrapped around
the otherwise unclad female lead.
Now, clearly they're weren't contemplating this the same way Nagel
contemplated his bats, but let's imagine that someone actually posed
this question more seriously.
Suppose you have two dualists (or, if you don't want them to be
dualist, pick any other univerality-denying position). One of them
claims that it actually would be like something to be a towel, whereas
the other denies this. How exactly would these non-universalists
resolve the question of
whether the towel had experiences or not?
------------
If they don't make the universality assumption, then they wouldn't be
expecting it to be like anything to be the bath towel if its behaviour
could be predicted using the same laws of physics as are used to
predict the behaviour of things it wasn't anything to be like.
Are you willing to acknowledge that even when for a particular build
of robot, theory B proponents posit that it would be like something to
be the robot, and theory A proponents posit that it wouldn't be like
anything to be the robot, that they will never expect a difference in
expected behaviour? Don't change the scenario, or pretend to
misunderstand, as it is disingenious. Just answer the question as it
was put. If you do simply just 'misunderstand' or avoid answering, I
shall assume you have no interest in a proper discussion, and are just
trying to avoid admitting the point. As such, I'll not bother
continuing the converstation.
Ymir
<3ba0b470-ae4e-4749...@a26g2000yqn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> On 31 July, 19:27, Ymir <inva...@invalid.invalid> wrote:
> > In article
> > Property dualism is not, IMO, a form of physicalism. It purports to be a
> > form of monism, but I've never found this claim to be particularly
> > compelling. That aside, though, property dualists would face the same
> > problems regarding the robot as they would face with the towel above if
> > they adopt your particular definition of consciousness. The problem you
> > claim exists follows from your definition, not from the universality
> > assumption.
> >
>
> Dennet didn't just refer to the definition at one point in his
> article, he used it throughout his article, and quoted several other
> philosophers that also used it. He doesn't dispute that it is like
> something to be a human, since he considers what it would be like to
> be the robot, his issue is with the ontological significance it being
> like something to be human has.
He refers to it at many points, but isn't particularly sympathetic to it.
> Property dualism is a form of physicalism, because they hold that only
> the physical exists. That they hold this makes them physicalists.
Except that they surreptitiously sneak in causal devices which aren't
physical, and which therefore calls into question any commitment to
physicalism which they might claim to hold.
> ------------
> If you adopt your definition without operationalising it in some way,
> then its not clear how any theory -- universalist or not -- would be
> able to make any predictions.
>
> To illustrate this, consider the following: While flipping through
> channels the other night I caught a brief portion of some rather
> crappy comedy in which two not-very-academically inclined college
> students were contemplating *what it would be like* to be the
> bathtowel wrapped around
> the otherwise unclad female lead.
>
> Now, clearly they're weren't contemplating this the same way Nagel
> contemplated his bats, but let's imagine that someone actually posed
> this question more seriously.
>
> Suppose you have two dualists (or, if you don't want them to be
> dualist, pick any other univerality-denying position). One of them
> claims that it actually would be like something to be a towel, whereas
> the other denies this. How exactly would these non-universalists
> resolve the question of
> whether the towel had experiences or not?
> ------------
>
> If they don't make the universality assumption, then they wouldn't be
> expecting it to be like anything to be the bath towel if its behaviour
> could be predicted using the same laws of physics as are used to
> predict the behaviour of things it wasn't anything to be like.
A towel follows the laws of physics, but that doesn't mean its behaviour
is entirely predictable (if I throw a towel into the air it is not
possible to predict exactly how it will land -- there are simply too
many variables involved), so does that mean that they claim it would be
like something to be a towel or not? If you are asserting that simply
following the laws of physics is sufficient to rule out
being-like-something, then they are effectively denying that anything
has this property since there is nothing we are aware of (human beings
included) which don't follow said laws.
> Are you willing to acknowledge that even when for a particular build
> of robot, theory B proponents posit that it would be like something to
> be the robot, and theory A proponents posit that it wouldn't be like
> anything to be the robot, that they will never expect a difference in
> expected behaviour?
It's quite possible, as I've already stated, that the two theories might
predict identical behaviours, but that the theorists would disagree over
whether those behaviours provided sufficient evidence for consciousness,
but in such a case we are dealing with two different *definitions* of
consciousness.
What I cannot acknowledge is that they could make identical predictions
regarding its behaviour, *and* hold identical views on what
consciousness is, *and* not agree on whether it is conscious, because
that scenario simply doesn't make sense within a physicalist framework.
[put more plainly, the answer to your question above is 'no'.]
Contrary to your claims to the contrary I am unaware of *any* theory of
consciousness, including physicalist ones, which denies that
consciousness has effects on our behaviour, so no physicalist would ever
expect it to be possible to hold the behaviour constant while varying
the state of consciousness.
Your argument which you claim demonstrates that adopting the
universalist assumption leads to the view that consciousness doesn't
effect behaviour rests on the assumption that it *is* possible to vary
consciousness without affecting behaviour.
> Don't change the scenario
When I and others have offered alternate scenarios, it is an attempt to
drive home the logical flaw in your argument -- it is hoped that
constructing scenarios which are clearly absurd, but which have the same
internal logical structure as your argument, might make it more clear to
you that your argument is a petitio.
> or pretend to
> misunderstand, as it is disingenious.
Please do not attribute motives to others which are not there.
No one here is pretending to misunderstand you. When we claim not to
understand, it is because we genuinely don't understand exactly what you
are saying and want clarification. If we misunderstand you, it is
because those clarifications have either been themselves unclear, or
have not been forthcoming. On a variety of occasions, I and others have
asked you questions which you have ignored, even when those questions
have been posed several times by several different people. Those
questions have, in fact, been genuine attempts to get you to clarify
your position.
Andr�
Eric Root
> In article
someoneN always gets to this point: he simply asserts that the two
positions expect no difference in behavior, and when everybody denies
that, he starts calling names. Every time he shows up, it gets to
this same point, and he never learns anything. He'll reword his
arguments, but ultimately, will never present an argument to show that
"physicalists" expect conscious and unconscious entities to behave the
same.
Eric Root
someone2
> In article
You change the scenario, and avoid answering the question, and have
repeatedly done so.
Note that the question didn't use the term consciousness, to avoid you
trying to disingeniously answer avoiding the point being made, by
bringing in a different definition of consciousness, or suggestions as
you did that the theorists might use different definitions of
consciousness.
The question was asking whether the proponents of theory A and theory
B would ever expect the robot build to behave differently, if they
disagreed about whether it would be like anything to be that robot.
The answer is that they couldn't, since both would expect the build to
follow the same laws of physics regardless of whether theory A was
true, and it wasn't like anything to be the robot, or whether theory B
was true, and it was like something to be the robot.
Ymir
<4a4e8079-59ed-463c...@e27g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
If, as you claim, theories A and B are *physicalist* theories, then
'being like something' is (to proponents of these theories) a physical
property which follows from the laws of physics. I stand by my above
paragraph even if you replace 'conscious' with 'being like something' in
all instances (modulo issues of grammar). If you want to define 'being
like something' as something intrinsically non-physical which does not
play any causal role in determining behaviour and which does not follow
from the laws of physics, then you are describing a (IMO imaginary)
property regarding which *physicalist* theories would make no
predictions at all.
Andr�
someone2
> In article
Being like something would an identity of the the physical reality, if
the physicalists were right. So as physicalists the proponents aren't
considering being like something to be non-physical.
One is explaining it as being (an identity of) certain biological
activity, the other is explaining it as being an identity of
function.
Theory A is a version of the Identity Theory of Mind:
http://plato.stanford.edu/entries/mind-identity/
Whereas Theory B is a version of functionalism:
http://plato.stanford.edu/entries/functionalism/
Both are considering that the behaviour of something that it is like
something to be, can have its behaviour described using the same laws/
terms of physics as something that it isn't anything to be like. Thus,
they will expect the build to follow the whatever the known laws of
physics are at the time. Neither will expect the build not to follow
the known laws of physics, and thus will always expect the same
behaviour from the build, even though they posit a different reality
regarding whether it is like something to be the robot or not.
Ymir
<086ef1c4-075d-445c...@32g2000yqj.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
Your use of the word 'explain' here is a bit misleading, since you are
stating the conclusions which your theorists reach without discussing
the actual details of the theories themselves or showing how those
conclusions are reached. I've avoided discussing the specifics of
theories A and B thus far because you assert in your paper that these
are only meant as examples and that the same claims could be made of any
physicalist theories (though you don't really demonstrate how the
general case follows from the specifics).
> Theory A is a version of the Identity Theory of Mind:
> http://plato.stanford.edu/entries/mind-identity/
There are many identity theories of mind, and their crucial property is
that they all claim that specific mental states are identified with
specific brain states. They do not necessarily, however, state that this
relationship is isomorphic. Let's assume, however, that a particular
theory does assert an isomorphism between brain states and mental
states. It does not follow from this that the theory denies the
possibility of a non-brain having mental states, merely that those
mental states could not be identical to the mental states which a brain
possesses. Or, if you prefer, it maintains that it is possible for it to
be like something to be a robot -- it just insists that
being-like-a-robot and being-like-a-person are not the same (just as
being like person X and being like person Y are not the same, though
they are presumably more similar to one another than they are to the
robot), which I don't think anyone here would argue against.
> Whereas Theory B is a version of functionalism:
> http://plato.stanford.edu/entries/functionalism/
Functionalism, on the other hand, is not a physicalist position at all
-- it is a methodological programme which is entirely *consistent* with
physicalism but which explicitly rejects the utility of reducing mental
states to their underlying physical structures since it argues that that
would end up missing various important generalisations regarding mental
states. Because you and I do not have identical brain states, we do not,
according to identity theory, have identical mental states, but
functionalists would argue that we have mental states which are
sufficiently similar that we can treat them as being the same and
discuss consciousness in terms of the functional relationships between
those states without worrying about their physical instantiations, much
the same as a C programmer can multiply two numbers without worrying
about differences between how the logic gates which perform
multiplication are organised on a PDP-11 and a Intel Core 2 Duo (or how
transistors are arranged to form those gates, or how the properties of
semiconductors allow those transistors to function, etc. etc. etc.).
Under this view, a very different physical substrate, such as a
conscious robot might still have its mental states described using
similar functional relations despite the differences in their physical
realisation. Note that the functionalist notion of 'multiple
realisability' doesn't entail that any arbitrary system can instantiate
given functions -- merely that their may be different physical systems
capable of doing so.
If your identity theorist were to claim that *only* brains can exhibit
mental states, they are in effect stating that the functions relevant to
consciousness are dependent on physical peculiarities of the nervous
system, and that it is not possible to duplicate those functions on
other types of physical systems. But this is a claim about the
*physical* behaviours of both biological and non-biological systems.
A functionalist who argues that it *is* possible to instantiate those
functions in silicon (or whatever) and the identity theorist above are
therefore making radically different claims concerning the types of
behaviours which one expects silicon to exhibit (presumably not just
with respect to consciousness), so it is *not* the case that they will
have identical expectations regarding the behaviour of our robot.
If the identity theorist is simply asserting the impossibility of being
like something in a non-biological system without relating that to the
underlying physical behaviours of biological systems and how those
behaviours differ from those of non-biological systems, then this claim
*doesn't* actually follow from their theory, nor is it a claim that is
rooted in physicalism, the universality assumption, or anything else for
that matter.
If you want to argue that two theorists may agree on the behaviour of a
system yet disagree with respect to some property of those systems which
isn't actually addressed by their theories, then I will agree with you.
However, if you claim that the theories themselves can agree on all
aspects of the behaviour, yet disagree with respect to some property
*which the theories actually purport to account for*, then I will
continue to maintain that this position is nonsensical (provided the
theories agree on their definitions).
Andr�
someone2
> In article
The theories are agreeing on the definition of something consciously
experiencing meaning that it would be like something to be that thing.
The theories are theories about the nature of consciously
experiencing.
They differ in their beliefs on its nature. Such that they will
disagree about the behaviours to which they think the term could be
correctly applied. Though don't make the mistake of thinking that they
define consciously experiencing as being the behaviour they thought it
would be something to be like. They can understand that they could be
wrong about which behaviour the term could be correctly applied to.
If either theory were correct, it could be thought to explain why the
term could be correctly applied to some behaviours and not others.
Even though given the robot build they might disagree over whether the
term 'consciously experiencing' can be correctly applied to the robot,
they are never going to disagree about the expected behaviour of the
robot. Nor would either expect any difference in behaviour if they
were wrong about whether the term could be correctly applied to the
robot.
Ymir
<c8df9f5f-72ad-4321...@v36g2000yqv.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
What you are describing is a situation where they do *not* agree on the
definition of the term. Their definitions might significantly overlap,
but they are not identical.
Incidentally, though, why do you think this sort of issue only arises
for universalists? Could two dualists not disagree over whether the term
'consciousness' could be applied to a given entity? As a concrete
example, some attribute consciousness to the existence of a soul, but
there is a large disagreement over whether (e.g.) pets possess souls.
This disagreement, however, isn't one about how dogs and cats behave.
If the existence of definitional disagreements among theorists renders
physicalism implausible, wouldn't it equally render soul-based accounts
implausible?
Andr�
someone2
> In article
No, they both share the same definition. If it wasn't like something
to be the robot, then both agree that it wouldn't be correct to
identify it as consciously experiencing, and if it was like something
to be the robot, then both agree that it would be correct to identify
it as consciously experiencing.
So to repeat:
---------
Even though given the robot build they might disagree over whether the
term 'consciously experiencing' can be correctly applied to the robot,
they are never going to disagree about the expected behaviour of the
robot. Nor would either expect any difference in behaviour if they
were wrong about whether the term could be correctly applied to the
robot.
---------
Ymir
<dcc6483c-7b89-4f5e...@w6g2000yqw.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
Saying the both agree that being like something is the defining
attribute of consciousness doesn't mean they agree on the definition --
the fact that they attribute this to different things suggests they have
a different concept of what 'being like something' means -- scientific
definitions need to be stated in terms of observable properties, and
clearly they do not agree on the observable properties involved if they
are suggesting that it should be applied to different things.
Is there some reason why you didn't address the question which I posed
regarding why only universalist positions are vulnerable to the
criticism which you are attempting to level?
someone2
> In article
The reason I didn't go onto other points is that we over the time I
should think exchanged over 100 posts over the current point. I'm sure
I have explained before that I want to concentrate on this point, as
you seem to have had such difficulty in understanding it.
The definition isn't stated in terms of properties observable from a
third person perspective. The reason it isn't is that if it wasn't
like something to be the robot, then both agree that it wouldn't be
correct to identify it as consciously experiencing, and if it was like
something to be the robot, then both agree that it would be correct to
identify it as consciously experiencing. They can't define it in terms
of properties observable from a third person perspective without there
being some scientific way to establish which observable properties it
would be something to be like, and as I have been trying to explain,
if the universality assumption were true, there would be no way of
scientifically distinguishing whether theory A or theory B was
correct. As I mentioned to you before "... don't make the mistake of
thinking that they define consciously experiencing as being the
behaviour they thought it would be something to be like. They can
understand that they could be wrong about which behaviour the term
could be correctly applied to."
So to repeat:
---------
Even though given the robot build they might disagree over whether the
term 'consciously experiencing' can be correctly applied to the robot,
they are never going to disagree about the expected behaviour of the
robot. Nor would either expect any difference in behaviour if they
were wrong about whether the term could be correctly applied to the
robot.
---------
Kermit
> On 30 July, 16:11, Kermit <unrestrained_h...@hotmail.com> wrote:
>
> > On Jul 30, 6:35 am, someone2 <glenn.spig...@btinternet.com> wrote:
>
> > > On 30 July, 05:14, Ymir <inva...@invalid.invalid> wrote:
<snip>
On the contrary, you do not address this (and I am not the first, nor
the last, to bring it up):
If the two theories both predict the same behavior under the same
circumstances, they are not different scientifically.
I also have difficulty in imagining what you mean to say that
conscious and non-conscious things would behave the same. Even if they
are superficially different (say, a conscious human *completely
paralyzed, and a brain-dead human still breathing) it would simply be
because our instruments are insufficient. Until they behave observably
differently (the paralyzed person heals, or we bring in an MRI
scanner) we cannot say which is conscious and which is not, or both,
or neither.
Unless the theories predict different observable behavior under
specified circumstances, they are the same.
>
> [Though for your general information here is a paper from Francis
> Crick and Koch
> http://www.klab.caltech.edu/~koch/Elsevier-NCC.html
> where you'll notice consciousness is not defined 'operationally',
> since the whole point is they don't know the neural correlate to it. I
> think Francis Crick counts as a scientist who has some specialisation
> in the brain.http://en.wikipedia.org/wiki/Francis_Crick
Yes, thank you.
But they can only study observable evidence. How would we know whether
a non-human is conscious - whether dog or robot - except by observing
its behavior?
>
> Further more your assertions that the universality assumption is true,
> isn't fully accepted, for example if you read property dualism in
> section 9 ofhttp://consc.net/papers/nature.html, you'll notice
> scientists like Roger Penrose are proponents of it.http://en.wikipedia.org/wiki/Roger_Penrose. You understand presumably
> that with property dualism, though things which consciously experience
> and things that don't both would follow laws of physics, they wouldn't
> be following the same ones.
That is not at all obvious. An airplane and a toaster both follow the
same laws of physics, but do not behave the same way. Structures which
are different can easily be imagined to be following the same laws of
physics, but behave differently. If, say, quantum effects were
important in understanding consciousness, but not in toasters, we
might ignore a certain subset of the laws of physics when designing
toasters, but they *all apply to everything at the same time*. It may
simply be that for our purposes, we ignore some of them some of the
time.
Remember, too, that the "laws" of physics are simply human
descriptions of universal behavior as we perceive and understand it.
The laws manifest differently (if only in degree) under various
circumstances, but they still apply, and the same laws they are. For
instance, we think that the velocity of an object affects its mass,
but this relativistic effect is only significant at very high speeds.
But the relativistic effects of mass, length. etc. are there all the
time.
> You couldn't describe the behaviour of
> something that consciously experiencing soley using the laws of
> physics which explain the behaviour of non-conscious things.]
Nobody can describe them yet, completely. That's what the paper you
linked to was about. But there's no reason to think we won't be able
to eventually, or that it can't be in principle. And when we can
describe them completely, we be will using the laws of physics (more
likely, chemistry, information theory, genetics, and perhaps some
others. We may ignore physics altogether, except in certain very
simple ways. (Like the caloric requirements of the brain, or the light
frequencies that eyes can react to.))
Physicalists would not postulate behavior as independent of the brain
design or structure.The mind is a subset of behavior of the brain
interacting with the environment surrounding the brain. You may think
otherwise, but you are a dualist.
Kermit
Ymir
<5bd342f2-0050-431f...@b14g2000yqd.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I understand it fine. I just don't buy it. Which suggests it might be a
good idea to move on and then revisit this point later -- you can't
enter a discussion forum and reasonably expect people to address your
questions if you will not address theirs. I've indicated what I perceive
to be a problem above, and it would be nice if you would address it.
> The definition isn't stated in terms of properties observable from a
> third person perspective. The reason it isn't is that if it wasn't
> like something to be the robot, then both agree that it wouldn't be
> correct to identify it as consciously experiencing, and if it was like
> something to be the robot, then both agree that it would be correct to
> identify it as consciously experiencing. They can't define it in terms
> of properties observable from a third person perspective without there
> being some scientific way to establish which observable properties it
> would be something to be like, and as I have been trying to explain,
> if the universality assumption were true,
As I tried to point out above, in the part which you didn't address,
this would be true regardless of whether you adopt the universality
assumption or not -- if you define something in a way that makes it
inherently inaccessible, then you effectively are declaring it to be off
limits to investigation. So I'll ask again -- why do you think that
approaches which *don't* adopt the universality assumption are somehow
immune from this?
> there would be no way of
> scientifically distinguishing whether theory A or theory B was
> correct. As I mentioned to you before "... don't make the mistake of
> thinking that they define consciously experiencing as being the
> behaviour they thought it would be something to be like. They can
> understand that they could be wrong about which behaviour the term
> could be correctly applied to."
>
>
> So to repeat:
> ---------
> Even though given the robot build they might disagree over whether the
> term 'consciously experiencing' can be correctly applied to the robot,
> they are never going to disagree about the expected behaviour of the
> robot. Nor would either expect any difference in behaviour if they
> were wrong about whether the term could be correctly applied to the
> robot.
I've already addressed this paragraph, and my position on it remains the
same. There's no need to keep appending it to the end of your messages.
Andr�
someone2
> In article
You haven't addressed it. You simply stated that you didn't think they
were using the same definition, which is why I spent time over the
last couple of posts explaining to you that they were. You didn't just
contest that were using the same definition, or claim not to
understand why they can't use a observational definition and be sure
it would be like something to be the chosen observational properties.
In the hope that you would be able to make such progress, I posted it
on the end again, so that you could say whether you agree that they
wouldn't be expecting any difference in behaviour if they were wrong
about wehtehr the term could be correctly applied to the robot.
Ymir
<45bb197f-a965-4559...@h31g2000yqd.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
And I still do not agree with you. Repeating the same points isn't going
to make them any more convincing since they are predicated on
ill-conceived definitions and the assumption that your 'theories' make
assertions which don't actually follow from the theories themselves
(e.g. you assume a 'physicalist' theory which posits only brains can
experience, but which doesn't provide any physical basis for this claim)
I'm happy to try to clarify this further, but only after you make at
least some attempt to address some of the questions which have been
asked of you.
In particular:
(1) Why do you think that your criticism doesn't equally apply to
non-physicalist/universalist theories given the fact that there are
disagreements among dualists over which entities consciousness can be
ascribed to?
(2) Can you identify a single aspect of human behaviour which violates
physical laws?
Andr�
someone2
> In article
I think we can stop the converstation, from your weak response, I
think you got the point.
someone2
Actually that was a bit harsh, I'll answer your points.
1) because if they don't make the universality assumption, then the
expected behaviour for the build consciously experiencing will not be
the same as the expected behaviour for if it isn't.
2) No, but if Penrose and Hammerhof were right for example it could be
at the quantum level. If the universality assumption isn't correct
though... oh yeah, you still don't understand why it isn't the
universality assumption.
I had said:
---------
Even though given the robot build they might disagree over whether the
term 'consciously experiencing' can be correctly applied to the robot,
they are never going to disagree about the expected behaviour of the
robot. Nor would either expect any difference in behaviour if they
were wrong about whether the term could be correctly applied to the
robot.
----------
You say you disagree, now could you please clarify why. The definition
of consciously experiencing that is being used is that if something is
consciously experiencing it will be like something to be that thing.
Furthermore, the theories are contrived to show the point. So given
the definition and those theories, please explain how you dispute the
point.
Ymir
<080002cc-6849-4830...@c29g2000yqd.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
But so far the only difference which you've pointed to is that they
wouldn't expect the conscious entity to follow the laws of physics, but
below you acknowledge that you haven't observed conscious entities
failing to follow the laws of physics, so you're not really predicting
anything here.
Let's take the example I gave above. Two people, neither of whom is a
dualist, argue over whether a cat is conscious. They are not in
disagreement over how the cat behaves, simply over whether cats possess
souls as for them, consciousness rests on having a soul. What
differences in behaviour would these two expect to find *which they can
actually point out to one another*. If you can't provide a description
of this, then it would seem that their views are, by your own reasoning,
equally implausible as any predictions made by those adopting the
universalist assumption.
> 2) No, but if Penrose and Hammerhof were right for example it could be
> at the quantum level. If the universality assumption isn't correct
> though... oh yeah, you still don't understand why it isn't the
> universality assumption.
>
> I had said:
> ---------
> Even though given the robot build they might disagree over whether the
> term 'consciously experiencing' can be correctly applied to the robot,
> they are never going to disagree about the expected behaviour of the
> robot. Nor would either expect any difference in behaviour if they
> were wrong about whether the term could be correctly applied to the
> robot.
> ----------
>
> You say you disagree, now could you please clarify why. The definition
> of consciously experiencing that is being used is that if something is
> consciously experiencing it will be like something to be that thing.
You haven't stated how they define what it is like to be like something.
If these people are actually engaged in creating theories, they will
need to define more clearly what that actually means -- I'm suggesting
that they do not, in fact, share the same definition of this if they
associate consciousness with different sets of observed behaviours
(which must be the case if one attributes consciousness to the robot and
one does not).
> Furthermore, the theories are contrived to show the point. So given
> the definition and those theories, please explain how you dispute the
> point.
The theories are most definitely contrived. They fail to show the point,
however, because you don't actually specify the contents of these
theories. You don't suggest what observable fact about the universe has
lead the one research to conclude that only organic brains are capable
of conscious experience. Without being told where this claim comes from,
we are not in a position to know what predictions this theory would or
wouldn't make. You keep asserting that it would predict the same
behaviour as the second (equally undefined) theory. Moreover, you keep
asserting that the contents of this theory are physicalist in nature,
but insist that the theorist is going to insist on holding themselves to
a definition of consciousness which is neither operationalised nor
grounded in the physical.
I already addressed this in an earlier post, but you didn't respond to
any of the specifics of that post. Why don't you reread my earlier
message and respond to it point by point rather than by simply repeating
the same vague argument at the end of each reply.
<http://groups.google.ca/group/talk.origins/msg/93bfd6b606996997?hl=en&dm
ode=source>
Andr�
someone2
> In article
Look this is taking ages, so much so, that I have openly questioned
whether you were being a disingenious intellectual coward. Though if
you aren't let me take you through the argument step by step, to see
what steps you can follow, and what step is giving you a problem. Then
we can discuss the problem, or if in agreement move onto the next
step, and thus slowly make progress. Assuming you weren't previously
attempting to obfuscate the issue with disingenious replies, I can't
see why you'd object, as it would save us time. So we can keep the
correspondance short and to the point.
Step 1: Are you ok with the physicalists which have made the
universality assumption claiming that what it is like to be human has
a physical basis, and making suggestions about what was significant
about the brain behaviour such that it was like something to be a
human?
Ymir
<28ba3d7d-727b-4d2a...@g6g2000vbr.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
That's rather brazen considering that once again you have entirely
skipped the questions which I asked.
> Though if
> you aren't let me take you through the argument step by step, to see
> what steps you can follow, and what step is giving you a problem. Then
> we can discuss the problem, or if in agreement move onto the next
> step, and thus slowly make progress. Assuming you weren't previously
> attempting to obfuscate the issue with disingenious replies, I can't
> see why you'd object, as it would save us time. So we can keep the
> correspondance short and to the point.
I'll answer the below questions, but if you want me to continue with
this discussion you will have to start answering my questions as well.
> Step 1: Are you ok with the physicalists which have made the
> universality assumption claiming that what it is like to be human has
> a physical basis
Physicalists by definition assume that everything ultimately has a
physical basis, so yes they would assume that 'what it is like' to be
human would have a physical basis.
> and making suggestions about what was significant
> about the brain behaviour such that it was like something to be a
> human?
Anyone can make suggestions. Whether I accept those suggestions would
depend on their actual content and the emipirical evidence which they
bring forward to back them up.
I'm not going to address anymore of your questions until you discuss the
example I raise regarding the cat above. What distinguishes that
situation from debate between your hypothetical 'physicalists'?
Andr�
Glenn Spigel
> In article
I not sure I fully got your cat example. Neither are dualists, and yet
they are considering whether the cat has a soul. I'll assume then that
they are monists, and considering there to only be the spiritual. So
either they make the universality assumption, or they don't. Since you
are using this as an example to contrast against those that make the
universality assumption, I'll assume that you didn't intend them to.
If that is correct, then the one that thinks the cat is conscious
would be considering that the cats behaviour cannot be predicted as a
result of the cat's build following soley laws of physics which are
used to describe the behaviour in non-conscious things. Due to the
complexity, it may well be that neither can show themselves to be
correct. It might be as mentioned that it is at the quantum level that
the difference occurs. An inability to show the difference in expected
behaviour is different from there not being expected to be any
difference.
Regarding the step by step explanation, you seemed ok with step 1,
pointing out that the first part had to be true, since they were
physicalists. Though on the second part, you didn't specifically say
you agreed or disagreed. Instead, feeling perhaps that it was
relevant, you just commented that you personally might not agree with
their suggestions. Can I take that you are ok with what has been said
in step 1? I've pasted it below as a reminder:
-----------
Step 1: Are you ok with the physicalists which have made the
universality assumption claiming that what it is like to be human has
a physical basis and making suggestions about what was significant
about the brain behaviour such that it was like something to be a
human?
-----------
If you are ok with step 1, we can move onto step 2 if you are ready,
or would you prefer to avoid that for a bit?
someone2
> In article
I not sure I fully got your cat example. Neither are dualists, and yet
they consider whether the cat has a soul. I'll assume then that they
are monists, and considering there to only be the spiritual. So either
they make the universality assumption, or they don't. Since you are
using this as an example to contrast against those that make the
universality assumption, I'll assume that you didn't intend them to.
If that is correct, then the one that thinks the cat is conscious
would be considering that the cats behaviour cannot be predicted as a
result of the cat's build following soley laws of physics which are
used to describe the behaviour in non-conscious things. Due to the
complexity, it may well be that neither can show themselves to be
correct. It might be as mentioned that it is at the quantum level that
the difference occurs. An inability to show the difference in
behaviour is different from there not being expected to be any
difference.
You seemed ok with step 1, pointing out that the first part had to be
true, since they were physicalists, though on the second part, you
didn't specifically say you agreed or disagreed. Instead, feeling
perhaps that it was relevant, just commented that you personally might
not agree with their suggestions. Can I take that you are ok with what
has been said in step 1? I've pasted it below as a reminder:
-----------
Step 1: Are you ok with the physicalists which have made the
universality assumption claiming that what it is like to be human has
a physical basis and making suggestions about what was significant
about the brain behaviour such that it was like something to be a
human?
-----------
If you are, we can move onto step 2 if you are ready, or would you
Ymir
<46e707b3-eba3-4df3...@s6g2000vbp.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> I not sure I fully got your cat example. Neither are dualists,
My fault -- I wrote 'dualist' when I had intended to write
'physicalist'. Sorry about that.
> and yet
> they consider whether the cat has a soul. I'll assume then that they
> are monists, and considering there to only be the spiritual. So either
> they make the universality assumption, or they don't. Since you are
> using this as an example to contrast against those that make the
> universality assumption, I'll assume that you didn't intend them to.
> If that is correct, then the one that thinks the cat is conscious
> would be considering that the cats behaviour cannot be predicted as a
> result of the cat's build following soley laws of physics which are
> used to describe the behaviour in non-conscious things. Due to the
> complexity, it may well be that neither can show themselves to be
> correct. It might be as mentioned that it is at the quantum level that
> the difference occurs.
I generally don't take claims about the 'quantum level' very seriously
unless they are actually claims about the realm of quantum physics. The
reason for this is simply that 'quantum' has become somewhat of a
buzzword which is frequently misused and abused.
> An inability to show the difference in
> behaviour is different from there not being expected to be any
> difference.
But 'expecting a difference' generally entails stating what that
difference will be in ways which makes it possible to look for it.
Followup question: Do you consider a cat to be conscious, and if so why?
> You seemed ok with step 1, pointing out that the first part had to be
> true, since they were physicalists, though on the second part, you
> didn't specifically say you agreed or disagreed. Instead, feeling
> perhaps that it was relevant, just commented that you personally might
> not agree with their suggestions. Can I take that you are ok with what
> has been said in step 1? I've pasted it below as a reminder:
>
> -----------
> Step 1: Are you ok with the physicalists which have made the
> universality assumption claiming that what it is like to be human has
> a physical basis and making suggestions about what was significant
> about the brain behaviour such that it was like something to be a
> human?
> -----------
>
> If you are, we can move onto step 2 if you are ready, or would you
> prefer to avoid that for a bit?
Why don't you simply present all your numbered steps in a single post. I
can then point out which of your steps are sound and which strike me as
unsound. I'm fine with your first step.
Andr�
someone2
> In article
As for whether it is like anything to be a cat, I don't know, but I
treat the cat as though it is to be on the safe side, since it doesn't
display any behaviour which gives me cause to think that it isn't.
I'd rather go through the steps 1 by 1 if that is ok, as I might vary
them, according to how you get on. I'm going to start off trying with
a different physicalist suggestion from the ones in the paper, and see
how you get on with it.
Step 2: One physicalist suggestion is that what it is like to be human
is an identity of the neuron arrangement in the human brain. As such
they suggest that if something such as a robot doesn't have the
biological neurons arranged in a similar fashion, then either it
wouldn't be like anything to be the robot, or that it would be like
something to be it, but it wouldn't be the same as being a human. Do
you understand what they are suggesting?
Ymir
<837942f5-fd5c-4a06...@g6g2000vbr.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
This, though, tends to reinforce that it isn't simply physicalists who
use behavioural criteria to establish consciousness -- we need
operational criteria in order to investigate *anything*, physicalist or
not, universalist or not. That are operational definitions may not
adequately capture the thing which we are trying to explain is a pitfall
which any theory must deal with, but science isn't about absolute
certainty; it's about providing the best explanation possible given the
data available to us.
> I'd rather go through the steps 1 by 1 if that is ok, as I might vary
> them, according to how you get on. I'm going to start off trying with
> a different physicalist suggestion from the ones in the paper, and see
> how you get on with it.
>
> Step 2: One physicalist suggestion is that what it is like to be human
> is an identity of the neuron arrangement in the human brain. As such
> they suggest that if something such as a robot doesn't have the
> biological neurons arranged in a similar fashion, then either it
(A)
> wouldn't be like anything to be the robot, or
(B)
> that it would be like
> something to be it, but it wouldn't be the same as being a human. Do
> you understand what they are suggesting?
I understand the suggestion, but as stated it is not specifically a
*physicalist* suggestion because no underlying physical rational is
provided for the claim. I'm not here calling into question your
researcher's commitment to physicalism, merely pointing out that as
stated nothing about this claim is dependent on physicalism.
To make this into a full fledged theory, they would need to provide an
account of how consciousness arises in the brain and what physical
peculiarities of the nervous system lead to the conclusion that only
humans (or things possessing biological nervous systems/whatever their
specific claim is) can be conscious (especially if they adopt the much
stronger position which I have labelled as A above rather than the
weaker position labelled as B). Without such details you are right to
claim that it makes no actual predictions about behaviour (human,
robotic, or otherwise), but your original argument revolved around
physicalist theories rather than pretheoretical speculations which might
initially guide one in forming a theory, but which may or may not prove
to be accurate or useful in the long run.
Andr�
someone2
> In article
It is only a suggestion, and so doesn't have to meet the criteria you
dictate for what you call a full fledged theory. Also the suggestion
that what it is like to be human is an identity of the neuron
arrangement in the human brain, has been put forward by physicalists
such as Place. If your point is that it doesn't rule out other people
who aren't physicalists suggesting such a thing, then I'll reword step
2.
----------
Step 2: One physicalist suggests that all that exists is the physical,
and that what it is like to be human is an identity of the neuron
arrangement in the human brain. As such they suggest that if something
such as a robot doesn't have the biological neurons arranged in a
similar fashion, then either it wouldn't be like anything to be the
robot, or that it would be like
something to be it, but it wouldn't be the same as being a human. So
did you understand what was being suggested?
----------
Anyway, you said you understood it, so I'll continue.
Step 3: For a robot that passes the Turing Test, those believing the
suggestion put forward in step 2, consider that either it would be
like something to be the robot, or that it wouldn't.
Are you ok with step 3?
Ymir
<fde2bfb7-648f-4033...@d4g2000vbm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> Step 3: For a robot that passes the Turing Test, those believing the
> suggestion put forward in step 2, consider that either it would be
> like something to be the robot, or that it wouldn't.
>
> Are you ok with step 3?
>
Step three boils down to either A or not A. Anyone who doesn't accept
this needs a remedial logic course. So yes I accept it.
Andr�
someone2
> In article
>
> someone2 <glenn.spig...@btinternet.com> wrote:
> > Step 3: For a robot that passes the Turing Test, those believing the
> > suggestion put forward in step 2, consider that either it would be
> > like something to be the robot, or that it wouldn't.
>
> > Are you ok with step 3?
>
> Step three boils down to either A or not A. Anyone who doesn't accept
> this needs a remedial logic course. So yes I accept it.
>
Step 4: 'behaviour X' will be defined as the resultant behaviour of a
robot build following the same laws of physics as things that it isn't
anything to be like. Those believing in the suggestion put forward in
step 2 would expect behaviour X (given the robot build) if reality was
that it wasn't like anything to be the robot.
Are you ok with step 4?
Ymir
<3b15903f-4a8f-41cc...@g1g2000vbr.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> On 7 Aug, 17:30, Ymir <inva...@invalid.invalid> wrote:
> > In article
> > <fde2bfb7-648f-4033-9a15-a725c730a...@d4g2000vbm.googlegroups.com>,
> >
> > �someone2 <glenn.spig...@btinternet.com> wrote:
> > > Step 3: For a robot that passes the Turing Test, those believing the
> > > suggestion put forward in step 2, consider that either it would be
> > > like something to be the robot, or that it wouldn't.
> >
> > > Are you ok with step 3?
> >
> > Step three boils down to either A or not A. Anyone who doesn't accept
> > this needs a remedial logic course. So yes I accept it.
> >
>
> Step 4: 'behaviour X' will be defined as the resultant behaviour of a
> robot build following the same laws of physics as things that it isn't
> anything to be like.
This is where things stop making much sense.
Let me define 'behaviour' Y as the resultant behaviour of a toaster
built following the same laws of physics as a microwave oven. If I
expect the toaster to exhibit Y, am I claiming simply that the same laws
of physics are involved for both toasters and ovens, or am I claiming
that toasters should somehow behave as ovens? Physicalism assumes that
there is a single unified set of physical laws, so they assume that a
conscious robot and a non-conscious robot (which would have to differ
from one another physically in some way) would follow the same laws of
physics, but that doesn't mean that they would expect them to behave in
the same way.
> Those believing in the suggestion put forward in
> step 2 would expect behaviour X (given the robot build) if reality was
> that it wasn't like anything to be the robot.
Now, in your second step you left it open whether the theorist took the
robot to be conscious or not, but here you are claiming that they will
take it to have the behaviour of a robot which isn't conscious
regardless of whether they take the robot to be conscious or not. Would
you similarly think it reasonable to assume that someone would expect
behaviour Y from a toaster if reality was such that it was actually an
oven? That has the same form as the position which you are attributing
to the physicalist, and yet it clearly makes no sense with respect to
toasters and ovens. I fail to see why you think it makes sense with
respect to conscious robots and non-conscious robots.
> Are you ok with step 4?
Clearly I am not OK with it.
Andr�
someone2
> In article
Regarding your question about toasters and ovens you asked:
-------
If I expect the toaster to exhibit Y, am I claiming simply that the
same laws of physics are involved for both toasters and ovens, or am I
claiming that toasters should somehow behave as ovens?
-------
The first one, you are "claiming simply that the same laws of physics
are involved for both toasters and ovens". Something built like a
toaster will not behave as an oven if it follows the same laws of
physics as an oven, because it is built differently.
I'll break what I put as step 4 down into step 4 and step 5.
Step 4: 'behaviour X' will be defined as the resultant behaviour of a
robot build following the same laws of physics as things that it isn't
anything to be like.
Given the clarification are you ok with step 4?
Ymir
<ce1c51cb-fd3d-4ddc...@p28g2000vbn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> > �someone2 <glenn.spig...@btinternet.com> wrote:
No.
I'd have been fine with the above if you had rephrased it simply as
"they'd expect the robot to follow the laws of physics."
The problem lies in this part:
> Those believing in the suggestion put forward in
> step 2 would expect behaviour X (given the robot build) if reality was
> that it wasn't like anything to be the robot.
Here, you're not merely claiming that the robot will follow the same
laws of physics as a toaster or a microwave oven. You are claiming that
your theorist expects it to have the behaviour of a non-conscious robot,
*regardless* of whether it is conscious or not (since both options were
left open).
That would be like saying that I would expect behaviour Y (toaster-like
behaviour) of my hypothetical toaster if reality was such that the
toaster was actually a microwave. Which is of course nonsense.
Either the robot is or isn't conscious. According to physicalism, this
would be determined by its construction. Both construction C (a
construction associated with a conscious robot (or, if you prefer, the
construction associated with a human) and construction U (one associated
with an unconscious robot) will follow the same laws of physics, but
that does not entail that they will behave the same way any more than a
toaster will be expected to behave in the same way as a microwave.
By remaining noncommital regarding whether the robot is conscious or
not, you are basically saying that this robot might have construction C
or it might have construction U. Behaviour X is thus the behaviour
associated with a robot whose construction has not been identified as
being either C or U, but you then go on to state that that we will
expect behaviour X *if reality were such that it had construction U*. So
in actuality, you are committing your theorist to the view that the
robot is not conscious from the get-go.
Andr�
someone2
> In article
You didn't seem to follow that step 4 is now simply
----------------
Step 4: 'behaviour X' will be defined as the resultant behaviour of a
robot build following the same laws of physics as things that it isn't
anything to be like.
----------------
Are you ok with step 4?
If you are, then
----------------
Step 5: If for a particular robot construction behaviour X would be
for it to pass the Turing Test, then those believing the suggestion
put forward in step 2 consider that reality would either be that would
be like something to be the robot, or that it wouldn't.
----------------
Are you ok with step 5?
[It is quite similar to step 3:
Step 3: For a robot that passes the Turing Test, those believing the
suggestion put forward in step 2, consider that either it would be
like something to be the robot, or that it wouldn't]
Ymir
<08aa4708-64ea-45e1...@r36g2000vbn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
From a physicalist standpoint, there is only one set of physical laws
which governs both conscious and non-conscious entities, so the above
could simply be stated as 'behaviour X will be defined as the resultant
behaviour of a (particular) robot determined by the laws of physics).
But, that wasn't my objection. It was to the below:
> > > Those believing in the suggestion put forward in
> > > step 2 would expect behaviour X (given the robot build) if reality was
> > > that it wasn't like anything to be the robot.
Saying something follows the same physical laws as nonconscious entities
IS NOT the same thing as saying that something is a nonconscious entity.
Given a particular construction of a robot, behaviour X would be the
behaviour which the laws of physics demand for that particular
construction. It would not be the behaviour which the laws of physics
demanded if reality was such that the robot wasn't conscious, because
that presupposes that the build in question isn't conscious, which is
begging the question.
As stated earlier, if a toaster's behaviour can be described using the
same laws of physics used to describe a microwave, and 'behaviour Y' is
the behaviour which the laws of physics determine for a toaster, we
would not expect behaviour Y if reality was such that the toaster was
actually a microwave (whatever that means). We expect toasters and
microwaves to behave entirely differently despite the fact that the same
laws of physics are involved. Similarly, we expect a robot whose
construction leads to consciousness to behave differently from one that
doesn't despite the fact that the same laws are involved. If the robot
in question is conscious, and behaviour X is the behaviour which the
laws of physics predict for that particular construction, we certainly
don't expect the behaviour to be the same in a reality in which it
wasn't conscious (which would entail either a different construction, or
a different set of physical laws at work in that alternate reality).
If the theorist doesn't accept the possibility of a conscious robot, we
can rephrase the above in terms of a robot and a human. Since you claim
the theorist is a physicalist, they accept that the same physical laws
govern the behaviour of both humans and robots. If Y is the behaviour
which they expect from a human given those laws, they certainly wouldn't
expect that behaviour if reality were such that the human were actually
a robot (however you want to interpret such a counterfactual).
I'm raising this despite your new breakdown of your step into 4 and 5
because the same odd leap has occurred in many of your posts (the leap
from 'following the same laws' to 'behaving the same way'. I want to
make sure you understand this objection and see the problem inherent in
the leap you made in your original formulation so you will avoid
sneaking the same error in at subsequent points in your argument.
So now lets turn to your new formulation:
> ----------------
> Step 4: 'behaviour X' will be defined as the resultant behaviour of a
> robot build following the same laws of physics as things that it isn't
> anything to be like.
> ----------------
>
> Are you ok with step 4?
As stated above, this is unnecessarily wordy. Can we simply rephrase it
as I did above?
> If you are, then
>---------------- Step 5:
> If for a particular robot
> construction behaviour X would be for it to pass the Turing Test, then
> those believing the suggestion put forward in step 2 consider that
> reality would either be that would be like something to be the robot, or
> that it wouldn't.
> ----------------
>
> Are you ok with step 5?
Step 5 is a tautology, so it doesn't really add any information to your
argument, unless you plan on having the claim regarding X above to carry
through to your subsequent steps. In which case we can reformulate the
above as:
Step 5: The behaviour of the robot under consideration (which follows
from the laws of physics) satisfies the Turing Test.
The disjunction about the person either accepting it as conscious or not
is analytically true, so we can dispense with that altogether.
Are you OK with these reformulations?
Andr�
someone2
> In article
Ok, you are bringing in bits to complain about which aren't even in
the step seems like an attempt to obfuscate the issue. Though I'll
give you the benefit of the doubt and assume for the second reply in a
row, you weren't able to follow what step 4 was, and need extra help.
----------------
Step 4: 'behaviour X' will be defined as the resultant behaviour of a
robot build following the same laws of physics as things that it isn't
anything to be like.
----------------
You have said:
----------------
From a physicalist standpoint, there is only one set of physical laws
which governs both conscious and non-conscious entities, so the above
could simply be stated as 'behaviour X will be defined as the
resultant behaviour of a (particular) robot determined by the laws of
physics).'
----------------
This isn't strictly true, as there are physicalist perspectives such
as property dualism as mentioned. I understand that you might have
issues with property dualism as being physicalist, though if they
aren't claiming anything other than the physical exists they are
physicalists. Anyway, to avoid any confusion, I have worded it so as
to cut out physicalist accounts such as property dualism.
In case you aren't clear on what step 4 is, it is written again below:
someone2
> In article
Ignore my previous post, as I hadn't fully read the bit where you were
talking about your objection to text that wasn't in step 4 or step 5.
I had assumed you hadn't followed that the text you were objecting to
wasn't in either of them.
As for the reformulations of steps 4 and 5, no I'll keep them as they
are.
----------------
Step 4: 'behaviour X' will be defined as the resultant behaviour of a
robot build following the same laws of physics as things that it isn't
anything to be like.
----------------
----------------
Step 5:
If for a particular robot construction behaviour X would be for it to
pass the Turing Test, then those believing the suggestion put forward
in step 2 consider that reality would either be that would be like
something to be the robot, or that it wouldn't.
----------------
Though in case I have misunderstood, are you ok with step 4 and 5.
Please don't obfuscate the issue, by bringing in text which isn't in
steps 4 and 5 and saying you have objections to that.
Eric Root
It is rude and dishonest of you to characterize Ymir's efforts to UN-
obfuscate your prose, as obfuscations.
Eric Root
someone2
It obfuscates that he has no objections to points 4 and 5, if the
majority of his post is an objection, but to text he brought in from
previous posts, which wasn't in either point 4 or 5. Such that I
responded incorrectly firstly, and had to repost. It also contributes
to why I have had to ask him again to confirm that he didn't have any
objections to points 4 and 5 (the other factor being that although he
didn't object to points 4 and 5, he offered suggestions as to how they
could be worded, and since I am not taking those suggestions, I need
him to confirm that he wasn't going to later claim that he hadn't
objected to what was said without his suggestions). So you are wrong.
someone2
(Assuming that you were considering Ymir adding in text from previous
posts and objecting to what it said an attempt to un-obfuscate my
prose. Though I'm not quite clear on why you'd view it as that.)
Ymir
<bef74f3f-b3ba-4a7b...@d34g2000vbm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> Ignore my previous post, as I hadn't fully read the bit where you were
> talking about your objection to text that wasn't in step 4 or step 5.
> I had assumed you hadn't followed that the text you were objecting to
> wasn't in either of them.
It was in your original formulation.
> As for the reformulations of steps 4 and 5, no I'll keep them as they
> are.
>
> ----------------
> Step 4: 'behaviour X' will be defined as the resultant behaviour of a
> robot build following the same laws of physics as things that it isn't
> anything to be like.
> ----------------
>
> ----------------
> Step 5:
> If for a particular robot construction behaviour X would be for it to
> pass the Turing Test, then those believing the suggestion put forward
> in step 2 consider that reality would either be that would be like
> something to be the robot, or that it wouldn't.
> ----------------
>
> Though in case I have misunderstood, are you ok with step 4 and 5.
Yes, though I still don't understand why you insist on making them
unnecessarily wordy. Please present your next steps.
Andr�
someone2
> In article
Step 6:
Whether or not reality was that it was like something to be the robot
construction mentioned in step 5 those believing the suggestion put
forward in step 2 would expect it to pass the Turing Test.
Ymir
<56465df4-5eed-4325...@z34g2000vbl.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
Not necessarily. You've stated that 'behaviour X' (which includes
passing the Turing test) is the behaviour which the laws of physics
would determine for this robot. Whether the theorist is able to predict
this behaviour is another matter. You've said that this theorist claims
that only organic brains can be conscious, but you haven't provided any
details on the model which they use to reach this conclusion.
Even if our understanding of the fundamental laws of physics (presumably
the ones which determine behaviour X) were complete and correct, we
normally cannot use these laws to predict the behaviours of complex
systems -- there are simply too many variables involved. Instead, we
employ models which attempt to approximate more complex systems as best
we can. Whatever model your hypothetical identity theorist is using may
or may not approximate physical law to the extent necessary to make
predictions regarding the robot.
Given that you've implied that the identity theorist doesn't attribute
consciousness to the robot (since it lacks a human brain), then I would
assume that either the theorist rejects the Turing test as a reliable
indicator of consciousness (in which case it is incumbent on them to
provide some alternative test), or their model in fact does not predict
that the robot will exhibit behaviour X (in which case their model makes
an incorrect prediction in this instance). What would not be consistent,
however, would be to assert that they simultaneously believe that (i),
non-biological entities cannot be conscious, (ii) that the turing test
is a valid test of consciousness, and (iii) that this (I am assuming)
non-biological robot will exhibit behaviour X which includes passing the
Turing test.
Andr�
someone2
> In article
You are incorrect to state that the theorists claim that it wouldn't
be like anything to be the robot brain. As mentioned all they are
saying is that if it was like something to be the robot brain, it
wouldn't be the same as being a human brain.
----------
Step 2: One physicalist suggests that all that exists is the physical,
and that what it is like to be human is an identity of the neuron
arrangement in the human brain. As such they suggest that if something
such as a robot doesn't have the biological neurons arranged in a
similar fashion, then either it wouldn't be like anything to be the
robot, or that it would be like something to be it, but it wouldn't be
the same as being a human. So did you understand what was being
suggested?
----------
That it might be like something to be the robot was reiterated in step
5:
----------------
Step 5:
If for a particular robot construction behaviour X would be for it to
pass the Turing Test, then those believing the suggestion put forward
in step 2 consider that reality would either be that would be like
something to be the robot, or that it wouldn't.
----------------
Also clearly from step 5 you can see that they weren't considering the
Turing Test to be a valid test of whether it was like something to be
the robot, since they consider that it might not be like something to
be the robot that passes the Turing Test.
Regarding you concern that they might not be able to predict how they
would expect the robot to behave, let me reword it for you:
Step 6:
Whether or not reality was that it was like something to be the robot
construction mentioned in step 5 those believing the suggestion put
forward in step 2 would, if they had the capability of predicting
behaviour X, expect it to pass the Turing Test.
Are you ok with step 6 now?
Ymir
<bc33ad8b-524e-45cb...@q14g2000vbi.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
OK -- that's fine then.
No.
Once you give the theorist effective omniscience with respect to making
behavioural predictions, the above can only be correct if the Turing
Test is not a valid indicator of consciousness (regardless of whether
the researcher accepts it as such or not).
By stating 'whether or not reality was that it was like something to be
the robot', you are implying that the turing test would be passed even
in cases where the robot isn't conscious, which would indicate that
behaviour X is unrelated to consciousness.
Thus, unless you have an argument that the Turing Test isn't a
legitimate test, six cannot be maintained.
Andr�
someone2
> In article
Certainly the Turing Test would either be a reliable indicator of
whether the robot was consciously experiencing or not, and certainly
the robot would be consciously experiencing or not. There is no
assertion or argument as to whether the Turing Test is reliable or
not.
Though step 6:
----------------
Whether or not reality was that it was like something to be the robot
construction mentioned in step 5 those believing the suggestion put
forward in step 2 would, if they had the capability of predicting
behaviour X, expect it to pass the Turing Test.
----------------
it is simply stating whether reality was be that it was like something
to be the robot construction mentioned in step 5 (as would be the case
if the Turing Test was a reliable test), or whether reality was that
it was not like something to be the robot construction mentioned in
step 5 (as might be the case if the Turing Test wasn't a realiable
test) they would still be expecting it to pass the Turing Test.
To write it another way; they consider reality to either be scenario
A, it is like something to be the robot, or scenario B, it isn't like
something to be the robot. They expect it to pass the Turing Test
whether reality was scenario A or whether reality was scenario B.
Do you see that step 6 is ok, regardless of whether the Turing Test is
reliable or not?
Ymir
<ba408635-8b8d-4b42...@g1g2000vbr.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
No, because *if* the Turing Test is reliable, then it *cannot* be the
case that the robot will pass the test and yet not be conscious.
However, you are asking us to contemplate two scenarios: one in which
the robot is conscious and one in which the robot isn't conscious. In
the latter case, the robot will not pass the turing test. It's
construction must differ from the robot in the first scenario in a way
which causes it not to be conscious, and your omniscient
researcher,using the same laws of physics as used in the first scenario,
isn't going to predict that it will exhibit behaviour X because it won't
exhibit behaviour X.
Andr�
someone2
> In article
There is no suggestion that reality is other than as it is. It is
*either* that it is like something to be the robot or it isn't.
Nor is the question about two different robot constructions, one
construction which it was something to be like, and one construction
which it wouldn't. For that in itself would be begging the question
that there could be a robot construction it would be something to be
like. Certainly there is no question of two identical robot
constructions, and it being like something to be one, and yet it not
being like something to be the other.
So given the clarification here, please don't try to make out that any
such things are being suggested, in order to be able to object to
them, for that would simply be creating strawmen.
You seemed to suggest looking at it from the basis of "*if* the Turing
Test is reliable". I've outlined the conditional statements below.
R represents reality for the robot.
if (the universality assumption is correct and behaviour X would be
for the robot to pass the Turing Test)
{
if (the Turing Test is reliable)
{
it is like something to be the robot and it is behaving as
they would expect.
}
else (the Turing Test is not reliable)
{
if(R = it is like something to be the robot)
{
it is like something to be the robot and it is behaving as
they would expect.
}
else(R = it isn't like something to be the robot)
{
it is not like something to be the robot and it is
behaving as they would expect.
}
}
}
Whereas point 6 simplifies the issue with the following conditional
statements.
if (R = it is like something to be the robot)
{
they expect it to pass the Turing Test.
}
else (R = it isn't like something to be the robot)
{
they expect it to pass the Turing Test.
}
If the Turing Test is reliable, the condition of the first if
statement would be met, and the second condition would never be
reached. If the Turing Test wasn't reliable the first condition might
or might not be met, and if it wasn't the second condition would be.
There is no assertion as to whether the first condition or the second
condition is correct.
Ymir
<8be9745a-8d0a-40a5...@f37g2000yqn.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
Except that that isn't what your step 6 (repeated below) says.
> > > Whether or not reality was that it was like something to be the robot
> > > construction mentioned in step 5 those believing the suggestion put
> > > forward in step 2 would, if they had the capability of predicting
> > > behaviour X, expect it to pass the Turing Test.
The above to me reads as
((if the robot is conscious then they expect it to pass the turing test)
AND
(if the robot is not conscious then they expect it to pass the turing
test))
Since the second conjunct would be false if the turing test is reliable,
the entire conjunction will also be false if the turing test is reliable.
Therefore, accepting your step six is only possible if the turing test
is not reliable. You have not established this.
Andr�
someone2
> In article
You seem to be ignoring that the robot is the one referenced in step
5, in which behaviour X is for it to pass the Turing Test.
I have written out the conditions more fully below, including the
reference in step 5 to behaviour X being that it will pass the Turing
Test.
if (behaviour X would be for the robot to pass the Turing Test)
{
Ymir
<3ac4b723-f92b-4124...@c14g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I'm not ignoring this at all.
If a robot is not conscious and the turing test is a valid test of
consciousness, then the laws of physics will not lead to behaviour X,
since behaviour X could only occur in something whose construction
generates consciousness. If they did lead to behaviour X, the turing
test would not be a valid test.
> I have written out the conditions more fully below, including the
> reference in step 5 to behaviour X being that it will pass the Turing
> Test.
>
> if (behaviour X would be for the robot to pass the Turing Test)
> {
> if (R = it is like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> else (R = it isn't like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> }
That'e exactly the same thing as what I wrote above. My objection to it
remains.
Andr�
someone2
> In article
In your previous post you had said you read it like:
----------
((if the robot is conscious then they expect it to pass the turing
test)
AND
(if the robot is not conscious then they expect it to pass the turing
test))
----------
And had objected that if the Turing Test was reliable, then if it
wasn't like anything to be the robot, they wouldn't expect it to pass
the Turing Test, as behaviour X being to pass the Turing Test would
only be for robots it was like something to be. Such that it would be
the case that if the robot wasn't conscious, then it wouldn't have
behaviour X being to pass the Turing Test, so it would be false to
claim that would expect such a robot to pass the Turing Test.
I was simply pointing out, that there is no such false claim, and
thought that you might have not realised that the conditional
statement only applies to a robot which has behaviour X being to pass
the Turing Test. As such it is wrong claim that it could be
interpreted as a false claim being made about a robot doesn't have
behaviour X being for it pass the Turing Test. I had assumed you had
ignored that the robot is the one referenced in step 5, in which
behaviour X is for it to pass the Turing Test. Though you claim this
wasn't why you made the mistake.
You say:
--------
If a robot is not conscious and the turing test is a valid test of
consciousness, then the laws of physics will not lead to behaviour X,
since behaviour X could only occur in something whose construction
generates consciousness.
--------
All that would mean is that the first if condition, that behaviour X
was for the robot to pass the Turing Test, would not be met. So what?
if (behaviour X would be for the robot to pass the Turing Test)
{
if (R = it is like something to be the robot)
{
they expect it to pass the Turing Test.
}
else (R = it isn't like something to be the robot)
{
they expect it to pass the Turing Test.
}
}
There no assertions as to whether reality meets any of the conditions,
as such there are no assertions that a particular condition will ever
be met.
Ymir
<f7eecf99-6e8a-49f7...@w41g2000yqb.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
If the first if statement is not met below, then the second one *will*
be met. So in that instance you are stating that they will expect it to
pass the turing test when in fact it will not. However, you previously
stated that they were in operating from a position in which they could
accurately predict whether behaviour X would be realised, which leads to
a contradiction.
> if (behaviour X would be for the robot to pass the Turing Test)
> {
> if (R = it is like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> else (R = it isn't like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> }
>
> There no assertions as to whether reality meets any of the conditions,
> as such there are no assertions that a particular condition will ever
> be met.
But one of them must always obtain since the two conditional exhaust all
of the logical possibilities.
Are you simply trying to say that the turing test may or may not be a
valid test? If that's your point why not simply state it as such?
Andr�
someone2
> In article
When I said:
----------
All that would mean is that the first if condition, that behaviour X
was for the robot to pass the Turing Test, would not be met. So what?
----------
You replied:
----------
If the first if statement is not met below, then the second one *will*
be met.
-----------
There is no statement that will be met if the first if statment
(behaviour X would be for the robot to pass the Turing Test) is not
met.
Ymir
<5d8d1c86-7ab3-459b...@c1g2000yqi.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
I meant if the following statement is not met, then the else statement
which follows it will be met. I ignored the above line since you'd
already *defined* X as passing a turing test so it doesn't serve any
purpose.
> if (R = it is like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> else (R = it isn't like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
You're saying here that if it turns out that it isn't the case that the
robot is conscious, then they will still expect it to pass the turing
test. I'm saying that that is not possible unless the turing test is not
a valid test. But you've already stated that your theorist isn't
necessarily denying the validity of that test. Therefore, there's no
possible justification for the content of your else clause.
Andr�
someone2
> In article
If the Turing Test was reliable, it will never enter the else clause.
The content of the clause can only be reached, and therefore is only
applicable if the Turing Test wasn't reliable
[I pretty much explained this to you a few posts ago, but you snipped
and ignored it.
-------------------------------
In your previous post you had said you read it like:
----------
((if the robot is conscious then they expect it to pass the turing
test)
AND
(if the robot is not conscious then they expect it to pass the turing
test))
----------
And had objected that if the Turing Test was reliable, then if it
wasn't like anything to be the robot, they wouldn't expect it to pass
the Turing Test, as behaviour X being to pass the Turing Test would
only be for robots it was like something to be. Such that it would be
the case that if the robot wasn't conscious, then it wouldn't have
behaviour X being to pass the Turing Test, so it would be false to
claim that would expect such a robot to pass the Turing Test.
I was simply pointing out, that there is no such false claim, and
thought that you might have not realised that the conditional
statement only applies to a robot which has behaviour X being to pass
the Turing Test. As such it is wrong claim that it could be
interpreted as a false claim being made about a robot doesn't have
behaviour X being for it pass the Turing Test. I had assumed you had
ignored that the robot is the one referenced in step 5, in which
behaviour X is for it to pass the Turing Test. Though you claim this
wasn't why you made the mistake.
-------------------------------]
Ymir
<e4f8f586-c712-4e07...@d32g2000yqh.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
Perhaps, then, you could offer a scenario in which the else clause would
apply. That might help clarify exactly what you are trying to say above.
The problem I'm having following this is that you are adding in many
seemingly pointless clauses to your various steps (i.e. clauses which
are tautological and thus contribute nothing to the argument, e.g.
'either it would be like something to be the robot or it wouldn't). I'm
then stuck trying to figure out what purpose those clauses serve, and
end up constantly second-guessing whatever I *think* you might be saying
while trying to find some relevance to those clauses.
In the above case, I don't quite understand why you insist on having
your if...else clauses rather than simply replacing it with 'they expect
it to pass the turing test' since they would mean the same thing.
That would leave us with
if (behaviour X would be for the robot to pass the Turing Test)
Then they would expect it to pass the turing test.
What exactly is it that you are trying to convey with these additional
clauses?
Andr�
someone2
> In article
If the Turing Test was reliable it would never reach the else clause,
if it wasn't it might.
The purpose of the extra clauses will become apparent as we move onto
the later steps.
So are you ok with step 6 now?
----------------
Whether or not reality was that it was like something to be the robot
construction mentioned in step 5 those believing the suggestion put
forward in step 2 would, if they had the capability of predicting
behaviour X, expect it to pass the Turing Test.
----------------
Ymir
<b66fc031-b5ec-46c3...@g10g2000yqh.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
In that case, it would be better to introduce them during those later
steps rather than now.
> So are you ok with step 6 now?
> ----------------
> Whether or not reality was that it was like something to be the robot
> construction mentioned in step 5 those believing the suggestion put
> forward in step 2 would, if they had the capability of predicting
> behaviour X, expect it to pass the Turing Test.
> ----------------
No. I am not OK with it. If your argument constitutes a rejection of the
validity of the turing test, then state so up front. If it doesn't
constitute such a rejection, then why talk about 'whether reality was
like something to be the robot'?
I would accept the following, which thus far is the only meaningful
interpretation I can make of your convoluted argument:
Step 6: If a given robot passes the turing test, then it is conscious
provided the turing test is a valid test of consciousness. If the turing
test is not a valid test of consciousness, then the question remains
open as to whether or not a robot passing this test is conscious.
Andr�
someone2
> In article
It doesn't make any assertions either way whether the Turing Test is
reliable or not. If you are claiming it does, point out where. If you
aren't claiming it does, then what is your objection, its lack of
assertions?
Ymir
<cf64d008-0b1b-473d...@e27g2000yqm.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> >
> > �someone2 <glenn.spig...@btinternet.com> wrote:
Nor does the formulation I give above.
> If you are claiming it does, point out where. If you
> aren't claiming it does, then what is your objection, its lack of
> assertions?
>
> if (behaviour X would be for the robot to pass the Turing Test)
> {
> if (R = it is like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> else (R = it isn't like something to be the robot)
> {
> they expect it to pass the Turing Test.
> }
> }
Look. Consider the following analogous argument. Let's say we have a
theorist who, like yours, has the infallible ability to predict whether
some behaviour will occur, call it Y. (using your pseudocode -- btw
pseudocode is more appropriate for representing algorithms rather than
declarative statements).
If (behaviour Y is for a particular solution to cause an acid indicator
to turn red)
{ if (R = the solution is an acid)
{
they expect the solution to turn red.
}
else (R = the solution is not an acid)
{
they expect the solution to turn red.
}
}
Do you accept the above? If so, please explain to me what it is actually
supposed to convey, and more importantly, what the point of the 'else'
clause is in the above statement? Most people, myself included, don't
accept the polarity in that else clause. They're not going to expect the
same behaviour from the solution if it turns out to be an acid as they
would if it turns out to be a non-acid.
Andr�
Inez
> In article
As an aside, I'm not sure he's proved that to be the case. I don't
know that "what it is to be like" is rigorously enough defined to
definitely categorize something one way or another. A human is
conscious, a rock isn't, but what about a dog? Or an earthworm? Or a
ficus? Where is the sharp line below which things are 100% not
conscious and above which they are 100% conscious?
someone2
> In article
It isn't that the people were thought to have an infallible ability to
predict by guessing. It was simply that since it is a thought
experiment, we can consider that they can work out how the build would
be expected to behave, if it followed the same laws of physics as
things that it isn't anything to be like. All us considering them not
being able to work it out would do would enable us to avoid
considering the implications, and why would we want to do that?
You analogy isn't analogous. Not only would they know the reason they
expected the non-acid solution to turn the acid indicator red, but the
reason they expected the acid indicator to have turned red, would have
told them R. Which isn't the case in the scenario you were trying to
find an analogy to.
I had said:
-------------
It doesn't make any assertions either way whether the Turing Test is
reliable or not.
-------------
To which you replied:
-------------
Nor does the formulation I give above.
-------------
So you admit that it doesn't make any assertions either way, whether
the Turing Test is reliable or not. So what is your objection, its
lack of assertions? The reason I ask is that it seems like you would
like me to assert that the Turing Test is an indicator of whether it
is like something to be the robot or not.
Ymir
<9f708146-f457-49b7...@j21g2000yqe.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
> if it followed the same laws of physics as
> things that it isn't anything to be like. All us considering them not
> being able to work it out would do would enable us to avoid
> considering the implications, and why would we want to do that?
>
> You analogy isn't analogous. Not only would they know the reason they
> expected the non-acid solution to turn the acid indicator red, but the
> reason they expected the acid indicator to have turned red, would have
> told them R. Which isn't the case in the scenario you were trying to
> find an analogy to.
So why do you think consciousness is any different from this? If your
theoretician has the ability to completely predict whether the robot
will exhibit X, it seems likely they would also understand the reason
and thus understand R.
You've been asked similar questions on a variety of occasions. You've
presented an argument, and someone has presented an argument which
follows a formally identical structure, and you've dismissed it by
stating that the analogy doesn't hold -- but if two arguments are
formally identical, either they are both valid or they are both invalid.
You're claim that consciousness is somehow different rests on a hidden
premise which, as far as I can tell, is simply that consciousness
doesn't follow from the laws of physics. But that is also the conclusion
which you purport to be trying to show, so you cannot incorporate this
into your argument as a buried assumption or you are simply begging the
question.
>
> I had said:
> -------------
> It doesn't make any assertions either way whether the Turing Test is
> reliable or not.
> -------------
>
> To which you replied:
> -------------
> Nor does the formulation I give above.
> -------------
>
> So you admit that it doesn't make any assertions either way, whether
> the Turing Test is reliable or not. So what is your objection, its
> lack of assertions? The reason I ask is that it seems like you would
> like me to assert that the Turing Test is an indicator of whether it
> is like something to be the robot or not.
My objection is simply that I can't figure out a coherent interpretation
of your claim -- getting back to my acid analogy, putting aside for the
moment whether it is an appropriate analogy, can you explain to me what
the claim regarding acids actually *means*? What purpose does the
else-clause serve given that it has the same polarity as the if-clause
it follows. I can't interpret this statement regarding acids any more
than I can interpret your statement regarding passing the turing test.
I'm not looking for a statement concerning whether you agree with my
statement regarding the acid test. I'm asking you to explain what it
actually *says* -- since it follows the same structure as your argument,
that might clarify for me what it is that you are actually claiming,
since I cannot come up with a coherent interpretation of your claim.
If you can provide this clarification and explain the acid test such
that it actually makes some sense and isn't contradictory, then fine. If
you cannot provide a non-contradictory interpretation of this, then
please explain how your consciousness claim isn't equally contradictory
without relying on any tacit premises regarding the nonphysicality of
consciousness.
Andr�
someone2
> In article
You say:
-----------
If your theoretician has the ability to completely predict whether the
robot will exhibit X, it seems likely they would also understand the
reason and thus understand R.
-----------
They expect that the build would be following certain laws, but they
don't know which resultant behaviours it would be like something to
be. So simply knowing the laws doesn't entail that they would know
whether R was that it was like something to be the robot or not. There
is no premise that what it is like to be the robot doesn't follow from
the laws of physics. It is simply that if the universality assumption
were correct there would be no scientific experiment to distinguish
whether those in step 2 were correct in their suggestion of what
physical activity it would be like something to be, or functionalists
such as those putting forward theory B in the paper, or those putting
forward theory A in the paper. So while I not using any premise that
what it is like to be the robot doesn't follow from the laws of
physics, I am pointing out that there would be no scientific way of
discovering how it followed if the universality assumption were true.
You said:
----------
My objection is simply that I can't figure out a coherent
interpretation of your claim -- getting back to my acid analogy,
putting aside for the moment whether it is an appropriate analogy, can
you explain to me what the claim regarding acids actually *means*
----------
In the acid case, if they were expecting behaviour Y, then since being
an acid is defined in terms of behaviour, the expectations that they
worked out would determine their expectation of R. There would to
their knowledge be only one potential value to R given behaviour Y.
With the robot, assuming the universality assumption was true, but us
being willing to take a leap an consider that they knew which
behaviours were associated with it being like something to be the
robot, then they too could define what it was like to be somthing as a
behaviour. Whatever behaviour occured would have determined their
expectation of R. There would to their knowledge be only one potential
value to R given behaviour X.
However with the robot situation you are being stepped through, since
they are not omniscient, and as mentioned couldn't scientifically
establish R with regards to the robot if the universality assumption
were true, to their knowledge R being that the robot was consciously
experiencing, or wasn't consciously experiencing cannot be ruled out.
Given their knowledge, their expectation of R hasn't been determined
by the behaviour. As far as they know, R could be that it was like
something to be the robot, or R could be that it wasn't like something
to be the robot. There would to their knowledge be two potential
values of R given behaviour X.
They can be asked a question about each of the potential values of R.
They can be asked the question *if* reality was that the robot was
consciously experiencing, would you expect it to pass the Turing Test.
Likewise the question *if* reality was that the robot wasn't
consciously experiencing, would you expect it to pass the Turing
Test.
Would you object to them being asked about what are potential values
of R, or object to them answering?
Ymir
<76e9d299-9949-4963...@d32g2000yqh.googlegroups.com>,
someone2 <glenn....@btinternet.com> wrote:
That there is no scientific experiment is what you purport to be showing
with your argument, but you haven't gotten there yet. So you can't use
that conclusion to justify your premises.
> , or those putting
> forward theory A in the paper. So while I not using any premise that
> what it is like to be the robot doesn't follow from the laws of
> physics, I am pointing out that there would be no scientific way of
> discovering how it followed if the universality assumption were true.
By your criteria, there would also be no scientific way of discovering
it if the universality assumption were false.
>
> You said:
> ----------
> My objection is simply that I can't figure out a coherent
> interpretation of your claim -- getting back to my acid analogy,
> putting aside for the moment whether it is an appropriate analogy, can
> you explain to me what the claim regarding acids actually *means*
> ----------
>
>
> In the acid case, if they were expecting behaviour Y, then since being
> an acid is defined in terms of behaviour,
An acid is a proton donor. We can't directly observe this anymore than
we can directly observe the brain activities which define consciousness.
We can only infer it from more indirect observations. There's no
difference between the acid case and the consciousness case.
> the expectations that they
> worked out would determine their expectation of R. There would to
> their knowledge be only one potential value to R given behaviour Y.
> With the robot, assuming the universality assumption was true, but us
> being willing to take a leap an consider that they knew which
> behaviours were associated with it being like something to be the
> robot, then they too could define what it was like to be somthing as a
> behaviour. Whatever behaviour occured would have determined their
> expectation of R. There would to their knowledge be only one potential
> value to R given behaviour X.
Again, given a sufficiently detailed theory of consciousness, there's no
reason to assume it would be any different from this.
> However with the robot situation you are being stepped through, since
> they are not omniscient, and as mentioned couldn't scientifically
> establish R with regards to the robot if the universality assumption
> were true,
Again, you're begging the question. I don't accept this, and since you
have yet to get to the point in the argument where you actually
demonstrate this claim about the universality assumption, you cannot use
it now to justify your premises.
> to their knowledge R being that the robot was consciously
> experiencing, or wasn't consciously experiencing cannot be ruled out.
> Given their knowledge, their expectation of R hasn't been determined
> by the behaviour. As far as they know, R could be that it was like
> something to be the robot, or R could be that it wasn't like something
> to be the robot. There would to their knowledge be two potential
> values of R given behaviour X.
>
> They can be asked a question about each of the potential values of R.
> They can be asked the question *if* reality was that the robot was
> consciously experiencing, would you expect it to pass the Turing Test.
> Likewise the question *if* reality was that the robot wasn't
> consciously experiencing, would you expect it to pass the Turing
> Test.
> Would you object to them being asked about what are potential values
> of R, or object to them answering?
I object to your claim that they would give the *same* answer for
different values of R.
Andr�
someone2
> In article
The argument isn't purporting to show that if functionalism were true
that there would be no scientific experiment to distinguish between
the suggestion in step 2 and functionalism, that is simply the case.
They would both always predict the mechanism to follow the same laws
of physics that can be used to describe things that aren't consciously
experiencing. If you could suggest a scientific experiment where one
of them wouldn't expect behaviour X then you would have shown me to be
wrong. If they both would always expect behaviour X then how can you
experimentally distinguish between them?
There is a difference in the acid case and the consciousness case. The
donation of a proton is a known behaviour. Whereas even if the
universality assumption were true, the physical activity which it is
like something to be, isn't known. You are trying to bring in
assertions that whether it was like something to be the robot can be
inferred from its behaviour, which is begging the question.
I pointed out that
------------
to their knowledge R being that the robot was consciously
experiencing, or wasn't consciously experiencing cannot be ruled out.
Given their knowledge, their expectation of R hasn't been determined
by the behaviour. As far as they know, R could be that it was like
something to be the robot, or R could be that it wasn't like something
to be the robot. There would to their knowledge be two potential
values of R given behaviour X.
They can be asked a question about each of the potential values of R.
They can be asked the question *if* reality was that the robot was
consciously experiencing, would you expect it to pass the Turing Test.
Likewise the question *if* reality was that the robot wasn't
consciously experiencing, would you expect it to pass the Turing Test.
Would you object to them being asked about what are potential values
of R, or object to them answering?
------------
To which you replied:
------------
I object to your claim that they would give the *same* answer for
different values of R.
------------
There is only one value of R for the robot. They don't know what it is
though, thus there would to their knowledge be two potential values of
R given behaviour X for the robot. Though only one actual value of R.
So it isn't a question about different actual values of R.
They can be asked a question about what are to them potential values
of R. They can be asked the question *if* reality was that the robot
was consciously experiencing, would you expect it to pass the Turing
Test. Likewise the question *if* reality was that the robot wasn't
consciously experiencing, would you expect it to pass the Turing Test.
Would you object to them being asked about what are to them potential
values of R, or object to them answering, or misundertand the question
to avoid answering?