Mitch
-----Original Message-----
From: Bruno Marchal <[email protected]>
To: everything-list <[email protected]>
Sent: Tue, Sep 16, 2014 12:21 pm
Subject: MGA revisited paper + supervenience
Hi Russell, Hi Others,
Sorry for the delay. Some comments on your (Russell) MGA paper appear
below.
On 25 Aug 2014, at 00:30, Russell Standish wrote:
> On Sun, Aug 24, 2014 at 01:22:51PM -0700, meekerdb wrote:
>> On 8/24/2014 12:55 AM, Russell Standish wrote:
>>>>
>>> I don't think that can be the case. I don't see how it can be
>>> anything
>>> to be like a tree, yet trees are clearly DNA-based beings. So you
>>> would get skewed results if you were to reason as though you could
>>> be
>>> a tree.
>>
>> Exactly. It's a reductio on the pattern of argument you used to
>> prove ants can't be conscious. I used it to prove ants can't be
DNA
>> based.
>
> I don't understand. How is having DNA relevant to having
> consciousness? It is quite plausible that non-DNA-based forms are
> conscious (eg a computer running a suitable AI program), and that
> some DNA-based forms are not conscious (trees, for example).
I agree.
Incidentally, when you see the complexity of the interaction between
the roots of trees and the soils, chemicals and through bacteria, and
when you believe, as some experiences suggest, that trees and plant
communicate, I am not so sure if trees and forest, perhaps on
different time scale, have not some awareness, and a self-awareness
of some sort. (I take awareness as synonymous with consciousness,
although I change my mind below!).
>
>
>>
>>>
>>> The reference class cannot be larger than the class of conscious
>>> beings. Obviously it can be quite a bit smaller, but there must
be a
>>> maximal reference class for which anthropic reasoning is valid,
>>> although it is quite controversial what it is - some suggest it
may
>>> even be as small as those people capable of understanding the
>>> anthropic argument, a sizable fraction of which inhabits this
list!
>>
>> That's what bothers me. If you exclude ants because they're not
>> conscious (and I assume you've read "Godel, Escher, and Bach") and
>> hence can't understand the argument, why not exclude people who
>> can't understand the argument?
>>
>
> "Ant Fugue" is about the possibility that ant _colonies_ might be
> conscious. My argument has nothing to say about ant colonies, even
> though I consider "Ant Fugue" to be just an interesting speculation,
> rather than a serious claim about ant colonies.
I am a bit agnostic on this. But I have few doubt that individual ants
have some consciousness, though.
>
> Oh - perhaps you mean "can't understand the argument" as in
organisms
> that can't understand the anthropic argument must be excluded from
the
> reference class. This seems a rather implausible claim - just
because
> anthropic argument has not occurred to you yet, shouldn't really
> exclude you. The idea that self-awareness is a necessary requirement
> of the reference class is a perhaps more believable claim - in order
> to even
> think anthropically requires a concept of self - but then I'm still
> not sure what it even means to be conscious, but not self-aware.
What
> does it even mean to "be an amoeba", as Bruno seems to think is
> possible.
OK. I will make a try. Awareness in its most basic forms comes from
the ability to distinguish a good feeling from a bad feeling. The
amoeba, like us, knows (in a weak sense) that eating some paramecium
is good, but that hot or to cold place are bad, and this makes it
reacts accordingly with some high degrees of relative self-referential
correctness. The genome of the amoeba, which is really a collection of
cooperating many genomes (lot of "nucleus") is Turing universal or
"complete", and the amoeba incarnates it relatively to her (our)
probable lower substitution level (which defines by the FPI the
physical reality). So she get a life, a first person life, of some
sorts. Little consciousness, if you want, because from the first
person view of the amoeba it is the whole big thing. The life of
protozoans are similar to ours. They keep moving for eating, try to
avoid the possible predators, get sleepy (very deeply so) when it get
cold (the cell transforms into a sort of egg), and they really dislike
when being eaten, and try to avoid it instinctively, but with a
possible "bad" experience.
here an amoeba eats two paramecia: https://www.youtube.com/watch?v=pvOz4V699gk
Now, amoeba are universal, but not Löbian, and so they lack the Kp ->
KKp law, and are not self-aware. Nor do have them memories, or only
few one, so they live in the instant present, happy when eating,
unhappy when being eaten. At least they will not philosophize and be
unhappy when eating because they know they *might* be eaten, nor happy
when being eaten because they got the point that it is part of the
game of life and be serene about this, or because they believe in
christ or someone. You need to be Löbian to develop those form of
craziness. I think this came with lower invertebrate, like jumping
spiders and cuttlefishes. But they are lucky, their brain are not
enough big to develop much of the craziness. They probably live a
little bit less in the present, but still don't get the point of the
existential question.
To be aware is to feel the cold, the hot, the yummy, the acidity
level, and capable of interpreting it "self-referentially", and
reacting.
To be self-aware add the memories and one more reflexive loop (which
you get in RA when adding the induction axioms, leading to PA). As
long as you are correct, you obey the modal logic G and G* in that
case. But the 1p views obeys the intensional variants.
>
>
>
>>> But that smacks of parochialism, much like the notion of
>>> geocentrism. I just haven't found a convincing argument that the
>>> maximal reference class is not just the class of conscious
>>> organisms,
>>> of beings for whom there is a something it is like to be.
>>>
>>> But my question (which you haven't answered) is what you think
this
>>> maximal reference class is from your four part classification of
>>> consciousness.
>>
>> If I had to pick, I'd say it was those entities who were aware of
>> their own thoughts and had sufficient language to formulate
Bayesian
>> inference.
>>
>
> The Bayesian theory is a bit stringent don't you think. There are
> plenty of formulations of the doomsday argument that don't use
> Bayesian reasoning. Take Gott's version for example.
>
> Self-awareness, as I mentioned, is more defensible property. The
> question is whether non-self-aware consciousness (your koi) is a
> coherent concept.
I agree that to have awareness, you need a self, a third person self.
But that is well played by the relative body (actually bodies,
incarnate through the UD).
Maybe we should define consciousness by self-awareness, and then self-
consciousness would be the higher form of self-self-awareness? That
makes one "self" per reflexive loop.
>
> Attacks on anthropic reasoning will work better by choosing a
> reference class which is indisputably a subset of the reference
class,
> such as all human beings, and then demonstrating a contradiction. I
> thought I had come up with such an example with my "Chinese
paradox",
> but it turned out anthropic reasoning was rescued from that by the
> peculiar distribution of country population sizes that happens to
hold
> in reality. AR has proved remakably resilient to empirical tests.
I am still a bit agnostic for its use in the fundamentals, as the
probability, with computationalism, are always relative. It is the
same in quantum mechanics, where the probabilities are not on states,
but on relative states: they have always the form <a I b>^2, the
probability for finding b when being in the state a.
But we can extract useful information from the Anthropic principle,
and even from the most general Turing-thropic. Just saying that the
laws of physics should be a calculus of relative probabilities.
PS I have printed your MGA paper, and so read it and comment it
despite being in a busy period.
Let me say here, as we are in the good thread, two main points, where
we might have vocabulary issue, or perhaps disagree on something? So
you might think about this and be prepared :)
The first point concerns the relation between counterfactualness and
modal realism, that you link in a way which makes me a bit uneasy. I
do believe in some links between them, though, but it might not
correspond to yours. Examples will follow later.
The second point is the one we have already discussed, and concerns
the definition of supervenience. We do both agree on the Stanford
definition, but I am still thinking you are misusing it when apply to
the Alice and Bob in the classroom situation.
You agree that
C supervenes on B if to change C it is necessary to change B.
For example, consciousness C supervenes on a brain activity B, because
to change that consciousness you need to change that brain activity.
Now take the (physical) union of B and A: B-and-A.
If you change B, you automatically change B-and-A.
So if consciousness C supervenes on B, you need to make a change to B-
and-A (indeed to the B part), and so automatically C will supervenes
on the union of B and A.
So when Alice and Bob are in the classroom, we have that
- Alice's consciousness supervenes on Alice's brain activity
- Alice's consciousness supervenes on Alice + the entire room
(including Bob's brain activity)
- Bob consciousness supervenes on Bob's brain activity.
- Bob consciousness supervenes on Bob + the entire room (including
Alice's brain activity).
It is not a problem that both Bob's and Alice's consciousness
supervenes on the same classroom, as to make a change in either Alice
or Bob's consciousness, you need to make a change of the A+B system.
It is the same for UD*. My and your consciousness supervenes on UD*.
To make a change to my or your's consciousness here and now, we would
need to make the (impossible) change in the UD*.
Are you OK with:
(C supervenes on B) entails (C supervenes on B+D), with B+D being some
physical union of B and some C.
I have to go, sorry for the delays, and possible other delays.
September-october are particularly heavy this year, but we have all
the time, OK?
Best,
Bruno
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google
Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google
Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it,
send an email to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
