On 9 October 2017 at 16:12, Bruno Marchal <marc...@ulb.ac.be> wrote:
>
> On 09 Oct 2017, at 15:29, David Nyman wrote:
>
> On 5 October 2017 at 16:06, Bruno Marchal <marc...@ulb.ac.be> wrote:
>
>>
>> On 05 Oct 2017, at 16:02, David Nyman wrote:
>>
>> On 5 October 2017 at 13:50, Bruno Marchal <marc...@ulb.ac.be> wrote:
>>
>>>
>>> On 02 Oct 2017, at 15:07, David Nyman wrote:
>>>
>>>
>>>
>>> On 2 Oct 2017 1:58 p.m., "Bruno Marchal" <marc...@ulb.ac.be> wrote:
>>>
>>> Dear Evgenii,
>>>
>>>
>>> On 28 Sep 2017, at 21:30, Evgenii Rudnyi wrote:
>>>
>>>
>>>
>>>
>>> Long time ago you have discussed Maudlin's paper. At that time I somehow
>>> did not get interested. Yet, other day I have got strong feeling that I
>>> must read Maudlin's paper right now. I guess this could be explained by
>>> peculiarities of the universal dovetailer.
>>>
>>> Anyway, I have read Maudlin's paper, then I have read Hoffmanm's
>>> Sandman, and once more Maudlin's paper. I have enjoyed reading, the paper
>>> is nicely written. I guess I have understood the argument. Thank you.
>>>
>>>
>>>
>>> You are welcome. Don't hesitate to ask any further questions. There are
>>> slight differing nuances between the Movie Graph Argument (MGA) and
>>> Maudlin's argument. Also, I have found a way to cut the UDA at step 7
>>>
>>>
>>> More details please.
>>>
>>>
>>> Maudlin focuses on the counterfactual issue, which the MGA somehow
>>> avoids. Both argument shows the inadequacy of materialism and mechanism,
>>> but the MGA assumes that we have already a good idea of what a
>>> (mathematical) computation is, and that it is a special logical relation,
>>> not something in need of any physical assumption. Then Maudlin's analysis
>>> can be used to answer the "counterfactual objection", instead of reminding
>>> implicitly the logical nature of what is a computation. In the context of
>>> the UDA, the rôle of the MGA is only to show that the move in step 7
>>> consisting in assuming a small primary universe, is isomorphic to
>>> creationist last rebuttal when saying eventually that they agree with the
>>> evidences for evolution, but that God was needed to make that evolution
>>> possible. That was already clear with Peter Jones old objection to UDA that
>>> only a computation supported by primary matter can be conscious. That is a
>>> sort of magic way of thinking, by rebutting a theory (experimentally
>>> testable) by invoking a god or a magic substance which a priori is not
>>> testable ... to avoid the search of an a posteriori test, given here by the
>>> theory.
>>>
>>
>>
>>> , or I could add a simpler step 8.
>>>
>>>
>>> And here.
>>>
>>>
>>> So, it is enough just to NEVER assumes a *primary* physical universe to
>>> start with, like I do. Then, we can avoid the MGA by explaining directly
>>> (in step 7) that the arithmetical reality implements all computations,
>>> which follows from what we can find in all textbook on theoretical computer
>>> science, like Davis chapter 4 (Turing machine self-applied). I am currently
>>> using that material to explain that very points to my students, so I might
>>> make a try to explain a bit here, to give the taste of it. The advantage of
>>> the Turing machine formalism, is that it is close to "physical computer",
>>> and yet simple enough to make the theory not too much hidden in technical
>>> details. I will think how to do that.
>>>
>>
>>
>> In effect then one argues as follows. Beginning with the assumption of
>> CTM, we can agree in principle that the existence of a computational
>> device, instantiated in a primary physical reality capable of computing
>> sufficient of the trace of the UD, would permit the UDA to go through. This
>> is the initial assumption of Step 7 of the UDA. Then any objection that
>> went to a presumed insufficiency of such a primary reality to implement
>> such a computer would be a merely contingent supposition of its 'actual'
>> non-existence. But the question of what is 'actual' with respect to the
>> theory is precisely what is being asked. So any such "Show me the computer"
>> type of objection begs that very question.
>>
>>
>> Absolutely.
>>
>>
>>
>> This is really forced by the initial assumption of CTM, which puts any
>> theory relying on it in the position of justifying the appearance of any
>> possible physics, including physical computational devices, on the basis of
>> the existence of an arithmetical, not physical, basis of computation.
>> 'Arithmetic' here just stands for any theoretically irreducible and
>> sufficient basis for computation. The relevant sense of 'existence', as in
>> any fundamental theory, essentially equates to explanatory power. It would
>> of course be open to anyone to additionally assume the existence of a
>> putatively more 'explanatorily primitive' physical reality. But this could
>> only weaken the theory by arbitrarily invoking the 'preselection' of
>> undetectable mechanisms that then had no further explanatory role in what
>> followed. Hence this move should be abandoned in favour of greater
>> explanatory parsimony.
>>
>>
>> I think you see the point very well. It makes "primary or primitive
>> matter" exactly like "invisible horse". It adds something "invisible" and
>> gives to it magical abilities having virtually no sense in the frame of the
>> hypothesis.
>>
>
> A small additional point. It strikes me that insistence on an ultimately
> 'physical' basis for computation is related to the intuitionist tendency
> in mathematics, i.e. the notion that mathematics is in fact secondary to
> certain primary relations between physical objects.
>
>
> The common naturalism, I would say? usually intuitionism in mathematics is
> seen as an an idealism, even a solipisme: it is: "the first person view
> *only*". (at the depend of the other).
>
> But I cut the air. I continue the distinction between Bp & p (S4grz, the
> soul, the first person singular) and Bp & Dt & p (the feeler, sensible
> matter).
>
>
>
>
>
> If so, computation in this view would similarly have to be seen as
> deriving from, or more simply *being*, such relations between
> particularised physical objects contrived for the purpose.
>
>
> What would that mean? We can define computation assuming elementary
> arithmetic. To define computation in physics, without using a subsystem of
> physics provably equivalent to the universal number/machine (the
> arithmetical notion) I am not sure the idea of defining computation in
> physical sense makes sense. You might say, with David Deutsch, "unitary
> transformation", but that will only mean "computation" for the reason that
> you will prove that a function from N to N is computable iff it can be
> coded in that "programming" language, actually implemented also by
> arithmetic. It is selecting one universal number among all the other.
>
> When actually with mechanism we must justify the "physical universal
> number" (the Gödel number of a first order specification of unitary
> transformation theory) from the sum on all computations.
>
>
>
>
> Trouble is, were that to be the case, the same arguments thus used to
> dismiss any 'existence', independent of physics, for computation would also
> dispose of any analogous existence for consciousness as its presumed
> consequence.
>
>
> Yes? Arguably so when keeping the mechanist assumption in the mind. To
> remain consistent here, it is important at some point to see that if true,
> the Mechanist hypothesis cannot be claimed to be true. The most we can get
> is a first person confirmation, but it proves nothing (even to oneself as
> the anosognosia illustrates).
>
>
>
> IOW, if computation need be considered nothing more or less than the
> implied relations between certain physical objects then consciousness may
> be conceived in exactly the same way. But then we would be left with a bare
> identity theory: physical relations=consciousness, with no elementary
> connection to computation per se.
>
>
> It is the belief that one universal machine won against all the other.
>
> With mechanism, we have the fight between a finite number of universal
> machine (from the colleagues to some bacteria ...) above our substitution
> level, and infinitely many universal numbers below our substitution level
> (the results of the global FPI on the infinitely many relative
> computations).
>
> The chance is that this is structred by the logic of self-reference.
>
>
>
>
>
> Also since 'existence' here essentially equates, as I remarked, to
> explanatory power, what will have occurred, on the foregoing assumptions,
> is the withdrawal of any such power from either computation or
> consciousness not already attributable to a primitively physical causality.
>
>
> That could have made sense, but by assuming the digitalness mechanism (and
> most of its weakening with oracle) just made the invocation of the physical
> like an invalid use of an ontological commitment to avoid a problem.
>
> If the physical play a role, it needs some magic ability to make
> consciousness, if consciousness is not in the trueness of the relation
> defining this or that computation. mechanism makes it into an inoperative
> god or a selector of computation, ignoring the first person indexical
> selection.
>
>
>
>
>
> Since explanatory entities should not needlessly be proliferated, the
> assumption of primitive physicality should equally entail the dismissal of
> such supernumerary hypotheses. AFAICS this is what Dennett, obfuscatory
> language apart, actually seeks to do. Trouble is, these 'supernumerary'
> hypotheses subsume the entire spectrum of phenomenal reality, or what we
> are pleased to call the ('actual') world. Ah well, so much the worse for
> something.
>
>
> It cannot work. This avoids (implicitly) the measure problem. With
> mechanism, the "hard problem of consciousness" is more easy: it is solved
> or meta-solved by the proof that any machine which observe itself is led to
> an undoubtable & non-rationally-justifiable truth (not necessarily having a
> unique form/content). But the "matter" problem becomes harder, as we can no
> more point to one big universal number (the physical world) and invoke it,
> we have to justify it by computing the relative sum on all histories which
> brought our relative states.
>
> Dennett seems to take Matter for granted, and then explain Consciousness
> away. He is inconsistent, like all people believing in weak materialism and
> mechanism. That works locally, but makes no sense in the possible "big
> picture".
>
> Note that "primary matter" is never assumed in physics. It is an
> assumption in metaphysics/theology.
>
I'm not sure that I can wholly agree with you about this. Or rather I
agree that it's a metaphysical or theological assumption but it's
nevertheless one that is probably assumed, even if unreflectively, as a
physical one by many practitioners. For me, 'primary' implies the
assumption of irreducibility in terms of the relevant theory. I suspect
that most physicists - Deutsch for one with the notion of 'computable' as
ultimately coterminous with physically-transformable - consider physics to
be in a rather strong sense the 'rock bottom' of reality, not further
reducible to something 'not-yet-physical'. Krauss comes to mind as someone
who proposes physical law as the ultimate 'reason' there is something
rather than nothing. Tegmark may be an exception, albeit a
not-always-consistent one.
David
> Bruno
>
>
>
>
> David
>
>
>> To be sure, this assumes Mechanism, in some strong sense. When the
>> argument is translated in mathematics, we get a more constructive view of
>> what physics can be (the logic of Bp & Dt (& p) with p semi-computable
>> (sigma_1, partial computable), and that can be tested. And if the test
>> violates the logic Bp & Dt (& p), it would mean that we have the following
>> disjunction:
>>
>> CTM is refuted OR we are in a malevolent simulation OR there is some
>> magic at play.
>>
>> Then, what the MGA shows, is that the third disjunct is basically an
>> element of the first or second disjunct. It is not really necessary, except
>> for those who don't know really what is a computation and like to cut the
>> air. This makes also Maudlin's contribution more interesting, because it
>> relates the difficulty of defining what is a physical instantiation of a
>> computation. But with CTM, it is the whole "physical" idea which can no
>> more be instantiated by any computation or even non-computation: it really
>> becomes a view from inside arithmetic.
>>
>> Note that in the original long text in french (Conscience & Mécanisme)
>> the UDA and the MGA were used only to motivate the "& Dt" addition to Bp,
>> to get a probability notion. When Brent asks us to take the environment
>> into account, he is using a similar intuition. The logic G fails on
>> knowledge, because it lacks Bp -> p, and it fails on "probability" because
>> it lacks Bp -> Dp. Adding Dt avoids the cul-de-sac world/state, and
>> provides sense to the idea of betting on alternative results possible.
>>
>> Bruno
>>
>>
>>
>>
>> David
>>
>>
>>>
>>> Bruno
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> David
>>>
>>>
>>> Kind regards,
>>>
>>> Bruno
>>>
>>>
>>>
>>>
>>>
>>> Best wishes,
>>>
>>> Evgenii
>>>
>>> --
>>> 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 everything-list+unsubscr...@googlegroups.com.
>>> To post to this group, send email to everything-list@googlegroups.com.
>>> Visit this group at https://groups.google.com/group/everything-list.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>>
>>> 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 everything-list+unsubscr...@googlegroups.com.
>>> To post to this group, send email to everything-list@googlegroups.com.
>>> Visit this group at https://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 everything-list+unsubscr...@googlegroups.com.
>>> To post to this group, send email to everything-list@googlegroups.com.
>>> Visit this group at https://groups.google.com/group/everything-list.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>>
>>> 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 everything-list+unsubscr...@googlegroups.com.
>>> To post to this group, send email to everything-list@googlegroups.com.
>>> Visit this group at https://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 everything-list+unsubscr...@googlegroups.com.
>> To post to this group, send email to everything-list@googlegroups.com.
>> Visit this group at https://groups.google.com/group/everything-list.
>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>> 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 everything-list+unsubscr...@googlegroups.com.
>> To post to this group, send email to everything-list@googlegroups.com.
>> Visit this group at https://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 everything-list+unsubscr...@googlegroups.com.
> To post to this group, send email to everything-list@googlegroups.com.
> Visit this group at https://groups.google.com/group/everything-list.
> For more options, visit https://groups.google.com/d/optout.
>
>
> 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 everything-list+unsubscr...@googlegroups.com.
> To post to this group, send email to everything-list@googlegroups.com.
> Visit this group at https://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 everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
