On 21 Mar -1, Marchal wrote:
> Gisle Reigstad Tangenes wrote:
>
> >I guess it's time to reveal the terrible secret: Your list is infected by
> >a biological naturalist.
>
> Terrible indeed !
> > My Searlean objection to the above brand of functionalism
> >is, How can computation as such be sufficient to generate consciousness,
> >when it obviously isn't an intrinsic process of any physical system?
>
> >To clarify: There is a distinction between intrinsic and
> >observer-relative features of reality. The former include all properties
> >
> >that are logically independent of the intentional attributions of
> >observers, such as the molecular structure of the object I am sitting on.
> >The latter are properties that exist only relative to such attributions,
> >such as being a chair. The problem here is that computation, to quote
> >Searle, is "not a machine process that like neuron firing or internal
> >combustion; rather, computation is an abstract mathematic process that
> >exists only relative to conscious observers and interpreters" (*The
> >Mystery of Consciousness*, 17).
>
> I am not sure that the distinction between intrinsic and
> observer-relative is not itself observer-relative.
I don't see how the distinction can _itself_ be observer-relative?
Let Observer Independence (OI) and Observer Relativity (OR) be
higher order properties of any property P of any substrate s. Then surely,
(-(p&-p)) --> (s(P)(OI)), for s(P) = OI/OR? I.e, it can
hardly be in the eye of the beholder whether any given feature of the
world is in the eye of the beholder or not. For instance, in a solipsist
world W, the fact of solipsism is an intrinsic (=OI) fact about W. This
does not preclude that observer-relativity is a recursice property, of
course. For example, semantics may be an OR feature of syntax, which is
itself an OR feature of the physical world.
Do you mean that the correct *application* of this distinction is itself
observer-relative for some P, where this observer-relativity is somehow
comparable to e.g. the relativity of simultaneity in Relativity theory?
Not that I really understand how this analogy might be possible. But then
I make no claim to understand your metaphysics, which to me (a diehard
external realist) appears exotic in an interesting way. Shall take a
look on your thesis soon.
Second thought: Just possibly we need to distinguish the case of
(i) [s = reality as a whole] from (ii) those of [s = any element of
reality], so that for (ii) but not (i) it is in some sense OR whether s(P)
is OR or OI. But I have no clear idea of what exactly this sense might
be. This deserves some hard thought (*damn*).
>You take "the
> molecular structure" as intrinsic and I will tell you below why
> I don't. It seems you take the notion of "physical things" as granted,
> but I don't.
Right. My point here is that even you must assume _some_ intrisic
features of reality. The metaphysical bucket must stop somewhere.
> But even if there were intrinsic features of physical reality, it is
> hard for me to understand how consciousness would rely on it.
> What in the brain/body/universe do you think to be intrinsically
> necessary for consciousness ? Sodium ions? Carbon atoms? Quark?
> Quantum phenomena in microtubules?
To the naturalist, this is a poorly understood empirical question: The
inclusive disjunction of causal requirements simply isn't known by any
member of this species (and will probably never be, because of the limited
light cone). All we know at present is that certain sorts of
electrochemical pulses in certain sorts of carbonbased neural tissue are
jointly sufficient for certain sorts of consciousness. As I see it, the
case for the computationalist hypothesis is weaker then the case for, say,
the possibility of zero point propulsion, which seems at least predicted
by QM. I.e., it makes good SciFi, which is no mean thing. (It can't be
said about the microtubule speculations, for example).
> (Searle's comparison between aware-thinking and photosynthesis is
> really a philosophical category mistake, I think).
Yeah, so it would presumably seem to a computationalist; but can it be
shown to be so on logical grounds?
> >Computation and all other syntax is
> >observer-relative, and in one sense exists only from a 1. person point of
> >view.
>
> Computation has almost nothing to do with syntax, and is not
> observer-relative, at all. To my knowledge it is even the less
"observer-relative" thing I can think about.
(1) On syntax vs. computation: Maybe I have been vague here: By
'syntax' I meant not merely linguistic grammar but
any manipulation of symbols according to a set of well-defined
procedures specifying legal and illegal operations. But don't let us be
bogged down in any terminological dispute: I'll avoid the term hereafter.
(2) On the observer-relativity of computation: When a brain computes
some function or other, that process is observer-independent as
it does not depend on how anything _outside_ that physical system treats
or interprets it. But if one somehow uses an electronic device, the
surface of a soap-bubble, or an elephants' trunk to the same end, then
there is only a brute physical process unless a computational
interpretation is assigned to it by some transformation function. This
OR, however, does not imply what Searle would call epistemic
subjectivity (ES): The truth value of a computational result R is not
determined by individual attitudes to R. More on mathematical epistemic
objectivity (EO) below.
How, then, does the _brain_ compute anything? Clearly there is no
conscious interpretation of non-intentional neuronics? I guess the brain
simulates a subset of the relations between a subset of abstract patterns
by means of its symbolic capability: It uses its unique physical
ability to treat certain physical objects (inks on paper, wrinkles on an
elephant's trunk, or even mental pics) *as* something else (patterns and
metapatterns). A special case of this special case of toolusing is what
computation boils down to for the naturalist, I suppose.
>Perhaps you confuse a computation with your own possible
> interpretations of that computation? or with a possible concrete
> implementation of that computation.
If so, I am absolutely unaware of that...
> First I *assume* computationnalist as a working hypothesis. I am
> open to the idea that comp is false. Perhaps biological naturalism
> is the correct approach, despite Searle!
OK, one can't quarrel with that, and it's openminded to (almost surely
more than, e.g., John Searle).
> By comp I mean: 1) the idea that there is a level of substitution
> such that I survive with a functionnal and digital substitution made
> at that level. Unlike so-called functionnalist I don't put any
> restriction on the level. My 'brain' could be the whole quantum
> multi-universe, my reasoning will still go through;
OK
> 2) the classical Church thesis. This is very important because it is
> Church thesis which make computations intrinsical and
> syntax independent;
OK. I don't quite see how the Church-Turing thesis can make computations
_intrinsic_ to anything, as opposed to merely intertransformable. In
general, I understand few of the grand claims made on behalf of this
thesis. But then I'm pretty far away from home here.
3) a minimal amount of arithmetical platonism. I
> guess this is maybe unnatural for a biological naturalist. It is
> nevertheless the simple idea that arithmetical truth is independant of
> me. For exemple the idea that the statement "there is an infinite
> number of primes" is true independently of me.
It is indeed, and that is not so unnatural. As suggested above, math
objects must be seen as the class of possible patterns and relations
between those, some of which are physically implemented (or all, according
to the everything idea.) I don't think we disagree here. As to the
location of non-implemented abstract patterns: I don't agonize so much
over this, but I guess it adds to the attraction of the everything idea.
That will have to do for now.
Best,
--
Gisle
Received on Wed Dec 22 1999 - 14:09:19 PST