Re: How would a computer know if it were conscious?

From: David Nyman <david.nyman.domain.name.hidden>
Date: Wed, 20 Jun 2007 11:44:39 -0000

On Jun 20, 3:35 am, Colin Hales <c.ha....domain.name.hidden> wrote:

> Methinks you 'get it'. You are far more eloquent than I am, but we talk of
> the same thing..

Thank you Colin. 'Eloquence' or 'gibberish'? Hmm...but let us
proceed...

> where I identify <<<???>>> as a "necessary primitive" and comment that
> 'computation' or 'information' or 'complexity' have only the vaguest of an
> arm waving grip on any claim to such a specific role. Such is the 'magical
> emergence' genre.

Just so. My own 'meta-analysis' is also a (foolhardy?) attempt to
identify the relevant 'necessity' as *logical*. The (awesome) power
of this would be to render 'pure' 3-person accounts (i.e. so-called
'physical') radically causally incomplete. Some primitive like yours
would be a *logically necessary* foundation of *any* coherent account
of 'what-is'.

Strawson, and Chalmers, as I've understood them, make the (IMO)
fundamental mis-step of proposing a superadded 'fundamental property'
to the 'physical' substrate ('e.g. 'information'). This has the fatal
effect of rendering such a 'property' *optional* - i.e. it appears
that everything could proceed just as happily without it in the 3-
person account, and hence 'consciousness' can (by some) still airily
be dismissed as an 'illusion'. The first move here, I think, is to
stop using the term 'consciousness' to denote any 'property'.

My own meta-analysis attempts to pump the intuition that all
processes, whether 0, 1, or 3-person, must from *logical necessity* be
identified with 'participative encounters', which are unintelligible
in the absence of *any* component: namely 'participation', 'sense',
and 'action'. So, to 'exist' or 'behave', one must be:

1) a participant (i.e. the prerequisite for 'existence')
2) sensible (i.e. differentiating some 'other' in relationship)
3) active (i.e. the exchange of 'motivation' with the related 'other')

and all manifestations of 'participative existence' must be 'fractal'
to these characteristics in both directions (i.e. 'emergence' and
'supervention'). So, to negate these components one-by-one:

1) if not a participant, you don't get to play
2) if not sensible, you can't relate
3) if not active in relationship, you have no 'motivation'

These logical or semantic characteristics are agnostic to the
'primitive base'. For example, if we are to assume AR as that base,
then the 'realism' part must denote that we 'participate' in AR, that
'numbers' are 'mutually sensible', and that arithmetical relationship
is 'motivational'. If I've understood Bruno, 'computationalism'
generates 'somethings' at the 1-person plural level. My arguments
against 'software uploading' then apply at the level of these
'emergent somethings', not to the axiomatic base. This is the nub of
the 'level of substitution' dilemma in the 'yes doctor' puzzle.

In 'somethingist' accounts, 'players' participate in sensory-
motivational encounters between 'fundamental somethings' (e.g.
conceived as vibrational emergents of a modulated continuum).

The critical move in the above argument is that by making the relation
between 0,1, and 3-person accounts and the primitives *self-relation*
or identity, we jettison the logical possibility of 'de-composing'
participative sensory-motivational relationship. 0,1, and 3-person
are then just different povs on this:

0 - the participatory 'arena' itself
1 - the 'world' of a differentiated 'participant'
3 - a 'proxy', parasitising a 1-person world

'Zombies' and 'software' are revealed as being category 3: they
'parasitise' 1-person worlds, sometimes as 'proxies' for distal
participants, sometimes 'stand-alone'. The imputation of 'soft
behaviour' to a computer, for example, is just such a 'proxy', and has
no relevance whatsoever to the 1-person pov of the distal
'participatory player'. Such a pov can emerge only fractally from its
*participative* constitution.

> A
> principle of the kind X must exist or we wouldn't be having this
> discussion. There is no way to characterise explanation through magical
> emergence that enables empirical testing. Not even in principle. They are
> impotent at all prediction. You adopt the position and the whole job is
> done and is a matter of belief = NOT SCIENCE.

Well, I'm happy on the above basis to make the empirical prediction:

No 'computer' will ever spontaneously adopt a 1-person pov in virtue
of any 'computation' imputed to it.

You, of course, are working directly on this project. My breath is
bated!

For me, one of the most important consequences of the foregoing
relates to our intuitions about ourselves. We hear from various
directions that our 1-person worlds are 'epiphenomenal' or 'illusory'
or simply that they don't 'exist'. But this can now be seen to be
vacuous, deriving from a narrative fixation on the 'proxy', or
'parasite', rather than the participant. In fact, it is the tacit
assumption of sense-action to the parasite (e.g. the 'external world')
that is illusory, epiphenomenal and non-existent. Real players -
participators - inherit the precursors of *all* their characteristics
from the primitives on which they supervene: hence the fundamental
'spontaneity' (i.e. 'given-ness') of the primitives must also emerge,
mutatis mutandis, at the 1-person level. This is crucial, because we
can now stop gibbering about 'illusion' (which of course isn't to say
that we can never be mistaken). Our personal worlds really are
'something like' - sensorily, motivationally - the explanatory
primitives on which they supervene. After all, sans magic, how else
could it be?

Cheers

David

> Dear David,
> (see below.. I left your original text here...
> =====================================================
>
> > 4) Belief in 'magical emergence' .... qualitative novelty of a kind
> > utterly unrelated to the componentry.
>
> Hi Colin
>
> I think there's a link here with the dialogue in the 'Asifism' thread
> between Bruno and me. I've been reading Galen Strawson's
> "Consciousness and its place in Nature", which has re-ignited some of
> the old hoo-hah over 'panpsychism', with the usual attendant
> embarrassment and name-calling. It motivated me to try to unpack the
> basic semantic components that are difficult to pin down in these
> debates, and for this reason tend to lead to mutual incomprehension.
>
> Strawson refers to the 'magical emergence' you mention, and what is in
> his view (and mine) the disanalogy of 'emergent' accounts of
> consciousness with, say, how 'liquidity' supervenes on molecular
> behaviour. So I started from the question: what would have to be the
> case at the 'component' level for such 'emergence' to make sense (and
> I'm aiming at the semantics here, not 'ultimate truth', whatever that
> might be). My answer is simply that for 'sensing' and 'acting' to
> 'emerge' (i.e. supervene on) some lower level, that lower level must
> itself 'sense' and 'act' (or 'grasp', a word that can carry the
> meaning of both).
>
> What sense does it make to say that, for example, sub-atomic
> particles, strings, or even Bruno's numbers, 'grasp' each other?
> Well, semantically, the alternative would be that they would shun and
> ignore each other, and we wouldn't get very far on that basis. They
> clearly seem to relate according to certain 'rules', but we're not so
> naive (are we?) as to suppose that these are actually 'laws' handily
> supplied from some 'external' domain. Since we're talking 'primitives
> here', then such relating, such mutual 'grasping', must just *be*.
> There's nothing wrong conceptually here, we always need an axiomatic
> base, the question is simply where to situate it, and semantically IMO
> the buck stops here or somewhere closely adjacent.
>
> The cool thing about this is, that if we start from such primitive
> 'grasping', then higher-level emergent forms of full sense-action can
> now emerge organically by (now entirely valid) analogy with purely
> action-related accounts such as liquidity, or for that matter, the
> emergence of living behaviour from 'dead matter'. And the obvious
> complexity of the relation between, say quantum mechanics and, say,
> the life cycle of the sphex wasp, should alert us to an equivalent
> complexity in the relationship between primitive 'grasp' and its fully
> qualitative (read: participatory) emergents - so please let's have no
> (oh-so-embarrassing) 'conscious electrons' here.
>
> Further, it shows us in what way 'software consciousness' is
> disanalogous with the evolved kind. A computer, or a rock for that
> matter, is of course also a natural emergent from primitive grasping,
> and this brings with it sense-action, but in the case of these objects
> more action than sense at the emergent level. The software level of
> description, however, is merely an imputation, supplied externally
> (i.e. by us) and imposed as an interpretation (one of infinitely many)
> on the fundamental grasped relations of the substrate components. By
> contrast, the brain (and here comes the research programme) must have
> evolved (crucially) to deploy a supremely complex set of 'mirroring'
> processes that is (per evolution) genuinely emergent from the
> primitive 'grasp' of the component level.
>
> From this comes (possibly) the coolest consequence of these semantics:
> our intrinsic 'grasp' of our own motivation (i.e. will, whether 'free'
> or not), our participative qualitative modalities, the relation of our
> suffering to subsequent action, and so forth, emerge as indeed
> 'something like' the primitive roots from which they inherit these
> characteristics. This is *real* emergence, not magical, and at one
> stroke demolishes epiphenomenalism, zombies, uploading fantasies and
> all the other illusory consequences of confusing the 'external
> world' (i.e. a projection) with the participatory one in which we are
> included.
>
> ===========================================
> Methinks you 'get it'. You are far more eloquent than I am, but we talk of
> the same thing..
>
> Liquidity is to H2O
> as
> <<<???>>> is to consciousness (qualia)
>
> where I identify <<<???>>> as a "necessary primitive" and comment that
> 'computation' or 'information' or 'complexity' have only the vaguest of an
> arm waving grip on any claim to such a specific role. Such is the 'magical
> emergence' genre.
>
> I have a viable candidate for the 'necessary primitive' of the kind you
> seek. Note that regardless of what anyone's suggestion for such a thing
> might be, the process of declaring it valid must arise in the form of (as
> you intuit above) an axiom. That is, it must come in the form of a
> statement such as
>
> X = "It is a fundamental principle of the natural world that <<< such and
> such a thing>> is the ultimate necessary primitive state of affairs that
> must be the case for consciousness to a) deliver a faculty of observation
> at all and then b) for biological delivery that, when observing its own
> processes delivering observation, looks like brain material when that it
> is happening and NOT when it isn't"
>
> This is the generic form of the kind of 'fundamental principle' that
> Chalmers called for in his 1996 book. You cannot escape the reality of
> some form of X, regardless of <<<such and such a thing>>. In terms of X:
>
> liquidity is to H2O
> as
> The property X is to consciousness (qualia)
>
> Such a statement, in one fell swoop, eliminates the
> circularity/self-fulfilment of the magical emergence kind...that which I
> have been at pains to point out (dialog with Russel) is inherent at
> multiple levels in the 'computationalist' or 'eliminativist' or
> 'functionalist' or 'representationalist' flavours of magical emergence. A
> principle of the kind X must exist or we wouldn't be having this
> discussion. There is no way to characterise explanation through magical
> emergence that enables empirical testing. Not even in principle. They are
> impotent at all prediction. You adopt the position and the whole job is
> done and is a matter of belief = NOT SCIENCE.
>
> Remaining at the 'meta' level of general discussion of statements of the
> kind X, we can make a couple of important observations:
> 1) X must be consistent with all empirical (neuroscience/cognitive) evidence.
> 2) X must at least in principle be able to make verifiable novel
> predictions of the natural world critically dependent on X being the case.
>
> If it cannot do this then it is not viable. If it can make even ONE then
> it is a better proposition than magical emergence.
>
> Note there is a unique character to X that is unlike anything science has
> ever encountered. Direct evidence for X is not any specific observation.
> It is the cause of the mere possibility of any observation at all. As I
> have said in previous posts.... "that which is seen (contents of
> consciousness) is our traditional demanded source of explicit scientific
> evidence. However..."Seeing" (any observation/consciousness at all) is
> intrinsic/implicit evidence of the existence of circumstances that
> necessarily exist or no explicit observation would exist. Every explicit
> act of scientific observation also delivers implicit evidence of an
> ability to observe. In other words evidence of the lack of qualia is the
> failure to observe.
>
> The only third party verifyable circumstance where failure to observe can
> be counted as evidence is in the failure of the scientific act. Hence my
> insistence that science and the scientific act be the evidentiary
> circumstances of a viable empirical science directed at proving X.
>
> Other than this strangeness (scientists themselves are evidence), the
> process is just like any other scientific process. Business as usual.
>
> If Galen Strawson is setting fire to the magical emergence camp....GOOD!
>
> I have an X proposition that delivers what looks like panpsychism, but
> isn't actually....well... if X is a principle that applies universally
> (something that exists at all spatial scales... inherent in the whole
> universe) and is intrinsic, logically unavoidable and implicit...then does
> that count as panpsychism? I don't know. Frankly I don't care!...X is a
> member of the class "Law of Physics". Where it fits philosophically is
> someone else's problem. Not such characterisation has any impact on the
> empirical work... and I am fiendishly empirical to the bitter end...
>
> Before I re-deliver my X... I'd like to leave the discussion at the META-X
> level (about any X or about all possible Xs)....over to you....
>
> cheers
> colin hales


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Everything List" group.
To post to this group, send email to everything-list.domain.name.hidden
To unsubscribe from this group, send email to everything-list-unsubscribe.domain.name.hidden
For more options, visit this group at http://groups.google.com/group/everything-list?hl=en
-~----------~----~----~----~------~----~------~--~---
Received on Wed Jun 20 2007 - 07:44:45 PDT

This archive was generated by hypermail 2.3.0 : Fri Feb 16 2018 - 13:20:14 PST