Bennett and Hacker: Village Idiots or Philosophers?

Alex Green dralexgreen at yahoo.co.uk
Mon Feb 16 12:16:42 EST 2004


"DRS_IsAnIdiotAsWellAsStupid" <DRS_IsAnIdiotAsWellAsStupid at yahoo.com> wrote in message news:<53ada801aa04f469194b1642a436efc9 at news.teranews.com>...
> "Eray Ozkural exa" <erayo at bilkent.edu.tr> wrote in message
> news:fa69ae35.0402151239.53570361 at posting.google.com...
> > "AlphaOmega2004" <OmegaZero2003 at yahoo.com> wrote in message
>  news:<59dd837546b2b1f4ee6231123f513845 at news.teranews.com>...
> > > "Eray Ozkural exa" <erayo at bilkent.edu.tr> wrote in message
> > > news:fa69ae35.0402121807.4090deb8 at posting.google.com...
> > > > I don't support that position, either. One does not need to be
> > > > "eliminative" to be  materialist.
> > >
> > > It does help when you have to have a bowel movement though.
> > >
> > > ;^))
> > >
> > > Seriously, there are conceptions of mind/brain_processes that rely
>  totally
> > > on time_as_primary; synchronization theories etc. The notion that brain
> > > experiences all environmental events in their past, the synchrony  (or
>  lack
> > > thereof) of neuronal group firings *as* representational vehicles. Etc.
>  etc.
> > >
> >
> > Empirical data seems to suggest that "timing" is all there is to
> > information in the neural code, see the Flynculus threads!
> >
> > > >
> > > > Omega might have something to say on Churchland, I'm curious of what
> > > > he thinks.
> > >
> > > I have always liked Pat's take on several issues. Her Computational
>  Brain
> > > (with Terry Sejnowski - one person in the field who deserves every
> > > accolade...)  IMHO, set a standard for turning philosophical meanderings
> > > into empirically-verifiable hypotheses and hard data (based on ANN
> > > observations). Modelling neurobiological structures with computational
> > > structures has provided a great deal of insight into how brain
>  categorizes
> > > and communicates between NGs - two essential characteristics of what it
> > > means to be a mind - or cognizing entity. Cognition of difference ( the
> > > function that co-operates with or operates just before actual
> > > categorization) is of material primacy to consciousness; separating self
> > > from non-self as the second-most (and secondly-adopted) function of
>  brain.
> > > E.g., infants prior to a few months develop the ability to cognize
> > > difference in the environment; thence they develop the cognition of
> > > self-other differences as part of the development of their selves -
>  their
> > > self-consciousness follows.
> >
> > Do you observe how that elates "difference" as a first class abstract
> 
> Yes!
> 
> That seems to be primary - a process of "seeing" thence "cognizing"
> difference.
> 

"Difference", La DifferAnce, is really about linguistics and literary
criticism. Experiments that rely on differences are likely to be
biased towards linguistics where such effects are of paramount
importance. This is because 'difference' is about the reduction of a
multidimensional state into a a single channel (language, or the tape
of a Turing Machine). Most states are not single channel, they are
multidimensional entities.  A single channel can be used to evaluate
functions and to describe a state but it cannot itself be a state of
more than one dimension. Specifically, a graph that is a two
dimensional object, can be described by the function y=mx+k but the
series of numbers that are the evaluation of this function are not the
graph, they are a 1D description of the graph.

The application of differences is valid in the field of linguistics
but the user must beware of confusing a one dimensional description
with the states that are being described. If a baby is tested for
recognition of differences between two things the experimenter should
test if the baby can sense each of them (otherwise the experiment is
confounded). If it can sense each one then it senses two things. If
the baby reacts in some post sensory way to the two things as if they
are one then this is remarkable because they each occupy different
positions in space and/or time. The conceptual achievement would seem
to be 'sameness' rather than 'difference'!

> > concept, not unlike what Lester suggests? (This is not to say I agree
> > with his arguments, only that difference could be treated more
> > carefully)
> >
> > Minsky also argues that much of subjective experience is a direct
> > consequence of analysis of differences, i.e. if the input is static
> > subjective experience dissolves, both in reactive and reflective
> > states. There may be a subtle issue of knowledge and experience here,
> > but I cannot quite pin it down.

Experience is at least a (3+1)D state so I cannot see how a linguistic
description can encapsulate it, although it may encapsulate a
subject's spoken description of it.  The description of experience
needs proper science.

> And the meditative arts teach just such a dissolution of difference (at
> least the cognition of part) such that all that is left is the cognizing
> instrument - pure consciousness.

I agree, in meditative states processes are observed until they
disappear. The "Pure consciousness" of meditation appears to be a
processless state.

> >
> > > But back to Churchland: her stand on materialsm (based on the reduction
>  of
> > > phsychological facets of mind/brain to physiological/computational
> > > mechanisms) is well-taken as far as it goes in "eliminating" the "spooky
> > > stuff" like souls and spirits (and with them - the elimination of the
> > > Cartesian/dualist conundrums).  Note that she admits that this is an
> > > hypothesis (in good standing - one which I agree with - as I have not
>  seen
> > > any disembodied minds yet), but strong enough to proceed as though it
>  were a
> > > fact of nature - a particularly strong position philosophically.
> >
> > That is well taken, indeed.

Also agreed in part, but this is the 21st century. Even in the 18th
century leading philosophers and scientists were dismissing spooky
stuff as humbug. My problem with reductionism is that it is senseless
without empiricism. It just gives rise to a pile of symbols without
empiricism. Reductionists commit a fatal error in neuroscience when
they believe that the symbols are reality, symbols can only ever
describe reality whereas observation is reality itself.  Empirical
science is reductionist with one extra, crucial, ingredient: the state
called 'observation'.  Observation always involves data arranged in
more than one dimension.

> > > In reading some of her other position papers, I understand that her
>  top-down
> > > reductionist approach does *not* mean that bottom-up appraches are not
> > > important.  She has an engineering/reverse-engineering bent - so it is
> > > natural for her to assume the reductionist approach - take it apart and
>  see
> > > how the components work and constitute the whole (with more emphasis on
>  the
> > > former than the latter - which IMO has some problems - one of which is
>  that
> > > the notion of emergence is relegated to the back burner, whereas I think
>  it
> > > should be front and center, as reductionism tends to eliminate
>  *information*
> > > and is a lossy process (as I have discussed before.)
> >
> > I have a problem with such theories when they don't lead to a
> > mathematical insight. I try to back up my arguments with some small
> > complexity "fact" / "hypothesis", if you recall how I presented a
> > mathematical interpretation of the irreducibility of mental processes.
> 

A mathematical insight is only valuable if it provides a description
that can provide the basis for an experiment. The experiment may be a
conscious machine or stimulation of a brain etc.  Mathematics itself
is just a description and to confuse a description with reality is a
category mistake. As Mary, in the Knowledge Argument would know, the
three letters RED are not red any more than the mathematical equations
that describe the absorbtion of light in cones and the neural
processing thereof. The equations are just descriptions. When these
descriptions are constructed into a state of the world they acquire a
multidimensional character that can be a 'thing' or an observation
etc.

We use scientific reductionism to observe, build, cure and predict.
The flow of symbols from a mathematician's pen or through a computer
just become rubbish in a waste paper basket or lost as heat.

Best Wishes

Alex Green



More information about the Neur-sci mailing list