Re: A new methodology for the human sciences

So much of what you have written is utter bs, but you are on to something.
Unfortunately what you propose will never be accomplished by one person -- 
it's just too much work.

"Jorn Barger" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Over the last 5000 years of recorded history, there's
> been a continual debate about _what we should do_ and
> _what we should believe_.  Innumerable religions have
> claimed authentic insight into God's will and God's
> truth.
>
> For at least 2500 years, some religions have strongly
> asserted that some or all human desires are contrary
> to God's will, that God wants us to deny our desires,
> while others have seen our desires as direct
> expressions of God's will.
>
> In the last 400 years, the new methodology of natural
> science has used observation and experiment to mount
> an overwhelming challenge to all religious doctrines
> of God's truth (what we should believe), but only
> scant progress has been made about what we should do
> (God's will).
>
> This new challenge of _scientific_ truth has inspired
> religious thinkers to scour the imagery of science for
> metaphors upon which new or old ethical doctrines can
> be anchored-- eg, whether the Universe will ever end,
> whether life begins at conception, and whether
> consciousness exists outside the neural reflex arc.
>
> Over the last 30 years, sociobiology has been building
> a strong case that evolution designed our desires to
> promote the survival of our genes.  But this
> circumstantial argument for God's will carries little
> weight in the ongoing debate because it still lacks
> particular models of how evolution implemented
> specific desires.  Until science achieves these models,
> religions will continue to assert the superiority of
> denial.
>
> Twentieth-century attempts to apply the methodology
> of natural science to the human sciences have been a
> complete failure.  Pseudo-experiments have been
> designed to take pseudo-measurements of pseudo-
> variables in pseudo-models.  When Will Wright sat down
> to create "The Sims" he had to start from scratch--
> there weren't any existing psychological models he
> could really use.
>
> In fact, most existing theories of the mind are
> closer to the old religious assertions of God's will
> than the new scientific assertions of God's truth.
> So we need to re-think the scientific method if we're
> going to successfully extend it into the psychological
> realm.
>
> To build a successful model you have to analyse a
> system into well-defined parts and show how they
> interact.  Models of the mind that consider neurons
> as the basic 'part' have had a small success in
> inspiring the 'neural network' model of computing.
>
> But most psychological models instead postulate
> 'parts' whose names are taken from the ill-defined
> intuitions of natural language-- memory, knowledge,
> understanding, plan, will, conscience, etc.
> Descriptions of human behavior made using such
> constrained vocabularies tend to be clumsy and
> artificial.
>
> But there is a highly evolved vocabulary for
> describing human behavior-- the vocabulary of
> novelists.  The best novelists strive to capture new
> subtleties in the most effective language available.
>
> The philosopher JL Austin made a remarkable series of
> arguments that ordinary language is somehow _smarter
> than we are_, and that to benefit from this hidden
> wisdom we have to approach it with humility.  The
> absence of this humility towards language is surely
> one of the flaws undermining 20thC approaches to the
> human sciences, but experimental computer platforms
> like "The Sims" should offer a remedy.
>
> To extend "The Sims", we'll need to collate tens of
> thousands of literary descriptions of human behavior,
> respecting the subtleties of the novelists' language,
> and extract from this observational/descriptive level
> the patterns that can refine the awkward errors of
> our Sims models.
>
> The sort of fake laboratory experiments that 20thC
> social science squandered its efforts on must remain
> meaningless until we have a computer model that
> begins to resemble the behaviors we're trying to
> measure.
>
> Another unexplored angle of attack is the challenge
> that Roget took on with his Thesaurus-- finding a
> systematic arrangement of _all_ ordinary vocabulary
> words.  Roget worked without the benefit of
> evolutionary theory, and it's high time that his
> categories be reworked along sociobiological lines.
>
> Each of the difficult psychological categories must
> have emerged in some unknown sequence, and
> reconstructing that sequence should prove a useful
> challenge for our model building.
>
>
> general: http://www.robotwisdom.com/issues/fiasco.html
> timeline: http://www.robotwisdom.com/ai/timeline/0000.html
> notation: http://www.robotwisdom.com/ai/antimath.html
> data structure: http://www.robotwisdom.com/ai/thicketfaq.html
> knowledgebase: http://www.robotwisdom.com/solace/
> etymogeny: http://www.robotwisdom.com/ai/etymogeny.html
>            http://www.robotwisdom.com/ai/universals.html




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----==  Over 100,000 Newsgroups - 19 Different Servers! =-----