[Top] [All Lists]

Re: [ontolog-forum] Watch out Watson: Here comes Amazon Machine Learning

To: ontolog-forum@xxxxxxxxxxxxxxxx
From: John F Sowa <sowa@xxxxxxxxxxx>
Date: Sun, 19 Apr 2015 02:50:31 -0400
Message-id: <55335037.5090306@xxxxxxxxxxx>
On 4/18/2015 8:11 PM, Thomas Johnston wrote:
> Jerry Fodor, of Language of Thought fame, believes connectionism
> to be a challenge to the idea that ideas are embodied as discrete
> representations in the brain, and that transformations (inferences,
> etc.)  are embodied as discrete processes in the brain.    (01)

Yes, but so what?  Jerry has some interesting hypotheses, but he also
has a "fetish" for a discrete set of options.  I was at a conference
where he presented a list of alternatives.  Then he went down the list
one by one to show problems with each one.  Then he proudly claimed
that the last one must therefore be true.    (02)

At the end of that performance, a colleague next to me stood up
and said, "Jerry, you have left a burning field of strawmen."    (03)

For a review of the issues about psycholinguistics, neuroscience,
and their relationships to language and thought, see
http://www.jfsowa.com/talks/goal2.pdf .    (04)

Various slides in that presentation mention the hypothesis of a
"language of thought".  The evidence for it is slim to nonexistent.
As the neuroscientist Antonio Damasio says, images are "the main
currency of our minds."    (05)

People remember images far better than they can remember the exact 
wording of prose.  But note that they can remember multimedia data
best of all.  They remember poetry better than prose, and they have
a much more exact memory for poetry when it's set to music and dance.    (06)

For more about NL semantics and its relationship to imagery, see
the next "chapter":  http://www.jfsowa.com/talks/goal3.pdf .    (07)

> To them [Paul and Patricia Churchland], Jerry's LOT is a
> formalization of what they call "folk psychology", which they
> regard as equivalent to earlier beliefs in phlogiston    (08)

Fodor and the Churchlands are at opposite extremes.  That's not
a good recommendation for either.    (09)

> [folk psychology] will not be refuted, but will simply wither away
> as it becomes less and less plausible    (010)

Guess what.  The Churchlands' ideas haven't withered away.  They're
ignored.  Neuroscientists today don't use the pejorative term "folk
psychology".  They use human experience and what people say about it
as key to understanding and interpreting what they find in fMRI scans.    (011)

> follow the references in the IEP (Internet Encyclopedia of Philosophy)
> article on Connectionism, and several of the articles in the SEP
> (Stanford Encyclopedia of Philosophy), such as the articles on Mental
> Representation and the Language of Thought.    (012)

Those are good reference sources.  They ask partisans of the many
areas to write good summaries and histories of the various positions.
But they present them all on an equal basis.  You have to look at
ongoing research and successful applications to see what works.    (013)

Critical issues:  The world is a continuum, but the words of language
are discrete.  People use an open-ended variety of different methods
of reasoning, and they're not even aware of the flexibility of
language in moving seamlessly from one type of reasoning to another.    (014)

People are multi-paradigm reasoners.  Anybody who thinks that they
have to pick and choose one method in preference to another is
hopelessly misguided.  That's the point I was trying to get across
in the MICAI talk:  http://www.jfsowa.com/talks/micai.pdf .    (015)

For more background, read the "7 chapters" of goal.pdf ... goal7.pdf.    (016)

John    (017)

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (018)

<Prev in Thread] Current Thread [Next in Thread>