Re: a request / Connectionism

John St. Julien (stjulien who-is-at UDel.Edu)
Tue, 24 Mar 1998 09:09:24 -0400

Mike,

Your remark:

>As I understand the form of connectionism proposed by Elman, Bates, et al
>in *Rethinking innateness: they are proposing a form of emergentism. It
>is best captured for me in Bates' metaphor of the relateion of behives
>to the genetic makeup of bees. Bees have founds, but they produce hexagonal
>hives structures. They do this as they pack honeycombe material into as small
>a space as possible with their round heads. The result of packing lots of
>circular units into small spaces is a hexagnal hive. But there is no
>"hexagon making" module in the brain, nor is their a UG grammar module,
>etc. Rather, UG is the emergent solution of packing lots of stuff (
>meanings) into a tiny channel (the speech stream).
>
>Have I got this right?

I do think you recognize the basic pattern. I don't think this way of
reasoning is all as new and amazing as it is sometimes made out to be. It
was roughly introduced into "science" by Darwin and made more explicit by
ecology. What is different is that a dynamic, history-of-particularity
explanation is being applied in so many places. Like social insects and
brains and societies and immune systems. I do think we make a mistake,
though, if we were to identify the various disciplines using this pattern
of reason with computers-as-mechanical simply because they use computers as
tools to model that immensly complex history of development.

---------------------------------------------------------------------------
John St. Julien (stjulien who-is-at udel.edu)
Department of Educational Development
University of Delaware