Re: Fundamental problems in Physics

[Tim Gwinn <***>]

Howard,

The difference is that you consider a failure of the modelling relation to
commute as being "not often a fatal error". I read this, and your other
comments, as indicating that a commuting modelling relation is somewhat
optional. By contrast, for a formalism to be a model in the Rosennean view,
the modelling relation must commute, otherwise that formalism simply is not
a model. Granted, determining when a model is in a commuting modelling
relation requires subjective determination. Still, the requirement for a
modelling relation to commute is fundamental to maintaining a clear
distionctiion between models and simulations, as well as laying the basis
for all the arguments for complexity, complementary models, etc. that flow
from that requirement.

Regards,
Tim


> -----Original Message-----
> From: ROSEN Forum [mailto:*** Behalf Of Howard
> Pattee
> Sent: Thursday, December 16, 2004 6:27 PM
> To: ***
> Subject: Re: Fundamental problems in Physics
>
>
> Tim,
> I would appreciate more explanation.
>
> TG: Ah, now we come to the crux of the matter. What you
> previously seemed to
> assert to be the same things - Hertz's discussion in the
> Introduction of the
> Principles of Mechanics and the Rosen Modelling Relation - as
> represented in
> your diagram "ELABORATED HERTZ/ROSEN MODEL COMMUTATION DIAGRAM" combining
> the two - are now very different things.
>
> HP: My elaborated diagram is labeled differently than Rosen's
> usage, but I
> do not understand why it is so essentially different. It is meant to
> express Hertz's basic concept of a good model. Could you explain what you
> think is "entirely different."
>
> TG: If you wish to promote an entirely different epistemological
> basis for
> science founded on your interpretation of Hertz, then fine, but
> this really
> isn't a useful forum for that.
>
> HP: I do not wish to promote anything. I wish to understand what
> we mean by
> good models. If complex systems require complementary models, as Rosen
> says, I think perhaps there may be more than one good
> epistemology for such
> irreducible models. In any case, I would like to understand how Rosen's
> epistemology is different from Hertz's.
>
> Howard