[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index

Re: Rosen & Ashby



Hi JohnM,
 
See interposed.
 
Regards,
Tim
-----Original Message-----
From: ROSEN Forum [mailto:***On Behalf Of John M
Sent: Sunday, November 23, 2003 11:55 AM
To: ***
Subject: Re: Rosen & Ashby
 
[JM] 
Tim, let me erase all your long post (it DOES make good sense however) and quote ONE LINE of you quoting RR, which causes IMO the most misunderstanding - even controversy - against all the Rosenesque idea (True, however in the RR sense):
"A system is called complex if it has a nonsimulable model." [EL 306]
"The rest of the world" speaks, theorizes. calculates about "complex systems" as the ominous complicated, hard-to-decipher constructs, MOSTLY consisting only of the RR's "simple" ones.
No "nonsimulables" included.
Would it be reasonable to add to the quoted text "According to the RR terminology" or something similar?
 
[TG]
Especially for the purposes of usage outside of this group....yes, it would. 
 
 
[JM]
Now a question to your conclusion: is a "model of a system" an abstraction of it, or an extension which includes the system? In the first case it is reductionistically a part of a reductionist model, ie. the system, while in the second sense it does not make sense (going
into the undefined natural connections). Model up or down?
 
Could you give a SHORT explanation?
 
[TG]
I'd say a model of a system (call it S) is definitely an abstraction of S. Especially in the etymological sense of abstracted as "drawn away from". When S is a natural system, and the qualities of S are encoded into a model, there is an inevitable abstraction occurring in the encoding of some quality of the material world to some fixed symbolic entity, like a number. A formal model also has a finite number of degrees of freedom, which is almost certainly not representative of the actual degrees of freedom in the material world.
 
The only case where the notion of "abstraction" becomes trivial or non-existent is when S is a formal system. In that case, S has a largest model, which is S itself [LI 55], and so there no loss of any qualities in the model with respect to the original system as one would normally expect with an abstraction. But that is a rather trivial kind of "model".
 
 
JohnM