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

Re: Causality vs Entailment



Judith,
 
First of all, I never said or intimated what you seem to attribute to me here:
JR:This is a far cry from saying that one must never look for causal entailments from system behaviors!
 
I did not use the phrase "never...". Instead, I said "...one cannot reliably argue backwards from behavior (effects) to causes" and "Analytic models are not generically the inverse of synthetic models." (Arrgh, I just noticed a typo in my reply to Steve, where I left out the "not" on the prior sentence -  should be: "And an accretion of behaviors does not generically entail a particular accretion of causes.") The use of the more cautious phrases "cannot reliably" and "are not generically" is because one might in fact be discussing a simple sytem where 1) analysis and synthesis are inverses, and 2) where subjective discrimination issues (such as in the Gibbs paradox referenced after the quotes I previously mentioned, on p. 43 and 124) do not inject themselves into the study of that system. Such situations will be nongeneric, but may exist. So, it is not possible to universally rule out such backward reasoning from from behaviors to causal underpinnings. The quotes from Essays p. 42 and 123 both take that same cautious tone, using the terms "unsound" and "cannot generally argue" respectively, rather than "never".
 
We certainly both agree on this:
JR:In that chapter, my father is talking about how foolish it is to study the behavior of simulacra and apply principles learned from that study to the real systems that were being simulated. He further said that simulacra, which mimic only the observable behavior of the real system because of the false premise that the underlying entailments will follow, are a waste of time, scientifically.
 
I disagree that in the quote on p.42 "the physical world" refers only to "contemporary physics".  I assert he is simply citing a well-known historical observation about the general failure of such backwards reasonings about the external world. The reductionist paradigm does not have a way of accounting for this observation (thus it is that the "Gibbs paradox" is a "paradox" at all), but the Rosennean paradigm does. Thus it is all the more explicitly apparent in the Rosennean paradigm why it is unsound to argue from behaviors to causal bases; and therefore, why it is generally a waste of time to study simulacra of a natural system as a way to learn more (i.e., be able to answer more "why" questions about) that natural system.
 
Regards,
Tim