Re: Inequivalence of models

JohnM,

I pretty much agree. I don't really like the term "inequivalent" because any two models which are not perfectly identical are obviously 'inequivalent' in some sense(s). But what do we mean by 'inequivalent'? Rosen from AS: "Thus, for us, a system will be complex to the extent that it admits non-equivalent encodings; encodings which cannot be transformed or reduced to one another." [AS p. 322, bold added] So, to me, if we are to speak about complexity in terms of "inequivalence" (or "non-equivalence") of models, then the criterion is that the models "cannot be transformed or reduced to one another". I think the term "incommensurability" captures that criterion economically.

> -----Original Message-----
> From: ROSEN Forum [mailto:***]On Behalf Of John M
> Sent: Sunday, January 16, 2005 11:59 AM
> To: ***
> Subject: Re: Inequivalence of models
>
>
> Tim,
> sorry, I sent my reply to Steve before I received this - your one.
> I did not take the engineering view (in wording), but I think the meanings
> are not far away.
>
> One remark on your last lines here:
> I feel 'inequivalent' models are aspectually different,
> commensurability may
> be one characteristic within such. An inequivalent (by meaning!)
> model pair
> does not have to be impredicative (though it can be). The "state-based"
> model-pair (1st time I read this name) is IMO aspect-based limited as well
> (if I understand its meaning right).
> I deem (as you know) the engineering thinking sort of reductionistic.
>
> Otherwise have a good Sunday
>
> John M