Re: Question

[Dan Fiscus <***>]

Howard,

I would say that just looking at these two, they may appear
the same, but that the more we add additional works by
Rosen and likely by Langton to fill in the broader contexts
we'd see more and more evidence for fundamental
philosophical differences. I don't have specific quotes or
references, but I would suggest these areas as good bets for
illuminating the major formal and physical (left and right
hand sides) differences, these all from Rosen:

1. context dependence - crucial issue for life systems
2. closed to efficient but open to material cause (similar to
#1 - local material context is causal, integral)
3. anticipatory always - clock in model (formal) runs faster
than clock/time in environment (phyical)
4. unfractionable - especially re: hardware vs software

The latter may be the biggest with Langton et al. and it
suggests a major caveat or complement to the quote of Rosen
you have below. It may be his approach was the opposite of
reductionist/mechanist, but it was also large enough to
subsume the material/physical as well and I think he wrote
this many places. The formal and organizational parts may
be discussable or even modelable as separate from the
material, but I don't think he ever suggested actually
separable in realized systems.

In summary, I'd say Rosen was trying to found a study or
theory of authentic life and authentic intelligence as very
different from the artificial schools we have now.

One other idea we have discussed before - an important test
that might aid the AI/AL folks and generalize the Turing
Test would be to test for criticality of context dependence.
To do this, unplug any simulation system from its stream of
energy and or material flows and then see if that
disconnection from local material context matters. If not,
if you can just plug it back in after an arbitrary time and it
comes back to artificial "life", that seems qualitatively
different from most if not all life systems. For life, to
disconnect from continual supply of energy, air, water,
food beyond well-defined and fairly short limits is to cause
a non-reversible change in the system. Caveats here are
for things like dormant cysts, seeds, etc. But even these
would require a special set of environmental conditions to
re-activate and bring them back to authentic life, and they
also require special conditions for maintenance of the
dormancy or stasis.

Dan


Howard Pattee wrote:
> Here is a question that I suspect will elicit several opinions:
>
> Rashevsky’s relational biology is the study of life at a level of abstraction that does 
> not address any particular material physical realization of life, but looks at its most general 
> logical organization. Rosen contrasts relational biology with reductionist biology in the 
> following words:
> “In any case, I can epitomize the reductionist approach to organization in general, and life in 
> particular, as follows: throw away the organization and keep the underlying matter. “The relational 
> alternative to this says the exact opposite, namely: when studying an organized material system, throw away 
> the matter and keep the underlying organization.” (LI, p. 119)
>
> Langton’s and other’s view of Artificial Life is that they also want to get beyond 
> particular material realizations of life. Langton says:
>  “Of course, the principle assumption made in Artificial Life is that the ‘logical form’ of an organism can be 
> separated from its material basis of construction, and that ‘aliveness’ will be found to be a property of the former, 
> not of the latter.” (Artificial Life, Langton, ed., Addison-Wesley, 1989, p.11.)
>
> Question: What substantial philosophical differences do you see here, if any. Of course, the actual research programs are quite different.
>
> Howard