Re: Empirics and Life

[Howard Pattee <***>]

At 09:09 AM 1/27/05 -0500, Judith wrote:
> Howard,
> That wasn't really what I was asking. My question is: How does science
> empirically verify life as a quality or property of a living organism? It
> cannot be measured directly, so how do they empirically verify it? As far
> as I know, they don't. 

HP: We just try all the tests we can think of. I don't see how your
question can be answered without also answering the big question: What is
a living organism? Rosen and I were on one of several NASA committees
trying to decide what detectors to put on Viking 1, the first soft-lander
on Mars, to see if there was life. There were landscape pictures,
microscopic pictures, chemical reactors, spectroscopes, culture media,
etc. It is a pattern recognition problem, but nobody exactly agrees on
the pattern. My previously stated problem is that nobody knew how to
detect "closure to efficient causation."

> Judith: In medical circles, where
> "human life" is considered to be so different (so much more
> important) from other life, they've put a lot of work into trying to
> benchmark this. Ultimately, in medicine, certain physical processes are
> used to verify life. Many a time it has been reported in the news that
> someone was declared dead and was zipped in a body bag, and they weren't
> dead after all-- giving folks an X-Files moment, I'm sure!

HP: Detecting death is just a special case of detecting life.

> Judith: But I also have a new
> question, based on your post:
>
> HP:  I think of it like the epistemic principles in physics, like
> the requirement that all empirically verifiable models (laws) obey
> invariance principles.
> Judith: How would you define an "invariance principle"?

HP: This is what the epistemology of physics requires for ideal objective
models. Objective just means freedom from subjective influence or effects
on the lawful properties of the system being modeled. Invariant means
invariant with respect to the state (time frame, location, speed, state
of mind, language, culture, etc.) of individual subjects or observers.
Clearly, all our individual observations are subjective just because it
is the observer who must decide what, when, and how to make measurements,
and the resulting data will depend on the state of the observer. Laws are
discovered by finding the invariant relations among the data from
individual subjective measurements.

> Judith: The reason I'm asking is
> connected to the first question above. Basically, empirics are slippery.
> They're context dependent, just like everything else. Literally: In the
> eye of the beholder. 

HP: That is true of individual measurements. But that is not true of the
laws, the invariant relations between measurements.

> Judith: Empirics in science
> consists of a set of rules, isn't that correct? 

HP: No, that's not the idea. There are no rules for Empirics. This is
close to what Rosen calls encoding and what physics calls the measurement
process. Designing experiments and the measuring instruments is entirely
a creative art. It is also by far the most expensive part of physics
research. The ideal is to create a single experiment that might disprove
an entire theory. That means it is a crucial test and not just more data.

> Judith: Very much like our legal
> system. Being able to prove or disprove something in a court of law is
> not the same as arriving at truth or achieving justice. Empirics in
> science is a lot like that, it seems to me. It's the substitution of
> syntax for semantics, because syntax is seen as somehow more
> "objective" and "invariant".

HP: Legal decisions assume the laws are correct and try to apply them to
individual cases. Physics uses individual cases (measurements) to try to
discover new laws or disprove the old laws.

I would say you have syntax and semantics backwards. Physical laws are
expressed as syntactic formalisms. That is where invariance can be tested
by using the symmetries of the formalism (like time-reversal). The
meanings given to the symbols in the model, the semantics, is what the
empirics (encodings, observations, measurements) provide. 

Howard