Re: Function, Symbiosis, Parasitism

["Dan Fiscus" <***>]

Tim,

Hi! Jumping in on your list...wanted to reply to your interesting round-up
of complex issues in ecology...

Tim Gwinn wrote:
>  Mainly I was wondering about the question of
> "improving" our (human) interaction with the biosphere; loosely, "how do we
> become more symbiotic and less parasitic?", along the lines of an earlier
> remark by Judith.

snip

> There seems to be little in the way of first principles or
> absolutes to rely upon; instead, all these notions are relative to some
> context.

snip

> When we expand into thousands and thousands of species, it seems
> mind-boggling: they are all interacting indirectly via the environment (to
> one degree or another), and in so doing they are *cultivating* that
> environment, making changes in some cases to it, and in other cases
> providing some kinds of stabilizing effects. Some species may also be
> interacting directly. How do we begin to describe this?   (Recall that systems
> in isolation do not provide the information needed - they need to be
> observed in their functional context.)

One point here that is like the inverse or complement of your notice
of the difficulty of modeling, addressing, understanding a system with
so many variables (ecosystem, biosphere) is that with just three
variables, components or perhaps functions we can model a system or
system + context that is infinitely complex as in completely
unpredictable.

I posted this idea before on the other Rosen list saying the the threshold
of near infinite complexity is very low - just 3. The easiest analogy or
evidence is the 3-body problem in physics.

Taking this as a positive or useful observation, we can say that we can
depict perhaps the vast majority of dynamics of atleast some aspect
of any complex system with a model with just three variables or
interdependent functions. Especially if we choose the three functions
or components with care and informed by Rosen (and I could add
Ulanowicz and the Odum brothers and others) and also Kercel's and
others' work on impredicativity, multi-level causal entanglement, etc.,
we can go very far with a small number of variables or functions.

It is kind of like turning a derogative statement like "statistics can be
made to say anything" into a positive.

My attempt at such a general model has been to work with the three
functions of 1) molecular string composition (as in the autotrophy or
"self-feeding" of plants), 2) molecular string decomposition (as in the
heterotrophy or other-feeding of animals) and 3) coupled
complementary processes (as in the interdependence of 1 and 2. From
this simple set of functions I have suggested that early eecosystemic
life generated the higher order functions of metabolism and genetics,
both of which I see as essentially based on variations on the them of
molecular string composition coupled to decomposition. As Howard
Odum has written, this ecosystemic set of functions (he called them
production and respiration; Rosen called them metabolism and repair
although with different wording, context) could also plausibly
generate cells.

One other way to modify the same model would be to choose the
three functions as 1) entropic evolution - as  in the dissipation and
disordering of radiation, 2) syntropic evolution - as in the
aggregation and ordering of gravitation and 3) coupled
complementary process of 1 and 2. In this scheme, functions 1
and 2 are both abiotic and are both closed future forms of
evolution - they end at stable attractors of heat death or "ice
death" (frozen crystalline lattice) respectively. But via function
3 the whole system becomes biotic and represents open-ended
or open-future evolution.

> And if we shift from individuals in species to an evolutionary perspective,
> considering the generational trail of a species as the "individual" under
> study, then relationships that appeared symbiotic may now appear parasitic,
> particularly if the species have widely different generational timeframes.
> And in this evolutionary perspective, it would be unusual if E had no
> notable changes, which in turn will affect the species in that ecosystem,
> etc.
>
> When are we humans guilty of wiping out a species? When it is a passenger
> pigeon, Tasmanian tiger, etc.
> When are we humans "successful" at wiping out a species? When it is polio or
> SARS.(At least, that is our goal.)
>
> How do we characterize this value judgement? And, how do we then make policy
> generally about the biosphere, and our place in it, given these kinds of
> considerations?

Value and good vs bad are very hard indeed to link to science. I have
struggled with this since 1993 when I started a research project to
develop an indicator of soil health for agricultural ecosystems. This
indicator was supposed to be able to take measurements of biotic
and/or physical/chemical aspects of a soil system and produce an
evaluation as to whether the soil was good or bad for agriculture.
It was also supposed to be quantitative - to say how good or how bad,
to indicate levels or degrees of goodness.

Precisely because of the difficulties you have outliuned and juxtaposed,
I was not able to bridge this chasm between supposedly valueless
science and measurement and necessarily value-loaded assessment of
soil health, condition, goodness for (sustained, sustainable) agriculture.
Or, I was not able to bridge this gulf given the set of science, biology
and ecology facts, first principles, assumptions, models I had learned
about thus far in grad. school. The only solution to this dilemma that
I found, the only way to bridge science and value, was to redefine life
as not a propery of an organism, but of an ecosystemic, autocatalytic
network of multiple inter-dependent functional aspects. I bit later I
found Rosen, Ulanowicz and Lotka and their works all served to
corroborate what I suspected - that mechanistic, reductionist biology,
ecology, science were insufficient for such conceptual models that
unified measurement and value.

The central function or issue that I focus on as basis for value and
measurement is open-ended evolution. This is essentially the same as
survival, continuation, endurance, sustained existence. If we adopt
this as the central issue, I think it is easier to make value judgements.
But it is still very hard to do. Two factors that complicate this
approach are 1) death is required for continued life (the deomposer
and entropic functions require death, decomposition, recycling as part
of the cycle of generating new forms to deal with environmental
uncertainty and perpetual change) and 2) this approach only deals
with quantity of life (as in time duration quantity) and does not
address quality of life. Clearly at some point and on some scales
quality of life trumps quantity - some humans would rather die than
be slaves, for example.

 > There are so many questions at this point, so many possible directions to
 > go. For example, suppose X & Y form some strong symbiotic relation.
Suppose
 > also that Y is in a parasitic relation with Z, that may ultimately
lead to
 > the demise of Y. In this case, X's survival may also be in jeopardy if it
 > depends too strongly on Y. In such a situation, there is nothing inherent
 > about components E, X, Y or Z that is "bad", it is simply a result of
their
 > contextual relationships. (In the spirit of Hamlet: "there is nothing
either
 > good or bad in a functional component, but the contextual
relationships make
 > it so.")
 >
 > Importantly, there is nothing inherently good or bad about the specific
 > behavioral *mechanisms* that a species might have. What is important
is how
 > the species performs in a functional role.

I agree with these points, yet the urge is still their to label some kinds
of dynamics or relations as good. I would say one key is that actions
that enable life - all of life, life itself, taken as a unified whole - to
continue, persist, survive are good, while those that reduce life's
ability to live on ("live long and prosper") are bad. Transferring the
3-function model to more common human issues, we could discuss
1) renewable energy, 2) recycling materials processes, and 3) the
coupling of 1 and 2. Actions that enhance one of these functions are
ones that I predict aid life's abilty to survive and thrive (good) while
actions that ignore, lessen or weaken these functions threaten life's
abilty to survive and thrive (bad). Again with caveat of entire
missing piece of "quality of life".  I would not think a nazi police
state that managed to do 1+2+3 above at the expense of human
dignity and kindness would be good at all. But for the biophysical
realm, I think these three are pretty darn "good" for modeling good
in terms of a human relationship to a planetary home environment,
assuming quality of life is also OK.

I will do a separate post that has two
quotes that may add to the picture you are painting here. One from
Aldo Leopold who gives an assertive and confident definition of good
and bad as relating to biology and humans. The other from notes in
the back of the Tao Te Ching (Stephen Mitchell version) in which is
implied the impossibility and wrong-headedness of trying to make such
distinctions and definitions.

Thanks for your thoughts,

Dan