Red Queen Inching Forward? Does Collective life have a direction?
Tom Hendricks
If life, meaning all living things together, is better at surviving
and evolving now than it was before, then could there be selection on
all life?
If all life or collective life is better adapted to its environment
then life of 1,2, or 3 billion years ago,
If biomes are now more stabile
if life is better adapted to its various environments,
If life is more responsive to changes in its various environments,
If there are collectively more defenses against environmental shifts
and changes - example better gene regulation,
If life can better survive a catastrophic events now then in its past
- example the two great extinctions,
Is there selection on collective life?
Dawkins says selection is at the gene level. But perhaps the gene is
working for collective life - life over time that better fits the
environment of the earth.
To put it another way, is the Red Queen inching forward?
This suggests 'progress' in evolution which is highly controversial.
William Morse
I have no problem with progress in evolution, and have previously argued
for it.
As to the level of selection, the answer is that selection _will_ act at
whatever level it _can_ act. It has to, because it is simply the result
of excess reproduction in combination with heritable difference. So your
question becomes that of whether biomes function as a level of
selection. Are there combinations of organisms which try to spread out
of their current geographic region as a group, and is their success in
doing so attributable to characteristics which belong to the group or to
individual characters.
The standard response is that the success is due to individual
characters, that the groups only form due to combinations of the
individual characters, and that therefore there is no difference between
groups that is heritable at the level of the group and thus nothing on
which selection can work.
This response may be correct. but I think it is worth a second look.
E.O. Wilson is convinced that group selection works in the eusocial
insects, and if it can work there I don't see why it can't work at the
level of the ecosystem.
William L Hunt
<tom-he...@att.net> wrote:
>There are many levels of selection. Here's one to consider.
>
>If life, meaning all living things together, is better at surviving
>and evolving now than it was before, then could there be selection on
>all life?
It is at least logically possible for selection to operate at the
level of gene, organism, group, species, ecosystem, etc. But all these
examples logically require that there be many such entities (or at
least more than one) so that there is some choice possible. You must
be able to choose some and not choose others. That is what selection
means.
You are proposing a highest level that is "all living things". But
no selection is even logically possible at this level because there is
only one entity (all life) and no alternate choices are even possible.
So the answer to your question is "no", there cannot be selection on
all life.
William L Hunt
.....
[snip]
.....
Guy A Hoelzer
biosphere, then natural selection cannot act on it as a whole. On the other
hand, processes of self-organization leading to coherent system function at
the biosphere level is plausible. In my personal view, function of the
biosphere is best understood as thermodynamic (increasing the rate of
entropy production).
Guy
in article hfjaq2$1ncj$1...@darwin.ediacara.org, Tom Hendricks at
tom-he...@att.net wrote on 12/7/09 8:34 AM:
John Edser
> As to the level of selection, the answer is that selection _will_ act
> at whatever level it _can_ act. It has to, because it is simply the
> result of excess reproduction in combination with heritable
> difference.
JE:-
Natural selection is not the result of "excess reproduction" it is the
result of MAXIMAL reproduction which is not at all the same. Maximized
reproduction can only be successfully achieved at ONE level because more
than one reproductive maximand per level always provides a conflict. The
result: just the one fitness reproductive maximand per level remains. If
the proposed levels of selection are NESTED then only the largest, i.e.
the most outer level which contains all the others as nested (proper)
fitness sub sets can possibly provide this critical, single reproductive
maximand.
Contained nested levels are selected to provide reproductive MINIMANDS
(they are selected to always reproduce themselves a little as possible
not as much as possible). This is because only a minimand/maximand
nested fitness configuration represents the most efficient for the
selection of each nested subset. Since resources always remain finite,
any excess reproduction of nested levels can only reduce the single
reproductive maximand which is always found at the largest level which
contains (set nests) all the others. The net result of excess production
of any nested in fitness part is to LOWER it's own reproduction where
it really matters: at the Darwinian, single maximand level. Excessive
nested set reproduction can only provides a REDUCTION in the total
number of fertile forms reproduced per parent per population and thus
itself, because it squanders limited resources. There is simply, no
selective gain for any nested part to reproductively increase within
it's own selective whole at the cost of reproducing that whole.
The reproductive minimand/maximand association between nested levels of
selection remains typical of any organism structure. When this critical
association becomes broken, e.g. via cancer cells which excessively
reproduce themselves within a nested reproductive configuration, they
eventually cause the death/debilitation of the nested structure they are
all selected within, _inevitably reducing their own Darwinian fitness in
the process_.
Natural selection will indeed "act at whatever level it _can_ act".
However, empirically, these "can do" selective levels are restricted to
just two:
(1) Contesting fitness maximands.
(2) Cooperating fitness minimands.
A typical example of (1) is any Darwinian population, i.e. any
population of fertile forms, each contesting the other by simple default
via the total number of fertile forms each independently reproduces per
population. An example of (2) is the limited and critically structured
reproduction of any body part. This INCLUDES genomic genes. While the
excessive reproduction of dependent (nested) in fitness body parts can
and does occur, it reduces the fitness of the fertile organism whole it
remains just a part of, REDUCING IT'S OWN DARWINIAN FITNESS. In short,
(1) represents a population while (2) represents the organism.
Populations evolve but only because adult individuals are selected. IOW,
populations are not selected and individuals do not evolve. Within
Darwinism (but not within Neo Darwinism) these represent critical
falsifications.
The distinction between organisms and the populations they
(reversibly) form has become hopelessly blurred within Neo Darwinism.
This is because mathematics cannot distinguish between nested and
intersecting sets of fitness. An example of the ongoing confusion this
can produce is represented by Wallace's group selection (the independent
selection of populations of adult forms). This allowed a MODEL of
fitness altruism at the Darwinian fertile organism level which was and
remains prohibited by Darwinian THEORY as a falsification. Because the
fitness of any Darwinian population is always the simple sum of the
Total Darwinian Fitness of each population member, populations cannot be
individually selected simply because they are NOT the fitness nested
entities represented by (1). If you attempt to select any population,
all you end up doing is selecting each member of that population on a
individual-by-individual basis only because the fitness of
a population is the simple sum of the fitness of each member (additive).
The fitness of each Darwinian individual within a population only
reversibly intersects (it does not irreversibly set nest) allowing (1)
NOT (2). Additive in fitness entities are not selectable entities. This
is because they represent (1) NOT (2) which is non additive. Via a
reverse of the same argument, fitness nested genes cannot provide a
reduction of total fitness at the fertile organism level because, if
they do, they are selected against at the single, fertile organism
maximand level.
Selection for Darwinian fitness altruism cannot be selected for from
above via group selection or from below via Haldane and Hamilton's
inclusive fitness simply because (1) and (2) do not represent the same
biological entity. However, within non falsifiable, oversimplified
models of falsifiable Darwinian theory (1) and (2) have become
equivalent. This is only possible when not a single fitness constant
remains introduced into evolutionary theory. In Darwinian theory this
critical constant is implicitly represented by one, single fitness
maximand per nested fitness set per population: the total number of
strictly fertile forms reproduced per parent per population. At this
single selective level, everything without exception is selected within
falsifiable Darwinism.
Regards,
John Edser
Independent Researcher
John Edser
>> There are many levels of selection. Here's one to consider.
>>
>> If life, meaning all living things together, is better at surviving
>> and evolving now than it was before, then could there be selection
>> on all life?
> It is at least logically possible for selection to operate at the
> level of gene, organism, group, species, ecosystem, etc.
JE:-
Hi William,
While the above remains (uncritically) today's Neo Darwinian wisdom,
empirically it was and remains, false. If the fitnesses of the proposed
levels are NESTED then it is only logically possible to select the
single largest nested level which contains all of the others. The
result: only fitness at just the ONE level (the largest nested level)
can provide a fitness maximand. Only the one nested fitness maximand is
possible because more than one, then they necessarily contest, reducing
one in favor of the other. Contained nested levels constitute dependent
parts which are selected to remain minimal and not maximal. For example,
the fitness of a single genomic gene's phenotype remains 100% nested
within a single fertile organism fitness maximand. This means that gene
replication within a fertile organism will be selected to remain minimal
and not maximal. Excessive gene or cell replication within fertile forms
only constitutes waste reducing fitness at the highest nested level, _on
which these same gene replicates remain 100% fitness dependent_. Note
that a gene cannot possibly be selected via its genotype (as Neo
Darwinistic models allow) because the non reversible logic of the
revised central dogma only allows DNA/RNA to code for polypeptides
excluding the reverse as a falsification.
If fitness levels do not nest then they can only intersect. The
difference: while all nested fitnesses remain dependently selected at
just the one, largest nested level (which alone can present a
falsifiable fitness maximand) intersected fitness remain 100% fitness
independent of each other. Within the biological sciences nested fitness
constitute bona fide organisms. OTOH, intersecting fitness constitute a
population of organisms. Since only fertile forms can even possibly pass
on their genes, fitness at the adult level can alone present a
falsifiable maximand fitness within nature.
The reason why group selection fails is because groups of independent in
fitness fertile organisms DO NOT constitute a single selectable entity.
To be so selectable, their fitnesses must nest, i.e. not just intersect.
If you attempt to select a single group of intersecting fitnesses all
you can possibly achieve is to select each nested fitness individually
but equally (for or against). IOW, selection _always_ operates at the
single Darwinian fertile organism level within group selection because
the fitness of any one population remains the simple sum of the fitness
of each fertile form within it. The net result is that group selection
cannot possibly provide fertile organism fitness altruism as was
originally proposed.
Virgil
Guy A Hoelzer <hoe...@unr.edu> wrote:
> Selection requires a set of objects to select among. If there is only one
> biosphere, then natural selection cannot act on it as a whole.
The bioshpere now, as always, is sufficiently non-homogeneous to allow
for differing selections depending on local conditions.
Guy A Hoelzer
would perhaps be selection among ecosystems.
Guy
in article hg1o35$1q6l$1...@darwin.ediacara.org, Virgil at Vir...@home.esc
wrote on 12/12/09 7:46 PM:
John Edser
Yes, but that would not be selection at the level of the biosphere. It
would perhaps be selection among ecosystems.
JE:-
Hi Guy,
I think you will agree that a valid evolutionary theory requires a
minimum of two _contradictory_ biological objects:-
(1) A discrete entity on which natural selection can _independently_ act.
(2) A group of the above, i.e. a minimum of two, which evolve.
At some stage, 2 (a population), cannot also be 1 (a unit of selection)
otherwise _this basic minimum of two contradictory biological objects is
not achieved_ disallowing a valid theory of evolution by natural selection.
Therefore my question: what population type, i.e. biological object type
2 do you (or anybody else here) define which cannot also constitute an
independent selectee i.e. biological object type 1 ?
Tom Hendricks
bigger and bigger brains as you go through time,
but it's a stepwise pattern, not a gradual thing. The raw statement of
punctuated equilibria removed the old
convenient element of directionality for long term trends in the
fossil record.
But if you concede that there's a directionality over time, what is
the explanation?
Guy A Hoelzer
I agree with your two points. However, we disagree on your subsequent
claim. In my opinion a population can also be a unified entity upon which
selection can act. In fact, I think this accurately described every level
of biological organization that could conceivably be a target of selection,
including the individual organism (a population of genes; a population of
cells for multicellular organisms). Ecosystems can (at least theoretically)
qualify as selectable objects. In fact, I would argue that most individual
organisms are actually ecosystems themselves. For example, you could not be
a functional living system without the functional integration of many other
species in and on your tissues. All eukaryotic cells are functionally
integrated ecosystems composed of the remnants of distinct species. These
are just familiar examples, but my view is that all levels of biological
organization represent populations or ecosystems that are functionally
unified as objects. The functional integration of ecosystems is a long
standing controversy in ecology, and I think it can be very tricky
identifying the bounds of objectified ecosystem units. I do not believe
this represents a theoretical problem with selection at the ecosystem level,
though.
Cheers,
Guy
in article hgbe16$2rv8$1...@darwin.ediacara.org, John Edser at
ed...@ozemail.com.au wrote on 12/16/09 11:56 AM: