Re: [asa] Falsification of Darwinian mechanisms

> What would falsify the random mutation/natural selection hypothesis, or is
> it simply a Universal Explanatory Mechanism which can explain everything by
> simply inferring from present knowledge that the large gaps in our knowledge
> may someday be filled? I've got some suggestions.

Arguments in both directions often fall into an all or nothing error.
It is practically impossible to prove that there are absolutely no
exceptions to random mutation+natural selection, or that random
mutation+natural selection never works, because we cannot investigate
every detail of every organism that ever lived. Antievolutionary
arguments routinely say "here's a problem for evolution, therefore all
evolution is bunk" and anti-ID arguments often jump from "this ID
argument is no good" to "no id".

> 1. If species were found deviating significantly from the predicted "tree of
> life", meaning that descendents appeared to be disconnected from their
> ancestors, would that falsify the evolutionary model?

If a species were found that could not be fit into the tree of life
(i.e., not merely that we had it on the wrong branch nor horizontal
transfer or other complicating pattern within the tree), that would
pose a significant problem for explaining its origins via natural
selection and random mutation, though it would not negate the fact
that a whole lot of other organisms fit in quite well.

> Isn't it true that the "predicted sequence" has been a work in progress over the years, requiring some adjustment as new data is discovered? If so, how much could this be seen as retrofitting the predicted model to the data as a UEM, as suggested above?<

If you look at how well a particular grouping is characterized and
supported, you have a pretty good idea of whether it will continue to
be supported. For example, my DNA data suggests that the current
genus-level classification of freshwater mussels needs a lot of work.
How did things get assigned to their current genera? Mostly on one or
two features, without regard to other morphology, and in many cases
particular genera were a holding bin for things that didn't fully
qualify as another genus-not a good way to characterize an
evolutionarily coherent whole. On the other hand, most of the
species assigned to most of the genera are fairly closely related to
each other.

Working with DNA data for bivalves generally over a period of years,
certain groups were consistently supported, while others show up in
one analysis (usually with weak support) and not in another. Go back
to morphological features and/or the fossil record, and you can find
good connections.

However, it is true that a given analytical technique may make a tree
out of any data. You need to pay attention and judge whether the
results are actually meaningful (plus wathcing out for contamination,
etc.).

> 2. If the genetic/molecular evidence showed an entirely different hierarchy
> of ancestor/descendent relationship and nearness to one another from the
> phylogenies previously determined from other means, that might constitute
> strong evidence against the standard model of common descent. As it is, the
> confirmation from genetics does seem to be strong evidence for the standard
> model, because it confirms the previously developed "tree of life" sequence.
> But does it really? How much adjustment is there being made even now, as
> genetic evidence is discovered and added to the present body of knowledge?
> I'm not asking for the scientific model to be fixed and unmovable in order
> for it to be valid, because that is unrealistic. But I am asking, what (if
> any) changes are made to the so-called "standard model" as new evidence is
> discovered? If the standard model is being corrected by reference to
> genetic evidence, how can this be considered a "proof" of the previously
> existing model? What kind of falsification of the standard model could be
> anticipated as evidence against the hypothesis of common descent, if the
> model is continually modified by new information? Could it potentially ever
> be falsified?

Again, traditional classifications are strong on some points and more
debated on others. Molecular data have pointed to new groups, not
previously recognized, and have contradicted some popular
classifications (I've been referring back to an older but longer
invertebrate zoology textbook on some points and cringe at some of the
pronouncements on relationships that are now strongly contradicted by
molecular data), but going back generally turns up overlooked
similarities that parallel the molecular pattern.

> 3. Once specific evolutionary steps in the predicted model are identified
> and well-defined, could the hypothesis of naturalistic evolution be
> falsified by showing the improbability of making the transition(s) from
> point A to point D? This seems to be the approach used by many in the IDM,
> however I don't see that this method of falsification is likely to be very
> effective. It always relies on gaps in our knowledge to disprove
> naturalistic explanations, and it's very difficult to prove that the gaps
> will never be filled. However, could evolution of various specific
> structures from primitive to complex forms be potentially falsified at some
> point, by showing that the complex forms could not possibly have been
> derived from simpler forms? What sort of evidence would be necessary to
> demonstrate beyond reasonable doubt that such transitions could never have
> occurred naturally? We have seen statistical approaches which rely solely
> on the probabilities of random mutations to arrive by fiat from random
> molecules to the final configuration, which isn't realistic because it
> doesn't take into account transitional form and function of more primitive
> structures. If that approach is invalid, are there means by which specific
> evolutionary steps could be absolutely ruled out as impossible to occur
> naturally, and therefore by inference constitute a failure of the
> naturalistic evolutionary model? Or will scientists continue to fall back
> on explanations such as Gould's punctuated equilibrium to justify the
> speculation of dramatic but unexplainable jumps in complexity?

If it's possible to accurately characterize the probability of a given
mutation, the probability of each set of mutations that could produce
a reasonable functional equivalent, and the probability of different
ways one could make life without that particular feature, then we can
estimate the probability of something happening randomly. I am not
optimistic that it's feasible, but can't think of any absolute barrier
to getting that sort of data.

Punctuated equilibrium is a pattern in which organisms change little
for much of the time and then show rapid change to a new state. This
is to be expected whenever selection favors the existing features
instead of favoring change until environmental change or new
competitors or predators or deenses in something else necessitate a
shift. It's not a cover for dramatic but unexplainable jumps in
complexity.

> 4. Could Mike Behe's approach, given much greater development of scientific
> knowledge, someday provide sufficient falsification of naturalistic
> evolutionary explanations? I.e., given much more knowledge of the
> underlying biochemical and genetic structures, evolutionary history at both
> macro and micro levels, and knowledge of form and function of primitive
> forms that were probable precursors to the complex versions, would it be
> possible at some point to prove that the intermediate structures and
> functions for a given organ could never have survived numerous transitional
> forms, because the organism would not have been able to survive, reproduce,
> etc. in the interim? To my way of thinking, as in the previous paragraph, I
> don't see our knowledge ever expanding to the point that we could absolutely
> rule out unknown natural mechanisms, and thus we can never by this means
> construct a suitably definitive falsification, except through the argument
> from incredulity (we can't possibly imagine how it could have arisen
> naturally, so therefore it must have been designed).

It seems to me that, in theory, it would be possible, but in practice
there are both the problems of getting the required knowledge and the
distraction of all the bad arguments made by ID in a determined effort
to prove that complex biochemical systems are "Designed" rather than a
serious effort at identifying what truly distinguishes non-natural
design.

> Are there any other means by which the hypothesis of common descent and its
> mechanisms could truly be falsified? If there isn't any way it could
> realistically be falsified, what does this say about its status as a
> scientific theory? What does it say about the claims that other
> explanations (such as ID) are not science, because they can't be falsified?

If something fits into the total tree of life, but were to turn up at
the wrong time, it would be a problem. E.g., Cambrian mammals, modern
humans in the Eocene, etc. This is complicated by caveats about the
quality of the fossil record, but significant deviation from the
pattern of primitve earlier and advanced added later would be
anomalous.

If there were no genetic material, or if survival did not vary between
individuals in a way that relates to the genetic material, biological
evolution would not work.

-- 
Dr. David Campbell
425 Scientific Collections
University of Alabama
"I think of my happy condition, surrounded by acres of clams"
To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.