Showing posts with label frontloading. Show all posts
Showing posts with label frontloading. Show all posts

Tuesday, February 24, 2009

On the Falsifiability of Evolution



I would give absolutely nothing for the theory of Natural Selection, if it requires miraculous additions at any one stage of descent.

Charles Darwin (in a letter to Charles Lyell)

********************************
To reject one paradigm without simultaneously substituting another is to reject science itself.

Thomas Kuhn (The Structure of Scientific Revolutions)

********************************

One of the websites I most love to hate, "Uncommon Descent", has a post regarding the falsifiability of evolution. Many "intelligent design" proponents (IDiots) are sophisticated enough to claim to concede "common descent", so the poster argues that biologists can't simply invoke Haldane's "rabbit in the pre-Cambrian" argument to drive a wedge between evolution and ID (as the discovery of a rabbit in the Cambrian would falsify both views).

What numerous ID proponents seem to argue for is "common descent with meddling", the meddling being of a conscious nature. But is that really common descent? Such a view suggests discontinuity, which could/should manifest in the fossil record. When arguing with "common descent with meddling" advocates, why shouldn't a "rabbit in the pre-Cambrian" be invoked as a falsification of evolution and a confirmation of ID? The rabbit is admittedly a dramatic example of discontinuity, but it differs with myriad other possibilities (e.g. a mammoth in Australia) only in degree. When an IDiot says that evidence for common descent can never refute ID, an appropriate response may be, "what bizarre version of common descent are you imagining?" Ordinary sexual/asexual reproduction plus occasional/constant creative input from a third/second party designer is quite a distortion of the historical concept. Visualize a "tree of life" with disconnected branches held up by skyhooks. Alternatively, you can place a fairy on every fork.

Another ID view, the front-loading view, is that evolution was somehow programmed from the very beginning to be driven, or accelerated toward some end; a single, primordial instance of meddling. Here, a "rabbit in the pre-Cambrian" would clearly falsify both this view of ID and evolution. So how does a researcher falsify either of these views? I'd say that the IDiots are playing a sneaky game here: they do their damndest to conform to standard theory, sprinkle in a tad of supernaturalism/teleology, and then ask biologists to falsify one view over the other.

Obviously, the more closely a parasitic view wraps itself around the other, the harder it is to falsify one view while leaving the other intact. Given a choice of views, it seems eminently reasonable to opt for that which is supported by peer-reviewed research, practicing biologists, and which dispenses of superhuman entities. Tediously, the "Uncommon Descent" crowd spends a great deal of energy refuting this, invoking worldwide academic conspiracies, the wisdom of engineers who aren't constrained by mere biology, and the Designer Without Inferable Motives or Identity.

Above, one sees that ID takes a number of forms. The rabbit in the pre-Cambrian falsifies some versions, but not others. But proponents like Dembski and Behe fastidiously avoid formulating specific theories of ID, preferring to harp on every perceived weakness in standard evolutionary theory. They wouldn't want to favor the frontloaders versus the tinkerers versus the constant interventionists versus the creationists. At the same time, when arguing that evolution is difficult to falsify over ID, the IDiots can conveniently pull out the version of ID that best makes the case.

Given the fact that some versions of ID are hugely parasitic on standard theory, piggybacking, free-loading at every opportunity, biologists are sometimes challenged to show how evolutionary theory provides any practical benefits that competing "theories" couldn't. The responses of biologists run the gamut, from Dobzhansky's famous "nothing in biology makes sense except in the light of evolution", to Coyne's recent argument that evolutionary theory, like cosmology, need only expand human knowledge to justify its existence.

Personally, it does seem that hardcore biologists sometimes exaggerate the case for evolutionary theory's utility in a clinical setting. A compartmentalizing creationist crystallographer could conceivably concoct cofactors (sorry, couldn't resist) that inhibit enzyme function, believing all the while that the active site was the result of conscious design. However, it occurs to me that one area that ID, in most of its incarnations (we can never say "all", given ID's aforementioned slipperiness) would fail horribly is in that of "ultra-selfish" DNA. We're talking about transposons, retroviral sequences, etc., that function solely to perpetuate themselves through the host's DNA. These sequences are problematic for ID, as they confirm an abundance of "junk DNA" in the genome, something most IDiots dispute with great tenacity. At the same time, these entities are strongly implicated in cancers and other maladies. Certainly, modeling and proposed treatments for such disease pathways should be influenced by whether the sequences are primarily "selfish" or "functional". To simplify, in one case it might be possible to "attack" the sequence and its [selfish] activity; in the other, one would expect any number of side effects.

My own dogma is that scientific insight arises in the minds of those who are already strongly grounded in correct views. Is it any surprise, then, that "ID theorists" contribute nothing to biological understanding?

****************

Below, I'll try to collect articles that do support the utility of evolutionary theory in medicine:

On the Utility of Evolution in Experimental Biology and Medicine

Hacking Evolution

Of What Value is Evolutionary Biology in Medicine?

Evolution as Policy, not Symbolism or Critical Thinking

Does Evolutionary Biology Make Predictions?


Of course, I understand that the poobahs of "Intelligent Design" may not be impressed. They'll argue that they don't deny "microevolution", so evolution of antibiotic resistance is no proof of standard theory's superiority to ID. They'll tell you that experiments in synthetic biology are an example of design, not evolution. And, of course, medical inferences based on "common descent" are taken to be "exactly what ID predicts".

The lack of creativity of ID proponents is sometimes mind-blowing. In addition to the aforementioned inability to come up with their own theories and research, there's the consistent flow of arguments that boil down to incredulity...I can't imagine it, so it can't be true. There's also this childish tendency to try to parrot back the criticisms of their opponents: evolution is a religious belief, evolution requires too much faith, blah, blah, blah. This is the tact adopted by children who, having been insulted, can't conjure up a decent comeback.

****************

Click here for a more artful "rabbit in the Cambrian" image.


****************

Saturday, August 23, 2008

Frontloading, Freeloading



Today's organisms are phylogenetically descended from others which were vastly simpler than they are, so much simpler, in fact, that it's inconceivable how any kind of description of the later, complex organisms could have existed in the earlier one. John Von Neumann

There are numerous grades of creationist arguments, from those that willfully ignore the convergent evidence from multiple branches of science, to those that make every effort to conform to available evidence without dispensing of that added element, the cosmic designer. The notion of "frontloading" falls into the latter. The idea seems to be that:

*Evolution did indeed proceed from a primordial common ancestor.

*Further tinkering may not have been necessary, but the potential for all sorts of otherwise impossible complexity was built into the DNA from the very beginning. Over geological time, organisms unfold into more complex, higher forms. Evolution, in other words, was directed from the very beginning.

Typically, a frontloader gets excited if/when some journal reports on a gene that serves no immediately obvious function in a creature that resides in a low position on the evolutionary tree, but is important for higher organisms. An example would be precursors for nerve tissue found in some sponges. Unlike the folks who do the research and write the articles, these frontloaders are quick to speculate that the gene in question was somehow embedded for future generations and is worthless for the sponge.

Despite the attempts to move the ID goalposts right out of the stadium, there are big problems with frontloading:

1) Exactly how many genes needed to be loaded into the primordial organism to encompass all of life's diversity over the last 4 billion years? Presumably, any genes that IDers typically label "irreducibly complex" (e.g. those for the blood clotting cascade) must be included here.

Michael Behe takes a particularly, uh, nuanced view on the subject. Some of his remarks seem to suggest he's not averse to a frontloading scenario. And he claims to endorse "common descent". On the other hand, he also argues that the majority of the 10,000 or so protein/protein interactions in a typical mammalian cell exhibit irreducible complexity. That's just one species! For a Behe-frontloading scenario to work, tens of thousands of different protein coding sequences would have to be packed into the primordial organism.

2) How were these useless genes preserved over 3 billion years or more? Useless genes, of course, have a habit of getting excised from the genome, so we should see evidence of some sort of mutation-checking system that is so powerful that a wide variety of genes can be passed through trillions of generations without so much as a single base pair substitution. We don't see that, and it's not for lack of trying...biochemists would love to get their hands on enzymes that copy with such fantastic fidelity. It's funny to see the same folks who have offered up so much blather on the topic of "genetic entropy" reverse course and speculate that certain useless genes have been passed through lineages unscathed for eons.

Then again, there seems to be another breed of frontloader who argues that none of these ancient genes were actually useless. They simply served functions other than what we see today. That's great...it's the same argument that biochemists have offered the IDiots to dispute "irreducible complexity" for the last 20 years (e.g. on its own, the spring mechanism of a mouse trap won't catch mice, but it will serve nicely as a tie clip).

3) Where are all these buried genes now?

I can see some "outs" for the above questions. As for the buried genes, IDers can wiggle out of that dilemma by fantasizing that we've reached the evolutionary juncture where all such genes have been jettisoned and are thus, conveniently, not available for study. With the advent of humanity, you might say, all biological innovation pooped out. Awfully sci-fi.

It's possible to imagine some sort of DNA or RNA software package that compresses a huge diversity of genetic outcomes into relatively few base pairs. I'd call this program "evolution", but the frontloaders must also explain how their pet examples of "irreducible complexity" get compressed and maintained in the primordial .zip file. If you respect the authority of Von Neumann, that's one hell of a hurdle.

Should I try to make that bacterium sweat, or grunt with effort?