googlef87758e9b6df9bec.html A Sure Word: Talk Origins Says Evolution Can Be Falsified

Monday, February 13, 2012

Talk Origins Says Evolution Can Be Falsified (TO) describes themselves on their homepage as “a Usenet newsgroup devoted to the discussion and debate of biological and physical origins.” That sounds somewhat benign. I would say they are more like a Usenet group devoted to ridiculing Christians in general and creationists specifically. I used to frequent their site fairly regularly but they no longer post new material so once I had read everything, I stopped going there.

Just the other day, I happened across their site again while researching something else. The article that I stumbled upon was in their “Index to Creationist Claims” and was a short post meant to rebut the claim that evolution cannot be falsified. You can read it here.

A good scientific theory should falsifiable. That's not my rule but one agreed on by the scientific establishment. The complaint raised by creationists (including myself on this blog) is that ToE cannot be falsified and so is not a good theory. No piece of evidence, no matter how damning to the theory, is sufficient to dissuade evolutionists from their beliefs. Any new find, no matter how much it might contradict previous understandings of evolution, is simply worked into the theory.

Normally, evolutionists are reluctant to suggest ways evolution might be falsified. I believe it's because they have been burned too many times in the past when some unexpected find upsets some point of their cherished theory. If they put something down in writing, they are putting themselves at risk that such a thing might actually be found someday. If they had been sincere, I would have to give kudos to TO for at least trying. However, the possible items they have suggested could disprove evolution cannot be serious. We'll look at them one by one.


I suppose they are saying that if the fossil record did not show any progression of life or change in species, then evolution would be falsified. I think it's funny they would suggest such a thing. We've already found millions of fossils and evolutionists have built their “nested hierarchy” based on what they have found. In other words, they've already spelled out what they identify as a progression in the fossil record so how could the fossil record ever be used to falsify the theory? It's a sort of prediction after the fact. This would be like me saying that the existence of God could be falsified by a lack of written revelation. Well, we already have the written revelation from God (the Bible) so this could not ever be used to falsify the existence of God. Therefore, it can't be a serious suggestion.

I once had an evolutionist suggest that evolution could be falsified by showing that animals don't reproduce. Yes, he was serious! Of course, animals are already known to reproduce so evos are safe from having their theory falsified by this test. In that same manner, “progression” has already been identified in the fossil record so TO can suggest that a static fossil record could falsify evolution without worrying such a thing would ever be found.


In Greek mythology, the chimera was a creature with a goat's head, a lion's body, and a tail ending in a snake's head. In this context, a chimera is any creature that is a composite of other creatures. A centaur, for example, was a composite of a human and a horse.

Bizarre creatures like centaurs or mermaids would be difficult to fit into evolution's precious “nested hierarchy” but neither would falsify evolution. Evolution uses similarities between animals as evidence of their relatedness. Therefore, if a chimera were found, it would not be evidence against evolution but would actually be evidence of a previously unknown relationship between different groups.

Let's be honest, TO knows we're not ever going to find a centaur. But say we found a creature that shared features with... oh, I don't know... say a reptile and a bird. Oh, wait! We've found that already. Have you ever heard of Archaeopteryx? Per Wikipedia, “Despite its small size, broad wings, and inferred ability to fly or glide, Archaeopteryx has more in common with other small Mesozoic dinosaurs than it does with modern birds. In particular, it shares the following features with the deinonychosaurs (dromaeosaurs and troodontids): jaws with sharp teeth, three fingers with claws, a long bony tail, hyperextensible second toes ("killing claw"), feathers (which also suggest homeothermy), and various skeletal features.

If we ever find a fish with hair, the headline the next day would not be, “Evolution proven wrong.” It will be, “New find shows fish more closely related to mammals than previously believed.”


This item suffers from the same flaw as the first item above. It's a test that has failed in advance. We already know that mutations accumulate in the genes (known as genetic burden). And since mutations already accumulate, TO knows that no mechanism preventing them from accumulating will ever be found. Evolution is safe from this test.


I can't tell if TO is making a joke here or if they're trying to be serious. How can they be serious? I mean, if God created an animal in front of my eyes, how does that disprove evolution? Even though God created that animal, everything else could have evolved! This isn't even close to disproving evolution.

Perhaps they are attempting to make a concession. They concede that, if God appeared and created something before their eyes, they would be forced to acknowledge that creation as revealed in the Bible was true. However, I don't believe even that would convince some people. After all, Jesus did turn water into wine and multiplied the loaves and fish. Even though He performed these acts of creation in front of literally thousands of witnesses, they still did not all believe.

IN CONCLUSION, this is merely more of the same. New finds in science have overturned previously held theories about evolution but nothing will ever threaten THE theory of evolution.


Steven J. said...

The theory of common descent implies a number of things: that some fish are more closely related to mammals than others are (e.g. lungfish are more closely related than trout, which are more closely related than sharks), and that no fish is more closely related to mammals than that fish is to birds, and that birds are more closely related to mammals (both being amniotes, a subgroup of tetrapods, a subgroup of vertebrates) than to any true fish.

Fish with true hair would be problematic (though there are enough hairlike structures in animals that this is probably not the best example, especially if we're discussing fossils). Deciding that, e.g. birds are more closely related to dinosaurs than to crocodiles is a small change to the "tree of life;" deciding that some fish are more closely related to mammals than, say, Dimetrodon (a "mammal-like reptile") is would be vastly more troublesome, and leave one to wonder whether anything about the tree was reliable.

Note that birds share more than feathers with some archosaurs (a group including crocodilians, birds, pterosaurs, and several more obscure and extinct groups): they share features ranging from gizzards to antorbital fenestrae in their skulls. It's been obvious for over a century that there must have been feathered archosaurs of some sort; that many of them (not just Archaeopteryx, but Anchiornis, Microraptor, Caudipteryx, etc.) are theropod dinosaurs doesn't make them chimeras but simply intermediates. Mammalian features, like the aforementioned three bones in the middle ear or a left but not a right aortic arch (okay, that's not likely to show up clearly in a fossil) would be chimeric features.

Steven J. said...

First, let me concede your last point: an instance of special creation would not disprove the evolution of other organisms -- although if the freshly-created organism fit neatly into the existing nested hierarchy of life, it would cast doubt on the value of that hierarchy as evidence for common descent.

Second, your overall argument seems to suggest that, as more and more evidence for a theory accumulates, and it becomes harder and harder to think of new observations that would be consistent with already-existing evidence but which would contradict the theory, the theory becomes less scientific, less falsifiable. What, at this late date, could astronomers observe that would falsify the idea that the Earth orbits the sun? What could chemists observe that would make them doubt whether atoms really existed? Yet is either theory unfalsifiable, merely because they have survived so many attempts at falsification?

There was certainly a time when it was quite conceivable that examination would turn up no mechanism for mutations. In the early years of geology, it was entirely conceivable that the lowest, apparently oldest rocks, would contain no species not known from living specimens (indeed, that they did not was for years a vexation to creationist geologists). For that matter, given that new fossil species are discovered every year, some of them quite large, I don't see how Mark Isaak can know that centaur fossils, or feathered bats, or birds with three bones in the middle ear and one in the lower jaw, will never be discovered.

Third, pedantic side note: scientists, of course, don't normally think of themselves as "trying to falsify" theories, even when they think of themselves as testing them. It's philosophers of science, not working scientists, who think of an attempt to find predicted evidence as an attempt to falsify a hypothesis (by finding contrary evidence). And philosophers of science don't really set the rules for science (neither does some formally-organized "scientific establishment). They're less like James Naismith writing down the rules of a new game "basketball," and more like people following scientists around to figure out what they're doing and why it works.

RKBentley said...

Steven J,

Evolutionary trees are drawn according to the imagined relatedness already existing according to the theory. It's a type of circular reasoning when evos then use the tree as evidence for their theory. Birds and dinos have their place on the tree because they are believed to have a common ancestor (or be parent/daughter). A fish with hair is only problematic because of where it is currently located on the tree of life. Such a discovery would not disprove evolution but merely require evos to redraw the tree and move the fish to a new spot.

While growing up, I remember reading that pandas were more closely related to racoons than bears. It wasn't until 1987 (I believe) that evos changed their minds and decided it was truly a bear. What exactly changed except having to redrawn the nested hierarchy?

You said, “your overall argument seems to suggest that, as more and more evidence for a theory accumulates, and it becomes harder and harder to think of new observations that would be consistent with already-existing evidence but which would contradict the theory, the theory becomes less scientific, less falsifiable.”

In another subject, I could see your point. I would have to think a while before I came up with a possible way to disprove heliocentrism. But is it really that difficult to think of a possible way to disprove evolution? Here's a suggestion: if we found a rabbit in the Cambrian, would that disprove evolution? I would say yes but I've published a quote from TO that disqualifies that possibility in advance. In summary, they said it might upset our understanding of how certain animals evolved but it's not enough to disprove the entire theory.

What about “irreducible complexity”? Darwin said that if a structure were ever found that could not have evolved in stepwise fashion, then his theory would “breakdown.” When Behe offered a few candidates for IC structures, he was assailed by the scientific community. His examples were attacked (with varying degrees of success) but ultimately, it was argued that IC is an argument from ignorance. Not knowing how such a structure evolved is not evidence that it couldn't evolve; it merely means we don't know how it evolved. So no structure, no matter how complex, is enough to meet this criterion.

My question still remains: if evolution is a good theory, how might it be falsified? I've heard some good ways, like the few I just suggested. However, evos don't like those. Instead, they recommend far weaker methods like the ones detailed on TO.

You said, “given that new fossil species are discovered every year, some of them quite large, I don't see how Mark Isaak can know that centaur fossils, or feathered bats, or birds with three bones in the middle ear and one in the lower jaw, will never be discovered.”

Hmmm. Perhaps you're right. Maybe someday we'll discover a centaur – though I doubt it. But I repeat my assertion that it will not dissuade the truly devout from their theory. It will merely force them to redrawn the nested hierarchies.

You said, “Third, pedantic side note: scientists, of course, don't normally think of themselves as "trying to falsify" theories.”

I am aware that scientists aren't thinking of ways to disprove evolution. But since the question is raised, can't any scientist think of some ways? Are the examples from TO really the best they have?

You said, “And philosophers of science don't really set the rules for science (neither does some formally-organized "scientific establishment).”

According to Wikipedia, “theories are preferably described in such a way that any scientist in the field is in a position to understand and either provide empirical support ("verify") or empirically contradict ("falsify") it.”

Also, see my post, “Evolution is Easy to Falsify.”

Thanks for your comments. God bless!!


Steven J. said...

Thank you for your response, and for tolerating the length of my replies.

Regarding giant pandas (there is a "lesser panda" that actually is more closely related to racoons than bears, and looks it), what changed was, I believe, genetic tests. Both bears and raccoons are "caniforme" members of the order Carnivora, more closely related to each other than either is to cats, to say nothing of horses or humans. Like the distinctions between which archosaur lineage gave rise to birds, this was a minor tweak, not a huge revision, to the tree of life.

Back in the 1920s, the geneticist Hermann Mueller argued that irreducible complexity (he called it "interlocking complexity," but it's the same idea) was a possible outcome of Darwinian evolution. Behe's argument that it contradicts neo-Darwinism assumes that [a] mutations can only add components, not delete or modify the function of existing components and [b] that structures must serve the same function throughout their evolution: that a precursor structure could not be functional for something other than what its descendant structure does. The first assumption is bizarre indeed; the second assumes that the evolution of complex molecular systems cannot work like the evolution of organs and limbs (where, e.g. a penguin's wing is a modified flying limb, which is a modified grasping limb, which is a modified walking limb, which is a modified swimming fin: repeated changes of function. Behe has moved beyond "irreducible complexity," and so have his critics, in response to this.

I cut, from my original response, a brief note on the Quine-Duhem thesis (a critique of the "silver bullet," one ugly fact killing a beautiful theory version of falsificationism) that "hypotheses cannot be tested in isolation." If a paleontologist finds a rabbit fossil in Cambrian layers, he is really testing many hypotheses at once besides "placental mammals evolved from, and hence not before, lobe-finned fish." He is testing hypotheses as obvious as "these are really undisturbed Precambrian sediments (e.g. the rabbit is not modern and just reworked into Ediacaran rocks) to the epistemologically nigh-paranoid ("I can tell a rabbit when I see one") to hypotheses so seemingly obvious only such an observation would make you realize you were testing them ("no future scientist will build a time machine and send a rabbit back to the Precambrian").

That the entire fossil record of vertebrate evolution is so weirdly and spottily preserved that every other group except lagomorphs first evolved hundreds of millions of years before the first fossil members of that group show up (which would be necessary to reconcile the current evolutionary tree with Precambrian rabbits) is arguably less plausible than bunny chrononauts.

For any theory, not just for evolution, you would need a pattern of facts, not just a single "silver bullet" disconfirmation, to overturn it.