The Logic of Science

This is one of the most common arguments that I hear from creationists. The claim is that if evolution is true, then in the fossil record we should see lots of intermediate species. In other words, we should see transitional fossils showing the changes from one group of organisms to the next. So far, we are on the right track. Evolution certainly does make that prediction. The problem is that creationists then proceed to make the factually incorrect claim that no transitional fossils have ever been found. The reality is that we have hundreds of transitional fossils, and examining them not only provides extremely strong evidence for evolution, but it shows a critical logical flaw in creationists’ thinking.

First, it is important to realize that the definition of “transitional fossil” is totally arbitrary. Any species that does not go extinct is an intermediate. In other words, evolution generally does not reach “end points” (living fossils being one debatable exception). To put this another way, we tend to think of the species that are alive today as the final products of evolution, but they aren’t. If we could go forward several million years into the future, we would see very few of today’s species but we would see plenty of their ancestors. So everything is an intermediate between the previous generations and the future generations. Nevertheless, I understand creationists’ point that we should see obvious transitions between major groups, so I’ll focus this post on them.

To understand the full story of transitional fossils, we have to go all the way back to Charles Darwin. In Darwin’s day, paleontology was a budding field, and the fossil record was extremely incomplete, so he did not have any transitional fossils, but he predicted that if evolution was true, we should find the intermediates once the fossil record is more complete. Further, being a good scientist, Darwin proposed that the fossil record should be a good way to test his theory. You see, science works by making testable predictions and proposing falsifiable hypotheses. Transitional fossils meet both of those criteria: if evolution is true, we should find numerous intermediates, and failing to find them would discredit (falsify) the theory. Creationism, of course, makes the opposite prediction. If all organisms were created basically in their present form, and if evolution is limited to occurring within a “kind,” then there should not be any intermediates between major groups.

Now, let’s jump back to the present and see which predictions have come true. Before I go any further, I want you to picture in your mind what an intermediate between a dinosaur and a bird would look like. No doubt, you have pictured an animal that is hallway between the two: an animal that has some of the features of a dinosaur and some of the features of a bird. That is also what evolution predicts that we should find. Now, the question is, “do we find such creatures in the fossil record?” The answer is, of course, a resounding YES!

The most famous intermediate between dinosaurs and birds is of course the Archaeopteryx, but it’s not alone. We have other examples such as Confuciusornis, Jeholornis, and Sinornis just to name a few. All of these fossils not only display some traits of dinosaurs and some of birds, but they show a spectrum of change. This is critically important. If creationism is true, and organisms are relegated to distinct kinds, then there should be clear gaps between each kind, but if evolution is true, then there should be a gradual transition resulting not in two distinct kinds, but in a spectrum with multiple intermediate species such that there is no clear break between the two groups. Evolution’s prediction is, of course, precisely what we find. On the one extreme, we have dinosaurs such as Triceratops and Stegosaurus that are clearly reptiles, but the group of dinosaurs know as Theropods already show some bird-life features. For example, the furculum (commonly known as a wishbone) was once thought to belong exclusively to birds, but we now know that some Theropods had it. Further, many of these dinosaurs had feathers (again a trait that was once thought to be unique to birds). Dinosaurs known as Oviraptors are even suspected to have guarded their nests (a trait that is rare among reptiles). In fact, many of these dinosaurs are so bird-like that scientists are flummoxed about whether they are birds or reptiles. There is no really clear point at which they stop being reptiles and start being birds (exactly as we would expect if evolution was true).

Jumping ahead a few million years, we arrive at the famous Archaeopteryx. At this point, everyone is in agreement that it is a bird and could fly for short periods, but it is still distinctly different from our modern birds. For example, it had a skull with teeth, long forearms, a long tail, a small sternum, and largely unfused digits on the hands. All of these features are found in dinosaurs and modern reptiles, but not in modern birds. In fact, despite being considered a bird, the skeleton of Archaeopteryx is so very dinosaur like that scientists once misidentified an Archaeopteryx skeleton as the skeleton of a Theropod dinosaur know as Compsognathus!

Archaeopteryx is as clear of an intermediate as we could ever hope to find. Although it is considered to be a bird, it is so dinosaur like that scientists once misidentified one of its fossils as a Compsagnathus (a dinosaur). Image via The Pterosaur Heresies

As we move past Archaeopteryx and closer to modern birds, we slowly accumulate more bird-like features. Confuciusornis, for example, had a beak very much like a modern bird, and Jeholornis had forearms with flight specializations that were not found on Archaeopteryx. One of the best examples of these advancements is Sinornis. It was far more bird-like than Archaeopteryx, possessing a more well developed sternum for flight muscle attachment, greatly reduced tail, and fused digits. Nevertheless, it still retained some dinosaur features such as a skull with teeth.

Finally, we arrive at our modern birds. These are clearly distinct from dinosaurs, but they still retain many reptilian features such as scales. Take a look at the feet of birds such as chickens and herons, for example. The structure of their feet is nearly identical to the structure of the feet of Theropods such as Tyrannosaurus rex. Further, we can test this relationship using genetics. If birds did evolve from reptiles as the fossil record suggests, then we would expect the genetic codes of birds to be much more similar to reptiles than to any other group of animals. Again, this is a testable prediction that evolution makes and creationism does not make. The results are, of course, that the DNA of birds is more similar to the DNA of reptiles than it is to the DNA of any other group of modern animals. So both the fossil record and genetics are in agreement with the predictions of evolution.

An even more astounding test of evolution’s predictions came in 2007 when scientists were able to obtain soft tissue from a fossilized Tyrannosaurus rex. They weren’t able to actually get DNA, but there were able to get protein sequences. Proteins are made of chains of amino acids. There are 21 different amino acids and their order determines the protein that is made, but for most proteins there are multiple amino acid sequences that will result in the same protein. Because of this, they are very useful for determining the relatedness among organisms because closely related species tend to have similar amino acid chains (e.g., all mammals have protein sequences that are more similar to each other than they are to the protein sequences of amphibians). This gives us another clear prediction. If birds evolved directly from dinosaurs, then the protein sequence of the T. rex should be more similar to the protein sequences of birds than to reptiles, but if, as creationists maintain, dinosaurs are a uniquely created “kind” of organism that are distinctly different from birds and are in fact a type of reptile, then the protein sequence should most closely resemble a reptiles. As you probably guessed, however, the sequence of the T. rex is more similar to a bird’s than it is to a reptile’s (or to any other modern animal). Again, evolution’s predictions consistently come true.

I realize I spent a lot of time on the evolution of birds, but I want to stress that they are in no way a cherry picked example. I could show a similar sequence between fish and amphibians, amphibians and reptiles, reptiles and mammals, etc. (I’m generally loath to cite Wikipedia, but they actually have a pretty good list of intermediates). Further, genetic comparisons agree with these fossil trends. This is an extraordinary number of predictions that evolution got right and creationism got wrong.

How do creationists explain these fossils?
So, given this wealth of transitional fossils, how do creationists continue to maintain the belief that there are no missing links? They simply claim that all of these fossils are not evolutionary intermediates, but are in fact uniquely created organism that just happen to appear to be intermediates. So, for example, Archaeopteryx is not an intermediate, rather, God just felt like creating something that had half the features of a bird and half the features of a dinosaur. As I will explain, however, this view is logically incoherent and extremely problematic.

First, I want to stress the difference between what the scientists are doing and what the creationists are doing. As I previously explained, creationists like to argue the scientists and creationists have the same amount of evidence and they just interpret it differently. The fossil record offers an excellent way to illustrate why that is simply untrue and misrepresents how science actually works. You see, creationists are proposing that scientists are looking at the fossils, then interpreting them to fit their preconceived notion that evolution is true. When we look at the history of science, however, we find that evolution predicted the existence of these fossils beforehand. In other words, scientists aren’t looking at the fossils and concluding that they are intermediates because they want evidence of evolution. Rather, evolution predicted that we should find an organism with half the features of a bird and half the features of a dinosaur, an organism with half the features of a fish and half the features of an amphibian, etc. In contrast, creationism predicted that these organisms should not exist, but since they clearly do exist, they have simply modified their view so that they now claim that these are not actually intermediates but are uniquely created kinds. There is obviously a huge difference between these two approaches. Evolution correctly predicted what the fossils should look like beforehand, whereas creationism made an incorrect prediction, then made an ad hoc change to try to make the fossils fit the creationist view. Really think about this for a second. If these creatures were uniquely created, isn’t it astounding that evolution managed to successfully predict their existence? Further, evolution didn’t just predict the existence of one bizarre intermediate, it has successfully predicted the existence of hundreds of intermediates! This is why evolution is so powerful: it makes accurate predictions.

To further illustrate the problem here, I’m going to demonstrate that creationists are committing a circular logic fallacy (sometimes it is actually question begging or no true Scotsman depending on exactly how they word it). All of these fossils are exactly what evolution predicted beforehand. If you just imagine what an intermediate between these groups would look like, you’ll no doubt imagine something exactly like the fossils that we have, and the intermediates agree with the genetic data. So scientifically, there is no reason not to consider these to be intermediates because they match evolution’s a priori predictions. So the creationist claim that they are unique kinds is not based on scientific grounds, but on the grounds of personal belief (i.e., its pseudoscience). To illustrate, imagine the following dialogue between a scientist and a creationist (I have personally had this conversation with creationists):

• Creationist: Evolution isn’t true because there are no missing links.
• Scientist: What about Archaeopteryx, Basilosaurus, etc.
• Creationist: None of those are intermediates, they are actually specially created kinds.
• Scientist: How do you know that they aren’t intermediates?
• Creationist: Because evolution isn’t true.

You no doubt see the problem here. There only reason that you would think that these aren’t intermediates is if you have already convinced yourself that intermediates don’t exist. Lest you think that I am committing a straw man fallacy, ask yourself this: what would it take to convince a creationist that something was a transitional fossil? To put it another way, can you imagine an organism that more perfectly embodies the traits of two major groups than Archaeopteryx? Archaeopteryx is as perfect a match for the predictions of what an intermediate should look like as you could ever hope to find. The problem is that no matter what intermediate we find, creationists will always assert that it is just a specially created organism. This is why creationism is pseudoscience and why the claim that “creationists and scientists are both interpreting the evidence” is nonsense. Creationists have decided beforehand that evolution isn’t true, so any evidence to the contrary will be “interpreted” to show that evolution isn’t true no matter how absurd those interpretations are. This is why science cannot ever start with a conclusion. It must always start with the evidence, then draw a conclusion.

Again, to avoid any accusations of a straw man fallacy, consider the history of our knowledge of the evolution of turtles. For many years, we didn’t have any fossils of a proto-turtle (i.e., an early evolutionary step with a partially formed shell). The earliest fossil we had was Proganochelys, which was distinctly different from modern turtles, but still was obviously a turtle with a fully formed shell. Creationists were adamant that turtles disproved evolution because if evolution was true we should see proto-turtles (this is of course an argument from ignorance fallacy given that the fossil record is still far from complete). To quote one of their pages on this topic,

“Given the amazingly unique structure of turtles, it should be a rather easy task to find the transitional forms to trace the evolutionary path from ancestral reptile to turtle, if that is in fact what has happened. He [Dr. Gish] explains that the changes would not be subtle, but obvious, even to someone with no training in anatomy or paleontology…. The biblical account of Creation in Genesis 1—animals created to reproduce after their kinds—would mean that turtles should be instantly recognizable as turtles, with the shell and other unique features fully formed from the start.”

Once again, we have creationism and evolution making opposing predictions. That quote came from 1999, but if you fast-forward to 2008, you’ll find that paleontologists discovered several species of proto-turtle. Odontochelys is the most striking of these fossils (in my opinion). Its tail and parts of its body and skull are very lizard like, but on its stomach, it has the unmistakable plastron (bottom half of a shell) of a turtle. The ribs, however, have not fully fused into a carapace (top half of the shell). So although it clearly has a partial shell, it also clearly does not have the fully formed shell of a modern turtle. This is a near perfect match for what we would expect a proto-turtle to look like, and it certainly meets creationists’ prediction that a proto-turtle would be, “obvious, even to someone with no training in anatomy or paleontology.” Nevertheless, despite the fact that evolution’s predictions were clearly met and creationists’ predictions clearly failed, creationists refuse to acknowledge Odontochelys as an intermediate. Rather, they insist that it is a specially created organism which, for unknown reasons, God decided to make look just like a proto-turtle (this is an ad hoc fallacy at its finest). Rather than grappling with the obvious logical fallacy that they just committed, creationists instead spend their time trying to discredit the find by claiming that scientists predictions were actually wrong because of disagreements about where turtles originated (geographically and land vs. sea). What they are missing is that these disagreements are simply healthy academic discussions about the details of turtle evolution. They are not debates about whether or not turtles evolved or whether or not these fossils are one of the steps in turtle evolution. I won’t waste any more time on their “reasoning” about turtles, but the creationists’ arguments on this topic are laughably ridiculous (such as the blatantly and demonstrably false claim that, “we have never observed the development of new genetic information”).

Finally, these intermediate fossils completely shatter creationists’ notion of distinct “kinds.” Go back to the bird fossils for a minute and try to give a definitive answer for when dinosaurs stop and birds begin. You can’t, and neither can scientists! There is this massive grey area where we aren’t sure what to call a bird and what to call a dinosaur. As a result, many scientists now refer to “avian dinosaurs” and “non-avian dinosaurs” (avian meaning “bird”). This makes absolutely no sense if God created discrete kinds, but it is exactly what evolution predicted. Dinosaurs and birds aren’t distinct “kinds,” rather, they represent two ends of a continuum, and that continuum only makes sense if evolution is true.

To conclude this post, I want to directly address any creationists that are reading this. Seriously ask yourself what it would take to convince you that something was an intermediate fossil. In other words, what scientific criteria could you possibly propose for determining if something is an intermediate that would not result in the conclusion that fossils like Archaeopteryx are transitional fossils? We have hundreds of fossils which clearly posses half the features of two different groups, so if you cannot give a scientific reason why you think that none of these are truly intermediates, then you must admit that you are deciding that they are not intermediates based solely on your religious convictions, and, as I previously explained, that is not logically valid and it places you well within the realm of pseudoscience.

Note: creationists often harp on the fact that some of these intermediates (such as Archaeopteryx) are not actually in the direct lineage of our modern birds. Although this fact is true, it does nothing to discredit their stance as bird intermediates, nor does it minimize their evolutionary importance. The evolutionary tree is full of dead branches. In other words, many evolution paths result in a dead end (i.e., extinction). So, in the case of Archaeopteryx, we think that it was a sister taxa to the group that actually evolved into modern birds, rather than being in that group itself. In other words, it was closely related to the group that became modern birds, and, as such, it still gives us good insight to the evolutionary history of birds and represents a transitional fossil, but it is technically an intermediate to a group of birds that went extinct rather than the group of birds that survived to the present day.

Note on sources: I apologize for citing papers for which the full text is not freely available to the general public, but, as a scientist, the peer-reviewed literature is my source for information because it is the most accurate. If you want the free copy of any of the papers I cited, you can usually email the author and ask for one, and sometimes you can get a copy through your library (especially if you go through a university library).