Header 1

Our future, our universe, and other weighty topics


Sunday, July 26, 2015

We Do Not Know That Evolution Occurs Mainly Because of Natural Selection

The history of science contains many cases of theories that were oversold by eager apostles. One of the main cases today involves the overselling of the theory of natural selection. In this case many of our biologists are representing as “settled science” the idea that natural selection is the main cause for evolution. The evidence for evolution is good, but the evidence is quite weak that natural selection is the main cause of evolution in large organisms.

Let me clarify the difference between evolution and natural selection. Evolution is the idea that over very long periods of time, species very gradually undergo changes, developing novel adaptions to their environment, with such adaptions sometimes becoming sufficient for one species to evolve into a different species. Natural selection is a hypothesis about the cause of evolution. Natural selection says that evolution occurs mainly because organisms that are more fit to survive tend to reproduce more. Darwin introduced both the basic idea of evolution and the theory of natural selection at the same time, in a book entitled On the Origin of Species by Means of Natural Selection. But it is quite possible to believe in evolution without believing that natural selection is its main cause.

The modern theory of natural selection is all centered around the idea of random mutations. The idea is that the blueprint of a species (found in its DNA) undergoes random changes called mutations, which might be produced by something like a cosmic ray hitting the DNA stored in a cell, or a random “copying error” when the DNA is not copied exactly right. In some cases, it is claimed, these random changes result in a change that is actually beneficial. When such a thing happens, it is claimed, natural selection causes the organisms that had such mutations to reproduce more often than some other organisms in that species which did not have such a mutation. It is claimed that this causes the mutation to be inherited more and more often in later organisms belonging to that species. Such a process, it is claimed, it so powerful that it can account for all kinds of astonishing adaptions in organisms. 
 
Now the question is: what evidence is there for this theory that natural selection is the main force behind evolution? Although there is lots of historical evidence for evolution (found in the fossil record), there is no significant historical or fossil evidence that natural selection has been the main cause of evolution in large organisms. Think for a moment of the difficulties of ever getting such evidence. In almost all cases when we have a fossil record of some organism that existed eons ago, we don't have DNA from that organism. Even if we had that DNA (from a few dead organisms), it would still not be enough to figure out the history of DNA changes in a species from long ago, and whether or not  natural selection was the cause of evolution in a particular species.

It turns out that getting evidence for natural selection requires an incredibly hard procedure. This procedure has been attempted on some small organisms with very short lifespans (such as bacteria and fruit flies). The procedure might go something like this:
  1. You start out with a population of organisms of some type, and you carefully record the DNA of all of those organisms.
  2. You track the evolution of these organism over many generations, while applying some factor that causes lots of mutations. The DNA of these organisms must be checked at very many times.
  3. You attempt to determine whether useful new adaptions are being evolved, and attempt to determine whether such adaptions (this new functionality of evolutionary novelty) are being produced by natural selection and random mutations.
The experiments that have been done along these lines have produced very modest results. The study here studied 600 generations of fruit flies, finding no change more dramatic than a 20% shorter life cycle. 

Another experiment is Richard Lenski's long-term experiment on bacteria, which has tracked more than 60,000 generations of bacteria. The results of this experiment have been modest, with the main result merely being that the bacteria seem to have developed an ability to digest citrate that it did not have before. Documenting such an adaption is “peanuts” compared to demonstrating that a significant and complicated structural innovation (such as the human eyeball or a wing) did occur because of natural selection (something that no one has ever demonstrated).

Even Lenski's meager result is subject to doubt, as a large fraction of scientific experiments are not replicated when another experimenter tries to replicate them (and in the case of Lenski's experiments, it will be decades before we can determine whether someone is able to replicate his results).

There is a simple reason why experiments of such a type can never show that natural selection is the main cause of evolution in large organisms. The shorter the life cycle of an organism, the more likely it might be for it to evolve beneficial changes because of natural selection. Bacteria such as those in Lenski's experiment have a lifespan about 130,000 times shorter than a human lifespan, and can double their population in only about an hour. It is all too possible that natural selection is sufficient to cause useful adaptions in very short-lived organisms such as bacteria, but is not sufficient to cause useful adaptions in large, long-lived organisms such as animals as large as a dog or larger.

But isn't there some way, in theory, that you could prove natural selection by using large long-lived animals such as mammals? Yes, in theory there is. But it has never been done, and would be a nightmare to do, as it would take ages.

Imagine how the project might be executed. After building some special testing environment (perhaps some special large building), you would start out with a population of some large species with a lifespan of more than decade. You would take samples of the DNA of each organism in such a population. You would then monitor such a population over many generations, frequently taking DNA samples to see how the DNA was changing. Since a generation for such organisms would take at least a year, the project would have to probably last for thousands of years. All in all, it would be a project more difficult than landing men on Mars. No one has ever done such a project, or even one tenth of such a project.

It would seem that for reasons such as these, the theory that natural selection is the main explanation for the evolution of large animals is one that simply is not very susceptible to experimental verification. But what about some other approach? What about some approach in which we get the predictions of the natural selection theory, and then try to verify that such predictions are coming true?

But that doesn't work either. The reason is that the modern theory of natural selection is all centered upon the idea of blind chance. The theory assures us that natural selection will do random stuff we can't predict. So there is not much of a way to match up reality with the predictions of natural selection. For example, natural selection does not give us any predictions about what a particular organism will evolve to in the future. So there is no way to exactly match up predictions and reality when trying to get proof for natural selection.

So from the standpoint of being verified, the theory that natural selection is the main driver of evolution in large animals is on incomparably weaker ground than other scientific theories which do make a host of exact numerical predictions that are repeatedly verified to the letter. Using a theory such as the theory of gravitation, one can make very precise predictions such as the prediction that a particular object released from a particular height will hit the ground 14.5 seconds after it has been dropped. Such predictions have been verified countless times. But the theory of natural selection has no such record of predictive success. Some of the things that are sometimes claimed as “successful predictions” of the theory (such as the discovery of something like DNA) are not actually predictions uniquely predicted by the theory. A nonbeliever in natural selection would have been just as likely to have predicted that something like DNA existed before it was discovered.

Eager to try to prove that natural selection is an important determinant of human traits (or the main determinant), some scientists have resorted to statistical analysis of DNA (the genome). But such studies may not be mathematically sound. In 2009 Phys.org published this article stating the following:

Scientists at Penn State and the National Institute of Genetics in Japan have demonstrated that several statistical methods commonly used by biologists to detect natural selection at the molecular level tend to produce incorrect results. "Our finding means that hundreds of published studies on natural selection may have drawn incorrect conclusions," said Masatoshi Nei, Penn State Evan Pugh Professor of Biology and the team's leader.

Likewise the paper here states, “Many of the statistical methods for detecting natural selection are unreliable.”

The study Genome-Wide Scans for Footprints of Natural Selection notes, “a puzzlement arises when we inspect how modest is the replication for discovery of different genomic regions between algorithmic approaches or between different studies.”

Given these items, one must wonder whether scientists scanning the genome, eagerly looking for faint traces of natural selection, are much different from UFO enthusiasts scanning photos of Mars and occasionally claiming to have found something important. Given a mountain of data, sufficient time and a huge toolkit of statistical methods to choose from, it is not too unlikely that you may be able to find “faint traces” of exactly whatever it is that you were hoping to find. I may also note that finding some statistical trace of natural selection would not by itself prove that natural selection is the main thing driving evolution in larger organisms such as man. It is entirely possible that natural selection is a relatively minor effect in species with long lifetimes, and the main thing propelling evolution is something else. 

There are also very substantial reasons for doubting that natural selection is the main thing that caused evolution in humans. The best reason I know of is the inability of natural selection to explain dozens of human mental faculties and traits that do not seem to be adaptions that contribute to reproductive success. Consider this question: what is it intellectually that makes a human different from a monkey? There are many things: we have good language ability; we're good at math; we have morality; we are spiritual; we have esthetic abilities that allow us to create and appreciate art; we have inner lives and introspection; we can form abstract ideas and ponder philosophical questions; and so forth. But none of these things are biological adaptions that improve an organism's likelihood of surviving until reproduction. So none of these things can be explained by natural selection. It would seem, in fact, that natural selection theory predicts that such things as these should not even exist. 

 Human traits hard to explain by assuming natural selection

So far from just being a case of natural selection not making predictions that we can verify, the problem seems to be that natural selection theory would seem to make counterfactual predictions about human nature – that you should not have any important characteristics unless they make you more likely to survive until reproduction (in other words, that you should not have most of the things that make you different from a monkey).

The difficulty of using natural selection to explain the origin of man's mental capabilities is compounded by the fact that about 200,000 years ago the population of humanity was believed to be small, as few as 10,000. The fewer organisms there are in a population, the more unlikely that there will be enough mutations for natural selection to produce something useful.

I may point out here that you could not counter these arguments merely by pointing out some evidence that natural selection occurs. Let's consider three different ideas that can be listed in a table.

Hypothesis 1 Natural selection occurs.
Hypothesis 2 Natural selection occurs, and is the main cause of new adaptions or features in small organisms with short lifespans (bacteria, fruit flies, etc.)
Hypothesis 3 Natural selection occurs, and is the main cause of new adaptions or features in large organisms such as humans.

You do not prove Hypothesis 3 by merely proving Hypothesis 1 or Hypothesis 2. Since the reproduction rate of tiny organisms may be thousands or tens of thousands of times greater than the reproduction rate of large organisms, you don't prove Hypothesis 3 by proving Hypothesis 2. So you can't successfully rebut this post by merely citing something that supports Hypothesis 1 or Hypothesis 2.

Given all these problems with the idea of assuming that natural selection is the main cause of evolution in large organisms, why is such an idea being sold as “settled science”? Why are we being dogmatically assured by so many biologists that they understand what caused the appearance of man's higher traits? I think it's a case of premature triumphalism. After first figuring out that evolution is occurring, our biologists should have merely said to themselves: we have completed one lap, but there are still 10,000 laps to go before we understand this. But instead scientists crowned their heads with laurel leaves (and put gold medals around their necks) after completing the first lap – by assuming that the first major book on evolution had also figured out what the cause of it is. It is all too plausible that the actual cause of evolution is some principle or principles far more complex and vastly more deep than the simple idea of natural selection. If we ever understand such a principle or principles, I suspect we will find that it is also the explanation of the origin of life, something entirely unexplained by the theory of natural selection.

It is entirely possible that the main thing driving evolution is some natural mechanism far deeper and more sophisticated than natural selection – some mechanism involving undiscovered laws of nature or undiscovered information processes (or both) that tend to work in a teleological or anti-chaotic manner, causing more and more complex things to emerge from simpler things. Nature could have embedded within it some kind of programming (or something that acts like programming), some wellspring of emergence that acts as an antithesis to the Second Law of Thermodynamics, causing ever-increasing order as time passes. Such a principle might be behind the evolution of man's faculties, the origin of life, and the physical ordering of the universe since the time of the Big Bang. 


Postscript: See the link here for a 2-sentence statement signed by hundreds of scientists and PhD's.  The statement states exactly the following:

We are skeptical of claims for the ability of random mutation and natural
selection to account for the complexity of life. Careful examination of the
evidence for Darwinian theory should be encouraged