• Proving Darwin: Prior Probability

    I understand that much of what I write here will be controversial. I know that creationists will reject out of hand the geologic column and that they will see an apparent inconsistency with the theory of evolution and the theory that life only comes from life once life exists. I will address those concerns later on in this post. All I ask is that people read this in its entirety and with an open mind.

    A necessary part of judging Darwin’s theory with Bayes’ Theorem is finding the prior probability, that is, the probability of evolution occurring given our background knowledge. To find it, I will call on two important pieces of background knowledge.

    The first fact is that it is nearly certain that spontaneous generation has never happened. Spontaneous generation is the idea that complex life can originate from nonlife even when life is present. The idea was widely believed in ancient times and even through the middle ages: people thought that maggots originated directly from rotting meat, that mice could be generated from old hay, and so on. Saint Augustine even interpreted spontaneous generation as present day creations of God Almighty (The Emergence of Life on Earth, p.18). I will define spontaneous generation as new organisms coming into existence without the sexual or asexual reproduction of a previous organism (such spontaneous generation might happen by the will of a supernatural power, extraterrestrial creation, or unknown natural processes) in a time period in which life already exists (meaning that we won’t be considering the hypothesis that aliens “seeded” the first life on earth or God creating it, we will only be considering the hypothesis that such events happened after the first living species existed).

    Today it is known, of course, that maggots originate from eggs that flies have laid, that baby mice come from the sexual reproduction of previous mice, and so on. That living things tend to come from other living things has been a universally accepted observation for over a hundred years.

    The second fact is that life was very different in the past than it is today. The US Geologic Survey provides the following graph to show the history of life on earth:

    Diagram showing stratigraphic ranges<br /><br /><br /><br /><br /><br /><br />
and origins of some major animal and plant groups

    Given the fact that new life forms have appeared on earth throughout its history, and given that life can only come from life after life has begun, it follows logically that newer species must be descended from older ones. Now this line of reasoning is so strong that few would resist it, it is a logically necessary conclusion from two strongly established premises. What chance is there that it is wrong? Is there a chance that spontaneous generation (via natural mechanisms, supernatural creation, or alien engineering) has occurred in the past? I suppose there is a chance, albeit an absurdly small one. I don’t know exactly what the chances are, but it is safe to say, I think, the conclusion “newer species must be descended from older ones” must have at least a fifty percent chance of being true, and arguably a much better chance than that.*

    Bayes’ theorem demands that our prior probability considers all possible theories. I’ve discussed common ancestry and descent with modification in a previous post, and so the possibilities we have before as are:

    1. Common Ancestry is true but Descent With Modification is Not.

    2. Common Ancestry is not true but Descent with Modification is true.

    3. Neither Common Ancestry nor Descent with Modification are true.

    4. Both Common Ancestry and Descent with Modification are true.

    Possibility 1 says that Common ancestry is true but descent with modification is not. That means that every living thing on earth has a common ancestor, but there has not been any change at all from the original species. Obviously this false, since there are lots of species with all kinds of differences between them, so we can rule it out automatically. Possibility 2 says that living things do not share common ancestors but that there has been descent with modification. Look at the geologic column. There is no way that present day life could have developed out of the species of microbes that existed 3 billion years ago without some of them having common ancestors. So this possibility can be ruled out. That leaves us with just two possibilities that fit the evidence so far: either new species have developed by descent with modification from previous species or new species came about some other way. I am therefore assigning a fifty-percent probability to each. This is an absurdly generous estimate favoring the creationist side of the argument, as our background knowledge clearly offers massive support to the statement “life only comes from life after life exists.” It’s the only thing we have ever observed, and so it is better supported. Nonetheless, I will treat the two possibilities as equal. Now that we have determined our prior probability for Evolution as 0.5, we will reach a verdict in future posts through examining the evidence and coming up with an evidential probability.

    Does the Geologic Column accurately represent the History of Life?

    Many creationists do not believe that the geologic column exists or that it can be taken as an accurate chronological record of the history of life. I will not here give an extensive refutation of that claim, but the geologic column is real, there are locations that have rocks from nearly every geological period, it is a perfectly valid construct. Elementary knowledge of how geology works tells us that much. One might claim that even though it exists, it does not show a true chronology of life on earth as fish appearing before amphibians, and so on. Creationists often argue that the geologic column was created by the flood, and that there was some type of filtering mechanism that caused fish to appear before reptiles, for example. This is complete nonsense, falsified extensively and repeatedly as Paleontologist Stephen Jay Gould showed.

    An old-earth creationist once floated the idea that maybe fish and mammals did not get preserved for so many hundreds of millions of years, but this strikes me as incredibly implausible. How could thousands and thousands of species exist for hundreds of millions of years without leaving a trace, either of fossil coprolite or fossilized bone?

    So the geologic column does represent a true history of life. That is an established fact in our background knowledge.

    Life Can’t Come From Life?

    A favorite creationist argument is that the first life could not have come from the chemicals that existed on the early earth because we’ve never seen life come from nonliving chemicals. Arguing from a lack of observation is valid in many cases, but it is not in this one.

    If I were to say that it is impossible for someone to make a sphere of gold a mile wide because I had never seen such a thing, I would be laughed at. Why? Because we know that gold can be melted and molded into all kinds of shapes of a broad variety of sizes, and have no real reason to suspect that it could not possibly be molded into a mile-wide sphere. Even though we’ve never seen a mile-wide sphere of gold, we have knowledge that entails that one could be made, and the fact that we haven’t seen it does not make it impossible.

    Making a valid argument from absence to impossibility requires, at the very least, that one does not have any knowledge that such things are possible. A “living thing” is a thing that can reproduce, has metabolism, etc. Its reproductive and metabolic abilities exist because of the the arrangement of the chemicals it is made out of (life is made only out of chemicals, as biologists have been confirming for the past century). Of course it is possible, it could happen, that chemicals not arranged in such a way as to carry out these functions could be thrown into an arrangement that does have such functions. Such an event must have some frequency of occurrence, even if that frequency is extremely low. I won’t argue about how low or how high it is, only that it does have a frequency.

    If it is the case that life exists, however, the spontaneous origin of new life is effectively impossible. Darwin provides the reasoning here:

    “It is often said that all the conditions for the first production of a living organism are now present, which could ever have been present. But if (and oh! what a big if!) we could conceive in some warm little pond, with all sorts of ammonia and phosphoric salts, lights, heat, electricity, etc. present, that a protein compound was chemically formed ready to undergo still more complex changes, at the present day such matter would be instantly devoured or absorbed, which would not have been the case before living creatures were formed.”

    In other words, before life existed it was possible for the building blocks of life to join together and form the first living thing. After life existed, the building blocks of life (chains of amino acids, for example) would be instantly consumed before they could join up and form a primitive living thing. Even if some primitive living thing were to form nowadays, it would almost certainly be outgunned by the well-adapted and highly evolved life that exists in the present day. Last but not least, it is also known that conditions on earth are different than they once were in such a way as to render the origin of new life impossible; for instance, the atmosphere contains large amounts of oxygen which degrade and break-up organic compounds. In summary: the origin of life from inanimate matter is possible if life does not already exist, but after life exists, life can only come from life.

    Of course, creationists and IDers probably already agree that naturalistic spontaneous generation is impossible. Their theory of spontaneous generation, though, requires postulating that some sort of designing entity or entities exist. We do not know that such an entity exists a priori, and without that knowledge, we cannot even say for sure that spontaneous generation via alien or supernatural creation is possible. This means that the creationist theory is far weaker a priori than evolution, because we have knowledge entailing that evolution at least can plausibly happen (we’ve observed species changing in the present day in all sorts of interesting ways, such as the Lenski experiment).

    Ad-Hocness and Mutual Exclusivity

    It might be objected that the two possibilities we are testing (Either “new species in the past were produced by the reproduction of previous species” [evolution] or they were produced by some other means [spontaneous generation]) are not really mutually exclusive. After all, you could have a mixture; some intelligent design proponents might happily admit that some new species evolved but some were created. This is an ad-hoc theory. IDers have a theory that has almost infinite wiggle room. If you can imagine a dial that specified how many species were evolved and how many were spontaneously generated, such a dial could be set on any one of millions and millions of different levels for the ID theorist, whereas under the theories of complete spontaneous generation or complete evolution the dial is locked into place at 0 and 100% respectively. This is problematic for IDers for the following reason: their theory is so adjustable that its predictive power is very weak (I recommend this paper on Ockham’s razor for an explanation of why). The theory can be adjusted to accommodate man’s shared ancestry with chimps and gorillas or with a theory of separate ancestry, and the same goes for every other proposed relationship between plants and animals. Perhaps more importantly, every example of shared ancestry that creationists admit to strengthens the case for evolution and weakens the case for ID and every other form of spontaneous generation, because every single example of evolution that we have strengthens the generalization that “new species come from previous ones.” In a greyhound race, every single past victory that a greyhound has achieved adds something to the probability that it will win the present race. The same principle applies here for the theory of evolution. This concludes the a priori objections that I  have for this thesis. In future posts I will put the evidential nail in the coffin of all doctrines of spontaneous generation.

    Category: Bayes Theorem


    Article by: Nicholas Covington

    I am an armchair philosopher with interests in Ethics, Epistemology (that's philosophy of knowledge), Philosophy of Religion, Politics and what I call "Optimal Lifestyle Habits."