Today’s post will be an off-the-wall post about evolution, God, probability theory and parallel universes (among other things). So strap on your seat belt. It’s gonna be a wild ride.
Professor Douglas Futuyma (famed evolutionary biologist) once said: “Creation and evolution, between them, exhaust the possible explanations for the origin of living things. Organisms either appeared on the earth fully developed or they did not. If they did not, they must have developed from preexisting species by some process of modification. If they did appear in a fully developed state, they must indeed have been created by some omnipotent intelligence, for no natural process could possibly form inanimate molecules into an elephant or redwood tree in one step.”
I’m inclined to agree with him. However, it also been pointed out that it is logically possible for sheer, random chance (as opposed to evolution, which isn’t chance) to have brought about all life. No one believes that this actually happened, of course, it is simply that connection between sudden chance appearance and divine creation isn’t absolute. So, the question I seek to answer is this: how is it valid to infer creationism if we found out that all life came into existence suddenly? Likewise, how would it be valid to infer atheism if evolution is true?
Answering this question will require laying down a foundation and building on it, so here goes.
First, there’s the issue of probablistic reasoning. Probability is really nothing more than frequency. There have been a number of questions raised about whether probablistic reasoning can be used to evaluate theological claims, and it’s important to address those before we move on. Matt McCormick once asked whether it would be meaningful to speak of the probability of God’s existence, since that would translate into “the frequency of god’s existence.” This strikes me as a confusion: obviously what we’re talking about is the frequency with which a claim like that of God is accurate. If a claim doesn’t seem to do a good job of predicting the features of the universe we live in, if it violates Ockham’s razor, and so on, is that type of claim likely to be true? If not, then we can deduce that God’s existence isn’t likely since it belongs to a category of claims that are not likely to be true. It has also been said by many believers that God’s existence is necessary; that the probability he should exist is exactly 1. The problem is, they cannot demonstrate that the probability of a disembodied perfect mind is 1. The next best thing is to discuss whether it is likely that a spirit being existing necessarily is true. It might sound like probablistic thinking about necessary truths is contradictory; but a moment’s thought reveals that this is flawed. Example: you might believe in the truth of a mathematical theorem that you cannot understand because every mathematician believes it to be true. Though mathematical statements are either necessarily true or necessarily false, there’s nothing wrong with affirming such a statement as being probably true of necessity. Think about it: it is possible that the mathematicians are wrong, but it is not likely, and if it is not likely, then the mathematical statement is not likely to be a necessary falsehood.
You could even model the probability of the mathematical statement like this: suppose that instead of the statement being true in every possible universe or false in every possible universe, we fudge this condition and suppose that the statement is true in only half of the possible universes. As long as it is the case that mathematicians will usually recognize whether the statement is true in their universe or not, it will remain the case that whenever you find yourself in a universe that has a consensus of mathematicians believing the statement, the statement will probably be true. Hold up ten fingers to represent ten universes. That represents ten universes in which the statement is true. Put down two fingers to represent the universes in which the whole mathematical community made a mistake and thought the statement was false. That’s eight. Now hold up ten fingers to represent the universes in which the statement is false. Put down all except two to represent the times when the whole mathematical community made a mistake and thought the statement was true even though it wasn’t. That’s two. So, altogether there are ten universes out of those twenty in which the mathematicians affirmed the statement to be true. In eight of those, they were right, in two of those, they were wrong. It’s a fictitious but perfectly legitimate way to model the probability.
The only thing that I can think of that might be contentious here is that I’ve assumed that there is an equal ratio of universes for which the statement is true as false (this is known as the principle of indifference). Although some would quarrel with such an assumption, I see nothing wrong with assuming two hypotheses are equally likely (equally frequent) if you have no reason to think otherwise. If you answered thousands of multiple choice questions for which you no idea what the answer was, around 25% of your answers would be correct (if there were only four choices). Hence, the principle of indifference seems to be a reasonable way of assigning probabilities.
All of that brings me back to the questions I posed earlier in the post. Let’s run a thought experiment: Suppose that there were infinite universes. Half of them have a god, half of them do not. Since we’re dealing with infinities, anything that can happen will, and so it will be the case that on super-rare occasions some of the godless universes generate a whole host of complex plants and animals in a single step. It will be relatively more frequent for those godless universes to randomly generate some type of simple single-celled organism or self-replicating molecule, simply because doing that depends upon so much less luck than generating complex plants and animals. When a simple self-replicator like that develops, it will, at least occasionally, evolve into a host of more complex species. We don’t know precisely how frequent any of these things are, but any plausible guestimates that are constrained by real facts is bound to show that when complex life exists in a godless universe, it will nearly always be the result of some type of cumulative, natural-selection process. So, if you were in a godless universe, you’d expect to find that life had evolved. Let’s say that at least nine out of ten godless universes with complex life would have come about by evolution, and one out of ten would have complex life that jumped into existence by sheer chance.
On the other side of the coin, what about the god-having universes? There are two possibilities: either God will directly create or he will not. I have argued that there are many reasons that a God would prefer direct creation: it would provide more evidence of his existence so that people could know the truth (which a good God would surely want), it would avoid the millions of years of animal death and suffering that evolution entails, and it would also make more sense for a rational agent with all-power to go the most direct route to his goals. So, god-having universes ought to be represented as having evolution happen less than half of the time. In order to be generous to the theist, I’ll grant that it evolution would happen in 50% of god-having universes, just to give their hypothesis as much predictive power as it could reasonably have.
If you hold up ten fingers to represent the god-having universes, you can put five down and keep five up to represent the god having universes in which life has come about by evolution. If you hold up ten fingers to represent godless universes, and put down one (the universe in which life came about by sheer chance) then the nine you have left up represent the the godless universes in which evolution has happened. So, out of the fourteen universes in which evolution has happened (the five from the god-having universes and the nine from the godless universes), nine out of the fourteen are godless. Which means evolution means it’s a good deal more likely than not that god does not exist. You could come to a similar conclusion regarding the sudden appearance of all life: out of all the universes in which that happened, most would be god-having, not godless.
Is there any reasonable way to get out of this conclusion? I can’t see that there is. As long as complex life more typically comes from evolution in godless universes than in god-having ones, the probability of god’s existence will go down, at least a little bit, from what it would be otherwise. The probability might go down from, let’s say, 99 percent to 98 percent, but it still goes down. How could you get a conclusion like that? Well, if you don’t doubt the italicized text (and I don’t think it can be doubted) then here are the ways you could still doubt that evolution renders the probability of god below fifty-percent:
1. Argue that the volume of god-having universes ought to be modeled as being much bigger than godless universes. I modeled the volume as “for every one godless universe, there’s one god-having universe” but some might argue that we ought to say for every one godless universe there are ten god-having universes. I’m inclined to think it ought to be the other way around, just because the theistic hypothesis violates Ockham’s razor. Nonetheless, if you modeled the probability that way, it would come out that the likelihood of god’s existence would still be over 50%.
2. Argue that although evolution counts against god’s existence, there is other evidence that counts for god’s existence. This might be the case, but I don’t know of any such evidence, at least not any that is not incredibly weak.
3. Argue that most godless universes would not have life at all, whereas a higher number of god-having universes would. I think this is a better objection, but it still fails. The only two conditions that need to be met in order to have life in a godless universe is: 1) the laws of physics allow life and 2) the origin of simple life is frequent enough to happen at least once per universe. Although the first condition may only hold for a small volume of godless universes, I’d argue that the volume in question is at least as high as the volume of universes that would have a god, a la Richard Dawkins’ argument in The God Delusion. As for the second condition, if we limit our comparisons to only those universes that resemble our own in size and age and have the evidence for abiogenesis that ours does, I’m pretty sure that condition 2 wouldn’t be one that favored theism.
I can add that there is at least one factor that would strengthen my argument: it might be argued that the probability of god creating life or allowing it to evolve is zero, because god would not do either of those things. The other possibilities are that god would not create anything or that he would create only other spiritual things. There is a pretty interesting argument for God not creating at all: Any action requires a desire. Any desire indicates a need or an incompleteness in the being that has it. Any being with needs or who is incomplete in any way is not perfect. Therefore, the action of creating anything could only ever be done by a being that was not perfect. God is supposed to be perfect. Therefore, the universe could not be created by God.
4. Doubt the validity of evolution. Although this is a possible position, it is my considered opinion that it has nothing going for it. The evidence for evolution is vast in quantity and quality, see Why Evolution is True or Evolution: What the Fossils Say and Why it Matters.
5. Natural Selection, Creation, and Chance might not be the only options. True, but there is no observed mechanism that has any bearing on the origin of well-adapted life besides those three. In fact, every hypothesis that has been proposed since ancient times reduces to one of those three. So it’s overwhelmingly likely that these are exhaustive, in my opinion at least.
I think that’s about it. Though some people that I have talked to find all of this unbelievable (that evolution counts as good evidence for atheism) I find it as plain as day. I should also mention that the idea is not my own, philosopher Paul Draper was the first to articulate the basic sentiment. Look forward to hearing your responses on this one.
And no, I am not on crack.
