Sye Ten Bruggencate on the Reasonableness of Belief in God

In a debate with Matt Dillahunty, Sye Ten Bruggencate made the following argument:

1. It’s reasonable to believe that which is true.

2. It’s true that God exists.

∴ It’s reasonable to believe that God exists.

The following video is based on the article that follows below it:

This looks like circular reasoning. After all, it’s going from God exists to It’s reasonable to believe that God exists. But these are actually two separate claims, and the problem with this argument is a different one. Putting aside the question of whether the second premise is true, this argument is unsound. Even if it is true that God exists, this is an unsound argument. To understand why, we need to first understand that the first premise has two possible interpretations. These are:

1. It is sometimes reasonable to believe that which is true.

2. It is always reasonable to believe everything that is true.

The first interpretation is true. For example, it is true that 1+1=2, and it is reasonable to believe this. But this interpretation does not make this argument valid. Just because some true things are reasonable to believe tells us nothing about whether other true things are reasonable to believe. Being invalid, this version of the argument is unsound.

The second interpretation does make the argument valid. If it is always reasonable to believe everything that is true, and if it is true that God exists, then it necessarily follows that it is reasonable to believe that God exists. But this second interpretation of the first premise is false. Having a false premise, the whole argument is unsound. So, whichever interpretation of premise 1 we use, this argument is unsound.

Here is one reason why the second interpretation is false. If it is always reasonable to believe everything that is true, then possessing reason should make us omniscient. Yet the consequent of this conditional is false, for we possess reason and are not omniscient. By modus tollens, it follows that it is not always reasonable to believe everything that is true.

Whether something is reasonable to believe depends on the evidence we have in support of it. When something is unsupported by evidence, it is not yet reasonable to believe it – even if it is true. Let’s consider some examples. One of the following two statements is true:

1. There is life on Mars.
2. There isn’t life on Mars.

Right now, there isn’t sufficient evidence to say which of these is more reasonable to believe. On the one hand, we have never found life on Mars, and Mars is inhospitable to the plant and animal life common to earth. On the other hand, large bodies of salty water have recently been discovered on Mars, and these may have provided a place for microbial life to evolve. Also, given that extremophile lifeforms inhabit areas of earth that are as inhospitable to our kind of life as Mars is, there could very well be extremophile lifeforms on Mars. We also know that meteors have come to earth from Mars, and perhaps parts of earth have gone to Mars as meteors. So, there is a chance that microbial terrestrial life has made it to Mars, where it has continued to evolve. At this point, though, we don’t know one way or the other. Each claim has some reason for believing it, but neither has enough evidence behind it to say for sure that it is true. So, at present, each of these claims is about as reasonable as the other to believe. In the future, when we have explored Mars more fully, we may have enough evidence to say that only one of these claims is reasonable to believe. But we’re not there yet.

Here is another example. The number π is an irrational number whose decimal representation never ends and never settles into a permanent repeating pattern.¹ For any specific digit after the decimal point, there is a fact about which digit this is. Let me pick a really large number: googolplex. This is 10^(10100).² As of 2014, π has been calculated for over 10¹³ digits.³ This is far, far fewer than googolplex, which is such a large number, I expect π will never be calculated that far before the sun extends to the orbit of the earth. Nevertheless, there is a fact of the matter concerning what the googolplexth digit of π is, and one of the following statements is true:

1. The googolplexth digit of π is 0.
2. The googolplexth digit of π is 1.
3. The googolplexth digit of π is 2.
4. The googolplexth digit of π is 3.
5. The googolplexth digit of π is 4.
6. The googolplexth digit of π is 5.
7. The googolplexth digit of π is 6.
8. The googolplexth digit of π is 7.
9. The googolplexth digit of π is 8.
10. The googolplexth digit of π is 9.

At this point, because the calculations for π have not been carried out this far, none of these claims are reasonable to believe. The true one could be made reasonable to believe only by being calculated.

The bare truth of something does not make it reasonable to believe. What makes something reasonable to believe is evidence, and we do not have adequate evidence for everything that happens to be true. There are so many truths we do not know, such as truths about history that are lost to us, truths about the conditions and inhabitants of other planets in the universe, truths about the future, etc. If it were reasonable to believe whatever is true, our reason alone would be sufficient for knowing these truths, but it is not. It takes not just reason but evidence to know a wide array of truths that are simply beyond our reach.

So far, we have seen two unsound versions of the argument Sye Ten Bruggencate made. One is invalid, and in the other, the first premise is false. The following is a variation on his argument that fixes these two problems:

1. It’s reasonable for me to believe that which I know.
2. I know God exists.
   ∴ It’s reasonable for me to believe that God exists.

This argument is valid, and the first premise is true, but I have modified both premises and the conclusion. I have replaced reasonable with reasonable for me, and I have replaced references to truth with references to knowledge. Truth is universal, but knowledge is particular to an individual. I can know something that you don’t know, or vice versa. Since knowledge is justified true belief, I am justified to believe that which I know. This is tautologically true. If I’m justified in believing something, then it is reasonable for me to believe it. So the first premise is true. So that the argument would be valid, I modified the second premise to be about knowledge of God’s existence, and I modified the conclusion to be about what it is reasonable for me to believe.

I’ll even assume, for the sake of argument, that Sye Ten Bruggencate knows that God exists. With this assumed, this argument is sound. Replacing the pronouns with a proper name, this is what the argument actually says:

1. It’s reasonable for Sye to believe that which Sye knows.
2. Sye knows God exists.
   ∴ It’s reasonable for Sye to believe that God exists.

But even if the second premise is true, making this a sound argument, it doesn’t demonstrate that it is reasonable for me, for Matt Dillahunty, or for anyone else to believe that God exists. For Sye to convince any of us that God exists, he needs to present evidence that would justify belief in God. But to do this, he would have to present evidence other than this argument that it is reasonable to believe that God exists. If he can do that, this argument, or any variation of it with another person, becomes superfluous. And if he cannot, it will not prove that it is reasonable for anyone still unconvinced of God’s existence to believe in God.

However, Sye does claim that everyone already knows that God exists. Given this, I’m sure he would accept this as a sound argument:

1. It’s reasonable for one to believe that which one knows.
2. Everyone knows that God exists.
   ∴ It’s reasonable for everyone to believe that God exists.

This is valid, and the first premise is true. Just to demonstrate its validity more clearly, I will write it out more formally:

1. For any given x and y, if x knows y, it is reasonable for x to believe y. (∀xy) (Kxy ⊃ Rxy)
2. For any given x, if x is a person, x knows that God exists. (∀x) (Px ⊃ Kxg)
   ∴ For any given x, if x is a person, it is reasonable for x to believe that God exists. (∀x) (Px ⊃ Rxg)

Proving the validity involves universally instantiating the two premises, then using hypothetical syllogism and universal generalization.

Only one problem remains with this argument. Despite Sye’s claim that premise 2 is true, it is not true. If the second premise were true, we would have universal consensus on its truth, and there would be nothing to debate. There is a debate at all because some people lack a belief in God. Yet anyone who lacks a belief in God fails to know that God exists. The existence of the debate itself, as well as my own self-reflection, refutes the second premise. But let’s presume, for the sake of argument, that everyone does believe in God, and those who say they don’t are lying or deluded.

Even given this, it has not been established that those who do believe in God are justified in this belief. Knowledge is justified true belief, which means that no one can have knowledge of God’s existence without their belief in God’s existence being justified. So, given the assumption that everyone does actually believe in God, it still cannot be established that everyone knows God exists unless the justification everyone has for believing in God can be produced. But to say that a belief is justified for a person is to say it is reasonable for the person to believe it. Thus, this argument begs the question. To establish the truth of 2, you have to establish the conclusion, that it is reasonable for those who do believe in God to hold this belief. So, after the argument has been fixed to make the first premise true and the argument valid, and after a generous assumption has been made toward acceptance of the second premise, the argument comes down to circular reasoning. Circular reasoning can be sound, but it still fails to establish its point. Evidence is still needed. So, Sye’s argument is a bust. It is unsound as he formulates it, and even fixed up, it begs the question, because it needs to establish the conclusion to establish the second premise.

1. Pi
2. Googolplex
3. y-cruncher – A Multi-Threaded Pi-Program