This is a response I wrote to a question in an online forum on atheism. The question was, which of the two do you prefer: to believe there is no God, or, to have no belief in God. Here is my response:
I prefer the first choice, which is “To believe there is no God”.
The second choice “To have no belief in God” would mean that I don’t believe in God, without specifically commenting on whether God exists. Some could argue that the implication is that God does exist, but you just lack belief for some reason. Others might argue that the implication is that God might exist. It becomes a word game, with different people making different guesses. In my opinion, this is poor communication. If you are agnostic, why not come right out and say so in words like these “I have no proof that God exists, therefore I do not believe in God”. If you want to be even more particular you can add “I am willing to change my mind if someone offers proof of God’s existence that I consider reasonable”.
For my part, I am an atheist, not an agnostic.
The reason for this is that I consider the concept of God to be neither true nor false, but arbitrary. To think of something as true, I would need some evidence that it exists. Similarly, to think of something as false, I would need some evidence of its falsity (remember, scientific hypotheses are falsifiable, if this is a scientific hypothesis, and if only scientific evidence can change my mind). No such evidence of either truth or falsehood exists, in reference to God. This makes the concept of God completely arbitrary, and therefore it has no cognitive value to me. It tells me nothing at all about the world, good or bad, right or wrong.
This is important to me for practical reasons. To illustrate with an example, suppose I am an electronics engineer, trying to debug some hardware I just built. I keep getting weird, flaky faults (this is common in hardware debugging). Although I have no idea what is causing these faults, I could make the assumption that something in the hardware is the cause. Perhaps a short somewhere; perhaps a bad circuit; perhaps overheating. I could carry out an investigation based on this assumption, and sooner or later discover the cause. This is a productive assumption, because it leads to a course of action that is fruitful. It is also a logical assumption, based on my knowledge of electronics. I know for a fact that hardware glitches can cause such faults.
On the other hand, I could make a completely arbitrary assumption, such as “there are gremlins in this machine that are thwarting me by switching around circuits”. An arbitrary assumption is one that (1) has no evidence to back it up, (2) that I can’t falsify, and (3) that produces no course of action. I could continue to tinker with the machine, and perhaps one day it might suddenly work perfectly. I could never be sure if it was because of something I did, or because the gremlins just got bored and went away to plague someone else.
It is the same for God. Nothing that anyone can do or say, no experiment, no logic whatsoever can disprove the existence of God. God, by definition, is beyond our capacity to understand. As such, the concept of God has no cognitive value. It gives us no course of action, no direction. The atheist and the agnostic both have exactly the same approach to the natural world — they both dismiss God and focus on natural explanations. The difference is that to the atheist the concept of God is arbitrary and meaningless, while to the agnostic it is some vague, amorphous and unknown “maybe”. In either case, it cannot be acted upon.
Only to the religious does God actually make a difference. This is because they have more than just a God. They have directions, in the form of written texts and revelations, of what God wants them to do. To them, God is an actionable concept. If you do not believe in these texts, why be agnostic?
Some may say that both atheism and agnosticism are basically the same, since neither produces any actionable intelligence. However, I would argue that atheism is more parsimonious, since it explicitly acknowledges God as arbitrary, and therefore assigns a cognitive status (zero) to it. This allows you to be clearer in your mind, and brings consistency to your thoughts, because there is no difference in how you treat this arbitrary concept and any other arbitrary concept. Even the most committed agnostics among us do not assume gremlins when they come across something unexplained. So why make an exception for God?
Another argument I hear often is that being agnostic is somehow “scientific”. People who follow this logic offer the argument “I only accept things that I have evidence for, or against. Since I cannot rule out that there is a God, I must choose to leave this question unanswered.” But what does this actually mean? What is the status of an unanswered question? Is it the same as that of any other unanswered question? I submit that it is not. Again, let me illustrate with an example.
Consider a biologist, grappling with some problem in his research. He may easily come to the conclusion that he just doesn’t have enough facts to understand the problem. He has an unanswered question. In honesty, he must classify it as such in his mind and admit that he does not have sufficient information to even make reasonable guesses. But there is a subtext to this. The subtext, that which is implicit in his understanding even when he has no answers is: there is an answer to this problem, there are facts (as yet undiscovered) which could point him towards an understanding, and that these facts are amenable to the same natural interpretations as any other facts, and will lead to some natural conclusion.
Contrast this to the unanswered question about God. When a person says that he must leave the idea of whether there is a God unanswered, he is not only saying that it is unanswered but also that it is unanswerable. There are no facts we could ever discover that would be evidence of a God. Even a messiah performing miracles or some creature from an alien civilization would be just that: some entity whose knowledge far exceeds our own. We only need to think of what it was like 2000 years ago when people regarded a broad range of phenomena as acts of God, or magic, for which we have natural explanations today. The same is true at any level of our civilization – that which appears to be magic today can be something we can understand and duplicate tomorrow. Therefore, to elevate the status of an entity from “something far in advance of our science” to “a God” requires an endless regression (to match his science in order to duplicate his miracles) to the logical endpoint – that of being Gods ourselves. Unless we are Gods, in possession of complete and total knowledge, whatever that might be – we cannot rule out that the entity isn’t just a more scientifically advanced version of ourselves.
I point this out to distinguish between the two kinds of unanswered questions: the kind we pose all the time in our work and daily lives, which actually have some cognitive value, and the question of God’s existence, which is not only unanswered, but also unanswerable. If we conflate these two types of questions, we are not being scientific; we are actually mixing two quite different things. In doing so, we hurt ourselves, because we have erased a real boundary in our minds, the boundary that separates the real or unreal from the arbitrary. You can argue “so what, neither atheists nor agnostics actually use the concept of God in their daily lives, so how is it hurting?” That may be true. I am no psychologist so I will not speculate on how our premises affect our actions. But I do call it sloppy thinking.