I'm aware that many things have been said and written on the burden of proof, and why it does not rest on the side of the atheist. That said, I'd like to put my two cents in.

I am often asked by religious folks to prove definitively that there is no god. Whenever I reply that I cannot absolutely and irrevocably prove that god does not exist I am often rebuked, scoffed at, and told that I must be wrong.

Incorrect.

Now I can go on about the burden of proof and how it's not my job to disprove god, but it always comes back to the rather superficial argument that I'm merely trying to back out of the argument and claim victory without doing any work. This is simply not so. I am following a standard of scientific process. If you make a claim, I'm going to ask you to prove it. If you are going to propose god, you've got to follow up accordingly. This is rarely the case, I am again told that I am the one who needs to provide evidence.

In this situation, an important piece of the scientific method is being ignored. It's called the null hypothesis. More or less, it's the default position of the scientific method. It's the argument against which we test our hypothesis. Most importantly, it must be noted that the null cannot be proven. It can only be accepted or rejected. With that stated, if your hypothesis is "God exists, all things in the universe are a result of him" then the null hypothesis by default is "God does not exist, all things in the universe are a result of chance."

In this argument, the atheists are on the side of the null hypothesis. Remembering that the null cannot be proven, only rejected or accepted, it is the task of believers to provide the proof that the null is false. Should significant evidence be submitted to prove the existence of God, we can reject the null. Otherwise, until such time that evidence is provided, we must accept it.

The problem now is evidence. What is evidence? The Oxford dictionary defines evidence as "the available body of facts or information indicating whether a belief or proposition is true or valid." But it is not the definition that is in question here, it is the identity. If you were to ask me for evidence of evolution, I would probably submit to you a fossil, an adapted body part on an animal, or perhaps a study of genetics. That would be evidence. If I were to ask for evidence of god, chances are I would be presented with some sacred text.

This is not evidence.

Texts like the Bible, the Torah, the Qu'ran are not evidence. They are claims made in writing. Furthermore, if I were to ask for evidence for the claims in these texts, there is a good chance I would be referred to specific entries from within them. This is a circular argument which holds no standing. You cannot use a text as evidence for claims within that same text. It is redundant. It proves nothing.

Likewise, I have seen this go even further than text. I've actually heard the argument that people have known the presence of god because they can "feel" it. I am asked to take this as solid, infallible evidence. But is this feeling testable? No, of course not. There is no meter with which the "feeling" of god's presence can be measured. Therefore it is just as much a claim as is the text, and it is inadmissible.

There is simply no tangible evidence which verifies the claims of sacred texts. We can provide real, empirical evidence of such things as evolution, universal expansion, and the Big Bang. We have observed things which, while not "proof" per se, are strong enough to support our claims. We have performed tests and repeated them with the same results. If the results change, we adapt our understanding. This is not the case with religion. Nothing truly changes in religion. It is by the book, even if the book is not a factual resource.

Consequently, it should not really come as a shock that thinking people are so quick to reject religious arguments. They are simply unevidenced claims, and claims without evidence can be rejected outright. Without question, I would reject any and all claims made to me without sufficient reason as to why they are true. To do otherwise would simply be bad science.