Atheist VS Naturalist - the latter sounds more appealing to me...

[Belacqua]

My personal view is that the term 'supernatural' is simply incoherent.

I gave a definition which is coherent.

No, but anything that cannot be observed, even in theory, might as well be non-existent.

By "observed," of course, you mean observed in the scientific sense. Repeatable, empirical. 

In the example I gave, we can observe the frog singing Mozart. We cannot discover the reasons for that through science -- in fact everything science tells us about frogs tells us it's impossible. 

I'm not saying that frogs can really sing, or that anything else supernatural happens. I am only pointing out that you affirm the consequent by determining the only acceptable methods.

The problem, ultimately, is in the definition of the concept of 'natural' and 'supernatural'. I think it much better to focus of 'testable' and 'untestable' in regard to hypotheses.

Right, because that starts out in a way that gives you the conclusion you like. Because "testable" means testable according to the method you prefer. 

This rules out anything which isn't testable according to science. It's begging the question. 

WAY too many philosophers are still stuck with an Aristotelian mindset.

Please explain why. Give examples. Why is talking about frogs having a nature overly Aristotelian? Would it be better for you if I just said that frogs are a certain way, and not another?

I'm looking for more than unsupported opinions.

This I agree with: that faith is not a part of science. Science requires testable hypotheses (which means there must be some way, in theory, that they could be shown to be wrong).

That's right. Testable according to the scientific methods. 

Metaphysical issues and supernatural ones, if they existed, would not be testable in this way. It is begging the question to assert that only such testable things exist.

nd, again, the problem is in the concept of 'natural'. What is *actually* required isn't 'natural' it is 'testable'.

This is the definition you prefer. By limiting the world to things that are testable in this way, you rule out anything else a priori. 

This is the metal detector issue: when your only tool is a metal detector, you only find metal. This doesn't allow you to conclude that only metal exists.

If a frog was able to sing Mozart, that would become the raw data about certain frogs from which we learn what frogs are capable of doing. Now, a deeper explanation: how do they do it without vocal chords, etc, is clearly also needed.

Frogs are not able to sing Mozart, especially simultaneously singing both parts of a duet. 

What I am saying is that science would be able to study it, to analyze it, and to find patterns in how it works. That is what science *is*, ultimately. The term 'natural' is, ultimately, just those things that show testable patterns.

This is a faith-based statement of metaphysics. Science can't prove that only things testable by science exist. 

I'm not saying that there are supernatural events. I'm only saying that your view amounts to a metaphysical commitment and faith which can't be proven. It may be correct, but we don't know.

What you're saying is that if something completely inexplicable to modern science happened, you have complete faith that it could be explained by science. Even before you know the explanation, you know it will be science. To me, this is the same as a Christian who says that in cases where we don't know the answer, the answer will surely be God. In both cases the believer is certain of things not proven.