4 ms·
> You are by saying there's no difference in the definitons. Chambers says atheists have a belief. Websters says it's the lack of belief. You're ignoring half
by CharlesColeman 8y ago
> You are by saying there's no difference in the definitons. Chambers says atheists have a belief. Websters says it's the lack of belief.
You're ignoring half of each of the sentences in question. Simplified: one's "belief of non-existence", the other's "non-belief of existence." They both express the same state of belief in different ways.
I understand you want the definitions to make different statements. They just don't, I'm sorry.
- Latty 8y agoWhy did you ignore me when I stated earlier that those two things are not equivalent? > They are arguing that this isn't boolean logic. > "god_exists == false" is not the same as "god_exists != true", because "not true" can mean "false", but also "almost certainly false" and many other things.
- CharlesColeman 8y ago“When I use a word,” Humpty Dumpty said, in rather a scornful tone, “it means just what I choose it to mean—neither more nor less.” If you want to express a concept of probability like you tried to, it's exceedingly unclear. If you want to be understood, you should make it clearer what you're trying to say. Going back to the definitions originally quoted, the only way you find probabilities like that in them is by reading them in yourself.
- Latty 8y agoWe don't walk around expecting gravity to suddenly change, or for elephants to suddenly appear around us. We are not talking about real probability - we are talking about an unfalsifiable thing having no evidence, which is functionally the same thing as proving it false, we can't prove it false, but we act like it is, because doing otherwise would mean acting like anything insane and unfalsifiable you can think of is true. The reason this distinction isn't clear is because no one needs to make it in the real world. Atheists don't believe in god. This doesn't mean that - if provided proof, they wouldn't change that belief. That's how normal people use "belief" in the real world, the only way that the weird definition twisting of "Atheist" makes sense here is trying to claim "belief" has to mean unwavering absolute truth. You can either be talking in absolute logic, and allow for "not false" to mean more than "true" because this isn't a boolean, or you can be talking in natural language where the central debate doesn't hold because no one uses belief like that. You can't pick and choose common use and holding to strict definitions.
- CharlesColeman 8y agoEven taking that perspective, the definitions are still semantically equivalent. > You can't pick and choose common use and holding to strict definitions. I never did that.
- amazingman 8y agoYou didn’t address the Boolean example, either. Do you actually think the following evaluates to `true`? `(god_exists == false) == (god_exists != true)`
- CharlesColeman 8y agoYes, especially when you write it like that. I get that some people really want to make a distinction, and such a distinction can exist, I just don't think it's reflected in the texts that they claimed it is, especially given their context.
- amazingman 8y agoAnd what if `god_exists` returns something other than a bool? Another way of asking this might be: what if `god_exists` is a black box with poorly understood/defined behavior? Because that’s basically the situation we’re in. Also consider: in the absence of theism, the word “atheist” wouldn’t mean anything at all.