10 ms·

I don't even think it's logically incompatible. Isn't this just ((A → B) ∧ (¬A → B)) → B?

by codeflo 3y ago

sowbug 3y ago
Perhaps, but it can be confusing if the listener interprets it as a witness making a factual statement, leading to the rhetorical question "were you lying when you said A, or else when you said not-A?" For legal arguments, it's more like "we contend that you can't prove A, and even if you can, you can't also prove B, and both A and B must be proven for legal liability." Which most people can understand isn't inconsistent at all. That's why the legal & ethical guidelines spell it out.
AnthonyMouse 3y ago
What this comes from is that sometimes you claim something that you can't prove. Suppose you got rid of your dog a month ago. If that's true then the non-existent dog certainly didn't bite anyone. But you still have dog food and leashes and there is a dog registered to you, so that may be tough for you to prove. Even if it's true. But now suppose you can establish that there was a fresh pie on the doorstep when the plaintiff claims to have been there getting bit by your dog. If there was a dog loose in your yard at the time, the dog would have eaten the pie. Since that didn't happen, if you had a dog then it must have been tied up. It's also perfectly consistent with you not having a dog, but it doesn't help you prove that because it's equally consistent with you having a dog that was tied up. The reason this makes people uncomfortable is that the system is supposed to work, but you can easily imagine a case where you in fact don't have a dog but there was also no pie, so the only way for you to win is to establish the thing the jury disbelieved. People don't want to have to conclude that the system would arrive at the wrong outcome in that case, therefore how dare you claim you don't have a dog when there is some evidence that you do.

10 ms·

I don't even think it's logically incompatible. Isn't this just ((A → B) ∧ (¬A → B)) → B?

by codeflo 3y ago

I don't even think it's logically incompatible. Isn't this just ((A → B) ∧ (¬A → B)) → B?

sowbug 3y ago
Perhaps, but it can be confusing if the listener interprets it as a witness making a factual statement, leading to the rhetorical question "were you lying when you said A, or else when you said not-A?" For legal arguments, it's more like "we contend that you can't prove A, and even if you can, you can't also prove B, and both A and B must be proven for legal liability." Which most people can understand isn't inconsistent at all. That's why the legal & ethical guidelines spell it out.
AnthonyMouse 3y ago
What this comes from is that sometimes you claim something that you can't prove. Suppose you got rid of your dog a month ago. If that's true then the non-existent dog certainly didn't bite anyone. But you still have dog food and leashes and there is a dog registered to you, so that may be tough for you to prove. Even if it's true. But now suppose you can establish that there was a fresh pie on the doorstep when the plaintiff claims to have been there getting bit by your dog. If there was a dog loose in your yard at the time, the dog would have eaten the pie. Since that didn't happen, if you had a dog then it must have been tied up. It's also perfectly consistent with you not having a dog, but it doesn't help you prove that because it's equally consistent with you having a dog that was tied up. The reason this makes people uncomfortable is that the system is supposed to work, but you can easily imagine a case where you in fact don't have a dog but there was also no pie, so the only way for you to win is to establish the thing the jury disbelieved. People don't want to have to conclude that the system would arrive at the wrong outcome in that case, therefore how dare you claim you don't have a dog when there is some evidence that you do.