If paranormal powers are shown to exist at some time in the future, they will be based on an entirely different set of observations than "observations made while allowing free rein to the effects of chance and bias".
But you cannot be one hundred percent certain about that.
You cannot be absolutely one hundred percent certain that present uncontolled observation does
not include some paranormal ability mixed in along with the "effects of chance and bias".
Maybe I can make this clearer if I put in some figures:
I'm not sure what success Connie claims, but let's assume it is 2/3. In other words, in uncontrolled tests that she conducts amongst family friends and clients, she claims a 2/3 success rate (instead of a 1/10 success rate) in picking 1 card out of 10. Whose to say - without a controlled trial of her claimed abilities - that 1/2 of this success rate is not due to paranormal powers and 1/2 due to the "effects of chance and bias". Then, if the "effects of chance and bias" are completely removed in the controlled trial, she should then have a success rate of 1/3. But this is 1/3
on average. In a single trial of her claimed ability, and assumiing she does has an average success rate of 1/3 with her paranormal powers, there is still a roughly 8/27 chance she will get all three cards wrong. So how can you even confidentally exclude her having paranormal powers as a result of that one test?
So isn't it more scientific, more correct, to say that "this test failed to disprove the null hypothesis that Connie has no paranormal powers", rather than that "Connie has no paranormal powers".
This means we can effectively ignore the possibility of future observations modifying current observations, because they have nothing to do with each other.
Who knows? There may be someone somewhere at sometime in the future who indeed has paranormal ability mixed in with the "effects of chance and bias".
You are acting like we cannot say that there is no effect from aspirin on hypertension when a controlled trial shows no difference between the aspirin and placebo group, because some future observations may show that aspirin has the effect of reducing headache.
Certainly, all we can say, given a negative result in a controlled trial of the effect of aspirin on hypertension, is that the null hypothesis that "aspirin has no effect on hypertension" has not been disproven.
But - and I may be missing something - I still don't see your analogy in shifting to the effect of aspirin on headache. I don't see that I'm doing anything like that with the question of paranormal powers.
How exactly does the latter exclude that possibility? Surely if evidence is found in the future, a reasonable person changes their opinion on X???!!!
That's right. That's my point. Your previous conclusion that "there is no X" was/is wrong. It would have been more accurate to have said "the evidence to date is that there is no X"
Are you serious? When working with an effect that is amenable to the effects of chance and bias, have we ever managed to forego the effects of chance and bias so that it is later discovered that all those who recovered from a headache were in the treatment group and all those who didn't were in the placebo group?
You said "If we discover something that can reasonably to taken to be paranormal abilities, and we look for the pattern of how it shows up on a macro level, that pattern
won't be "people who have little understanding of chance, and of cognitive and other biases, who claim unusual knowledge under uncontrolled conditions but fail to demonstrate anything under controlled conditions".
I've just demonstrated how a 1/3 paranormal ability to pick one card out of ten (in other words, the paranormal ability increases the chances of the person picking the correct card from an average of 1/10 to 1/3), means there is still a roughly 8/27 chance of getting none correct out of 3 trials.
So how can you say "
won't"
Based on the information available to us at the current time, the two reference frames appear to be different. This doesn't exclude the possibility that additional information will demonstrate that they are the same.
That could only be so if the point on Earth at which the M-M experiment was carried out was at rest relative to the absolute reference frame (only then* would the travel times** of the two light beams be the same). It would be simple to exclude this possibility by conducting the experiment at two locations simultaneously (or, if you want to deny Earth rotation, one on Earth and one on a rocket travelling out towards the edge of the solar system). They can't both be at rest relative to an absolute reference frame, so if all four travel times are the same, there cannot be an absolute reference frame.
*another possibility is that that point is travelling relative to the absolute reference frame, in a direction 45 degrees to both light beams. But that possibility would also be covered by the follow up experiment suggested.
**the travel times are not actually measured but laser interferometers can detect any difference in the travel times.
BillyJoe
