by Joy Christian » Thu Jul 20, 2023 2:32 am
Dirkman2 wrote: ↑Wed Jul 19, 2023 8:56 pm
Sabine Hossenfelder, in her 9th of July youtube video about Bell, said that the conclusion is we can either have a) or b), and the nobel was actually awarded for proving that we have either a) or b)
a)measurement independence=>violation of local causality
b)local causality=>violation of measurement independence
Sabine is a friend of mine, but I do not agree with her on this point. She, like 't Hooft, is a proponent of what is usually referred to as "superdeterminism", which amounts to assuming a "violation of statistical independence." Sabine has now invented a new name for "statistical independence" and calls it "measurement independence." That is a rhetorical trick. But "What's in a name? That which we call a rose by any other name would smell as sweet."
Whatever we call it, mathematically "measurement independence" means the following condition on the probability distribution:
p(h | a, b) = p(h),
where "h" stands for "hidden variables" (which I prefer to call "the initial state of the system"), "a" and "b" are the experimental settings, and p(h) is the probability distribution of the hidden variables (or the initial states) "h".
Unless Sabine allows violation of the above condition assumed by Bell, she cannot do her superdeterminism stuff. Indeed, Bell inequalities are derived assuming p(h | a, b) = p(h) explicitly. So Sabine wants to abandon this condition. And voila, Bell inequalities can no longer be derived!
That sounds pretty innocent but it is not. Because a violation of the above condition introduces a subtle form of non-locality. This is very well known. So, by violating "measurement independence" Sabine violates local causality after all. She has not got rid of the problem --- just renamed it. Sorry, Sabine.
.
[quote=Dirkman2 post_id=930 time=1689825402]
Sabine Hossenfelder, in her 9th of July youtube video about Bell, said that the conclusion is we can either have a) or b), and the nobel was actually awarded for proving that we have either a) or b)
a)measurement independence=>violation of local causality
b)local causality=>violation of measurement independence
[/quote]
Sabine is a friend of mine, but I do not agree with her on this point. She, like 't Hooft, is a proponent of what is usually referred to as "superdeterminism", which amounts to assuming a "violation of statistical independence." Sabine has now invented a new name for "statistical independence" and calls it "measurement independence." That is a rhetorical trick. But "What's in a name? That which we call a rose by any other name would smell as sweet." :)
Whatever we call it, mathematically "measurement independence" means the following condition on the probability distribution:
p(h | a, b) = p(h),
where "h" stands for "hidden variables" (which I prefer to call "the initial state of the system"), "a" and "b" are the experimental settings, and p(h) is the probability distribution of the hidden variables (or the initial states) "h".
Unless Sabine allows violation of the above condition assumed by Bell, she cannot do her superdeterminism stuff. Indeed, Bell inequalities are derived assuming p(h | a, b) = p(h) explicitly. So Sabine wants to abandon this condition. And voila, Bell inequalities can no longer be derived!
That sounds pretty innocent but it is not. Because a violation of the above condition introduces a subtle form of non-locality. This is very well known. So, by violating "measurement independence" Sabine violates local causality after all. She has not got rid of the problem --- just renamed it. Sorry, Sabine. :)
.
