Monthly Archives: April 2015

Fast note on Scientific Principles, Bell’s Inequality, Relative Simultaneity

I’ve not posted in a long time. There’s about seven draft posts still sitting there waiting to be finished. I don’t have much time since last October. I’ve a 9-6 non-academic job, and any time outside that I’ve been spending on projects with other people.

So, I’ve decided to do a few fast posts. The first one is about the physics of time. It’s a thought I have about it, involving changing my basic views of simultaneity.

At Christmas, I had two discussions with, first, a well-established very-good-at-it professional physics friend and, second, a friend who is equally a very good at it professional philosopher of physics. With both, I brought up a question about relative simultaneity and what Tim Maudlin tells me is the violation of Bell’s Inequality.

Without going into it too much–I am not a physicist and do not want to pretend that I fully understand the physics–the problem is this: One implication of quantum entanglement is that two particles created by the same event–e.g., in a lab–can have properties with values dependent on the values of the other. You can change the value of one particle’s properties by changing the value of at least one of the properties belonging to the other particle.

So, for example, one particle might have a spin of 1 and the other a spin of -1 (don’t worry what it means to have spin ‘-1’. It isn’t really spinning being discussed here).

The two particles – 1-spin and -1-spin – are created and speed away from each other into space. Say they’re now 90,000,000 miles apart. Now we change the spin of the one of the particles, e.g., the 1-spin to -1-spin. The other particle’s spin also changes.

The problem I had with this is it seems to violate relativistic physics. Not the maths as such, but the underlying interpretation.

Fundamental to relativistic physics is the principle of equivalence: the idea that the laws of physics are the same in all frames (Encyclopedia Britannica entry on the principle). Each frame is equally correct for describing physical phenomena.

Included in this equivalence are the descriptions of space and time, e.g., that two events are simultaneous. If two events are simultaneous in one frame, and not simultaneous in another frame, both are as equally correct. This is not incoherent. The events are relatively simultaneous.

This solves some problems. For example, one can discount a physical aether. It is also very like similar relative physical phenomena–for example, some objects are at rest and in motion without contradiction because they are at rest to relative to one frame and relatively in motion to another frame.

As Galison describes it, Poincaré and Lorentz came up with similar mathematical descriptions to the maths in Einstein’s relativity (see Galison’s book). The difference is that they still took there to be a matter of fact whether or not two events are really simultaneous beyond how they are described in a frame. This would make it the case that:

– There’s real simultaneity (or motion/rest) beyond frame-relative simultaneity (or motion/rest).

– There’s at least one frame–a privileged frame–which has values of simultaneity (or motion/rest) which correspond to the real simultaneity (motion/rest).

The point is that a cornerstone of Einstein’s revolutionary thinking is the abandonment of an aether and absolute motion/rest. To do it, and make sense of the experimental evidence, he also abandoned absolute simultaneity (and, as a consequence, any privileged frame).

I have a problem with this when it’s coupled with current quantum theories and the evidence that supports it–that is, experimental evidence from Aspect et al. (Aspect et al’s paper on the much-discussed experiment, discussion on the philosophy of Bell’s Theorem). After discussing it with my physicist friends, I then cold-mailed Professor Tim Maudlin. He sent me a quick, but thorough and helpful reply.

I described the problem to Maudlin as follows:

Part of the typical description of a scenario implied by [Bell’s Inequality] is the claim that, if the spin of an entangled particle is changed, then at the same time, the spin of the particle’s entangled partner also changes. As the particles are spatially separated from one another, this is a description of simultaneity between spatially separated events.

Yet, there seems to be no mention of relative simultaneity here. For example, there is no reference to some frame for the simultaneity and then, say, a further clarification that, to some other frame, the events would be non-simultaneous.

Physicists I’ve discussed it with tell me there is no problem. It is not possible for information about these states to be communicated faster than light. So the speed of light is not met or crossed. Although I can see how this might make the issue experimentally irrelevant [..] it does not, I think, solve the problem. The problem is not what we can know about the entangled state; it is that the state involves frameless simultaneity.

Then I asked him: if this is right, then don’t we have to give up relative simultaneity?

Maudlin replied to me, making two points:

1) He agreed that my problem is a problem–and the typical physicist response tends to miss the point.

2) It doesn’t mean you have to give up relative simultaneity–it’s not a matter of logical necessity. It’s just easier to re-introduce absolute simultaneity. By easier, I take him to mean that it is simpler.

There is at least one alternative which keeps relative simultaneity: Roderich Tumulka’s ‘Relativistic Flashy GRW’ theory.* But many theorists find such an alternative quite odd. Which I take to mean: it is less intuitive than re-introducing absolute simultaneity.

I won’t repeat what he said exactly in his mails to me (nothing problematic in it–it was very helpful. It’s just I didn’t tell him I would make a post from it, and am not comfortable about quoting a private mail from without informing him).

Scientific Principles

I generally hold the following (many do):

a) A scientific theory–including a physical theory– should satisfy scientific principles such as simplicity, intuitiveness and actually being supported by empirical evidence. Given two competing scientific theories, the simpler, more intuitive position with more evidence is better.

b) All of those principles in ‘a’ are trumped by necessity. If denying it leads to incoherence and nonsense, a scientific theory can be complex, counter-intuitive and have no evidence** for it.

However, where competing theories are coherent and somewhat sensible (even if weird), the principles in ‘a’ can select between them.

c) Between the principles in ‘a’, I take it that there is the following trump order:

– ‘Evidence’ trumps ‘intuition’: evidence for quantum physics beats its weirdness reputation).

– ‘Intuition’ beats ‘simplicity’: the simplest view is there is absolutely nothing at all, you’re hallucinating all existence, and this doesn’t need to be explained (that x doesn’t need to be explained is always the simplest view).

d) It can be much more complicated between theories. Choosing between theories, one has to make a kind of global choice (something Maudlin points out to me).

Still, a global choice is about using the set of principles together. It doesn’t involve an extra principle which trumps all of the more local/specific principles (unlike necessity). You weigh up competing theories in terms of simplicity, evidence, intuitiveness, and see what you get.

e) There may be other principles beyond simplicity, evidence, intuitiveness.

The following are not principles of science or physics for a theory: that the theory has been around a long time; that its theorist is charismatic; that the maths is hard; that you can explain it at a party; that it’s getting you published; that the experiments are carried out in the Bahamas (what are you? The ’80s Michael Caine of Science?).

f) If Maudlin is right, it is simpler and more intuitive to re-introduce absolute simultaneity than to take on the alternative position.

That is, it is better to go back on Einstein’s abandonment of absolute simultaneity and a non-privileged frame.

I’ve been working on the assumption of relative simultaneity in most of my work. So that is an important conclusion for me.


*Here’s a paper on that: (not by Tumulka). Here’s a video of a lecture on that:

**Although without evidence I’d tend to call that a metaphysical or philosophical theory–or, if evidence is possible, a speculative theory–but let’s ignore that for now.