I was mulling over some GR equations, when something hit me. We often say in physics that certain things can't be possible because they would violate causality, but just how well-founded is the invocation of causality to constrain a physical theory anyway?
Starting from classical physics, [math]F = ma[/math] really doesn't say anything, except to provide a meaning to the word "force", until we start attaching the implication that a force *causes* a mass to accelerate. Otherwise it, and everything else in Newtonian physics, is just a bunch of acausal couplings between expressions.
The problem gets hairier under relativity, because simultaneity now goes out the window. Two observers in different inertial frames cannot agree on the simultaneity of events, yet it is asserted that they can still agree on a partial ordering of some events, so long as a lightlike or timelike interval seperates them.
(cont.)
(cont'd)
While Newtonian gravity expressed the interaction of bodies in a basically acausal, yet simultaneous way, GR suggests that the bodies (represented by the stress-energy tensor) have an effect on the geometry of spacetime (the metric tensor), which then affects the bodies in turn (the geodetic equations). This might not be a problem, except for the fact that their exist valid solutions like the van Stocktum, Gödel, and Alcubierrie metrics allow you to create a circle of events which are still seperated by timelike intervals (closed timelike curves), which means you could have a seemingly paradoxical cyclic causality. To date, all responses to these metrics have been the suggestion of ad-hoc conjectures that invoke causality as some kind of universal force field, something which strikes me as desperate and unsound.
When one tries to natively combine the equations of relativity and QM, we get the timeless, acasual, and nonlocal Wheeler-DeWitt equation. This is presently interpreted as showing some essential incompatibility between GR and QM, but what if what that equation has been demonstrating, and we've been ignoring, is that this is precisely how the universe operates? That time, space, and therefore causality and locality, are merely emergent properties of laws which are have none of those things?
What would the implications be of ditching the reliance on causality? Is it even possible to make a coherent theory without it? Can we even prove that the universe is logically consistent in the first place?
because we have absolutely no experimental evidence for acausality
also metrics such as godel's do obey causality, and also just because it's a mathematical solution doesn't mean it's in any way a good physical solution
EFE break down at certain extremes, we already know this. is meme theory the way to fix that breakdown? probably not, but it sounds like you don't actually care about physics being a science (rather than physics being a study of mathematics), so i guess you'll be a great meme theorist
>>8093369
I didn't say acausality as such had been observed, certainly not at normal scales. My question is if it's possible that causality might a) be an emergent phenomenon of something which is acausal, and b) if that causality might itself break down under extreme conditions.
As for the "not all mathematical solutions are good physical solutions", well, then we still have to explain what those constraints are and why they exist. Physics tends to follow the math quite closely. No one had ever observed antimatter when Paul Dirac first wondered what would happen if he did't discard the negative solution to a particular quadratic term in the relativistic Schrödinger equation.
Philosophy doesn't belong here.
Also go and conduct a thought experiment and you'll see how retarded the idea of acausality is.
>>8093300
But you said it yourself. Causality is (likely) an emergent property.
So long as your model is modelling the emergent universe and not the unknown underlying laws you conjecture, you have to include causality or your model will obviously be wrong.
>>8093616
I'm talking about the effect a particular axiom has on the construction of physical laws. That's not philosophy, that's science.
>>8093624
Yeah, but if we drop our strict reliance on it as a fundamental postulate, then we open the possibility that certain extreme conditions can break it, even if those conditions are so extremely improbable in nature that they've never existed in the universe's history.
>>8093293
the mere fact that causality has a certain finite velocity means there is some underlying mechanism that tells it what it should be.
Maybe it's more obvious to me than to other people, but I firmly believe in this.
>>8093651
I'm not sure I follow you. You can theoretically create a causal connection between events that would normally have a space-like separation using a Krasnikov tube to provide a timelike worldline between them. That doesn't itself break causality, at least not until you use a second one to make a return trip, but it does break the limitation on the speed of interactions from the perspective of a distant observer. So there has to be something more fundamental going on.
>>8094361
I was just saying that since causality is not infinitely fast, I feel there must be an underlying mechanism that tells it how fast it should go.
>>8094373
maybe we're even in a universe that is a slice of a continuum of universes with each slice being defined by a different constant speed of causality :')
>>8094373
The speed of light is actually a semi-arbitrary scaling factor that connects our units for duration with our units for distance. In fact, right now it's not actually measured: it's *defined* by the SI, with the meter set from it by way of the caesium-transition second. Even though we percieve them as being different things, they're actually all just different components of spacetime.
>>8093369
>because we have absolutely no experimental evidence for acausality
We also haven't played with warp drives or wormholes yet.
I'm pretty sure causality is correct, but I'd really like to hear something beyond "we've never seen it be wrong".
>>8094390
That's the thing: I've never heard of a rigorous explanation for why it should apply in all situations, and in all spacetime topologies.
I'm not the first to notice this. http://philsci-archive.pitt.edu/4690/1/CausalFundam.pdf
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.129.191&rep=rep1&type=pdf
and
http://www.pitt.edu/~jdnorton/papers/003004.pdf
have all noticed this, but they've mostly come at it from a philosophical angle. I'm more interested in what the concrete implications are, assuming we define causality in a consistent way.
>>8094390
then it's not science
you know you're on /sci/, not /x/ right?
>>8094400
I don't have a response to what you're saying, but I'm just dropping in to say it's very, very interesting and I hope you keep posting about this here some more.
Reading these pdfs now.
>>8094451
>then it's not science
>you know you're on /sci/, not /x/ right?
Gregor Mendel never saw DNA, but he still studied genetics.
And I hope science will, someday, address OP's question.
But for now, all we have is: "i've never seen it not be so, so shut the fuck up".
And that's not science either.
>>8094400
Without causality effects predate cause.
If a particle could travel faster than causality, it would be rudimentary to send information backwards through time.
First of all, sending information to you past and future self breaks symmetry and that's the least of your problems because essentially all of physics is broken. Also if time travel was possible, we would expect to see advanced lifeforms everywhere and a bunch of other phenomena. The lack of such evidence makes non-causality incredibly unlikely.
Also, literally all experimental evidence and observation of all phenomena have always adhered to causality and it explains and predicts many observations including the speeds of mass-less particles and gravitational waves. It has an arbitrarily large order of magnitude more support than every other theory combined.
Of course it is not impossible but even more likely is that this universe is a simulation, you are God, my fart was the big bang, etc.
>>8094390
>I'm pretty sure causality is correct, but I'd really like to hear something beyond "we've never seen it be wrong".
"We've never seen it to be wrong" is not that bad of support; it is the basic support for all scientific theories, after all. That said, it does not mean that these bizarre metrics are forbidden, just that they do not arise in nature (which is the case for a lot of things with interesting properties, like two dimensional electron gases).
As for how to deal with these "acausal" solutions, I had thought that they weren't really a violation of causality as causality is essentially defined by the light cone of the system of interest and these things seriously fuck with the light cone. For example, I thought that the Alcubierrie metric got around this issue by arguing that nothing locally moves faster than light and that the apparent faster than light effects arise from messing with space-time geometry to make the light cones work out [I could be entirely wrong; I know fairly little about GR or HEP theory aside from what I use in CM].
>>8094617
What the Alcubierrie metric allows you to do is get around the infinite energy requirement to accelerate an object past the the speed of light, and the effects of length contraction and time dilation on the travelling object. It does *not* remedy the possible violations of causality that can result from completing a round-trip through different inertial frames. If anything, it turns the causality violations from an unphysical hypothetical, into a potentially real hypothetical.
>>8094400
Best waifu
>>8093293
causality is a convenient concept. It's real, but only in the same way money is. You do not need to worry about its influence on the universe, only on your decisions and calculations.
>>8095561
And the same can be said for Newtonian mechanics, and caloric theory, and classical electrodynamics. Yet we know all those theories break down under extremal conditions. Therefore, if causality is a correct description of the universe only in the limit of something, then it stands to reason that there are situations in which it no longer describes reality.
By the way, is there anyone here who's good with tensor calculus? I feel like I've been going in circles trying to understand the notation.
>>8096876
I understood the basics at least
What do you want to know?
Oh and where are you from?
>>8093339
This
