[Atomic-S (OP)]

Can it be proven from the Second Law of Thermodynamics that an observer is unable to influence anything in his past, and unable to observe anything in his future?

Please pose each topic as a question. You can change it to a more appropriate question - mod.

[Atomic-S (OP)]

To render the question intelligible, it is necessary to define what we are talking about. The most important concept here is "observer". We make the empirical observation that we cannot redo the last Presidential election, and cannot now produce an unambiguously authentic video of the next Presidential election. And the question is, is the Second Law of Thermodynamics the reason?

The problem begins with defining "observer".  An observer must be some system that is able to note and act upon the characteristics of an event. The action may be a simple as making available information about the event, but if there is no possible action based on the event, then in my judgment there has not been an observation. Thus, a person sitting in a dark theater viewing a dark stage is not observing the play even if the person is present in the theater.  Likewise, neither is a block of wood sitting in a lighted theater observing the play, because nothing happens to the block of wood as a result that would be of any use in, say, writing a review about the play. So, the essence of observation is that the event be connected to the observer as by beams of light, and the observer must be able to record the event as a result.

The reverse process, influencing an event, would appear to be the precise reverse:  The -- "observer" might be the wrong word here, let us say "influencer" -- must connected with the event in some manner, as by a beam of light; and this connection, based upon what the inflencer does, defines what the event does.

So it seems that the entire question boils down to this:  Can we prove thermodynamically that when any two events are connected by a signal,  the former can infuence the latter but never the reverse?

[yor_on]

What about two objects interacting?
Would you say that they 'observed' each other, colliding for example?

[Atomic-S (OP)]

My tentative thought about that is that yes, they are both observing each other. Of course, their collisions are simultaneous.

[Atomic-S (OP)]

Of course, if the objects are separated by a considerable distance, and signal each other simultaneously, that is no longer a single event. It is actually four events: The two sendings of the signals, and the two receivings of the signals.

One other thought:  we must be careful that we do not confuse "objects" and "events".

[Pincho]

I don't think it's true that you can't influence a future event. All I think you have to do is to propagate Quantum Physics ahead of a moving body.

Say for example you have a boat, and the water is the carrier. Then a flow is a carrier wave. Then a waterfall influences the boat ahead of time.

That's all I think you have to do. Influence the carrier wave ahead of the moving body.

So what would this do?

This would allow a spaceship to influence space to give the ship propulsion from spacetime itself.

And I think it's possible.

[evan_au]

It is true that the second law of thermodynamics provides a direction to the arrow of time in "Large" systems: entropy/disorder increases over time.

• This arrow is missing in "Small" systems governed by quantum theory: on this scale, every interaction seems reversible.
• Some proton collisions in the LHC clearly fall into the "Large" size, since it is incredibly unlikely that numerous unstable subatomic particles would simultaneously appear, converge on a single point in space and then merge together to form two protons heading in opposite directions at almost the speed of light.

However, the second law of thermodynamics does not explicitly forbid time travel - especially if an action initiated from the future increased entropy both now and in the future.

Causality is an assumption built into relativity - it is assumed that a system cannot be influenced by anything outside its "light cone", ie nothing can travel faster than light, and all inertial observers will agree on the sequence of events, once light from these events reaches them. This agrees with our experience that we cannot change the past or travel into the far future, but it does not prove that it is a fundamental law.

Some theoretical work done on Quantum Loop Gravity suggests that if you constrain the universe to obey causality, you can produce a model of the universe that looks something like the one we live in. If you remove this constraint, the models produce some bizarre behaviors that look nothing like the universe we experience.
See http://en.wikipedia.org/wiki/Causality_(physics)#Causal_dynamical_triangulation

[lightarrow]

Good post, evan_au, my compliments!

[yor_on]

Well, to me that relies on ones definition of proving something Evan. You can assume with a really large probability that there are no arrows more than the one we have found so far macroscopically.  What the sigma value for such a proposition would be I don't know, but it's huge. The sigma value for a proposition of time reversibility would then be represented relative what value you would give the first proposition. That we can find something to 'reverse' does not state that it must do so, inside the 'system'. To me it's instead a statement of that there is a logic to it. Imagine that we couldn't play a movie backwards, imagine that Feynman diagrams constantly gave us solutions that made no sense when reversed. Then the universe, small and big, would become more magic than it already is, because we still have to define it.

[Ethos_]

 Imagine that we couldn't play a movie backwards, imagine that Feynman diagrams constantly gave us solutions that made no sense when reversed. Then the universe, small and big, would become more magic than it already is, because we still have to define it.

Absolutely, and defining it is what the scientific method is all about. Excellent reply yor_on.....................

[Atomic-S (OP)]

However, the second law of thermodynamics does not explicitly forbid time travel - especially if an action initiated from the future increased entropy both now and in the future.

Interesting concept, although I don't quite understand it.

[Pmb]

Can it be proven from the Second Law of Thermodynamics that an observer is unable to influence anything in his past, and unable to observe anything in his future?

No. All the second law of thermodynamics states is that the entropy of a closed system never decreases with time. There is no possible way to influence past of future events merely because entropy of a closed system can only increase.

Why would you think otherwise?