Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Time travel paradox only due to choice

  1. Aug 22, 2010 #1
    The title is the best way I could group this thought. And the following is most likely sloppy so I do apologize.
    I get the 'feeling' (subjective I know), that the measurement effecting the outcome i.e. only being able to measure position over velocity, collapsing wave functions, etc... comes down to proof of choice.

    If the interactions of our choices are free to occur without any pre-destination then the concept of time travel is meaningless. I.e. in order for time to travel to work at any point it would mean a specific order of events would have to occur and thus the 'paradox' but also the forced realization that there is no choice. If there is a choice free from any future fixed point then time travel is simply not possible.

    In thinking of that it occurred to me that this holds greater implications than just 'time travel' but also effects the interactions and measurements of physics. for if I imagine the 'ordered' system of the universe without chaos by removing the human element then it seem to run in a predictable manner and 'future' events could always be predicted perfectly. Add the human element in to the system and predictability either becomes impossible with time travel impossible and any measurement of wave function or particle completely dependent upon that measurement and not translatable 'perfectly' to another or it becomes exponentially larger to calculate and time travel is possible and the results of all measurements are predetermined to keep the time line.

    So it's not really a paradox but a choice, you have one or the other not both as they are two separate operating systems. Physics and science in general seems to try and separate out and ignore the human element. We have a direct effect on the system but how does one measure that accurately? every measurement only supports what was trying to be measured when your part of the system being measured.
  2. jcsd
  3. Aug 23, 2010 #2


    User Avatar
    Science Advisor

  4. Aug 23, 2010 #3
    Thanks that was a good read. If my grammar is funny I've seemed to have lost some 'keys' in the firefo rowser. reoot didn't help.

    quik ? wouldn't FW per the artile 3e needed y the manipulator to get alie to think she has FW ?
    ughhh I need to fi this key prolem....
  5. Aug 23, 2010 #4
    Lol Madhatter

    Anyways, back to topic, I think that the concept of FW itself is where the problems come from. It seems to be a paradoxical notion. Let me explain.

    Suppose there is an agent X and an action A.
    X is free to do A iff:
    it is possible that X choses to do A
    it is possible that X choses not to do A.

    Ok. That makes sense for now. Now let's deal with the concept of a choice. A choice is a course of action agreed upon based on some self structure of the agent and the input it recieves. In other words, there is something TO a choice. It is not random and arbitrary, nor is it inherent (i.e. a falling ball does not fall due to a choice, but a person moves do to a choice). It is an interaction between the agent and it's environment. An output the result of COMPUTATION.

    Ok, still makes sense hopefully. Now the problem: if a choice is due to computation of weighted self structure, i.e. it makes sense, the structure will be deterministic, will return the same result every time. If it were not to, that would imply randomness and therefore eliminate the choosing. Therefore it is NOT possible that X can choose to do A and X can choose not to do A. Since that is a paradox, the concept of FW itself is a paradox.

    And a bottom line, what is 'Free Will' supposed to MEAN, anyway? It just seems like a flawed concept to me...
  6. Aug 23, 2010 #5


    User Avatar
    Science Advisor

    I agree that the notion of a "free will" which is supposed to be incompatible with determinism (the incompatibilist position), but also different from randomness, seems to be a fairly incoherent notion, but isn't this more a subject for the philosophy forum?
  7. Aug 23, 2010 #6
    alright got all my keys back! had to trim the flex connector a bit to get a better contact on the board.

    Okay in the above article there was the proposition that another 'observer' could induce a pseudo FW to 'Alice' however that seems to be a contradiction to the point because you now have 'someone/thing' acting outside of the environment while in it, how are we to know that it's actions are not also FW or not?

    My thought was that we either have FW choice and that automatically removes any ability for 'time travel' unless one invokes a theory of multiple splits that collapse but to me that still seems like a way to try and merge the FW choice with the ability to 'time travel' and thus the constant paradox's and circular theories.

    moving thru time at will would be counter-intuitive to that it isn't FW but has to occur to support the system. to have FW and time travel isn't possible as you would invariably end up with 'pinched off' time loops that would be inaccessible to any measurement or observation.
  8. Aug 23, 2010 #7
    I like your style. But, you didn't seem to include "will" in your argument at all. Someone can will themselves to do something, but obviously, that is not a choice. I can will myself to teleport, but i wont. It's not that I did not choose to teleport like nightcrawler, hell I even have the will! but I don't. A ball can will itself to fall, but not choose to. so the will to do something is not the same as choosing to do something. You have that choice is an action, however the context in which you were using the word, it seems to have little to do with will.
  9. Aug 24, 2010 #8


    User Avatar
    Science Advisor

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook