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

B What is the appropriate time/matter model for the universe?

  1. Feb 23, 2017 #1
    We see it frequently stated that in the subatomic/quantum world, there is no reason that the arrow of time points in only one direction, that either direction for particles is equally valid and workable. (If I am mis-stating this let me know.)

    But what is the model of the universe that is proper for a particle, or for that matter, anything with mass, regarding the particle itself. Is a particle to be seen as a long thin noodle that stretches from it's beginning to it's end, with a pointer as to where the current time you are observing and it is interacting, indicating the present according to your observation? And if this is the model, then would the particle's mass be the sum total of the entire noodle?

    Or is the proper model, a series of snapshots, perhaps trillions per second or more, each snapshot a complete picture of the universe and it's interaction with our particle (which way it's traveling through space, it's velocity, etc. This would be a huge snapshot since in essence that one particle interacts in at least a gravitational way with the rest of the universe.) What is the particle's real mass in this model? It's instantaneous mass of a single snapshot? A series of snapshots that make up one wavelength of whatever frequency is applicable to the wave function of the particel? Or is it only the mass in the present frame or a summation of all the snapshots combined?

    Or is the model for the particle something that only exists in the moment, that in what we call the present or past, that it no longer exists, like a traveling wave on a still pond, the wave only existing where it is at the moment, no hint of it behind it or ahead as it passes. In this model it would seem it has only the mass we observe. This seems to me to make some sense. But if this model is true, it seems to say there is no past or future to move to, that there is only the present for any particle, which I believe would hint that time travel is a fantasy.

    And finally are massive particles such as a neutron or proton, subject to different universal models of time and mass, than say a photon or maybe a neutrino?

    The reason I ask, is that it seems that to understand something like time travel in any serious way, or argue as to it's possibility or its possible complications (I went back in time and shot my grandfather), then one has to be working with the right model of what time is, especially when it comes to the relationship of time and mass, since after all that would be one of the more likely objectives of time travel, to be able to move through time as well as space.
     
  2. jcsd
  3. Feb 24, 2017 #2
    That's called a world line. It's just a way of visualizing the fourth dimension.
    Nope. The length of the world line is how long the particle exists, which isn't the mass.

    You talk about time "frames" but you should be aware that time is relative, and spacetime is curved, so it's not so easy to compare time between distant points. General relativity is the theory for the curvature of space time (which is also the source of gravity).
    Model is the word here. It's just a model. The models help us understand and predict reality. If two models make the same prediction, then we say they are the same model, just different "pictures". One picture might say that we are seeing one frame in a block universe that contains the full future and past. Another might say the future is continually created. If the math is the same, it's the same theory, just different picture.

    Nope. They have different masses, but the definition of mass is the same for them.

    Since you are still trying to learn, you shouldn't make assumptions that they are somehow related. If you want to be serious about time travel then you'll have to learn GR.
     
  4. Feb 25, 2017 #3
    Thank you for the response. You say that the world line of the particle is how long it will exist. It seems popular for some to offer that there is all the past universe in all it's moments of existence, still in existence simultaneously with all the other moments past, present and future. It makes sense that the world line represents the length of the existence of the particle instead of the particle existing at multiple points in time, all simultaneously. So what then would qualify as legitimate time travel? And how would one know if they had traveled into the past to a previous point. Obviously one can't set the time machine clock to a universal clock, as I am pretty sure it's agreed there isn't one. So what then? Have an exact snapshot of the universe at a previous point in time, with all information present for all particles in the universe? Then I travel back and keep comparing what I see of the universe around me to the snapshot, and when they agree, I know that I have gone back in time. But have I? It seems to me now that there is a bit of extra mass there, myself and the time machine, as well as the energy I brought with me, and now there are two of me, which if I met the conditions of a perfect replica of the past, two of me now would occupy the same space at the same time doing the same thing. And then I have to account for the time machine now that wasn't in the snapshot of the past. Wouldn't this violate a law of physics or two or three? That the universe had changed it's mass and that two identical particles (me) were occupying the same space at the same time? Are there other laws I have violated in doing this?

    The time travel to the future is easy, I think, since it supposedly has not happened (nothing is or has been there yet,) so there is only one of me there, and it would allow for the earlier invention of the time machine to move into the future as well, without violating any laws.

    Regarding the arrow of time, could it be due to the probabilistic nature of matter? That going backwards in time for anything more than a single particle would require that all particles involved also generated the same probabilistic outcome as they had before? That the dice get thrown the same for all particles at the same moment as they did previously? Though this might be much easier for a single particle, I would think it would be nearly impossible for any large quantity of particles that made up a macroscopic object. That it would be impossible for them all to take their previous state exactly as they were before at that point in time, which would be to reverse their entropy as well. Where movement into the future does not require a particle to meet a state to qualify as having moved into the future. Any state is allowable that is not a violation of the laws of physics. So movement into the future is natural, movement into the past, near impossible. Wouldn't this then dictate the arrow time only pointing in one direction as we see in the real world?
     
  5. Feb 27, 2017 #4
    Most science fiction depictions of time travel are impossible. They probably violate causality, thermodynamics, conservation of energy/momentum. The closest thing to time travel consistent with real physics is closed timelike curves, which are possible structures in general relativity.
    https://en.wikipedia.org/wiki/Closed_timelike_curve
    But it is doubtful that they really exist because they would violate causality.
    If the arrow of time is determined by the direction of increasing entropy, then it's impossible to have a loop in time because the gradient of entropy can't have a curl.
     
  6. Feb 27, 2017 #5

    Nugatory

    User Avatar

    Staff: Mentor

    It is indeed easy. I travelled a few hours into the future this morning while I drank my coffee and read the newspaper.... I wear on my wrist a clever device called a "wristwatch", which reports on the speed of my travel into the future.... seems like I'm pretty much always travelling into the future (and after more than a half-century of forward time travel, the novelty has pretty much worn off).

    The above may sound flippant, but it's not. There's a serious point here: what exactly do you mean by "time travel to the future"? How does the worldline of someone travelling into the future differ from any other timelike worldline? Unlike the hypothetical and almost certainly impossible time travel to the past, it won't form a closed timelike curve.
     
  7. Feb 27, 2017 #6
    Nugatory, I didn't take it as flippant. I understood an agree with what your statements entirely. I am not trying to invent a time machine, or figure out how. Or even prove if it is in fact possible or not. I have a strong feeling however, that subatomic particles may in fact 'drift' around the current time line with a bit of freedom, the less massive they are, the more free they are to drift away from the present 'now' moment. Which then also brings a few questions to mind, specifically if universal laws are broken, like thermodynamics or conservation laws. In the following questions I am really thinking of perhaps nanoseconds and picoseconds into the past or future, not so much years or centuries.

    The specific questions I have are:

    Does travel of a particle (person, photon, neutron, etc.) violate a law of conservation of energy if there ends up being a duplicate in the universe at that past point in time, of the very same particle or a particle that didn't exist at that time in the past previously? What happens when something pops into the past, interacts gaining some energy then returns to the future before that energy is returned to the system that provided it in the past?

    and:

    In moving into the past, all things would have to be exactly as it was before, in my opinion, for one to truly have moved into the past, and not just having moved forward in time into a newly created simulation of the past, or branched into an alternate past. Not only would one have to meet the requirements of particle positions, but all other information in the system would have to be identical. Now given the quantum world's uncertainty, this would mean that each particle's state would have to arise from the same cause it had before in the exact same manner. In other words if we look at a dice problem, and we needed to exactly duplicate the probabilities in an event as a roll of 5 dice, with dice #1, 2, & 3 as 4's and dice 4 & 5 as 3's, with a total of 18, that to truly duplicate the past, one can't just get the 5 die to total 18, but that you have to duplicate 1,2,&3 coming up as 4's and dies 4 & 5 as 3's again. Any other state would not work, though the outcome might appear the same.

    What then are the odds of all particles in the universe duplicating not only their spatial positions, quantum states, etc., that once before existed and are 'known' vs moving forward where the spatial positions and quantum states, etc., are not yet set. This seems to be a real cause for the arrow of time to only point one direction for large systems of particles, yet for an individual subatomic particle by itself, not so difficult to achieve as the statistical odds of duplicating a previous quantum state is many orders of magnitude less.

    This would still result in a much higher probability that the particle moves forward with the rest of the universe on the macroscopic scale, yet give the subatomic particles a small degree of freedom moving against the arrow of time to some degree (non-zero.)

    If correct, then this makes me wonder if on the subatomic level if particles (in either state, particle or wave) such as photons, (especially, since they have no rest mass,) and possibly neutrinos, haven't a higher degree of freedom of moving away from the current time line, a little bit, in both directions with some degree of freedom not enjoyed by large systems. And larger particles where you have more mass, like say an electron or proton are far less free. But in the case of electron/proton or neutron, the electron is a single point particle with much higher freedom statistically to vary from the time line in both directions (forward and past) while the proton or neutron, being made up of 3 quarks, for that to move into the past, all 3 quarks have to simultaneously agree to move into a past or future time, as well as change their locations in space as they move around in space-time.
     
  8. Feb 27, 2017 #7
    You should learn as much as you can about mainstream physics before you start formulating these personal theories. Your intuition is molded by what you know.
     
  9. Feb 27, 2017 #8
    There doesn't seem to be any objection to the 3 dimensional space-time location of a subatomic particle being a function of statistical probability, why then would there be an objection for the same particle having a similar statistical location in the 4th dimension?
     
  10. Feb 27, 2017 #9
    Quantum field theory already treats space and time on the same level. There is always some uncertainty in when a particle is created or destroyed, and when it interacts with another particle. Moreover, there is often uncertainty in how many particles there are and how many interactions have occurred.
     
  11. Feb 27, 2017 #10

    Drakkith

    User Avatar

    Staff: Mentor

    I suspect the objections are mostly a result of the way you've written up your question, as it is very close to what most here would consider a "personal theory". This post (the one I'm quoting) seems to be much more reasonable. You're just asking if the uncertainty principle applies to a particle's position in time as well as in space, correct?
     
  12. Feb 27, 2017 #11
    Basically the problem, how to adequately phrase my questions without being guilty of the crime of 'personal theories'. If quantum mechanics already treats time the same as space, then what about conservation in a particular frame, where a particle that say was interacting in a previous frame, isn't there in the frame any longer. Doesn't this introduce paradoxes?

    And can't things like the wave function be derived from statistical deviations of time alone rather than spatially? Example: Electron cloud distribution in 3 dimensions. Couldn't a single variation in time create the same result, making an electron appear or not appear at a particular location in the cloud at any particular moment, without positing that the path of location of an electron is mysteriously popping in and out at almost random locations? Or am I misunderstanding something there about the electron cloud.

    So then the question I have would be, that if at the quantum level there could be deviations in time, in both directions, (which I think one direction I prefer for reasons earlier stated) then sub atomic particles could be 'time travelers' to a degree? And here I am talking of time traveling of nanoseconds or picoseconds, etc.
     
  13. Feb 28, 2017 #12
    In relativity, conservation laws are written in the form of the continuity equation, which puts time and space on equal footing. The continuity equation looks like
    ##\partial_\mu J^\mu = 0##
    which is another way of writing
    ##\frac{\partial \rho}{\partial t} + \frac{\partial j_x}{\partial x} + \frac{\partial j_y}{\partial y} + \frac{\partial j_z}{\partial z} = 0##
    where ##J = (\rho, j_x, j_y, j_z)##
    Basically, it says that if there is a change in the density, it is due to a current which carried some in or away. You can replace density with energy density or charge density or whatever. There's no paradox.

    There is no conservation law for number of particles. So there is no problem if we are uncertain if the particle is still there or not at some time t. The energy is conserved, so if the particle has energy it means that there is some chance that the energy has changed to some other form. For example, if you have an electron and a positron at time t0. At time t1 in the future, we aren't sure if the electron and positron still exist, because they can annihilate. But if they annihilate, they are replaced by two photons which carry the same energy and momentum. There is an uncertainty in the time of the annihilation, so there is some spread in time of the {electron and positron} state and some spread in time of the {two photon} state, but the energy is conserved.

    You seem to be trying to mentally assign a time to an entire particle, rather than a time to events in the particle's worldline. Remember that a particle is extended in the time direction from when it was created to when it is destroyed. You can have an uncertainty in when the particle was created, or when it was destroyed, or when it collides with another particle. You could write down a probability cloud for the particle, g(t,x,y,z), which gives the probability of measuring the particle at a particular time and place. For an unstable particle, the probability will go down as you check higher values of t.
     
  14. Feb 28, 2017 #13
    Khahishi, I appreciate that explanation.

    But it raises questions. If I were able to know the exact location of an electron, say, at any and all moments, and my watching it could in fact not interfere with it in any way, so that there was no uncertainty as to it's location at any given point in time, would I watch the electron progress in linear fashion around the nucleus? Or would it seem to pop in and out randomly at different locations around the nucleus according to a probability curve?
     
  15. Feb 28, 2017 #14
    You are asking an impossible question. The electron does not have a point location. You could say, the electron's location is the probability cloud.
    Only when you make a measurement you get a random point from the probability cloud. And it will change when you make more measurements.
     
  16. Feb 28, 2017 #15
    I am not sure why it has no point location considering it is called a 'point particle' but if you mean that it exists as a wave around the nucleus, then OK I understand that. Then it would be a wave not a point particle then, correct? But even if it is a wave, the electron exists in a 4 dimensional cloud does it not? Not just a 3 dimensional one? How does the statistical variations in time manifest, considering time is linear? This is where my confusion lies I think. A good answer here could go a long way for me.
     
  17. Feb 28, 2017 #16
    Well, it's the whole duality thing. Whatever it is, we just call it a particle.
    What don't you understand about my previous responses? In particular, in post 12 I specifically said there was 4 dimensional probability cloud g(t,x,y,z). And in post 9 I stated there is uncertainty in when a particle is created and destroyed?
     
  18. Feb 28, 2017 #17
    In spatial references, randomly showing up at point c, then a then b, is just randomly showing up at different points in space. But not following a logical sequential forward motion in time, in the direction that the arrow of time is pointing, would be time travel would it not? Can the particle pop back in time and show up before it was created in an event? You are talking only about the uncertainty of when it is created or destroyed, not it's journey and interactions between if I understand what you are saying. In the spatial references there is equal freedoms in 3 dimensions in all directions from a reference point (center of the atom or whatever) and if the same freedom is available to the subatomic particle, then it should equally be able to move forward or back during it's existence, along the time axis as well shouldn't it? But then there is the pesky arrow of time kind of saying no.
     
  19. Mar 1, 2017 #18
    The particle doesn't randomly show up at different points in space. Rather, we have a chance of measuring it at various points in space. There's a difference. Likewise, the particle doesn't randomly show up at different points in time. Rather, if we make measurements at various points in time, we have chances to measure it at those times. This is a key point that you seem to not be grasping.

    You need to stop thinking of a particle as existing at only one point in time. It is an extended object in time (like a noodle, but fuzzy because of uncertainty). It is also extended in space. You should probably view the particle and the wavefunction as the same thing. If you are viewing the universe as a 4D object, then nothing is "moving". You already see the entire history of everything laid out in front of you.

    Let me rephrase this into a question that makes sense. Can we measure the existence of a particle at a time before we measure the creation of the particle? Yeah. Keep in mind that particles of the same type are indistinguishable from each other. So, we can't really say if the two measurements were measuring the same particle. It could be that we measured a particle, then it annihilated with an antiparticle, and it created some photons, and then the photons generated a new particle (and antiparticle) which we measured at a later time. Naively, you might view this as particles going back in time. Note that this is just one possible "path". What "actually happened" is uncertain. Now when I say uncertain, I don't mean that we are ignorant of what happened, but rather that the physical processes contain contributions from every possible path.
     
  20. Mar 1, 2017 #19
    I think that clears it up for me. Thank you.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: What is the appropriate time/matter model for the universe?
  1. Universe model (Replies: 0)

  2. Modelling universe. (Replies: 10)

Loading...