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

Time - a Dimension of Physics or Mathematics?

  1. Nov 30, 2008 #1
    I have read a number of threads that argue the ontological status of time in physics.
    One of the most concrete "descriptions" of time in physics is offered by Huw Price.
    Huw Price suggests time is symmetrical, but the order of events in time are not.

    He uses the analogy of a long cathedral hall representing time where all the chairs in this hall face one way representing the order of events.
    The common direction of the chairs is the direction of entropy, but not necessarily the "direction" of time. We can imagine all the chairs being turned to face the other direction, but the hall remains unchanged.
    The symmetry of time void of the order of events is then simply the symmetry of an operational time in the equations of physics.
    I think this analogy is at the root of many arguments about the nature of time and the "arrow" of time. It allows one to separate the dimension Time from the operational definition of time in physics. Which leaves the order of events (the chairs) separate from any causal relationship with the direction of time (the hall).
    As an operational definition, time in physics is concerned with the relationships between the chairs, the time (chair events) of light signals between chairs, etc.
    This "operational" definition of time presents a new problem though: if time is not the order of events then neither is the rate of events the rate of time.
    Such an operational treatment of the dimension time quite literally removes time as a "fundamental dimension" of physics and requires physics discover some new dynamic responsible for the order and rate of events.
    On the other hand, it is possible that present physical dynamics are each partial descriptions of a more fundamental dynamic where all are collectively responsible for the order and rate of events.
    It would then make sense to call such a new or collective dynamic - Time.

    Please feel free to show me the error/s in this line of reason. It is at the heart of my essay on the nature of time and as no one has commented on it, I don't know if its being read, dismissed, misunderstood or ignored.
  2. jcsd
  3. Dec 6, 2008 #2
    Huw Price is probably wrong. The overwhelming evidence is that time is asymmetric.

    You might think of time in terms of sets of objects mapped to some coordinate backdrop in order to unambiguously express any particular arrangement of the objects in some volume. Any particular time of a set of objects is its instantaneous configuration. The flow of time refers to the indexing of unique configurations. This is the kinematic ordering that you referred to in a post in another thread.

    Any dynamical hypotheses are inferred from the kinematics.

    The problem with the cathedral hall and chairs analogy is that there's no cathedral hall that can be called 'time' independent of the changing configurations of the chairs.

    The expansion of the universe is the fundamental motion. It circumscribes the behavior of all subsystems. And it's time asymmetric.
    Last edited: Dec 6, 2008
  4. Dec 6, 2008 #3
    I don't think there needs to be any discussion about whether time is symmetric because I don't think that Chrisc's argument there is talking about time. So I will refer to the word he uses as "timecc" and we can decide whether or not he is talking about what the rest of us would consider to be time, and only after that if we find that he's really talking about time need we discuss time's symmetry or asymmetry or whatever.

    Chrisc, you say that timecc is a thing which is symmetric but which has nothing to do with the order or rate of events. What is it about timecc that is symmetric?
  5. Dec 6, 2008 #4


    User Avatar
    Gold Member

    http://en.wikipedia.org/wiki/Arrow_of_time" [Broken] (A wiki article on the general concept with the seven arrows)

    I get the feeling the causal arrow is being disputed:
    Last edited by a moderator: May 3, 2017
  6. Dec 6, 2008 #5
    If not time, then what's he talking about?
  7. Dec 6, 2008 #6


    User Avatar
    Gold Member

    I'm pretty sure the thesis in there is:

    which relates to the causal arrow, as I posted above.
  8. Dec 6, 2008 #7
    It appears the order I presented the arguments has confused my interpretation with the others.

    Huw Price posits time as something other than the order of events, separating the hall from the
    direction of the chairs in the hall, (i.e.: separating the symmetry of time from the order [entropic] of events)

    Physics today posits time as an operational in the equations of the laws. While recognizing the
    thermodynamic arrow of time, physics upholds the time symmetry of the equations in the face of this
    evidence by claiming entropy amounts to a statistical probability of the order of events.

    When time is defined (handled) as an operational (which amounts to the kinematical
    quantification of the evolution of a system) clearly there is no problem in defining time symmetry.
    Time symmetry in this case is nothing more, or less, than the symmetry of the equations, which
    by their axiomatic structure defines the time symmetry of the laws as the replacement of t with -t.

    My point is that an operational definition of time is not a definition but a descriptive. It is
    the "kinematic" quantification of events where time is a comparative with some agreed system
    we call a clock (see Julian Barbour's and Carl Rovelli's more explicit versions).
    So we have removed any dynamic treatment of the dimension time from physics and replaced
    it with a strictly kinematic comparison and claimed whatever time is, it is not responsible for
    the order of events. But we cannot separate the order of events from the rate of events. It is nonsensical to
    claim the frequency of a bouncing ball is determined by the rate of local time but the
    order of the bounces has nothing to do with the direction of time.
    Separating the order and rate of events from the direction and rate of time leaves physics no alternative
    but to search for the dynamics responsible for the order and rate of events.
    If found, such a dynamic/s would demand the label Time, for at that point, time in its present
    definition will have vanished from the equations of physics.
    The model of time I presented in the FQXI essay contest, does exactly that. It shows the dynamical
    laws are collectively the dynamics of time which can be understood as a single dynamic that is the
    continuum space-time-mass. In treating time this way, the second law of thermodynamics is
    actually predicted, not statistically probable, but the causal mechanics of the dynamic that is time.

    Interestingly, it predicts the vanishing of time in QT just as Carlo Rovelli proposes. Not because
    we should "Forget Time" but because the dimension time is one of three aspects of a continuum
    that vanishes when we measure mass.
  9. Dec 6, 2008 #8
    I think that he essentially started with time as his model, then stripped away all of the characteristics that would cause any of the rest of us to consider what he's talking about to be time, resulting in timecc. That's why he can then make arguments about timecc that appear to be quandaries, because he's using the word "time" although he's no longer referring to anything like what the rest of us would consider time to be.
  10. Dec 7, 2008 #9
    No, that's not what he has said at all.
    He has said what the rest of you consider when you talk about time:
    A. - the evolution of quantifiable kinematical variables of events(sidereal time or clocks),
    B. - the cosmological arrow of time(expansion of the universe),
    C. - the thermodynamic arrow of time(statistical probability of events under the laws),
    D, - the time symmetry of the equations of the laws(axiomatic rules for + and - substitutions in A) ,
    E. - relativistic local time(the Lorentz invariance of A, C and D in all Frames),

    are ALL the evidence of the dynamics of time. Events do not just happen because time proceeds
    unless time is the evidence of the dynamics that drive the events.
    But physics has no dynamic model of time.
    Physics has only the operational definition of time which is a kinematical description driven by
    any number of separate and/or collective dynamics of the four forces of the SM.
    So what drives time - the weak, strong, electromagnetic or gravitational force?
    If you suspect more than one you are heading to a unification of forces.
    If you suspect all four as does physics, you are heading toward the unification of QT and GR.

    So, contrary to what CaptainQuasar has suggested, I am not stripping away all the characteristics
    of time, physics has stripped them away and attributed all our notions and measures of time as
    the kinematics arising from one or more of the physical dynamics of the four forces.

    I am suggesting we can now put them all back together.

    The unification of QT and GR which will require the unification of the four
    forces, arises from the model that unifies space, time and mass as a continuum.
    This continuum "predicts" the thermodynamic arrow, the cosmological arrow,
    the vanishing of time at micro scales and gives a logical, causal explanation
    for each. It does this in a model that lets "discrete" and "continuous" exist
    together as the kinematical measures of a "discrete" cessation of a "continuous" process.
  11. Dec 7, 2008 #10
    I finally found time to go to fqxi.org, download your essay and begin reading it. Thanks for the link. At first glance it looks like there are hours and hours of great reading wrt the submissions on the nature of time.

    As soon as I get through yours I'll post some comments and/or questions depending on how well I think I've understood what you've written. I'll only say now that I think we might have somewhat different ideas on how to approach the subject.
  12. Dec 7, 2008 #11
    Okay, all that stuff you said is very nice but I think you're simply avoiding my main point because I did not mention any of those characteristics of time.

    Once you strip away any relationship to the order or rate of events, what the heck does timecc have to do with what the rest of us would regard as time? And what is symmetric about it?

    You very blithely assume that a timecc with no relation to order or rate of events is perfectly suitable to swap into any physics equation as the t but that does not appear at all true to me. I certainly have a limited familiarity with physics compared to most of the people here but in all the physics I'm familiar with, swapping in some notion of timecc that specified no order to events wouldn't work, it would be useless. Once you could no longer relate a specific time, a specific value of t to an event / a set of other values in the equation the equation would basically be meaningless.

    Perhaps what you're saying is more nuanced and subtle but since you simply ignored me when I asked that question the first time I am not inclined to think so.

    (And of course, when I say "he" I'm talking about Chrisc, not Huw Price.)
  13. Dec 7, 2008 #12
    CaptainQuasar, you seemed to have misread both my posts.
    Why, after quoting my post where I said quite explicitly I am not stripping
    away all those characteristics, do you again ask what's left when they're all stripped away?
    I don't agree with Huw Price, I quoted his paper to make the point that philosophers and
    physicists have separated the order of events from the rate of events leaving the dimension
    time as nothing more than the kinematical measures of a system wrt any other system we
    define as a clock.

    I said "I am suggesting we can now put them all (A,B,C,D and E) back together".
    They are all evidence of time, so what could give rise to all this evidence?
    There is a way to model all of them as aspects or observables of a continuum that
    actually predicts them.

    You seem to be of the same opinion as me, but you don't seem aware
    that physics does not posit time as the order of events.

    I have not ignored you, I restated the whole OP because or your first post.
  14. Dec 7, 2008 #13
    Thanks ThomasT, there is a lot of essays to enjoy and think about.
  15. Dec 7, 2008 #14
    That may well be the issue. If you would be so kind as to furnish an example where time is presented in the context of physics, where any order to events may be assumed, I will do my best to follow along.

    Or possibly I am misunderstanding you in some more general way; I'm not sure I entirely grasp in what sense you are using the word "dynamic".
  16. Dec 8, 2008 #15
    CaptainQuasar, if I understand you, you are asking I offer an example that shows physics
    does not consider the order of events is the direction of time.

    The easiest to speak of is a series of bodies aligned on a radial of a major mass M.
    As we measure the velocity of the bodies approaching M we find those closest to M have
    the greatest v, while all have the same rate of acceleration.
    According to the time symmetry of the laws, gravitation will remain unchanged with a
    reversal of time. Which is to say it remains an "attractive" force or the curvature of space-time
    remains positive toward M, therefore the direction of motion of the bodies remains
    toward M under a reversal of time.
    Assign each body a letter in ascending order of proximity to M beginning with A.
    The order of events we want to observe is the order of the bodies A, B, C, etc.
    passing a point on their common axis (radial of M) of motion toward M. ( say an altitude of 100m)
    This order of events A, B, C, etc. remains unchanged with the reversal of time
    according to the time symmetry of the law of gravitation.
    As the time symmetry of the laws leaves this order unchanged, it is then reasoned
    the direction of time is not the order of events.

    I think this is incorrect. It appears to result from a limited consideration of the evidence
    of time, which results in the mathematical symmetry of the equations being attributed with
    too significant a role in defining the symmetry of time.
    That is not to say the math is wrong, but is too limited in scope to account for all
    the dynamics at play in the kinematics it takes into consideration as the evidence of time.

    To see this more clearly and fundamentally at work in the laws, (although a little more difficult
    to initially grasp) one can look at the founding principles of the Standard Model.
    SM is based on the conjecture (and well founded so far) that the four forces are evidence
    of a greater previous symmetry. The symmetry breaking of SM models the emergence
    of the four forces we observe today.
    This line of reasoning in layman terms can be understood as a series of symmetry breaking
    over time from the earliest time after the big bang, results in a Grand Symmetry of yet unknown
    design, has given way to the smaller and greater number of existing symmetries we observe
    today and which serve as the basis of present law.
    While physics occupies the great majority of its efforts in discovering the model that might
    unify present day symmetries (forces) it has apparently (as far as I've found) not noticed that
    the order of symmetry breaking is possibly the most fundamental order of physical "dynamics"
    predicting a direction of time.
    Far from the kinematical evidence that "describes" the operational treatment of time in physics
    today, the dynamics of symmetry breaking cannot possibly remain unchanged with a reversal
    of time, unless we are willing to throw out the SM and start again.
    A reversal of time must then lead to the unification of symmetries, a "dynamical" process that
    only appears to offer time symmetry when the kinematics are the only observable variables
    and are treated in context of the isolated and partial equations that define the (at present) isolated
    and partial laws. Isolated and partial as they are yet incapable of defining a greater symmetry.
    It is as if physics has written laws knowing full well they are incomplete, yet has neglected
    to consider their incomplete nature when declaring their validity.
    This does not mean the laws don't hold, but that when considering time, a dimension of all
    physics, the laws are too narrow in scope to effectively treat the history of the universe in
    context of the isolated, closed system mechanics they were designed to govern.
  17. Dec 8, 2008 #16
    Are you sure about this? Consider a movie of an apple falling from a tree. What does the time-reversed, rewind version show?

    Are you saying that GR says it's impossible for apples to spontaneously rise from the ground into the tree? That's a good thing, isn't it?

    I agree with you that the current mathematical models are probably too limited in scope, and that there are probably underlying dynamics that might taken into consideration. But this isn't exactly easy to do. :smile:

    I do think that you (vis Huw Price and others) are giving too much significance to the fact that the basic equations of motion are time-symmetric. If you call them time-independent instead, then it's easier to see that there's no conflict between their time-symmetry and the fact that nature appears to be time-asymmetric.

    I got through most of your paper. Not sure I understand it. The following comments have to with some of your statements on page 8 of your paper, The Heuristic Significance of the Principle of General Relativity on the Nature of Time. You write (your stuff is in italics, my comment is in regular type):

    This model leads to very interesting interpretations of classic quantum behavior. How can a propagating “wave’“ be measured as a“particle”, or what is the wave-particle duality of light? This is the most difficult notion of this model that requires considerable focus on the nature of light as a propagating cessation of the condensation of space as a measure of time.

    Photon detection is accomplished by detecting a change in the state of energy of the detector during collision. That this change is isolated to a small region of the surface of the detector is interpreted as evidence of the incident photon also being isolated to a small region of space-time upon colliding and not spread out as a wave-front affecting the entire surface of the detector. When a photon is understood to be a propagating, discrete and finite cessation of the processes we perceive as time, the entire wave front will strike the entire surface of the detector over a very short period of time as any spherical wave front must. But the detection of this collision can only be measured at the first, small region of contact with the detector.

    Because I don't really conceptually understand what "propagating cessation of the condensation of space" means yet, I had a problem with this until I imagined a photon as a propagating spherical disturbance, and that there is a first point of contact with any obstruction. The total kinetic energy of the expanding wavefront is carried by each point on the wavefront. When the wavefront contacts the detector, all of the kinetic energy of the incident wave is transferred to a concentrated area on the detector (surrounding the first point of contact with the detector) and the amplification process is initiated. With the initiation of the amplification process, because the rest of the incident wave is contacting the detector at very close to the same instant, the detection process can only register one detection for each incident wavefront.

    But wrt the two-slit experiment with single photon or electron emission, why isn't more than one detection at least sometimes registered for a single emission. My guess is that it's for the same reason that a single incident wavefront can't result in more than one detection event.

    In the two slit experiment, the kinetic energy of each of the resultant (and subsequently interfering) wavefronts is slightly different and slightly less than the original wavefront. Even though the instant of their initial contact with the detection apparatus is slightly different, it's close enough so that the same reasoning for why a single wavefront only produces one detection event also applies to the two resultant wavesfront and the results of their interference.

    Because the interference pattern is actually produced before any disturbance hits the detector, this pattern will emerge statistically wrt a large enough data sample.

    Now suppose a photon detection corresponds to a wavetrain? That is, that original emission events are some sort of vibratory phenomenon that produces a succession of wavefronts all of which correspond to the detection of one photon? Can the same reasoning be applied?
    Last edited: Dec 8, 2008
  18. Dec 8, 2008 #17
    Oh, okay, when you say that "physics does not consider the order of events" you're just saying that physics doesn't care whether the arrow of time points towards the future or the past. The events are still ordered, it just doesn't matter whether they go from Α to Ω or from Ω to Α.

    I misunderstood, I thought that you were describing some concept of time where events are unordered. Sorry about that.
  19. Dec 13, 2008 #18
    I am sure about this.
    A movie running backward does not represent time reversal, it "displays" the inverse kinematics of
    time forward events.

    No I am saying the reversal of time will result in GR defining the path of a body progresses "away" from M.
    An apple will not rise back onto the branch it grew on as the backward movie analogy suggests, it will
    move away from the Earth as will all mass.

    If we call the equations of motion time "independent" in order to prevent the conflict of time symmetry
    they are no longer equations of motion. If we call the metric of GR time independent, it is no longer
    a model of gravity, it is a ten equation geometry expressing a static state of nothing.

    Almost, when the wavefront contacts the detector the "change" can only be detected at the point of contact as the particles that change, literally change time wrt to local time of the experimental construct. Any additional, contact with the wave as it "washes" over the remaining area of the detector is "undetectable" as the change in time is already made.
    The remaining influence of the wave brings the rest of the detector into alignment with
    the changed time of the area of contact.
    You have to understand the geometry of the model or the propagating wave front of light as a cessation of
    time will make no sense at all.
    Consider a very crude analogy of a river. The flow of water is always in one direction, say left to right.
    Now imagine a very strong disturbance down stream (to the right of your position) that initiates a large
    wave propagating up stream. The river still flows down stream and all the fish in the river are unaware
    of any change in the motion or direction of the river, the force is always directing them down stream.
    When the wave coming up stream passes a fish it is momentarily stationary wrt to you, but it is
    momentarily accelerated wrt to all the other fish moving down stream. The wave imparts momentum
    wrt the flow of the river.
    Replace the water of the river with the constant condensation of space toward mass and replace the
    wave propagating up stream with a spherically expanding wave that momentarily ceases the condensation
    of space over its wavelength and that is what i have described as EM waves.
    Remember the condensation of space is what we measure as the passing of time and you will see the
    propagating wave is a cessation of time. A cessation that imparts momentum wrt local time of all masses
    and is itself an event void of time. i.e. a photon has no mass, but imparts momentum, local time of a photon
    has ceased yet it propagates through time.
    With this crude analogy in mind, the double slit experiment will make sense. There is no need for a
    probability wave or its collapse. You will see a model that is clearly deterministic and removes the strangeness
    of quantum action.
    The only difficulty people have with this model is the idea of space condensing to mass. It seems at
    first to be a nonsensical statement since space has no intrinsic physicality that can be measured to condense.
    This problem disappears when you consider this model a fully relational model of space, time and mass.
    Space condensing is still a measure of the rate of time as in GR, one does not have to imagine a physical
    medium or ether that has density in and of itself, one only has to recognize the density of space is the
    rate of acceleration of space which is a relative measure of the rate of time.
  20. Dec 16, 2008 #19
    Chris, I thought that I posted a reply yesterday, but I don't see it in the thread. Maybe I posted it in a different thread?

    Anyway, the gist of it was that I finished your paper, and need to read it again. :smile: I'll try to do that tonight and then ask some questions.

    I recently changed my mind wrt the meaning of the word, time. I think of it only as a certain sort of index now. I no longer equate time with motion (though motion might be evident wrt the contents of a time index). Not sure if this would be compatible with your conception or definition of time. See the thread in this forum "Does time really move forwards".

    Also, I might have a problem with the part where you mention the uncertainty principle.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Threads - Dimension Physics Mathematics Date
Holiday trip for physics student Saturday at 1:52 PM
Dimension, Temperature & Time: Semantically Misleading? Jun 10, 2014
Is there a fourth dimension? Dec 30, 2011
Higher Dimensions from Maxwell's Equations Aug 27, 2011