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

I A Question about the empirical evidence of time

  1. Oct 7, 2015 #1
    I want to put this thread in the right context. After reading a plethora of physics literature for many years I have no doubt the people with advanced degrees in physics who work on this stuff are much smarter than me. I don’t believe my observations haven’t already been considered and dismissed for good cause however; the connections and dismissal are not apparent in the volumes of literature I’ve read. This has left me with nagging questions to which I seek answers? I accept there may be unobservable dynamics of the related phenomena that are not translatable to the observable outcomes however; I do try to put empirical evidence, logical progression, and methodical reasoning ahead of enigmatic solutions, and I am able to put aside intuition where evidence indicates differently. It’s why I am so disturbed by my observations and the lack of literature or answers concerning them. Below is a brief description of what I believe is an accurate assessment of facts and observations. I don’t believe I’ve made speculative assumptions but please correct them if I have.


    I believe most people would agree that in the broadest sense time represents change. I also believe if we asked the question, what is the cause of change; most physicists would agree that fundamentally observable change necessitates (and results in) *energy changing position (*energy in a broad sense from particles to matter). All clocks I’ve studied including atomic cesium clocks base their time output on some increment (a piece) of motion [I can provide the functional details with references for Cesium clocks if needed]. While motion is defined as a change in position per time, the time standard itself is an increment of motion. Just as the phenomenon being measured for a mass standard is a quantity of energy, it seems apparent the phenomenon (and quantity) of the time standard is an increment of energy’s change in position (motion). What has necessitated us to pass over this simple empirical evidence that time is a measure of energy’s change in position, in lieu of our current enigmatic view of time?
     
  2. jcsd
  3. Oct 7, 2015 #2

    Dale

    Staff: Mentor

    This is not correct. Please provide the reference that leads you to believe this and explain what part of the reference you are interpreting as implying this.
     
  4. Oct 7, 2015 #3
    NIST Cesium clocks keep time by adjusting a microwave frequency derived from a quartz oscillator to the resonance frequency of the Cesium atom. As the Cesium atoms pass through a microwave cavity a detector is used to produce a feedback signal that continually tunes the quartz oscillator (from which the microwave frequency is being derived) in a way that maximizes the number of state changes (florescence of the Cesium atoms).When the maximum number of state changes is reached the quartz oscillator is locked and standard output frequencies are derived from it.

    Maximum florescence determines that the microwave frequency is tuned precisely, which in turn was derived from (and used to tune) the quartz oscillator that provides the output frequency of the clock. < (increments of energy's motion)

    Fundamentals of Time and Frequency, Michael A. Lombardi, National Institute of Standards and Technology
    17.3
    Cesium Oscillators http://tf.nist.gov/general/pdf/1498.pdf

     
    Last edited: Oct 7, 2015
  5. Oct 7, 2015 #4
    The time it takes for an excited electron to fall to the base energy level, does not involve any movement, at least as far as my understanding goes. It's not very useful in constructing accurate clock, but in theory can be used to measure time.

    Also I'm not sure what you're actually proposing, but consider the fact that the time seems to go at very similar speed throughout the universe, even if there is no or weak connection between various parts. For example, we don't see one star burning twice as fast as another (of the same mass). Does your concept of measuring time with movement explain, why this is so?
     
  6. Oct 7, 2015 #5

    Dale

    Staff: Mentor

    Reading your referenced material I can see how you would get the impression. An atomic clock does not base its time output on an increment of motion. There is certainly motion involved in the clock, but the time output is based on an increment of energy.

    Think about a pendulum clock. The time output is based on the periodicity of the pendulum. There are also a variety of gears and springs and other moving parts that are important for the practical implementation of the clock. But the thing on which the time output is based is the pendulum, not those other parts.

    In an atomic clock there are many "gears and springs" several of which involve motion, but the "pendulum" is the hyperfine transition which is an increment of energy not motion.
     
  7. Oct 8, 2015 #6
    The concept I outlined is not a predictive theory it is only a logical thought progression of empirical observations. We already have great predictive theories that address the rate at which a star burns. My post and questions revolve around, how I may have erred in those empirical observations, or even if empirically correct why can’t that understanding of time be applied to those existing theories?
     
    Last edited: Oct 8, 2015
  8. Oct 8, 2015 #7
    But what is that observation? You only say that moving things move. I can add that moving things keep their speed.
    But if time was defined by movement, nothing would prevent one star aging twice as fast as another star, since nothing moves between them. That's why we need a stronger definition of time, one that explains why all stars age in the same way.
    Can you clarify, what exactly are you trying to say?
     
  9. Oct 8, 2015 #8
    I think I see how you may have interpreted my statements differently than I intended. I didn't mean to imply time is defined by movement, this may seem a subtle difference however; what I said was empirically I've observed time to be a measure of movement. Trying to put that in proper context, just as mass is a measure for a quantity of energy, time appeared to be a measure of energy’s change in position (movement). Our theories define how energy behaves relative to its mass and movement.

    When I contemplated the time relationship to the empirical change I could observe or imagine, it always appeared time accounted for energies (any energy meaning particles to matter) motion which was the cause of that empirical change. That is what I am questioning here, are there errors in that contemplation or why may it be unworkable with our theories? My assumption is that this has already been considered and excluded for a reason, yet I find no reference to it in literature I've read?
     
  10. Oct 8, 2015 #9

    Dale

    Staff: Mentor

    That is not correct. Velocity is the measure of movement.
     
  11. Oct 8, 2015 #10
    I think you are missing a fundamental distinction. That distinction is between objects that occur naturally and properties that are invented by humans. Energy, position, and time are human inventions. We invent them in an attempt to be able to understand and predict how naturally-occurring objects behave. They are properties of objects.

    Energy can't change position. Objects have energy. Objects have position. We can change, or observe a change in, an object's position. And that change in position might be accompanied by a change in energy. And it may be accompanied by a change in time.

    We arrange objects (or observe objects already arranged) in such a way that a change in their position can be compared to another arrangement of objects' change in their position or energy. Then we compare those objects and use the comparison as a way of measuring what we call time. For example, the hand of a clock makes 24 revolutions while Earth rotates once relative to the sun. When we can observe those two things happening in a repeatable way we declare it to be a good way to measure time. When we notice discrepancies we look for better ways of doing it.
     
  12. Oct 8, 2015 #11

    pervect

    User Avatar
    Staff Emeritus
    Science Advisor

    If you are saying that the concept you are outlinging doesn't make physical predictions, I would agrre. I'm not sure I follow what you mean by "emperical observation", but to me the lack of physical consequences imples that you are basically arguing about philosophical issues. I don't happen to agree with what I can understand of your philosophy. While I could go into some more specifics as to where I disagree with your philosophy, I'm not sure its worthwhile. If we are all agreed on the physical predictions, that is as much agreement as can be reasonably expected.
     
  13. Oct 8, 2015 #12
    I don't believe I have brought any philosophy into this. What I have done is look more deeply at one aspect of physical change from, fusion, to aging, to the position of objects in our universe. I think we can agree everything is comprised of energy in one form or another, and the dynamics that predict and explain those changes are explained in theories like QM and GR. What I have concluded (right or wrong) is the time aspect always seems to define the change in position of those energies and forces? While I am questioning and asking about that conclusion, I don't believe any aspect of it is philosophical.
     
    Last edited: Oct 8, 2015
  14. Oct 8, 2015 #13
    I don’t follow how you arrive at the conclusion each apex of a pendulum swing is not an increment of its motion?

    What has caused you to conclude from the NIST documents they make any direct measurements of the energy in the hyperfine transition? They only discuss detecting the number of photons emitted from the transition.

    Excerpts from the previously linked documents:

    "…the microwave signal in the cavity is tuned to different frequencies. Eventually, a microwave frequency is found that alters the states of most of the cesium atoms and maximizes their fluorescence…."

    "….Those atoms whose atomic states were altered by the microwave signal emit light (a state known as fluorescence). The photons, or the tiny packets of light that they emit, are measured by a detector…"

    "…The detector produces a feedback signal that continually tunes the quartz oscillator in way that maximizes the number of state changes so that the greatest number of atoms reaches the detector…"


    From NIST FAQ (How Clocks Work 3rd paragraph): http://www.nist.gov/pml/div688/timekeeping.cfm

    If the correct frequency can be generated to make the atoms change, then that frequency can be counted or divided down and compared. In this case, the generated frequency is the 'tick

    Velocity is a change in position per time; you can see the reasoning will be circular if time is accounted for by the increment of a change in position of some other energy or object, i.e. a quartz crystal, earth rotation, etc. In such a case those things are a measure of energy's movement; velocity will be a movement in relationship to their measured movements.

    From the previous NIST links:

    "All time and frequency standards are based on a periodic event that repeats at a constant rate. The device that produces this event is called a resonator. In the simple case of a pendulum clock, the pendulum is the resonator. Of course, a resonator needs an energy source before it can move back and forth."
     
  15. Oct 8, 2015 #14

    Nugatory

    User Avatar

    Staff: Mentor

    The text that you've underlined is what happens when we use English to describe something that isn't easily written in natural language. There is no shortage of resonant systems in which nothing physical moves, but it is very hard to describe in English how these systems change from a state described by one point in phase space to another, and then return to the initial state, without using words like "move back and forth". (I managed it in the previous sentence, but it took me several tries and the wording is still awkward - it would have been far easier to have written ##\phi(t)=e^{i\omega{t}}## and then point out that ##\phi## isn't the position of anything).

    Consider that some of the more interesting resonant systems around are quantum mechanical in nature, and in these nothing has any identifiable position so it makes no sense at all to talk about something in the system "moving".

    It is true that most of our measuring instruments have to move some reasonable macroscopic object just so that we can read them. But that's just a matter of how we like to see measurement results presented.... The cesium atom that measures the passage of one second is going through its hyperfine transitions without anything moving.
     
    Last edited: Oct 8, 2015
  16. Oct 8, 2015 #15

    russ_watters

    User Avatar

    Staff: Mentor

    There are other predictable processes that do not involve movement; radioactive decay is one commonly used as a clock.
     
  17. Oct 8, 2015 #16
    Wait a second, but in case of radioactive decay, isn't something moving away from nucleus which just decayed?
     
  18. Oct 8, 2015 #17

    Nugatory

    User Avatar

    Staff: Mentor

    Yes, but that's not what we're using to measure time here. We're just looking at the transition from "undecayed" to "decayed", and the motion or not of the decay products is beside the point.
     
  19. Oct 8, 2015 #18
    Well, I was just trying to follow OP idea for sake of discussion. The transition is happening with movement of both nucleus and outbound particle/photon.
    Same happens with aging stars or black holes, or any other physical object.
     
  20. Oct 8, 2015 #19
    I don't agree. I don't even agree that that assertion makes sense. Energy is a property of everything, but those things are not comprised of energy. Some things, like electrons for example, might be fundamental particles. Fundamental in the sense that they have no internal structure, so in that sense to say they are comprised of anything is nonsense. They do have some properties, though. Like a magnetic moment, a mass, and yes, an energy.

    The measurement of time involves changes.

    It is entirely philosophical. It is one thing to understand physics, or any other science, but it is quite another thing to understand the philosophy of that science. Many college professors teach a science and know very little of the philosophy of that science. When students ask them questions about that philosophy, perfectly valid and highly relevant questions by the way, they are totally puzzled by them.
     
  21. Oct 8, 2015 #20

    Dale

    Staff: Mentor

    Sorry, I must have been very unclear. I agree that a pendulum clock measures an increment of motion, specifically the cyclic back and forth motion of the pendulum.

    What I am trying to show is the distinction between the motion on which pendulum clocks "base their time output" and the motions which are incidental. The pendulum motion is essence of the clock. It is the thing that is used as the frequency standard in a pendulum clock.

    There are also other motions, such as weights or springs, and gears and hands. Those motions, however important for the practical design of the clock, are not the frequency standard.

    In an atomic clock there are many motions which are important for the practical function of the clock, but they are all of the incidental kind. In an atomic clock the essence of the clock, the frequency standard, is the hyperfine transition. Unlike a pendulum, the hyperfine transition is not an increment of motion.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: A Question about the empirical evidence of time
Loading...