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

How does a clock measure time?

  1. Feb 19, 2012 #1
    Albert Einstein is quoted as having said "Zeit ist das, was man an der Uhr abliest" ["Time is what a clock measures"]. The question is, as per the title of the thread, how exactly does a clock measure time?

    When considering this question we need to consider a few things; namely:
    - what time is considered to be in the physical sciences
    - how the physical processes of a clock measure a secondary physical property called time (if that is what time is asserted to be)
    - how a clock demonstrates that there is a temporal dimension i.e. how it demonstrates the relationship between past and present, or present and future.

    Time in physics
    The first point is something I'm not abundantly clear on.
    Does time exist in Quantum Gravity?

    In Einsteinian relativity, time appears to be physical, dynamical, as well as relative; motion affects its passage for different frames of reference. It appears to be different in Quantum Mechanics, where it appears to take on a more absolute form. That appears to be similar to the view under the interpretation of Lorentzian relativity, which is equally supported by relativistic experiments, as Einsteinian relativity. All three view time as a physical property of the universe - or so I believe.

    A clock provides a regularly occurring, repetitive process which is used for the purpose of comparison. The repetitive process provides a standard unit in which other, different processes are expressed, and then compared to yet other processes, expressed using the same standard units.

    For example, if we take the standard atomic clock, the recurring process there is the oscillations of the caesium-133 atom; what appears to be actually measured, by the atomic clock, is the number of oscillations of the atom, not some secondary physical property called time. Theses oscillations are then used to compare different processes.

    For example; if we say that an object is displaced by a distance of X in 1 second, what we actually mean is that when 9,192,631,770 oscillations are counted, the distance traveled by the object will be X.

    We can then compare other objects using this standard unit of comparison; if an object is displaced by a distance of X+2 when the counter of the clock reaches 9,192,631,770, then we say that the second object has moved more quickly than the first.

    The question is, at what point in this process is a secondary (or tertiary), physical property called time measured - without, of course, simply assuming that it is?

    "Distance is what a ruler measures" is a comparison often used to attempt to explain how a clock measures time, but, much like time, "distance" is just a concept. If we talk about measuring the "distance" between ourselves and a remote object what we are actually saying is how many standard units of measurement - a metre stick for example - could we fit between [an arbitrarily defined point on] ourselves and [an arbitrarily defined point on] the object. If the object we are talking about is a coffee table in our living room for example, we might say how many metre sticks can we lay between ourselves and the coffee table; that number would correspond to the amount of floor between us and the coffee table. While we might say the floor exists, the coffee table exists, we exist, and the metre stick exists, "distance" is just a concept.

    "Length is what a ruler measures" is somewhat different to the notion of a ruler measuring distance, because here we are talking about the physical dimensions of an object, as opposed to the conceptual distance between objects.

    Time, however, is somewhat different to the spatial dimensions of an object; the three spatial dimensions are [in general] clearly observable, however, the question is how do we discern that an object, or a process, has a temporal dimension? The time co-ordinate of an object, or a process, will always be "now"; that is, any attempt to measure a temporal dimension can only be carried out in the present moment. While we may be able to recall a previous state of an object or process, this recollection is just a mental construct, a memory; we may also be able to project a future state of the object, but this too is just a mental contsrtuct, or a concept. The same applies to any mathematical representation of the "past" and "future" states.

    Given that this same reasoning applies to any clock, how can a clock demonstrate that there is a temporal dimension?

    A conclusion that could be drawn is that time is not actually a physical property; it is, however, dynamical and relative but only insofar as it is a mental construct for each individual, and our memories and projections can be distorted. It appears as though "time" is not so much something to be measured, as it is the system of measurement, or comparison.

    All that, of course, is based on the reasoning that a clock does not actually measure a physical property called time; but the question remains, how exactly does a clock measure time?
    Last edited: Feb 19, 2012
  2. jcsd
  3. Feb 19, 2012 #2
    How does a clock measure time?
    It counts a rate of change, or better said a length of change.
  4. Feb 19, 2012 #3


    User Avatar
    Staff Emeritus
    Science Advisor

    A clock is calibrated. The basic measures or units (seconds, minutes, hours) of time are arbitrary.

    Of course, the day is based on the rotation of the earth, and the year is based on the earth's period of revolution around the sun - and of course, the year is not some nice integer number of days.

    The number of degrees in a circle, and arcminutes/arcseconds are arbitrary. Like seconds and minutes, they are based on 60.

    Nevertheless, time is based on those things that affect human life.
  5. Feb 19, 2012 #4
    Again, if we relate this back to an atomic clock for example, what is actually counted is the number of oscillations of the caesium-133 atom; where does the physical, dynamical and relative property called time come into that equation?
  6. Feb 19, 2012 #5
    Indeed, time is the system of measurement that breaks those phenomena into units that can be used for comparison and communication; a secondary physical property called time is not measured by those phenomena.
  7. Feb 19, 2012 #6


    User Avatar
    Gold Member

    My wife is a genius, she can turn "Just a minute" into three minutes or more.:smile:
  8. Feb 19, 2012 #7
    "Again, if we relate this back to an atomic clock for example, what is actually counted is the number of oscillations of the caesium-133 atom; where does the physical, dynamical and relative property called time come into that equation?"

    Again a changing quantity, or a countable number of the changes. Many things can effect change including rate of decay of the atom, yet so far it is one of the more stable items. . This were the rest of it all comes into play.

    If I take the rate of decay as the unit to be measured by and accelerate it from a point and leave an identical one there( at the point) the forces of nature will have will have less effect on one, than the other.

    A relay crude way to explain this : A candle flame. Leave your hand in it you get burnt.
    Accelerate through it and it has less effect. very crude but might get the idea across.
  9. Feb 19, 2012 #8
    Apologies, I struggle to see where "time" comes into play in any of the above.
  10. Feb 19, 2012 #9


    User Avatar
    Gold Member

    A clock actually measures motion. That motion (if the clock is built correctly) will coincide with mans concept of time.

    I say mans concept of time because I have never seen an animal watching the clock.
  11. Mar 5, 2012 #10
    I would say that time is merely a concept, with no physical existence; this doesn't appear to be how time is treated in physical theories, however. Einsteinian relativity appears to treat it as being physical and dynamical. I'm wondering how a clock measures time in this context, without simply assuming that it does; or how a temporal dimension can be deduced from the processes of a clock.

    In relation to time being conceptual, I would say that the concept of a temporal dimension arises due to humankinds capacity for memory and projection, meaning that "past" and "future" are just mental constructs. It corresponds directly to motion, because man's concept of time is entirely based on motion and change; memory of a previous [no longer existing] state is labelled as "the past"; the current state (which is the only state that can ever really be said to exist) is the present; the projection of a state that has yet to materialise is labelled as "the future".

    This obviously isn't anything original, but just worth stating for the purpose of discussion; as I have yet to hear it satisfactorily addressed.

    But just to try and stick to the strict topic of the thread; if we consider the processes of a clock, where is the physical entity called "time" actually measured? The counter of an atomic clock counts the number of events (or oscillations) in the clock i.e. it measure the number of events; where does the measurement of the temporal (and physical) element of spacetime occur?
  12. Mar 5, 2012 #11


    User Avatar
    Science Advisor

    We measure it relative to some other process: this process could be mechanical in the form of a clock like using gravity and a pendulum to utilize a measure of periodicity that eventually translates into 'one second' or one 'time unit', or alternatively use a process like that found in the cesium atom way of measuring one second.

    From these things we use some kind of change to effectively measure time. The changes can be something as simple as the periodicity of a pendulum in a huge grandfather clock to something a little more complicated like the entropy increase law in thermodynamics as we currently understand it.

    Also its important to remember that all of the things that are usually utilized to measure time have a huge dynamic component. In other words if you have some kind of process that does not change, then its really hard to use that process to measure time so naturally we want to get some kind of process that is dynamic with properties that are well understood enough to extract the appropriate information about what kind of unit of time we are looking for.

    One final thing: in order to get any universal measurement it is a good idea to use any universal constant(s) that we can use. If we choose things that are not constants we run into trouble getting different answers for the same thing.

    Since in our current understanding the speed of light has so far looked to be a constant, this gives us a good candidate for measuring something in a more standardized way since so far it has passed the 'universality' test.

    As long as we have some kind of standardization (the pendulum utilizes gravity which is for the most part well understood in the context that it is in in terms of its mechanics), then we can be sure that to whatever appropriate level of accuracy, that there will be the right standardization so that it can be used in many reference frames and therefore have everyone agree on it.
  13. Mar 5, 2012 #12


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    No, it isn't.

    But why is this in General Discussion?
  14. Mar 5, 2012 #13
    Sometimes it is.
  15. Mar 5, 2012 #14


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    And if she is to spend "just a minute" in a shop?? :biggrin:
  16. Mar 5, 2012 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2018 Award

    Time? Like light, do we even know if it's discrete or continuous?
  17. Mar 5, 2012 #16


    User Avatar

    Staff: Mentor

    As I've said many times, there is nothing fundamentally wrong with our concept of time for this question to be useful. You wouldn't ask where (or when!) on a ruler the measurement of length occurs, would you? It's meaningless and you've got to stop seeing a problem with time in such questions: the problem isn't with time, it is with your meaningless questions.
  18. Mar 5, 2012 #17
    No memory = no time. The present is the only reality. How long is the present? For humans it's as long as the psychological smear of past/present we can hold in consciousness, and the vast bulk of that is, in fact, memory. In fact, the present is more like a dimensionless euclidian point: a 'location' with no actual dimensions. (A hypothetical being with consciousness but no memory would perceive the world as a static phenomenon.) We don't know, and I don't think there is a way to know, the authentic rate that point "travels", because we would need some other kind of time to compare it to, and there isn't any.
  19. Mar 5, 2012 #18
    The best established linguistic theories assert that words only have demonstrable meaning in specific contexts and the newest theories in quantum mechanics suggest this principle of contextualism applies to physical observations as well. Whether the cat is perceived to be dead, alive, or in superposition could be merely a question of the specific context in which we take the measurement. Relativity makes a similar assertion that whether we perceive something as time or space merely depends on the context of our relative motion. Thus the simplest and most demonstrable explanation to date is that time can be considered a physical property in some contexts and not one in others.

    Is the cat really dead, alive, or in superposition? Is time really a physical property or not? Who cares! We observe what we observe and the rest I leave to the metaphysicians and mystics to debate. A photon doesn't appear to experience time so its perfectly sensible to tell someone time is not a physical property of photons. Clocks measure time so its makes perfect sense to tell someone time is a physical property of clocks. What matters first and foremost is what we observe, and communicating effectively about what we observe.
  20. Mar 5, 2012 #19


    User Avatar
    Science Advisor
    Gold Member

    Clocks don't "measure" the passage of time directly, they estimate it through the measurement of an oscillating circuit (whether it's electrtical, mechanical, or atomic). This is an important distinction in my humble opinion.
  21. Mar 5, 2012 #20


    User Avatar

    Staff: Mentor

    Yes, clocks count periodic events. But I don't see how that makes the word "estimate" applicable.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook