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

Why do we judge time by light?

  1. Nov 3, 2005 #1
    I have been trying to figure this out in my head and I just can’t make sense of it. Why don’t we use matter, i.e. an oscillating spring, to keep track of the rate of time? I know light has a constant velocity in all reference frames, but does that make it a better source for an accurate rate of time? Any ways this was just buggin me and I thought I would hear what all you guys had to say. Thanks.
  2. jcsd
  3. Nov 3, 2005 #2
    Put your mind at ease. Matter, such as an oscillating spring, is used to keep track of the rate of time. Light is not used.
  4. Nov 3, 2005 #3
    Einstein uses light as a constant when he proves length contraction and time dilation doesn't he? I.e. if there is a light source L meters from a mirror it will take L / c seconds for the light to make the trip. Then from 2 reference frames d (distance the light traveled) and d' can be measured. If I’m not mistaken (which I am frequently) this setup constitutes a basic clock. If he would have used a spring when proving contraction and dilation there wouldn't be special relativity.
  5. Nov 3, 2005 #4
    He starts off saying the laws of physics are the same in all inertial reference frames. This leads to light being constant, which can then be used to derive the rest.
  6. Nov 3, 2005 #5

    This is the first postulate of special relativity, what leads to a Lorentz type transformation, not the actual Lorentz transformation.

    The speed of light being constant is an experimental fact, and is not explained by the first postulate. There are sevreral facts that could be taken as "second postulate", e.g. the famous E=mc^2. The laws of physics could be the same in all inertial reference frames and even speed of light different in any inertial frame, what would lead to a modification of _Maxwell laws, for example.
  7. Nov 3, 2005 #6
    How does that make a clock? Do you intend to stand there with a stop watch or something and time the light beam? The spring in the watch is what is measuring time, not the light beam.
  8. Nov 3, 2005 #7


    User Avatar
    Staff Emeritus
    Science Advisor

    The current acutal definition of time is a certain number of oscillations of a cesium atom. (This has been in use since 1967).

    see for instance

    This is not exactly a spring, but it is very close to a quantum equiavalent to it. A spring-mass clock will give very similar results, but with much less precision, since atoms are all identical and springs are not.

    For the units and defintions used when Einstein was formulating relativity take a look at


    Thus you will see that the second, in Einstein's time, the primary standard for time was based on astronomical measurements, and that the meter was based on a standard prototype bar in Paris.

    These were the defintions of units when relativity was formulated, and relativity applies to measurements in terms of these units, as well as more modern and accurate systems of units, which have changes as we have found more accurate ways to measure both time and distance (see the first wikipedia URL or the NIST site for current SI standards for time and distance).

    Spring clocks have absolutely nothing at all to do with the problem, I really have no idea why you think they would.
  9. Nov 3, 2005 #8
    This is true. I'm going to bed.

  10. Nov 4, 2005 #9


    User Avatar
    Science Advisor

    Einstein used the following reasoning:
    1) given the postulates, you can build a perfectly working clock with light rays
    2) we can calculate what this clock would show when it is moved
    3) since different types of clocks must change the same way with velocity, we can indirectly calculate what every type of clock would show - that is: time.

    of course you can use any clock you like, but afaik you can derive SR only from light clocks. That´s enough, because all sorts of clocks must agree.
  11. Nov 4, 2005 #10


    User Avatar

    Staff: Mentor

    You don't need a watch, you just count how many times the light has bounced back and forth compared to a pre-defined interval that that represents, a la pervect's statement that a second is defined as a specific number of "oscillations" of a cesium atom. You could do the same thing with a bouncing beam of light. You don't measure the oscillations, you just count 10 and define that 10 oscillations=1 second.
  12. Nov 4, 2005 #11
    What is wrong with this argument?

    Let a train be moving with velocity [tex]u[/tex] relative to a "stationary" person. Let one tick of a clock be when a ball, with a constant velocity [tex]v[/tex] (we ignore gravity) relative to the train, goes from the floor of the train to the ceiling which is [tex]d[/tex] meters. Relative to the train, its measurement of time [tex]t = d/v[/tex]. Relative to the stationary frame, the ball travels a distance of [tex]d'=\sqrt{u^2t'^2+d^2[/tex] with velocity [tex]v'=\sqrt{u^2+v^2}[/tex]. So the time [tex]t'[/tex] relative to the stationary frame is [tex]t'=d'/v'=\sqrt{\frac{u^2t'^2+d^2}{u^2+v^2}}=\sqrt{\frac{u^2t'^2+v^2t^2}{u^2+v^2}}[/tex]
    So [tex]t'^2=\frac{u^2t'^2+v^2t^2}{u^2+v^2} \rightarrow t'^2(u^2+v^2)-t'^2u^2=v^2t^2 \rightarrow t'=t [/tex].

    Therefore, since the laws of physics are the same in every reference frame, time is the same as well.
    Last edited: Nov 4, 2005
  13. Nov 4, 2005 #12


    User Avatar
    Staff Emeritus
    Science Advisor

    The problem with this argument is that the velocity of the ball is not v' = sqrt(u^2+v^2) at sufficiently high velocities.

    Note that the speed of light was observed to be constant before the idea of keeping time with a "light clock" was proposed. Until the speed of light was shown to be constant, there was no reason to propose using a light clock to measure time.
  14. Nov 4, 2005 #13


    User Avatar
    Staff Emeritus
    Science Advisor

    Just a quick note on timekeeping in Einstein's time. (No pun intended). While the primary standard for time in 1905 was astronomical time, chronometers were aparently used as a secondary standard.


    talks about how the US Naval observatory used a Frodsham clock (exact details of operation unspecified, but was probably a balance wheel) that was frequently reset based on astronomical events (the primary time standard).

    Frodsham has a website, but no information :-( - other sources indicate they were apparently highly respected clock-makers of the time, Frodsham being able to boast that their company made the clocks for the Queen of England.

    For the purposes of relativity, the primary astronomical time standard would not be particularly useful, as one cannot accelerate the entire solar system. Instead, one would have to rely on secondary time standards, like chronometers of the period, to measure the passage of time on a moving object.

    Quartz clocks apparently did not appear until the 1930's, according to
    http://inventors.about.com/library/weekly/aa072801a.htm, and from a later page in the same source, atomic clocks first appeared about 1949 with an amonia based clock from the National Institute of Standards.

    Pendulum clocks were also used in the period, and were among the most accurate clocks available in the 1900's according to this article, even as late as the 1920's. According to the first URL, however, the US Naval Observatory (USNO) did not use pendulum clocks on a routine basis in 1905, though apparently some pendulum clocks were in experimental use. Although accurate, pendulum clocks would probably not be suitable for use in a moving environement - the ship chronometer of the period would be best suited to any application that was not in a stationary, vibration-isolated room.

    That's as much as I could find out with a bit of googling - I'm not by any means an expert on history.
  15. Nov 11, 2005 #14
    What are the consequences in such a universe? Is it self consistent?


    I guess it would be a long way from our own....
    Last edited: Nov 11, 2005
  16. Nov 11, 2005 #15
    Isn't time just a measuring unit?! there is no real such thing as time, it is concentric/circular. Hence God saying he is alpha and omega. The reason we cannot fathom eternity is because of the way we've been taught to believe what time is. We always measure it in a straight line. Motion and energy. What happens if you have no motion or energy to measure? If light ceases to exist in the universe, does time? I don't know what physics says about this yet.. but it seems like quite the illusion to me at the moment.
  17. Nov 12, 2005 #16
    Your best taking this to the philosophy forums. Time is real, a measurement is real, therefore time is real. If time doesn't exist, then how can it be circular? Something isn't inheriently a circle if it doesn't exist. And when God said he was the alpha and the omega, it ment a beginning and an end, there is no beginning or end if time is a circle.

    What makes you think we can't fathom eternity? And we don't always measure time in a straight line, their are many theories that deal with circular time.

    Measuring what? Be more specific. Presumably if there is no energy or motion in a system, you will measure zero velocity and zero energy.

    Yeah, well stop being so philosophical for a second, look at the fact and you'll see the obvious conclusion.
  18. Nov 12, 2005 #17


    User Avatar

    Staff: Mentor

    No, seconds are a unit, time is a dimension like the three spatial dimensions, and it is, most certainly, real.
  19. Nov 12, 2005 #18


    User Avatar
    Staff Emeritus
    Science Advisor

    I agree, a question about "what is real" is definitely a philosophical question, not a scientific one.
  20. Nov 12, 2005 #19
    thanks for your guys feedback. You're right, but i don't get this last post... science is about what is real. We wouldn't even have science without questioning. We wouldn't even have physics, remember how Einstein came to his conclusions? Pure brainstoriming about what is real. Philosophy and science have to work together.
  21. Nov 13, 2005 #20
    No, he brainstormed about things that are real, he didn't brainstorm over whether or not they are real, like you are.
  22. Nov 13, 2005 #21
    I think the question "Is time real" is more a philosophical question, but a very fair and interesting one to ask in philosophy.

    What IS a scientific issue that has been answered clearly is to know that any statement about the time of an event in the universe has absolutely no meaning independently. The time of an event must reference a location (a reference body in space in motion) to have any meaning.

    My limited understanding is that that's one bottom line about time.


    Last edited: Nov 13, 2005
  23. Nov 13, 2005 #22
    This was my original preponderance. Light was shown to have a unique property (constant velocity in all internal reference frames), so why did we decide to use it as a clock? We knew it didn't follow the usual propagation rules of matter... If Einstein had used eok20's ball for a clock, why couldn't he have said that light has unique properties? This would lead to a time and space constant relativity.
  24. Nov 13, 2005 #23


    User Avatar
    Staff Emeritus
    Science Advisor

    Science is about what we can measure, and how to predict the outcome of experiments.

    We know how to measure time, so it's a suitable subject of the investigation of science, regardless of whether or not it's "real".

    I would say that scientists, as a whole, tend to be realists, because some philosphical positions don't expect the universe to be predictable and understandable. Since the goal of science is to predict and to understand the universe, one would expect that scientists would have a philosophy in which this was possible. Of course, this is somewhat of a meta-philosophical point, a philosophy about philosophies.

    As far as time goes, one can regard it as fundamental, which relativity, being a classical theory, does, or one can perhaps take the view that time is only an approximation, valid over large distance scales, something that doesn't work at a small enough scale - i.e. time might be like Wheeler's quantum foam. I gather that this is how at least some string theorists view time, though I'm not terribly familiar with string theory.

    The problem is that just assuming that time _could_ be an emergent phenomenon based on other more fundamental phenomenon is not sufficient to make predictions, thus it is not a scientific hypothesis. Remeber, the goal of science is to make predictions, not to engage in endless philosphical arguments.

    If you accept that time as measured by clocks is useful in everyday life, it becomes worth studying, regardless of ones philosophy about what it "really" is.
  25. Nov 17, 2005 #24


    User Avatar
    Homework Helper

    The first reply answered your question. We DO use matter to measure the rate of time...

    Light is NOT used to define time... Time is time. And light happens to go at the same speed in all inertial frames (in terms of that time). If we didn't already have a notion of time we wouldn't even be able to say that light has a constant speed in all inertial frames... If time is only defined in terms of the speed of light, then "the speed of light is constant in all inertial frames" turns into a truth by definition, and not an experimental truth. The concept of time is independent of the speed of light. And through experiment and observation we find that light has a constant speed in all inertial frames in terms of this notion of time that we ALREADY have.

    The laws of physics are the same in all inertial frames... Suppose we have a digital clock that passes from 2:00 to 2:01 in some inertial frame (suppose it is at rest). So 1 minute passes.

    Now we put the digital clock in a spaceship going at some speed v... we reset it at 2:00... now inside the spaceship the clock passes from 2:00 to 2:01... Now how much time has passed inside the spaceship? By the principle of relativity... everything should be exactly the same during the physical process of the clock going from 2:00 to 2:01... So if 1s passed when the clock was at rest... then 1s passed inside the spaceship... since we've repeated exactly the same experiment in the spaceship.

    Time is intrinsically connected (sort of by definition) to physical processes... a unit of time measures how long a physical process takes... The same physical process (such as a digital clock going from 2:00 to 2:01) in a different inertial frame by definition takes the same amount of time within that frame.

    We could redefine time... but why do so? There is no reason. And when we're on the spaceship it still feels to us like 1 minute passes when the clock passes from 2:00 to 2:01 (what we feel is identical I suppose because the physical process in our brains is identical). So it doesn't make sense to me to define time differently when we're in a spaceship as opposed to at rest.

    So to answer your question... time is already defined independent of the speed of light...

    Please correct me if I'm wrong here... Although the principle of relativity is an experimental truth... with regards to time it could be taken as a truth by definition of sorts: "Time is defined identically in all inertial frames." So if 1 min is defined as the time a digital clock takes to go from 2:00 to 2:01 in the rest frame... then 1 min is also the time it takes for the digital clock (now at rest in a moving frame) to go from 2:00 to 2:01 in the moving frame.

    It seems mighty inconvenient to redefine time when you're in a different inertial frame. I don't think there's anything theoretically wrong with doing it, but it seems impractical. You'd have to pick some absolute frame, and go from there...
    Last edited: Nov 17, 2005
  26. Nov 17, 2005 #25
    huh? Can i just ask a few questions here... Yes he did brainstorm about what was real.. because it wasn't real before he came along. First he brainstormed, then he proved it.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook