# How does relativity define time (as a constant or not?)

1. Jul 17, 2014

### amarshalltd

Hello everyone! My question is this: Do we believe time is a constant and forces such as gravity affect our perception of time? Or do we believe that time itself is being slowed down by forces such as gravity?

I ask this because I had an interesting thought while driving to work. If time is a requirement for change, then how would the big bang be possible if time was a property of our universe specifically (i.e gravity slowing down time)? Wouldn't time need to exist outside of our universe, and if so, how can our universe slow it down?

Last edited: Jul 17, 2014
2. Jul 17, 2014

### ghwellsjr

Relativity defines time as what a clock measures.

Since different clocks tick at different rates under different conditions of gravity and relative speed, time cannot be constant and is running at different rates depending on those influences.

Since there are no clocks outside our universe, we don't have a definition of time that would be applicable outside our universe.

3. Jul 17, 2014

### amarshalltd

I understand that. I guess what I was trying to get at was is it possible that time is actually a constant, but it is gravity or other forcers that are changing the clocks time, not time itself being slowed down?

4. Jul 17, 2014

### Staff: Mentor

How would you test this? If all clocks are affected, how could you tell that "time is actually a constant" when there is nothing physical that reflects this "constant time"?

5. Jul 17, 2014

### ghwellsjr

The problem with speculating "that time is actually a constant" is in defining exactly what that means and identifying under what conditions time is not slowed down. This is exactly what was being proposed by Lorentz and others prior to Einstein's 1905 paper introducing Special Relativity.

6. Jul 17, 2014

### Maxila

To even speculate more accurately about your question you need to define more precisely “what clock’s measure”… This I have studied quite a bit and every clock I’ve looked at from Cesium clocks to the Earth’s orbit and rotation (arguably the first clocks used by man) measure motion, or being more scientifically accurate, they measure a change of position and its energy as it relates to the position change within a distance. This is what we usually refer to as “speed”; however it is important to note that scientifically speed is defined as a change of position per time (dx/dt). Where x is distance, t is time and s is speed (d is a delta).

In empirical reality (that includes clocks), time and speed are coincident and reciprocal, you cannot observe one without the other, we see both t = dx/ds and s = dx/dt. A unit of time measured by a clock encompasses a speed as the ratio of x/s that it was derived from. An easy physical example of this is the Earth’s rotation where one rotation is both a day of time and a ratio of x/s at any given latitude; here they (time and speed) can easily be seen as coincident and reciprocal sides of the same physical phenomena (a change of position and its magnitude within a distance).

As far as this can relate to your question above, as best I can determine, this empirical reality is consistent with the math of SR including Lorentz transformations and Lorentz invariance; however it may be problematic with GR, yet if empirically tested, it is accurate and factual. You likely have to do a lot more research if you want to relate time as a constant, relative to both motion and gravity. Personally I think time is just the same as distance and speed in that they are all relative for a frame at rest, and time is exactly what it appears to be empirically, one way to describe the measurement of a change of position, with speed being another.

Last edited: Jul 17, 2014
7. Jul 17, 2014

### Staff: Mentor

This is not what Cesium clocks measure. Atomic clocks in general are based on the frequencies of electron energy level transitions within atoms; such transitions are not "changes of position" in any meaningful sense.

8. Jul 17, 2014

### Maxila

Cesium clocks count oscillations from the known frequency emitted for the hyperfine transation of cesium 133. The "frequency" unit or "oscillation" is merely a measurable increment for lights change of position, analogous to one Earth rotation (a day) being a measurable increment of its change in position. A physical process is taking place that is being measured.

As you can see the "frequency" is a known incremental change of position of light (x per c).

Last edited by a moderator: May 6, 2017
9. Jul 17, 2014

### MFPunch

To answer the OP's question no, most people do not think of time as a constant. There's all kinds of theories and metaphysical debates you can have about the true nature of time but it isn't a constant. Because it's relative. Literally that's the opposite of a constant.

10. Jul 17, 2014

### Staff: Mentor

No, it isn't. This is more a question of quantum mechanics than relativity, so if you want to go into this in more detail, you should probably post a separate thread in the quantum physics forum. But the term "frequency" as applied in this case does not mean what you think it means; it has nothing to do with any "oscillation" in the ordinary sense, where the position of something changes in a periodic fashion.

The quote you give does not say that; this is your (incorrect) interpretation of the term "frequency". It is true that one quote uses the unfortunate term "crests of a light wave"; this is what comes of trying to describe quantum phenomena in ordinary language. The photon measurements by a detector that they are talking about do *not* measure the motion of anything; they measure the energy of the photons, which is directly proportional to their frequency according to quantum mechanics.

11. Jul 17, 2014

### amarshalltd

Thank you to everyone for the replies!

I believe the root of my confusion comes from our definition of a constant (such as the speed of light, for example). If time is NOT constant, then how can the speed of light be constant? A black hole, for example, would change the speed at which light is traveling because it is changing time. If time was constant, however, then the speed of light would remain a constant, as its speed is unaffected by time changing. What would be affected is space by gravity. I would think that it is the extension or compression of space by forces such as gravity that causes the illusion of time changing speed, because it takes a longer or shorter period to get from point A to point B. Light approaching a black hole would always be traveling at the speed of light, but its destination (the center of the black hole) would be an enormous distance away because of the stretching of space.

Last edited: Jul 17, 2014
12. Jul 17, 2014

### Maxila

There are many credible references all over, such as this NIST publication showing the direct measurement of wavelength as a property for a frequency of light. http://nvlpubs.nist.gov/nistpubs/sp958-lide/191-193.pdf

13. Jul 17, 2014

### Staff: Mentor

This does not describe how they actually "directly measure" the wavelength. Do you have a source that does? And does that source say they actually measure the distance between wave crests? My understanding is that these "direct" wavelength measurements are really indirect, because they are not direct measurements of the light wave crests, they are measurements of something like the length of the cavity in which the waves are created. Once again, you can't go by ordinary language descriptions of something like this; the words don't always mean what you think they mean.

14. Jul 17, 2014

### Staff: Mentor

Because saying "the speed of light is constant" is a bad way to describe the actual underlying physics. A better way is: "all objects move on or within the local light cones", where the "light cones" at each event in spacetime are the set of null rays passing through that event. Since locally, any spacetime looks like the flat spacetime of special relativity, the speed of light is constant in any spacetime for the same reason it's constant in SR. In other words, "the speed of light" (and indeed "speed" in general) is a local concept.

Thinking in terms of light cones also gives a better way of thinking about what gravity does: gravity tilts the light cones at a given event, relative to the light cones at other events. For example, in the empty space surrounding a black hole, far away from the hole, the light cones look pretty much the same as they do in flat spacetime: light rays in a spacetime diagram are 45-degree lines. But as you get closer to the hole, the outgoing sides of the light cones tilt inwards: outgoing light rays on a spacetime diagram get more and more vertical. At the black hole's horizon, the outgoing sides of the light cones are exactly vertical: outgoing light rays stay at the same radius (the horizon radius) forever.

This means that the effect of gravity on light is not due to "changing time" *or* "changing space"; it's due to changing the light cones. I would recommend rethinking your views in that light (pun intended ); I think it will help to make things clearer.

15. Jul 17, 2014

### MFPunch

That's not what a constant means. Constant doesn't mean "unchangeable". If you are on the surface of a planet and I am traveling in a ship at 99% the speed of light away from you if I shine a beam of light out of the front of my ship I will observe it moving away form me at velocity c. (c being the speed of light in a vacuum) On the planet you will also observe it moving away from you at c. We will both observe it traveling at the same velocity c regardless of how fast we are moving relative to each other.

That is why it is a constant. It has a invariable value: 299 792 458 m / s. No matter what your particular frame of reference is relative to another you will both always observe light traveling at 299 792 458 m / s in a vacuum. Unless some outside force is acting on it there is no possible way to ever observe light traveling at any velocity other than c. Time has no constant value, there is no baseline. Light that is traveling at any velocity other than c has been influenced by some outside force. There is no corresponding value of time that you can say that about. Which is why it isn't a constant.

16. Jul 17, 2014

### amarshalltd

That makes perfect sense MFPunch and PeterDonis, Thanks for your time:).

17. Jul 18, 2014

### ghwellsjr

I'm always fascinated by statements like this saying that it is possible to observe light traveling at c out the front of a ship. Could you please elaborate on exactly what you mean? How do you observe the speed of light?

18. Jul 18, 2014

### Maxila

It doesn't matter, unless you are willing to claim light does not change position, then just as every clock in the past has done the Cesium clock uses an increment of motion (light) to derive time. The arguments you brought up obfuscate that simple truth; even the technologies that are being developed to create even more accurate clocks are mostly trying to measure smaller increments and shorter wavelengths of c. However light is the underlying change of position used and broken down into these small increments of time (this basic mechanic has been unchanged in every timekeeping device). The consistency of c and the ability to accurately measure small increments of it are what make these clocks so much more accurate than clocks of the past.

Last edited: Jul 18, 2014
19. Jul 18, 2014

### Staff: Mentor

This does not follow. The fact that light changes position does not mean that changing position is the only thing it does, or that position is the only change that can be used as a time reference.

No, they are trying to measure smaller increments of *energy*. Once again, do you have any specific reference that explicitly describes a measurement of "the position of light" used in a clock? I'm guessing the answer is "no" since you haven't given one up to now.

20. Jul 18, 2014

Is not the true constant "space-time"? a relative combination of motion and mass?

21. Jul 18, 2014

### Maxila

By now we both know we won't agree on the meaning of this time keeping process; that's no surprise considering the last survey I saw of accomplished physicist's noted how equally divided they were on the interpretations of QM and their own pet ideas.

I will provide a link to the National Physical Laboratory on the topic that I believe also supports my arguments http://www.npl.co.uk/educate-explore/what-is-time/how-do-atomic-clocks-work

Also note the function of prior clocks and any observable evidence of time support a much higher probability of my arguments being correct. As not to hijack this thread off topic any further, this is the last comment I will make in it on the matter.

22. Jul 18, 2014

### Staff: Mentor

Here's how this article says the frequencies of the energy level transitions in the cesium atoms are measured:

This has nothing to do with measuring the position of anything: "radiation patterns" just means measuring the energy of light emitted as electrons transition between energy levels.

In other words, your argument is: the clocks you're familiar with work by something changing position, therefore all clocks must work by something changing position. This is not a valid argument. "Probability" doesn't come into it: it's a simple question of what physical measurements are involved in atomic clocks. Every article I've seen, including the ones you've linked to, says those measurements are measurements of energy, not position.

23. Jul 21, 2014

### Maxila

You know what a hyper-fine transition is (the emission of a photon of a specific frequency by an electron), you know what a frequency is and what it is composed of (waves or an oscillation within a change of position). The official standard of a second is a frequency standard defined as, "The second is the duration of 9 192 631 770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium 133 atom" http://www.bipm.org/en/si/si_brochure/chapter2/2-1/second.html

Now look here: http://www3.nd.edu/~techrev/Archive/Winter2002/a4.html [Broken]

"A detector at the end of the tube gives an output according to the number of cesium atoms striking it and peaks when the frequency is absolutely correct. This peak is then used to make slight corrections to the crystal oscillator that controls the clocking mechanism, locking in the frequency. This locked frequency is then divided by 9,192,631,770 which results in the familiar one pulse per second."

Is the "detector" measuring time or is it the "crystal oscillator" that controls the clocking mechanism?

I know I said it was my last post prior to this, I just find it hard to refrain when I see facts obfuscated in support of an incorrect argument.

Last edited by a moderator: May 6, 2017
24. Jul 21, 2014

### Staff: Mentor

I would call focusing on the core physics of a clock "illuminating" rather than "obfuscated".

In a pendulum clock the core physics is the pendulum itself. The hands, face, winding mechanism, and housing are all necessary, but ancillary.

In an atomic clock the core physics is the hyperfine transition, which does not involve motion. The oscillators and detectors and so forth are all important components, but they are ancillary to the hyperfine transition.

Last edited: Jul 21, 2014
25. Jul 21, 2014

### Staff: Mentor

Yes. And as DaleSpam pointed out, that is the core physics of an atomic clock. The other stuff--the detector and the equipment that adjusts the flight of the cesium atoms--is all ancillary.

Wrong. That is "what a frequency is" for *some particular types of waves* (for example, vibrations of a string or sound waves in a medium). It is *not* a general definition of "what a frequency is" that applies to any type of wave. In particular, it doesn't apply to the "waves" corresponding to the photons emitted by the cesium atoms in an atomic clock. For those photons, "frequency" is just energy divided by Planck's constant (modulo a lot of technicalities involving what kind of wave function the actual photons have--are they energy eigenstates, coherent states, etc., etc.). It is *not* the periodicity in time of the "change of position" of anything. None of the references you have given have said anything different; they talk about "positions", yes, but they're the positions of detector elements; they're not the positions of "something that's changing its position as part of the oscillation of light".