## How does a clock measure time?

 Quote by Gokul43201 Why can't you measure time by a direct comparison as well? What your doctor does with his hand on your pulse and his eye on a wristwatch is essentially how we all measure time - by a direct comparison to some other calibrated time interval. I think most any predator that actively hunts prey (not the kind that ambush) has a pretty good intuitive sense of time.
In the example above, one process is compared to another process - where does physical "time" enter the equation?

Recognitions:
Gold Member
Science Advisor
Staff Emeritus
 Quote by mangaroosh In the example above, one process is compared to another process - where does physical "time" enter the equation?
In what way is the time interval between two ticks of a clock different from the length interval between two ends of a ruler? (Other than that the first interval is measured along a dimension we call 'time' and the second is measured along a dimension we call 'length'.)

Mentor
 Quote by lisab Me too. In a lab, one measures length by comparing it to another (calibrated) length; it's a direct comparison. But to measure time, there's more involved (time = distance/rate).
That's only true if you use a mechanical clock! As I said above, you (and M_E) have the dependency backwards when it comes to the definition/most accurate ways of measuring: Time is counted, length is calculated from measuring time and converting.

I've seen an awful lot of people saying they just don't like time or feel uncomfortable with it, but I've never seen it be based on a logical reason. It's just a feeling and that's not good enough to really say there is a problem with the concept of time.

Mentor
 Quote by Mech_Engineer And of course time isn't constant- its measurement depends on your speed, further complicating things!
No, that doesn't further complicate it any more than it complicates measuring length (due to length contraction). You're misunderstanding the point of Relativity.

Mentor
 Quote by mangaroosh The use of a periodic cycle simply gives us a common unit of comparison, in which we can express information about a process. We can then compare other processes by expressing them in terms of this common unit. The periodic cycle exists, and the processes that are expressed in terms of the cycle exist, but physical "time" cannot be deduced from that.
Why not!? Length is measured using a proxy as well. There is no difference.
 "Length", much like "time", is just a concept; what happens when we use a ruler to measure the physical dimensions of an object is, in a simplistic example, we take a standard unit and hold it beside a physical object, and see how many of those standard units can be held beside the object. This allows us to express the physical dimensions of the object in a standard unit, which allows us to compare the physical dimensions of other objects expressed in the same units. That objects have spatial dimensions is self-evident. The measurement of the temporal dimension is not quite as straight forward - impossible if it doesn't exist - because the attempted temporal measurement of an object can only ever be carried out in the present; that is, the actual time co-ordinate will always be "now". We may of course remember a past state, and project a future state, but those are just mental constructs. The object only ever exists in the present.
That is completely wrong. Both distance and time are only measured in a single reference at a time, using some anchor to another reference. For distance, the obvious example is with a tape measure, where it is fixed at one end and read at the other. For longer distances, you can completely lose the starting reference and still have an accurate measurement (such as with the odometer in your car).

Here and "now" are the same concept.
 Quote by mangaroosh Whether it is estimated or not, how does the counting of periodic events allow us to deduce that there is a temporal dimension; bearing in mind that the memory of a past event is just a mental construct, and only events in the present can be said to be real, without assuming that past or future events are real?
That is an assertion/assumption without basis and one that is wrong for the same reason as the last. How, for example, can you say all the places you visited in your car, recorded by your odometer, are real? They only exist in the past for you, both in terms of time and space.

 Quote by mangaroosh The question is about what do we actually observe. Physical theories, make ontological claims about the nature of things like time and space; to have more accurate physical theories we need to see if those claims are justifiable. Indeed, our subconscious belief in time, just as our other subconscious beliefs, can affect our experience of reality, or more pointedly, how we live our lives. Again, it is the emboldened in the last paragraph that is being questioned.
LOL ontology is metaphysics, not science, and Relativity can describe time entirely by its relationship to other phenomena without any ontological explanation or reference to metaphysics whatsoever. Reality, existence, being, and the subconscious are for mystics and philosophers to debate. Science can state unequivocally that rainbows are optical illusions, but not whether they exist or are real in any ontological sense.

Mentor
 Quote by russ_watters That's only true if you use a mechanical clock! As I said above, you (and M_E) have the dependency backwards when it comes to the definition/most accurate ways of measuring: Time is counted, length is calculated from measuring time and converting.
In my lab, we measure properties just like I said in my previous post. Nearly all of the clocks we use are mechanical clocks. We use tape measures and metersticks to measure distance, most of the time. Sometimes a caliper. When measuring the depth of a glulam beam, we don't ever invoke the length of the path travelled by light in vacuum in 1 ⁄ 299,792,458 of a second.

I work in an engineering lab that tests building materials. It ain't rocket science.

 Blog Entries: 2 Recognitions: Gold Member Science Advisor Well Russ you can explain till you're blue in the face, I still find the absolute definition of time to be a bit of a wishy-washy subject That's not to say I can't utilize it in equations or understand HOW it can be utilized, I'm just saying it's got a certain smell about it... It's probably just a lack of fundamental understanding of where there are/aren't circular arguments. If I was asked to prove the accuracy of a CMM or a micrometer or even the flatness of an optical surface that's probably not a problem (given access to NIST references haha). But if someone asked me to prove the accuracy of an atomic clock, not sure where I would go other than to say "well it's got cesium in it, and this definition of a second says it should be right..."

Recognitions:
Gold Member
Homework Help
 Quote by mangaroosh The question is, as per the title of the thread, how exactly does a clock measure time?
Since we've established (I hope) that the flow of time is now defined by counting the periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium 133 atom in atomic clocks, perhaps one should ask, "Okay, what makes the atomic clock 'tick.' What drives the electron to regularly, seemingly change state in the first place?"

Without thinking, my first thought might be the evolution of the wavefunction, of course. But that leads to a sort of chicken-or-the-egg problem here. The wavefunction is modeled by the time-dependent Schrödinger equation. And the $t$ in the $i \hbar \frac{\partial \Psi}{\partial t}$ term is based on the outcome of the evolution. So that doesn't really help us much here in this particular case.

Instead, perhaps there's another approach to understanding this. Consider a very long hallway. In this hallway there are many, many clones/copies of yourself, all lined up one after the other. Each of these clones has a big box containing a stack of loose-leaf papers, in varying levels of organization. The first clone in the line represents some copy of your much younger self. That clone's box of papers is fairly well organized. The box belonging to the next clone in line is identical to the first's, except one of the papers has been moved out of order. As the line continues back, the corresponding stack of papers belonging to that clone are slightly more disorderly than the preceding clone's stack. This goes all the way back to the last clone in line, an older version of yourself who's stack of papers is fairly disorganized.

Now suppose you go up to one of these clones, and ask, "are you in the past, present or future?" The answer that you will invariably get is: "I am in the present." And it doesn't matter which clone you ask. They all think that they are in the present. Each clone thinks that it is he/she that is the one in the present, and all the others are either past or future.

If you haven't figured out my analogy yet, the line of clones represents a line on the "time" dimension of spacetime. And on this 4-dimensional (or 10 or 11 dimensional, whatever) chunk of spacetime, all versions of you equally exist. No version is more valid or less valid that any other. And each version is fooling himself/herself into thinking that he/she is the only version in the present. What's really the case is that each version is in its own present, and is stuck there.

Remember that stack of papers? Each stack of papers represents a particular quantum state of the universe. And each version of yourself is associated with one and only one quantum state. Each version of yourself is able to recall/look at notes/look at records, etc. regarding versions ahead of it, because those versions are almost identical to the clone in question, except they have stacks of paper that are more orderly, not less.

What I'm getting at is this: time, whether one considers it an illusion or not, is suspected to be related to entropy. And the arrow of time is always in the direction of increasing entropy.

-------

There's more to it than that still. Take an atomic clock, put it in a refrigerator and let it cool down. The entropy of the atomic clock as decreased, and yet it certainly does not tick backwards. As a matter of fact, it doesn't even slow down or change its rate at all. So there's more to it than the entropy of only the atomic clock itself.

At the heart of the atomic clock is that caesium 133 atom which contains an electron which emits radiation as it transitions states. And when a photon of radiation is detected by a detector, an electron in an atom in that detector becomes entangled with the original electron in thecaesium 133 atom. Almost immediately both of those electrons become entangled with other particles in the detector, and then the apparatus holding onto the detector and then with the all the atoms in the atomic clock, then room containing the atomic clock, then the Earth, and then throughout the universe. This is the process of decoherence: quantum state leaking out into the universe via quantum entanglement, becoming entangled with ever more and more particles.

Decoherence happens fast. And it makes the wavefunction appear to collapse [almost] instantly. And the wavefunction collapses simultaneously (whether treated as instant or "almost" instant) across all space.

Decoherence is akin to the second law of thermodynamics acting at its most basic level. And I speculate that resulting quantum entanglement between the caesium 133 atom and the rest of the universe has a large role in evolution of the wavefunction, and the eventual caesium 133 atom's changing of states again (even if that is the 'illusion' of time by talking to a different clone in that hallway analogy: moving to a different quantum state within spacetime).

(Tangent: Historically, this idea of instantaneous and simultaneous-across-space of the wavefunction collapse has caused many debates and experiments. According to Einstein's relativity, there is no such thing as absolute simultaneity. Events that are simultaneous in one frame of reference are not simultaneous in another frame. Since then the arguments have been mostly worked out by realizing (a) the wavefunction "collapse" should not be thought of as a classical event: it's not valid to describe it that way. And (b) any real events involving a given wavefunction collapse cannot be brought together for comparison faster than the speed of light. With those realizations in mind, special relativity and quantum mechanics are not in conflict. But it's these realizations that have given birth to various interpretations of quantum mechanics.)

Although putting an atomic clock in a refrigerator won't cause it to change its rate of time, relativity will. Put two atomic clocks in airplanes and let one fly around the world toward East, and the other West, and when they return they will show a difference. Put an atomic clock in a gravity well, and it will slow down.

That last point I find quite interesting. Consider a simple, non-rotating black hole. The volume of space contained within the event horizon of a black hole is at maximum entropy. One simply cannot put more entropy into a black hole without making it bigger (btw, the entropy of a black hole is proportional to the event horizon's surface area). Now consider carefully lowering an atomic clock such that it hovers just above the event horizon (assume you the observer are a safe distance away). Or just let the clock fall toward the black hole, whatever. When the atomic clock approaches the event horizon, its rate of 'ticking' approaches zero. Of course its slower rate of time can be explained with general relativity. But is there a connection with entropy and how general relativity affects entropy, and thus time, or is that just a coincidence?

I'm betting one would have to find more connections to quantum entropy, quantum entanglement, decoherence, and how they are affected by special and general relativity, to really understand how a clock measures time.

Further reading:
http://arxiv.org/abs/quant-ph/0203033
http://www.fqxi.org/data/essay-conte..._McGucke_7.pdf
http://fqxi.org/data/essay-contest-files/Kiefer_fqx.pdf
http://lmgtfy.com/?q=quantum +entropy+relativity#

 Quote by Gokul43201 In what way is the time interval between two ticks of a clock different from the length interval between two ends of a ruler? (Other than that the first interval is measured along a dimension we call 'time' and the second is measured along a dimension we call 'length'.)
The interval between events is always in the present though.

If we caricature the operations of a caesium clock: as a man sitting on a chair counting the events (or whatever the more specific term is) as they appear in front of him. One pops into view, he counts one, and it disappears again; then another one pops into view, he counts two, and it disappears again; and so on.

Here, the temporal "relation" between events is only an imagined one. The man observes them in a sequence, but each one ceases to exist after he counts it. To say that one event exists in the past is simply to imagine that it does.

Alternatively, if we imagine that each event is on a conveyor belt such that, as it comes into view he counts it and it continues along the conveyor belt (with or without disappearing, or ceasing to exist). Again, there is no past/present/future relationship, other than in the imagination of the counter. As an event occurs i.e. as he counts the object on the conveyor belt, it continues to exist in the present, but he retains the memory of it having passed him and designates is as being "in the past" - again, this is only imagined, because the object continues to exist in the present and the physical relationship between it and the other events is entirely spacial, not temporal. The temporal relationship is imagined, on the basis of his capacity for memory.

 Thankyou for that detailed response collinsmark: very insightful.

 Quote by mangaroosh I would say that time is merely a concept, with no physical existence ...
Time refers to changes in physical configurations.

 Quote by mangaroosh ... 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.
Right.

 Quote by mangaroosh 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.
A clock is a more or less periodically changing physical configuration. The underlying assumption is that any and all physical evolutions are ultimately due to some fundamental physical dynamic(s). So, counting the vibrations of a quartz crystal, or counting the revolutions of the earth around the sun, or counting the ticks of a mechanical wound up clock, or counting the full moons, etc., are all ways of indexing the evolution of our universe.

 Quote by mangaroosh 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?
Time is the changing, the evolution, of configurations, from the largest to the smallest scale. So, one can measure time on virtually any scale.

 Quote by mangaroosh 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?
It occurs in the detection of the discrete oscillations. Which are enumerated. And which provide an index for comparison.

Time isn't some mysterious dimension. It's just the changing configurations of various ponderable objects. We, or instruments which augment our senses, record physical configurations. The changing of those configurations is what we call the passing of time.

 To add to what ThomasT said, back in the ancient days the point of references were things like the position of the earth relative to the solar system and things like relativity to the stars. This was before all of the computers, grandfather clocks and all of that: a lot of it was based on understanding various periodic mechanisms that eventually gave us the things like the calenders of which you can find in many ancient documents from all across the globe.

 Quote by russ_watters Why not!? Length is measured using a proxy as well. There is no difference.
An explanation of how physical time, or a temporal dimension cannot be deduced from counting the events of a periodic cycle is proffered here.

The difference between the physical dimensions of an object - bear in mind "length" is just a concept - and the temporal dimensions, is that it is possible to hold that proxy beside the object being measured and "measure its length". The same cannot be said of time, unless we assume that time exists a priori; we can't hold a clock beside time and measure the physical property of "time".

 Quote by russ_watters That is completely wrong. Both distance and time are only measured in a single reference at a time, using some anchor to another reference. For distance, the obvious example is with a tape measure, where it is fixed at one end and read at the other. For longer distances, you can completely lose the starting reference and still have an accurate measurement (such as with the odometer in your car). Here and "now" are the same concept.
Apologies, I was a little narrow in my explanation; the explanation, as the one above represents a possibility that is distinctly lacking when it comes to time. With regard to the odometer of a car, it must be highlighted again that "distance", like "lenght" and "time" is just a concept. Indeed, what is measured by the odometer of a car is the amount of road in between the two locations. Again, the measurement device (the wheel) is in direct contact with the object of measurement; something which is not possible with time.

 Quote by russ_watters That is an assertion/assumption without basis and one that is wrong for the same reason as the last. How, for example, can you say all the places you visited in your car, recorded by your odometer, are real? They only exist in the past for you, both in terms of time and space.
It is not so much an assertion or an assumption as reasoning based on lack of evidence. Have you ever existed in a moment that wasn't the present? Bear in mind that the memories you have of "the past" are just mental constructs of the present moment, which has subsequently changed. Is there any evidence, that doesn't exist in the present, of either "the past" or "the future" - bear in mind that antiques or ancient buildings have always existed in the present, and will continue to do so.

It is safe enough to say that no experiment has ever been, or will never be, conducted outside of the present, for anyone. This means that there is no possible evidence of the continuing existence of past or future events, only present events. To say that they exist is to assume they do.

The difference between the places we visit in our car and events in the past and future, is that we can return to the places we visit in our car, but we can never visit past or future events. We can even set up webcams in the places we visit, or make phonecalls to people located there, but we canntot do the same with past or future events.

 Quote by wuliheron LOL ontology is metaphysics, not science, and Relativity can describe time entirely by its relationship to other phenomena without any ontological explanation or reference to metaphysics whatsoever. Reality, existence, being, and the subconscious are for mystics and philosophers to debate. Science can state unequivocally that rainbows are optical illusions, but not whether they exist or are real in any ontological sense.
No doubt it can, but things like the existence of spacetime and "the block universe" are ontological claims.

Mentor
 Quote by lisab In my lab, we measure properties just like I said in my previous post. Nearly all of the clocks we use are mechanical clocks. We use tape measures and metersticks to measure distance, most of the time. Sometimes a caliper. When measuring the depth of a glulam beam, we don't ever invoke the length of the path travelled by light in vacuum in 1 ⁄ 299,792,458 of a second. I work in an engineering lab that tests building materials. It ain't rocket science.
That's fine as long as you recognize that you are using relatively crude instruments that work differently than the official definitions of the units. I wouldn't complain that a sundial doesn't have a second hand!

I just find it so odd for otherwise scientific people to analyze a scientific concept according to feelings.

Mentor
 Quote by mangaroosh An explanation of how physical time, or a temporal dimension cannot be deduced from counting the events of a periodic cycle is proffered here.
That is completely wrong and your conveyor belt example shows exactly the error you made earlier with "here" and "now". Both the time and position of the starting point are lost on a conveyor belt. They are only kept track of.
 The difference between the physical dimensions of an object - bear in mind "length" is just a concept - and the temporal dimensions, is that it is possible to hold that proxy beside the object being measured and "measure its length".
Sometimes yes, sometimes no. Don't fool yourself into thinking that since sometimes you can see both ends of a ruler at the same time that it makes a difference. It doesn't. I'm sure you wouldn't concede the point for time, that both ends of an interval of time that is shorter than human perception happen "now", would you?
 The same cannot be said of time, unless we assume that time exists a priori; we can't hold a clock beside time and measure the physical property of "time".
Nonsense. The concept of time was developed because it was needed. It was not assumed to exist, it was observed to exist.
 Again, the measurement device (the wheel) is in direct contact with the object of measurement; something which is not possible with time.
What do you mean "again"? That's a new objection that has nothing to do with anything. There is no requirement that a distance measuring device be in contact with what it is measuring.
 It is not so much an assertion or an assumption as reasoning based on lack of evidence. Have you ever existed in a moment that wasn't the present?
Have you ever measured a distance from the wrong end of a tape measure?
 Bear in mind that the memories you have of "the past" are just mental constructs of the present moment, which has subsequently changed. Is there any evidence, that doesn't exist in the present, of either "the past" or "the future" - bear in mind that antiques or ancient buildings have always existed in the present, and will continue to do so.
That is pseudophilosophical nonsense. Again, if that were true, which it isn't, the fact that you are always here would mean distance exists only in your head as well. Both time and distance are measured by establishing one end point, then tracking, with an instrument, to another.
 It is safe enough to say that no experiment has ever been, or will never be, conducted outside of the present, for anyone. This means that there is no possible evidence of the continuing existence of past or future events, only present events. To say that they exist is to assume they do.
Nonsense - that's "not even wrong" (it would have to improve in order to merely be wrong). There is no claim nor requirement for an event that happened in the past to still exist. They did exist. They don't anymore.

Moreover, not only do experiments occur in the past, it is actually more correct to say they all occur in the past -- not to mention, they all occur there, not here! But that's true of both time and distance, so again an irrelevant concern.
 The difference between the places we visit in our car and events in the past and future, is that we can return to the places we visit in our car, but we can never visit past or future events.
Not necessarily true. I once visited the World Trade Center twin towers, but I can't again. Clearly, this shows us that "location" really needs 4 coordinates, not three.

Yes, there is a difference between distance in time in that time flows and distance does not. But who are you to call that a flaw? It's just a difference and when you oversimplify examples to try to make it into a flaw, you typically only show that time is real.
 We can even set up webcams in the places we visit, or make phonecalls to people located there, but we canntot do the same with past or future events.
Both webcams and phone calls record in the past and bring the recordings to you in the present. You're arguing against yourself.

 Thread Tools

 Similar Threads for: How does a clock measure time? Thread Forum Replies Special & General Relativity 55 Special & General Relativity 11 General Physics 3 Special & General Relativity 4 Special & General Relativity 22