# From the Perspective of Light

1. Mar 9, 2013

### jalsck

I am trying to gain some understanding of what SR 'tells us' about the universe. The following statements are made from the imaginary perspective that light would have. Are they correct?

1. A light wave observes no time passing for any matter.

2. The light wave is absorbed or is at the 'edge' of the universe instantaneously. A secondary question is if the light wave is at the edge of the universe instantaneously, what happens to it there?

3. When switching from our perspective to light's perspective, the universe has collapsed into 2 dimensions. The plane is perpendicular to the path that we would observe the light to be travelling in.

2. Mar 9, 2013

### ghwellsjr

If you want to understand what SR 'tells us' about the imaginary perspective that light would have, you will have to do what we do when we want to switch from our perspective to any other perspective moving at any arbitrary speed with respect to us, and that is we use the Lorentz Transform. In that process, there is a factor called gamma (γ) which evaluates to 1/0 when we use c as the value for v. So if γ=1/0 that means that we have to find a number for gamma which when multiplied by zero equals one. Until you can 'tell us' the answer to that problem, SR cannot 'tell us' anything about the imaginary perspective that light would have.

3. Mar 9, 2013

### ZapperZ

Staff Emeritus
What you are doing here is transforming to the "rest frame of a photon", which is something you cannot do. Read the FAQ subforum in the Relativity forum.

https://www.physicsforums.com/forumdisplay.php?f=210 [Broken]

Zz.

Last edited by a moderator: May 6, 2017
4. Mar 9, 2013

### jalsck

Thanks George and Zz.

5. Mar 9, 2013

### michael879

I really hate those kinds of responses... You guys clearly know what he means, but you're choosing to take him too literally!

While you can't transform to the rest frame of a photon, you can look at the limit as v->c. In this limit a particle with infinitesimal mass would be very similar to light, and observe the things you said, which can be condensed into the following:
It would see its creation and annihilation (absorption) nearly instantaneously

In this sense it would see the rest of the universe as "frozen", and it would see its final state an infinitesimal distance away from its initial state.

6. Mar 9, 2013

### danR

This should be pinned, and every newcomer made/allowed to read it before they do anything else.

7. Mar 9, 2013

### Bill_K

michael879, We drew the distinction because there is a distinction. Your claim that they are the same, and we are therefore being unfair to the OP, is simply wrong.

A particle with nonzero (even infinitesimal) mass has proper time - a light ray does not. A particle with nonzero mass would see the same elapsed proper time between its creation and annihilation, no matter how fast it may be traveling. The prototype example is the ultrarelativistic muon that makes it through the atmosphere before decaying. From the "perspective of the muon", the decay takes exactly the same time it does at rest, namely 2 μsec.

Last edited: Mar 9, 2013
8. Mar 9, 2013

### michael879

Um, you're wrong about your second point.. a particle created at A, travelling near light speed to be destroyed at B, will see the distance between A and B shrink to 0 and as a result would not experience any time because it is created and destroyed almost simultaneously.

And also, the differences between a massless particle and a massive particle shrinks to nothing as the mass of the particle shrinks to 0. So while you can't talk about light's reference frame, you can take the limit as a massive particle becomes massless traveling at c.

I assume the OP is looking for an answer besides "it just can't be done", as 99% of the people asking this question are. So if you're going to take the time to answer them, why not provide a good answer? Notice I did specify that you couldn't transform to light speed, and I was talking about a limit in my post.

9. Mar 9, 2013

### Bill_K

We disagree.

10. Mar 9, 2013

### 1977ub

If you go chasing after a photon (let's say small periods of acceleration followed by checking out the universe from your new IRF) to find out what the universe is like from the perspective of a photon, the photon you're chasing continues to recede at c. So unfair. All other photons continue to move at c. The rest of the universe indeed contracts and slows down. So while you see all those effects in the universe, have you really approached the experience of "moving with a photon" ?

11. Mar 9, 2013

### ZapperZ

Staff Emeritus
Unfortunately, when replying to question such as this, one HAS to be literal. It is my experience in dealing with this type of question for the many years that I've been on here is that often time, when we try to be "too smart", often what we try to convey is NOT the message that the other person receive or understand. This is because, unlike you and I, the person who asked this often does NOT have the mathematical description in the back of his/her mind, and thus, able to decipher what the limit to c means! More often than not, your description of the speed approaching to c is typically lost, and all that sunk in is that, yes, the universe has shrunk when we are moving at c.

Don't believe me? Spend some time browsing through the thousands of threads in this forum and see how many instances occurred when what is written is not what is understood by the other party.

So we end up trying to be as direct and as clear as we can, without beating around the bush. Helen Quinn wrote a terrific article in trying to make sure our communication to the public is as direct and free of ambiguity.

Zz.

12. Mar 9, 2013

### Fredrik

Staff Emeritus
While we do understand that people who ask these questions have that limit in mind, most of us don't approve of the idea that this limit defines a photon's perspective. I don't consider such a definition meaningful, so like the others, I would have said that there's no such thing and linked to the FAQ. We get this question a lot, so it doesn't make sense to write a unique answer each time.

One of the reasons why we get that question a lot is that Brian Greene apparently disagrees with us and likes to think of photons that way.

13. Mar 9, 2013

### Ben Niehoff

The limit of a Lorentz transformation as the boost parameter $\beta \equiv \frac{v}{c}$ goes to 1 does not exist, so you cannot talk about the "limit of small mass" as approximating a massless particle in any way.

Another way you can argue this is to observe that light waves oscillate with some given frequency along their path. If there were any frame in which that path were traversed instantaneously, then there could be no oscillations. Therefore thinking of light as a "limit of small mass" is inappropriate.

14. Mar 9, 2013

### michael879

I would argue yes. Look at the example of 2 infinitesimal masses approaching the speed of light both in random directions. Unless the angle between the trajectories is in a very small region (which becomes infinitesimal as v->c), both masses will observe the other traveling at near-lightspeed! So in essence, I would say photons "see" other photons traveling at c (in the appropriate limit). This is consistent with the SR postulate, and any other result would be in violation.

fair enough. It just really frustrates me when I see stuff similar to this (this might be a bad example because its such a popular example).

*same comment to ZapperZ* I get it, you guys post on here constantly these questions must get tiring. Its just that I've been on his side of things (back when I was in high school), on these same forums and it used to frustrate me to no end to have people respond with null answers like those.

15. Mar 9, 2013

### michael879

Bill, I don't really see anything to disagree about here, this is basic SR... Use your example of cosmic muons: The higher the energy of the muon, the longer its lifetime appears to us. From it's point of view though, the explanation for this discrepancy is that the distance between its creation and destruction has shrunk.

16. Mar 9, 2013

### Staff: Mentor

How can a null question have anything other than a null answer?

17. Mar 9, 2013

### Staff: Mentor

As long as we're going to keep pointing newcomers who ask the "rest frame of light" question at the FAQ, this is a fair argument for further expanding the FAQ.

18. Mar 11, 2013

### ghwellsjr

Um, you're wrong, michael, everything Bill_K said is exactly right and everything you said is exactly wrong. The Proper Time for the particle between events A and B does not depend on any frame or any convention or anything else. The Coordinate Time between events A and B can change depending on the frame but it will never get shorter than the Proper Time, only longer. Furthermore, the particle already sees the distance between A and B equal to zero, there's no shrinking involved. The Coordinate Distance between events A and B is zero in the particle's rest frame, it only gets bigger in other frames.
There's always a huge difference between a massless particle and a massive particle, no matter how small the massive particle gets, the spacetime interval between the events describing its creation and annihilation (assuming it's inertial as we have been in this case) will always remain timelike while the spacetime interval for a photon is always lightlike.
He asked for the correct answer and that's what I gave him. And it was a good answer and so was Zz's and the OP thanked us for them.
Talking about taking the limit is the wrong answer because it's just as wrong as thinking that you can transform to light speed.

I'm sure my statements haven't persuaded you so let me try another approach. Consider the two events A and B describing the creation and annihilation of a muon. Let's assume the muon survives for 2 microseconds. We'll use units of microseconds for time and light-microseconds for distance and we'll put them in the format of [t,x]. We'll assume the muon was created at [0,0] and was annihilated at [2,0] in its rest frame. Now we want to increase the distance for the annihilation of a hypothetical particle and see how the type of the spacetime interval changes.

First we'll note that as long as the t parameter is greater than the x parameter, the spacetime interval is lightlike. If it is smaller, then it's spacelike and if the t and x parameters are equal, it's lightlike. We'll also note that a timelike spacetime interval can be measured by an inertial clock that is present at both events A and B and a spacelike spacetime interval can be measured with an inertial ruler present simultaneously (as defined by SR) at both events. A lightlike spacetime interval cannot be measured by anything inertial.

So as we increase the x parameter from zero and approach 2, the spacetime interval can always be measured by a clock, no matter how close we get to 2. But when we get to 2, the spacetime interval abruptly changes from being timelike to becoming lightlike. We could also start with the x parameter equal to 4 and decrease it as we approach 2. In this case, the spacetime interval is spacelike and could be measured with a ruler as long as we never hit 2, but the instant we hit 2, it can no longer be measured with a ruler as it becomes lightlike.

Now the question is: should we call the lightlike spacetime interval "zero microseconds" (because we approached it from the timelike side using a clock to measure it) or "zero light-microseconds" (because we approached it from the spacelike side using a ruler to measure it)? As a matter of fact, it is called a null interval because it is inherently different that either a timelike spacetime interval or a spacelike spacetime interval. We can neither measure it with either a clock or a ruler, neither can we characterize as either timelike or spacelike.

19. Mar 12, 2013

### jalsck

I've got a LOT of study to do before I ask anymore questions so this isn't a question . I'm interested in transforming from 3 spatial dimensions plus time to 2 spatial dimensions plus time. I've done some googling but didn't come up with what I was looking for. It's just a curiosity and I know that I need to understand SR and GR in depth first. Genuine thanks to everyone that has contributed to this thread it is truly appreciated.

20. Mar 12, 2013

### Fredrik

Staff Emeritus
That sort of transformation is never done in SR or GR. Maybe there are situations in which 2+1-dimensional GR is interesting, but then you simply choose to use it instead of standard GR.