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

Reference frame of light

  1. Feb 28, 2007 #1
    I've seen several responses (and links) to the question "What happens when something is traveling at the speed of light" and understand that the answer is twofold. First no object can travel that fast because it would require an infinite amount of energy to acelerate physical objects to that speed and second, light itself does not experience time or space.

    The thing I can't get my head around is that second part. I mean we can see, even at our macroscopic scales, that light very definitely IS affected by both time and space, no matter your relative observation point and yet I keep seeing this assertion that light doesn't experience either.

    To my concept of logic, this makes absolutely no sense. Put a lens in the path of a light beam and it bends, period. How can these two things be rectified? One a plain-as-day observation with our own eyes and the other a statement that basically says: No. What you're seeing does not in fact happen because light doesn't experience space or time.

    I understand that our senses can't always be completely trusted (optical illusions etc) but these effects (time delay through differing materials and refraction) are also measurable. It's not just our senses that are telling us that light slows or bends but every tool we use to measure these effects also indicate that these effects are real.

    Since our built in senses and artificial senses all agree on the fact that space and time exert real effects on light, how can it be that light "from it's frame" doesn't experience either.

    Last edited: Feb 28, 2007
  2. jcsd
  3. Feb 28, 2007 #2


    User Avatar

    Staff: Mentor

    Light bends in a lens according to you. What does the light think is happening?
    You are measuring the time in your reference frame, not the light's reference frame.
  4. Feb 28, 2007 #3
    christ... Just take the lorentz transforms & put in c for v. What do you get?
  5. Mar 1, 2007 #4
    To mind-meld with a photon for a minute...

    I'm zipping along headed directly for a square with the letter "x" on it. I know I'm heading toward the square because I analyzed the absorption spectrum of the light that reaches me and determined that it is blue shifted. All of a sudden, I enter a lens and the next thing I know, I'm headed toward a circle with a "y" on it's surface. From these observations, I would be forced to conclude that my direction has changed.
  6. Mar 2, 2007 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    Do you have experimental evidence that a photon can distinguish between an "x" and a "y"?:rolleyes:

    Or, perhaps more to the point, how can a photon "know" it is traveling toward anything when it does not "experience" time.
  7. Mar 2, 2007 #6


    User Avatar

    Staff: Mentor

    How does that follow? Even setting aside the time dilation issue, light on a reciprocal path must be on a reciprocal path: that photon won't ever think it is headed toward X - it will always think Y is directly in front of it.
  8. Mar 2, 2007 #7
    A photon no more 'knows' where it's headed than the middle of the atlantic 'knows' that its waves are breaking on the shores in sunny florida. I believe this is the interpretation of quantum field theory - a photon is the manifestation of the whole field at a certain point. An event happens which causes a quantised vibration of the electromagnetic field. This disturbance will translate to a different point in spacetime according to quantum probability. There is nothing in between.
  9. Mar 2, 2007 #8


    User Avatar
    Gold Member

    "...zipping along headed directly for..." This perception would require an experience of the passage of time would it not?

    "I analyzed the absorption spectrum of the light that reaches me and determined that it is blue shifted."
    And this would require some sort of perceptive ability that has nothing to do with a photon.

    "headed directly for a square with the letter "x" on it. "
    How does the photon know what it was headed toward?

    If a photon can experience anything at all, it experiences it as a photograph - unmoving. It's whole life consists of:

    "Well, isn't it a find moment to be alive. One of my ends has a flashlight attached to it, and the other of my ends has a 'y'."
    Last edited: Mar 2, 2007
  10. Mar 2, 2007 #9


    User Avatar
    Gold Member

    DaveC, I like your photon.

    If the flashlight and the 'y' were in relative motion, would the photon use this
    information to shift frequency ?

    Ktoz should look at Huygens principle and Feynman's 'QED'.
  11. Mar 2, 2007 #10
    Yes but it wouldn't fit in this margin :)
  12. Mar 2, 2007 #11
    My analogy clearly exploded in a ball of fire, but you lost me with the term "reciprocal path." What's that? And how does it relate to the smoldering shards of the analogy?
  13. Mar 3, 2007 #12


    User Avatar
    Gold Member

    OK, that's too funny to let slip by.
  14. Mar 3, 2007 #13


    User Avatar
    Gold Member

    He means that the light beam could just as easily go from "y" to the flashlight - no diff. What's why I said "flashlight at one end, 'y' at other". Beginning and end ae reversible.
  15. Apr 9, 2008 #14
    Light just travels it's path, instantaneously, things don't "happen to it" because they all happen to it at the same time, and then, nothing... it's only to you (or another non-light observer) that it gets diffracted over time... though it's an interesting idea of superposition, it's not really surprising given the lorentz factor, given that we are trying to solve for 1/0, for which no value really exists...
  16. Apr 9, 2008 #15
    Zeroes have kept things interesting for a long time. :smile:


  17. Apr 9, 2008 #16
    I always wonder about people who say this.

    Asymptotically, 1/0 is infinity. Which, given the nature of infinity, makes 1/0 infinity.

    Think of it like this. Pick any number you like and add 1 to it. Take the sum and add 1 to it again, repeat. Do you reach infinity? No. The reason for this is that there is no number for with x+1=infinity, except infinity, in which case you are already there.

    You can keep adding 1, getting larger and larger numbers and never getting any close to infinity.

    Try instead, starting with a number above 1 and squaring it. Then square the result. Repeat. You get larger and larger numbers, but never ever get to infinity. You don't even approach infinity "faster" than just adding 1s to the number you first thought of. You never get there, so you take the same amount of time to get to infinity with both methods ... forever.

    Now look at 1/0. Sure it is difficult to think about and computers used to throw in the claw if you divided anything by zero. But 1/(a very very very small number) = (a very very very big number) and as (a very very very small number) approaches zero, 1/(a very very very small number) approaches infinity.

    Personally I think that saying that 1/0 is undefined is just another way of saying that you can't actually reach infinity (even if you can manipulate transfinite numbers, if you felt like doing that) so 1/0 is as undefined as infinity is.

    If I travelled 1 metre in 0 seconds, I am pretty damn sure that I would be travelling infinitely fast, not undefinedly fast :smile:

    Of course in my own "rest" frame, I would be telling you that I was stationary the whole time, and that the rest of the universe was moving so fast that the ruler that moved past me shrank to zero length and that's why I appeared to cover 1 metre in 0 seconds. The rest of the universe only had to travel at light speed to acheive that - which I suppose is what the OP's photon thinks.


  18. Apr 10, 2008 #17
    sorry, but mathmatically, it is still undefined, consider the inverse operation, what is infinity times 0? certainly not one, or any other number that would be divided by 0... also...
    Does light view itself as infinitely dense? (because of the lorentz factor?)
  19. Apr 11, 2008 #18
    This is something I've never wrapped my head around: light doesn't experience time, but somehow in my reference frame light interacts with other things *in a timely fashion*. The two stories don't seem compatible.
  20. Apr 11, 2008 #19
    Amusingly enough I was thinking about this this very morning. 0/0 is indeed undefined, because you can't approach it asymptotically. The result of an attempted asymptotic approach will either stay locked on "almost infinity", stay locked on "almost zero" or sit on 1 the whole way - or it could swing wildly from "almost infinity" to "almost zero", depending on the approach you make.

    Think of a very very very small number divided by itself (the same very very very small number). This is going to be unity.

    Then a very very very small number, divided by a very very very very very very small number (exaggerating here to make my point). The result will be a very very very big number. Keep making the number you are dividing by smaller, until it approaches zero, and you keep approaching a result of infinity. This is what I mean by locked to "almost infinity".

    Say you now think that your two numbers are not equally close to zero, so you start making the numerator smaller. Your result will get smaller and smaller, reaching 1 eventually and, as you keep making the numerator smaller, the result will then approach zero. (If you started out with a much smaller numerator, and then made it smaller and smaller, then the result would be locked to "almost zero".)

    This 0 divided to 0 is clearly a slipperly character, not wanting to be tied to 0, 1 or infinity, so yes, 0/0 is undefined. I would lean towards saying that 0*infinity is 0, but that is probably because I give precendence to 0, since it is a more reasonable number.

    If you wanted to argue better, you could point out an unstated assumption that I originally made which, when brought to light and addressed properly, invalidates my argument about 1/0 being infinity - mathematically. Note that this does not invalidate my argument physically,

    In it's own rest frame, light has no reason to consider itself to be infinitely dense. It probably would consider the rest of the universe as being infinitely dense though (assuming it could consider anything, of course).


  21. Apr 11, 2008 #20


    User Avatar
    Gold Member

    Pretend you are in a service elevator in a building under construction. You get in the elevator and ascend to the top (it is a one way elevator).

    As you go up through level after level of floors, you realize that there are loooong strands of yarn tied to pillars from floor to floor at all sorts of angles. The first piece is tied to the the 2nd floor pillar right next to the elevator. The other end of that same piece is tied waaaaay up on the 6th floor to a pillar out on the north east corner of the building.

    You can only experience one floor at a time, so you only see sections of yarn that are horizontally in your line of sight. What you see is that the line of yarn (that is, only the short section you can see on the floor you're on) starts very close to you (right by the elevator) and "moves" away from you, ending up, four floors later, way out at the north east corner. When you were on the first floor there was no yarn to be seen, and when you passed the sixth floor, there was no yarn to be seen.

    You have experienced this bit of yarn as an apparent movement through the building's space as a function of your travel through the floors (and, incidentally, through time). It was "emitted" on the second floor, and "absorbed" on the sixth floor.

    All the while, the string has had no experience of "moving" from floor to floor, no experience of emission or absorption - or of any "time" whatsoever.
    Last edited: Apr 11, 2008
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?