Home Creators Posts Import Register Favorites Logout

Content

Hey! So I'm super excited for this video. It's about something I didn't know and have been excited to learn about. 

I am really interested to know what your feelings are - does the video make sense? Do you have unanswered questions? What are the things that pop into your head as you watch this? (maybe jot them down as you go) Other people will probably have the same questions so I'd love to answer them before they are even asked.

Thanks for your help and please keep this top sneaky - patrons only.

Files

Why no one has measured the speed of light

Top sneaky - tell no one but do let me know your comments on Patreon. Thank you!

Comments

Daniel Carroll

What if you stick with 2 clocks located in the same spot and move them away from each other at the same velocity for a predetermined amount of time... Could you measure the one way speed of light?

Kris Pockell

I see what you're doing with the Top Sneaky, Dirk.

Adam Berkan

I thought I knew special relativity, but I learned something today. Excellent video. I'll admit about 45 seconds in I was worried the catch was going to be something dumb, but it delivered.

Kenneth Hartman

Yay! Destin is in this video. I love your videos so much.

Veritasium

this was written into the script and I was waiting on an animation for it. I'll put this in.

Veritasium

I'm glad it delivered. That is what I want people to think coming away from this video. It's a big claim so I really have to back it up.

Stephen Connell

I think hiking up a hill would be a good analogy. It's faster hiking back down the hill.

Oscar Hesselberth

The explanation you give at 9:00 is in my opinion not easy to follows, I had to rewatch it a few times and I’m still not sure if I get it or not. While watching the video I was also wondering if this has implications that are here on earth rather than examples with extreme distances. Like measuring world records and if it would matter if the measurement was done in a different direction.

Veritasium

We are redoing the animation at 9:00 - hopefully that helps. Also thanks for raising that interesting point. I'll see if there's a good place to address it.

Jesse Thompson

Yeah, I'm not certain that it's helpful to try to "define" a 1-way speed of light, because it presumes a sense of simultaneity that special relativity doesn't properly support. Per SR, anything spacelike from an event can be viewed as "simultaneous" to it depending upon the relative velocities of the reference frames from which the measurements are taken. What I'm still waiting for is an empirical treatment of "distance" writ large in either relativity or quantum mechanics. Distance is important because space can only be said to be "curved" by dint of distances failing to line up with euclidean predictions, and because locality presupposes the maximum distance that the leading edge of a wave-function can propagate over a given duration of time (that said, duration is also nothing but a temporal distance). But every theory appears to presume distance and locality instead of rigorously defining it.

Stephen Connell

So you are saying we need the ability to communicate faster than the speed of light to calculate the speed of light. Warp speed?

Ken Schwarz

Fascinating. How would a hypothetical lopsided speed of light jive with our deep space observations? Wouldn’t one direction look young and the other old?

Matt Dunlap

Man, there's a lot to unpack here. I'm shocked by the idea that we haven't figured out how to measure the one way speed of light. I'm going to spend some time thinking about that one, it really seems like it ought to be possible. Even if we can't establish a one way speed, we can still rule out the case you talk most about: half speed one way and infinite speed the other. The experiment would be as follows: send one pulse through a beam splitter, then send one beam through a vacuum and the other through an optical fiber, and then measure the difference in the arrival times at the receiving end. If it's truly unlimited the pulse should arrive at the same time via both media. ... and I'm not sure why similar tricks can't be used to pin it down futher... congratulations, you've nerd sniped me. Edit: Nevermind, I've convinced myself that test doesn't work, I hadn't been thinking carefully enough about _why_ light slows down in various media.... still nerd sniped though.

Anonymous

Fascinating. The one scenario left unaddressed during the great animation at 12:00 is what if the slider keeps going further? Could a scenario of -10 minutes and +30 minutes be possible?

Kimberly Green

Can you set up an experiment using quantum entanglement? Create an entanglement, then separate the entangled particles by some significant known distance. Place a laser at the nearby particle. Fire the laser at the more distant particle to raise its energy level. And observe the resulting state change of the nearby particle. Wouldn't the time between firing the laser and the state change be the one-way time to travel a known distance?

Suramrit Singh

Really loved the video. The only part that bugs me is the "light-clock" at 7.37 mark. I could not understand what the context for that was and what the example with the trains was supposed to convey but it could be down to my own ignorance.

Peter Metzger

This was my thought as well - could both clocks start in the middle, be synchronized, and then move in opposite directions at the same speed until they reach the endpoints of the experimental range?

Joey Luther

Obviously not practical, but what if you take a measurement at the event horizon of a black hole? Would it be considered linear or round-trip, since the frame of reference for light would be linear (right?), although you would be seeing it travel away and hit you in the back as it orbits the EH

Ricky Lee

Paused right after the intro and do some personal stuff. Thinking about the title and feel like it seems to be related to relativity. And then Destin's reaction towards your question. Boom!

Matt Dunlap

Wait a second, if light could be different speeds in different directions, the event horizons of black holes wouldn't be spherical. Wouldn't that show up if we could image them well enough? Edit: and if it really was unlimited in one direction, that would be a naked singularity on one side...

Clayton Greenwell

Isn't using the term "speed" in it self false when thinking of light? I was under the impression from my extremely limited knowledge of physics that time does not exist as far as light is concerned. I thought as far as light is concerned it is just everywhere at the same time. So trying to use the term speed doesn't truly apply. Really we are just recording the measurement of our observation of the light? Or am I way off on this.

Scott Lowe

Great video! Something that comes to mind is your explanations with Earth and Mars communicating the time to each other is that they are modelled in the explanation as if they are static relative to each other (which they approximately are over the course of 20 minutes, of course). But since the bodies are in orbit, you could extend the experiment and repeat the clock synchronisation process lots of times at regular intervals from both Earth to Mars and from Mars to Earth over the course of 4 years. If the speed of light is anisotropic across the 2d plane within which Earth and Mars are orbiting, would you be able to tell? Would changes in the speed of light would cause a change in the interval between pings, in comparison to what is expected given the distance between the two celestial bodies and their relative velocities?

Myron Dietz

Well done. It is nice to have something that showcases critical thinking and not just contrarian thinking.

Candice Sheaffer

Great video. I never really ever thought about this topic that way.

Robert Blum

Kind of related to previous comments about accelerating in different directions, but is there really no bound to the one-way speed of light between c/2 and infinity? If the Mars astronaut goes back to earth with his clock, surely 10 minutes of difference between his clock and the bow tie NASA scientist 😃 could not all be explained by special relativity during his trip back? I know I’m missing some catch here, but wanted to double check. Great video, as usual!

marenoodles

The video does make sense, and all of the questions I had were ultimately addressed in the video. No negative feedback :)

Ricky Lee

Gavin and Den visited a lab that claims to be able to capture motions of light in their planet slowmo series. Is the lab also measuring the 2way speed of light?

Matt Dunlap

Here's a question, what would anisotropic lightspeeds mean for doppler shifts? If light was infinitely fast in one direction wouldn't it be immune to doppler shifts? Edit: Okay, so imagine a Very Large Clock on the moon. Wouldn't we expect to see a sinusoid in the rate of that clock as it orbits?

Jacek

Really nice video, pleasure to watch :-) Hmm, to add to the "if" comments: if light speed was significantly asymmetric, wouldn't we see it in the far space? Like the cosmos would look older on one side and young on the other? This, of course, assumes a super-simplistic scenario where it's just about some absolute "left" & "right" direction that's common to every observer in the universe, and of course we know such things are not likely to be the case :D

Norman Margolus

Very nice video! This is really just pointing out the arbitrariness of the idea that the laws of physics are the same in all inertial frames. The evolution of the universe might actually be computed in one particular frame, as we generally do in our simulations, but the laws have a symmetry that they look the same in all frames. A deep reason why this symmetry must be at least approximately true in our universe is that Darwinian evolution of complex organisms would be difficult if the laws of chemistry and biology changed depending on our state of motion. So the symmetry is important, but tells us nothing about how physics is actually being computed. This is indeed related to the problem of merging QM and GR, since how the computation is done tells us what microscopic time is.

Noelle Smiley

If we synchronised clocks together and then really slowly separated them, wouldn't that minimise the relativistic sync? Alternatively, might the half life of radioactive samples allow us to measure how out of sync our two clocks are?

Matt Dunlap

The same technique they use in that video appears in this video at 8:20 (although I think the clip from this video is from an MIT lab rather than the Cal Tech one)

C.J. Smith

I suspect this will be quite similar reasoning to what you're referring to in your response to Daniel Carroll, but what I'm currently working through in my head is... - Person A starts both a light signal and a simple motor-powered moving machine in the same direction towards person B. Person B measures time between the two arrivals. - Same thing the other direction: B sends light and identical motor object to A. The time-measuring devices are not moving, so no adjustments need to be made for time. Because the motors are identical, they "should" be moving at the same speed as each other, so if the light is moving at a different speed, the time-between-arrivals should be different at A and at B. Since apparently that is not the case, that means the speed of the motors would therefore be different, even though they are going through the same processes. I'm trying to think what has to change there - it seems since you can turn a ruler around that distance is constant (maybe?), which means I think time must be moving slower for the object moving in the direction of the faster light direction? I might have something wrong in there, but I have to go soon :P I'm also a little turned around as to what the root cause is there, and I'm not intimately familiar with the velocity-time relationship, but I guess that means the velocity-time relationship is dependent on only the speed of light in the same direction? I can think about this more, but you see my ponderings.

John Callahan

That was really good. Time went by quickly watching that! ;) While watching, I kept thinking of theory vs experiment. Are there any theoretical derivations of c? Do they match observations? Maxwell equations or Lorentz transformation? Or do all theoretical solutions of c require observations that a priori depend on c? I don't recall this idea mentioned in the video.

Veritasium

The tricky thing is that when did your local particle change? It won’t spontaneously indicate the change, you have to measure it. When you measure it, you change it. Either you measure too early or two late...

Veritasium

If objects are further apart - it all come down to that round trip delay distance

Martin S

The question I asked myself at the end was: "But what about the CMB?" (I.e. the cosmic microwave background radiation.) We measure essentially the same radiation in all directions, which is explained by it being emitted at pretty much the same time when the universe globally had cooled down enough to become transparent. If light is travelling faster in one direction that'd have to be compensated exactly by the universe becoming transparent later in that region, wouldn't it? And if that's the case these factors would only cancel each other out in one place in the whole universe, and we just happen to be right in that spot?

Kyle Nishioka

I remember a similar discussion / video about this area of metaphysics. The key question is not about the two-way measurement of light but what the true definition of simultaneous is in physics. All of the problems you outline go away if you can ever figure out a way to make two events synchronized perfectly. Then you setup two lasers and timers facing each other and fire them simultaneously.

Evan C Davies

Dr. Jason Lisle has been saying this for years. Maybe you could interview him on this topic.

Evan C Davies

You will likely disagree with him on some aspects of cosmology but many would enjoy the conversation whichever side they agree with more.

amir

the speed of light changes according to the medium it passes through and can have different speeds in different directions. for example, birefringent materials have 2 and sometimes 3 axes in which the speed is different. such materials are , for example, are liquid crystals, which are the core technology of most of the screens.

Adrian Byram

Thought provoking as always -- and extremely well-presented! Two questions / thoughts 1. Your space-time diagram nicely shows that the time taken for the signal from A to B and back to A has to be 2c (A's timeline), and what you are pointing out that 2c = c' + c" where c' and c" are the speeds A>B and B>A respectively. You claim that c' cannot be larger than 2c, but there's really no reason why not. E.g. c' = 3c and c" = -c. In other words, in one direction time is moving backwards compared to the other direction! Don't anti-particles do this in a Feynman diagram? 2. If c' and c" were different wouldn't this mess up E = m c^2 ? Which c would you use and why?

Kimberly Green

@Veritasium ... But isn't your concern one of observational precision and not strictly a problem with the experiment setup? If the particles are absolutely entangled and known to change state simultaneously, then measuring too early or late is something that can continue to be refined. But you'd still have a measure of the single-direction speed of light. And with greater precision of measuring the entangled state change, your speed precision would improve but not fundamentally vary. In fact, if the experiment were repeated many times and found to radically vary depending on place and distance between particles, that would strengthen your variable lightspeed conjecture.

Paul Looyen

Ok. Now your messing with my head. But I love it.

Rob Napier

I felt the video raised a lot more questions than it answered, for exactly this reason. It feels like there are a lot of observations that would at least put some strong boundaries on how asymmetric the speed can be. For example, if there's a consistent preferred direction, then synchronizing at one point in the year between Earth and Mars should lead to different offsets than a different point in the year. I should the be able to use two clocks at each site, synchronized several months apart, to detect a problem. Similarly, if I sent a clock to Mars, re-synchronized it, and then brought it back, I should at least be able to put some limits on the possible directional bias. Acceleration does impact time, but it's not arbitrary. We should be able to use observations to put limits on how much it can be over different directions. I'm absolutely certain that physicists have thought of all of this, and I'm missing the key points, but I spent the whole video asking these questions without feeling I got any closer to understanding.

^x3ro

Isn't the CMB moving "away" from us with (almost?) the speed of light? I mean isn't it radiation that was emitted from so war away that, due to the expansion of the universe, the thing emitting it was moving relative to us at (almost?) the speed of light? If so, then I think it also works because if the light travels 2C in one direction, then if we look in that direction, the stuff that emits the background radiation would travel away from us with 2C and its radiation would reach us instantaneously. So we won't notice a difference, right?

RayLast

I think he is referring to the same stream of photons and their potential speed direction preference relative to their origination point.

RayLast

Maybe both exist? E' = c'^2 and E" = c"^2?

Anonymous

Sorry, that's not quite how I meant it. The video illustrates the possibilty of the 20 minute duration not being split 50/50 send/receive split and potentially being an 80/20 or 100/0 split at a fixed distance. The question that popped into my head was whether there could also be a 150/-50 split. Basically negative time on the return trip, where it's still 20 minutes, but it actually takes 30 minutes to get there and the return trip takes a shortcut and takes negative 10 minutes to get home.

Dmitry Zinenko

This is true from the point of view of the photon itself. But for us time exists and the photon moves.

Jody Steele

Would it be possible to use the curvature of space time to "measure" the speed of light by measuring the angle a photon is altered by being shot past a known mass? I envision a setup where you have a photon emitter and detector some distance apart in a vacuum. You find out where the photon lands without the mass in place, position the mass, and measure the difference the mass has of the photon's position. My intuition is that a photon with a higher speed than we expect would curve less, but I will admit I only dabble in physics

Paul De Ronne

I've never studied physics so I apologize if I'm missing something obvious. I've always thought that when we look into space we see backwards in time. But if light was infinitely fast in one direction and c/2 in the other, wouldn't we notice differences (in star's brightness, or where gas clouds are concentrated, or something) if we looked in opposite directions? The line from the video that threw me was, under this assumption it's possible that "mars sees earth as it was 20mins ago, but Earth sees mars in real time." How do red and blue shifts fit into this? Wouldn't we not see shifts in a certain direction if the speed of light was instantaneous? Thanks in advance to anyone who takes the time to respond!

Dean Kyfiuk

Why could the one-way speed of light not be compared to a moving body of known velocity? Would the speeds of all moving bodies differ with direction of movement? Setup: Fire Destin's baseball cannon and a pulse of light simultaneously from point A. Observer at point B uses a stopwatch to measure the difference in arrival time between the light and the baseball. The stopwatch is only measuring delta t between the arrivals, so no synchronization with the launch is needed. If the speed of the baseball is known, the delta t measured would tell you the speed of light in that direction. Obviously this falls apart if the movement of macroscopic bodies are also affected by directional speed differences. Would the baseball somehow move more slowly in one direction than the other, like the light? If so, would it be worth mentioning this in the video?

Binal Bruno

As always an excellent video. Isn't the problem inherent in the definition itself. We define the speed of light in vacuum with the units that are defined by the speed of light. And according to our frame of reference, light itself never experience space or time. So I think the problem is inherent in our frame of reference. Like the axioms that we use to make math work. I am guessing we need to rethink or reimagine our frame of reference. Like the concept of time. We need to start with something that is unbiased by our concepts of time and distance.

RayLast

Some thoughts I had: I'm wondering what would need to be considered in the mental exercise and how it would apply to a single photon vs. a collection of photons and any probabilistic effects. Wouldn't a percentage of the photons prefer the faster (let's pick the instantaneous example) speed towards and the c/2 coming back and the rest the opposite? Or are we thinking there is an absolute direction where light in general, or any photon, "prefers?" Is it only relative to its origination point? If absolute, would we see areas of the sky where "instantaneous" light from "stacked" stars in a certain direction having different brightness? Would we see no or significantly varying amounts of redshift? I also thought about using perhaps a different medium to sync the clocks. Instead of splitting a wire and sending an electrical signal, say using (very) long rigid rods and sending a tap/vibration that travels at, even though it travels at the speed of sound, maybe equal lengths still means simultaneous? Maybe the noise or inaccuracies because of the slower speed doesn't tell you if back and forth are exactly equal but maybe you can eliminate the c/2 and instantaneous case?

Corey Gemme

on another note, I was able to post the above message using m iphone but was unable to post any more. when I tried to edit the original li had no "post" button . I’m typing this message on my mac. I’ll send this message to patreon too.

Veritasium

"Acceleration does impact time, but it's not arbitrary. " but it does depend on the one-way speed of light. So if you don't know the one way speed of light you can't actually put good bounds on how much it can be...

Veritasium

the claim is c' cannot be smaller than c/2. It can be as fast as you like - I assume for E=mc^2 you just use the round-trip average c, which is the one we use

Veritasium

these are great points and cosmological questions of isotropy might be the strongest circumstantial evidence on this point. But if the variation in c were a few percent, could we tell? Instantaneous is obviously the extreme example

Shawn Van Ness

Just move _both_ the clocks away from one another, symmetrically.. same distance, same acceleration/deceleration profile? It is a super interesting thought experiment, and I love these kinds of videos from you -- but I think this one needs more work up front to establish the premise as valid (or at least worthy of serious consideration) .. poor Destin is trying so hard not to ask what medication you're on, knowing he's being recorded. :P

Shawn Van Ness

+1 .. coming so close on the heels of your excellent (and also mind-melting) video about redshift and hubble expansion .. my mind also immediately went to wondering about the uniformity of cosmological redshift but it's not entirely uniform is it? we are orbiting the milky way, and our galaxy itself is moving something like 1,000 km/s toward the great attractor .. maybe there is no great attractor and really we're just seeing effects of slightly non-uniform 'c' in the photons coming from that direction? No idea if that holds water, empirically or analytically, but If that is a possible implication of asymmetry in 'c', it's probably worth calling out in the video

Shawn Van Ness

Or just setup the test near the equator .. oriented latitudinally, and run it once at noon and again at midnight? Someone from Singapore or Nairobi or etc should just do this, and let us know what they find out.

Kermideretvar

Nice topic and great video. What about this experiment: On earth you observe clock on satellite. You know the clock is ticking every second. Let's assume gravitational field of Earth to be perfectly spherical. Wouldn't this mean that as these two bodies rotate, the second pattern would widened and narrowed periodically? If the satellite makes one rotation every 90 minutes, after 45 minutes the transfer must be instantaneous while every 90minutes it must take distance/ (c/2) time. As the satellite approaches instantaneous direction, the ticks would be closer to each other.

Shawn Van Ness

I also began wondering if/how GPS satellites would work in a universe of non-uniform-c. A quick google for "gps speed of light" uncovers a rich field of people doing such analyses.. specifically trying to ascertain the uniformity of c in all directions.. and surprisingly it's not all debunked.. or is it? https://www.intechopen.com/books/new-approach-of-indoor-and-outdoor-localization-systems/gps-and-the-one-way-speed-of-light I don't understand yet how Michelson/Morley could have reached the conclusion they did, and how or why that doesn't square with what we're seeing from GPS signals .. Anyway, if this is not all hokum, it might be worth covering in the video.

l0b0

Would photons moving through a slower medium would be helpful? Then test once with the slower medium blocking the outgoing beam, and again with the medium blocking the incoming beam. I guess this would require that we could somehow determine that the speed multiplier in a medium is constant without knowing the actual speed of light.

JDG

Love it. Before saying the return speed of light is infinite, I'd add short section about why it's useful to talk about: Imagine you're on a car driving past a stationary wall at a speed of A. You throw a ball with a departing speed of B (your frame of reference). The ball will ricochet back at you at a speed of A+B---it's faster returning than going from your frame of reference. Of course, from the stationary frame of reference (the wall), the incoming and outgoing speeds would be the same. When talking about this idea for light, it'll simplify everything if we use the extreme case where we simply say the return trip speed is infinite.

Bryce Stoddart Esquire

as always great video, and so much fun watching Destine's brain almost melt as he is thinking about it.

Brendan J Scully

Great video, I've been loving the recent theme of relativity! All of my questions were answered by the end of the video, only things that stuck out to me were minor: - at 4:45 the two animations are out of sync for no apparent reason - at 5:21 the left end of the red bar jumps and covers the "0" line for no apparent reason

JDG

Also, you need a short description of what you mean at 8:35 with fiber optic cables. Some people may not realize that "round trips" refers to the light making many ricochets along the length cable before exiting the other side (assuming I understood correctly).

JDG

Woah, I'll keep that in mind. I just finished the second semester of physics in my math + physics degree, so everything is still kids' gloves. I'm trying to pay more attention to the assumptions of whatever thing I'm studying, but I find myself often just churning through the algebra, forgetting the underlying assumptions. Messing with those assumptions is often completely outside the scope of the course, but the assumptions are the foundation. Then again, maybe I'm getting ahead of myself and should just chill.

JDG

Okay, now you broke my head. That would be freaking fascinating. I'm studying relativity soon so this is something I'll try to remember to mess with.

JDG

Woah that's such an awesome idea. Although it's literally a round trip 100% of the way! 😂 I wonder if such an extreme circumstance would change things or if the results would still be relevant. They would be relevant, wouldn't they? As you said, from the light's perspective, its path would be linear . . .

Veritasium

fixed! those were temp graphics and thanks for spotting.

Veritasium

I spoke to an expert who told me there is a GPS issue related to the fact the frame of ref is rotating so there's actually a correction to account for this acceleration. That might be what people are incidentally measuring...

JDG

We could be in a region or frame that is conducive to life. I've recently heard the universe might be polar and exhibit varying physical timings depending on one's position with respect to the poles.

JDG

Thanks for mentioning Lisle. I've never heard of him (sounds like your right about disagreeing on cosmology), but his explanation is uniquely good, especially for someone not already knee deep in science videos. https://youtu.be/5vUel1AIM8w @Veritasium: I'm not sure what Lisle is referring to, but he mentions there are materials through which light travels differently in different directions. You might want to devote a short section to giving some examples of that as it would ground the entire discussion because, well, what if spacetime itself is a material with that property?

Shawn Van Ness

ps I don't recall if you mentioned Michelson/Morley in the video but probably should .. it was a watershed moment, that would have informed Einstein's thinking on special relativity .. and now it all kinda comes back full circle .. viz. could they have been wrong about the "aether" in 1887, but correct in their suspicion, more generally, that c is not isotropic? wow

Bob Van Valzah

When you were talking about clock sync and time dilation as a clock is moved, I wondered why the clock would need to move at relativistic speeds. My recollection from my modern physics class is that a clock had to move at more than 0.9C for a good long time before there was any significant movement relative to a stationary clock. Maybe say why two clocks in one spot could not be synchronized and then moved at classical speeds to be separated by a distance adequate to measure the time of flight to the required precision. Clock stability comes into play here, but I think cesium fountain clocks are getting about 1 part in 10 to the 12 these days.

Bob Van Valzah

Michelson/Morley entered my mind too. Agree it might be good to mention it quickly and dismiss it as looking for round trip differences at right angles. That would detect the luminous ether if it were there and the apparatus was moving through it, but I don't see how it'd detect any asymmetry in the legs of each round trip.

Shawn Van Ness

Watching again.. I hate to say but I think the whole Earth/Mars example confuses more than clarifies. 1) We just made a big deal about time-dilation effects trying to sync 2 nearby clocks here on earth.. but now we're ignoring the time-dilation effects due to differing rates of orbital motion, different gravitational field strength. Maybe throw in a quick hand-wavy disclaimer to put the viewers' mind at ease? 2) Imagine our astronaut stays on Mars for ~6 months .. now Mars is in a different position, different direction in the sky from to Earth .. but same distance. Does he still receive our messages at 12:10? Or did the time offset drift because now c is faster or slower? (again ignoring any cumulative time-dilation due to speed/gravity differential) 3) In the extreme case of c/2 one way.. if Earth and Mars are using a laser to communicate .. would the laser we aim at Mars even hit the target?

Veritasium

The Mars example is being reworked so hopefully that addresses some of your concerns. Regarding other concerns the difference between Mars and Earth tests was that Earth tests were only over a km whereas Mars is vastly further. We are only trying to sync the clocks for a day or a few days and our tolerance is minutes, not nanoseconds. The idea with Mars is just to get a more manageable scale for these effects,.

Shawn Van Ness

I feel like I'm learning a lot but boy this makes my head hurt. Mars is about 4200 miles wide, but orbits about 50,000 mph. If we did want to use a laser to communicate with Mars.. would we have to lead our shot? #WordProblemsFromHell More to the point .. could this be used to verify c is the same in both directions?

Norman Margolus

According to special relativity, our motion changes our notion of simultaneity, so wouldn't we just interpret a difference in the age of the universe in different directions as us moving very fast relative to the average motion of the stars? As for cosmic isotropy, isn't that again depending on the two-way speed of light?

JDG

Oh! Maybe at the end of the video, recommend folks look up the one way speed of light and dip into sections that seem interesting!

Wayne VanWeerthuizen

Since the speed of light can be calculated using Maxwell's equations and other measured constants, does this mean one or more of those other constants has a similar measurement problem?

Gordon Garmaise

When I took first year university physics I learned that one of the motivations for relativity arose from Michelson-Morley and subsequent experiments . The experiments demonstrated that light traveled at the same speed in every direction, which invalidated the theory of a "luniniferous aether". Those experiments must have taken the round trip element into account in order to persuade the physics community to abandon a long held hypothesis. Good enough for me.

Veritasium

those experiments showed the round trip speed was the same when comparing perpendicular round trips. Those experiments have no bearing on the one-way speed of light

Jonas

If light was travelling faster in one way than another, would the universe not look older in one direction than the other?

Jonas

How would it change our perception of the planets circulating our sun if light was travelling faster in one way than another? Would it appear that their speeds were defying their actual position wrt the sun due to them appearing to be at expected locations either too early/late?

Jonas

Also, great video as always, thanks a lot :)

Yann Vernier

We do have one way time measurement systems operating along a whole bunch of paths; for instance, while each GPS satellite runs its own clock, the receiver's determination of its position is reliant on the known relation of those clocks. That doesn't get us an accurate measurement of one-way C, but it gets us information about C in multiple different directions; specifically, I think it demonstrates that C matches isotropic speed for the orbits of these satellites. So if C is asymmetric *relative to gravity*, the same asymmetry would apply to matter moving too. Is this mistaken?

Stefan thyr woller

Fantastisk video.... 🤔🤔 A little scary to think about, as the first thing after you wake up

Greig Burges

Wow, that was thought provoking.

Daniel Wisehart

So why not put the clocks and light source on opposite ends of a bar, synchronize the clocks as if c is the same in both directions, then rotate the bar 180 degrees so the clocks and light source have swapped positions and measure the synchronization of the clocks again with the light source pointed in the opposite direction. If the light source was pointed north for the calibration step and the south to north travel time is c/2, when the bar is rotated so the light source is now pointed south where the travel time is instantaneous, the clocks should now say the light pulse arrived at the end before it left the start.

Michael Asaro

I wish you gave us a little more time to think about these questions before you plan to post the video since I also have to read through all the comments to make sure I'm not asking the same point as another poster. :-) That said, I did a quick search on the page and no one mentioned that the universe is inherently non-local and non-geometric. The fundamental theory's are being described in terms of quantum information (not that information can be anything but quantized). Sorry I don't have the time to look up the references I have in my head that I would need to actually ponder this line of thought, but if you stop thinking of the universe as a big geometric ball, and instead think of the universe as a large quantum computer, the speed of light then is the "clock speed" of that computer. Per your video, perhaps this would mean there is longer be a need to think about a "fixed" clock speed in the universe either - there is just that which is computed. It also helps to explain how entanglement allows for "spooky action at a distance". If it takes light time to travel, but then "information" can transferred instantly - you have to start thinking about differences between the photons themselves and the information they can give us about the structure of the universe.

Veritasium

or time would dilate differently for the two clocks during the rotation - a particular concern if the speed of light is anisotropic

Shawn Van Ness

I guess that's what I'm imagining when I think of GPS satelliites .. they are basically little atomic clocks in orbit, sending regular pulses of radio waves down to us. So it seems implausible that any directional variance could have gone noticed for 30+ years.. I'm still not buying into this premise at all .. Light moves about 1 ft per nanosecond .. per wikipedia the latest (US) GPS band signals are accurate to about 1 ft, and the (EU) Galileo system purports accuracy approaching 1 inch.

Sam Bebbington

Feel free to shut me down immediately as I don't have a great understanding of the universe... But: We use background radiation to look back in time in the universe, this is just another form of light that that is taking time to reach earth. If what we see in all directions appears to be similar, then at the very least we can prove light is not c/2 and instantaneous.

Shawn Van Ness

Answering my own question .. we regularly bounce lasers off reflectors left on the moon, which is about 2.5 sec roundtrip-light-time. The moon moves hundreds of meters in that time, along its orbit, relative to an observer on earth .. but those three-corner prism reflectors bounce the photons back precisely along the line they came in. There is no "leading the shot". :-/ I'm very slowly beginning to grok the problem statement .. and gaining a deeper intuition about special relativity in the process. (ps thanks Derek for continuing to produce such mind expanding content.)

Jerrad Pierce

Nevertheless, they deserved a mention and a more detailed explanation of why they may or may not be relevant than simple hand-waving to disregard them given that they were specifically devised and executed to determine whether or not the universe had a preferred direction of travel for light.

IVR

Hello, I just saw your video on youtube ( "Why no one has measured the speed of light") and thought of a possible solution so I came here. I hope you see this. In the video, you talked about using a stopwatch to measure how long it takes light to travel a certain distance and how difficult it is to do so. However, my question is why are you using a stopwatch? From my perspective, a better solution is to use two clocks that AREN’T sync and those two clocks would just count time as every regular clock does. You would then position Clock-1 at point A and Clock-2 at point B. When the light goes through at point A, Clock-1 records the time stamp and when the light goes through at point B, Clock-2 records this time stamp. Now you have both the starting time and the end time and you would use subtraction to determine the result which would be the time it took for light to reach those two points. Now that you have the time it takes light to travel one way, you can subtract it from the previous time you had when it traveled two ways and you will get your answer to see if the speed is constant when traveling both ways. P.S. I just turned 17 and I can’t help but feel naive proposing search a simple solution to a problem that has been debated for decades by some of the smartest thinkers including you. So I would be blissful if my proposal contributes to solving this problem. Also, a video response to this will be very interesting either if it explains why this proposal doesn’t work or if you do an experiment testing it (which I would love to see). Thanks

Kimberly Green

Naivety can be a strength. Don't apologize for it. There have been plenty of discoveries, and feats of engineering as well, by people your age. "Naivety" can bring a fresh perspective, new ways of looking at things, unconstrained by convention.

Kermideretvar

This looks like Wheeler's delayed choice experiment where the earlier measurement behaves according to final state. If I understand your setup right, the results would be the same no matter of space asymmetry.

Gordon Garmaise

So the speed of light would always be different on each leg of a return trip regardless of the direction? That hearkens to the anthropomorphic principle. The universe is such as it is because such a universe permits us to exist and observe it. I am skeptical of any hypothesis that posits a privileged place for humanity. A different speed of light in different directions would also break the principle of equivalence. Without the principle of equivalence we might as well study metaphysics as study physics.

Андрей Киреев

Just whatched this video. I think, I have an idea, how to check this isotropic property of light distortion, but we still need to make some assumption. Let us assume two thinks: 1) all the universe we see were initialized at the same time 2) at all the parts of the universe we see, the same physical laws applying Then we can measure the elements densities and if they are the same that means the difference of the speed of lights difference from two directions can be bounded significantly P.s. sorry for the bad grammar in this message (English is not my native language)

Dmitry Rudchenko

Can't you look at various directions in the sky and see what galaxies look like in all directions? Early universe looked quite different comparing to what it is now. If we're seeing Mars as it is now closer to one direction, won't we also not see early universe when looking very far? By knowing what we "expect to see" on average (stage of galaxy development, etc from the big bang) can't we do this?

Augusto Righetto

Imagine 6 identical laser devices capable of pulsing at identical frequencies at identical cadences. Attach each one of these laser devices to a face of a regular hexahedron. The lasers are pointing towards the regular hexahedron. Each face of the regular hexahedron have sensors detecting when the laser beans reach regular hexahedron. The regular hexahedron has an ultra precise single clock that individually registry when a laser beam reaches any of its faces. Launch these apparatus into space far enough from any significant gravitational body. A release mechanism is used to propel the 6 laser devices simultaneously with the same energy perpendicular to the hexahedron faces. The release of the laser devices should not affect the trajectory of the regular hexahedron since the lasers were propelled simultaneously with the same energy. If the sensors detect significant or increasing differences between registries, light does not follow the convention. Experiment has a limited lifetime since results would be invalid as soon as one of the laser devices get closer to a large gravitational body.

GottZ.de

well.. if you can sync two clocks with NTP.. shouldn't reverse NTP catch it? i'm not even kidding.. A remembers when it started the request, B replies with it's current time, A receives the time and knows how long A -> B -> A took. now if B would ask A for it's time the exact same way, it should be able to spot a offset. if you can run A -> B -> A multiple times and come to the same solution and then run B -> A -> B and find a difference, you found the proof AND know the difference between each direction.

IVR

Thanks for the kind words Kimberly. Have a nice day :)

G.Kendrick

Pump a pulse of light through a kilometer of fiber-optic cable?

BinkyNingy2

This might work. Use two different media (i.e. vacuum, glass). Create a pulse of light, split it, and one (P1) goes down the glass path to the mirror, and is reflected back through the vacuum. The other pulse (P2) goes down through the vacuum and back through the glass. Let’s say the round trip times for the two media are Tv and Tg, and the times from the source to the mirror are Tv1 and Tg1, and the times back are Tv2 and Tg2. Regardless, Tv = Tv1 + Tv2 and Tg = Tg1 + Tg2 (the round trip time is the sum of the travel time in each direction), and since the media is different, Tv != Tg (the speed of light is not the same for different media) Let Tp1 = round trip time for pulse 1, Tp2 is for pulse 2 Tp1 = Tg1 + Tv2 (glass on the way down, vacuum on the way back) Tp2 = Tv1 + Tg2 (vacuum on the way down, glass on the way back) If the pulses arrive at the same time, then Tp1 = Tp2, so Tg1 + Tv2 = Tv1 + Tg2 In the worst case of non-symmetry, Tv1 = Tv, Tv2 = 0, Tg1 = Tg, Tg2 = 0 So Tg + 0 = Tv + 0 But, we’ve already established that Tg != Tv, so this cannot be true. Of course, if the pulses arrive at different times, then we’ve proven the speed of light is different in each direction.

BinkyNingy2

this video has consumed half my day so far.... The whole concept of “right now” could be a great video (or two, or three…). For the first rabbit hole, you could then discuss why GPS is dependent on extremely accurate measurements of time delay, and how the GPS system estimates the changes in the speed of light caused by the Ionosphere, as well as the signals travelling from (near) vacuum to full atmospheric pressure, which changes the speed of light is changing during the entire trip. The next rabbit hole is how relativity causes GPS satellites need to periodically re-synchronize their clocks instead of just synchronizing them once. The Earth causes time dilation, so the clocks of the satellites travel through time at a different rate when compared to clocks on the ground. Then take a moment to discuss geodetic precession vs. frame dragging effect. Then raise that crazy to the power of crazy when discussing the trip of a signal from Earth to Mars. The time dilation caused by the mass of the Sun, the rotation of the sun, different orbital distances of both planets, the elliptical orbits of the planets, the mass of each planet, and the rotation of each planet, all affect the observed passage of time. The final brain-bender (for me, at least) is how useless the concept of right-now is. Using the space-travel experiment, where two people start on the ground, one travels in a spaceship some great distance and back at high speed. Both people start at the same “now”, and they both end up at the same “now”, but each has experienced a different amount of time between those two “now”. This really shows how “time” is actually another dimension. Just as two people can walk a different path between their house and the store, two people can travel different paths between two “now” locations in time. Actually, that’s not the final brain-bender. The final one is how the speed of light doesn’t change based on how fast the observer is moving relative to the light source, but the frequency of the light does change. After all that, sit back and enjoy the fact that you have not only destroyed the concept of “now”, but also completely decimated what it means to “observe”, and thrown the concept of “travel” into complete disarray.

Dr. Martin Borchert

But the cosmic microwave background looks the same - thus the same „age“ - from all directions. If it was instantaneous in one direction, wouldn’t the CMB be different in different direction ?