36
$\begingroup$

Since the speed of light is the speed of causality, the "now" state of the a faraway object for an observer should be the exact point at which its light reaches the observer, for literally all intents and purposes.

My question is whether purporting that something "happened a long time ago but its light is only just reaching us" and that there is a more up-to-date state of an object is a scientifically useful viewpoint, or if it serves no other purpose than to deliver that pang of weird existential nostalgia for the layperson when we say "an alien on planet XYZ would look at earth and see dinosaurs roaming around"?

$\endgroup$
9
  • 13
    $\begingroup$ It's certainly useful if you want to predict the timing of the eclipses of Jupiter's moons (as seen from Earth), since that is how the finite speed of light was discovered. But I agree that "we're seeing it as it was 100,000 years ago" is misleading, since it kind of implies that more recent information is available somehow, and it's not. $\endgroup$
    – antlersoft
    Oct 28 at 5:03
  • 7
    $\begingroup$ It's important to know even if you don't mention it all the time. Some people don't know the basics of your field (no matter which one it is). They might ask whether Earth is the center of the universe since things are arranged in concentric shells around it. They might ask why farther away things seem younger. There are many people, even new programmers, who don't know that computers do one thing at a time, and there are professors who fail to explain that, because to them it is so obvious! $\endgroup$
    – user253751
    Oct 28 at 12:37
  • 2
    $\begingroup$ I mean, this information is valuable when we're looking at a faraway galaxy which is similar to our own. It lets us look at our past, or at least something similar to our past. How would you describe the Cosmic Microwave Background with your terminology? $\endgroup$ Oct 28 at 12:48
  • 4
    $\begingroup$ @user253751 actually, modern computers are capable of doing multiple things at a time, both via pipelining inside one core, and by having multiple cores. $\endgroup$ Oct 28 at 20:29
  • 2
    $\begingroup$ @Michael look up "lie-to-children." People who understand multiple cores or GPUs also do understand the meaning of "computers do one thing at a time" and when it does and does not apply. $\endgroup$
    – user253751
    Oct 29 at 11:29
27
$\begingroup$

You're right that astronomers don't really care what's going on in a some galaxy right now; we care about how they evolve through time, and how its light has been altered during its journey (e.g. redshift and extinction). We don't know how a particular star or galaxy has evolved since it emitted the light we observe, except in a statistical sense, and that "statistical sense" has been obtained through observing similar objects at various distances and hence times, and building physical models that explain the observed properties.

That is not to say that the time it has taken light to reach us is not a concept used by astronomers, at least in cosmology and extragalactic astronomy. It is called the lookback time, and is cosmologically linked to the distance from us, the redshift of the light reaching us, and of course also the age of the Universe at the time of emission. It is not uncommon to see plots in the literature showing some property of galaxies as a function of lookback time.

Because of these relationships, however, astronomers typically use neither distance nor time, but simply use redshift when talking about some distant galaxy or phenomenon. For less distant phenomena, such as stars inside our own galaxy, which are not cosmologically redshifted, we use parsec (or kiloparsec) which is ~3.26 times a lightyear.

That said, I think there is in fact a virtue in "pangs of weird existential nostalgia" when trying to enthrall laypeople. Firstly, I think it's easier to relate to the difference between "when humans started cultivating the land" and "when dinosaurs roamed around" than to relate to "12 thousand lightyears" and "a hundred million lightyears", which are both ridiculously far away. Secondly, I think teaching/reminding people that we always look back in time is a fascinating feature of physics.

Furthermore, there are some specific experiments involving short distances where the exact delay between emission and observation is important. Examples include predicting the eclipses of Jupiter's moons, as commented by @antlersoft, and communicating with rovers on Mars, as described in @Nuclear Hoagie's answer.

$\endgroup$
9
  • 1
    $\begingroup$ I would add that to a layperson, this observation is an important point to bring up? Yes to us at our current level of technological advancement there is no more readily observable state, but hypothetically there could be, no? $\endgroup$ Oct 28 at 19:36
  • 2
    $\begingroup$ @SchrodingersStat: No, you really can't observe distant objects any faster than the speed-of-light limit. That's because there is a Lorentz transformation which shrinks the travel time arbitrarily small. In other words, you can always find an inertial reference frame in which the travel time is as short as you would like it to be. If you could observe faster than the speed of light, then that reference frame would involve information traveling into the past, which is problematic for a whole host of reasons. $\endgroup$
    – Kevin
    Oct 28 at 19:56
  • 5
    $\begingroup$ @SchrodingersStat: Instantaneous teleportation is just as problematic, for exactly the same reasons (i.e. if you can travel from A to B, then you can ferry information from A to B on a physical medium). I agree that an observer at the point B would care about "the present," but it is important to realize that their present and our present are spacelike-separated and causally independent of each other. $\endgroup$
    – Kevin
    Oct 28 at 20:06
  • 1
    $\begingroup$ @Kevin My point is really that this is an important distinction to make, especially to a layperson. I don't feel it's intended to elicit a pang of existential nostalgia. It may not be a useful way to observe the universe, but it seems integrally important to describing observations of the universe, no? "It's not important for our purposes" seems like a bad reason to not discuss our observations in this context. $\endgroup$ Oct 28 at 20:09
  • 2
    $\begingroup$ @SchrodingersStat The speed of light/causality appears to be integral to the universe itself. You could make the point that the effect is symmetrical; that someone looking at us from Alpha Centauri 4 light years away would see the Earth as we were in 2017. However, to speak of a universe where instantaneous teleportation existed would be like an ocean documentary speaking of water that isn't wet. You would only do so if you were making a point, like as a counter-factual; to show how teleportation violates all sorts of fundamental laws of physics. $\endgroup$
    – Schwern
    Oct 28 at 22:14
15
$\begingroup$

I agree that for very distant objects which are completely out of humanity's reach, the time delay of how we see the rest of the universe may not have much practical impact. But for nearer objects, there are meaningful implications for space exploration. A Mars rover, for example, cannot be driven in real-time like a remote controlled car, since light takes several minutes to travel between Earth and Mars. Anyone on Earth watching a "live" feed from the rover is actually seeing footage from several minutes ago. We have no possible way of seeing the rover as it currently is, we can only see the rover as it was in the past.

This has significant implications for space probe design and autonomy, since distant probes may need to make "decisions" in absence of human input. This is especially important for time-sensitive, dynamic situations such as landing. By the time anyone on Earth has any inkling that something has gone wrong, it may already be too late to fix the problem.

The notion that we can never see or interact with a distant object in real-time is particularly meaningful when you want to engage in two-way communication, rather than just passively observing. Watching a "live" broadcast with a delay isn't very different from watching it in real-time, but trying to hold a conversation with a delay fundamentally changes the conversation itself.

$\endgroup$
5
  • 4
    $\begingroup$ It was particular relevant during the "seven minutes of terror". We can't switch to manual control because of the delay. $\endgroup$
    – Barmar
    Oct 28 at 14:25
  • $\begingroup$ A remote controlled car on Earth can also not be driven in "real time" :^) $\endgroup$
    – Tvde1
    Oct 29 at 7:23
  • $\begingroup$ @Tvde1 this hints at the underlying problem. "Commonsense" teaches us to treat the speed of light as infinite. In human terms, 300 km/ms means we can remotely control a car on Earth, but it might get a bit dodgy from the other side of the planet. (I'm ignoring internet or similar latency and jitter which isn't directly a speed-of-light issue). $\endgroup$
    – nigel222
    Oct 29 at 9:50
  • 1
    $\begingroup$ @nigel222 Even if you were on the opposite side of the planet, light would only take 0.13 seconds to make a round trip, which is very close to the human limit of perceiving things as "instantaneous". Not truly "real time", but the difference from true instantaneous response is almost imperceptible. $\endgroup$ Oct 29 at 15:03
  • 2
    $\begingroup$ @Nuclear_Hoagie 0.13 seconds on top of 0.5(?) second human reaction time is "a bit dodgy" in my book. A car going at 100km/h travels 3.6m in 0.13s. $\endgroup$
    – nigel222
    Oct 29 at 15:15
7
$\begingroup$

It's useful for cosmology, which studies how the universe has changed over time. When we observe extremely distant galaxies, we're seeing them in the state they were in during the extreme past. By observing a number of galaxies at about the same distance, we can construct a decent picture of that time in the history of the universe. Then when we combine observations over different time frames, we can develop theories that explain the processes that lead to the changes we see.

It can also be useful in relation to the search for extraterrestrial intelligence (SETI). If we manage to detect a signal from a distant star, it would have come from a civilization that lived long ago, and might already be extinct. Or if we imagine intelligences out there trying to detect and communicate with us, they would have to be within about 50 light years to detect anything we've transmitted so far by and for us to receive a response. Even if there are intelligences out there, it's very unlikely that they're that close and also at comparable levels of technology with us. So while SETI might allow us to detect intelligences, the time delays mean that communication is virtually impossible.

$\endgroup$
4
$\begingroup$

Imagine that we see light from a star 1000 light-year away that has gone supernova and we calculate that atomic particles from the explosion will be traveling at 0.9c (I made this speed up, I don't know the actual speed).

We know (based on the light) that the explosion happened 1000 years ago, so we can calculate that the particles have traveled 900 light-years and will reach us in about 111 years.

I think it would be confusing in this case to say that the star exploded when we saw the light. Saying it exploded "now" would imply (to me at least) that the particles just started on their journey. Alternatively, if we want to say it exploded "now", then we would also have to say that the particles appeared 100 light-years away from us "just now".

$\endgroup$
3
  • 1
    $\begingroup$ Minor note: $0.9c$ is too fast for supernova ejecta, but $0.1c$ is not unheard of. $\endgroup$
    – pela
    Oct 29 at 9:31
  • 1
    $\begingroup$ @pela Do you count neutrinos as particles? ;-) $\endgroup$ Oct 29 at 15:20
  • $\begingroup$ @Peter-ReinstateMonica No. Neutrinos are stupid, I don't count them at all… $\endgroup$
    – pela
    Oct 30 at 10:21
4
$\begingroup$

The universe is believed, in the broadest possible terms, to be the same everywhere (isotropic homogeneous); that there's no real physical difference between the Milky Way and the furthest galaxies visible in the Hubble Deep Field images.

But we observe very real differences: because less time passed between the beginning of the universe and the emission of the light we see from those galaxies, they have different rates of star formation and galaxy collisions than we do now, 12 billion years later.

Aliens in those galaxies might look at the primordial Milky Way and come to the same conclusions about the early universe. So this is a very real effect.

$\endgroup$
1
  • 1
    $\begingroup$ Good answer, but note that "the same everywhere" is homogeneous. This is not the same as isotropic which means "the same in all directions". Our Universe is believed to be both (at least on the scales we can observe), but in fact neither of the two implies the other. $\endgroup$
    – pela
    Oct 30 at 10:20
2
$\begingroup$

Yes, there is a purpose to that, and that is any time when there can be interaction between two objects that are rather far away from each other, because it shows a hard limit on what kind of interaction is possible.

A light-speed round-trip to the ISS is basically instant. That means, real-time communication with the ISS is no issue.

A light-speed round-trip to the moon is ~2.5 seconds. That means, you will notice the delay, but with a bit of patience, almost real-time communication is possible.

Mars has a light-speed round-trip duration of 6-44 minutes. That means, real-time communication is impossible, but you can communicate at "email-speed". You can easily communicate multiple times a day. You can reasonably manually control a robot/rover that way, even though it will be moving quite slowly.

A light-speed round-trip to Neptune is around 8 hours. That means, at max three messages a day. Manually controlling a robot/rover is pretty much out of the question. It will need to be either quite autonomous or incredibly slow.

A light-speed round-trip to Alpha Centauri is almost 9 years. That means, if we somehow manage to get people there, there won't be any possibility to direct or govern them from Earth. Whoever goes there might be able to send a message 10 times in their life time if they are lucky.

A light-speed round-trip to the Galactic Center would be around 50 000 years. That means, even if we get a probe to go there at light-speed, humanity will most probably be extinct before we get a response.

And that's pretty much the limit to our theoretical sphere of influence. If we could theoretically manage to fly something to the farthest objects we know of (e.g. the galaxy GN-z11, which is 13.39 billion light years away), they might possibly be gone by that time. Or actually, they might be gone already.

$\endgroup$
3
  • $\begingroup$ Homo sapiens emerged about 300,000 years ago. You think the species might only have another 50,000 years left in it? I'd have thought this highly adaptive organism is particularly virulent and hard to wipe out. Luckily, it can only spread at sub-light speed... ;-) $\endgroup$ Oct 29 at 3:02
  • $\begingroup$ Judging how long humanity will still be around might be difficult, you are right. But I think it would be fair to say, that in 50k years there will be neither compatible technology, nor any knowledge about that probe at the galactic center. So when the signal returns, it would be most probably not decodable. $\endgroup$
    – Dakkaron
    Oct 29 at 8:32
  • $\begingroup$ Great answer, but minor quibble: While GN-z11 emitted the light we see today 13.4 billion years ago, its distance is now 32 billion lightyears due to the expansion of the Universe. $\endgroup$
    – pela
    Oct 29 at 10:36
2
$\begingroup$

"Since the speed of light is the speed of causality, the "now" state of the a faraway object for an observer should be the exact point at which its light reaches the observer, for literally all intents and purposes."

I'm guessing you mean that the "now" state of a faraway object for an observer should be the state at the exact time at which its light now reaching the observer set off.

But I'm unclear on whether you mean for 'the "now" state' to just be a way to label the latest possible state relevant for causality at the observer's "now", or whether you are extending the definition of "now" such that the faraway object's "now" is the same "now" as the observer's, and thus 'the "now" state' just means 'the faraway object's state right now'.

If the former, this is just a matter of non-standard terminology, and a matter of choice. Science aims to model the universe, and can do so with equal accuracy in many different languages. Whether a particular language choice is 'useful' is more a matter of psychology, not science. If the latter, you've got problems with trying to extend a singular "now" as a region across spacetime in a consistent way.

If you use this definition, your concept of "now" is not transitive or reflexive. Light travels from event A to event B and then in a different direction from event B to event C. B sees A happen "now", but A does not see B happen "now". And C sees B happen "now", while B sees A happen "now", but C does not consider A to be happening "now".

If you select one particular observer, you can simply divide the universe up into the past lightcones of the points along its worldline, and call each one a "now". It's (part of) a perfectly valid coodinate system. But each such "now" contains many pairs of events that are not causally related to one another. You can't say that states may be causally related if they are 'in the same "now"'.

The concept of events being null-separated is certainly useful, and so is that of one event being on the past lightcone of another. It may be worth coining a word for the concept. But using the word "now" to do this is probably not a good idea. It fits badly with the equivalence-relation-like intuition we have of events happening "at the same time" from Newtonian spacetime, and it doesn't even fit well with the way causality works. Although Newtonian gravity posited instantaneous action at a distance, Newton himself rejected the idea as philosophically incoherent. Causality is asymmetric with respect to time, so you can't use a symmetric relation like "at the same time" to describe it.

$\endgroup$
-4
$\begingroup$

General relativity is the same as special relativity at small distance scales, so for the purposes of this question, which is about a star only one light year away, it's perfectly OK to use special relativity. The answer would be different if the distance were cosmological.

Relative velocities of stars within our galaxy are small compared to c. Therefore there is a natural frame of reference to employ, which is the rest frame of the galaxy in general. Once this frame has been fixed, special relativity defines an unambiguous notion of simultaneity. By that definition of simultaneity, the event happened one year ago.

There is no frame of reference in which the event happens now. This is because special relativity doesn't allow frames of reference that move at the speed of light.

$\endgroup$
1
  • 4
    $\begingroup$ You criticise our site's answers from "people who don't actually know much about relativity" but then assert that "General relativity is the same as special relativity at small distance scales", ironically proving your point. Special relativity is restricted to Minkowski space, i.e. flat spacetime unaffected by significant gravitational effects. The distance doesn't have to be "cosmological": spacetime near massive local objects is curved. For example, GR explains Mercury's anomalous perihelion shift, and GPS wouldn't work without GR's time dilation in Earth's gravitational field. $\endgroup$ Oct 29 at 2:37

Not the answer you're looking for? Browse other questions tagged or ask your own question.