Because light can only travel so fast, all of the light we see in the sky was emitted at a previous moment in time. So if for example we see a supernova or some other great stellar event, by the time we see it, it maybe long over. That made me kind of curious, what is the most ancient light we can see from earth?

The universe is supposedly ~13+ billion years old, but we are probably not at the very edge of the known universe so all the light we see is probably less than 13 billion years old. So what is the oldest light we can see? and as an optional follow-up question how do we know the age of that light?

I guess the light itself may not actually be literally 'old', but its probably obvious what I'm asking here, put another way: what's the longest distance that now earth visible light emitted has traveled to reach the earth? Though that reformation of the question gets kind of tangled with lensing effects.

  • 24
    $\begingroup$ "but we are probably not at the very edge of the known universe" - We're in the exact middle of the visible universe, since we can see back to the emission of the CMB in all directions. $\endgroup$
    – JollyJoker
    Commented Apr 10, 2018 at 11:15
  • 10
    $\begingroup$ @JollyJoker But isn't everything in the exact "middle?" $\endgroup$ Commented Apr 10, 2018 at 15:47
  • 2
    $\begingroup$ There's no edge. $\endgroup$ Commented Apr 10, 2018 at 15:54
  • 11
    $\begingroup$ @DonBranson There may be no edge. We have no way of telling. There's little reason to assume there is an edge, but there's also little reason to assume there is no edge. Knowing the limits of our knowledge is important. $\endgroup$
    – Luaan
    Commented Apr 11, 2018 at 8:15
  • 6
    $\begingroup$ @PhilNDeBlanc Every place is in the middle of its own observable universe, yes $\endgroup$
    – JollyJoker
    Commented Apr 11, 2018 at 9:01

5 Answers 5


The oldest light in the universe is the cosmic microwave background. Roughly 380,000 years after the Big Bang, protons and electrons "recombined"1 into hydrogen atoms. Before this, any photons scattered off the free electrons in the plasma filling space, and the universe was essentially opaque to light. Once recombination occurred, however, photons were able to "decouple" from the electrons and move through space unimpeded. This relic radiation is still observable today; it has been redshifted and cooled.

We can detect light from very distant objects, and we have. It makes more sense to talk about distance in terms of redshift; the larger the redshift, the farther away an object is. There are a number of extremely high-redshift objects, some of which have had their measurements confirmed, and others of which have not. Candidates include

All of these objects would have formed some hundreds of millions of years after the Big Bang, however, so the light we see from them is much "younger" than that of the cosmic microwave background.

1 I've never liked the usage in this context, as this was the first time they combined; the "re" is kind of misleading.

  • $\begingroup$ I'd argue that "recombined" in this context is not "kind of misleading" but instead downright wrong. But that's not your fault. $\endgroup$ Commented Apr 12, 2018 at 9:57
  • $\begingroup$ I had to read a lot about redshift to finally find the point where they talk about the expansion of space - I guess that is what you are specifically referring to? $\endgroup$
    – Arsenal
    Commented Apr 12, 2018 at 15:33
  • 3
    $\begingroup$ I'm not sure where the idea comes from that the epoch of recombination is the "first time" the hydrogen has been neutral. Hydrogen ionization was in "ionization equilibrium" prior to that time, and its state of being neutral got "frozen in" during that epoch. What all that means is the timescale for hydrogen to ionize went from being less than the age of the universe at the time to be being more than the age of the universe at an age of about 400,000 years. It also means each electron had been captured and released by protons many many times prior to that. So yes, it's "re"-combination. $\endgroup$
    – Ken G
    Commented Apr 29, 2018 at 9:10

What's the oldest light that we can see?

The Cosmic Microwave Background is considered to be the oldest E-M radiation detectable to us. It's in the microwave spectrum, so it can't be seen with the naked eye but is picked-up by "radio telescopes". We call it "light" in the broad sense.

One remarkable aspect about this background radiation is its near-uniformity in all directions. Astronomers reason that the uniformity is too strong for the source to be a really big thing like a huge balloon... but that would be the case if it was all actually as far apart as it seems to be.

If it were really as big as it looks, it would take twice the age of the universe for one side to be affected by the other side! Instead, astronomers believe that what we see was a very small body, which has become bigger; that's why it looks the same in each direction. Some of the growth is called metric expansion of space and has a different meaning than ordinary growth.

How do we know the age of that light?

The age of the cosmic background light can only be determined indirectly, first by knowing how long ago the Big Bang happened, then by figuring when the light was emitted in the course of the Big Bang.

By comparing the rate at which everything seems to be getting bigger with how big everything seems to be, in the same way that you might estimate how long it would take to drive to a place given the speed of the road and the distance, we calculate the Hubble Constant. This helps us calculate how long ago the Big Bang happened.

Also, there are certain "sound waves" (baryonic acoustic oscillations) where old things we see, including the cosmic microwave background, get brighter and dimmer with a rhythm, like a clock's pendulum. They can be measured either left-right (for moving things) or by monitoring a video (for stationary things). Measuring these rhythms and comparing them to the Hubble Constant also helps to calculate how long ago the Big Bang happened.

Finally, the microwave background has physical qualities (like temperature and density) that make it possible for us to determine when it was emitted during the expansion and cooling of the Big Bang. Together using all these calculations is how we figure the age of the cosmic microwave background light.

Astronomers believe that this combined calculation (called "LCDM", "Lambda-CDM", or "Big Bang Cosmology") is very good because the different numbers do line up, for the most part*. They were pleased to report more good findings as recently as 2018 when a study called the Dark Energy Survey finished. Nevertheless, since LCDM includes certain assumptions that may never be validated, and since there are still some unexplained discrepancies, we don't know whether another kind of calculation would be better, provided that it still fits the measurements.

How do we know that this is the oldest light?

It is only by thinking about the physical qualities of the cosmic microwave background, and thinking about when during the Big Bang it must have emitted its light, that astronomers identified it as the oldest possible light in the universe, older than any stars or galaxies. It doesn't tell us how old it is by itself; in fact, astronomers are always making sure that it's not in fact just a layer of dust on the telescope!

How far away is the cosmic microwave background?

This is a really hard question to answer. According to Big Bang Cosmology, the cosmic microwave background wasn't "somewhere" but instead it was everywhere. And the distance it has traveled since the Big Bang is different than the time multiplied by light speed, because of metric expansion of space. This is a result of the relativistic length-contraction due to the speed at which everything is moving.

Is the observable universe younger than the greater universe, assuming that that exists?

Calculating the amount of time from the Big Bang to now gives the same result whether you consider our observable universe or the greater universe that may exist. That's why the age of "our" universe is the same as the age of "the" universe.

*Some different studies to determine the Hubble constant have given cosmologists pause (link 1, link 2); depending on which part of the universe you look, it may be close to 67 or it may be closer to 73 in the standard units.

  • $\begingroup$ This combined calculation, the ΛCDM, is built on several assumptions that are impossible to verify, but which astronomers agree to be sensible. One assumption is the "cosmological principle": every place in the universe is going to be pretty much the same as every other place. By this principle, the thought (for instance) that groups of galaxies closer to the Earth are all closer to each other than they are in the far corners of the observable universe, is ruled-out. $\endgroup$ Commented Apr 12, 2018 at 18:36
  • $\begingroup$ The cosmological principle is actively used today to interpret plain findings in astronomy that seem to show that the farther you look, the lighter and tighter all the galaxies are. $\endgroup$ Commented Apr 12, 2018 at 18:50
  • $\begingroup$ Without assuming the cosmological principle, researchers may "come up with alternate cosmologies" that compete against ΛCDM for acceptance with other astronomers. One such alternate is the "White Hole Cosmology" that was proposed by Russell Humphreys in his book, "Starlight & Time". $\endgroup$ Commented Apr 12, 2018 at 20:42
  • $\begingroup$ Your third paragraph seems to suggest that the CMB originated during cosmic inflation. That is incorrect. Inflation occurred within the first fraction of a second after the Big Bang; the CMB was created 380,000 years later. $\endgroup$ Commented Apr 12, 2018 at 22:00
  • 1
    $\begingroup$ You’re still conflating the metric expansion of space (observed and agreed) with inflation (theorised, unobserved, disputed). I’m happy to upvote if you remove the unnecessary reference to inflation. $\endgroup$ Commented Apr 12, 2018 at 22:44

Scientists have discovered a galaxy, named GN-z11 (already mentioned by HDE 226868), which existed a mere 400 million years after the Big Bang, or about 13.3 billion years ago:

Farthest Galaxy Yet Smashes Cosmic Distance Record

The discovery of a 10 billion year old star was announced just last week:

Hubble spots farthest star ever seen

Here is a list of distant astronomical objects on Wikipedia.

  • 1
    $\begingroup$ <shields_up>. The 13.7 B years universe age is based on a tail chasing definition which now does not "allow" the universe to be older than current received truth says it is. So BB-400 cm years if the 400 m years is in fact accurate :-). <shields_still_up> $\endgroup$ Commented Apr 11, 2018 at 9:52

You've made two questions using semantics:

  • "How old is the oldest light visible from Earth?"

From @Pela's answer to: Why is there a difference between the cosmic event horizon and the age of the universe? - So in ~100M years the most distant light will reach us, from over 116M light years away.

The 16 Gly that the distance to the event horizon is today is sort of a coincidence. It has nothing to do with the age of the Universe. It only depends on the future expansion of the Universe, which in turn depends on the densities of the components of the Universe (Ωb, ΩDM, ΩΛ, etc.). If the Universe has been dominated by matter (or radiation), then there would be no event horizon: No galaxy, ever-so far away would not be visible to us, if we just had the patience to wait. A galaxy is 10,000 billion lightyears away? Just wait long enough (exactly how long depends on the actual density).

However, our Universe happens to be dominated by dark energy, which accelerates the expansion without boundaries. This unfortunately means that the light leaving today from a galaxy 17 Gly away will be carried away by the expansion faster than it can travel toward us. In contrast, the light emitted today from a galaxy 15 Gly away will travel in our direction, but will nonetheless initially move away from us due to the expansion. However, its journey toward us makes this expansion rate smaller and smaller (since the expansion rate increases with distance from us), and after a period of time it will have traveled so far that it has overcome expansion and starts decreasing its distance from us and eventually reach us after 100 Gyr or so.

  • "I guess the light itself may not actually be literally 'old', but its probably obvious what I'm asking here, put another way: what's the longest distance that now earth visible light emitted has traveled to reach the earth? Though that reformation of the question gets kind of tangled with lensing effects?"

Yes, that's a totally different question ...

See one of the earliest papers: "Expanding Confusion: Common Misconceptions of Cosmological Horizons and the Superluminal Expansion of the Universe" by Davis and Lineweaver (2003).

Newer works:

"The Shared Causal Pasts and Futures of Cosmological Events", by Friedman, Kaiser, and Gallicchio (2013).

"Conclusion: ... While the observable spatial densities of galaxies, clusters, and thus quasars are thought to reflect correlations set up during inflation, it remains an open question whether inflationary era events at specific comoving locations — where quasar host galaxies later formed — could yield an observable correlation signal between pairs of eventual quasar emission events at those same comoving locations billions of years after the inflationary density perturbations were imprinted.

In closing, we note that all of our conclusions are based on the assumption that the expansion history of our observable universe, at least since the end of inflation, may be accurately described by canonical general relativity and a simply-connected, non-compact FLRW metric. These assumptions are consistent with the latest empirical search for non-trivial topology, which found no observable signals of compact topology for fundamental domains up to the size of the surface of last scattering.

Fig. 1

FiG. 1. Conformal diagram showing comoving distance, $R_0χ$ in Glyr, versus conformal time, $R_0τ /c$ in Gyr, for the case in which events A and B appear on opposite sides of the sky as seen from Earth (α = 180°). The observer sits at Earth at $χ = 0$ at the present conformal time $τ = τ0$. Light is emitted from A at ($χA, τA$) and from B at ($χB, τB$); both signals reach the Earth along our past lightcone at ($0,_{τ0}$). The past-directed lightcones from the emission events (red and blue for A and B, respectively) intersect at ($χAB, τAB$) and overlap for $0 < τ < τAB$ (purple region). For redshifts $z_A = 1$ and $z_B = 3$ and a flat ΛCDM cosmology with parameters given in Eq. (11), the events are located at comoving distances $R_{0χA} = 11.11$ Glyr and $R_{0χB} = 21.25$ Glyr, with emission at conformal times $R_{0τA}/c = 35.09$ Gyr and $R_{0τB}/c = 24.95$ Gyr. The past lightcones intersect at event AB at $R_{0χAB} = 10.14$ Glyr at time $R_{0τAB}/c = 13.84$ Gyr, while the present time is $R_{0τ0}/c = 46.20$ Gyr. Also shown are the cosmic event horizon (line separating yellow and gray regions) and the future-directed lightcones from events A and B (thin dashed lines) and from the origin (0,0) (thick dashed lines). In a ΛCDM cosmology like ours, events in the yellow region outside our current past lightcone are space-like separated from us today but will be observable in the future, while events in the gray region outside the event horizon are space-like separated from observers on Earth forever. Additional scales show redshift (top horizontal axis) and time as measured by the scale factor, $a(τ)$, and by proper time, $t$, (right vertical axis) as measured by an observer at rest at a fixed comoving location.

Also see: "Causal horizons in a bouncing universe", by Bhattacharya, Bari, and Chakraborty (2017):

"Conclusion: The present work shows that the causality problem in bouncing universe is intrinsically related to an understanding of the various phases of the universe during the contraction phase. As our understanding of the contraction phase is purely speculative at present the models we use to figure out the nature of particle horizon remains over simplistic. The present authors believe that although the causality problem in bouncing universe models are far from being solved the present article shows the qualitative and quantitative difficulties one must have to circumvent in the future to produce more meaningful results.".

Short answer: It's 46.9B light years. Another Wikipedia page says: 46.6B light years. The experts above calculate 46.2.


This April 2, 2018, CNN article says:

Scientists detect 'fingerprint' of first light ever in the universe"

Following the Big Bang, physicists believe there was only darkness in the universe for about 180 million years, a period known by scientists as Cosmic “Dark Ages.”

So I think your answer might be that Big Bang + 180 million years is the oldest light we can see.

  • 1
    $\begingroup$ Why not link to the Nature news article CNN cited? $\endgroup$
    – Mike G
    Commented Apr 10, 2018 at 15:19
  • 3
    $\begingroup$ I'm a bit colorblind, between that and the display on this computer I did not see the link. I had to go back and mouse over things to see it pop out. $\endgroup$
    – CrossRoads
    Commented Apr 10, 2018 at 15:23
  • $\begingroup$ @MikeG Isn't it more righteous to mention the source that first introduced the info to you ? $\endgroup$ Commented Apr 11, 2018 at 22:13
  • 3
    $\begingroup$ Sorry, but this answer is not good. First of all, 180 Myr is much later than the CMB discussed in the answer of HDE 226868. Second, the observations you describe is not light, it's the lack of light, i.e. absorption (in fact it's an absorption feature in the CMB). $\endgroup$
    – pela
    Commented Apr 12, 2018 at 15:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .