I love it when a seemingly simple question—one so simple that it feels silly to even ask it—leads to profound, even cosmic, insight.
For example: Why is the sky dark at night?
I can imagine you reading this and thinking, “Seriously? That’s profound?”
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Yes. Yes, it is. I can also imagine you thinking, “But the answer’s obvious! There are only so many stars to go around. Ones farther away are fainter. So there’s lot of black between them and nothing out there that they would illuminate. Of course the sky is dark at night.”
But that answer is wrong or, at best, incomplete, and the actual situation wasn’t obvious even to eminent scientists a century or so ago. And the person who first solved this mystery—or, at least, the first to give a correct scientific answer in writing—is almost certainly not who you’d think!
To be clear, the sky isn’t perfectly dark at night. Earth’s atmosphere glows faintly even from the darkest site. Here we’re talking cosmologically dark, however: the universe itself is in a state of being unlit. There are some sources of background light even then—distant stars and galaxies too small and faint to be seen individually—yet the sky looks pretty black compared with, say, the surface of the sun.
That may seem like an extreme comparison, but it’s not. That’s the heart of the problem.
Historically, astronomers held to the idea that the universe was infinite in time and space, stretching on forever. It always had been, and it always would be. It was static, unchanging.
There was reason to believe this, especially if you were an astronomer in, say, the 19th century. You’d think that the Milky Way was the whole universe and that, by Ptolemy’s ghost, it didn’t change from night to night. Oh, certainly the moon underwent phases, you’d say; each planet, true to the Greek etymology of that word, wandered the sky, and so on. But the stars were always there and always had been.
This idea leads to a problem, though: if the universe is infinite in space, with stars evenly distributed throughout, then the sky should be bright—very bright, based on its geometry.
Let’s say you count up all the stars in a thin spherical shell that’s centered on Earth, one with walls a light-year thick and 1,000 light-years away. If you look at another shell that’s twice the distance (2,000 light-years) away and the same thickness, the math dictates that it will have four times the volume and therefore four times the number of stars in it. The volume of a thin shell like this increases as the square of its distance.
It’s farther away, however, and that means the stars will be fainter. That follows the inverse square law: the brightness drops as the square of the distance. So a star twice as distant will be one quarter as bright.
That means any shell you pick will seem to be as bright as any other, no matter its distance. Near or far, the increasing volume and decreasing brightness exactly cancel out.
Okay, that’s fine. But remember, in our 19th-century thinking, the universe is infinite—that means there are an infinite number of shells and an infinite number of stars. In this scenario, no matter how far out you look, your line of sight will always hit a star, regardless of how small an area of the sky you observe. This means the entire sky should be as bright as a star!
Not to break it to you too harshly, but that’s not the case. The sky is dark. How can this be? Well, you might argue that there are nebulas, dust and gas clouds in space. These could block the light from the more distant stars, keeping the sky dark.
Sadly, this doesn’t work. Those clouds heat up from absorbing all that radiation and soon glow as brightly as any star. You haven’t solved the problem by adding nebulas; you’ve just delayed it by a bit. And an infinite universe is patient. Soon enough, it’ll be bright everywhere all the time, and you’re back to square one.
We call this seeming dark-sky contradiction Olbers’ paradox, after the German astronomer Heinrich Wilhelm Matthias Olbers, who wrote about it in his 1823 treatise “Über die Durchsichtigkeit des Weltraums” (“On the Transparency of Space”). I’ll note that in the grand tradition of Western nomenclature, it’s called this although Olbers was not the first person to think of or write about it—the question had been around for centuries, and the 16th-century English astronomer Thomas Digges discussed it when he proposed that the universe might be infinite in nature.
Various scholars tackled the issue, but none came up with a plausible scientific solution to Olbers’ paradox. That changed in 1848, however, when an American writer published an essay entitled Eureka: A Prose Poem. In it, he proposed that the universe was not infinite in extent and that, with the speed of light being finite as well, there has simply not been enough time for all that light to reach us:
The only mode, therefore, in which, under such a state of affairs, we could comprehend the voids which our telescopes find in innumerable directions, would be by supposing the distance of the invisible background so immense that no ray from it has yet been able to reach us at all.
Even though the essay doesn’t provide a quantitative mathematical solution—that would come a few decades later via the famous British physicist Lord Kelvin (William Thomson)—it is in principle correct, comprising one of two key parts of the correct explanation.
Who was the author of the essay? Edgar Allan Poe. I bet you’ll look at him the same way nevermore.
The other part of the answer is similar in that the universe has not always existed. In the early 20th century, astronomers, already suspicious that the universe was much larger than just the Milky Way, were becoming more certain. The discovery of the cosmic expansion discarded the static cosmos hypothesis, the view that the universe is unchanging, and, together with other observations, paved the way for the current big bang model. The big bang model doesn’t preclude a universe infinite in space, but, in support of Poe’s conjecture, it does predict a beginning, a cosmic Day One. In the end, we can only see so far, and the light from stars that are too far away cannot reach us.
Interestingly, the expansion of the cosmos would still explain the dark night even if it occurred in a steady-state universe. If the entire universe were growing larger, at some distance from Earth, that expansion would carry stars away too rapidly for their light to reach us. So it’s not so much that the dark sky proves the big bang model but rather that the big bang is a key aspect to understanding what’s going on.
Now we know that the universe got its start some 13.8 billion years ago, putting a limit to how far away we can see objects. The expansion dims their light, too, as they lose energy reaching us (creating the cosmic redshift). Also, stars have a finite lifetime, and once all the nebular gas is consumed to make them, the universe will dim, making the dark sky even darker over time.
Lord Kelvin himself said that there are no paradoxes in science; they are illusory conflicts that arise because of our limited understanding. The more we learn about the universe, the more so-called paradoxes themselves become dark and fall away.