In the late 1990s, astronomers discovered something mysterious pushing galaxies apart faster than gravity pulls them together. It seemed like every little bit of space had some amount of energy that spread it away from every other little bit of space, and that strange pushing came to be known as ' dark energy' - dark, because no one knows what it is.

And now a group of physicists have shown that dark energy could probably be explained - as long as we're willing to give up a fundamental piece of our understanding of light…

Most scientists think that dark energy exists because of what's known as a cosmological constant - something acting throughout the Universe that tells different bits of space to repel each other. It's sort of like an anti-gravity force, but it acts everywhere instead of just being between two things with mass and it always acts with the same strength.

The cosmological constant explanation works, but it's a hollow victory. Physicists don't like having numbers they can't explain - things like the mass of the electron, for example.

As far as we can tell, there isn't a way to really derive or predict an electron's mass using other physics. It's not because of something else; it's just that when the Universe formed some 14 billion years ago, it was able to generate these little bits of mass that, 14 billion years later, we started calling electrons. We just have to put this number (the mass) into our equations and begrudgingly accept that we can't explain it.

(The appeal of string theory is that it might be able to explain some of these numbers.)

Originally, it seemed like the cosmological constant wasn't like this. Using all of the physics we know right now, we can predict the strength of the cosmological constant. But instead of matching what we see, the result is known as the most embarrassing number in physics: if you multiply dark energy's measured strength by 10120 - that's a one with 120 zeros - you get the value we predict.

Hypotheses that have passed every single other test we've ever thrown at them predict a cosmological constant 100 million trillion googols stronger than dark energy. Dark energy is a tiny effect that doesn't match any of these giant predictions.

This prompted physicists to look for alternatives. If dark energy isn't caused by a cosmological constant, then the cosmological constant is actually equal to zero, and we don't have to worry as much about predicting it. So the hunt was on for something that might produce such a tiny repulsion.

One way to come up with something really small is to find something else that's really small and ask if the two might be related - like trying to explain a child's hair colour by asking what colour their parents' hair is.

This is what a team of physists led by Seyen Kouwn from the Korea Astronomy and Space Science Institute did - except they didn't just take something we already know to be small, like some previous groups have. Instead, they asked what would happen if light - which we've been assuming for a century and a half is massless - has some very tiny mass. And what they found was surprising.

Experiments have already proven that light can't be more massive than about 10-62 kilograms. That's a zero, a decimal point, 61 more zeros, and then a one, which is a really small number. But it's not zero, as Kouwn and his team point out.

They showed that if the photon's mass is 10 million times smaller than that limit, the way that photons interact with the different fields and forces in the Universe leads to a repulsive effect that looks an awful lot like what we've been calling dark energy. In other words, massive photons could cause dark energy.

Physicists probably won't be flocking to rewrite the textbooks, though. The proposed mass of the photon is, for all intents and purposes, immeasurably small, so using it instead of a cosmological constant to explain dark energy is trading something we can't verify for something we can't explain. And physicists dislike mechanisms they can't measure just as much as they dislike numbers they can't explain.

Plus, we're also trading one number we can't explain for another: why should photons have this exact mass?

It'll be interesting to see if groups working on other problems find that massive photons explain more than just dark energy. If light having a non-zero mass turns out to solve a bunch of other unsolved problems, physicists might warm up to this very strange idea.

The paper has been published in Physical Review D.