The universe is dead; long live the universe.

Not right at this moment, not yet. But one day everything we know will be gone. The cities we build, the lakes we swim in, the planet we live on, the solar system we inhabit, the star we orbit and every star we don’t – they’re all headed towards an inescapable finale.

At the end of it all, what happens? Some say our ever-expanding universe will slow down and then one day do a cosmic U-turn, undoing all the growth that has happened since the big bang. Eventually, everything will crunch together into the tiniest possible space and then explode out again in a riot of rebirth – that’s the idea we call cyclic cosmology, or the big bounce. It’s been around for a long time, and the idea itself has faced a trajectory that mirrors its contents. It was briefly popular in the mid-20th century, fell from favour, and now it may be making a comeback thanks to new data from the largest 3D map of the universe ever created, made by the Dark Energy Spectroscopic Instrument (DESI).

As is occasionally the case with grand cosmological hypotheses, proponents of cyclic cosmology mostly preferred it for its elegance: if the universe is cyclic, that means we probably don’t have to worry about what precipitated the big bang or what existed before it – those near-impossible questions are already answered. There’s a beautiful sense of symmetry to the whole thing. Catherine Heymans, the Astronomer Royal for Scotland, expressed it nicely during a recent New Scientist subscriber event I hosted, where she said, “It really gels with me that the universe sort of is created in a big bang, it expands, it slows down, gravity pulls it back in on itself, there’s a big crunch, there’s another big bang and it expands… This just makes me very happy.”

At the same event, Adam Riess, who won a Nobel for his and his colleagues’ discovery of dark energy, put forward one of the more concrete reasons for many cosmologists’ fondness for the idea. “We like it because it tells us that this is not a special time that we live in or the one-shot universe,” he said. In other words, in a cyclic universe it wouldn’t be quite such an unbelievable coincidence that we’re here at all to ponder these things. Personally, I don’t think the idea that times like this happen over and over again – perhaps not in every bounce but definitely in more than one – with all the right conditions for life and trees and rockets to the moon, makes it that much less special, but I digress. Maybe that’s more of an anthropocentric, emotional position than one based on the laws of physics.

For a long time, cyclic cosmology fell out of favour, driven partly by Riess’s work showing that the universe is expanding at an accelerating rate. If the space between the stars is growing faster and faster, it feels unlikely that it will eventually shrink back down to nothing again. Gravity just isn’t strong enough to counteract dark energy. “Unfortunately, all of the measurements that we make tell us that there just isn’t enough mass in the universe to pull it back together,” said Heymans. “At the moment, the evidence is pointing towards a very cold and sad and empty death for our universe.” This idea, called the heat death, is now the most accepted version of what is to come.

There are various other reasons that the big bounce faded into relative obscurity, largely to do with problems that arise when we try to sort out how matter, energy and entropy might be recycled or destroyed in the moment between bounces.

The second law of thermodynamics is a sticking point: it says that disorder, or entropy, in a closed system (such as the universe, as far as we know) can never decrease. With an expanding universe, that’s easy to square – we would just see a continual slow increase in entropy over the lifetime of the cosmos. But if the universe starts contracting again, entropy would correspondingly start to decrease. There are ways around this, generally involving pushing the problem off into the next cycle of expansion and contraction. If the universe gets bigger in each cycle, entropy is still increasing overall. But if you extrapolate backwards or forwards in time enough, you end up in the same situation as before. We still start with a big bang at the beginning of the universe and an eventual heat death at the end, it’s just a more complicated, stepwise path between the two.

Another way around the entropy problem was popularised in the 2010s by legendary theoretical physicist Roger Penrose, of Penrose triangle fame. His model is called conformal cyclic cosmology, and it would look exactly like an ever-expanding universe… right until the very end. As the universe expands and everything gets further and further from everything else, matter will decay into its composite parts, and eventually everything will just be leftover photons floating in the abyss. That’s not particularly controversial. But what Penrose proposed next is. His idea is that the extreme emptiness and uniformity of space-time at the end of one cycle, or aeon, is the same as the structure we’d expect at the very beginning of a new aeon. The idea behind conformal cyclic cosmology is that thanks to this functionally identical structure (and some very complicated maths), a new, expanding universe can be kicked off from the frigid remains of the previous one.

The idea is niche and difficult (bordering on impossible) to test. Penrose has proposed some potentially measurable bits of evidence for it, but on the whole, cosmologists tend to find them unconvincing. However, it hasn’t been disproven either, and the fact that it manages to get around the entropy problem means that it shouldn’t be simply discarded, even if it is widely viewed with scepticism. So, we’re stuck without much of a way to apply these ideas to the real universe we live in.

Enter DESI. Its enormous map of the universe has shown that dark energy, which previously looked like it would only grow in strength forever, seems to be weakening. That is, the outward acceleration of the universe appears to be slowing. As Heymans stressed during the event, this does not mean that the universe is coming back together – it’s still accelerating in its expansion, just not quite as quickly.  Still, this is a radical shift in our understanding of dark energy, and it could well kick off an era of new theories about the way our cosmos will spend its final days.

And among those new theories, cyclic cosmologies seem to be rising once again. “What could be causing dark energy to change could mean that in another 10 billion years’ time, dark energy weakens so much that it does reverse and it does pull everything back in on itself, which would be lovely,” said Heymans.

The problem with knowing what it all means is that we don’t understand a massive proportion of the universe. Dark energy makes up nearly 70 per cent of all the matter and energy in the entire cosmos; it controls the ultimate fate of everything, and yet we have no idea what it is or how it works. On cosmological time scales, we just met – Reiss and his colleagues only identified it less than 30 years ago.

“Without understanding the nature of the dark energy that’s driving the present acceleration, it’s very difficult to extrapolate it into the future. Will it weaken?” said Riess. “I would say all bets are off about the future.” Smart money may still be on a cold and empty end of the universe, but, for the first time in a century, it might also be worth placing a long-shot wager on the big bounce.