We can split an atom, simulate the Big Bang, and map a black hole’s gravitational field. We can track eclipses to the second. But as of 2025, physicists still don’t know what time actually is. They can’t even agree on whether it’s real or an illusion.
Meanwhile, you’re just out here… living. Brushing your teeth. Sleeping at night. Acting like the glowing digits on the microwave are gospel. All while the deepest rift in physics remains unsolved: Quantum mechanics and general relativity, the two pillars of modern science, do not agree on time. One says it’s fixed and absolute. The other says it’s bendy and local. Both can’t be right. How could you possibly eat at a time like this?!
“It’s kind of embarrassing,” says Chiara Marletto, Ph.D., an Italian theoretical physicist and a research fellow at Oxford University. “Time appears in equations. But physics doesn’t tell you what it actually is.”
Legendary British physicist Julian Barbour, Ph.D., has been publishing on this topic since the 1970s. Now 88, he doesn’t mince words: “People have been trying to solve the problem of time for 70 years. And they haven’t solved it. There are physicists who say, ‘Oh, it’s not a problem.’ I think they’re doing a disservice to science.”
Enter constructor theory, a radical physics rethink that’s been quietly gaining buzz over the past decade. A few months back, Marletto and her collaborator, David Deutsch, Ph.D., a British quantum physicist, aimed the theory at physics’ most persistent blindspot: time itself. Instead of just asking what time is in their new paper, they suggest we’ve been thinking backward.
“Laws of motion are reversible at the microscopic scale,” Marletto says. “And yet we live in an irreversible world. That’s a very old problem. Constructor theory is a new statement about it.”
This flips the usual narrative. In classical physics, time’s arrow is an afterthought, a statistical artifact of how disorder tends to increase. Entropy rises not because it has to, but because it’s just wildly improbable for entropy to do anything else.
Constructor theory doesn’t buy that. Instead of shrugging at improbability, it builds irreversibility into the foundations. If no machine (a “constructor”) can perform the reverse task, then that task is fundamentally impossible.
“A scrambled egg doesn’t reassemble because there’s no constructor that can do it reliably,” says Marletto.
Time has a direction because certain things are simply impossible to reverse. No clever machine or process can reliably undo them. It’s the difference between “won’t happen” and “can’t happen.” Traditional physics focuses on what won’t happen due to astronomical odds. Constructor theory identifies what actually can’t happen, period.
That’s also where clocks come in.
A clock is just a system that reliably performs a sequence of transformations, ticking forward in a predictable order. (Without clocks, things would still evolve and change. You just couldn’t measure when or how long.) You don’t start with “time” and then build a clock. You start with what physical transformations are possible. Clocks (and time) emerge from that. Yesterday, from constructor theory’s point of view, is nothing mystical, just the current arrangement of the universe plus the records that point backward.
If constructor theory’s take on time holds up, it could reshape the foundations of physics. The direction of time would be a built-in feature of the universe’s operating manual. That could change how we approach the biggest questions: How did the universe begin? What is consciousness? How does physical reality actually work?
Not everyone is convinced. “It sounds intelligent,” says Sabine Hossenfelder, Ph.D., a German theoretical physicist whose 1.7 million YouTube followers have made her one of science’s most influential and outspoken skeptics. “But I can’t say I understand what it means for a clock to ‘make’ time.” Hossenfelder agrees that physics may need new conceptual tools. She’s just not sure this is the one.
Then again, people said the same about Deutsch in the 1980s, when he first proposed quantum computing. His early papers were dismissed as too philosophical. Now he’s credited as the father of quantum computing, defining a field that’s gone from thought experiment to the multibillion dollar global race to build computers that could solve certain problems exponentially faster than any traditional machine. “It’s a little like that,” Marletto says. “At first you get strange looks. Then someone takes it seriously.”
Still, she admits that she and Deutsch could be wrong. “The problem with science right now is that we use the same mathematical frameworks over and over,” Marletto says. “Everything becomes a nail.” Constructor theory is an attempt to escape that by embracing possibility, including the possibility of failure. “You have to be able to be wrong,” she says. “Otherwise we just stay in the same place.”
In a moment of deep skepticism about science, it feels almost subversive to hear a physicist embrace the possibility of being wrong. “It’s a good way to look at things,” says Marletto. “You don’t stop asking just because you don’t have the right tools yet.”
But until constructor theory is validated, as Barbour says, “all we can prove about time is the present, you and me, here, now. The rest is memory, or speculation. So carpe diem.”