In 2023 the Nobel committee handed the physics prize to three people who had, in effect, built a camera with a shutter fast enough to photograph an electron in mid-motion.1 The shutter speed is measured in attoseconds. An attosecond is a billionth of a billionth of a second, and the number you're meant to be staggered by is this: there are about as many attoseconds in a single second as there have been seconds since the Big Bang.2 We can now freeze a slice of time that thin. We can sit and watch the smallest things there are do the thing they do.
I think this is genuinely thrilling, and I want that on the record before I spoil it. Because here's what happens when you finally get down to the actual floor of reality. The floor stops behaving like a floor.
The bottom is strange, and it's leaking
The quantum weirdness we like to keep quarantined inside physics departments has, it turns out, been seeping into living things the whole time. A European robin finds its way to Africa partly by sensing the Earth's magnetic field, and the leading explanation is that it does this through a quantum process running in a protein in its eye. Light knocks a pair of electrons into entangled spin states, and the chemistry that follows depends on the angle to the magnetic field. The bird is, in a real and almost unbearable sense, reading a quantum instrument.3
Photosynthesis may pull a related trick, shuttling energy through quantum coherence on its way to the reaction centre. I'll be straight with you: that one has been fought over for fifteen years, and the field has quietly retreated from the boldest version of the claim. The coherences are real; whether the plant is "using" them as computation is very much not settled.4 I mention it because honesty is the brand, and because the careful version is strange enough.
Reality starts looking like information
Here's the part that does something to people. The deeper you drill, the less the bottom of the world resembles tiny billiard balls and the more it resembles information. John Wheeler, one of the great physicists of the last century, spent his final decades on a four-word slogan: it from bit. The notion that the universe is, at its root, made of the answers to yes-or-no questions.5 There is a whole respectable-adjacent field that takes seriously the idea that reality is something being computed. And the moment "reality is computation" is on the table, another idea strolls in behind it, grinning.
In 2003 the philosopher Nick Bostrom set the argument out cleanly. If any civilisation could ever run detailed simulations of conscious minds, it would likely run an enormous number of them. The simulated minds would then vastly outnumber the un-simulated ones. So, statistically, you should bet you're one of the many and not one of the rare few in base reality.6 Elon Musk does a cartoon of this at parties: the odds we're in base reality, he likes to say, are billions to one.
Strip out the vocabulary and look at what's left. A vast intelligence designed the reality you live in, for purposes you cannot access, and the world in front of your face is not the base layer. That is not a new idea. That is Aquinas with a server farm.
Notice who's saying it
The simulation hypothesis is most at home in exactly the rooms where Genesis gets the indulgent smile. People far too sophisticated to think a mind made the world will tell you, evenly, that the universe was authored, designed, and is being run on purpose by an intelligence outside it. They just won't say God. They'll say the simulators.
That's the cheap shot, and I put it in the headline, so let me argue against myself for a moment. Bostrom is not quietly smuggling God back in. His argument is a piece of probability: if conscious simulations are possible, and if someone runs a great many of them, then most minds like ours are simulated, so you should weight your bet that way. That is not a creation myth, it's a counting argument, and it stands or falls on premises about computation and consciousness that nobody can currently check. "Aquinas with a server farm" is a good line. I'm genuinely not sure it's a fair one.
The objection that actually slows me down is a different one. A simulation good enough to fake all of our physics would, by design, leave no trace inside that physics for us to find. So the claim may not be testable even in principle, and a claim that nothing could ever count against is not really a scientific claim at all. Sabine Hossenfelder has made roughly this case bluntly: the popular version isn't daring science, it's faith in a lab coat, and so far nobody can reproduce even known physics from any step-by-step computation.7 But that knife has two edges. If "we are simulated" can't be falsified, then neither can my tidy reply that it's "just religion." Both of us are doing theology and giving it a different name.
So here is where I actually come down, which is not on a punchline. Attoseconds and quantum robins show that reality is stranger and finer-grained than animals built to dodge leopards can intuit. They do not show that it was built. The step from "weird at the bottom" to "someone made it" is not in the data, it's in us. And I can't always tell, from the inside, whether I'm following a real clue or just running the same machinery that once put a god in every thunderstorm. I notice the pull. I don't trust it. I also can't get rid of it, and I've stopped believing the people who claim they have.
❧
Notes & Sources
- The 2023 Nobel Prize in Physics went to Pierre Agostini, Ferenc Krausz and Anne L'Huillier "for experimental methods that generate attosecond pulses of light for the study of electron dynamics in matter." NobelPrize.org.
- An attosecond is 10-18 s. There are 1018 of them in one second; the universe is roughly 4.3 × 1017 seconds old, hence the (order-of-magnitude) comparison used in the Nobel materials.
- On avian magnetoreception and the radical-pair mechanism in cryptochrome: Hore, P. J. & Mouritsen, H. (2016), "The Radical-Pair Mechanism of Magnetoreception," Annual Review of Biophysics. Link.
- Coherence in photosynthesis was reported by Engel et al. (2007), Nature 446, 782. The interpretation has since been heavily debated; for the more sober current view see Cao et al. (2020), "Quantum biology revisited," Science Advances. Link.
- Wheeler, J. A. (1990), "Information, Physics, Quantum: The Search for Links" — the "it from bit" essay.
- Bostrom, N. (2003), "Are You Living in a Computer Simulation?", Philosophical Quarterly 53(211):243–255. Full text.
- For the counter-case: Hossenfelder, S., "The Simulation Hypothesis is Pseudoscience" (2021), and Existential Physics (2022). No one can presently reproduce general relativity or the Standard Model from a step-by-step computation, and an untestable claim is not science. Link.