🫖 Kettle

Appearance

The Ancient Question

What is light? For centuries, physicists fought over the answer.

Newton said light was made of particles — tiny bullets shooting through space. Huygens said it was a wave, rippling through an invisible medium. They couldn't both be right.

Then came the 20th century, and quantum mechanics delivered a verdict that satisfied no one: both answers are correct. Light is neither purely wave nor purely particle. It's something stranger — something that seems to choose its nature based on how you look at it.

But what if you could delay that choice? What if you could wait until after the light had already traveled, and only then decide how to observe it?

Wave or Particle?

The evidence for wave-particle duality is undeniable, and deeply unsettling.

Send a photon through two slits, and it creates an interference pattern on the screen behind — bright and dark bands, the unmistakable signature of a wave. But put a detector at one slit to see which way the photon went, and the interference vanishes. Now you see two simple bands, as if particles were passing through one slit or the other.

The act of observing which path the photon took destroys the wave behavior. It's as if the photon knows whether you're watching.

This is complementarity: you can see the wave nature or the particle nature, but never both at once. The universe seems to enforce this rule absolutely.

Wheeler's Thought Experiment

In 1978, physicist John Wheeler asked a disturbing question: what if you make the choice after the photon has already passed the slits?

Imagine a Mach-Zehnder interferometer: a photon hits a beam splitter, which puts it in superposition — traveling both paths at once. The paths converge at a second beam splitter, which recombines them. If both paths are open, interference occurs. The photon exits predictably.

But what if you could remove that second beam splitter at the last instant? Without it, there's no recombination, no interference. You get which-path information instead — the photon reveals which route it took.

Here's the twist: you make this choice after the photon has already passed the first beam splitter. After it should have "decided" to be a wave or particle. After the past should be fixed.

The Paradox

How can a choice made now affect what the photon did before?

If you insert the second beam splitter, you see interference. The photon behaved like a wave, traveling both paths. If you remove it, you see which path it took. The photon behaved like a particle, traveling one path only.

But the photon already passed the first beam splitter! It already "chose" to go both ways or one way. Didn't it?

Wheeler's delayed-choice experiment seems to suggest that your choice now reaches back in time and determines what the photon did then. The past isn't fixed until you observe it.

This isn't just a thought experiment. It's been performed in laboratories around the world, with photons, atoms, and even molecules. The results always confirm quantum mechanics. The "delayed choice" works exactly as predicted.

See It Yourself

Run the experiment. Watch the photon "decide" based on a choice you make after it's already in flight.

With the second beam splitter in place, interference occurs. The photon always exits the same way — the waves from both paths combine constructively in one direction, destructively in the other. You'll see all 0s. Deterministic. Predictable. Wave behavior.

With the second beam splitter removed, there's no interference. The photon reveals which path it took: 50% chance of 0, 50% chance of 1. Pure randomness.

The classical mind rebels: "But it already went through the first beam splitter! It must have been on one path or both paths!" Quantum mechanics replies: that question has no answer until you choose how to measure.

A True Delayed Choice

The examples above have a problem: the choice is made before the code runs. Can we do better?

In the real experiment, the choice happens after the photon is already in superposition. Our earlier code passes the choice as a parameter — it's decided before the quantum operations even begin.

Here's a more faithful simulation: we use classical randomness (via Kettle's Random effect) to make the choice mid-execution, after the photon has already passed through the first beam splitter.

This is admittedly a simulation — we're using classical randomness to stand in for a human decision — but it captures the essential weirdness: the choice genuinely happens after the photon is in superposition, yet the photon's behavior still depends on that choice.

The Resolution

The paradox dissolves once you abandon a classical assumption: that the photon has a definite history before you observe it.

We instinctively believe the photon "really" went one way or both ways, and our measurement merely reveals this pre-existing fact. But quantum mechanics says otherwise. The photon doesn't have a path until the experiment is complete.

There's no backwards-in-time causation. There's no signal from the future to the past. The photon simply doesn't have a classical history. The question "which path did it take?" is meaningless until you define the entire experimental setup, including the final measurement.

As Wheeler himself put it: "No phenomenon is a phenomenon until it is an observed phenomenon." The past isn't determined until it's recorded.

Reality Waits

Wheeler's delayed-choice experiment isn't just a curiosity. It reveals something profound about the nature of reality.

The universe doesn't decide what happened until you ask. History isn't written until it's observed. This isn't philosophy — it's experimental fact, confirmed countless times with ever-increasing precision.

We live in a universe where the past is as uncertain as the future, where observation creates reality rather than merely revealing it, where a choice you make now determines what happened then.

The delayed choice is in your hands.