How to confuse yourself with quantum physics in 3 easy steps
None of the experts got it — why not you, too?
Quantum physics is weird. That’s what you’re supposed to think, anyway. But, also, it’s over one hundred years old and is the most accurate scientific theory ever created. Moreover, it provides the basis for all modern technology. Surely, then, someone understands what’s going on, right? …right?
I’ll let you decide for yourself. I’m going to show you the simplest set of experimental facts about quantum physics which displays the problem in its interpretation. Fair warning, though — there will be no answers here that will satisfy you. If nothing else, quantum physics shows us that Nature is much richer than the world constructed by our senses — and our sensibilities don’t like it.
Step 1: light is a wave
If you shine a laser in the following arrangement of mirrors and beam splitters — called a Mach-Zehnder interferometer — here is what will happen.
Because laser light — in fact, all light — is an electromagnetic wave, it can experience constructive and destructive interference. Depending on which way the laser light passes through the beam splitter, it may flip its orientation or not. For the top observer, the light taking the top path has reflected off the “back” of the beam splitter twice, while the light from the bottom path has not reflected off any beam splitter, passing through both. So the two reflections in the top path cancel, and the light “adds up” — that’s constructive interference.
For the bottom observer, the opposite is true. The light taking the top path reflects off the back of the first beam splitter but passes through the second — only one inversion. The light in the bottom path reflects off the second beam splitter, but it hits the “front” and does not invert. So, one path has inverted light for the bottom observer, and the other does not, forcing the light to “cancel out” — destructive interference.
By the way, there is nothing particularly special about this setup — it’s a fairly routine undergraduate physics lab. That the light gets inverted when reflecting off the back of a beam splitter and not the front might seem strange at first, but it has been well-understood since 1821. A simple analogy is in the reflection of a wave in a rope when it bounces off an end held fixed or let free. Don’t let this part confuse you because we haven’t even gotten to the quantum stuff yet!
Step 2: light is a particle
OK, the quantum stuff. As Einstein demonstrated in 1905, light is made of particles — later called photons — the discovery of which led to his Nobel Prize in Physics. So, let’s dial down the laser power until the observers detect one photon at a time — a brief flash of light. We could arrange it so that only once per minute a detection is observed. What do you expect will happen?
The first beam splitter will reflect or let pass each photon at random. Consider the case where a photon bounces off the first beam splitter. So, it’s on the upper path. When it hits the second beam splitter, it has a 50:50 chance to bounce or pass. That means both the top and bottom observers should report flashes — albeit periodically and at random. The same would be true if the photon originally passed through the first beam splitter.
Step 3: huh?
However, this is not what happens. Still, only the top observer sees the light. Once per minute, they report a flash, and the bottom observer stays forever in the dark. Somehow, the photon must be taking both paths and interfering with itself. We could verify this precisely if we wanted to by adjusting the “phase” in one path. For example, we could delay the photon in the top path by making the path longer. Then, the bottom observer would start to see flashes, but only according to exactly how much the light in the new paths overlaps. The light definitely takes both paths and interferes at the second beam splitter.
Just to throw a wrench in this, suppose a third observer is sitting on the bottom path watching for the photon. Suppose a photon is fired through, and the new observer doesn’t see anything. OK, no big deal, right? They didn’t disturb the photon or anything, so this shouldn’t change the result. But it does! They don’t see a photon which means no light is in the lower path. That means no light can interfere at the second beam splitter, and the final observers both see flashes at random. If the new observer does detect a photon, then the other two observers see nothing. This is entirely consistent with the photon having taken one path or the other. What gives?
Asking questions
If we try to spy on the photon, it acts like a particle. If we don’t look, it acts like a wave. In quantum physics, this is called wave-particle duality — light has wave-like and particle-like behavior. Which is it? Neither, both — depends who you ask. It’s definitely not exclusively one or the other. But to say it is one and the other is just nonsense. It’s something different, something we don’t really have a good everyday analogy for — nothing to directly compare it to in our experiences.
In over one hundred years, we still can’t agree on an answer to the question of how best to think about what’s happening here. The uncontroversial thing, the thing that all physicists do agree on, is how to very accurately predict what the observers will see. That can be done to astonishing precision. Explain the experimental setup to me, and I will predict exactly what will be observed. What happens in the intervening time? No one can say because to test it would require a different experimental setup, which quantum physics would then predict new outcomes for.
We could ask “what would have happened if…” — but no one agrees on answers to those questions, or even if such questions are worth asking. Students of quantum physics are simply told not to ask such questions — embodied by the famous phrase “shut up and calculate.” Calculations, of course, do need to be done, but maybe there is some deeper truth to that point of view that is not so dismissive. Maybe there is no answer to “what really happened” to the photon because our concepts of what constitutes acceptable answers are too limited.
The subtle irony is that — scientifically speaking — answers to “what really happened” are only useful because they give us models of the world that we can use to make accurate predictions. But, we can already do that within quantum physics. Yet, we demand more, asking questions we cannot provide empirical answers to. What’s really going on in our minds when we demand more? Well, there’s an interesting question.
