Heisenberg's Uncertainty Principle, Properly

In 1940, the physicist George Gamow published a thought experiment that has been haunting physics education ever since. In his book Mr. Tompkins in Wonderland, a professor places a billiard ball inside a wooden triangle on a table. Planck's constant, in this dream world, is enormous. Instead of sitting still, the ball races around inside its confinement. The lesson: restrict a particle's position and it must acquire a spread of momentum. Kinetic energy, from nowhere, forced into existence by confinement.

It's a vivid picture. It's also, as the Physics Explained channel lays out in a recent 47-minute video, not quite right, and the gap between Gamow's metaphor and the actual mathematics is worth taking seriously.

The shortcuts that work, and why that's a problem

The heuristic version of Heisenberg's uncertainty principle gets used constantly, and it genuinely earns its keep. Treat an electron as confined to an atom with a characteristic size of around 10^-10 meters, set delta-x equal to that size, rearrange the uncertainty relation, and you recover a momentum scale that gives you a kinetic energy of roughly four electron volts. That's the right ballpark for atomic physics. Push the same logic to nucleons inside a nucleus, tens of thousands of times smaller, and you correctly predict that protons and neutrons carry substantial kinetic energies even when the nucleus as a whole is at rest. Go further still, confining an electron to ever-smaller regions, and the energy scale eventually exceeds the rest mass energy needed for particle-antiparticle creation, which is precisely where single-particle quantum mechanics breaks down and quantum field theory has to take over.

These are not trivial results. The shortcuts work. But their success, as the Physics Explained video puts it, "can make it easy to forget what the symbols in the uncertainty principle actually mean."

What does delta-x actually mean? Is it the physical size of the confinement region? Is it some vague sense of how uncertain we are about where the particle is? And when we say the momentum uncertainty is delta-p, does that mean the electron has a definite momentum we just don't know, and is it literally bouncing around like Gamow's billiard ball?

The answer to all three questions is no, and the video works through why with unusual care.

Standard deviations, not error bars

The single-slit diffraction experiment is the vehicle. Fire electrons, one at a time, through a narrow slit. On a distant screen, they build up a pattern: most arrive near the center, fewer toward the edges, the familiar diffraction envelope. Because the distance to the screen is known and the forward momentum of each electron is fixed, the position where an electron lands on the screen can be translated directly into a transverse momentum value. The screen, in this framing, is the momentum distribution.

Run the experiment a million times. You get a histogram. Let the bins go infinitesimally narrow and you get a continuous probability distribution over transverse momentum values. The average of that distribution, because it's symmetric around zero, is zero. But two distributions can both average to zero while looking completely different: one tightly peaked, one spread wide. The average doesn't tell you which you're dealing with. The standard deviation does.

That's delta-p. Not a vague uncertainty, not an error bar, not a measure of how much some measurement has disturbed the electron. It's the standard deviation of the momentum probability distribution, calculated from the wave function describing the quantum state. Similarly, delta-x is the standard deviation of the position distribution, also calculated from the same wave function.

This is the point the video keeps returning to, and it's worth sitting with: "Delta X and Delta P are not error bars, they are not fundamentally measures of how one measurement disturbs another, they are standard deviations of the position and momentum distributions predicted by the same quantum state."

The Stanford Encyclopedia of Philosophy's entry on the uncertainty principle makes a related point: the uncertainties don't arise exclusively from measurement disturbances, but from the fundamental wave-particle nature of quantum systems. The Philosophy Institute puts it cleanly: "The uncertainty is not something added by observation; it is an intrinsic property of the quantum state."

Why the shape of the wave function matters

Here's where the video gets genuinely interesting, and where most popularizations don't go.

Take the idealized case: a wave function that's perfectly flat across the slit and drops abruptly to zero at the edges. This is a "top hat" or rectangular wave function. The position uncertainty for this state works out to the slit width divided by the square root of 12. Textbooks approximate this as roughly equal to the slit width, and that's fine for order-of-magnitude purposes.

But compute the momentum uncertainty for this sharp-edged wave function and you hit a problem. The momentum spread is determined by how the wave function changes with position, specifically by the gradient of the wave function. In the flat interior, the gradient is zero. Outside the slit, the gradient is also zero. The entire contribution to the momentum spread comes from the edges, where the wave function drops to zero. For a perfectly sharp rectangular wave function, those edges are infinitely steep. The gradient diverges. And so does the momentum uncertainty.

This isn't a quirk of the math. It's physically meaningful. Sharp edges in position space correspond to an enormous spread in momentum space. The video works through a family of smoothed wave functions, parameterized by a quantity epsilon that controls how gradually the edges taper to zero. As epsilon increases and the edges become smoother, delta-p decreases. As epsilon shrinks toward zero and the edges become sharper, delta-p diverges. The position uncertainty changes too, but comparatively modestly. The momentum distribution is far more sensitive to the shape of the wave function's edges.

This connects directly to what the uncertainty principle is actually constraining. Different wave function shapes give different values of delta-x and delta-p. The uncertainty principle says only that their product can never fall below h-bar over two. For the family of rounded top-hat wave functions in the video, the smoothest possible version (where the flat central region has vanished entirely) gives a product of about 0.568 h-bar. The principle is satisfied, but not saturated. The only wave function that actually achieves the minimum exactly is a Gaussian, which doesn't have compact support at all; its tails extend indefinitely in both directions.

That's a detail most introductions skip, and it matters. The uncertainty bound isn't something you can hit with just any reasonable-looking state. It's a hard floor approached asymptotically by increasingly optimal shapes.

This same structural argument, incidentally, is what prevents atoms from collapsing: quantum uncertainty keeps atoms stable not because electrons are buzzing around in some classical sense, but because a fully collapsed electron would require an infinitely sharp position distribution, driving the momentum spread, and therefore the kinetic energy, to infinity.

The measurement disturbance story is real but it's the wrong story

There's a version of the uncertainty principle that shows up in pop science constantly: you can't know both position and momentum precisely because measuring one disturbs the other. Look for an electron with a photon, the photon kicks the electron, you've scrambled the momentum. True enough as a description of a real experimental effect.

But it's not what the uncertainty relation means, and the Physics Explained video is unambiguous on this. Imagine preparing two million electrons in identical quantum states. Give the first million to one experimentalist, who measures only position. Give the second million to another, who measures only momentum. No particle ever has both measurements performed on it. There is no kick, no disturbance, no first measurement available to disturb a second. And yet the position distribution from the first batch and the momentum distribution from the second batch must still satisfy Heisenberg's inequality.

The constraint is on the distributions themselves, not on the act of measuring. It's baked into the wave function before anyone touches a detector.

This distinction matters beyond the philosophical. It's been at the center of real debates in foundations of physics, and the broader history of how the principle has been taught and misrepresented suggests the disturbance story has done genuine damage to public understanding of what quantum mechanics is actually claiming.

What the heuristics are actually doing

So where does that leave Gamow's billiard ball, and the standard textbook move of treating delta-p as a characteristic momentum?

Mostly intact, but reframed. In many confined or bound states, the average momentum is zero. When that's true, the standard deviation of the momentum distribution equals the root-mean-square momentum, which is a reasonable proxy for a "typical" momentum scale. So when a textbook writes p ~ h-bar / delta-x, it's not saying the electron has that specific momentum; it's saying that momentum scale is typical of what you'd measure if you sampled the momentum distribution many times. The estimate works because the physics cooperates. It's not because delta-p literally equals the momentum.

The billiard ball is a metaphor for the spread of outcomes, not a description of a trajectory. The electron is not racing back and forth. It doesn't have a path. What it has is a wave function that assigns probabilities to position and momentum measurements, and those two distributions are coupled by the mathematics of quantum mechanics in a way that prevents both from being simultaneously narrow.

That coupling, precise and quantitative and derivable from the wave function, is what Heisenberg's uncertainty principle actually says. The billiard ball is just how we make it feel less alien.

Amelia Nwofor is Science Desk Editor at Buzzrag.