Mass Cannot Turn You

Contents

1. The ladder, once more
2. A real number cannot turn you
3. Quantum writes the turn down
4. Momentum is how fast the phase winds
5. A boost is a rotation through an imaginary angle
6. The ghost axis does real work
7. So is collapse a hack?
8. What is load-bearing, and what is not
9. On “math is why”
10. Exercises

01 The ladder, once more

Take position and differentiate it against time, over and over. Velocity, acceleration, jerk — the third one also goes by jolt — then snap, crackle, pop. Seven rungs, all of them respectable mathematics. In part three we asked which of them a human body can feel, and the answer came back in stripes.

Write the motion the way part one taught us, as a point going around the complex plane: x = A e^iωt. Every derivative multiplies by iω, and multiplying by i is a quarter turn. So the ladder does not march off in one direction. It rotates, ninety degrees per rung, and after four rungs it is back where it started.

Figure 01

Seven derivatives, alternating between the felt axis and the ghost axis

1. 0thposition× 1felt
2. 1stvelocity× iωghost
3. 2ndacceleration× −ω²felt
4. 3rdjerk · jolt× −iω³ghost
5. 4thsnap× ω⁴felt
6. 5thcrackle× iω⁵ghost
7. 6thpop× −ω⁶felt

• Real multiple of x — reaches the body
• Imaginary multiple of x — perpendicular to feeling

That alternation is the whole inheritance from part three. Now we spend it.

02 A real number cannot turn you

Momentum is p = mv. Mass is a real, positive scalar — a kilogram is not an angle. And multiplying a complex number by a real number does exactly one thing: it stretches or shrinks the arrow. It cannot rotate it by so much as a degree. Only i rotates.

So momentum does not get a vote on which axis it lives on. It inherits velocity’s, exactly:

Read the second line slowly, because it answers a question most people have never thought to ask. Why is force something you feel and momentum something you only ever calculate? Not because one is fundamental and the other derived. Because they are ninety degrees apart. Force is momentum rotated onto the axis your body can read.

Check it against your own experience. You have never once felt a momentum. In a collision you feel the crumple — the force over the milliseconds of contact. On a train you feel the brakes, never the eighty tons. Momentum is bookkeeping that only ever pays out through a derivative. It is exactly as unfeelable as velocity, for exactly the same reason, and the reason is arithmetic: a real coefficient cannot move you off the axis you were already on.

Mass is not what makes momentum heavy. Mass is what makes momentum long. The direction was decided one rung earlier, and nothing classical ever changes it back.

03 Quantum writes the turn down

Here is where the lesson pays. Open any quantum mechanics textbook to the momentum operator and this is what you find:

Undergraduates are told the i is there “so the operator comes out Hermitian, so its eigenvalues are real.” That sentence is correct, and it is also the entire argument of this series stated in professional vocabulary. Unpack it.

The bare derivative ∂/∂x is anti-Hermitian. Integrate by parts and it picks up a minus sign; its expectation values come out pure imaginary. In the language we have been using: the derivative operator lives on the ghost axis. Nothing there can ever be the outcome of a measurement, because measurements produce real numbers.

So we multiply by −i. One quarter turn. And the object lands on the axis where answers are allowed to be.

Figure 02

Anatomy of the momentum operator

What you start with

∂/∂x

Anti-Hermitian. Expectation values pure imaginary. On the ghost axis, where nothing can be measured.

× (−iħ)  —  one quarter turn  ↓

What you end with

p̂ = −iħ ∂/∂x

Hermitian. Spectrum real. Now it is permitted to be an answer.

Classical mechanics performs this same rotation and never mentions it. It writes p = mv, quietly uses only the magnitude, and cashes everything out through F = dp/dt — which, as we saw, is the turn back to the real axis. The rotation happens; it is simply never written down. Quantum mechanics is the first formalism that puts the turn in the notation where you can see it. That is the honest answer to “why is momentum complex in quantum physics.” It was always complex. Quantum mechanics is where we stopped hiding it.

04 Momentum is how fast the phase winds

Push once more. A state of definite momentum is a plane wave, ψ = e^ipx/ħ, and de Broglie’s relation is p = ħk. Look at what that says. The exponent is a phase — an angle — and p is how quickly that angle turns as you walk along x.

Momentum is not a quantity that happens to be represented by a rotation. In quantum mechanics it is the rotation rate. High momentum is tight winding; low momentum is a lazy sweep. Nothing else is going on.

Figure 03

Two momenta, drawn as what they actually are

And that last caveat is not a retreat. It is the bridge to the classical world, and we will cross it in section 08.

05 A boost is a rotation through an imaginary angle

Before quantum mechanics gets all the credit, relativity said something almost identical, and said it first.

In special relativity a change of velocity — a boost — is not a shift. It is a hyperbolic rotation of spacetime, by an angle called rapidity, with tanh φ = v/c. Rotations in space use ordinary angles. Boosts use hyperbolic ones. And the two are the same function with one substitution, because cos(iφ) = cosh φ and sin(iφ) = i sinh φ.

Poincaré noticed this in 1905, published in 1906, and wrote the fourth coordinate as ct√−1 so that a Lorentz transformation became literally a rotation in a four-dimensional space. Minkowski adopted the device and made it famous with x₄ = ict. In that formalism the sentence “velocity is an imaginary rotation angle” is not a metaphor. It is the notation.

06 The ghost axis does real work

If the imaginary axis were mere bookkeeping you would expect it to stay out of the laboratory. It does not.

Send a particle at a barrier taller than its energy. Classically it stops. Quantum mechanically, solve for the momentum inside the barrier and you get p = √(2m(E − V)) with E < V — the square root of a negative number. The momentum in that region is imaginary. And the wave does not oscillate there; it decays, ψ ∼ e^−κx. Then it comes out the far side, smaller, oscillating again.

Figure 04

Where momentum goes imaginary, and what happens to the wave

The pattern generalises, and it is worth stating precisely because it is the strongest form of the intuition this series has been building. Complex momenta and complex energies show up throughout real physics — in tunnelling, in the instanton paths that describe it, in Gamow’s decaying nuclear states with energy E − iΓ/2, in bound states appearing as poles at imaginary momentum in scattering theory, in the ringdown frequencies of colliding black holes that LIGO measures. In every case the imaginary part encodes decay or attenuation, and in no case is it the number a momentum detector reports.

07 So is collapse a hack?

Now the claim that started this. Look at what a measurement actually does to a wavefunction. The Born rule says the probability of finding the particle at x is |ψ(x)|² — modulus squared. Write ψ = R e^iS/ħ and the operation is naked: |ψ|² = R². The phase is deleted. And the phase, as section 04 established, is precisely where momentum lives.

So the instinct is right about the mechanics. Measurement is the step at which the ghost axis is thrown away and a real number is handed to you. If you have been following the series, collapse is not a mysterious extra physical event; it is the moment the formalism stops carrying the part you cannot feel.

And “hack” puts you in serious company. John Bell spent the last years of his life saying so in print. In Beables for quantum field theory (1984) he wrote that conventional formulations of quantum theory are “unprofessionally vague and ambiguous,” adding that professional theoretical physicists ought to be able to do better. In Against ‘measurement’ (1990) he named the specific offence: the axioms anchor there “the shifty split of the world into ‘system’ and ‘apparatus’” — shifty because nothing in the theory says where to put the line.

But three things have to be said or the lesson becomes propaganda.

Decoherence explains most of what collapse was invented to explain, without collapse. A system entangled with its environment loses interference terms extremely fast, and this is measured, not conjectured. What decoherence does not deliver is a single outcome: it turns “this and that” into a list, not into “this or that.” That residue is the measurement problem, and it is still open.

Serious theories exist in which collapse never happens at all. Everett’s and Bohm’s formulations are fully unitary; nothing is ever discarded, and the appearance of a single result is explained another way. If those are right, then collapse is not a hack, it is a fiction — a stronger claim than the one being made here.

And there are theories in which collapse is entirely physical. GRW and continuous spontaneous localisation add a real stochastic term to the Schrödinger equation. They make different predictions from standard quantum mechanics, experiments have been squeezing their parameter space for years, and they have not been ruled out. If one of those is right, collapse is not a hack either — it is a law we had not written down yet.

So: a live and respectable position, held by people who thought about it harder than anyone. Not a settled fact. Teach it as the first.

08 What is load-bearing, and what is not

Every lesson in this series should be checkable, so here is the ledger for this one.

Load-bearing — this is just true

Mass is a real scalar and cannot rotate a complex quantity, so p = mv puts momentum on exactly the axis velocity was on. F = dp/dt is one quarter turn back onto the felt axis. ∂/∂x really is anti-Hermitian and p̂ really is that operator rotated by −i so its spectrum is real. The Born rule really does delete the phase. Boosts really are rotations by an imaginary angle in the ict formalism.

Convention, not physics

The minus sign in −iħ ∂/∂x is a choice; +iħ ∂/∂x gives identical physics. And ict is one representation of Lorentz geometry among several — the one the field abandoned.

Where the shorthand breaks

“Momentum is a complex number” is false. Every momentum ever measured is real, and it has to be, or probability would not be conserved. What is complex is the operator, the state, and the phase — never the reading.

Where I would push back on myself

Complexness is not what separates quantum from classical. Classical mechanics can be written on a complex Hilbert space too (Koopman–von Neumann), and Hamilton had phase and wavefronts in the 1830s with no ħ anywhere. The real dividing line is superposition and interference; complex numbers are the efficient way to write those, not the source of them. Anyone who tells you the i is why quantum mechanics is strange has skipped a step.

And the word “collapse to the real plane” is doing too much

Classical mechanics is not the real part of quantum mechanics. Taking Re(ψ) is meaningless — the phase convention is arbitrary and the result is not a state. The classical limit is an asymptotic limit, not a projection.

That last one deserves its replacement, because the true version is better than the slogan.

Write ψ = R e^iS/ħ and let ħ become small against the action. What survives is the Hamilton–Jacobi equation of classical mechanics, with classical momentum given by p = ∂S/∂x. Classical momentum is the gradient of the quantum phase. And in Feynman’s formulation the reason a classical trajectory exists at all is that everywhere else the phases from neighbouring paths cancel; along the classical path they agree, and the cancellation stops.

So the felt world is not the real axis of the complex world. It is the ridge where the phases stop cancelling — the stationary-phase skeleton of a complex amplitude. That is a more interesting claim than the one I set out to defend, it is what the mathematics actually says, and it keeps the shape of the intuition intact: what you can feel is a thin residue of a much larger structure, and the structure is not optional.

09 On “math is why”

This series runs on a claim borrowed from Max Tegmark: the physical world does not obey a mathematical structure, it is one, and what we call physical properties are positions inside it. His Mathematical Universe Hypothesis is the sharpest modern statement of the idea — The Mathematical Universe in Foundations of Physics (2008), then the 2014 book.

He is not alone in it, and the students should know the lineage: Pythagoras and Plato in the obvious sense; Eugene Wigner’s 1960 essay on the unreasonable effectiveness of mathematics in the natural sciences, which posed the problem Tegmark answers; John Wheeler’s “it from bit”; Roger Penrose’s three worlds; and, in contemporary philosophy of science, structural realism — John Worrall’s cautious 1989 version, on which structure is what we can know, and the stronger ontic version named by James Ladyman in 1998 and developed with Steven French, on which structure is all there is. Tegmark’s version is the most committed and the most willing to take the consequences on the chin. But he is the sharpest point of a long argument, not the only one making it.

10 Exercises

1. Extend the ladder Add momentum, force, and the time derivative of force (“yank”) to Figure 01 as rungs. Confirm each lands where the multiplier says. What does yank being on the ghost axis predict about whether you can feel it?
2. Break the argument with kinetic energy Kinetic energy is ½mv². Squaring an imaginary quantity gives a real one. So energy sits on the felt axis while momentum does not — and energy is conserved alongside momentum. Is that a problem for the picture, or a prediction of it?
3. Find the real eigenvalue Show directly that ∂/∂x is anti-Hermitian by integrating by parts, and that multiplying by −i fixes it. Then find where the boundary terms you discarded went, and say what physical assumption you just made.
4. Kill the metaphor Take a particle in a box in its ground state. The wavefunction is real — zero phase, zero winding — and yet ⟨p²⟩ is not zero. Explain how that is consistent with section 04, and rewrite section 04’s claim so it survives this case.
5. Argue the other side Take the Everettian position — nothing ever collapses — and make the strongest case you can in one paragraph. Then say what it costs you.